{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.4081971510967404,
  "eval_steps": 3000,
  "global_step": 39000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.637994861602783,
      "epoch": 0.0004369483527047103,
      "grad_norm": 1.0390625,
      "learning_rate": 2e-06,
      "loss": 7.5197,
      "mean_token_accuracy": 0.0836352527141571,
      "num_tokens": 10478.0,
      "step": 5
    },
    {
      "entropy": 7.623341083526611,
      "epoch": 0.0008738967054094206,
      "grad_norm": 1.1015625,
      "learning_rate": 4.5e-06,
      "loss": 7.5301,
      "mean_token_accuracy": 0.08603608533740044,
      "num_tokens": 21152.0,
      "step": 10
    },
    {
      "entropy": 7.645789051055909,
      "epoch": 0.0013108450581141308,
      "grad_norm": 1.1484375,
      "learning_rate": 7e-06,
      "loss": 7.552,
      "mean_token_accuracy": 0.08498126044869422,
      "num_tokens": 31246.0,
      "step": 15
    },
    {
      "entropy": 7.679685497283936,
      "epoch": 0.0017477934108188413,
      "grad_norm": 0.97265625,
      "learning_rate": 9.5e-06,
      "loss": 7.5895,
      "mean_token_accuracy": 0.07903653383255005,
      "num_tokens": 41934.0,
      "step": 20
    },
    {
      "entropy": 7.792749309539795,
      "epoch": 0.0021847417635235517,
      "grad_norm": 0.9375,
      "learning_rate": 1.2e-05,
      "loss": 7.5444,
      "mean_token_accuracy": 0.08353051990270614,
      "num_tokens": 52830.0,
      "step": 25
    },
    {
      "entropy": 7.786162900924682,
      "epoch": 0.0026216901162282617,
      "grad_norm": 0.9375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 7.5,
      "mean_token_accuracy": 0.08886836394667626,
      "num_tokens": 62706.0,
      "step": 30
    },
    {
      "entropy": 7.815877103805542,
      "epoch": 0.003058638468932972,
      "grad_norm": 1.015625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 7.5486,
      "mean_token_accuracy": 0.08427426405251026,
      "num_tokens": 72619.0,
      "step": 35
    },
    {
      "entropy": 7.870150184631347,
      "epoch": 0.0034955868216376825,
      "grad_norm": 1.390625,
      "learning_rate": 1.95e-05,
      "loss": 7.5448,
      "mean_token_accuracy": 0.08383063822984696,
      "num_tokens": 82903.0,
      "step": 40
    },
    {
      "entropy": 7.800839138031006,
      "epoch": 0.003932535174342393,
      "grad_norm": 0.921875,
      "learning_rate": 2.2e-05,
      "loss": 7.5341,
      "mean_token_accuracy": 0.09498083963990211,
      "num_tokens": 94243.0,
      "step": 45
    },
    {
      "entropy": 7.756237649917603,
      "epoch": 0.004369483527047103,
      "grad_norm": 1.015625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 7.4765,
      "mean_token_accuracy": 0.08740946426987647,
      "num_tokens": 103618.0,
      "step": 50
    },
    {
      "entropy": 7.747218179702759,
      "epoch": 0.004806431879751813,
      "grad_norm": 0.98046875,
      "learning_rate": 2.7e-05,
      "loss": 7.5363,
      "mean_token_accuracy": 0.08639938347041606,
      "num_tokens": 112180.0,
      "step": 55
    },
    {
      "entropy": 7.8047871589660645,
      "epoch": 0.005243380232456523,
      "grad_norm": 1.015625,
      "learning_rate": 2.95e-05,
      "loss": 7.5337,
      "mean_token_accuracy": 0.08548481613397599,
      "num_tokens": 122462.0,
      "step": 60
    },
    {
      "entropy": 7.848284912109375,
      "epoch": 0.005680328585161234,
      "grad_norm": 0.97265625,
      "learning_rate": 3.2e-05,
      "loss": 7.5602,
      "mean_token_accuracy": 0.08545567914843559,
      "num_tokens": 132741.0,
      "step": 65
    },
    {
      "entropy": 7.7641541957855225,
      "epoch": 0.006117276937865944,
      "grad_norm": 0.90625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 7.3675,
      "mean_token_accuracy": 0.09678877294063568,
      "num_tokens": 142973.0,
      "step": 70
    },
    {
      "entropy": 7.738112068176269,
      "epoch": 0.006554225290570655,
      "grad_norm": 1.0625,
      "learning_rate": 3.7e-05,
      "loss": 7.5937,
      "mean_token_accuracy": 0.08664538338780403,
      "num_tokens": 153199.0,
      "step": 75
    },
    {
      "entropy": 7.7370462894439695,
      "epoch": 0.006991173643275365,
      "grad_norm": 1.0546875,
      "learning_rate": 3.95e-05,
      "loss": 7.4445,
      "mean_token_accuracy": 0.09458591938018798,
      "num_tokens": 163079.0,
      "step": 80
    },
    {
      "entropy": 7.807093095779419,
      "epoch": 0.0074281219959800755,
      "grad_norm": 1.0234375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 7.4872,
      "mean_token_accuracy": 0.08863211423158646,
      "num_tokens": 173308.0,
      "step": 85
    },
    {
      "entropy": 7.725804328918457,
      "epoch": 0.007865070348684786,
      "grad_norm": 1.078125,
      "learning_rate": 4.45e-05,
      "loss": 7.5445,
      "mean_token_accuracy": 0.09568502977490426,
      "num_tokens": 183856.0,
      "step": 90
    },
    {
      "entropy": 7.763433265686035,
      "epoch": 0.008302018701389496,
      "grad_norm": 1.046875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.4676,
      "mean_token_accuracy": 0.08726829290390015,
      "num_tokens": 193368.0,
      "step": 95
    },
    {
      "entropy": 7.7914244651794435,
      "epoch": 0.008738967054094207,
      "grad_norm": 1.0,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 7.5028,
      "mean_token_accuracy": 0.08547773025929928,
      "num_tokens": 202686.0,
      "step": 100
    },
    {
      "entropy": 7.707356786727905,
      "epoch": 0.009175915406798917,
      "grad_norm": 1.0390625,
      "learning_rate": 5.2e-05,
      "loss": 7.4556,
      "mean_token_accuracy": 0.08999297693371773,
      "num_tokens": 213237.0,
      "step": 105
    },
    {
      "entropy": 7.722362422943116,
      "epoch": 0.009612863759503626,
      "grad_norm": 1.0,
      "learning_rate": 5.45e-05,
      "loss": 7.4589,
      "mean_token_accuracy": 0.09247651025652885,
      "num_tokens": 222406.0,
      "step": 110
    },
    {
      "entropy": 7.7970898151397705,
      "epoch": 0.010049812112208336,
      "grad_norm": 1.1328125,
      "learning_rate": 5.7e-05,
      "loss": 7.5187,
      "mean_token_accuracy": 0.0885585032403469,
      "num_tokens": 232139.0,
      "step": 115
    },
    {
      "entropy": 7.722547149658203,
      "epoch": 0.010486760464913047,
      "grad_norm": 1.0703125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 7.4757,
      "mean_token_accuracy": 0.08926801681518555,
      "num_tokens": 242438.0,
      "step": 120
    },
    {
      "entropy": 7.695586252212524,
      "epoch": 0.010923708817617757,
      "grad_norm": 0.984375,
      "learning_rate": 6.2e-05,
      "loss": 7.4684,
      "mean_token_accuracy": 0.08990140855312348,
      "num_tokens": 253082.0,
      "step": 125
    },
    {
      "entropy": 7.861955642700195,
      "epoch": 0.011360657170322468,
      "grad_norm": 1.1171875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.4714,
      "mean_token_accuracy": 0.09366248250007629,
      "num_tokens": 264067.0,
      "step": 130
    },
    {
      "entropy": 7.758950090408325,
      "epoch": 0.011797605523027178,
      "grad_norm": 0.9296875,
      "learning_rate": 6.7e-05,
      "loss": 7.4561,
      "mean_token_accuracy": 0.08862813860177994,
      "num_tokens": 274571.0,
      "step": 135
    },
    {
      "entropy": 7.681786012649536,
      "epoch": 0.012234553875731888,
      "grad_norm": 0.984375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.4413,
      "mean_token_accuracy": 0.09196381121873856,
      "num_tokens": 284561.0,
      "step": 140
    },
    {
      "entropy": 7.749648571014404,
      "epoch": 0.012671502228436599,
      "grad_norm": 0.9296875,
      "learning_rate": 7.2e-05,
      "loss": 7.3464,
      "mean_token_accuracy": 0.09177442193031311,
      "num_tokens": 295366.0,
      "step": 145
    },
    {
      "entropy": 7.664951229095459,
      "epoch": 0.01310845058114131,
      "grad_norm": 1.296875,
      "learning_rate": 7.45e-05,
      "loss": 7.5296,
      "mean_token_accuracy": 0.08601538985967636,
      "num_tokens": 305423.0,
      "step": 150
    },
    {
      "entropy": 7.767955684661866,
      "epoch": 0.01354539893384602,
      "grad_norm": 1.015625,
      "learning_rate": 7.7e-05,
      "loss": 7.4251,
      "mean_token_accuracy": 0.08934596404433251,
      "num_tokens": 315182.0,
      "step": 155
    },
    {
      "entropy": 7.662592172622681,
      "epoch": 0.01398234728655073,
      "grad_norm": 1.078125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.459,
      "mean_token_accuracy": 0.08355334773659706,
      "num_tokens": 324780.0,
      "step": 160
    },
    {
      "entropy": 7.84730715751648,
      "epoch": 0.01441929563925544,
      "grad_norm": 0.93359375,
      "learning_rate": 8.2e-05,
      "loss": 7.5123,
      "mean_token_accuracy": 0.0899798959493637,
      "num_tokens": 335104.0,
      "step": 165
    },
    {
      "entropy": 7.775908184051514,
      "epoch": 0.014856243991960151,
      "grad_norm": 1.1875,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.4946,
      "mean_token_accuracy": 0.08976865485310555,
      "num_tokens": 345073.0,
      "step": 170
    },
    {
      "entropy": 7.828866195678711,
      "epoch": 0.015293192344664861,
      "grad_norm": 1.109375,
      "learning_rate": 8.7e-05,
      "loss": 7.4994,
      "mean_token_accuracy": 0.08743183836340904,
      "num_tokens": 354342.0,
      "step": 175
    },
    {
      "entropy": 7.731208658218383,
      "epoch": 0.015730140697369572,
      "grad_norm": 0.9140625,
      "learning_rate": 8.95e-05,
      "loss": 7.4269,
      "mean_token_accuracy": 0.08921334892511368,
      "num_tokens": 364441.0,
      "step": 180
    },
    {
      "entropy": 7.805326890945435,
      "epoch": 0.01616708905007428,
      "grad_norm": 1.015625,
      "learning_rate": 9.2e-05,
      "loss": 7.4345,
      "mean_token_accuracy": 0.08910228461027145,
      "num_tokens": 373918.0,
      "step": 185
    },
    {
      "entropy": 7.738444805145264,
      "epoch": 0.016604037402778993,
      "grad_norm": 1.0859375,
      "learning_rate": 9.45e-05,
      "loss": 7.5209,
      "mean_token_accuracy": 0.08797775506973267,
      "num_tokens": 383273.0,
      "step": 190
    },
    {
      "entropy": 7.697931146621704,
      "epoch": 0.0170409857554837,
      "grad_norm": 1.140625,
      "learning_rate": 9.7e-05,
      "loss": 7.357,
      "mean_token_accuracy": 0.09655481651425361,
      "num_tokens": 393262.0,
      "step": 195
    },
    {
      "entropy": 7.8257646560668945,
      "epoch": 0.017477934108188414,
      "grad_norm": 1.1171875,
      "learning_rate": 9.95e-05,
      "loss": 7.4976,
      "mean_token_accuracy": 0.08942469507455826,
      "num_tokens": 402389.0,
      "step": 200
    },
    {
      "entropy": 7.703541660308838,
      "epoch": 0.017914882460893122,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000102,
      "loss": 7.4401,
      "mean_token_accuracy": 0.09125433415174485,
      "num_tokens": 412997.0,
      "step": 205
    },
    {
      "entropy": 7.828316783905029,
      "epoch": 0.018351830813597834,
      "grad_norm": 1.125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.4592,
      "mean_token_accuracy": 0.0881546139717102,
      "num_tokens": 422222.0,
      "step": 210
    },
    {
      "entropy": 7.572913932800293,
      "epoch": 0.018788779166302543,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000107,
      "loss": 7.4023,
      "mean_token_accuracy": 0.09697630628943443,
      "num_tokens": 432539.0,
      "step": 215
    },
    {
      "entropy": 7.832680416107178,
      "epoch": 0.019225727519007252,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0001095,
      "loss": 7.4286,
      "mean_token_accuracy": 0.09254593029618263,
      "num_tokens": 442311.0,
      "step": 220
    },
    {
      "entropy": 7.6697183609008786,
      "epoch": 0.019662675871711964,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000112,
      "loss": 7.3919,
      "mean_token_accuracy": 0.0872996911406517,
      "num_tokens": 452727.0,
      "step": 225
    },
    {
      "entropy": 7.759046792984009,
      "epoch": 0.020099624224416673,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001145,
      "loss": 7.4466,
      "mean_token_accuracy": 0.09114575162529945,
      "num_tokens": 462461.0,
      "step": 230
    },
    {
      "entropy": 7.685237169265747,
      "epoch": 0.020536572577121385,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.4258,
      "mean_token_accuracy": 0.08979431167244911,
      "num_tokens": 472362.0,
      "step": 235
    },
    {
      "entropy": 7.747664165496826,
      "epoch": 0.020973520929826094,
      "grad_norm": 1.15625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.4573,
      "mean_token_accuracy": 0.08877176120877266,
      "num_tokens": 482718.0,
      "step": 240
    },
    {
      "entropy": 7.68355770111084,
      "epoch": 0.021410469282530806,
      "grad_norm": 1.171875,
      "learning_rate": 0.000122,
      "loss": 7.3976,
      "mean_token_accuracy": 0.09123231843113899,
      "num_tokens": 492947.0,
      "step": 245
    },
    {
      "entropy": 7.7995494365692135,
      "epoch": 0.021847417635235514,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001245,
      "loss": 7.4648,
      "mean_token_accuracy": 0.08322979286313056,
      "num_tokens": 502957.0,
      "step": 250
    },
    {
      "entropy": 7.740732431411743,
      "epoch": 0.022284365987940227,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000127,
      "loss": 7.4177,
      "mean_token_accuracy": 0.08514684960246086,
      "num_tokens": 513220.0,
      "step": 255
    },
    {
      "entropy": 7.732230281829834,
      "epoch": 0.022721314340644935,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001295,
      "loss": 7.3827,
      "mean_token_accuracy": 0.08738726824522018,
      "num_tokens": 523217.0,
      "step": 260
    },
    {
      "entropy": 7.6845520496368405,
      "epoch": 0.023158262693349647,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000132,
      "loss": 7.4167,
      "mean_token_accuracy": 0.0847338579595089,
      "num_tokens": 533789.0,
      "step": 265
    },
    {
      "entropy": 7.76469612121582,
      "epoch": 0.023595211046054356,
      "grad_norm": 1.046875,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.3767,
      "mean_token_accuracy": 0.09299462959170342,
      "num_tokens": 544146.0,
      "step": 270
    },
    {
      "entropy": 7.644073247909546,
      "epoch": 0.02403215939875907,
      "grad_norm": 1.109375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.4063,
      "mean_token_accuracy": 0.09182784631848336,
      "num_tokens": 554215.0,
      "step": 275
    },
    {
      "entropy": 7.711794137954712,
      "epoch": 0.024469107751463777,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001395,
      "loss": 7.4101,
      "mean_token_accuracy": 0.08731252029538154,
      "num_tokens": 564587.0,
      "step": 280
    },
    {
      "entropy": 7.801581239700317,
      "epoch": 0.024906056104168486,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.4581,
      "mean_token_accuracy": 0.08540895506739617,
      "num_tokens": 574819.0,
      "step": 285
    },
    {
      "entropy": 7.712231540679932,
      "epoch": 0.025343004456873198,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001445,
      "loss": 7.3978,
      "mean_token_accuracy": 0.08907487243413925,
      "num_tokens": 585178.0,
      "step": 290
    },
    {
      "entropy": 7.57398829460144,
      "epoch": 0.025779952809577907,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000147,
      "loss": 7.3785,
      "mean_token_accuracy": 0.09169624000787735,
      "num_tokens": 595086.0,
      "step": 295
    },
    {
      "entropy": 7.753789281845092,
      "epoch": 0.02621690116228262,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001495,
      "loss": 7.4015,
      "mean_token_accuracy": 0.08911786824464799,
      "num_tokens": 604942.0,
      "step": 300
    },
    {
      "entropy": 7.740912914276123,
      "epoch": 0.026653849514987327,
      "grad_norm": 1.171875,
      "learning_rate": 0.000152,
      "loss": 7.365,
      "mean_token_accuracy": 0.09200693666934967,
      "num_tokens": 616082.0,
      "step": 305
    },
    {
      "entropy": 7.668862152099609,
      "epoch": 0.02709079786769204,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.4392,
      "mean_token_accuracy": 0.0856046225875616,
      "num_tokens": 626375.0,
      "step": 310
    },
    {
      "entropy": 7.629800462722779,
      "epoch": 0.027527746220396748,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000157,
      "loss": 7.3741,
      "mean_token_accuracy": 0.0859305128455162,
      "num_tokens": 637563.0,
      "step": 315
    },
    {
      "entropy": 7.764728212356568,
      "epoch": 0.02796469457310146,
      "grad_norm": 1.125,
      "learning_rate": 0.0001595,
      "loss": 7.3935,
      "mean_token_accuracy": 0.09147773832082748,
      "num_tokens": 647256.0,
      "step": 320
    },
    {
      "entropy": 7.690927219390869,
      "epoch": 0.02840164292580617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000162,
      "loss": 7.4189,
      "mean_token_accuracy": 0.0867665283381939,
      "num_tokens": 657247.0,
      "step": 325
    },
    {
      "entropy": 7.781694221496582,
      "epoch": 0.02883859127851088,
      "grad_norm": 1.125,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.3252,
      "mean_token_accuracy": 0.08868500515818596,
      "num_tokens": 667613.0,
      "step": 330
    },
    {
      "entropy": 7.653351449966431,
      "epoch": 0.02927553963121559,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.3939,
      "mean_token_accuracy": 0.09035046473145485,
      "num_tokens": 678048.0,
      "step": 335
    },
    {
      "entropy": 7.669305658340454,
      "epoch": 0.029712487983920302,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.3691,
      "mean_token_accuracy": 0.09391902461647987,
      "num_tokens": 688468.0,
      "step": 340
    },
    {
      "entropy": 7.645442485809326,
      "epoch": 0.03014943633662501,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.3628,
      "mean_token_accuracy": 0.09365171119570732,
      "num_tokens": 699809.0,
      "step": 345
    },
    {
      "entropy": 7.640151834487915,
      "epoch": 0.030586384689329723,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.3112,
      "mean_token_accuracy": 0.0941974550485611,
      "num_tokens": 709227.0,
      "step": 350
    },
    {
      "entropy": 7.659973907470703,
      "epoch": 0.03102333304203443,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000177,
      "loss": 7.4041,
      "mean_token_accuracy": 0.08795862942934037,
      "num_tokens": 720178.0,
      "step": 355
    },
    {
      "entropy": 7.749969959259033,
      "epoch": 0.031460281394739144,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001795,
      "loss": 7.3598,
      "mean_token_accuracy": 0.0944575697183609,
      "num_tokens": 730514.0,
      "step": 360
    },
    {
      "entropy": 7.634895277023316,
      "epoch": 0.03189722974744385,
      "grad_norm": 1.25,
      "learning_rate": 0.000182,
      "loss": 7.4221,
      "mean_token_accuracy": 0.09183157607913017,
      "num_tokens": 741345.0,
      "step": 365
    },
    {
      "entropy": 7.632371330261231,
      "epoch": 0.03233417810014856,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001845,
      "loss": 7.3275,
      "mean_token_accuracy": 0.09244873151183128,
      "num_tokens": 751163.0,
      "step": 370
    },
    {
      "entropy": 7.697724151611328,
      "epoch": 0.03277112645285327,
      "grad_norm": 1.0625,
      "learning_rate": 0.000187,
      "loss": 7.3143,
      "mean_token_accuracy": 0.08760180100798606,
      "num_tokens": 762090.0,
      "step": 375
    },
    {
      "entropy": 7.604908990859985,
      "epoch": 0.033208074805557986,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001895,
      "loss": 7.299,
      "mean_token_accuracy": 0.09711357578635216,
      "num_tokens": 772049.0,
      "step": 380
    },
    {
      "entropy": 7.570729303359985,
      "epoch": 0.03364502315826269,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000192,
      "loss": 7.2903,
      "mean_token_accuracy": 0.0953090362250805,
      "num_tokens": 781280.0,
      "step": 385
    },
    {
      "entropy": 7.673049736022949,
      "epoch": 0.0340819715109674,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001945,
      "loss": 7.3234,
      "mean_token_accuracy": 0.09910958260297775,
      "num_tokens": 791642.0,
      "step": 390
    },
    {
      "entropy": 7.579310750961303,
      "epoch": 0.034518919863672115,
      "grad_norm": 1.125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.3467,
      "mean_token_accuracy": 0.09289622604846955,
      "num_tokens": 801224.0,
      "step": 395
    },
    {
      "entropy": 7.6497889995574955,
      "epoch": 0.03495586821637683,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.2947,
      "mean_token_accuracy": 0.092826696485281,
      "num_tokens": 810727.0,
      "step": 400
    },
    {
      "entropy": 7.73345537185669,
      "epoch": 0.03539281656908153,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000202,
      "loss": 7.3824,
      "mean_token_accuracy": 0.08923227414488792,
      "num_tokens": 819943.0,
      "step": 405
    },
    {
      "entropy": 7.517602396011353,
      "epoch": 0.035829764921786245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.2848,
      "mean_token_accuracy": 0.09139850735664368,
      "num_tokens": 829474.0,
      "step": 410
    },
    {
      "entropy": 7.638075828552246,
      "epoch": 0.03626671327449096,
      "grad_norm": 1.1875,
      "learning_rate": 0.000207,
      "loss": 7.2934,
      "mean_token_accuracy": 0.0908023789525032,
      "num_tokens": 839288.0,
      "step": 415
    },
    {
      "entropy": 7.616179609298706,
      "epoch": 0.03670366162719567,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002095,
      "loss": 7.2845,
      "mean_token_accuracy": 0.09343387261033058,
      "num_tokens": 850064.0,
      "step": 420
    },
    {
      "entropy": 7.491023874282837,
      "epoch": 0.037140609979900374,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000212,
      "loss": 7.2923,
      "mean_token_accuracy": 0.09214133247733117,
      "num_tokens": 860732.0,
      "step": 425
    },
    {
      "entropy": 7.562995529174804,
      "epoch": 0.037577558332605086,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002145,
      "loss": 7.2648,
      "mean_token_accuracy": 0.09335138648748398,
      "num_tokens": 870076.0,
      "step": 430
    },
    {
      "entropy": 7.482726287841797,
      "epoch": 0.0380145066853098,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.1828,
      "mean_token_accuracy": 0.09552065506577492,
      "num_tokens": 880201.0,
      "step": 435
    },
    {
      "entropy": 7.537555170059204,
      "epoch": 0.038451455038014504,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002195,
      "loss": 7.2454,
      "mean_token_accuracy": 0.09459976553916931,
      "num_tokens": 890344.0,
      "step": 440
    },
    {
      "entropy": 7.510470342636109,
      "epoch": 0.038888403390719216,
      "grad_norm": 1.171875,
      "learning_rate": 0.000222,
      "loss": 7.2468,
      "mean_token_accuracy": 0.09458372145891189,
      "num_tokens": 900443.0,
      "step": 445
    },
    {
      "entropy": 7.594428253173828,
      "epoch": 0.03932535174342393,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002245,
      "loss": 7.2734,
      "mean_token_accuracy": 0.09156661406159401,
      "num_tokens": 910284.0,
      "step": 450
    },
    {
      "entropy": 7.5778489112854,
      "epoch": 0.03976230009612864,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.3465,
      "mean_token_accuracy": 0.09012126699090003,
      "num_tokens": 920986.0,
      "step": 455
    },
    {
      "entropy": 7.58350772857666,
      "epoch": 0.040199248448833345,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.2518,
      "mean_token_accuracy": 0.09617682099342346,
      "num_tokens": 930481.0,
      "step": 460
    },
    {
      "entropy": 7.555335330963135,
      "epoch": 0.04063619680153806,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.2733,
      "mean_token_accuracy": 0.09550615251064301,
      "num_tokens": 940554.0,
      "step": 465
    },
    {
      "entropy": 7.51325364112854,
      "epoch": 0.04107314515424277,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.2389,
      "mean_token_accuracy": 0.09795199707150459,
      "num_tokens": 950005.0,
      "step": 470
    },
    {
      "entropy": 7.518522882461548,
      "epoch": 0.04151009350694748,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000237,
      "loss": 7.2443,
      "mean_token_accuracy": 0.0910925917327404,
      "num_tokens": 959414.0,
      "step": 475
    },
    {
      "entropy": 7.530915355682373,
      "epoch": 0.04194704185965219,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002395,
      "loss": 7.259,
      "mean_token_accuracy": 0.09173793643712998,
      "num_tokens": 969352.0,
      "step": 480
    },
    {
      "entropy": 7.570693063735962,
      "epoch": 0.0423839902123569,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000242,
      "loss": 7.2917,
      "mean_token_accuracy": 0.08883628323674202,
      "num_tokens": 980294.0,
      "step": 485
    },
    {
      "entropy": 7.439288902282715,
      "epoch": 0.04282093856506161,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002445,
      "loss": 7.1819,
      "mean_token_accuracy": 0.09918189346790314,
      "num_tokens": 990453.0,
      "step": 490
    },
    {
      "entropy": 7.474934530258179,
      "epoch": 0.04325788691776632,
      "grad_norm": 1.390625,
      "learning_rate": 0.000247,
      "loss": 7.1861,
      "mean_token_accuracy": 0.10081736147403716,
      "num_tokens": 1000509.0,
      "step": 495
    },
    {
      "entropy": 7.503030824661255,
      "epoch": 0.04369483527047103,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002495,
      "loss": 7.3407,
      "mean_token_accuracy": 0.08743650615215301,
      "num_tokens": 1011747.0,
      "step": 500
    },
    {
      "entropy": 7.507284879684448,
      "epoch": 0.04413178362317574,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000252,
      "loss": 7.2963,
      "mean_token_accuracy": 0.086452054977417,
      "num_tokens": 1021511.0,
      "step": 505
    },
    {
      "entropy": 7.546348524093628,
      "epoch": 0.04456873197588045,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002545,
      "loss": 7.302,
      "mean_token_accuracy": 0.08823695406317711,
      "num_tokens": 1032116.0,
      "step": 510
    },
    {
      "entropy": 7.467855310440063,
      "epoch": 0.04500568032858516,
      "grad_norm": 1.125,
      "learning_rate": 0.000257,
      "loss": 7.1938,
      "mean_token_accuracy": 0.09948761016130447,
      "num_tokens": 1041356.0,
      "step": 515
    },
    {
      "entropy": 7.516360902786255,
      "epoch": 0.04544262868128987,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002595,
      "loss": 7.2624,
      "mean_token_accuracy": 0.09001169204711915,
      "num_tokens": 1051314.0,
      "step": 520
    },
    {
      "entropy": 7.522716856002807,
      "epoch": 0.04587957703399458,
      "grad_norm": 1.078125,
      "learning_rate": 0.000262,
      "loss": 7.2206,
      "mean_token_accuracy": 0.09518473073840142,
      "num_tokens": 1060857.0,
      "step": 525
    },
    {
      "entropy": 7.510475397109985,
      "epoch": 0.046316525386699295,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.2541,
      "mean_token_accuracy": 0.095691280066967,
      "num_tokens": 1071744.0,
      "step": 530
    },
    {
      "entropy": 7.459283924102783,
      "epoch": 0.046753473739404,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.2165,
      "mean_token_accuracy": 0.09503942206501961,
      "num_tokens": 1080777.0,
      "step": 535
    },
    {
      "entropy": 7.485194826126099,
      "epoch": 0.04719042209210871,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.2191,
      "mean_token_accuracy": 0.09791651666164398,
      "num_tokens": 1090951.0,
      "step": 540
    },
    {
      "entropy": 7.405230283737183,
      "epoch": 0.047627370444813424,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.1784,
      "mean_token_accuracy": 0.09603556096553803,
      "num_tokens": 1101667.0,
      "step": 545
    },
    {
      "entropy": 7.466827440261841,
      "epoch": 0.04806431879751814,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002745,
      "loss": 7.3023,
      "mean_token_accuracy": 0.08709683492779732,
      "num_tokens": 1112652.0,
      "step": 550
    },
    {
      "entropy": 7.492521619796753,
      "epoch": 0.04850126715022284,
      "grad_norm": 1.21875,
      "learning_rate": 0.000277,
      "loss": 7.232,
      "mean_token_accuracy": 0.09334004074335098,
      "num_tokens": 1122518.0,
      "step": 555
    },
    {
      "entropy": 7.511386775970459,
      "epoch": 0.048938215502927554,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002795,
      "loss": 7.2466,
      "mean_token_accuracy": 0.09488805904984474,
      "num_tokens": 1132867.0,
      "step": 560
    },
    {
      "entropy": 7.544691753387451,
      "epoch": 0.049375163855632266,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.33,
      "mean_token_accuracy": 0.08940334022045135,
      "num_tokens": 1142524.0,
      "step": 565
    },
    {
      "entropy": 7.499535369873047,
      "epoch": 0.04981211220833697,
      "grad_norm": 1.25,
      "learning_rate": 0.0002845,
      "loss": 7.1969,
      "mean_token_accuracy": 0.1014377273619175,
      "num_tokens": 1152983.0,
      "step": 570
    },
    {
      "entropy": 7.482742214202881,
      "epoch": 0.050249060561041684,
      "grad_norm": 1.0625,
      "learning_rate": 0.000287,
      "loss": 7.2659,
      "mean_token_accuracy": 0.09228436276316643,
      "num_tokens": 1164271.0,
      "step": 575
    },
    {
      "entropy": 7.432041788101197,
      "epoch": 0.050686008913746396,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002895,
      "loss": 7.1369,
      "mean_token_accuracy": 0.0982656441628933,
      "num_tokens": 1173567.0,
      "step": 580
    },
    {
      "entropy": 7.360566711425781,
      "epoch": 0.05112295726645111,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000292,
      "loss": 7.194,
      "mean_token_accuracy": 0.09685536473989487,
      "num_tokens": 1183619.0,
      "step": 585
    },
    {
      "entropy": 7.399991750717163,
      "epoch": 0.05155990561915581,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002945,
      "loss": 7.1451,
      "mean_token_accuracy": 0.09111583679914474,
      "num_tokens": 1193922.0,
      "step": 590
    },
    {
      "entropy": 7.478466176986695,
      "epoch": 0.051996853971860525,
      "grad_norm": 1.203125,
      "learning_rate": 0.000297,
      "loss": 7.2197,
      "mean_token_accuracy": 0.09398753717541694,
      "num_tokens": 1203665.0,
      "step": 595
    },
    {
      "entropy": 7.44222354888916,
      "epoch": 0.05243380232456524,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002995,
      "loss": 7.23,
      "mean_token_accuracy": 0.09689271599054336,
      "num_tokens": 1214119.0,
      "step": 600
    },
    {
      "entropy": 7.460528373718262,
      "epoch": 0.05287075067726995,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000302,
      "loss": 7.2427,
      "mean_token_accuracy": 0.09010300114750862,
      "num_tokens": 1225275.0,
      "step": 605
    },
    {
      "entropy": 7.30951566696167,
      "epoch": 0.053307699029974655,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003045,
      "loss": 7.1342,
      "mean_token_accuracy": 0.09279084131121636,
      "num_tokens": 1235199.0,
      "step": 610
    },
    {
      "entropy": 7.454565858840942,
      "epoch": 0.05374464738267937,
      "grad_norm": 1.109375,
      "learning_rate": 0.000307,
      "loss": 7.1556,
      "mean_token_accuracy": 0.0968701608479023,
      "num_tokens": 1245648.0,
      "step": 615
    },
    {
      "entropy": 7.305359888076782,
      "epoch": 0.05418159573538408,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003095,
      "loss": 7.1208,
      "mean_token_accuracy": 0.10165121629834176,
      "num_tokens": 1254793.0,
      "step": 620
    },
    {
      "entropy": 7.404820203781128,
      "epoch": 0.05461854408808879,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000312,
      "loss": 7.1588,
      "mean_token_accuracy": 0.0971425749361515,
      "num_tokens": 1264429.0,
      "step": 625
    },
    {
      "entropy": 7.337283086776734,
      "epoch": 0.055055492440793496,
      "grad_norm": 1.125,
      "learning_rate": 0.0003145,
      "loss": 7.1419,
      "mean_token_accuracy": 0.0944690503180027,
      "num_tokens": 1274449.0,
      "step": 630
    },
    {
      "entropy": 7.475021743774414,
      "epoch": 0.05549244079349821,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000317,
      "loss": 7.2603,
      "mean_token_accuracy": 0.09432755038142204,
      "num_tokens": 1284987.0,
      "step": 635
    },
    {
      "entropy": 7.418339729309082,
      "epoch": 0.05592938914620292,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003195,
      "loss": 7.2474,
      "mean_token_accuracy": 0.08950235322117805,
      "num_tokens": 1295984.0,
      "step": 640
    },
    {
      "entropy": 7.26748685836792,
      "epoch": 0.056366337498907626,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000322,
      "loss": 7.1544,
      "mean_token_accuracy": 0.09836009740829468,
      "num_tokens": 1305738.0,
      "step": 645
    },
    {
      "entropy": 7.405171012878418,
      "epoch": 0.05680328585161234,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.0997,
      "mean_token_accuracy": 0.09074313640594482,
      "num_tokens": 1315691.0,
      "step": 650
    },
    {
      "entropy": 7.292159748077393,
      "epoch": 0.05724023420431705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.1171,
      "mean_token_accuracy": 0.10072707086801529,
      "num_tokens": 1326195.0,
      "step": 655
    },
    {
      "entropy": 7.338443660736084,
      "epoch": 0.05767718255702176,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.1405,
      "mean_token_accuracy": 0.09912769198417663,
      "num_tokens": 1336916.0,
      "step": 660
    },
    {
      "entropy": 7.30134425163269,
      "epoch": 0.05811413090972647,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.1773,
      "mean_token_accuracy": 0.09219239056110382,
      "num_tokens": 1347386.0,
      "step": 665
    },
    {
      "entropy": 7.414137363433838,
      "epoch": 0.05855107926243118,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.1208,
      "mean_token_accuracy": 0.09784709438681602,
      "num_tokens": 1357959.0,
      "step": 670
    },
    {
      "entropy": 7.263702297210694,
      "epoch": 0.05898802761513589,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000337,
      "loss": 7.1354,
      "mean_token_accuracy": 0.09500606879591941,
      "num_tokens": 1368135.0,
      "step": 675
    },
    {
      "entropy": 7.372778415679932,
      "epoch": 0.059424975967840604,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003395,
      "loss": 7.1653,
      "mean_token_accuracy": 0.0939582608640194,
      "num_tokens": 1378702.0,
      "step": 680
    },
    {
      "entropy": 7.19251708984375,
      "epoch": 0.05986192432054531,
      "grad_norm": 1.1875,
      "learning_rate": 0.000342,
      "loss": 7.0456,
      "mean_token_accuracy": 0.09847457632422447,
      "num_tokens": 1387902.0,
      "step": 685
    },
    {
      "entropy": 7.3456488132476805,
      "epoch": 0.06029887267325002,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.0601,
      "mean_token_accuracy": 0.10286817029118538,
      "num_tokens": 1397288.0,
      "step": 690
    },
    {
      "entropy": 7.266854667663575,
      "epoch": 0.060735821025954734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000347,
      "loss": 7.0971,
      "mean_token_accuracy": 0.10038763061165809,
      "num_tokens": 1407416.0,
      "step": 695
    },
    {
      "entropy": 7.384059572219849,
      "epoch": 0.061172769378659446,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003495,
      "loss": 7.1717,
      "mean_token_accuracy": 0.09187826812267304,
      "num_tokens": 1417733.0,
      "step": 700
    },
    {
      "entropy": 7.187748289108276,
      "epoch": 0.06160971773136415,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000352,
      "loss": 7.0684,
      "mean_token_accuracy": 0.09974993579089642,
      "num_tokens": 1426723.0,
      "step": 705
    },
    {
      "entropy": 7.310714626312256,
      "epoch": 0.06204666608406886,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003545,
      "loss": 7.1289,
      "mean_token_accuracy": 0.0992344006896019,
      "num_tokens": 1437290.0,
      "step": 710
    },
    {
      "entropy": 7.245534753799438,
      "epoch": 0.062483614436773575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000357,
      "loss": 7.0715,
      "mean_token_accuracy": 0.09555437043309212,
      "num_tokens": 1446756.0,
      "step": 715
    },
    {
      "entropy": 7.302251720428467,
      "epoch": 0.06292056278947829,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003595,
      "loss": 7.1289,
      "mean_token_accuracy": 0.09620894640684127,
      "num_tokens": 1457074.0,
      "step": 720
    },
    {
      "entropy": 7.223730421066284,
      "epoch": 0.06335751114218299,
      "grad_norm": 1.265625,
      "learning_rate": 0.000362,
      "loss": 7.2065,
      "mean_token_accuracy": 0.08963397517800331,
      "num_tokens": 1467785.0,
      "step": 725
    },
    {
      "entropy": 7.307017707824707,
      "epoch": 0.0637944594948877,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003645,
      "loss": 7.0911,
      "mean_token_accuracy": 0.09854559674859047,
      "num_tokens": 1477532.0,
      "step": 730
    },
    {
      "entropy": 7.1604907512664795,
      "epoch": 0.06423140784759242,
      "grad_norm": 1.09375,
      "learning_rate": 0.000367,
      "loss": 7.0269,
      "mean_token_accuracy": 0.10236552879214286,
      "num_tokens": 1487102.0,
      "step": 735
    },
    {
      "entropy": 7.2588645935058596,
      "epoch": 0.06466835620029712,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003695,
      "loss": 7.1184,
      "mean_token_accuracy": 0.0926779955625534,
      "num_tokens": 1497459.0,
      "step": 740
    },
    {
      "entropy": 7.252214527130127,
      "epoch": 0.06510530455300184,
      "grad_norm": 1.015625,
      "learning_rate": 0.000372,
      "loss": 7.1014,
      "mean_token_accuracy": 0.09459270238876342,
      "num_tokens": 1508756.0,
      "step": 745
    },
    {
      "entropy": 7.249363708496094,
      "epoch": 0.06554225290570655,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003745,
      "loss": 7.021,
      "mean_token_accuracy": 0.10529743134975433,
      "num_tokens": 1518735.0,
      "step": 750
    },
    {
      "entropy": 7.237057495117187,
      "epoch": 0.06597920125841125,
      "grad_norm": 1.09375,
      "learning_rate": 0.000377,
      "loss": 7.0688,
      "mean_token_accuracy": 0.09851523265242576,
      "num_tokens": 1528632.0,
      "step": 755
    },
    {
      "entropy": 7.1528160572052,
      "epoch": 0.06641614961111597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003795,
      "loss": 7.0595,
      "mean_token_accuracy": 0.09594285860657692,
      "num_tokens": 1538180.0,
      "step": 760
    },
    {
      "entropy": 7.243620729446411,
      "epoch": 0.06685309796382068,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000382,
      "loss": 7.0977,
      "mean_token_accuracy": 0.0974743902683258,
      "num_tokens": 1547788.0,
      "step": 765
    },
    {
      "entropy": 7.187331247329712,
      "epoch": 0.06729004631652538,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003845,
      "loss": 7.017,
      "mean_token_accuracy": 0.100215582549572,
      "num_tokens": 1558008.0,
      "step": 770
    },
    {
      "entropy": 7.164813470840454,
      "epoch": 0.0677269946692301,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.081,
      "mean_token_accuracy": 0.09390448704361916,
      "num_tokens": 1567728.0,
      "step": 775
    },
    {
      "entropy": 7.256727600097657,
      "epoch": 0.0681639430219348,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.0475,
      "mean_token_accuracy": 0.09773643538355828,
      "num_tokens": 1578488.0,
      "step": 780
    },
    {
      "entropy": 7.129490995407105,
      "epoch": 0.06860089137463951,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.0126,
      "mean_token_accuracy": 0.10168901681900025,
      "num_tokens": 1588070.0,
      "step": 785
    },
    {
      "entropy": 7.209209394454956,
      "epoch": 0.06903783972734423,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.0928,
      "mean_token_accuracy": 0.09427050650119781,
      "num_tokens": 1598572.0,
      "step": 790
    },
    {
      "entropy": 7.211539697647095,
      "epoch": 0.06947478808004894,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.0894,
      "mean_token_accuracy": 0.09861121252179146,
      "num_tokens": 1607982.0,
      "step": 795
    },
    {
      "entropy": 7.200946855545044,
      "epoch": 0.06991173643275365,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003995,
      "loss": 7.1132,
      "mean_token_accuracy": 0.09658989906311036,
      "num_tokens": 1618785.0,
      "step": 800
    },
    {
      "entropy": 7.177917003631592,
      "epoch": 0.07034868478545836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000402,
      "loss": 7.0225,
      "mean_token_accuracy": 0.09970952942967415,
      "num_tokens": 1627988.0,
      "step": 805
    },
    {
      "entropy": 7.200639438629151,
      "epoch": 0.07078563313816306,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004045,
      "loss": 7.1263,
      "mean_token_accuracy": 0.09254502132534981,
      "num_tokens": 1638024.0,
      "step": 810
    },
    {
      "entropy": 7.2738360404968265,
      "epoch": 0.07122258149086778,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.11,
      "mean_token_accuracy": 0.09046216607093811,
      "num_tokens": 1649215.0,
      "step": 815
    },
    {
      "entropy": 7.09494981765747,
      "epoch": 0.07165952984357249,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004095,
      "loss": 7.0047,
      "mean_token_accuracy": 0.09915160611271859,
      "num_tokens": 1659198.0,
      "step": 820
    },
    {
      "entropy": 7.139633274078369,
      "epoch": 0.0720964781962772,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000412,
      "loss": 7.0459,
      "mean_token_accuracy": 0.0924466960132122,
      "num_tokens": 1668972.0,
      "step": 825
    },
    {
      "entropy": 7.046590662002563,
      "epoch": 0.07253342654898191,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004145,
      "loss": 7.0373,
      "mean_token_accuracy": 0.0979261752218008,
      "num_tokens": 1679807.0,
      "step": 830
    },
    {
      "entropy": 7.146584796905517,
      "epoch": 0.07297037490168662,
      "grad_norm": 1.234375,
      "learning_rate": 0.000417,
      "loss": 7.0786,
      "mean_token_accuracy": 0.09333261549472809,
      "num_tokens": 1690567.0,
      "step": 835
    },
    {
      "entropy": 7.060258579254151,
      "epoch": 0.07340732325439134,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004195,
      "loss": 6.9708,
      "mean_token_accuracy": 0.10236304700374603,
      "num_tokens": 1700939.0,
      "step": 840
    },
    {
      "entropy": 7.0893957138061525,
      "epoch": 0.07384427160709604,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000422,
      "loss": 7.038,
      "mean_token_accuracy": 0.10244481042027473,
      "num_tokens": 1711181.0,
      "step": 845
    },
    {
      "entropy": 7.2199750423431395,
      "epoch": 0.07428121995980075,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004245,
      "loss": 7.0547,
      "mean_token_accuracy": 0.09390221312642097,
      "num_tokens": 1721392.0,
      "step": 850
    },
    {
      "entropy": 7.097149181365967,
      "epoch": 0.07471816831250547,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000427,
      "loss": 6.9883,
      "mean_token_accuracy": 0.1020311638712883,
      "num_tokens": 1731942.0,
      "step": 855
    },
    {
      "entropy": 7.16186032295227,
      "epoch": 0.07515511666521017,
      "grad_norm": 1.0,
      "learning_rate": 0.0004295,
      "loss": 7.0872,
      "mean_token_accuracy": 0.09683374986052513,
      "num_tokens": 1743409.0,
      "step": 860
    },
    {
      "entropy": 7.026008224487304,
      "epoch": 0.07559206501791488,
      "grad_norm": 0.96875,
      "learning_rate": 0.000432,
      "loss": 6.9834,
      "mean_token_accuracy": 0.10218272209167481,
      "num_tokens": 1753425.0,
      "step": 865
    },
    {
      "entropy": 7.078909397125244,
      "epoch": 0.0760290133706196,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004345,
      "loss": 7.0062,
      "mean_token_accuracy": 0.09942305907607078,
      "num_tokens": 1762722.0,
      "step": 870
    },
    {
      "entropy": 7.281267595291138,
      "epoch": 0.0764659617233243,
      "grad_norm": 1.15625,
      "learning_rate": 0.000437,
      "loss": 7.0547,
      "mean_token_accuracy": 0.09092987477779388,
      "num_tokens": 1771895.0,
      "step": 875
    },
    {
      "entropy": 7.075471496582031,
      "epoch": 0.07690291007602901,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004395,
      "loss": 6.9906,
      "mean_token_accuracy": 0.10108151286840439,
      "num_tokens": 1782175.0,
      "step": 880
    },
    {
      "entropy": 7.135920286178589,
      "epoch": 0.07733985842873373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000442,
      "loss": 7.0323,
      "mean_token_accuracy": 0.09571060985326767,
      "num_tokens": 1792409.0,
      "step": 885
    },
    {
      "entropy": 7.1416253566741945,
      "epoch": 0.07777680678143843,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004445,
      "loss": 7.0772,
      "mean_token_accuracy": 0.09155471846461297,
      "num_tokens": 1802721.0,
      "step": 890
    },
    {
      "entropy": 7.018156719207764,
      "epoch": 0.07821375513414315,
      "grad_norm": 1.109375,
      "learning_rate": 0.000447,
      "loss": 6.9152,
      "mean_token_accuracy": 0.1009690523147583,
      "num_tokens": 1812064.0,
      "step": 895
    },
    {
      "entropy": 7.185165739059448,
      "epoch": 0.07865070348684786,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.0243,
      "mean_token_accuracy": 0.0946855939924717,
      "num_tokens": 1822284.0,
      "step": 900
    },
    {
      "entropy": 7.019971942901611,
      "epoch": 0.07908765183955256,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.9408,
      "mean_token_accuracy": 0.10249921903014184,
      "num_tokens": 1832150.0,
      "step": 905
    },
    {
      "entropy": 7.135112857818603,
      "epoch": 0.07952460019225728,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.0546,
      "mean_token_accuracy": 0.09617897793650627,
      "num_tokens": 1842870.0,
      "step": 910
    },
    {
      "entropy": 7.019263792037964,
      "epoch": 0.07996154854496199,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.8767,
      "mean_token_accuracy": 0.10999749973416328,
      "num_tokens": 1852771.0,
      "step": 915
    },
    {
      "entropy": 6.987358617782593,
      "epoch": 0.08039849689766669,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.0028,
      "mean_token_accuracy": 0.09920822754502297,
      "num_tokens": 1863095.0,
      "step": 920
    },
    {
      "entropy": 7.018406534194947,
      "epoch": 0.08083544525037141,
      "grad_norm": 1.234375,
      "learning_rate": 0.000462,
      "loss": 7.0192,
      "mean_token_accuracy": 0.10134783908724784,
      "num_tokens": 1873060.0,
      "step": 925
    },
    {
      "entropy": 7.069212102890015,
      "epoch": 0.08127239360307612,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004645,
      "loss": 6.9396,
      "mean_token_accuracy": 0.10620549395680427,
      "num_tokens": 1882574.0,
      "step": 930
    },
    {
      "entropy": 7.033281898498535,
      "epoch": 0.08170934195578082,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000467,
      "loss": 7.0066,
      "mean_token_accuracy": 0.10208514109253883,
      "num_tokens": 1892227.0,
      "step": 935
    },
    {
      "entropy": 7.052687931060791,
      "epoch": 0.08214629030848554,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004695,
      "loss": 7.0234,
      "mean_token_accuracy": 0.09568553268909455,
      "num_tokens": 1902868.0,
      "step": 940
    },
    {
      "entropy": 7.065501070022583,
      "epoch": 0.08258323866119024,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000472,
      "loss": 6.9832,
      "mean_token_accuracy": 0.09660054594278336,
      "num_tokens": 1912781.0,
      "step": 945
    },
    {
      "entropy": 7.0425121784210205,
      "epoch": 0.08302018701389496,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004745,
      "loss": 6.9728,
      "mean_token_accuracy": 0.10843080580234528,
      "num_tokens": 1923638.0,
      "step": 950
    },
    {
      "entropy": 7.016868972778321,
      "epoch": 0.08345713536659967,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000477,
      "loss": 6.9288,
      "mean_token_accuracy": 0.10849572941660882,
      "num_tokens": 1933498.0,
      "step": 955
    },
    {
      "entropy": 7.000980377197266,
      "epoch": 0.08389408371930437,
      "grad_norm": 1.0,
      "learning_rate": 0.0004795,
      "loss": 6.9509,
      "mean_token_accuracy": 0.10340929329395294,
      "num_tokens": 1943798.0,
      "step": 960
    },
    {
      "entropy": 7.054853391647339,
      "epoch": 0.0843310320720091,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000482,
      "loss": 6.9966,
      "mean_token_accuracy": 0.0966202974319458,
      "num_tokens": 1954235.0,
      "step": 965
    },
    {
      "entropy": 7.01042628288269,
      "epoch": 0.0847679804247138,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004845,
      "loss": 6.9834,
      "mean_token_accuracy": 0.09832858666777611,
      "num_tokens": 1964421.0,
      "step": 970
    },
    {
      "entropy": 6.955400609970093,
      "epoch": 0.0852049287774185,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000487,
      "loss": 6.976,
      "mean_token_accuracy": 0.10213669240474701,
      "num_tokens": 1975862.0,
      "step": 975
    },
    {
      "entropy": 7.061838102340698,
      "epoch": 0.08564187713012322,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004895,
      "loss": 6.9598,
      "mean_token_accuracy": 0.09855997413396836,
      "num_tokens": 1985758.0,
      "step": 980
    },
    {
      "entropy": 6.923084402084351,
      "epoch": 0.08607882548282793,
      "grad_norm": 1.015625,
      "learning_rate": 0.000492,
      "loss": 6.9431,
      "mean_token_accuracy": 0.09963723346590996,
      "num_tokens": 1996904.0,
      "step": 985
    },
    {
      "entropy": 7.119592952728271,
      "epoch": 0.08651577383553263,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004945,
      "loss": 7.023,
      "mean_token_accuracy": 0.0975315362215042,
      "num_tokens": 2006114.0,
      "step": 990
    },
    {
      "entropy": 7.007775974273682,
      "epoch": 0.08695272218823735,
      "grad_norm": 1.0625,
      "learning_rate": 0.000497,
      "loss": 6.9555,
      "mean_token_accuracy": 0.10550742074847222,
      "num_tokens": 2016349.0,
      "step": 995
    },
    {
      "entropy": 6.926405000686645,
      "epoch": 0.08738967054094206,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995,
      "loss": 6.9055,
      "mean_token_accuracy": 0.09952122941613198,
      "num_tokens": 2026701.0,
      "step": 1000
    },
    {
      "entropy": 6.909988212585449,
      "epoch": 0.08782661889364678,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999999986190023,
      "loss": 6.8942,
      "mean_token_accuracy": 0.10107116997241974,
      "num_tokens": 2036950.0,
      "step": 1005
    },
    {
      "entropy": 7.06274824142456,
      "epoch": 0.08826356724635148,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999999930086991,
      "loss": 6.9046,
      "mean_token_accuracy": 0.10389338061213493,
      "num_tokens": 2047655.0,
      "step": 1010
    },
    {
      "entropy": 6.927995491027832,
      "epoch": 0.08870051559905619,
      "grad_norm": 1.125,
      "learning_rate": 0.000499999983082778,
      "loss": 6.8846,
      "mean_token_accuracy": 0.10662774741649628,
      "num_tokens": 2056800.0,
      "step": 1015
    },
    {
      "entropy": 6.8703173160552975,
      "epoch": 0.0891374639517609,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999999688412394,
      "loss": 6.9257,
      "mean_token_accuracy": 0.10540272071957588,
      "num_tokens": 2066862.0,
      "step": 1020
    },
    {
      "entropy": 7.05763463973999,
      "epoch": 0.08957441230446561,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999999502840836,
      "loss": 6.9449,
      "mean_token_accuracy": 0.1028741791844368,
      "num_tokens": 2076585.0,
      "step": 1025
    },
    {
      "entropy": 6.949811029434204,
      "epoch": 0.09001136065717032,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999999274113107,
      "loss": 7.0015,
      "mean_token_accuracy": 0.09622879326343536,
      "num_tokens": 2087768.0,
      "step": 1030
    },
    {
      "entropy": 6.970440816879273,
      "epoch": 0.09044830900987504,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999999002229214,
      "loss": 6.912,
      "mean_token_accuracy": 0.1016117550432682,
      "num_tokens": 2097027.0,
      "step": 1035
    },
    {
      "entropy": 7.010776376724243,
      "epoch": 0.09088525736257974,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999998687189161,
      "loss": 6.8502,
      "mean_token_accuracy": 0.11224842295050622,
      "num_tokens": 2107055.0,
      "step": 1040
    },
    {
      "entropy": 6.8836586475372314,
      "epoch": 0.09132220571528446,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999998328992954,
      "loss": 6.7809,
      "mean_token_accuracy": 0.10591298639774323,
      "num_tokens": 2116771.0,
      "step": 1045
    },
    {
      "entropy": 6.847192192077637,
      "epoch": 0.09175915406798917,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00049999979276406,
      "loss": 6.9104,
      "mean_token_accuracy": 0.1076812945306301,
      "num_tokens": 2126881.0,
      "step": 1050
    },
    {
      "entropy": 6.95491886138916,
      "epoch": 0.09219610242069387,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999997483132108,
      "loss": 6.8986,
      "mean_token_accuracy": 0.10368494689464569,
      "num_tokens": 2136964.0,
      "step": 1055
    },
    {
      "entropy": 6.9707235336303714,
      "epoch": 0.09263305077339859,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999996995467483,
      "loss": 6.9526,
      "mean_token_accuracy": 0.10105137750506402,
      "num_tokens": 2147263.0,
      "step": 1060
    },
    {
      "entropy": 6.893505430221557,
      "epoch": 0.0930699991261033,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999996464646738,
      "loss": 6.9698,
      "mean_token_accuracy": 0.09723725020885468,
      "num_tokens": 2157474.0,
      "step": 1065
    },
    {
      "entropy": 6.928754138946533,
      "epoch": 0.093506947478808,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999995890669882,
      "loss": 6.8968,
      "mean_token_accuracy": 0.10273590087890624,
      "num_tokens": 2167701.0,
      "step": 1070
    },
    {
      "entropy": 6.855544805526733,
      "epoch": 0.09394389583151272,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999995273536925,
      "loss": 6.8695,
      "mean_token_accuracy": 0.10916284620761871,
      "num_tokens": 2178581.0,
      "step": 1075
    },
    {
      "entropy": 6.926065063476562,
      "epoch": 0.09438084418421742,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499999461324788,
      "loss": 6.8665,
      "mean_token_accuracy": 0.10301437005400657,
      "num_tokens": 2189529.0,
      "step": 1080
    },
    {
      "entropy": 6.931471347808838,
      "epoch": 0.09481779253692213,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999993909802759,
      "loss": 6.9188,
      "mean_token_accuracy": 0.10070269852876663,
      "num_tokens": 2199805.0,
      "step": 1085
    },
    {
      "entropy": 6.864389324188233,
      "epoch": 0.09525474088962685,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999993163201577,
      "loss": 6.867,
      "mean_token_accuracy": 0.10666536092758179,
      "num_tokens": 2210108.0,
      "step": 1090
    },
    {
      "entropy": 6.986412286758423,
      "epoch": 0.09569168924233155,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999992373444347,
      "loss": 6.8837,
      "mean_token_accuracy": 0.10290357768535614,
      "num_tokens": 2220325.0,
      "step": 1095
    },
    {
      "entropy": 6.831525468826294,
      "epoch": 0.09612863759503627,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999991540531084,
      "loss": 6.8743,
      "mean_token_accuracy": 0.10060080140829086,
      "num_tokens": 2230521.0,
      "step": 1100
    },
    {
      "entropy": 7.026464128494263,
      "epoch": 0.09656558594774098,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999990664461804,
      "loss": 6.9131,
      "mean_token_accuracy": 0.0995321400463581,
      "num_tokens": 2240372.0,
      "step": 1105
    },
    {
      "entropy": 6.830053663253784,
      "epoch": 0.09700253430044568,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999989745236524,
      "loss": 6.8651,
      "mean_token_accuracy": 0.11102000325918197,
      "num_tokens": 2250522.0,
      "step": 1110
    },
    {
      "entropy": 6.86471962928772,
      "epoch": 0.0974394826531504,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999988782855261,
      "loss": 6.8068,
      "mean_token_accuracy": 0.10345931202173234,
      "num_tokens": 2260142.0,
      "step": 1115
    },
    {
      "entropy": 6.8818775653839115,
      "epoch": 0.09787643100585511,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999987777318035,
      "loss": 6.8529,
      "mean_token_accuracy": 0.10523375123739243,
      "num_tokens": 2270372.0,
      "step": 1120
    },
    {
      "entropy": 6.768800354003906,
      "epoch": 0.09831337935855981,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999986728624865,
      "loss": 6.7754,
      "mean_token_accuracy": 0.11057863682508469,
      "num_tokens": 2280517.0,
      "step": 1125
    },
    {
      "entropy": 6.968469667434692,
      "epoch": 0.09875032771126453,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499998563677577,
      "loss": 6.8773,
      "mean_token_accuracy": 0.10483333468437195,
      "num_tokens": 2290778.0,
      "step": 1130
    },
    {
      "entropy": 6.8761578559875485,
      "epoch": 0.09918727606396924,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999984501770771,
      "loss": 6.928,
      "mean_token_accuracy": 0.0982240878045559,
      "num_tokens": 2301692.0,
      "step": 1135
    },
    {
      "entropy": 6.879663705825806,
      "epoch": 0.09962422441667394,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999983323609891,
      "loss": 6.8015,
      "mean_token_accuracy": 0.10966127961874009,
      "num_tokens": 2312759.0,
      "step": 1140
    },
    {
      "entropy": 6.873649406433105,
      "epoch": 0.10006117276937866,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999982102293153,
      "loss": 6.7811,
      "mean_token_accuracy": 0.10762107968330384,
      "num_tokens": 2322016.0,
      "step": 1145
    },
    {
      "entropy": 6.836494588851929,
      "epoch": 0.10049812112208337,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999980837820578,
      "loss": 6.7932,
      "mean_token_accuracy": 0.11270353570580482,
      "num_tokens": 2332476.0,
      "step": 1150
    },
    {
      "entropy": 6.899167490005493,
      "epoch": 0.10093506947478809,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999979530192192,
      "loss": 6.8019,
      "mean_token_accuracy": 0.10367468595504761,
      "num_tokens": 2342914.0,
      "step": 1155
    },
    {
      "entropy": 6.799224042892456,
      "epoch": 0.10137201782749279,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999978179408019,
      "loss": 6.8876,
      "mean_token_accuracy": 0.10129361897706986,
      "num_tokens": 2353381.0,
      "step": 1160
    },
    {
      "entropy": 6.881389856338501,
      "epoch": 0.1018089661801975,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999976785468085,
      "loss": 6.8543,
      "mean_token_accuracy": 0.10723742172122001,
      "num_tokens": 2364207.0,
      "step": 1165
    },
    {
      "entropy": 6.8175615787506105,
      "epoch": 0.10224591453290222,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999975348372418,
      "loss": 6.768,
      "mean_token_accuracy": 0.10937329679727555,
      "num_tokens": 2374985.0,
      "step": 1170
    },
    {
      "entropy": 6.797830724716187,
      "epoch": 0.10268286288560692,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999973868121044,
      "loss": 6.8479,
      "mean_token_accuracy": 0.1104587584733963,
      "num_tokens": 2385684.0,
      "step": 1175
    },
    {
      "entropy": 6.961680555343628,
      "epoch": 0.10311981123831163,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999972344713994,
      "loss": 6.7834,
      "mean_token_accuracy": 0.10461923331022263,
      "num_tokens": 2395177.0,
      "step": 1180
    },
    {
      "entropy": 6.685928773880005,
      "epoch": 0.10355675959101635,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999970778151295,
      "loss": 6.8172,
      "mean_token_accuracy": 0.11517063304781913,
      "num_tokens": 2405154.0,
      "step": 1185
    },
    {
      "entropy": 6.829747676849365,
      "epoch": 0.10399370794372105,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999969168432976,
      "loss": 6.8064,
      "mean_token_accuracy": 0.10844132006168365,
      "num_tokens": 2415818.0,
      "step": 1190
    },
    {
      "entropy": 6.826361274719238,
      "epoch": 0.10443065629642576,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999967515559071,
      "loss": 6.7788,
      "mean_token_accuracy": 0.10787174105644226,
      "num_tokens": 2425072.0,
      "step": 1195
    },
    {
      "entropy": 6.7768299102783205,
      "epoch": 0.10486760464913047,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499996581952961,
      "loss": 6.8137,
      "mean_token_accuracy": 0.11103299036622047,
      "num_tokens": 2435055.0,
      "step": 1200
    },
    {
      "entropy": 6.840562725067139,
      "epoch": 0.10530455300183518,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999964080344625,
      "loss": 6.7297,
      "mean_token_accuracy": 0.1094184011220932,
      "num_tokens": 2444434.0,
      "step": 1205
    },
    {
      "entropy": 6.745257711410522,
      "epoch": 0.1057415013545399,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999962298004151,
      "loss": 6.7444,
      "mean_token_accuracy": 0.10750698521733285,
      "num_tokens": 2453860.0,
      "step": 1210
    },
    {
      "entropy": 6.85921721458435,
      "epoch": 0.1061784497072446,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999960472508219,
      "loss": 6.8699,
      "mean_token_accuracy": 0.1019215852022171,
      "num_tokens": 2464235.0,
      "step": 1215
    },
    {
      "entropy": 6.823530101776123,
      "epoch": 0.10661539805994931,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999958603856869,
      "loss": 6.7813,
      "mean_token_accuracy": 0.10817842781543732,
      "num_tokens": 2474561.0,
      "step": 1220
    },
    {
      "entropy": 6.743971061706543,
      "epoch": 0.10705234641265403,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999956692050132,
      "loss": 6.763,
      "mean_token_accuracy": 0.11355147212743759,
      "num_tokens": 2484574.0,
      "step": 1225
    },
    {
      "entropy": 6.859746837615967,
      "epoch": 0.10748929476535873,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999954737088048,
      "loss": 6.7401,
      "mean_token_accuracy": 0.11617439091205597,
      "num_tokens": 2494146.0,
      "step": 1230
    },
    {
      "entropy": 6.72626371383667,
      "epoch": 0.10792624311806344,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999952738970652,
      "loss": 6.8298,
      "mean_token_accuracy": 0.10861490145325661,
      "num_tokens": 2504157.0,
      "step": 1235
    },
    {
      "entropy": 6.863236904144287,
      "epoch": 0.10836319147076816,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999950697697983,
      "loss": 6.7584,
      "mean_token_accuracy": 0.1099944420158863,
      "num_tokens": 2512979.0,
      "step": 1240
    },
    {
      "entropy": 6.742759895324707,
      "epoch": 0.10880013982347286,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999948613270081,
      "loss": 6.8293,
      "mean_token_accuracy": 0.10701432153582573,
      "num_tokens": 2523202.0,
      "step": 1245
    },
    {
      "entropy": 6.795160913467408,
      "epoch": 0.10923708817617758,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999946485686987,
      "loss": 6.7757,
      "mean_token_accuracy": 0.1092297151684761,
      "num_tokens": 2532790.0,
      "step": 1250
    },
    {
      "entropy": 6.7600242614746096,
      "epoch": 0.10967403652888229,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499994431494874,
      "loss": 6.7616,
      "mean_token_accuracy": 0.10790274366736412,
      "num_tokens": 2542337.0,
      "step": 1255
    },
    {
      "entropy": 6.730962944030762,
      "epoch": 0.11011098488158699,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999942101055381,
      "loss": 6.6937,
      "mean_token_accuracy": 0.11463424563407898,
      "num_tokens": 2552762.0,
      "step": 1260
    },
    {
      "entropy": 6.775425910949707,
      "epoch": 0.11054793323429171,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999939844006955,
      "loss": 6.7362,
      "mean_token_accuracy": 0.11280569657683373,
      "num_tokens": 2563261.0,
      "step": 1265
    },
    {
      "entropy": 6.780911922454834,
      "epoch": 0.11098488158699642,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999937543803503,
      "loss": 6.7208,
      "mean_token_accuracy": 0.11466270163655282,
      "num_tokens": 2573089.0,
      "step": 1270
    },
    {
      "entropy": 6.763316488265991,
      "epoch": 0.11142182993970112,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499993520044507,
      "loss": 6.8021,
      "mean_token_accuracy": 0.10308952704071998,
      "num_tokens": 2584358.0,
      "step": 1275
    },
    {
      "entropy": 6.851671457290649,
      "epoch": 0.11185877829240584,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999932813931702,
      "loss": 6.7804,
      "mean_token_accuracy": 0.11292212307453156,
      "num_tokens": 2594981.0,
      "step": 1280
    },
    {
      "entropy": 6.707735347747803,
      "epoch": 0.11229572664511055,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999930384263441,
      "loss": 6.7698,
      "mean_token_accuracy": 0.1120561957359314,
      "num_tokens": 2603508.0,
      "step": 1285
    },
    {
      "entropy": 6.87262110710144,
      "epoch": 0.11273267499781525,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499992791144034,
      "loss": 6.7188,
      "mean_token_accuracy": 0.11393993496894836,
      "num_tokens": 2613666.0,
      "step": 1290
    },
    {
      "entropy": 6.819432783126831,
      "epoch": 0.11316962335051997,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999925395462439,
      "loss": 6.8095,
      "mean_token_accuracy": 0.10437591895461082,
      "num_tokens": 2624441.0,
      "step": 1295
    },
    {
      "entropy": 6.761607503890991,
      "epoch": 0.11360657170322468,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999922836329793,
      "loss": 6.7573,
      "mean_token_accuracy": 0.11246527284383774,
      "num_tokens": 2635261.0,
      "step": 1300
    },
    {
      "entropy": 6.718364334106445,
      "epoch": 0.1140435200559294,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999920234042446,
      "loss": 6.7283,
      "mean_token_accuracy": 0.11169910728931427,
      "num_tokens": 2645493.0,
      "step": 1305
    },
    {
      "entropy": 6.7557297706604,
      "epoch": 0.1144804684086341,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999917588600451,
      "loss": 6.6748,
      "mean_token_accuracy": 0.11324728205800057,
      "num_tokens": 2655430.0,
      "step": 1310
    },
    {
      "entropy": 6.731936120986939,
      "epoch": 0.1149174167613388,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999914900003857,
      "loss": 6.7619,
      "mean_token_accuracy": 0.1131984330713749,
      "num_tokens": 2665675.0,
      "step": 1315
    },
    {
      "entropy": 6.8105064868927006,
      "epoch": 0.11535436511404352,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999912168252717,
      "loss": 6.7183,
      "mean_token_accuracy": 0.10891555175185204,
      "num_tokens": 2675962.0,
      "step": 1320
    },
    {
      "entropy": 6.6981361389160154,
      "epoch": 0.11579131346674823,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999909393347083,
      "loss": 6.7696,
      "mean_token_accuracy": 0.11046200320124626,
      "num_tokens": 2687015.0,
      "step": 1325
    },
    {
      "entropy": 6.805249977111816,
      "epoch": 0.11622826181945294,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999906575287006,
      "loss": 6.8003,
      "mean_token_accuracy": 0.10271921083331108,
      "num_tokens": 2698065.0,
      "step": 1330
    },
    {
      "entropy": 6.699923133850097,
      "epoch": 0.11666521017215765,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999903714072544,
      "loss": 6.6888,
      "mean_token_accuracy": 0.1208715133368969,
      "num_tokens": 2709046.0,
      "step": 1335
    },
    {
      "entropy": 6.805598878860474,
      "epoch": 0.11710215852486236,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999900809703749,
      "loss": 6.7648,
      "mean_token_accuracy": 0.1078792616724968,
      "num_tokens": 2719705.0,
      "step": 1340
    },
    {
      "entropy": 6.71959080696106,
      "epoch": 0.11753910687756706,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004999897862180678,
      "loss": 6.639,
      "mean_token_accuracy": 0.1174460805952549,
      "num_tokens": 2729556.0,
      "step": 1345
    },
    {
      "entropy": 6.650531578063965,
      "epoch": 0.11797605523027178,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999894871503386,
      "loss": 6.7566,
      "mean_token_accuracy": 0.11257789805531501,
      "num_tokens": 2741144.0,
      "step": 1350
    },
    {
      "entropy": 6.823479223251343,
      "epoch": 0.11841300358297649,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999891837671932,
      "loss": 6.6674,
      "mean_token_accuracy": 0.11613549664616585,
      "num_tokens": 2751384.0,
      "step": 1355
    },
    {
      "entropy": 6.634953927993775,
      "epoch": 0.11884995193568121,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999888760686375,
      "loss": 6.6928,
      "mean_token_accuracy": 0.11121985912322999,
      "num_tokens": 2761044.0,
      "step": 1360
    },
    {
      "entropy": 6.683824157714843,
      "epoch": 0.11928690028838591,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499988564054677,
      "loss": 6.6573,
      "mean_token_accuracy": 0.11309800446033477,
      "num_tokens": 2770946.0,
      "step": 1365
    },
    {
      "entropy": 6.737067604064942,
      "epoch": 0.11972384864109062,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999882477253182,
      "loss": 6.7161,
      "mean_token_accuracy": 0.10762240588665009,
      "num_tokens": 2781247.0,
      "step": 1370
    },
    {
      "entropy": 6.665458488464355,
      "epoch": 0.12016079699379534,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999879270805667,
      "loss": 6.5612,
      "mean_token_accuracy": 0.12452048733830452,
      "num_tokens": 2792078.0,
      "step": 1375
    },
    {
      "entropy": 6.6383239269256595,
      "epoch": 0.12059774534650004,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999876021204291,
      "loss": 6.6665,
      "mean_token_accuracy": 0.11566043272614479,
      "num_tokens": 2802458.0,
      "step": 1380
    },
    {
      "entropy": 6.732123756408692,
      "epoch": 0.12103469369920475,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999872728449112,
      "loss": 6.7962,
      "mean_token_accuracy": 0.10792736932635308,
      "num_tokens": 2813352.0,
      "step": 1385
    },
    {
      "entropy": 6.750059413909912,
      "epoch": 0.12147164205190947,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999869392540195,
      "loss": 6.692,
      "mean_token_accuracy": 0.10860167518258095,
      "num_tokens": 2822954.0,
      "step": 1390
    },
    {
      "entropy": 6.739627456665039,
      "epoch": 0.12190859040461417,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999866013477605,
      "loss": 6.712,
      "mean_token_accuracy": 0.11105184629559517,
      "num_tokens": 2833375.0,
      "step": 1395
    },
    {
      "entropy": 6.681881904602051,
      "epoch": 0.12234553875731889,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999862591261407,
      "loss": 6.654,
      "mean_token_accuracy": 0.11549825221300125,
      "num_tokens": 2843103.0,
      "step": 1400
    },
    {
      "entropy": 6.7078999996185305,
      "epoch": 0.1227824871100236,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999859125891664,
      "loss": 6.72,
      "mean_token_accuracy": 0.11280344799160957,
      "num_tokens": 2853801.0,
      "step": 1405
    },
    {
      "entropy": 6.723126173019409,
      "epoch": 0.1232194354627283,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999855617368445,
      "loss": 6.712,
      "mean_token_accuracy": 0.1107217937707901,
      "num_tokens": 2863851.0,
      "step": 1410
    },
    {
      "entropy": 6.639747476577758,
      "epoch": 0.12365638381543302,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999852065691816,
      "loss": 6.6556,
      "mean_token_accuracy": 0.1157209224998951,
      "num_tokens": 2875073.0,
      "step": 1415
    },
    {
      "entropy": 6.819645643234253,
      "epoch": 0.12409333216813773,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999848470861846,
      "loss": 6.6923,
      "mean_token_accuracy": 0.11399516686797143,
      "num_tokens": 2885676.0,
      "step": 1420
    },
    {
      "entropy": 6.66095814704895,
      "epoch": 0.12453028052084243,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999844832878602,
      "loss": 6.6744,
      "mean_token_accuracy": 0.11909589692950248,
      "num_tokens": 2896663.0,
      "step": 1425
    },
    {
      "entropy": 6.682529258728027,
      "epoch": 0.12496722887354715,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999841151742156,
      "loss": 6.6855,
      "mean_token_accuracy": 0.11644814759492875,
      "num_tokens": 2907555.0,
      "step": 1430
    },
    {
      "entropy": 6.654642724990845,
      "epoch": 0.12540417722625186,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999837427452578,
      "loss": 6.6823,
      "mean_token_accuracy": 0.11288027614355087,
      "num_tokens": 2917173.0,
      "step": 1435
    },
    {
      "entropy": 6.634070777893067,
      "epoch": 0.12584112557895658,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499983366000994,
      "loss": 6.6819,
      "mean_token_accuracy": 0.11158063933253289,
      "num_tokens": 2928268.0,
      "step": 1440
    },
    {
      "entropy": 6.777620124816894,
      "epoch": 0.12627807393166127,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999829849414312,
      "loss": 6.6361,
      "mean_token_accuracy": 0.1173455186188221,
      "num_tokens": 2938203.0,
      "step": 1445
    },
    {
      "entropy": 6.595511102676392,
      "epoch": 0.12671502228436599,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499982599566577,
      "loss": 6.666,
      "mean_token_accuracy": 0.1175123281776905,
      "num_tokens": 2948094.0,
      "step": 1450
    },
    {
      "entropy": 6.724627923965454,
      "epoch": 0.1271519706370707,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999822098764385,
      "loss": 6.6761,
      "mean_token_accuracy": 0.1126259833574295,
      "num_tokens": 2957865.0,
      "step": 1455
    },
    {
      "entropy": 6.633754539489746,
      "epoch": 0.1275889189897754,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999818158710234,
      "loss": 6.582,
      "mean_token_accuracy": 0.11640499681234359,
      "num_tokens": 2967880.0,
      "step": 1460
    },
    {
      "entropy": 6.671688032150269,
      "epoch": 0.12802586734248012,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999814175503392,
      "loss": 6.6542,
      "mean_token_accuracy": 0.11618517264723778,
      "num_tokens": 2979537.0,
      "step": 1465
    },
    {
      "entropy": 6.653942728042603,
      "epoch": 0.12846281569518483,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999810149143935,
      "loss": 6.6421,
      "mean_token_accuracy": 0.11047034710645676,
      "num_tokens": 2989700.0,
      "step": 1470
    },
    {
      "entropy": 6.558183765411377,
      "epoch": 0.12889976404788953,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499980607963194,
      "loss": 6.6535,
      "mean_token_accuracy": 0.11763448640704155,
      "num_tokens": 2999141.0,
      "step": 1475
    },
    {
      "entropy": 6.711502933502198,
      "epoch": 0.12933671240059424,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999801966967486,
      "loss": 6.5868,
      "mean_token_accuracy": 0.11921618357300759,
      "num_tokens": 3008259.0,
      "step": 1480
    },
    {
      "entropy": 6.674244356155396,
      "epoch": 0.12977366075329896,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999797811150652,
      "loss": 6.648,
      "mean_token_accuracy": 0.10815014466643333,
      "num_tokens": 3018281.0,
      "step": 1485
    },
    {
      "entropy": 6.719975137710572,
      "epoch": 0.13021060910600368,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999793612181516,
      "loss": 6.6448,
      "mean_token_accuracy": 0.11819751039147378,
      "num_tokens": 3029062.0,
      "step": 1490
    },
    {
      "entropy": 6.508638143539429,
      "epoch": 0.13064755745870837,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999789370060161,
      "loss": 6.6111,
      "mean_token_accuracy": 0.11788376942276954,
      "num_tokens": 3038142.0,
      "step": 1495
    },
    {
      "entropy": 6.793538331985474,
      "epoch": 0.1310845058114131,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999785084786667,
      "loss": 6.7071,
      "mean_token_accuracy": 0.10659723281860352,
      "num_tokens": 3049123.0,
      "step": 1500
    },
    {
      "entropy": 6.612226247787476,
      "epoch": 0.1315214541641178,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999780756361114,
      "loss": 6.619,
      "mean_token_accuracy": 0.11126089543104171,
      "num_tokens": 3059031.0,
      "step": 1505
    },
    {
      "entropy": 6.67326545715332,
      "epoch": 0.1319584025168225,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999776384783589,
      "loss": 6.6532,
      "mean_token_accuracy": 0.11633370667696,
      "num_tokens": 3068830.0,
      "step": 1510
    },
    {
      "entropy": 6.579953718185425,
      "epoch": 0.13239535086952722,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999771970054174,
      "loss": 6.6592,
      "mean_token_accuracy": 0.1143224373459816,
      "num_tokens": 3080124.0,
      "step": 1515
    },
    {
      "entropy": 6.664220142364502,
      "epoch": 0.13283229922223194,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999767512172953,
      "loss": 6.6545,
      "mean_token_accuracy": 0.10695429295301437,
      "num_tokens": 3089607.0,
      "step": 1520
    },
    {
      "entropy": 6.688190889358521,
      "epoch": 0.13326924757493663,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999763011140013,
      "loss": 6.7211,
      "mean_token_accuracy": 0.10691676661372185,
      "num_tokens": 3099363.0,
      "step": 1525
    },
    {
      "entropy": 6.751109886169433,
      "epoch": 0.13370619592764135,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999758466955439,
      "loss": 6.6214,
      "mean_token_accuracy": 0.11871336251497269,
      "num_tokens": 3109552.0,
      "step": 1530
    },
    {
      "entropy": 6.567575359344483,
      "epoch": 0.13414314428034607,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999753879619319,
      "loss": 6.6033,
      "mean_token_accuracy": 0.1198199823498726,
      "num_tokens": 3119536.0,
      "step": 1535
    },
    {
      "entropy": 6.637157773971557,
      "epoch": 0.13458009263305076,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999749249131741,
      "loss": 6.6466,
      "mean_token_accuracy": 0.11405760124325752,
      "num_tokens": 3130552.0,
      "step": 1540
    },
    {
      "entropy": 6.744642210006714,
      "epoch": 0.13501704098575548,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999744575492793,
      "loss": 6.5619,
      "mean_token_accuracy": 0.11781792342662811,
      "num_tokens": 3140723.0,
      "step": 1545
    },
    {
      "entropy": 6.585478115081787,
      "epoch": 0.1354539893384602,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999739858702567,
      "loss": 6.6016,
      "mean_token_accuracy": 0.11881603077054023,
      "num_tokens": 3151031.0,
      "step": 1550
    },
    {
      "entropy": 6.587906789779663,
      "epoch": 0.1358909376911649,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499973509876115,
      "loss": 6.6036,
      "mean_token_accuracy": 0.11454507485032081,
      "num_tokens": 3161373.0,
      "step": 1555
    },
    {
      "entropy": 6.673034143447876,
      "epoch": 0.1363278860438696,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999730295668634,
      "loss": 6.6232,
      "mean_token_accuracy": 0.11494075879454613,
      "num_tokens": 3171094.0,
      "step": 1560
    },
    {
      "entropy": 6.599486541748047,
      "epoch": 0.13676483439657433,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999725449425114,
      "loss": 6.5751,
      "mean_token_accuracy": 0.11703467667102814,
      "num_tokens": 3181646.0,
      "step": 1565
    },
    {
      "entropy": 6.574843311309815,
      "epoch": 0.13720178274927902,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499972056003068,
      "loss": 6.5467,
      "mean_token_accuracy": 0.12450342401862144,
      "num_tokens": 3190727.0,
      "step": 1570
    },
    {
      "entropy": 6.643044805526733,
      "epoch": 0.13763873110198374,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999715627485426,
      "loss": 6.6619,
      "mean_token_accuracy": 0.10692948997020721,
      "num_tokens": 3200316.0,
      "step": 1575
    },
    {
      "entropy": 6.6441529273986815,
      "epoch": 0.13807567945468846,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499971065178945,
      "loss": 6.5991,
      "mean_token_accuracy": 0.11431368961930274,
      "num_tokens": 3211830.0,
      "step": 1580
    },
    {
      "entropy": 6.572905778884888,
      "epoch": 0.13851262780739318,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999705632942843,
      "loss": 6.6204,
      "mean_token_accuracy": 0.11557646915316581,
      "num_tokens": 3222711.0,
      "step": 1585
    },
    {
      "entropy": 6.611633920669556,
      "epoch": 0.13894957616009787,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999700570945704,
      "loss": 6.5282,
      "mean_token_accuracy": 0.12471159398555756,
      "num_tokens": 3232810.0,
      "step": 1590
    },
    {
      "entropy": 6.591669034957886,
      "epoch": 0.1393865245128026,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999695465798128,
      "loss": 6.6326,
      "mean_token_accuracy": 0.11323555037379265,
      "num_tokens": 3243366.0,
      "step": 1595
    },
    {
      "entropy": 6.661644554138183,
      "epoch": 0.1398234728655073,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999690317500215,
      "loss": 6.6116,
      "mean_token_accuracy": 0.11405737772583961,
      "num_tokens": 3254000.0,
      "step": 1600
    },
    {
      "entropy": 6.700826120376587,
      "epoch": 0.140260421218212,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999685126052063,
      "loss": 6.642,
      "mean_token_accuracy": 0.11356792375445365,
      "num_tokens": 3263601.0,
      "step": 1605
    },
    {
      "entropy": 6.5335101127624515,
      "epoch": 0.14069736957091672,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999679891453772,
      "loss": 6.6709,
      "mean_token_accuracy": 0.1109616331756115,
      "num_tokens": 3273243.0,
      "step": 1610
    },
    {
      "entropy": 6.65966911315918,
      "epoch": 0.14113431792362144,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499967461370544,
      "loss": 6.6082,
      "mean_token_accuracy": 0.11456717923283577,
      "num_tokens": 3284208.0,
      "step": 1615
    },
    {
      "entropy": 6.625819253921509,
      "epoch": 0.14157126627632613,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999669292807172,
      "loss": 6.5819,
      "mean_token_accuracy": 0.11917855739593505,
      "num_tokens": 3294216.0,
      "step": 1620
    },
    {
      "entropy": 6.636405372619629,
      "epoch": 0.14200821462903085,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999663928759066,
      "loss": 6.6316,
      "mean_token_accuracy": 0.11559324115514755,
      "num_tokens": 3304576.0,
      "step": 1625
    },
    {
      "entropy": 6.641275072097779,
      "epoch": 0.14244516298173557,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999658521561229,
      "loss": 6.5962,
      "mean_token_accuracy": 0.11600318327546119,
      "num_tokens": 3314214.0,
      "step": 1630
    },
    {
      "entropy": 6.633429431915284,
      "epoch": 0.14288211133444026,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999653071213763,
      "loss": 6.6071,
      "mean_token_accuracy": 0.1085868775844574,
      "num_tokens": 3325527.0,
      "step": 1635
    },
    {
      "entropy": 6.5328723907470705,
      "epoch": 0.14331905968714498,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999647577716771,
      "loss": 6.5292,
      "mean_token_accuracy": 0.12012957409024239,
      "num_tokens": 3334511.0,
      "step": 1640
    },
    {
      "entropy": 6.605924606323242,
      "epoch": 0.1437560080398497,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999642041070361,
      "loss": 6.5238,
      "mean_token_accuracy": 0.1242619700729847,
      "num_tokens": 3343922.0,
      "step": 1645
    },
    {
      "entropy": 6.595180988311768,
      "epoch": 0.1441929563925544,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999636461274637,
      "loss": 6.5324,
      "mean_token_accuracy": 0.11857725381851196,
      "num_tokens": 3353895.0,
      "step": 1650
    },
    {
      "entropy": 6.566598129272461,
      "epoch": 0.1446299047452591,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999630838329706,
      "loss": 6.6033,
      "mean_token_accuracy": 0.1127896174788475,
      "num_tokens": 3364637.0,
      "step": 1655
    },
    {
      "entropy": 6.610513353347779,
      "epoch": 0.14506685309796383,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999625172235679,
      "loss": 6.5719,
      "mean_token_accuracy": 0.11518365815281868,
      "num_tokens": 3374308.0,
      "step": 1660
    },
    {
      "entropy": 6.5965320587158205,
      "epoch": 0.14550380145066852,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499961946299266,
      "loss": 6.5453,
      "mean_token_accuracy": 0.11517407447099685,
      "num_tokens": 3383637.0,
      "step": 1665
    },
    {
      "entropy": 6.590162563323974,
      "epoch": 0.14594074980337324,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999613710600763,
      "loss": 6.6251,
      "mean_token_accuracy": 0.11774841398000717,
      "num_tokens": 3394150.0,
      "step": 1670
    },
    {
      "entropy": 6.680703973770141,
      "epoch": 0.14637769815607796,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999607915060094,
      "loss": 6.5903,
      "mean_token_accuracy": 0.11480792313814163,
      "num_tokens": 3404445.0,
      "step": 1675
    },
    {
      "entropy": 6.5011927604675295,
      "epoch": 0.14681464650878268,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999602076370768,
      "loss": 6.4964,
      "mean_token_accuracy": 0.12533730790019035,
      "num_tokens": 3415183.0,
      "step": 1680
    },
    {
      "entropy": 6.560801219940186,
      "epoch": 0.14725159486148737,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999596194532895,
      "loss": 6.4879,
      "mean_token_accuracy": 0.1252271644771099,
      "num_tokens": 3424824.0,
      "step": 1685
    },
    {
      "entropy": 6.547294235229492,
      "epoch": 0.1476885432141921,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999590269546588,
      "loss": 6.6166,
      "mean_token_accuracy": 0.11056721806526185,
      "num_tokens": 3434789.0,
      "step": 1690
    },
    {
      "entropy": 6.4686205863952635,
      "epoch": 0.1481254915668968,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999584301411961,
      "loss": 6.495,
      "mean_token_accuracy": 0.13793878480792046,
      "num_tokens": 3445810.0,
      "step": 1695
    },
    {
      "entropy": 6.66889123916626,
      "epoch": 0.1485624399196015,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999578290129129,
      "loss": 6.6183,
      "mean_token_accuracy": 0.11348513811826706,
      "num_tokens": 3455510.0,
      "step": 1700
    },
    {
      "entropy": 6.594088411331176,
      "epoch": 0.14899938827230622,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999572235698206,
      "loss": 6.5905,
      "mean_token_accuracy": 0.11757459565997123,
      "num_tokens": 3465785.0,
      "step": 1705
    },
    {
      "entropy": 6.552883243560791,
      "epoch": 0.14943633662501093,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999566138119308,
      "loss": 6.652,
      "mean_token_accuracy": 0.11335393786430359,
      "num_tokens": 3477568.0,
      "step": 1710
    },
    {
      "entropy": 6.600546979904175,
      "epoch": 0.14987328497771563,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999559997392554,
      "loss": 6.4769,
      "mean_token_accuracy": 0.13354343175888062,
      "num_tokens": 3487703.0,
      "step": 1715
    },
    {
      "entropy": 6.499100685119629,
      "epoch": 0.15031023333042035,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999553813518061,
      "loss": 6.5618,
      "mean_token_accuracy": 0.11586256697773933,
      "num_tokens": 3497757.0,
      "step": 1720
    },
    {
      "entropy": 6.590635681152344,
      "epoch": 0.15074718168312506,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999547586495946,
      "loss": 6.5057,
      "mean_token_accuracy": 0.1161200724542141,
      "num_tokens": 3507594.0,
      "step": 1725
    },
    {
      "entropy": 6.565278720855713,
      "epoch": 0.15118413003582976,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999541316326329,
      "loss": 6.5941,
      "mean_token_accuracy": 0.11325425505638123,
      "num_tokens": 3517889.0,
      "step": 1730
    },
    {
      "entropy": 6.582718229293823,
      "epoch": 0.15162107838853447,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999535003009332,
      "loss": 6.5919,
      "mean_token_accuracy": 0.1146755650639534,
      "num_tokens": 3528158.0,
      "step": 1735
    },
    {
      "entropy": 6.656514263153076,
      "epoch": 0.1520580267412392,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999528646545074,
      "loss": 6.5854,
      "mean_token_accuracy": 0.12044015303254127,
      "num_tokens": 3539099.0,
      "step": 1740
    },
    {
      "entropy": 6.604732513427734,
      "epoch": 0.15249497509394389,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999522246933678,
      "loss": 6.5961,
      "mean_token_accuracy": 0.1175947681069374,
      "num_tokens": 3549819.0,
      "step": 1745
    },
    {
      "entropy": 6.599875211715698,
      "epoch": 0.1529319234466486,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999515804175268,
      "loss": 6.596,
      "mean_token_accuracy": 0.11611012145876884,
      "num_tokens": 3560022.0,
      "step": 1750
    },
    {
      "entropy": 6.581331825256347,
      "epoch": 0.15336887179935332,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999509318269964,
      "loss": 6.5371,
      "mean_token_accuracy": 0.11462760418653488,
      "num_tokens": 3569681.0,
      "step": 1755
    },
    {
      "entropy": 6.580878448486328,
      "epoch": 0.15380582015205801,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999502789217894,
      "loss": 6.6168,
      "mean_token_accuracy": 0.11628764718770981,
      "num_tokens": 3580731.0,
      "step": 1760
    },
    {
      "entropy": 6.541976404190064,
      "epoch": 0.15424276850476273,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999496217019181,
      "loss": 6.5798,
      "mean_token_accuracy": 0.11936882063746453,
      "num_tokens": 3590482.0,
      "step": 1765
    },
    {
      "entropy": 6.561248731613159,
      "epoch": 0.15467971685746745,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999489601673952,
      "loss": 6.5756,
      "mean_token_accuracy": 0.11308965906500816,
      "num_tokens": 3600988.0,
      "step": 1770
    },
    {
      "entropy": 6.729976606369019,
      "epoch": 0.15511666521017214,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999482943182334,
      "loss": 6.6527,
      "mean_token_accuracy": 0.11284584775567055,
      "num_tokens": 3611226.0,
      "step": 1775
    },
    {
      "entropy": 6.494902324676514,
      "epoch": 0.15555361356287686,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999476241544454,
      "loss": 6.5584,
      "mean_token_accuracy": 0.11853229030966758,
      "num_tokens": 3621324.0,
      "step": 1780
    },
    {
      "entropy": 6.614112901687622,
      "epoch": 0.15599056191558158,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999469496760441,
      "loss": 6.4845,
      "mean_token_accuracy": 0.11637181118130684,
      "num_tokens": 3631650.0,
      "step": 1785
    },
    {
      "entropy": 6.576256132125854,
      "epoch": 0.1564275102682863,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999462708830425,
      "loss": 6.5023,
      "mean_token_accuracy": 0.1310523718595505,
      "num_tokens": 3641173.0,
      "step": 1790
    },
    {
      "entropy": 6.535652542114258,
      "epoch": 0.156864458620991,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999455877754535,
      "loss": 6.5203,
      "mean_token_accuracy": 0.11737008094787597,
      "num_tokens": 3651846.0,
      "step": 1795
    },
    {
      "entropy": 6.526522207260132,
      "epoch": 0.1573014069736957,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999449003532903,
      "loss": 6.5324,
      "mean_token_accuracy": 0.11961890906095504,
      "num_tokens": 3661174.0,
      "step": 1800
    },
    {
      "entropy": 6.528283977508545,
      "epoch": 0.15773835532640043,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999442086165661,
      "loss": 6.487,
      "mean_token_accuracy": 0.11840147376060486,
      "num_tokens": 3671042.0,
      "step": 1805
    },
    {
      "entropy": 6.552019643783569,
      "epoch": 0.15817530367910512,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000499943512565294,
      "loss": 6.4648,
      "mean_token_accuracy": 0.11766612604260444,
      "num_tokens": 3681206.0,
      "step": 1810
    },
    {
      "entropy": 6.509097480773926,
      "epoch": 0.15861225203180984,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999428121994874,
      "loss": 6.5826,
      "mean_token_accuracy": 0.11521901562809944,
      "num_tokens": 3691748.0,
      "step": 1815
    },
    {
      "entropy": 6.594895744323731,
      "epoch": 0.15904920038451456,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999421075191599,
      "loss": 6.4523,
      "mean_token_accuracy": 0.1187455803155899,
      "num_tokens": 3701730.0,
      "step": 1820
    },
    {
      "entropy": 6.495048093795776,
      "epoch": 0.15948614873721925,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004999413985243249,
      "loss": 6.5631,
      "mean_token_accuracy": 0.11405811235308647,
      "num_tokens": 3712525.0,
      "step": 1825
    },
    {
      "entropy": 6.652899694442749,
      "epoch": 0.15992309708992397,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499940685214996,
      "loss": 6.5304,
      "mean_token_accuracy": 0.12245537713170052,
      "num_tokens": 3722016.0,
      "step": 1830
    },
    {
      "entropy": 6.5295939445495605,
      "epoch": 0.1603600454426287,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999399675911869,
      "loss": 6.4453,
      "mean_token_accuracy": 0.11833713129162789,
      "num_tokens": 3731299.0,
      "step": 1835
    },
    {
      "entropy": 6.487839317321777,
      "epoch": 0.16079699379533338,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999392456529113,
      "loss": 6.5011,
      "mean_token_accuracy": 0.1233910709619522,
      "num_tokens": 3740922.0,
      "step": 1840
    },
    {
      "entropy": 6.48659315109253,
      "epoch": 0.1612339421480381,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999385194001831,
      "loss": 6.4982,
      "mean_token_accuracy": 0.12182619720697403,
      "num_tokens": 3750339.0,
      "step": 1845
    },
    {
      "entropy": 6.537278699874878,
      "epoch": 0.16167089050074282,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999377888330164,
      "loss": 6.4728,
      "mean_token_accuracy": 0.12529798746109008,
      "num_tokens": 3761379.0,
      "step": 1850
    },
    {
      "entropy": 6.524621915817261,
      "epoch": 0.1621078388534475,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999370539514249,
      "loss": 6.5474,
      "mean_token_accuracy": 0.11859968900680543,
      "num_tokens": 3771649.0,
      "step": 1855
    },
    {
      "entropy": 6.492616367340088,
      "epoch": 0.16254478720615223,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999363147554229,
      "loss": 6.4554,
      "mean_token_accuracy": 0.12341330200433731,
      "num_tokens": 3782393.0,
      "step": 1860
    },
    {
      "entropy": 6.598450231552124,
      "epoch": 0.16298173555885695,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999355712450244,
      "loss": 6.5257,
      "mean_token_accuracy": 0.1240497075021267,
      "num_tokens": 3792849.0,
      "step": 1865
    },
    {
      "entropy": 6.5042177677154545,
      "epoch": 0.16341868391156164,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999348234202438,
      "loss": 6.4764,
      "mean_token_accuracy": 0.12349856197834015,
      "num_tokens": 3803848.0,
      "step": 1870
    },
    {
      "entropy": 6.288655757904053,
      "epoch": 0.16385563226426636,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999340712810956,
      "loss": 6.4098,
      "mean_token_accuracy": 0.13501542136073114,
      "num_tokens": 3814704.0,
      "step": 1875
    },
    {
      "entropy": 6.517355871200562,
      "epoch": 0.16429258061697108,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999333148275939,
      "loss": 6.5381,
      "mean_token_accuracy": 0.11455504298210144,
      "num_tokens": 3824955.0,
      "step": 1880
    },
    {
      "entropy": 6.5474401950836185,
      "epoch": 0.1647295289696758,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999325540597535,
      "loss": 6.5202,
      "mean_token_accuracy": 0.12205357849597931,
      "num_tokens": 3836540.0,
      "step": 1885
    },
    {
      "entropy": 6.465744876861573,
      "epoch": 0.1651664773223805,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999317889775887,
      "loss": 6.4075,
      "mean_token_accuracy": 0.1288439191877842,
      "num_tokens": 3846472.0,
      "step": 1890
    },
    {
      "entropy": 6.521185493469238,
      "epoch": 0.1656034256750852,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999310195811143,
      "loss": 6.5135,
      "mean_token_accuracy": 0.11601202338933944,
      "num_tokens": 3857187.0,
      "step": 1895
    },
    {
      "entropy": 6.53255524635315,
      "epoch": 0.16604037402778993,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999302458703451,
      "loss": 6.4815,
      "mean_token_accuracy": 0.12292045503854751,
      "num_tokens": 3866259.0,
      "step": 1900
    },
    {
      "entropy": 6.447905969619751,
      "epoch": 0.16647732238049462,
      "grad_norm": 1.265625,
      "learning_rate": 0.000499929467845296,
      "loss": 6.5408,
      "mean_token_accuracy": 0.11351652964949607,
      "num_tokens": 3876869.0,
      "step": 1905
    },
    {
      "entropy": 6.610359334945679,
      "epoch": 0.16691427073319934,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999286855059818,
      "loss": 6.5203,
      "mean_token_accuracy": 0.12371717020869255,
      "num_tokens": 3888278.0,
      "step": 1910
    },
    {
      "entropy": 6.551647138595581,
      "epoch": 0.16735121908590406,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004999278988524177,
      "loss": 6.5777,
      "mean_token_accuracy": 0.11743155568838119,
      "num_tokens": 3898648.0,
      "step": 1915
    },
    {
      "entropy": 6.493818187713623,
      "epoch": 0.16778816743860875,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999271078846184,
      "loss": 6.4898,
      "mean_token_accuracy": 0.120925023406744,
      "num_tokens": 3908282.0,
      "step": 1920
    },
    {
      "entropy": 6.4657470226287845,
      "epoch": 0.16822511579131347,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999263126025994,
      "loss": 6.4982,
      "mean_token_accuracy": 0.11906409412622451,
      "num_tokens": 3918333.0,
      "step": 1925
    },
    {
      "entropy": 6.558496046066284,
      "epoch": 0.1686620641440182,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999255130063759,
      "loss": 6.4876,
      "mean_token_accuracy": 0.12188828215003014,
      "num_tokens": 3927403.0,
      "step": 1930
    },
    {
      "entropy": 6.521365690231323,
      "epoch": 0.16909901249672288,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999247090959633,
      "loss": 6.5066,
      "mean_token_accuracy": 0.12119645178318024,
      "num_tokens": 3937352.0,
      "step": 1935
    },
    {
      "entropy": 6.453633642196655,
      "epoch": 0.1695359608494276,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999239008713767,
      "loss": 6.3406,
      "mean_token_accuracy": 0.12555457800626754,
      "num_tokens": 3947164.0,
      "step": 1940
    },
    {
      "entropy": 6.530312156677246,
      "epoch": 0.16997290920213232,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999230883326319,
      "loss": 6.5191,
      "mean_token_accuracy": 0.11990117952227593,
      "num_tokens": 3956887.0,
      "step": 1945
    },
    {
      "entropy": 6.510148668289185,
      "epoch": 0.170409857554837,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999222714797445,
      "loss": 6.4802,
      "mean_token_accuracy": 0.118320432305336,
      "num_tokens": 3966542.0,
      "step": 1950
    },
    {
      "entropy": 6.5808569431304935,
      "epoch": 0.17084680590754173,
      "grad_norm": 1.109375,
      "learning_rate": 0.00049992145031273,
      "loss": 6.4445,
      "mean_token_accuracy": 0.1298608437180519,
      "num_tokens": 3976127.0,
      "step": 1955
    },
    {
      "entropy": 6.371415281295777,
      "epoch": 0.17128375426024645,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999206248316043,
      "loss": 6.4302,
      "mean_token_accuracy": 0.12315708696842194,
      "num_tokens": 3987371.0,
      "step": 1960
    },
    {
      "entropy": 6.443069648742676,
      "epoch": 0.17172070261295114,
      "grad_norm": 1.25,
      "learning_rate": 0.000499919795036383,
      "loss": 6.4361,
      "mean_token_accuracy": 0.12713590562343596,
      "num_tokens": 3997924.0,
      "step": 1965
    },
    {
      "entropy": 6.529845285415649,
      "epoch": 0.17215765096565586,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999189609270823,
      "loss": 6.5188,
      "mean_token_accuracy": 0.12592333555221558,
      "num_tokens": 4009150.0,
      "step": 1970
    },
    {
      "entropy": 6.494536066055298,
      "epoch": 0.17259459931836058,
      "grad_norm": 1.375,
      "learning_rate": 0.000499918122503718,
      "loss": 6.5196,
      "mean_token_accuracy": 0.12174422815442085,
      "num_tokens": 4018899.0,
      "step": 1975
    },
    {
      "entropy": 6.504218292236328,
      "epoch": 0.17303154767106527,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999172797663064,
      "loss": 6.4728,
      "mean_token_accuracy": 0.11927963867783546,
      "num_tokens": 4028652.0,
      "step": 1980
    },
    {
      "entropy": 6.526062726974487,
      "epoch": 0.17346849602376999,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999164327148634,
      "loss": 6.5272,
      "mean_token_accuracy": 0.12078682556748391,
      "num_tokens": 4039819.0,
      "step": 1985
    },
    {
      "entropy": 6.509147644042969,
      "epoch": 0.1739054443764747,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999155813494053,
      "loss": 6.4176,
      "mean_token_accuracy": 0.12781703993678092,
      "num_tokens": 4050029.0,
      "step": 1990
    },
    {
      "entropy": 6.421483421325684,
      "epoch": 0.17434239272917942,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999147256699487,
      "loss": 6.4811,
      "mean_token_accuracy": 0.12011851966381074,
      "num_tokens": 4060497.0,
      "step": 1995
    },
    {
      "entropy": 6.563172960281372,
      "epoch": 0.17477934108188412,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999138656765096,
      "loss": 6.4376,
      "mean_token_accuracy": 0.12605830207467078,
      "num_tokens": 4071477.0,
      "step": 2000
    },
    {
      "entropy": 6.576144123077393,
      "epoch": 0.17521628943458883,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999130013691049,
      "loss": 6.5454,
      "mean_token_accuracy": 0.11874540522694588,
      "num_tokens": 4081869.0,
      "step": 2005
    },
    {
      "entropy": 6.570287227630615,
      "epoch": 0.17565323778729355,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999121327477509,
      "loss": 6.5252,
      "mean_token_accuracy": 0.12263748943805694,
      "num_tokens": 4092797.0,
      "step": 2010
    },
    {
      "entropy": 6.444070291519165,
      "epoch": 0.17609018613999824,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999112598124642,
      "loss": 6.4571,
      "mean_token_accuracy": 0.11857560724020004,
      "num_tokens": 4103688.0,
      "step": 2015
    },
    {
      "entropy": 6.4113997459411625,
      "epoch": 0.17652713449270296,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999103825632618,
      "loss": 6.4368,
      "mean_token_accuracy": 0.12056360468268394,
      "num_tokens": 4114037.0,
      "step": 2020
    },
    {
      "entropy": 6.5229727268219,
      "epoch": 0.17696408284540768,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004999095010001605,
      "loss": 6.4223,
      "mean_token_accuracy": 0.12138459011912346,
      "num_tokens": 4124164.0,
      "step": 2025
    },
    {
      "entropy": 6.470005750656128,
      "epoch": 0.17740103119811237,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499908615123177,
      "loss": 6.5271,
      "mean_token_accuracy": 0.11661396995186805,
      "num_tokens": 4134225.0,
      "step": 2030
    },
    {
      "entropy": 6.5835813045501705,
      "epoch": 0.1778379795508171,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999077249323284,
      "loss": 6.5804,
      "mean_token_accuracy": 0.1160724326968193,
      "num_tokens": 4145962.0,
      "step": 2035
    },
    {
      "entropy": 6.540454864501953,
      "epoch": 0.1782749279035218,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999068304276319,
      "loss": 6.4559,
      "mean_token_accuracy": 0.11698207259178162,
      "num_tokens": 4156572.0,
      "step": 2040
    },
    {
      "entropy": 6.494862699508667,
      "epoch": 0.1787118762562265,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999059316091045,
      "loss": 6.5379,
      "mean_token_accuracy": 0.117926075309515,
      "num_tokens": 4166960.0,
      "step": 2045
    },
    {
      "entropy": 6.504702520370484,
      "epoch": 0.17914882460893122,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999050284767634,
      "loss": 6.4799,
      "mean_token_accuracy": 0.1223859153687954,
      "num_tokens": 4177833.0,
      "step": 2050
    },
    {
      "entropy": 6.444409275054932,
      "epoch": 0.17958577296163594,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999041210306261,
      "loss": 6.4169,
      "mean_token_accuracy": 0.12724354416131972,
      "num_tokens": 4187858.0,
      "step": 2055
    },
    {
      "entropy": 6.524564933776856,
      "epoch": 0.18002272131434063,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999032092707098,
      "loss": 6.5382,
      "mean_token_accuracy": 0.11748412624001503,
      "num_tokens": 4198207.0,
      "step": 2060
    },
    {
      "entropy": 6.572392845153809,
      "epoch": 0.18045966966704535,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004999022931970323,
      "loss": 6.4191,
      "mean_token_accuracy": 0.1241879142820835,
      "num_tokens": 4207594.0,
      "step": 2065
    },
    {
      "entropy": 6.38447756767273,
      "epoch": 0.18089661801975007,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999013728096107,
      "loss": 6.4312,
      "mean_token_accuracy": 0.125095746666193,
      "num_tokens": 4217334.0,
      "step": 2070
    },
    {
      "entropy": 6.526350355148315,
      "epoch": 0.18133356637245476,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499900448108463,
      "loss": 6.4185,
      "mean_token_accuracy": 0.12915206849575042,
      "num_tokens": 4226407.0,
      "step": 2075
    },
    {
      "entropy": 6.489539337158203,
      "epoch": 0.18177051472515948,
      "grad_norm": 1.9296875,
      "learning_rate": 0.000499899519093607,
      "loss": 6.3787,
      "mean_token_accuracy": 0.12649197205901147,
      "num_tokens": 4237818.0,
      "step": 2080
    },
    {
      "entropy": 6.521691370010376,
      "epoch": 0.1822074630778642,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004998985857650603,
      "loss": 6.4944,
      "mean_token_accuracy": 0.11592372059822083,
      "num_tokens": 4247783.0,
      "step": 2085
    },
    {
      "entropy": 6.413587856292724,
      "epoch": 0.18264441143056892,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004998976481228409,
      "loss": 6.4355,
      "mean_token_accuracy": 0.1198279656469822,
      "num_tokens": 4258049.0,
      "step": 2090
    },
    {
      "entropy": 6.4406171321868895,
      "epoch": 0.1830813597832736,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998967061669668,
      "loss": 6.4463,
      "mean_token_accuracy": 0.11746001914143563,
      "num_tokens": 4269357.0,
      "step": 2095
    },
    {
      "entropy": 6.498240995407104,
      "epoch": 0.18351830813597833,
      "grad_norm": 1.25,
      "learning_rate": 0.0004998957598974559,
      "loss": 6.3522,
      "mean_token_accuracy": 0.12721729502081872,
      "num_tokens": 4279576.0,
      "step": 2100
    },
    {
      "entropy": 6.463101100921631,
      "epoch": 0.18395525648868305,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004998948093143266,
      "loss": 6.4257,
      "mean_token_accuracy": 0.12523012161254882,
      "num_tokens": 4289761.0,
      "step": 2105
    },
    {
      "entropy": 6.442949867248535,
      "epoch": 0.18439220484138774,
      "grad_norm": 1.59375,
      "learning_rate": 0.000499893854417597,
      "loss": 6.3533,
      "mean_token_accuracy": 0.12625055760145187,
      "num_tokens": 4299231.0,
      "step": 2110
    },
    {
      "entropy": 6.447075605392456,
      "epoch": 0.18482915319409246,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998928952072854,
      "loss": 6.5121,
      "mean_token_accuracy": 0.11597891449928284,
      "num_tokens": 4308687.0,
      "step": 2115
    },
    {
      "entropy": 6.5224504470825195,
      "epoch": 0.18526610154679718,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998919316834102,
      "loss": 6.4698,
      "mean_token_accuracy": 0.11885222047567368,
      "num_tokens": 4318183.0,
      "step": 2120
    },
    {
      "entropy": 6.467560958862305,
      "epoch": 0.18570304989950187,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00049989096384599,
      "loss": 6.5139,
      "mean_token_accuracy": 0.12279015704989434,
      "num_tokens": 4328482.0,
      "step": 2125
    },
    {
      "entropy": 6.529058790206909,
      "epoch": 0.1861399982522066,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004998899916950431,
      "loss": 6.3821,
      "mean_token_accuracy": 0.1252948209643364,
      "num_tokens": 4339313.0,
      "step": 2130
    },
    {
      "entropy": 6.460389947891235,
      "epoch": 0.1865769466049113,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998890152305885,
      "loss": 6.4944,
      "mean_token_accuracy": 0.12029476463794708,
      "num_tokens": 4350299.0,
      "step": 2135
    },
    {
      "entropy": 6.528489065170288,
      "epoch": 0.187013894957616,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998880344526447,
      "loss": 6.4391,
      "mean_token_accuracy": 0.12706440985202788,
      "num_tokens": 4361424.0,
      "step": 2140
    },
    {
      "entropy": 6.482869291305542,
      "epoch": 0.18745084331032072,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998870493612306,
      "loss": 6.407,
      "mean_token_accuracy": 0.12022507637739181,
      "num_tokens": 4371480.0,
      "step": 2145
    },
    {
      "entropy": 6.387491750717163,
      "epoch": 0.18788779166302544,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998860599563649,
      "loss": 6.4422,
      "mean_token_accuracy": 0.12101913392543792,
      "num_tokens": 4381951.0,
      "step": 2150
    },
    {
      "entropy": 6.489475870132447,
      "epoch": 0.18832474001573013,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004998850662380668,
      "loss": 6.5016,
      "mean_token_accuracy": 0.1167457014322281,
      "num_tokens": 4392290.0,
      "step": 2155
    },
    {
      "entropy": 6.4344056129455565,
      "epoch": 0.18876168836843485,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004998840682063553,
      "loss": 6.2632,
      "mean_token_accuracy": 0.135141958296299,
      "num_tokens": 4403350.0,
      "step": 2160
    },
    {
      "entropy": 6.418215799331665,
      "epoch": 0.18919863672113957,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998830658612496,
      "loss": 6.4718,
      "mean_token_accuracy": 0.11757678985595703,
      "num_tokens": 4413171.0,
      "step": 2165
    },
    {
      "entropy": 6.456563186645508,
      "epoch": 0.18963558507384426,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004998820592027688,
      "loss": 6.4049,
      "mean_token_accuracy": 0.1265949308872223,
      "num_tokens": 4424371.0,
      "step": 2170
    },
    {
      "entropy": 6.361171054840088,
      "epoch": 0.19007253342654898,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998810482309322,
      "loss": 6.3929,
      "mean_token_accuracy": 0.11713957265019417,
      "num_tokens": 4434040.0,
      "step": 2175
    },
    {
      "entropy": 6.585212564468383,
      "epoch": 0.1905094817792537,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998800329457595,
      "loss": 6.4795,
      "mean_token_accuracy": 0.11658665388822556,
      "num_tokens": 4444663.0,
      "step": 2180
    },
    {
      "entropy": 6.493283939361572,
      "epoch": 0.1909464301319584,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998790133472697,
      "loss": 6.4233,
      "mean_token_accuracy": 0.11703696846961975,
      "num_tokens": 4454733.0,
      "step": 2185
    },
    {
      "entropy": 6.449396371841431,
      "epoch": 0.1913833784846631,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998779894354827,
      "loss": 6.409,
      "mean_token_accuracy": 0.12359335049986839,
      "num_tokens": 4464286.0,
      "step": 2190
    },
    {
      "entropy": 6.488334655761719,
      "epoch": 0.19182032683736783,
      "grad_norm": 1.125,
      "learning_rate": 0.000499876961210418,
      "loss": 6.3958,
      "mean_token_accuracy": 0.13071504160761832,
      "num_tokens": 4475005.0,
      "step": 2195
    },
    {
      "entropy": 6.437582492828369,
      "epoch": 0.19225727519007255,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998759286720953,
      "loss": 6.3682,
      "mean_token_accuracy": 0.12397781834006309,
      "num_tokens": 4484606.0,
      "step": 2200
    },
    {
      "entropy": 6.396815586090088,
      "epoch": 0.19269422354277724,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004998748918205345,
      "loss": 6.4733,
      "mean_token_accuracy": 0.11907115429639817,
      "num_tokens": 4494930.0,
      "step": 2205
    },
    {
      "entropy": 6.471502208709717,
      "epoch": 0.19313117189548196,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004998738506557554,
      "loss": 6.4905,
      "mean_token_accuracy": 0.11480139791965485,
      "num_tokens": 4505304.0,
      "step": 2210
    },
    {
      "entropy": 6.591741514205933,
      "epoch": 0.19356812024818668,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499872805177778,
      "loss": 6.4211,
      "mean_token_accuracy": 0.12898224964737892,
      "num_tokens": 4514333.0,
      "step": 2215
    },
    {
      "entropy": 6.402223110198975,
      "epoch": 0.19400506860089137,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998717553866225,
      "loss": 6.4659,
      "mean_token_accuracy": 0.1265234135091305,
      "num_tokens": 4525175.0,
      "step": 2220
    },
    {
      "entropy": 6.485518169403076,
      "epoch": 0.1944420169535961,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998707012823087,
      "loss": 6.3523,
      "mean_token_accuracy": 0.12841272950172425,
      "num_tokens": 4534212.0,
      "step": 2225
    },
    {
      "entropy": 6.408768939971924,
      "epoch": 0.1948789653063008,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998696428648573,
      "loss": 6.5188,
      "mean_token_accuracy": 0.12075343132019042,
      "num_tokens": 4545141.0,
      "step": 2230
    },
    {
      "entropy": 6.482947492599488,
      "epoch": 0.1953159136590055,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004998685801342881,
      "loss": 6.3669,
      "mean_token_accuracy": 0.1291575625538826,
      "num_tokens": 4554410.0,
      "step": 2235
    },
    {
      "entropy": 6.375760126113891,
      "epoch": 0.19575286201171022,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998675130906217,
      "loss": 6.3699,
      "mean_token_accuracy": 0.12993818074464797,
      "num_tokens": 4564397.0,
      "step": 2240
    },
    {
      "entropy": 6.3663304328918455,
      "epoch": 0.19618981036441493,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998664417338787,
      "loss": 6.3751,
      "mean_token_accuracy": 0.12797396183013915,
      "num_tokens": 4573986.0,
      "step": 2245
    },
    {
      "entropy": 6.509523153305054,
      "epoch": 0.19662675871711963,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004998653660640794,
      "loss": 6.3699,
      "mean_token_accuracy": 0.13107719346880914,
      "num_tokens": 4583449.0,
      "step": 2250
    },
    {
      "entropy": 6.419308996200561,
      "epoch": 0.19706370706982435,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004998642860812447,
      "loss": 6.4144,
      "mean_token_accuracy": 0.12404663264751434,
      "num_tokens": 4593637.0,
      "step": 2255
    },
    {
      "entropy": 6.3758574485778805,
      "epoch": 0.19750065542252906,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004998632017853951,
      "loss": 6.4012,
      "mean_token_accuracy": 0.12606620118021966,
      "num_tokens": 4603865.0,
      "step": 2260
    },
    {
      "entropy": 6.584481143951416,
      "epoch": 0.19793760377523376,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004998621131765514,
      "loss": 6.4357,
      "mean_token_accuracy": 0.12628217190504074,
      "num_tokens": 4613502.0,
      "step": 2265
    },
    {
      "entropy": 6.38882474899292,
      "epoch": 0.19837455212793847,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998610202547345,
      "loss": 6.3999,
      "mean_token_accuracy": 0.12597004026174546,
      "num_tokens": 4623163.0,
      "step": 2270
    },
    {
      "entropy": 6.471518421173096,
      "epoch": 0.1988115004806432,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998599230199654,
      "loss": 6.4193,
      "mean_token_accuracy": 0.13149505183100701,
      "num_tokens": 4632850.0,
      "step": 2275
    },
    {
      "entropy": 6.44390549659729,
      "epoch": 0.19924844883334789,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004998588214722653,
      "loss": 6.413,
      "mean_token_accuracy": 0.12612975835800172,
      "num_tokens": 4642858.0,
      "step": 2280
    },
    {
      "entropy": 6.517856550216675,
      "epoch": 0.1996853971860526,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998577156116551,
      "loss": 6.444,
      "mean_token_accuracy": 0.12016589418053628,
      "num_tokens": 4653083.0,
      "step": 2285
    },
    {
      "entropy": 6.466843843460083,
      "epoch": 0.20012234553875732,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998566054381561,
      "loss": 6.4486,
      "mean_token_accuracy": 0.11985157281160355,
      "num_tokens": 4663353.0,
      "step": 2290
    },
    {
      "entropy": 6.445455694198609,
      "epoch": 0.20055929389146204,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004998554909517895,
      "loss": 6.3183,
      "mean_token_accuracy": 0.12755734249949455,
      "num_tokens": 4673878.0,
      "step": 2295
    },
    {
      "entropy": 6.427143669128418,
      "epoch": 0.20099624224416673,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004998543721525769,
      "loss": 6.3877,
      "mean_token_accuracy": 0.12570685148239136,
      "num_tokens": 4683661.0,
      "step": 2300
    },
    {
      "entropy": 6.355686902999878,
      "epoch": 0.20143319059687145,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004998532490405397,
      "loss": 6.445,
      "mean_token_accuracy": 0.12695046067237853,
      "num_tokens": 4693817.0,
      "step": 2305
    },
    {
      "entropy": 6.415957927703857,
      "epoch": 0.20187013894957617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998521216156991,
      "loss": 6.3672,
      "mean_token_accuracy": 0.1261789694428444,
      "num_tokens": 4703550.0,
      "step": 2310
    },
    {
      "entropy": 6.463042449951172,
      "epoch": 0.20230708730228086,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998509898780771,
      "loss": 6.4915,
      "mean_token_accuracy": 0.1210303895175457,
      "num_tokens": 4714872.0,
      "step": 2315
    },
    {
      "entropy": 6.400016498565674,
      "epoch": 0.20274403565498558,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998498538276954,
      "loss": 6.3665,
      "mean_token_accuracy": 0.12931155189871787,
      "num_tokens": 4725745.0,
      "step": 2320
    },
    {
      "entropy": 6.441245460510254,
      "epoch": 0.2031809840076903,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004998487134645755,
      "loss": 6.4628,
      "mean_token_accuracy": 0.12229650542140007,
      "num_tokens": 4735560.0,
      "step": 2325
    },
    {
      "entropy": 6.49150595664978,
      "epoch": 0.203617932360395,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004998475687887396,
      "loss": 6.44,
      "mean_token_accuracy": 0.1181219644844532,
      "num_tokens": 4746304.0,
      "step": 2330
    },
    {
      "entropy": 6.52930097579956,
      "epoch": 0.2040548807130997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998464198002095,
      "loss": 6.3942,
      "mean_token_accuracy": 0.12358200997114181,
      "num_tokens": 4756200.0,
      "step": 2335
    },
    {
      "entropy": 6.371595859527588,
      "epoch": 0.20449182906580443,
      "grad_norm": 1.25,
      "learning_rate": 0.0004998452664990072,
      "loss": 6.4228,
      "mean_token_accuracy": 0.12109901309013367,
      "num_tokens": 4766249.0,
      "step": 2340
    },
    {
      "entropy": 6.5394433498382565,
      "epoch": 0.20492877741850912,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004998441088851548,
      "loss": 6.4503,
      "mean_token_accuracy": 0.11931437700986862,
      "num_tokens": 4776510.0,
      "step": 2345
    },
    {
      "entropy": 6.455130910873413,
      "epoch": 0.20536572577121384,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998429469586748,
      "loss": 6.4275,
      "mean_token_accuracy": 0.12399638965725898,
      "num_tokens": 4786798.0,
      "step": 2350
    },
    {
      "entropy": 6.386494255065918,
      "epoch": 0.20580267412391856,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000499841780719589,
      "loss": 6.3515,
      "mean_token_accuracy": 0.12153764739632607,
      "num_tokens": 4796001.0,
      "step": 2355
    },
    {
      "entropy": 6.400313091278076,
      "epoch": 0.20623962247662325,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004998406101679202,
      "loss": 6.335,
      "mean_token_accuracy": 0.13208253532648087,
      "num_tokens": 4807634.0,
      "step": 2360
    },
    {
      "entropy": 6.462019729614258,
      "epoch": 0.20667657082932797,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004998394353036906,
      "loss": 6.3527,
      "mean_token_accuracy": 0.1344490632414818,
      "num_tokens": 4817840.0,
      "step": 2365
    },
    {
      "entropy": 6.37156572341919,
      "epoch": 0.2071135191820327,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998382561269229,
      "loss": 6.3228,
      "mean_token_accuracy": 0.1289958320558071,
      "num_tokens": 4827186.0,
      "step": 2370
    },
    {
      "entropy": 6.284419775009155,
      "epoch": 0.20755046753473738,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998370726376395,
      "loss": 6.3715,
      "mean_token_accuracy": 0.12209030091762543,
      "num_tokens": 4837140.0,
      "step": 2375
    },
    {
      "entropy": 6.49763560295105,
      "epoch": 0.2079874158874421,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004998358848358634,
      "loss": 6.3981,
      "mean_token_accuracy": 0.12760360389947892,
      "num_tokens": 4846907.0,
      "step": 2380
    },
    {
      "entropy": 6.359625816345215,
      "epoch": 0.20842436424014682,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000499834692721617,
      "loss": 6.4259,
      "mean_token_accuracy": 0.12730940505862237,
      "num_tokens": 4856165.0,
      "step": 2385
    },
    {
      "entropy": 6.4935239315032955,
      "epoch": 0.2088613125928515,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004998334962949237,
      "loss": 6.2773,
      "mean_token_accuracy": 0.13658949360251427,
      "num_tokens": 4865308.0,
      "step": 2390
    },
    {
      "entropy": 6.358750247955323,
      "epoch": 0.20929826094555623,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004998322955558059,
      "loss": 6.3561,
      "mean_token_accuracy": 0.13445974960923196,
      "num_tokens": 4876286.0,
      "step": 2395
    },
    {
      "entropy": 6.389959907531738,
      "epoch": 0.20973520929826095,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998310905042869,
      "loss": 6.3916,
      "mean_token_accuracy": 0.12824709340929985,
      "num_tokens": 4887227.0,
      "step": 2400
    },
    {
      "entropy": 6.459248208999634,
      "epoch": 0.21017215765096567,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998298811403898,
      "loss": 6.4181,
      "mean_token_accuracy": 0.1229472927749157,
      "num_tokens": 4896994.0,
      "step": 2405
    },
    {
      "entropy": 6.416028738021851,
      "epoch": 0.21060910600367036,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004998286674641378,
      "loss": 6.3954,
      "mean_token_accuracy": 0.1258566915988922,
      "num_tokens": 4907623.0,
      "step": 2410
    },
    {
      "entropy": 6.339293336868286,
      "epoch": 0.21104605435637508,
      "grad_norm": 1.234375,
      "learning_rate": 0.000499827449475554,
      "loss": 6.3075,
      "mean_token_accuracy": 0.12986938655376434,
      "num_tokens": 4917040.0,
      "step": 2415
    },
    {
      "entropy": 6.543447732925415,
      "epoch": 0.2114830027090798,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499826227174662,
      "loss": 6.4469,
      "mean_token_accuracy": 0.12280156686902047,
      "num_tokens": 4928390.0,
      "step": 2420
    },
    {
      "entropy": 6.359825563430786,
      "epoch": 0.2119199510617845,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998250005614852,
      "loss": 6.4135,
      "mean_token_accuracy": 0.12152730897068978,
      "num_tokens": 4939040.0,
      "step": 2425
    },
    {
      "entropy": 6.393878936767578,
      "epoch": 0.2123568994144892,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000499823769636047,
      "loss": 6.4349,
      "mean_token_accuracy": 0.12271395474672317,
      "num_tokens": 4949915.0,
      "step": 2430
    },
    {
      "entropy": 6.448448228836059,
      "epoch": 0.21279384776719393,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998225343983712,
      "loss": 6.3378,
      "mean_token_accuracy": 0.12833459079265594,
      "num_tokens": 4960127.0,
      "step": 2435
    },
    {
      "entropy": 6.440428781509399,
      "epoch": 0.21323079611989862,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998212948484814,
      "loss": 6.3871,
      "mean_token_accuracy": 0.12280435562133789,
      "num_tokens": 4970152.0,
      "step": 2440
    },
    {
      "entropy": 6.411177682876587,
      "epoch": 0.21366774447260334,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998200509864012,
      "loss": 6.3458,
      "mean_token_accuracy": 0.12693380266427995,
      "num_tokens": 4979745.0,
      "step": 2445
    },
    {
      "entropy": 6.479267740249634,
      "epoch": 0.21410469282530806,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998188028121547,
      "loss": 6.3747,
      "mean_token_accuracy": 0.13292665034532547,
      "num_tokens": 4989924.0,
      "step": 2450
    },
    {
      "entropy": 6.365407466888428,
      "epoch": 0.21454164117801275,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998175503257658,
      "loss": 6.3248,
      "mean_token_accuracy": 0.13266007974743843,
      "num_tokens": 4999698.0,
      "step": 2455
    },
    {
      "entropy": 6.399653434753418,
      "epoch": 0.21497858953071747,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004998162935272585,
      "loss": 6.433,
      "mean_token_accuracy": 0.12551234066486358,
      "num_tokens": 5010625.0,
      "step": 2460
    },
    {
      "entropy": 6.387683725357055,
      "epoch": 0.2154155378834222,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998150324166568,
      "loss": 6.3143,
      "mean_token_accuracy": 0.13393508419394493,
      "num_tokens": 5019740.0,
      "step": 2465
    },
    {
      "entropy": 6.487478494644165,
      "epoch": 0.21585248623612688,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499813766993985,
      "loss": 6.4293,
      "mean_token_accuracy": 0.12056878060102463,
      "num_tokens": 5030601.0,
      "step": 2470
    },
    {
      "entropy": 6.497150993347168,
      "epoch": 0.2162894345888316,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004998124972592674,
      "loss": 6.3906,
      "mean_token_accuracy": 0.1267626941204071,
      "num_tokens": 5040567.0,
      "step": 2475
    },
    {
      "entropy": 6.3080087184906,
      "epoch": 0.21672638294153632,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004998112232125283,
      "loss": 6.2682,
      "mean_token_accuracy": 0.13081765845417975,
      "num_tokens": 5049512.0,
      "step": 2480
    },
    {
      "entropy": 6.32685604095459,
      "epoch": 0.217163331294241,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499809944853792,
      "loss": 6.4609,
      "mean_token_accuracy": 0.12319278120994567,
      "num_tokens": 5060986.0,
      "step": 2485
    },
    {
      "entropy": 6.42396092414856,
      "epoch": 0.21760027964694573,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004998086621830833,
      "loss": 6.3755,
      "mean_token_accuracy": 0.12850141301751136,
      "num_tokens": 5071492.0,
      "step": 2490
    },
    {
      "entropy": 6.4019676685333256,
      "epoch": 0.21803722799965045,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998073752004267,
      "loss": 6.4115,
      "mean_token_accuracy": 0.12014096081256867,
      "num_tokens": 5081722.0,
      "step": 2495
    },
    {
      "entropy": 6.498205232620239,
      "epoch": 0.21847417635235516,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004998060839058467,
      "loss": 6.3881,
      "mean_token_accuracy": 0.12541978657245637,
      "num_tokens": 5091906.0,
      "step": 2500
    },
    {
      "entropy": 6.3498509407043455,
      "epoch": 0.21891112470505986,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004998047882993683,
      "loss": 6.3482,
      "mean_token_accuracy": 0.12800950780510903,
      "num_tokens": 5102025.0,
      "step": 2505
    },
    {
      "entropy": 6.381645059585571,
      "epoch": 0.21934807305776458,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998034883810162,
      "loss": 6.3475,
      "mean_token_accuracy": 0.12698294296860696,
      "num_tokens": 5112354.0,
      "step": 2510
    },
    {
      "entropy": 6.373659706115722,
      "epoch": 0.2197850214104693,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998021841508155,
      "loss": 6.289,
      "mean_token_accuracy": 0.13288134709000587,
      "num_tokens": 5121695.0,
      "step": 2515
    },
    {
      "entropy": 6.395002651214599,
      "epoch": 0.22022196976317399,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004998008756087911,
      "loss": 6.4074,
      "mean_token_accuracy": 0.1230700857937336,
      "num_tokens": 5132561.0,
      "step": 2520
    },
    {
      "entropy": 6.474684524536133,
      "epoch": 0.2206589181158787,
      "grad_norm": 1.828125,
      "learning_rate": 0.000499799562754968,
      "loss": 6.4335,
      "mean_token_accuracy": 0.12292742729187012,
      "num_tokens": 5141981.0,
      "step": 2525
    },
    {
      "entropy": 6.371050119400024,
      "epoch": 0.22109586646858342,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004997982455893716,
      "loss": 6.3251,
      "mean_token_accuracy": 0.1293874688446522,
      "num_tokens": 5151550.0,
      "step": 2530
    },
    {
      "entropy": 6.47653021812439,
      "epoch": 0.22153281482128812,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004997969241120271,
      "loss": 6.3947,
      "mean_token_accuracy": 0.1197041004896164,
      "num_tokens": 5161679.0,
      "step": 2535
    },
    {
      "entropy": 6.365097761154175,
      "epoch": 0.22196976317399283,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004997955983229597,
      "loss": 6.4342,
      "mean_token_accuracy": 0.1268290475010872,
      "num_tokens": 5171575.0,
      "step": 2540
    },
    {
      "entropy": 6.401581716537476,
      "epoch": 0.22240671152669755,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000499794268222195,
      "loss": 6.2896,
      "mean_token_accuracy": 0.13048343136906623,
      "num_tokens": 5180152.0,
      "step": 2545
    },
    {
      "entropy": 6.347995901107788,
      "epoch": 0.22284365987940224,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004997929338097583,
      "loss": 6.362,
      "mean_token_accuracy": 0.12662460505962372,
      "num_tokens": 5190518.0,
      "step": 2550
    },
    {
      "entropy": 6.484149408340454,
      "epoch": 0.22328060823210696,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004997915950856755,
      "loss": 6.3986,
      "mean_token_accuracy": 0.12541723176836966,
      "num_tokens": 5201729.0,
      "step": 2555
    },
    {
      "entropy": 6.404265975952148,
      "epoch": 0.22371755658481168,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004997902520499721,
      "loss": 6.4016,
      "mean_token_accuracy": 0.12332973778247833,
      "num_tokens": 5211940.0,
      "step": 2560
    },
    {
      "entropy": 6.325415325164795,
      "epoch": 0.22415450493751637,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004997889047026738,
      "loss": 6.3135,
      "mean_token_accuracy": 0.1277544841170311,
      "num_tokens": 5221028.0,
      "step": 2565
    },
    {
      "entropy": 6.368482255935669,
      "epoch": 0.2245914532902211,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004997875530438066,
      "loss": 6.3189,
      "mean_token_accuracy": 0.13291172459721565,
      "num_tokens": 5230498.0,
      "step": 2570
    },
    {
      "entropy": 6.430984592437744,
      "epoch": 0.2250284016429258,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004997861970733963,
      "loss": 6.3663,
      "mean_token_accuracy": 0.1215190939605236,
      "num_tokens": 5240112.0,
      "step": 2575
    },
    {
      "entropy": 6.458737134933472,
      "epoch": 0.2254653499956305,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000499784836791469,
      "loss": 6.3192,
      "mean_token_accuracy": 0.12820565402507783,
      "num_tokens": 5249201.0,
      "step": 2580
    },
    {
      "entropy": 6.335276365280151,
      "epoch": 0.22590229834833522,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004997834721980507,
      "loss": 6.331,
      "mean_token_accuracy": 0.12517586797475816,
      "num_tokens": 5258541.0,
      "step": 2585
    },
    {
      "entropy": 6.543854236602783,
      "epoch": 0.22633924670103994,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004997821032931676,
      "loss": 6.4331,
      "mean_token_accuracy": 0.12591790556907653,
      "num_tokens": 5268524.0,
      "step": 2590
    },
    {
      "entropy": 6.359928607940674,
      "epoch": 0.22677619505374463,
      "grad_norm": 1.25,
      "learning_rate": 0.000499780730076846,
      "loss": 6.3493,
      "mean_token_accuracy": 0.12371724024415016,
      "num_tokens": 5280702.0,
      "step": 2595
    },
    {
      "entropy": 6.4017528057098385,
      "epoch": 0.22721314340644935,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997793525491122,
      "loss": 6.3274,
      "mean_token_accuracy": 0.12179224193096161,
      "num_tokens": 5291214.0,
      "step": 2600
    },
    {
      "entropy": 6.364180421829223,
      "epoch": 0.22765009175915407,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004997779707099926,
      "loss": 6.2881,
      "mean_token_accuracy": 0.1345026545226574,
      "num_tokens": 5300729.0,
      "step": 2605
    },
    {
      "entropy": 6.363022375106811,
      "epoch": 0.2280870401118588,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004997765845595138,
      "loss": 6.441,
      "mean_token_accuracy": 0.12474732324481011,
      "num_tokens": 5310551.0,
      "step": 2610
    },
    {
      "entropy": 6.430810880661011,
      "epoch": 0.22852398846456348,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997751940977024,
      "loss": 6.3122,
      "mean_token_accuracy": 0.13267467617988588,
      "num_tokens": 5320038.0,
      "step": 2615
    },
    {
      "entropy": 6.3052386283874515,
      "epoch": 0.2289609368172682,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997737993245848,
      "loss": 6.3353,
      "mean_token_accuracy": 0.129467436671257,
      "num_tokens": 5329753.0,
      "step": 2620
    },
    {
      "entropy": 6.376943922042846,
      "epoch": 0.22939788516997292,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004997724002401881,
      "loss": 6.3323,
      "mean_token_accuracy": 0.1261540874838829,
      "num_tokens": 5340546.0,
      "step": 2625
    },
    {
      "entropy": 6.386794996261597,
      "epoch": 0.2298348335226776,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004997709968445391,
      "loss": 6.4117,
      "mean_token_accuracy": 0.12256940975785255,
      "num_tokens": 5352097.0,
      "step": 2630
    },
    {
      "entropy": 6.436126089096069,
      "epoch": 0.23027178187538233,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004997695891376644,
      "loss": 6.3473,
      "mean_token_accuracy": 0.12921667471528053,
      "num_tokens": 5361646.0,
      "step": 2635
    },
    {
      "entropy": 6.376198720932007,
      "epoch": 0.23070873022808705,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004997681771195912,
      "loss": 6.3273,
      "mean_token_accuracy": 0.12919537127017974,
      "num_tokens": 5371695.0,
      "step": 2640
    },
    {
      "entropy": 6.378803825378418,
      "epoch": 0.23114567858079174,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004997667607903466,
      "loss": 6.3975,
      "mean_token_accuracy": 0.12181895822286606,
      "num_tokens": 5381978.0,
      "step": 2645
    },
    {
      "entropy": 6.467701435089111,
      "epoch": 0.23158262693349646,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004997653401499578,
      "loss": 6.4007,
      "mean_token_accuracy": 0.12531357929110526,
      "num_tokens": 5391891.0,
      "step": 2650
    },
    {
      "entropy": 6.402388381958008,
      "epoch": 0.23201957528620118,
      "grad_norm": 1.28125,
      "learning_rate": 0.000499763915198452,
      "loss": 6.3555,
      "mean_token_accuracy": 0.12459794357419014,
      "num_tokens": 5403172.0,
      "step": 2655
    },
    {
      "entropy": 6.378549528121948,
      "epoch": 0.23245652363890587,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004997624859358565,
      "loss": 6.3646,
      "mean_token_accuracy": 0.1256065085530281,
      "num_tokens": 5413130.0,
      "step": 2660
    },
    {
      "entropy": 6.3742975234985355,
      "epoch": 0.2328934719916106,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004997610523621988,
      "loss": 6.3917,
      "mean_token_accuracy": 0.12258691787719726,
      "num_tokens": 5423510.0,
      "step": 2665
    },
    {
      "entropy": 6.373180770874024,
      "epoch": 0.2333304203443153,
      "grad_norm": 1.375,
      "learning_rate": 0.0004997596144775063,
      "loss": 6.2241,
      "mean_token_accuracy": 0.13480209782719613,
      "num_tokens": 5433707.0,
      "step": 2670
    },
    {
      "entropy": 6.2156439304351805,
      "epoch": 0.23376736869702,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997581722818067,
      "loss": 6.3541,
      "mean_token_accuracy": 0.12960029989480973,
      "num_tokens": 5444381.0,
      "step": 2675
    },
    {
      "entropy": 6.525657272338867,
      "epoch": 0.23420431704972472,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004997567257751275,
      "loss": 6.2937,
      "mean_token_accuracy": 0.12946442216634751,
      "num_tokens": 5453905.0,
      "step": 2680
    },
    {
      "entropy": 6.361790466308594,
      "epoch": 0.23464126540242944,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004997552749574965,
      "loss": 6.3146,
      "mean_token_accuracy": 0.12517919018864632,
      "num_tokens": 5464839.0,
      "step": 2685
    },
    {
      "entropy": 6.367331504821777,
      "epoch": 0.23507821375513413,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004997538198289416,
      "loss": 6.3483,
      "mean_token_accuracy": 0.12339176088571549,
      "num_tokens": 5475459.0,
      "step": 2690
    },
    {
      "entropy": 6.353179883956909,
      "epoch": 0.23551516210783885,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004997523603894907,
      "loss": 6.3649,
      "mean_token_accuracy": 0.12877027690410614,
      "num_tokens": 5484886.0,
      "step": 2695
    },
    {
      "entropy": 6.352494192123413,
      "epoch": 0.23595211046054357,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004997508966391717,
      "loss": 6.355,
      "mean_token_accuracy": 0.1291891060769558,
      "num_tokens": 5494803.0,
      "step": 2700
    },
    {
      "entropy": 6.36141185760498,
      "epoch": 0.2363890588132483,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997494285780129,
      "loss": 6.3053,
      "mean_token_accuracy": 0.12495973929762841,
      "num_tokens": 5504997.0,
      "step": 2705
    },
    {
      "entropy": 6.327673482894897,
      "epoch": 0.23682600716595298,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997479562060421,
      "loss": 6.2937,
      "mean_token_accuracy": 0.13582547903060913,
      "num_tokens": 5514492.0,
      "step": 2710
    },
    {
      "entropy": 6.405963373184204,
      "epoch": 0.2372629555186577,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004997464795232877,
      "loss": 6.3269,
      "mean_token_accuracy": 0.12349909618496895,
      "num_tokens": 5524419.0,
      "step": 2715
    },
    {
      "entropy": 6.358639526367187,
      "epoch": 0.23769990387136242,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997449985297782,
      "loss": 6.3633,
      "mean_token_accuracy": 0.12754026502370835,
      "num_tokens": 5534289.0,
      "step": 2720
    },
    {
      "entropy": 6.355934572219849,
      "epoch": 0.2381368522240671,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004997435132255418,
      "loss": 6.2981,
      "mean_token_accuracy": 0.12902715429663658,
      "num_tokens": 5543034.0,
      "step": 2725
    },
    {
      "entropy": 6.414120054244995,
      "epoch": 0.23857380057677183,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004997420236106071,
      "loss": 6.3655,
      "mean_token_accuracy": 0.12629004567861557,
      "num_tokens": 5553291.0,
      "step": 2730
    },
    {
      "entropy": 6.336919546127319,
      "epoch": 0.23901074892947655,
      "grad_norm": 1.375,
      "learning_rate": 0.0004997405296850026,
      "loss": 6.3536,
      "mean_token_accuracy": 0.13027266338467597,
      "num_tokens": 5563772.0,
      "step": 2735
    },
    {
      "entropy": 6.426336717605591,
      "epoch": 0.23944769728218124,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004997390314487569,
      "loss": 6.3971,
      "mean_token_accuracy": 0.12313945814967156,
      "num_tokens": 5574348.0,
      "step": 2740
    },
    {
      "entropy": 6.372211456298828,
      "epoch": 0.23988464563488596,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997375289018989,
      "loss": 6.242,
      "mean_token_accuracy": 0.12484033331274987,
      "num_tokens": 5583491.0,
      "step": 2745
    },
    {
      "entropy": 6.295545482635498,
      "epoch": 0.24032159398759068,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004997360220444574,
      "loss": 6.3951,
      "mean_token_accuracy": 0.12214951664209366,
      "num_tokens": 5594273.0,
      "step": 2750
    },
    {
      "entropy": 6.42808051109314,
      "epoch": 0.24075854234029537,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004997345108764611,
      "loss": 6.2957,
      "mean_token_accuracy": 0.12530528381466866,
      "num_tokens": 5604529.0,
      "step": 2755
    },
    {
      "entropy": 6.33855619430542,
      "epoch": 0.2411954906930001,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004997329953979391,
      "loss": 6.2975,
      "mean_token_accuracy": 0.13347954750061036,
      "num_tokens": 5614036.0,
      "step": 2760
    },
    {
      "entropy": 6.406115388870239,
      "epoch": 0.2416324390457048,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004997314756089206,
      "loss": 6.3418,
      "mean_token_accuracy": 0.1278097912669182,
      "num_tokens": 5624036.0,
      "step": 2765
    },
    {
      "entropy": 6.385963106155396,
      "epoch": 0.2420693873984095,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004997299515094347,
      "loss": 6.3345,
      "mean_token_accuracy": 0.13025815337896346,
      "num_tokens": 5633987.0,
      "step": 2770
    },
    {
      "entropy": 6.340337467193604,
      "epoch": 0.24250633575111422,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004997284230995105,
      "loss": 6.2052,
      "mean_token_accuracy": 0.13974855989217758,
      "num_tokens": 5643905.0,
      "step": 2775
    },
    {
      "entropy": 6.351865005493164,
      "epoch": 0.24294328410381894,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004997268903791773,
      "loss": 6.4087,
      "mean_token_accuracy": 0.12293170467019081,
      "num_tokens": 5654922.0,
      "step": 2780
    },
    {
      "entropy": 6.3397914409637455,
      "epoch": 0.24338023245652363,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004997253533484647,
      "loss": 6.2413,
      "mean_token_accuracy": 0.13280420899391174,
      "num_tokens": 5664407.0,
      "step": 2785
    },
    {
      "entropy": 6.353033638000488,
      "epoch": 0.24381718080922835,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004997238120074021,
      "loss": 6.3079,
      "mean_token_accuracy": 0.13058429807424546,
      "num_tokens": 5674524.0,
      "step": 2790
    },
    {
      "entropy": 6.383109712600708,
      "epoch": 0.24425412916193306,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004997222663560188,
      "loss": 6.3565,
      "mean_token_accuracy": 0.12487837448716163,
      "num_tokens": 5684691.0,
      "step": 2795
    },
    {
      "entropy": 6.3418395042419435,
      "epoch": 0.24469107751463778,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004997207163943449,
      "loss": 6.2672,
      "mean_token_accuracy": 0.12925235480070113,
      "num_tokens": 5693683.0,
      "step": 2800
    },
    {
      "entropy": 6.346609592437744,
      "epoch": 0.24512802586734248,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004997191621224098,
      "loss": 6.3505,
      "mean_token_accuracy": 0.12702021300792693,
      "num_tokens": 5704394.0,
      "step": 2805
    },
    {
      "entropy": 6.343639373779297,
      "epoch": 0.2455649742200472,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004997176035402435,
      "loss": 6.2929,
      "mean_token_accuracy": 0.13152689412236213,
      "num_tokens": 5715480.0,
      "step": 2810
    },
    {
      "entropy": 6.425011825561524,
      "epoch": 0.2460019225727519,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004997160406478758,
      "loss": 6.2805,
      "mean_token_accuracy": 0.13279203623533248,
      "num_tokens": 5725973.0,
      "step": 2815
    },
    {
      "entropy": 6.252740144729614,
      "epoch": 0.2464388709254566,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004997144734453367,
      "loss": 6.2817,
      "mean_token_accuracy": 0.13347256407141686,
      "num_tokens": 5735798.0,
      "step": 2820
    },
    {
      "entropy": 6.413337850570679,
      "epoch": 0.24687581927816132,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004997129019326562,
      "loss": 6.3644,
      "mean_token_accuracy": 0.12277695909142494,
      "num_tokens": 5746623.0,
      "step": 2825
    },
    {
      "entropy": 6.432828378677368,
      "epoch": 0.24731276763086604,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004997113261098645,
      "loss": 6.3132,
      "mean_token_accuracy": 0.1289552219212055,
      "num_tokens": 5756526.0,
      "step": 2830
    },
    {
      "entropy": 6.36118860244751,
      "epoch": 0.24774971598357073,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004997097459769919,
      "loss": 6.3125,
      "mean_token_accuracy": 0.13007271513342858,
      "num_tokens": 5766670.0,
      "step": 2835
    },
    {
      "entropy": 6.405504179000855,
      "epoch": 0.24818666433627545,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004997081615340685,
      "loss": 6.3703,
      "mean_token_accuracy": 0.12305036559700966,
      "num_tokens": 5776473.0,
      "step": 2840
    },
    {
      "entropy": 6.457984447479248,
      "epoch": 0.24862361268898017,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004997065727811249,
      "loss": 6.3868,
      "mean_token_accuracy": 0.12714115232229234,
      "num_tokens": 5787059.0,
      "step": 2845
    },
    {
      "entropy": 6.364512252807617,
      "epoch": 0.24906056104168486,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004997049797181915,
      "loss": 6.3791,
      "mean_token_accuracy": 0.1261816143989563,
      "num_tokens": 5797367.0,
      "step": 2850
    },
    {
      "entropy": 6.3602512836456295,
      "epoch": 0.24949750939438958,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004997033823452988,
      "loss": 6.2839,
      "mean_token_accuracy": 0.12893701419234277,
      "num_tokens": 5806704.0,
      "step": 2855
    },
    {
      "entropy": 6.357887887954712,
      "epoch": 0.2499344577470943,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004997017806624774,
      "loss": 6.3511,
      "mean_token_accuracy": 0.13046250268816947,
      "num_tokens": 5817797.0,
      "step": 2860
    },
    {
      "entropy": 6.452028465270996,
      "epoch": 0.250371406099799,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004997001746697582,
      "loss": 6.378,
      "mean_token_accuracy": 0.1331723690032959,
      "num_tokens": 5829270.0,
      "step": 2865
    },
    {
      "entropy": 6.366587162017822,
      "epoch": 0.2508083544525037,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004996985643671719,
      "loss": 6.3137,
      "mean_token_accuracy": 0.1306256264448166,
      "num_tokens": 5839214.0,
      "step": 2870
    },
    {
      "entropy": 6.356428956985473,
      "epoch": 0.2512453028052084,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004996969497547495,
      "loss": 6.2981,
      "mean_token_accuracy": 0.1351346492767334,
      "num_tokens": 5848171.0,
      "step": 2875
    },
    {
      "entropy": 6.351561784744263,
      "epoch": 0.25168225115791315,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004996953308325217,
      "loss": 6.2169,
      "mean_token_accuracy": 0.13406590819358827,
      "num_tokens": 5858399.0,
      "step": 2880
    },
    {
      "entropy": 6.331246757507325,
      "epoch": 0.25211919951061784,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004996937076005198,
      "loss": 6.2302,
      "mean_token_accuracy": 0.13163061663508416,
      "num_tokens": 5868212.0,
      "step": 2885
    },
    {
      "entropy": 6.3308275699615475,
      "epoch": 0.25255614786332253,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004996920800587749,
      "loss": 6.2959,
      "mean_token_accuracy": 0.13326340839266776,
      "num_tokens": 5877883.0,
      "step": 2890
    },
    {
      "entropy": 6.429145336151123,
      "epoch": 0.2529930962160273,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004996904482073181,
      "loss": 6.3208,
      "mean_token_accuracy": 0.12892481535673142,
      "num_tokens": 5887090.0,
      "step": 2895
    },
    {
      "entropy": 6.335492897033691,
      "epoch": 0.25343004456873197,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004996888120461808,
      "loss": 6.3525,
      "mean_token_accuracy": 0.12223572880029679,
      "num_tokens": 5897491.0,
      "step": 2900
    },
    {
      "entropy": 6.407157039642334,
      "epoch": 0.25386699292143666,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004996871715753943,
      "loss": 6.3709,
      "mean_token_accuracy": 0.12866056635975837,
      "num_tokens": 5908494.0,
      "step": 2905
    },
    {
      "entropy": 6.483530569076538,
      "epoch": 0.2543039412741414,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00049968552679499,
      "loss": 6.4113,
      "mean_token_accuracy": 0.11444444879889488,
      "num_tokens": 5919708.0,
      "step": 2910
    },
    {
      "entropy": 6.394390058517456,
      "epoch": 0.2547408896268461,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004996838777049998,
      "loss": 6.2742,
      "mean_token_accuracy": 0.1324307791888714,
      "num_tokens": 5930122.0,
      "step": 2915
    },
    {
      "entropy": 6.266551733016968,
      "epoch": 0.2551778379795508,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000499682224305455,
      "loss": 6.2714,
      "mean_token_accuracy": 0.1286554180085659,
      "num_tokens": 5940033.0,
      "step": 2920
    },
    {
      "entropy": 6.341103506088257,
      "epoch": 0.25561478633225554,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004996805665963874,
      "loss": 6.2955,
      "mean_token_accuracy": 0.12507626488804818,
      "num_tokens": 5949335.0,
      "step": 2925
    },
    {
      "entropy": 6.422477149963379,
      "epoch": 0.25605173468496023,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004996789045778287,
      "loss": 6.3136,
      "mean_token_accuracy": 0.13033667132258414,
      "num_tokens": 5959299.0,
      "step": 2930
    },
    {
      "entropy": 6.385528469085694,
      "epoch": 0.2564886830376649,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004996772382498111,
      "loss": 6.2879,
      "mean_token_accuracy": 0.12810725048184396,
      "num_tokens": 5970114.0,
      "step": 2935
    },
    {
      "entropy": 6.352691411972046,
      "epoch": 0.25692563139036967,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996755676123662,
      "loss": 6.3671,
      "mean_token_accuracy": 0.12160494327545165,
      "num_tokens": 5979630.0,
      "step": 2940
    },
    {
      "entropy": 6.448770380020141,
      "epoch": 0.25736257974307436,
      "grad_norm": 1.625,
      "learning_rate": 0.0004996738926655262,
      "loss": 6.3475,
      "mean_token_accuracy": 0.12364518791437148,
      "num_tokens": 5990190.0,
      "step": 2945
    },
    {
      "entropy": 6.390782070159912,
      "epoch": 0.25779952809577905,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004996722134093231,
      "loss": 6.3783,
      "mean_token_accuracy": 0.12294622734189034,
      "num_tokens": 6000706.0,
      "step": 2950
    },
    {
      "entropy": 6.32075514793396,
      "epoch": 0.2582364764484838,
      "grad_norm": 1.375,
      "learning_rate": 0.0004996705298437894,
      "loss": 6.3407,
      "mean_token_accuracy": 0.1283108413219452,
      "num_tokens": 6011274.0,
      "step": 2955
    },
    {
      "entropy": 6.437451887130737,
      "epoch": 0.2586734248011885,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004996688419689572,
      "loss": 6.2534,
      "mean_token_accuracy": 0.13058432415127755,
      "num_tokens": 6021481.0,
      "step": 2960
    },
    {
      "entropy": 6.3496260166168215,
      "epoch": 0.25911037315389324,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004996671497848589,
      "loss": 6.3367,
      "mean_token_accuracy": 0.12986429557204246,
      "num_tokens": 6032091.0,
      "step": 2965
    },
    {
      "entropy": 6.342704248428345,
      "epoch": 0.25954732150659793,
      "grad_norm": 1.5,
      "learning_rate": 0.0004996654532915269,
      "loss": 6.3442,
      "mean_token_accuracy": 0.12034625858068466,
      "num_tokens": 6043371.0,
      "step": 2970
    },
    {
      "entropy": 6.40626015663147,
      "epoch": 0.2599842698593026,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004996637524889937,
      "loss": 6.2854,
      "mean_token_accuracy": 0.1287955202162266,
      "num_tokens": 6052812.0,
      "step": 2975
    },
    {
      "entropy": 6.317795324325561,
      "epoch": 0.26042121821200737,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004996620473772921,
      "loss": 6.2722,
      "mean_token_accuracy": 0.13428313955664634,
      "num_tokens": 6064033.0,
      "step": 2980
    },
    {
      "entropy": 6.403295278549194,
      "epoch": 0.26085816656471206,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004996603379564546,
      "loss": 6.4065,
      "mean_token_accuracy": 0.12221458777785302,
      "num_tokens": 6075837.0,
      "step": 2985
    },
    {
      "entropy": 6.480539703369141,
      "epoch": 0.26129511491741675,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004996586242265142,
      "loss": 6.3028,
      "mean_token_accuracy": 0.13012727349996567,
      "num_tokens": 6085514.0,
      "step": 2990
    },
    {
      "entropy": 6.413984298706055,
      "epoch": 0.2617320632701215,
      "grad_norm": 1.75,
      "learning_rate": 0.0004996569061875036,
      "loss": 6.2375,
      "mean_token_accuracy": 0.1320499710738659,
      "num_tokens": 6094992.0,
      "step": 2995
    },
    {
      "entropy": 6.2930519580841064,
      "epoch": 0.2621690116228262,
      "grad_norm": 1.625,
      "learning_rate": 0.0004996551838394559,
      "loss": 6.2941,
      "mean_token_accuracy": 0.12757463231682778,
      "num_tokens": 6104841.0,
      "step": 3000
    },
    {
      "epoch": 0.2621690116228262,
      "eval_entropy": 6.144934864984054,
      "eval_loss": 6.327561855316162,
      "eval_mean_token_accuracy": 0.1354727728229144,
      "eval_num_tokens": 6104841.0,
      "eval_runtime": 27.9402,
      "eval_samples_per_second": 1324.76,
      "eval_steps_per_second": 165.604,
      "step": 3000
    },
    {
      "entropy": 6.289577436447144,
      "epoch": 0.2626059599755309,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000499653457182404,
      "loss": 6.1846,
      "mean_token_accuracy": 0.13535229861736298,
      "num_tokens": 6114830.0,
      "step": 3005
    },
    {
      "entropy": 6.366774415969848,
      "epoch": 0.2630429083282356,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004996517262163811,
      "loss": 6.3078,
      "mean_token_accuracy": 0.13021773472428322,
      "num_tokens": 6125165.0,
      "step": 3010
    },
    {
      "entropy": 6.333688974380493,
      "epoch": 0.2634798566809403,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004996499909414204,
      "loss": 6.2553,
      "mean_token_accuracy": 0.1315462201833725,
      "num_tokens": 6135670.0,
      "step": 3015
    },
    {
      "entropy": 6.301775789260864,
      "epoch": 0.263916805033645,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004996482513575551,
      "loss": 6.2316,
      "mean_token_accuracy": 0.1403089925646782,
      "num_tokens": 6146055.0,
      "step": 3020
    },
    {
      "entropy": 6.33044810295105,
      "epoch": 0.26435375338634975,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004996465074648186,
      "loss": 6.2685,
      "mean_token_accuracy": 0.1325012370944023,
      "num_tokens": 6155941.0,
      "step": 3025
    },
    {
      "entropy": 6.416541910171508,
      "epoch": 0.26479070173905445,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004996447592632444,
      "loss": 6.3498,
      "mean_token_accuracy": 0.1273452140390873,
      "num_tokens": 6166291.0,
      "step": 3030
    },
    {
      "entropy": 6.322501277923584,
      "epoch": 0.26522765009175914,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004996430067528661,
      "loss": 6.3076,
      "mean_token_accuracy": 0.12802767083048822,
      "num_tokens": 6175294.0,
      "step": 3035
    },
    {
      "entropy": 6.430583572387695,
      "epoch": 0.2656645984444639,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004996412499337172,
      "loss": 6.2871,
      "mean_token_accuracy": 0.1301487609744072,
      "num_tokens": 6185273.0,
      "step": 3040
    },
    {
      "entropy": 6.367474174499511,
      "epoch": 0.2661015467971686,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004996394888058314,
      "loss": 6.2536,
      "mean_token_accuracy": 0.14053120017051696,
      "num_tokens": 6195075.0,
      "step": 3045
    },
    {
      "entropy": 6.221368312835693,
      "epoch": 0.26653849514987327,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004996377233692425,
      "loss": 6.1577,
      "mean_token_accuracy": 0.1434023678302765,
      "num_tokens": 6206018.0,
      "step": 3050
    },
    {
      "entropy": 6.334922933578492,
      "epoch": 0.266975443502578,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004996359536239843,
      "loss": 6.2924,
      "mean_token_accuracy": 0.1288744553923607,
      "num_tokens": 6216577.0,
      "step": 3055
    },
    {
      "entropy": 6.441603183746338,
      "epoch": 0.2674123918552827,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004996341795700908,
      "loss": 6.3002,
      "mean_token_accuracy": 0.12710044384002686,
      "num_tokens": 6226521.0,
      "step": 3060
    },
    {
      "entropy": 6.268930816650391,
      "epoch": 0.2678493402079874,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004996324012075961,
      "loss": 6.2947,
      "mean_token_accuracy": 0.1343860723078251,
      "num_tokens": 6236509.0,
      "step": 3065
    },
    {
      "entropy": 6.361376142501831,
      "epoch": 0.26828628856069214,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004996306185365341,
      "loss": 6.2844,
      "mean_token_accuracy": 0.13270404413342476,
      "num_tokens": 6246584.0,
      "step": 3070
    },
    {
      "entropy": 6.384239864349365,
      "epoch": 0.26872323691339683,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004996288315569393,
      "loss": 6.2926,
      "mean_token_accuracy": 0.1306929111480713,
      "num_tokens": 6256881.0,
      "step": 3075
    },
    {
      "entropy": 6.3746906280517575,
      "epoch": 0.2691601852661015,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004996270402688457,
      "loss": 6.2599,
      "mean_token_accuracy": 0.1351035661995411,
      "num_tokens": 6266380.0,
      "step": 3080
    },
    {
      "entropy": 6.383773946762085,
      "epoch": 0.2695971336188063,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004996252446722877,
      "loss": 6.314,
      "mean_token_accuracy": 0.13512232452630996,
      "num_tokens": 6274872.0,
      "step": 3085
    },
    {
      "entropy": 6.327398204803467,
      "epoch": 0.27003408197151096,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004996234447673,
      "loss": 6.3424,
      "mean_token_accuracy": 0.1278736785054207,
      "num_tokens": 6286167.0,
      "step": 3090
    },
    {
      "entropy": 6.377889776229859,
      "epoch": 0.27047103032421566,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004996216405539169,
      "loss": 6.2842,
      "mean_token_accuracy": 0.1276158906519413,
      "num_tokens": 6295898.0,
      "step": 3095
    },
    {
      "entropy": 6.3654858589172365,
      "epoch": 0.2709079786769204,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004996198320321728,
      "loss": 6.301,
      "mean_token_accuracy": 0.12989438697695732,
      "num_tokens": 6306543.0,
      "step": 3100
    },
    {
      "entropy": 6.359067678451538,
      "epoch": 0.2713449270296251,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004996180192021029,
      "loss": 6.2831,
      "mean_token_accuracy": 0.12411871701478958,
      "num_tokens": 6317709.0,
      "step": 3105
    },
    {
      "entropy": 6.350419044494629,
      "epoch": 0.2717818753823298,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004996162020637415,
      "loss": 6.2444,
      "mean_token_accuracy": 0.13357833847403527,
      "num_tokens": 6326874.0,
      "step": 3110
    },
    {
      "entropy": 6.339018774032593,
      "epoch": 0.27221882373503453,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004996143806171238,
      "loss": 6.3442,
      "mean_token_accuracy": 0.12774320989847182,
      "num_tokens": 6337943.0,
      "step": 3115
    },
    {
      "entropy": 6.367568111419677,
      "epoch": 0.2726557720877392,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004996125548622845,
      "loss": 6.4008,
      "mean_token_accuracy": 0.12949984148144722,
      "num_tokens": 6348442.0,
      "step": 3120
    },
    {
      "entropy": 6.405733299255371,
      "epoch": 0.2730927204404439,
      "grad_norm": 1.5,
      "learning_rate": 0.0004996107247992587,
      "loss": 6.2846,
      "mean_token_accuracy": 0.13143943548202514,
      "num_tokens": 6358675.0,
      "step": 3125
    },
    {
      "entropy": 6.400189399719238,
      "epoch": 0.27352966879314866,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004996088904280815,
      "loss": 6.3301,
      "mean_token_accuracy": 0.13247563317418098,
      "num_tokens": 6368631.0,
      "step": 3130
    },
    {
      "entropy": 6.404301691055298,
      "epoch": 0.27396661714585335,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004996070517487881,
      "loss": 6.2082,
      "mean_token_accuracy": 0.12956621274352073,
      "num_tokens": 6378709.0,
      "step": 3135
    },
    {
      "entropy": 6.3348571300506595,
      "epoch": 0.27440356549855804,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004996052087614138,
      "loss": 6.3512,
      "mean_token_accuracy": 0.12001562491059303,
      "num_tokens": 6389686.0,
      "step": 3140
    },
    {
      "entropy": 6.3280986785888675,
      "epoch": 0.2748405138512628,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004996033614659938,
      "loss": 6.3014,
      "mean_token_accuracy": 0.1311795823276043,
      "num_tokens": 6399447.0,
      "step": 3145
    },
    {
      "entropy": 6.403613090515137,
      "epoch": 0.2752774622039675,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004996015098625638,
      "loss": 6.3174,
      "mean_token_accuracy": 0.13122527301311493,
      "num_tokens": 6408891.0,
      "step": 3150
    },
    {
      "entropy": 6.3696483135223385,
      "epoch": 0.2757144105566722,
      "grad_norm": 2.0625,
      "learning_rate": 0.000499599653951159,
      "loss": 6.2713,
      "mean_token_accuracy": 0.13232835605740548,
      "num_tokens": 6419471.0,
      "step": 3155
    },
    {
      "entropy": 6.342033958435058,
      "epoch": 0.2761513589093769,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004995977937318153,
      "loss": 6.2585,
      "mean_token_accuracy": 0.12938153818249704,
      "num_tokens": 6430445.0,
      "step": 3160
    },
    {
      "entropy": 6.420746564865112,
      "epoch": 0.2765883072620816,
      "grad_norm": 1.59375,
      "learning_rate": 0.000499595929204568,
      "loss": 6.297,
      "mean_token_accuracy": 0.12924976646900177,
      "num_tokens": 6440821.0,
      "step": 3165
    },
    {
      "entropy": 6.337243461608887,
      "epoch": 0.27702525561478636,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004995940603694533,
      "loss": 6.2071,
      "mean_token_accuracy": 0.13440316244959832,
      "num_tokens": 6450958.0,
      "step": 3170
    },
    {
      "entropy": 6.311748075485229,
      "epoch": 0.27746220396749105,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004995921872265067,
      "loss": 6.3223,
      "mean_token_accuracy": 0.13291849121451377,
      "num_tokens": 6462106.0,
      "step": 3175
    },
    {
      "entropy": 6.371618556976318,
      "epoch": 0.27789915232019574,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004995903097757643,
      "loss": 6.1892,
      "mean_token_accuracy": 0.13572993725538254,
      "num_tokens": 6471890.0,
      "step": 3180
    },
    {
      "entropy": 6.294724369049073,
      "epoch": 0.2783361006729005,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004995884280172622,
      "loss": 6.2298,
      "mean_token_accuracy": 0.1322424106299877,
      "num_tokens": 6481959.0,
      "step": 3185
    },
    {
      "entropy": 6.308170366287231,
      "epoch": 0.2787730490256052,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004995865419510362,
      "loss": 6.2535,
      "mean_token_accuracy": 0.13232620432972908,
      "num_tokens": 6492086.0,
      "step": 3190
    },
    {
      "entropy": 6.3365153789520265,
      "epoch": 0.27920999737830987,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004995846515771227,
      "loss": 6.2954,
      "mean_token_accuracy": 0.13309518322348596,
      "num_tokens": 6502530.0,
      "step": 3195
    },
    {
      "entropy": 6.349178171157837,
      "epoch": 0.2796469457310146,
      "grad_norm": 3.609375,
      "learning_rate": 0.000499582756895558,
      "loss": 6.2884,
      "mean_token_accuracy": 0.13106891587376596,
      "num_tokens": 6511600.0,
      "step": 3200
    },
    {
      "entropy": 6.388007688522339,
      "epoch": 0.2800838940837193,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004995808579063782,
      "loss": 6.2566,
      "mean_token_accuracy": 0.125956416875124,
      "num_tokens": 6522712.0,
      "step": 3205
    },
    {
      "entropy": 6.198056030273437,
      "epoch": 0.280520842436424,
      "grad_norm": 1.5,
      "learning_rate": 0.00049957895460962,
      "loss": 6.1415,
      "mean_token_accuracy": 0.14635513052344323,
      "num_tokens": 6532472.0,
      "step": 3210
    },
    {
      "entropy": 6.329718112945557,
      "epoch": 0.28095779078912875,
      "grad_norm": 1.5,
      "learning_rate": 0.0004995770470053197,
      "loss": 6.1952,
      "mean_token_accuracy": 0.14441470950841903,
      "num_tokens": 6541778.0,
      "step": 3215
    },
    {
      "entropy": 6.376521205902099,
      "epoch": 0.28139473914183344,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004995751350935139,
      "loss": 6.1604,
      "mean_token_accuracy": 0.13274973705410958,
      "num_tokens": 6552368.0,
      "step": 3220
    },
    {
      "entropy": 6.273313760757446,
      "epoch": 0.28183168749453813,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004995732188742395,
      "loss": 6.2168,
      "mean_token_accuracy": 0.13165887892246247,
      "num_tokens": 6562593.0,
      "step": 3225
    },
    {
      "entropy": 6.357030725479126,
      "epoch": 0.2822686358472429,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004995712983475329,
      "loss": 6.289,
      "mean_token_accuracy": 0.13038937225937844,
      "num_tokens": 6572695.0,
      "step": 3230
    },
    {
      "entropy": 6.311696243286133,
      "epoch": 0.28270558419994757,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004995693735134313,
      "loss": 6.2646,
      "mean_token_accuracy": 0.13381487503647804,
      "num_tokens": 6581671.0,
      "step": 3235
    },
    {
      "entropy": 6.300328207015991,
      "epoch": 0.28314253255265226,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004995674443719713,
      "loss": 6.2492,
      "mean_token_accuracy": 0.12872557416558267,
      "num_tokens": 6590460.0,
      "step": 3240
    },
    {
      "entropy": 6.371827697753906,
      "epoch": 0.283579480905357,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00049956551092319,
      "loss": 6.2629,
      "mean_token_accuracy": 0.12990283593535423,
      "num_tokens": 6599892.0,
      "step": 3245
    },
    {
      "entropy": 6.347928094863891,
      "epoch": 0.2840164292580617,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004995635731671247,
      "loss": 6.3503,
      "mean_token_accuracy": 0.12329007610678673,
      "num_tokens": 6610530.0,
      "step": 3250
    },
    {
      "entropy": 6.301224994659424,
      "epoch": 0.2844533776107664,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004995616311038124,
      "loss": 6.2437,
      "mean_token_accuracy": 0.13091473653912544,
      "num_tokens": 6621117.0,
      "step": 3255
    },
    {
      "entropy": 6.368579530715943,
      "epoch": 0.28489032596347114,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004995596847332901,
      "loss": 6.2965,
      "mean_token_accuracy": 0.12649882435798646,
      "num_tokens": 6631521.0,
      "step": 3260
    },
    {
      "entropy": 6.446327972412109,
      "epoch": 0.2853272743161758,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004995577340555956,
      "loss": 6.3704,
      "mean_token_accuracy": 0.12478309124708176,
      "num_tokens": 6642180.0,
      "step": 3265
    },
    {
      "entropy": 6.373644304275513,
      "epoch": 0.2857642226688805,
      "grad_norm": 1.625,
      "learning_rate": 0.0004995557790707661,
      "loss": 6.307,
      "mean_token_accuracy": 0.12872309386730194,
      "num_tokens": 6652735.0,
      "step": 3270
    },
    {
      "entropy": 6.359430027008057,
      "epoch": 0.28620117102158527,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000499553819778839,
      "loss": 6.3187,
      "mean_token_accuracy": 0.13014050871133803,
      "num_tokens": 6663135.0,
      "step": 3275
    },
    {
      "entropy": 6.323476886749267,
      "epoch": 0.28663811937428996,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000499551856179852,
      "loss": 6.2859,
      "mean_token_accuracy": 0.12836155891418458,
      "num_tokens": 6673375.0,
      "step": 3280
    },
    {
      "entropy": 6.437594795227051,
      "epoch": 0.28707506772699465,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004995498882738428,
      "loss": 6.2728,
      "mean_token_accuracy": 0.13003526180982589,
      "num_tokens": 6683101.0,
      "step": 3285
    },
    {
      "entropy": 6.372012615203857,
      "epoch": 0.2875120160796994,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004995479160608491,
      "loss": 6.3349,
      "mean_token_accuracy": 0.13348986431956292,
      "num_tokens": 6692763.0,
      "step": 3290
    },
    {
      "entropy": 6.357828855514526,
      "epoch": 0.2879489644324041,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004995459395409085,
      "loss": 6.3056,
      "mean_token_accuracy": 0.12744921073317528,
      "num_tokens": 6702833.0,
      "step": 3295
    },
    {
      "entropy": 6.400131702423096,
      "epoch": 0.2883859127851088,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004995439587140593,
      "loss": 6.2848,
      "mean_token_accuracy": 0.1267162047326565,
      "num_tokens": 6713083.0,
      "step": 3300
    },
    {
      "entropy": 6.3027080535888675,
      "epoch": 0.2888228611378135,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004995419735803392,
      "loss": 6.2384,
      "mean_token_accuracy": 0.12887218371033668,
      "num_tokens": 6722501.0,
      "step": 3305
    },
    {
      "entropy": 6.312692928314209,
      "epoch": 0.2892598094905182,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004995399841397865,
      "loss": 6.2842,
      "mean_token_accuracy": 0.12585734203457832,
      "num_tokens": 6731961.0,
      "step": 3310
    },
    {
      "entropy": 6.3658750534057615,
      "epoch": 0.2896967578432229,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004995379903924392,
      "loss": 6.2646,
      "mean_token_accuracy": 0.12715218290686608,
      "num_tokens": 6742576.0,
      "step": 3315
    },
    {
      "entropy": 6.382916116714478,
      "epoch": 0.29013370619592765,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004995359923383356,
      "loss": 6.2524,
      "mean_token_accuracy": 0.1332198791205883,
      "num_tokens": 6753066.0,
      "step": 3320
    },
    {
      "entropy": 6.364381551742554,
      "epoch": 0.29057065454863235,
      "grad_norm": 1.5625,
      "learning_rate": 0.000499533989977514,
      "loss": 6.3998,
      "mean_token_accuracy": 0.12284245267510414,
      "num_tokens": 6764455.0,
      "step": 3325
    },
    {
      "entropy": 6.441966438293457,
      "epoch": 0.29100760290133704,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004995319833100128,
      "loss": 6.2923,
      "mean_token_accuracy": 0.1344316452741623,
      "num_tokens": 6774655.0,
      "step": 3330
    },
    {
      "entropy": 6.308348751068115,
      "epoch": 0.2914445512540418,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004995299723358705,
      "loss": 6.2407,
      "mean_token_accuracy": 0.13696040362119674,
      "num_tokens": 6785942.0,
      "step": 3335
    },
    {
      "entropy": 6.244423246383667,
      "epoch": 0.2918814996067465,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004995279570551257,
      "loss": 6.2162,
      "mean_token_accuracy": 0.13762748166918753,
      "num_tokens": 6795355.0,
      "step": 3340
    },
    {
      "entropy": 6.344776153564453,
      "epoch": 0.29231844795945117,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000499525937467817,
      "loss": 6.2924,
      "mean_token_accuracy": 0.12889732420444489,
      "num_tokens": 6805579.0,
      "step": 3345
    },
    {
      "entropy": 6.384791278839112,
      "epoch": 0.2927553963121559,
      "grad_norm": 1.5,
      "learning_rate": 0.0004995239135739833,
      "loss": 6.3162,
      "mean_token_accuracy": 0.12749789729714395,
      "num_tokens": 6814878.0,
      "step": 3350
    },
    {
      "entropy": 6.368048191070557,
      "epoch": 0.2931923446648606,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004995218853736631,
      "loss": 6.1847,
      "mean_token_accuracy": 0.14038095027208328,
      "num_tokens": 6824533.0,
      "step": 3355
    },
    {
      "entropy": 6.190872669219971,
      "epoch": 0.29362929301756535,
      "grad_norm": 1.625,
      "learning_rate": 0.0004995198528668956,
      "loss": 6.1205,
      "mean_token_accuracy": 0.14549067616462708,
      "num_tokens": 6835529.0,
      "step": 3360
    },
    {
      "entropy": 6.32953577041626,
      "epoch": 0.29406624137027004,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004995178160537197,
      "loss": 6.3262,
      "mean_token_accuracy": 0.12966961190104484,
      "num_tokens": 6845679.0,
      "step": 3365
    },
    {
      "entropy": 6.41126766204834,
      "epoch": 0.29450318972297473,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004995157749341744,
      "loss": 6.2922,
      "mean_token_accuracy": 0.12410857155919075,
      "num_tokens": 6855876.0,
      "step": 3370
    },
    {
      "entropy": 6.376305389404297,
      "epoch": 0.2949401380756795,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004995137295082989,
      "loss": 6.3439,
      "mean_token_accuracy": 0.12616318017244338,
      "num_tokens": 6866200.0,
      "step": 3375
    },
    {
      "entropy": 6.401220417022705,
      "epoch": 0.2953770864283842,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004995116797761324,
      "loss": 6.2191,
      "mean_token_accuracy": 0.13245865926146508,
      "num_tokens": 6875191.0,
      "step": 3380
    },
    {
      "entropy": 6.234606075286865,
      "epoch": 0.29581403478108886,
      "grad_norm": 1.625,
      "learning_rate": 0.0004995096257377143,
      "loss": 6.2016,
      "mean_token_accuracy": 0.13304308578372,
      "num_tokens": 6885352.0,
      "step": 3385
    },
    {
      "entropy": 6.220990896224976,
      "epoch": 0.2962509831337936,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004995075673930839,
      "loss": 6.2853,
      "mean_token_accuracy": 0.12391391023993492,
      "num_tokens": 6895585.0,
      "step": 3390
    },
    {
      "entropy": 6.4373351573944095,
      "epoch": 0.2966879314864983,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004995055047422808,
      "loss": 6.2569,
      "mean_token_accuracy": 0.13108095303177833,
      "num_tokens": 6905532.0,
      "step": 3395
    },
    {
      "entropy": 6.287919569015503,
      "epoch": 0.297124879839203,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004995034377853443,
      "loss": 6.2526,
      "mean_token_accuracy": 0.1321505032479763,
      "num_tokens": 6915835.0,
      "step": 3400
    },
    {
      "entropy": 6.225392484664917,
      "epoch": 0.29756182819190774,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004995013665223143,
      "loss": 6.3016,
      "mean_token_accuracy": 0.12643175646662713,
      "num_tokens": 6925964.0,
      "step": 3405
    },
    {
      "entropy": 6.40485520362854,
      "epoch": 0.29799877654461243,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004994992909532305,
      "loss": 6.246,
      "mean_token_accuracy": 0.13027336671948433,
      "num_tokens": 6936145.0,
      "step": 3410
    },
    {
      "entropy": 6.382689380645752,
      "epoch": 0.2984357248973171,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004994972110781326,
      "loss": 6.2152,
      "mean_token_accuracy": 0.1278024733066559,
      "num_tokens": 6945557.0,
      "step": 3415
    },
    {
      "entropy": 6.2249171257019045,
      "epoch": 0.29887267325002187,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004994951268970606,
      "loss": 6.261,
      "mean_token_accuracy": 0.1286550961434841,
      "num_tokens": 6955147.0,
      "step": 3420
    },
    {
      "entropy": 6.312950229644775,
      "epoch": 0.29930962160272656,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004994930384100545,
      "loss": 6.2045,
      "mean_token_accuracy": 0.1307476803660393,
      "num_tokens": 6964024.0,
      "step": 3425
    },
    {
      "entropy": 6.370791673660278,
      "epoch": 0.29974656995543125,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004994909456171541,
      "loss": 6.2855,
      "mean_token_accuracy": 0.129286939650774,
      "num_tokens": 6974399.0,
      "step": 3430
    },
    {
      "entropy": 6.292686939239502,
      "epoch": 0.300183518308136,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004994888485183998,
      "loss": 6.2188,
      "mean_token_accuracy": 0.13459003791213037,
      "num_tokens": 6984324.0,
      "step": 3435
    },
    {
      "entropy": 6.31501612663269,
      "epoch": 0.3006204666608407,
      "grad_norm": 2.046875,
      "learning_rate": 0.000499486747113832,
      "loss": 6.2332,
      "mean_token_accuracy": 0.13288868591189384,
      "num_tokens": 6995659.0,
      "step": 3440
    },
    {
      "entropy": 6.416201114654541,
      "epoch": 0.3010574150135454,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004994846414034905,
      "loss": 6.3531,
      "mean_token_accuracy": 0.1243766687810421,
      "num_tokens": 7006463.0,
      "step": 3445
    },
    {
      "entropy": 6.35507755279541,
      "epoch": 0.30149436336625013,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000499482531387416,
      "loss": 6.2482,
      "mean_token_accuracy": 0.13327691406011583,
      "num_tokens": 7016188.0,
      "step": 3450
    },
    {
      "entropy": 6.353025627136231,
      "epoch": 0.3019313117189548,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004994804170656489,
      "loss": 6.2932,
      "mean_token_accuracy": 0.1287926346063614,
      "num_tokens": 7028435.0,
      "step": 3455
    },
    {
      "entropy": 6.418145370483399,
      "epoch": 0.3023682600716595,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004994782984382299,
      "loss": 6.2713,
      "mean_token_accuracy": 0.1334073044359684,
      "num_tokens": 7037992.0,
      "step": 3460
    },
    {
      "entropy": 6.302688789367676,
      "epoch": 0.30280520842436426,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004994761755051995,
      "loss": 6.2772,
      "mean_token_accuracy": 0.1256247043609619,
      "num_tokens": 7048743.0,
      "step": 3465
    },
    {
      "entropy": 6.353930807113647,
      "epoch": 0.30324215677706895,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004994740482665982,
      "loss": 6.2428,
      "mean_token_accuracy": 0.13198975548148156,
      "num_tokens": 7059524.0,
      "step": 3470
    },
    {
      "entropy": 6.3610532760620115,
      "epoch": 0.30367910512977364,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004994719167224672,
      "loss": 6.2449,
      "mean_token_accuracy": 0.12525714486837386,
      "num_tokens": 7069742.0,
      "step": 3475
    },
    {
      "entropy": 6.25976300239563,
      "epoch": 0.3041160534824784,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004994697808728472,
      "loss": 6.3001,
      "mean_token_accuracy": 0.1322442226111889,
      "num_tokens": 7079909.0,
      "step": 3480
    },
    {
      "entropy": 6.398858690261841,
      "epoch": 0.3045530018351831,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004994676407177791,
      "loss": 6.33,
      "mean_token_accuracy": 0.12571098953485488,
      "num_tokens": 7089136.0,
      "step": 3485
    },
    {
      "entropy": 6.40047402381897,
      "epoch": 0.30498995018788777,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004994654962573043,
      "loss": 6.3137,
      "mean_token_accuracy": 0.12729762643575668,
      "num_tokens": 7099621.0,
      "step": 3490
    },
    {
      "entropy": 6.3163258075714115,
      "epoch": 0.3054268985405925,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004994633474914634,
      "loss": 6.1798,
      "mean_token_accuracy": 0.13531483858823776,
      "num_tokens": 7109599.0,
      "step": 3495
    },
    {
      "entropy": 6.30221791267395,
      "epoch": 0.3058638468932972,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004994611944202979,
      "loss": 6.2133,
      "mean_token_accuracy": 0.13482960239052771,
      "num_tokens": 7120733.0,
      "step": 3500
    },
    {
      "entropy": 6.414334535598755,
      "epoch": 0.3063007952460019,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004994590370438492,
      "loss": 6.2994,
      "mean_token_accuracy": 0.1258564293384552,
      "num_tokens": 7129871.0,
      "step": 3505
    },
    {
      "entropy": 6.349997615814209,
      "epoch": 0.30673774359870665,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004994568753621585,
      "loss": 6.2083,
      "mean_token_accuracy": 0.1295277401804924,
      "num_tokens": 7139167.0,
      "step": 3510
    },
    {
      "entropy": 6.272922849655151,
      "epoch": 0.30717469195141134,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004994547093752673,
      "loss": 6.3118,
      "mean_token_accuracy": 0.12376317158341407,
      "num_tokens": 7150534.0,
      "step": 3515
    },
    {
      "entropy": 6.316261863708496,
      "epoch": 0.30761164030411603,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004994525390832171,
      "loss": 6.2146,
      "mean_token_accuracy": 0.12593079432845117,
      "num_tokens": 7160472.0,
      "step": 3520
    },
    {
      "entropy": 6.373012065887451,
      "epoch": 0.3080485886568208,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004994503644860497,
      "loss": 6.2688,
      "mean_token_accuracy": 0.13120559900999068,
      "num_tokens": 7171636.0,
      "step": 3525
    },
    {
      "entropy": 6.322237157821656,
      "epoch": 0.30848553700952547,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004994481855838067,
      "loss": 6.2184,
      "mean_token_accuracy": 0.12979013845324516,
      "num_tokens": 7182163.0,
      "step": 3530
    },
    {
      "entropy": 6.2787556648254395,
      "epoch": 0.30892248536223016,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004994460023765298,
      "loss": 6.2565,
      "mean_token_accuracy": 0.13765134364366532,
      "num_tokens": 7191694.0,
      "step": 3535
    },
    {
      "entropy": 6.318243455886841,
      "epoch": 0.3093594337149349,
      "grad_norm": 1.328125,
      "learning_rate": 0.000499443814864261,
      "loss": 6.277,
      "mean_token_accuracy": 0.13471179008483886,
      "num_tokens": 7202813.0,
      "step": 3540
    },
    {
      "entropy": 6.37991418838501,
      "epoch": 0.3097963820676396,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004994416230470422,
      "loss": 6.2705,
      "mean_token_accuracy": 0.1303069233894348,
      "num_tokens": 7213226.0,
      "step": 3545
    },
    {
      "entropy": 6.387282371520996,
      "epoch": 0.3102333304203443,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004994394269249155,
      "loss": 6.2918,
      "mean_token_accuracy": 0.12414048239588737,
      "num_tokens": 7223520.0,
      "step": 3550
    },
    {
      "entropy": 6.270621490478516,
      "epoch": 0.31067027877304904,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004994372264979231,
      "loss": 6.2038,
      "mean_token_accuracy": 0.13948841840028764,
      "num_tokens": 7232651.0,
      "step": 3555
    },
    {
      "entropy": 6.371720647811889,
      "epoch": 0.3111072271257537,
      "grad_norm": 1.5,
      "learning_rate": 0.0004994350217661069,
      "loss": 6.2742,
      "mean_token_accuracy": 0.12946462705731393,
      "num_tokens": 7242064.0,
      "step": 3560
    },
    {
      "entropy": 6.361967611312866,
      "epoch": 0.3115441754784585,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004994328127295095,
      "loss": 6.2906,
      "mean_token_accuracy": 0.1276524022221565,
      "num_tokens": 7253048.0,
      "step": 3565
    },
    {
      "entropy": 6.290471410751342,
      "epoch": 0.31198112383116317,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004994305993881731,
      "loss": 6.2542,
      "mean_token_accuracy": 0.13387253284454345,
      "num_tokens": 7262516.0,
      "step": 3570
    },
    {
      "entropy": 6.304445266723633,
      "epoch": 0.31241807218386786,
      "grad_norm": 2.125,
      "learning_rate": 0.0004994283817421402,
      "loss": 6.1703,
      "mean_token_accuracy": 0.13668953701853753,
      "num_tokens": 7272079.0,
      "step": 3575
    },
    {
      "entropy": 6.320972061157226,
      "epoch": 0.3128550205365726,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004994261597914534,
      "loss": 6.1808,
      "mean_token_accuracy": 0.1378669910132885,
      "num_tokens": 7282150.0,
      "step": 3580
    },
    {
      "entropy": 6.280206489562988,
      "epoch": 0.3132919688892773,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004994239335361552,
      "loss": 6.2746,
      "mean_token_accuracy": 0.1271176241338253,
      "num_tokens": 7291782.0,
      "step": 3585
    },
    {
      "entropy": 6.338173151016235,
      "epoch": 0.313728917241982,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004994217029762884,
      "loss": 6.2368,
      "mean_token_accuracy": 0.13136749416589738,
      "num_tokens": 7301472.0,
      "step": 3590
    },
    {
      "entropy": 6.331088447570801,
      "epoch": 0.31416586559468673,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004994194681118957,
      "loss": 6.207,
      "mean_token_accuracy": 0.1352943629026413,
      "num_tokens": 7311456.0,
      "step": 3595
    },
    {
      "entropy": 6.34141173362732,
      "epoch": 0.3146028139473914,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00049941722894302,
      "loss": 6.2813,
      "mean_token_accuracy": 0.12667939737439154,
      "num_tokens": 7321106.0,
      "step": 3600
    },
    {
      "entropy": 6.367776727676391,
      "epoch": 0.3150397623000961,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004994149854697042,
      "loss": 6.2758,
      "mean_token_accuracy": 0.1336451880633831,
      "num_tokens": 7331232.0,
      "step": 3605
    },
    {
      "entropy": 6.341811943054199,
      "epoch": 0.31547671065280086,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004994127376919916,
      "loss": 6.2605,
      "mean_token_accuracy": 0.1278167925775051,
      "num_tokens": 7341560.0,
      "step": 3610
    },
    {
      "entropy": 6.405181121826172,
      "epoch": 0.31591365900550555,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004994104856099249,
      "loss": 6.2619,
      "mean_token_accuracy": 0.13169389516115187,
      "num_tokens": 7350923.0,
      "step": 3615
    },
    {
      "entropy": 6.224088668823242,
      "epoch": 0.31635060735821025,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004994082292235476,
      "loss": 6.1756,
      "mean_token_accuracy": 0.14111646711826326,
      "num_tokens": 7360121.0,
      "step": 3620
    },
    {
      "entropy": 6.254033422470092,
      "epoch": 0.316787555710915,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004994059685329027,
      "loss": 6.2583,
      "mean_token_accuracy": 0.13028259351849555,
      "num_tokens": 7370823.0,
      "step": 3625
    },
    {
      "entropy": 6.385916757583618,
      "epoch": 0.3172245040636197,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000499403703538034,
      "loss": 6.2119,
      "mean_token_accuracy": 0.13265296518802644,
      "num_tokens": 7380102.0,
      "step": 3630
    },
    {
      "entropy": 6.30387601852417,
      "epoch": 0.3176614524163244,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004994014342389845,
      "loss": 6.2108,
      "mean_token_accuracy": 0.13178636208176614,
      "num_tokens": 7389909.0,
      "step": 3635
    },
    {
      "entropy": 6.3403831958770756,
      "epoch": 0.3180984007690291,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004993991606357979,
      "loss": 6.295,
      "mean_token_accuracy": 0.1265092261135578,
      "num_tokens": 7401092.0,
      "step": 3640
    },
    {
      "entropy": 6.286354827880859,
      "epoch": 0.3185353491217338,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004993968827285179,
      "loss": 6.1916,
      "mean_token_accuracy": 0.12839968428015708,
      "num_tokens": 7412155.0,
      "step": 3645
    },
    {
      "entropy": 6.294556522369385,
      "epoch": 0.3189722974744385,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004993946005171881,
      "loss": 6.1561,
      "mean_token_accuracy": 0.14158833622932435,
      "num_tokens": 7421981.0,
      "step": 3650
    },
    {
      "entropy": 6.278389358520508,
      "epoch": 0.31940924582714325,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004993923140018523,
      "loss": 6.1845,
      "mean_token_accuracy": 0.13419949114322663,
      "num_tokens": 7432298.0,
      "step": 3655
    },
    {
      "entropy": 6.25642614364624,
      "epoch": 0.31984619417984794,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004993900231825542,
      "loss": 6.2386,
      "mean_token_accuracy": 0.13513367772102355,
      "num_tokens": 7442575.0,
      "step": 3660
    },
    {
      "entropy": 6.27722692489624,
      "epoch": 0.32028314253255263,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004993877280593381,
      "loss": 6.2116,
      "mean_token_accuracy": 0.13987505286931992,
      "num_tokens": 7453072.0,
      "step": 3665
    },
    {
      "entropy": 6.380948162078857,
      "epoch": 0.3207200908852574,
      "grad_norm": 1.75,
      "learning_rate": 0.0004993854286322477,
      "loss": 6.1814,
      "mean_token_accuracy": 0.13585382252931594,
      "num_tokens": 7463173.0,
      "step": 3670
    },
    {
      "entropy": 6.2719017505645756,
      "epoch": 0.32115703923796207,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004993831249013272,
      "loss": 6.2875,
      "mean_token_accuracy": 0.1299177125096321,
      "num_tokens": 7473751.0,
      "step": 3675
    },
    {
      "entropy": 6.251084613800049,
      "epoch": 0.32159398759066676,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004993808168666207,
      "loss": 6.0862,
      "mean_token_accuracy": 0.14155191406607628,
      "num_tokens": 7483233.0,
      "step": 3680
    },
    {
      "entropy": 6.301878023147583,
      "epoch": 0.3220309359433715,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004993785045281726,
      "loss": 6.2041,
      "mean_token_accuracy": 0.12450918033719063,
      "num_tokens": 7493152.0,
      "step": 3685
    },
    {
      "entropy": 6.313766288757324,
      "epoch": 0.3224678842960762,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004993761878860272,
      "loss": 6.2031,
      "mean_token_accuracy": 0.13442298620939255,
      "num_tokens": 7502463.0,
      "step": 3690
    },
    {
      "entropy": 6.39133095741272,
      "epoch": 0.3229048326487809,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000499373866940229,
      "loss": 6.239,
      "mean_token_accuracy": 0.13316251635551452,
      "num_tokens": 7512301.0,
      "step": 3695
    },
    {
      "entropy": 6.224301195144653,
      "epoch": 0.32334178100148564,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004993715416908225,
      "loss": 6.2931,
      "mean_token_accuracy": 0.13126731812953948,
      "num_tokens": 7522591.0,
      "step": 3700
    },
    {
      "entropy": 6.3102202892303465,
      "epoch": 0.32377872935419033,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004993692121378521,
      "loss": 6.2684,
      "mean_token_accuracy": 0.13121233657002448,
      "num_tokens": 7532400.0,
      "step": 3705
    },
    {
      "entropy": 6.348808860778808,
      "epoch": 0.324215677706895,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004993668782813629,
      "loss": 6.1861,
      "mean_token_accuracy": 0.12972399443387986,
      "num_tokens": 7542163.0,
      "step": 3710
    },
    {
      "entropy": 6.283147525787354,
      "epoch": 0.32465262605959977,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004993645401213992,
      "loss": 6.1776,
      "mean_token_accuracy": 0.13527768924832345,
      "num_tokens": 7551638.0,
      "step": 3715
    },
    {
      "entropy": 6.33964433670044,
      "epoch": 0.32508957441230446,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004993621976580061,
      "loss": 6.2465,
      "mean_token_accuracy": 0.13232119008898735,
      "num_tokens": 7561230.0,
      "step": 3720
    },
    {
      "entropy": 6.32242784500122,
      "epoch": 0.32552652276500915,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004993598508912285,
      "loss": 6.1333,
      "mean_token_accuracy": 0.140650524944067,
      "num_tokens": 7570962.0,
      "step": 3725
    },
    {
      "entropy": 6.260022592544556,
      "epoch": 0.3259634711177139,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004993574998211114,
      "loss": 6.2948,
      "mean_token_accuracy": 0.1350044935941696,
      "num_tokens": 7581454.0,
      "step": 3730
    },
    {
      "entropy": 6.327285718917847,
      "epoch": 0.3264004194704186,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004993551444476998,
      "loss": 6.185,
      "mean_token_accuracy": 0.13983666077256202,
      "num_tokens": 7591497.0,
      "step": 3735
    },
    {
      "entropy": 6.31549973487854,
      "epoch": 0.3268373678231233,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004993527847710392,
      "loss": 6.1903,
      "mean_token_accuracy": 0.1394244320690632,
      "num_tokens": 7600719.0,
      "step": 3740
    },
    {
      "entropy": 6.272513103485108,
      "epoch": 0.32727431617582803,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004993504207911744,
      "loss": 6.2102,
      "mean_token_accuracy": 0.1353791318833828,
      "num_tokens": 7610550.0,
      "step": 3745
    },
    {
      "entropy": 6.323500490188598,
      "epoch": 0.3277112645285327,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004993480525081511,
      "loss": 6.2471,
      "mean_token_accuracy": 0.1340168222784996,
      "num_tokens": 7620472.0,
      "step": 3750
    },
    {
      "entropy": 6.342343807220459,
      "epoch": 0.3281482128812374,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004993456799220145,
      "loss": 6.1823,
      "mean_token_accuracy": 0.13188303336501123,
      "num_tokens": 7630354.0,
      "step": 3755
    },
    {
      "entropy": 6.272470760345459,
      "epoch": 0.32858516123394216,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004993433030328103,
      "loss": 6.2206,
      "mean_token_accuracy": 0.1355465531349182,
      "num_tokens": 7639689.0,
      "step": 3760
    },
    {
      "entropy": 6.333719682693482,
      "epoch": 0.32902210958664685,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004993409218405839,
      "loss": 6.256,
      "mean_token_accuracy": 0.12586364150047302,
      "num_tokens": 7649319.0,
      "step": 3765
    },
    {
      "entropy": 6.390874624252319,
      "epoch": 0.3294590579393516,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004993385363453812,
      "loss": 6.2469,
      "mean_token_accuracy": 0.12913736402988435,
      "num_tokens": 7659008.0,
      "step": 3770
    },
    {
      "entropy": 6.306790447235107,
      "epoch": 0.3298960062920563,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004993361465472477,
      "loss": 6.1865,
      "mean_token_accuracy": 0.13965204283595084,
      "num_tokens": 7669064.0,
      "step": 3775
    },
    {
      "entropy": 6.321382665634156,
      "epoch": 0.330332954644761,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004993337524462295,
      "loss": 6.2582,
      "mean_token_accuracy": 0.13454772233963014,
      "num_tokens": 7679122.0,
      "step": 3780
    },
    {
      "entropy": 6.278156852722168,
      "epoch": 0.3307699029974657,
      "grad_norm": 3.875,
      "learning_rate": 0.0004993313540423722,
      "loss": 6.2661,
      "mean_token_accuracy": 0.13367949202656745,
      "num_tokens": 7688373.0,
      "step": 3785
    },
    {
      "entropy": 6.318772506713867,
      "epoch": 0.3312068513501704,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004993289513357221,
      "loss": 6.1879,
      "mean_token_accuracy": 0.1309869632124901,
      "num_tokens": 7697879.0,
      "step": 3790
    },
    {
      "entropy": 6.3288311004638675,
      "epoch": 0.3316437997028751,
      "grad_norm": 1.625,
      "learning_rate": 0.0004993265443263251,
      "loss": 6.2505,
      "mean_token_accuracy": 0.12642839550971985,
      "num_tokens": 7708104.0,
      "step": 3795
    },
    {
      "entropy": 6.307876062393189,
      "epoch": 0.33208074805557986,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004993241330142274,
      "loss": 6.214,
      "mean_token_accuracy": 0.14225138276815413,
      "num_tokens": 7717626.0,
      "step": 3800
    },
    {
      "entropy": 6.292093658447266,
      "epoch": 0.33251769640828455,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004993217173994754,
      "loss": 6.2507,
      "mean_token_accuracy": 0.1357093021273613,
      "num_tokens": 7728221.0,
      "step": 3805
    },
    {
      "entropy": 6.376062059402466,
      "epoch": 0.33295464476098924,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004993192974821154,
      "loss": 6.2349,
      "mean_token_accuracy": 0.13234030231833457,
      "num_tokens": 7739083.0,
      "step": 3810
    },
    {
      "entropy": 6.372753095626831,
      "epoch": 0.333391593113694,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004993168732621935,
      "loss": 6.2732,
      "mean_token_accuracy": 0.12840995863080024,
      "num_tokens": 7748573.0,
      "step": 3815
    },
    {
      "entropy": 6.367612409591675,
      "epoch": 0.3338285414663987,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004993144447397565,
      "loss": 6.2329,
      "mean_token_accuracy": 0.13663119226694107,
      "num_tokens": 7758741.0,
      "step": 3820
    },
    {
      "entropy": 6.333526277542115,
      "epoch": 0.33426548981910337,
      "grad_norm": 1.546875,
      "learning_rate": 0.000499312011914851,
      "loss": 6.2413,
      "mean_token_accuracy": 0.1300497069954872,
      "num_tokens": 7768504.0,
      "step": 3825
    },
    {
      "entropy": 6.343638324737549,
      "epoch": 0.3347024381718081,
      "grad_norm": 1.875,
      "learning_rate": 0.0004993095747875234,
      "loss": 6.3875,
      "mean_token_accuracy": 0.12843759283423423,
      "num_tokens": 7779017.0,
      "step": 3830
    },
    {
      "entropy": 6.344084548950195,
      "epoch": 0.3351393865245128,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004993071333578207,
      "loss": 6.1865,
      "mean_token_accuracy": 0.13624552488327027,
      "num_tokens": 7787630.0,
      "step": 3835
    },
    {
      "entropy": 6.292089986801147,
      "epoch": 0.3355763348772175,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004993046876257898,
      "loss": 6.1929,
      "mean_token_accuracy": 0.13584847524762153,
      "num_tokens": 7797105.0,
      "step": 3840
    },
    {
      "entropy": 6.232599401473999,
      "epoch": 0.33601328322992224,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004993022375914773,
      "loss": 6.2056,
      "mean_token_accuracy": 0.13672226890921593,
      "num_tokens": 7807070.0,
      "step": 3845
    },
    {
      "entropy": 6.373867988586426,
      "epoch": 0.33645023158262694,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004992997832549303,
      "loss": 6.2153,
      "mean_token_accuracy": 0.13373004645109177,
      "num_tokens": 7818090.0,
      "step": 3850
    },
    {
      "entropy": 6.312876987457275,
      "epoch": 0.3368871799353316,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004992973246161961,
      "loss": 6.1596,
      "mean_token_accuracy": 0.13836875334382057,
      "num_tokens": 7827562.0,
      "step": 3855
    },
    {
      "entropy": 6.281669950485229,
      "epoch": 0.3373241282880364,
      "grad_norm": 1.75,
      "learning_rate": 0.0004992948616753215,
      "loss": 6.2139,
      "mean_token_accuracy": 0.1298152059316635,
      "num_tokens": 7837581.0,
      "step": 3860
    },
    {
      "entropy": 6.370991325378418,
      "epoch": 0.33776107664074106,
      "grad_norm": 2.0,
      "learning_rate": 0.0004992923944323541,
      "loss": 6.291,
      "mean_token_accuracy": 0.13142600283026695,
      "num_tokens": 7848729.0,
      "step": 3865
    },
    {
      "entropy": 6.291586875915527,
      "epoch": 0.33819802499344576,
      "grad_norm": 1.59375,
      "learning_rate": 0.000499289922887341,
      "loss": 6.2113,
      "mean_token_accuracy": 0.13233808651566506,
      "num_tokens": 7857379.0,
      "step": 3870
    },
    {
      "entropy": 6.4027495861053465,
      "epoch": 0.3386349733461505,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004992874470403296,
      "loss": 6.1882,
      "mean_token_accuracy": 0.13714056983590125,
      "num_tokens": 7867200.0,
      "step": 3875
    },
    {
      "entropy": 6.301979207992554,
      "epoch": 0.3390719216988552,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004992849668913674,
      "loss": 6.1906,
      "mean_token_accuracy": 0.13526173904538155,
      "num_tokens": 7876709.0,
      "step": 3880
    },
    {
      "entropy": 6.296885204315186,
      "epoch": 0.3395088700515599,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004992824824405021,
      "loss": 6.2745,
      "mean_token_accuracy": 0.1300392933189869,
      "num_tokens": 7886256.0,
      "step": 3885
    },
    {
      "entropy": 6.3121579647064205,
      "epoch": 0.33994581840426463,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004992799936877813,
      "loss": 6.216,
      "mean_token_accuracy": 0.13202250376343727,
      "num_tokens": 7896497.0,
      "step": 3890
    },
    {
      "entropy": 6.3092265129089355,
      "epoch": 0.3403827667569693,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004992775006332526,
      "loss": 6.2109,
      "mean_token_accuracy": 0.13259809389710425,
      "num_tokens": 7906608.0,
      "step": 3895
    },
    {
      "entropy": 6.359441232681275,
      "epoch": 0.340819715109674,
      "grad_norm": 1.953125,
      "learning_rate": 0.000499275003276964,
      "loss": 6.2779,
      "mean_token_accuracy": 0.1314925841987133,
      "num_tokens": 7916359.0,
      "step": 3900
    },
    {
      "entropy": 6.261974716186524,
      "epoch": 0.34125666346237876,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004992725016189632,
      "loss": 6.1721,
      "mean_token_accuracy": 0.13453363105654717,
      "num_tokens": 7925800.0,
      "step": 3905
    },
    {
      "entropy": 6.283708906173706,
      "epoch": 0.34169361181508345,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004992699956592984,
      "loss": 6.2553,
      "mean_token_accuracy": 0.12666851356625558,
      "num_tokens": 7935424.0,
      "step": 3910
    },
    {
      "entropy": 6.38649730682373,
      "epoch": 0.34213056016778814,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004992674853980175,
      "loss": 6.2616,
      "mean_token_accuracy": 0.12863119691610336,
      "num_tokens": 7945766.0,
      "step": 3915
    },
    {
      "entropy": 6.412109518051148,
      "epoch": 0.3425675085204929,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004992649708351689,
      "loss": 6.2922,
      "mean_token_accuracy": 0.12698930352926255,
      "num_tokens": 7955622.0,
      "step": 3920
    },
    {
      "entropy": 6.338596963882447,
      "epoch": 0.3430044568731976,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004992624519708004,
      "loss": 6.3246,
      "mean_token_accuracy": 0.1264619618654251,
      "num_tokens": 7965750.0,
      "step": 3925
    },
    {
      "entropy": 6.415570306777954,
      "epoch": 0.3434414052259023,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004992599288049607,
      "loss": 6.2778,
      "mean_token_accuracy": 0.1360319010913372,
      "num_tokens": 7977568.0,
      "step": 3930
    },
    {
      "entropy": 6.382330083847046,
      "epoch": 0.343878353578607,
      "grad_norm": 1.734375,
      "learning_rate": 0.000499257401337698,
      "loss": 6.3174,
      "mean_token_accuracy": 0.12833578512072563,
      "num_tokens": 7988411.0,
      "step": 3935
    },
    {
      "entropy": 6.2563676834106445,
      "epoch": 0.3443153019313117,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004992548695690608,
      "loss": 6.175,
      "mean_token_accuracy": 0.1387489154934883,
      "num_tokens": 7998315.0,
      "step": 3940
    },
    {
      "entropy": 6.279330539703369,
      "epoch": 0.3447522502840164,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004992523334990977,
      "loss": 6.2307,
      "mean_token_accuracy": 0.12642429322004317,
      "num_tokens": 8007133.0,
      "step": 3945
    },
    {
      "entropy": 6.331608390808105,
      "epoch": 0.34518919863672115,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004992497931278574,
      "loss": 6.1766,
      "mean_token_accuracy": 0.13578565120697023,
      "num_tokens": 8016882.0,
      "step": 3950
    },
    {
      "entropy": 6.30817813873291,
      "epoch": 0.34562614698942584,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004992472484553886,
      "loss": 6.1846,
      "mean_token_accuracy": 0.13525984510779382,
      "num_tokens": 8026893.0,
      "step": 3955
    },
    {
      "entropy": 6.32960057258606,
      "epoch": 0.34606309534213053,
      "grad_norm": 1.640625,
      "learning_rate": 0.00049924469948174,
      "loss": 6.2121,
      "mean_token_accuracy": 0.13157317489385606,
      "num_tokens": 8036240.0,
      "step": 3960
    },
    {
      "entropy": 6.3944062232971195,
      "epoch": 0.3465000436948353,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004992421462069606,
      "loss": 6.1882,
      "mean_token_accuracy": 0.1298525132238865,
      "num_tokens": 8046086.0,
      "step": 3965
    },
    {
      "entropy": 6.226625156402588,
      "epoch": 0.34693699204753997,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004992395886310993,
      "loss": 6.2141,
      "mean_token_accuracy": 0.13688210174441337,
      "num_tokens": 8056294.0,
      "step": 3970
    },
    {
      "entropy": 6.290466213226319,
      "epoch": 0.3473739404002447,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004992370267542052,
      "loss": 6.2412,
      "mean_token_accuracy": 0.1284775547683239,
      "num_tokens": 8067329.0,
      "step": 3975
    },
    {
      "entropy": 6.355967235565186,
      "epoch": 0.3478108887529494,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004992344605763275,
      "loss": 6.2075,
      "mean_token_accuracy": 0.13838136419653893,
      "num_tokens": 8077106.0,
      "step": 3980
    },
    {
      "entropy": 6.3079071044921875,
      "epoch": 0.3482478371056541,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004992318900975153,
      "loss": 6.1156,
      "mean_token_accuracy": 0.13575755059719086,
      "num_tokens": 8087766.0,
      "step": 3985
    },
    {
      "entropy": 6.22908730506897,
      "epoch": 0.34868478545835885,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004992293153178179,
      "loss": 6.2289,
      "mean_token_accuracy": 0.13779617995023727,
      "num_tokens": 8098649.0,
      "step": 3990
    },
    {
      "entropy": 6.301569175720215,
      "epoch": 0.34912173381106354,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004992267362372848,
      "loss": 6.2339,
      "mean_token_accuracy": 0.1273333542048931,
      "num_tokens": 8109237.0,
      "step": 3995
    },
    {
      "entropy": 6.2999763011932375,
      "epoch": 0.34955868216376823,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004992241528559653,
      "loss": 6.2344,
      "mean_token_accuracy": 0.13048239648342133,
      "num_tokens": 8119731.0,
      "step": 4000
    },
    {
      "entropy": 6.388835811614991,
      "epoch": 0.349995630516473,
      "grad_norm": 1.5,
      "learning_rate": 0.0004992215651739092,
      "loss": 6.2518,
      "mean_token_accuracy": 0.12731548473238946,
      "num_tokens": 8129448.0,
      "step": 4005
    },
    {
      "entropy": 6.251416540145874,
      "epoch": 0.35043257886917767,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000499218973191166,
      "loss": 6.131,
      "mean_token_accuracy": 0.1354080356657505,
      "num_tokens": 8138345.0,
      "step": 4010
    },
    {
      "entropy": 6.277962684631348,
      "epoch": 0.35086952722188236,
      "grad_norm": 1.5,
      "learning_rate": 0.0004992163769077853,
      "loss": 6.1982,
      "mean_token_accuracy": 0.13461976125836372,
      "num_tokens": 8148380.0,
      "step": 4015
    },
    {
      "entropy": 6.3026374816894535,
      "epoch": 0.3513064755745871,
      "grad_norm": 1.6875,
      "learning_rate": 0.000499213776323817,
      "loss": 6.2011,
      "mean_token_accuracy": 0.13807606995105742,
      "num_tokens": 8159359.0,
      "step": 4020
    },
    {
      "entropy": 6.384302949905395,
      "epoch": 0.3517434239272918,
      "grad_norm": 1.9375,
      "learning_rate": 0.000499211171439311,
      "loss": 6.3019,
      "mean_token_accuracy": 0.13113811537623404,
      "num_tokens": 8168586.0,
      "step": 4025
    },
    {
      "entropy": 6.348558378219605,
      "epoch": 0.3521803722799965,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004992085622543172,
      "loss": 6.2021,
      "mean_token_accuracy": 0.13641840294003488,
      "num_tokens": 8178836.0,
      "step": 4030
    },
    {
      "entropy": 6.198421144485474,
      "epoch": 0.35261732063270124,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004992059487688859,
      "loss": 6.1999,
      "mean_token_accuracy": 0.1357082948088646,
      "num_tokens": 8188933.0,
      "step": 4035
    },
    {
      "entropy": 6.274309825897217,
      "epoch": 0.35305426898540593,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004992033309830668,
      "loss": 6.2121,
      "mean_token_accuracy": 0.13952750116586685,
      "num_tokens": 8199366.0,
      "step": 4040
    },
    {
      "entropy": 6.392431402206421,
      "epoch": 0.3534912173381106,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004992007088969105,
      "loss": 6.181,
      "mean_token_accuracy": 0.1361783489584923,
      "num_tokens": 8209806.0,
      "step": 4045
    },
    {
      "entropy": 6.390678024291992,
      "epoch": 0.35392816569081537,
      "grad_norm": 2.078125,
      "learning_rate": 0.000499198082510467,
      "loss": 6.2629,
      "mean_token_accuracy": 0.12851823195815087,
      "num_tokens": 8219367.0,
      "step": 4050
    },
    {
      "entropy": 6.26587610244751,
      "epoch": 0.35436511404352006,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004991954518237868,
      "loss": 6.1572,
      "mean_token_accuracy": 0.13181626051664352,
      "num_tokens": 8228918.0,
      "step": 4055
    },
    {
      "entropy": 6.2422205924987795,
      "epoch": 0.35480206239622475,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004991928168369205,
      "loss": 6.2616,
      "mean_token_accuracy": 0.13364648148417474,
      "num_tokens": 8239380.0,
      "step": 4060
    },
    {
      "entropy": 6.414402294158935,
      "epoch": 0.3552390107489295,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004991901775499183,
      "loss": 6.3145,
      "mean_token_accuracy": 0.13284917771816254,
      "num_tokens": 8249131.0,
      "step": 4065
    },
    {
      "entropy": 6.40028018951416,
      "epoch": 0.3556759591016342,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004991875339628311,
      "loss": 6.2099,
      "mean_token_accuracy": 0.12965109348297119,
      "num_tokens": 8258550.0,
      "step": 4070
    },
    {
      "entropy": 6.293116044998169,
      "epoch": 0.3561129074543389,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004991848860757095,
      "loss": 6.2298,
      "mean_token_accuracy": 0.13586684167385102,
      "num_tokens": 8269089.0,
      "step": 4075
    },
    {
      "entropy": 6.273027086257935,
      "epoch": 0.3565498558070436,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004991822338886044,
      "loss": 6.2034,
      "mean_token_accuracy": 0.13297911509871482,
      "num_tokens": 8279735.0,
      "step": 4080
    },
    {
      "entropy": 6.304439830780029,
      "epoch": 0.3569868041597483,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004991795774015665,
      "loss": 6.1622,
      "mean_token_accuracy": 0.1390426389873028,
      "num_tokens": 8289693.0,
      "step": 4085
    },
    {
      "entropy": 6.297573852539062,
      "epoch": 0.357423752512453,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004991769166146469,
      "loss": 6.2431,
      "mean_token_accuracy": 0.13711095005273818,
      "num_tokens": 8300909.0,
      "step": 4090
    },
    {
      "entropy": 6.24781174659729,
      "epoch": 0.35786070086515775,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004991742515278965,
      "loss": 6.2253,
      "mean_token_accuracy": 0.13772813826799393,
      "num_tokens": 8311479.0,
      "step": 4095
    },
    {
      "entropy": 6.389146137237549,
      "epoch": 0.35829764921786245,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004991715821413665,
      "loss": 6.2126,
      "mean_token_accuracy": 0.1380111902952194,
      "num_tokens": 8321267.0,
      "step": 4100
    },
    {
      "entropy": 6.372566509246826,
      "epoch": 0.35873459757056714,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004991689084551082,
      "loss": 6.1705,
      "mean_token_accuracy": 0.13093933686614037,
      "num_tokens": 8331445.0,
      "step": 4105
    },
    {
      "entropy": 6.255347871780396,
      "epoch": 0.3591715459232719,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004991662304691727,
      "loss": 6.2525,
      "mean_token_accuracy": 0.13012968823313714,
      "num_tokens": 8340519.0,
      "step": 4110
    },
    {
      "entropy": 6.285258197784424,
      "epoch": 0.3596084942759766,
      "grad_norm": 3.0,
      "learning_rate": 0.0004991635481836114,
      "loss": 6.2245,
      "mean_token_accuracy": 0.1317976050078869,
      "num_tokens": 8350467.0,
      "step": 4115
    },
    {
      "entropy": 6.332366991043091,
      "epoch": 0.36004544262868127,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004991608615984757,
      "loss": 6.2055,
      "mean_token_accuracy": 0.1296989619731903,
      "num_tokens": 8360067.0,
      "step": 4120
    },
    {
      "entropy": 6.346688556671142,
      "epoch": 0.360482390981386,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004991581707138175,
      "loss": 6.2587,
      "mean_token_accuracy": 0.13267388194799423,
      "num_tokens": 8370509.0,
      "step": 4125
    },
    {
      "entropy": 6.327474784851074,
      "epoch": 0.3609193393340907,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004991554755296878,
      "loss": 6.2102,
      "mean_token_accuracy": 0.1328631594777107,
      "num_tokens": 8381489.0,
      "step": 4130
    },
    {
      "entropy": 6.310758972167969,
      "epoch": 0.3613562876867954,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004991527760461389,
      "loss": 6.2163,
      "mean_token_accuracy": 0.13652996942400933,
      "num_tokens": 8392287.0,
      "step": 4135
    },
    {
      "entropy": 6.322778129577637,
      "epoch": 0.36179323603950014,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004991500722632223,
      "loss": 6.2219,
      "mean_token_accuracy": 0.13221596032381058,
      "num_tokens": 8402433.0,
      "step": 4140
    },
    {
      "entropy": 6.261847448348999,
      "epoch": 0.36223018439220483,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004991473641809896,
      "loss": 6.2622,
      "mean_token_accuracy": 0.1302519731223583,
      "num_tokens": 8412944.0,
      "step": 4145
    },
    {
      "entropy": 6.351562118530273,
      "epoch": 0.3626671327449095,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004991446517994932,
      "loss": 6.1566,
      "mean_token_accuracy": 0.13552847281098365,
      "num_tokens": 8422511.0,
      "step": 4150
    },
    {
      "entropy": 6.261544895172119,
      "epoch": 0.3631040810976143,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004991419351187849,
      "loss": 6.1887,
      "mean_token_accuracy": 0.14115940481424333,
      "num_tokens": 8433094.0,
      "step": 4155
    },
    {
      "entropy": 6.267559051513672,
      "epoch": 0.36354102945031896,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004991392141389169,
      "loss": 6.2792,
      "mean_token_accuracy": 0.13087793439626694,
      "num_tokens": 8444430.0,
      "step": 4160
    },
    {
      "entropy": 6.380277824401856,
      "epoch": 0.36397797780302366,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004991364888599412,
      "loss": 6.3375,
      "mean_token_accuracy": 0.12574921697378158,
      "num_tokens": 8455171.0,
      "step": 4165
    },
    {
      "entropy": 6.351542806625366,
      "epoch": 0.3644149261557284,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004991337592819103,
      "loss": 6.1583,
      "mean_token_accuracy": 0.13929566666483878,
      "num_tokens": 8464724.0,
      "step": 4170
    },
    {
      "entropy": 6.366502380371093,
      "epoch": 0.3648518745084331,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004991310254048763,
      "loss": 6.2187,
      "mean_token_accuracy": 0.13644032552838326,
      "num_tokens": 8474732.0,
      "step": 4175
    },
    {
      "entropy": 6.324858379364014,
      "epoch": 0.36528882286113784,
      "grad_norm": 2.375,
      "learning_rate": 0.0004991282872288919,
      "loss": 6.181,
      "mean_token_accuracy": 0.13449523076415063,
      "num_tokens": 8484731.0,
      "step": 4180
    },
    {
      "entropy": 6.249287128448486,
      "epoch": 0.36572577121384253,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004991255447540095,
      "loss": 6.1791,
      "mean_token_accuracy": 0.13690031841397285,
      "num_tokens": 8494509.0,
      "step": 4185
    },
    {
      "entropy": 6.355121803283692,
      "epoch": 0.3661627195665472,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004991227979802817,
      "loss": 6.2417,
      "mean_token_accuracy": 0.13381650745868684,
      "num_tokens": 8504260.0,
      "step": 4190
    },
    {
      "entropy": 6.368691539764404,
      "epoch": 0.36659966791925197,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004991200469077612,
      "loss": 6.2896,
      "mean_token_accuracy": 0.12882014065980912,
      "num_tokens": 8514222.0,
      "step": 4195
    },
    {
      "entropy": 6.362591314315796,
      "epoch": 0.36703661627195666,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004991172915365008,
      "loss": 6.2393,
      "mean_token_accuracy": 0.1345599003136158,
      "num_tokens": 8523979.0,
      "step": 4200
    },
    {
      "entropy": 6.345672225952148,
      "epoch": 0.36747356462466135,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004991145318665532,
      "loss": 6.1827,
      "mean_token_accuracy": 0.13316255062818527,
      "num_tokens": 8533516.0,
      "step": 4205
    },
    {
      "entropy": 6.354747772216797,
      "epoch": 0.3679105129773661,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004991117678979716,
      "loss": 6.3054,
      "mean_token_accuracy": 0.12466351985931397,
      "num_tokens": 8544444.0,
      "step": 4210
    },
    {
      "entropy": 6.271222448348999,
      "epoch": 0.3683474613300708,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004991089996308089,
      "loss": 6.2132,
      "mean_token_accuracy": 0.13266286626458168,
      "num_tokens": 8555571.0,
      "step": 4215
    },
    {
      "entropy": 6.358844470977783,
      "epoch": 0.3687844096827755,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004991062270651179,
      "loss": 6.2838,
      "mean_token_accuracy": 0.12845440730452537,
      "num_tokens": 8566486.0,
      "step": 4220
    },
    {
      "entropy": 6.400605630874634,
      "epoch": 0.36922135803548023,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004991034502009522,
      "loss": 6.193,
      "mean_token_accuracy": 0.13754271492362022,
      "num_tokens": 8575324.0,
      "step": 4225
    },
    {
      "entropy": 6.3530802726745605,
      "epoch": 0.3696583063881849,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004991006690383649,
      "loss": 6.1839,
      "mean_token_accuracy": 0.13819497525691987,
      "num_tokens": 8586015.0,
      "step": 4230
    },
    {
      "entropy": 6.285164785385132,
      "epoch": 0.3700952547408896,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004990978835774093,
      "loss": 6.1634,
      "mean_token_accuracy": 0.13958604261279106,
      "num_tokens": 8594748.0,
      "step": 4235
    },
    {
      "entropy": 6.222214555740356,
      "epoch": 0.37053220309359436,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004990950938181389,
      "loss": 6.2382,
      "mean_token_accuracy": 0.12688292562961578,
      "num_tokens": 8605805.0,
      "step": 4240
    },
    {
      "entropy": 6.301893901824951,
      "epoch": 0.37096915144629905,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004990922997606071,
      "loss": 6.2171,
      "mean_token_accuracy": 0.13133906796574593,
      "num_tokens": 8616591.0,
      "step": 4245
    },
    {
      "entropy": 6.398301029205323,
      "epoch": 0.37140609979900374,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004990895014048677,
      "loss": 6.2327,
      "mean_token_accuracy": 0.1328352488577366,
      "num_tokens": 8627063.0,
      "step": 4250
    },
    {
      "entropy": 6.483572244644165,
      "epoch": 0.3718430481517085,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004990866987509741,
      "loss": 6.292,
      "mean_token_accuracy": 0.12863529548048974,
      "num_tokens": 8637041.0,
      "step": 4255
    },
    {
      "entropy": 6.353319072723389,
      "epoch": 0.3722799965044132,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004990838917989802,
      "loss": 6.2176,
      "mean_token_accuracy": 0.14009865373373032,
      "num_tokens": 8647485.0,
      "step": 4260
    },
    {
      "entropy": 6.2800273418426515,
      "epoch": 0.37271694485711787,
      "grad_norm": 1.9375,
      "learning_rate": 0.00049908108054894,
      "loss": 6.1856,
      "mean_token_accuracy": 0.12896187603473663,
      "num_tokens": 8657185.0,
      "step": 4265
    },
    {
      "entropy": 6.349277353286743,
      "epoch": 0.3731538932098226,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004990782650009071,
      "loss": 6.201,
      "mean_token_accuracy": 0.13796407282352446,
      "num_tokens": 8667639.0,
      "step": 4270
    },
    {
      "entropy": 6.400645351409912,
      "epoch": 0.3735908415625273,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004990754451549358,
      "loss": 6.2424,
      "mean_token_accuracy": 0.12953674867749215,
      "num_tokens": 8678003.0,
      "step": 4275
    },
    {
      "entropy": 6.356324338912964,
      "epoch": 0.374027789915232,
      "grad_norm": 2.375,
      "learning_rate": 0.0004990726210110798,
      "loss": 6.2695,
      "mean_token_accuracy": 0.1289951339364052,
      "num_tokens": 8688837.0,
      "step": 4280
    },
    {
      "entropy": 6.252076530456543,
      "epoch": 0.37446473826793675,
      "grad_norm": 2.125,
      "learning_rate": 0.0004990697925693938,
      "loss": 6.0801,
      "mean_token_accuracy": 0.14513857066631317,
      "num_tokens": 8698395.0,
      "step": 4285
    },
    {
      "entropy": 6.324973058700562,
      "epoch": 0.37490168662064144,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004990669598299316,
      "loss": 6.2089,
      "mean_token_accuracy": 0.12990765050053596,
      "num_tokens": 8708031.0,
      "step": 4290
    },
    {
      "entropy": 6.288310861587524,
      "epoch": 0.37533863497334613,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004990641227927478,
      "loss": 6.1754,
      "mean_token_accuracy": 0.13533689379692077,
      "num_tokens": 8717825.0,
      "step": 4295
    },
    {
      "entropy": 6.375963544845581,
      "epoch": 0.3757755833260509,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004990612814578967,
      "loss": 6.253,
      "mean_token_accuracy": 0.12898548021912576,
      "num_tokens": 8728873.0,
      "step": 4300
    },
    {
      "entropy": 6.33867506980896,
      "epoch": 0.37621253167875557,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004990584358254327,
      "loss": 6.1909,
      "mean_token_accuracy": 0.13240848779678344,
      "num_tokens": 8737917.0,
      "step": 4305
    },
    {
      "entropy": 6.232820701599121,
      "epoch": 0.37664948003146026,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004990555858954107,
      "loss": 6.1558,
      "mean_token_accuracy": 0.13440656289458275,
      "num_tokens": 8747466.0,
      "step": 4310
    },
    {
      "entropy": 6.307000064849854,
      "epoch": 0.377086428384165,
      "grad_norm": 2.3125,
      "learning_rate": 0.000499052731667885,
      "loss": 6.1712,
      "mean_token_accuracy": 0.13426090329885482,
      "num_tokens": 8757037.0,
      "step": 4315
    },
    {
      "entropy": 6.33053412437439,
      "epoch": 0.3775233767368697,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004990498731429105,
      "loss": 6.2313,
      "mean_token_accuracy": 0.1316849783062935,
      "num_tokens": 8766780.0,
      "step": 4320
    },
    {
      "entropy": 6.317747402191162,
      "epoch": 0.3779603250895744,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004990470103205421,
      "loss": 6.19,
      "mean_token_accuracy": 0.13137859404087066,
      "num_tokens": 8777063.0,
      "step": 4325
    },
    {
      "entropy": 6.304370450973511,
      "epoch": 0.37839727344227914,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004990441432008347,
      "loss": 6.2286,
      "mean_token_accuracy": 0.13164292871952057,
      "num_tokens": 8786479.0,
      "step": 4330
    },
    {
      "entropy": 6.3518133640289305,
      "epoch": 0.3788342217949838,
      "grad_norm": 2.0,
      "learning_rate": 0.0004990412717838431,
      "loss": 6.2033,
      "mean_token_accuracy": 0.1347177192568779,
      "num_tokens": 8795663.0,
      "step": 4335
    },
    {
      "entropy": 6.343950986862183,
      "epoch": 0.3792711701476885,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004990383960696226,
      "loss": 6.1799,
      "mean_token_accuracy": 0.1344247914850712,
      "num_tokens": 8805854.0,
      "step": 4340
    },
    {
      "entropy": 6.261251306533813,
      "epoch": 0.37970811850039327,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004990355160582283,
      "loss": 6.1557,
      "mean_token_accuracy": 0.13314540237188338,
      "num_tokens": 8816068.0,
      "step": 4345
    },
    {
      "entropy": 6.227701425552368,
      "epoch": 0.38014506685309796,
      "grad_norm": 2.25,
      "learning_rate": 0.0004990326317497153,
      "loss": 6.1711,
      "mean_token_accuracy": 0.13875154703855513,
      "num_tokens": 8826301.0,
      "step": 4350
    },
    {
      "entropy": 6.319527101516724,
      "epoch": 0.38058201520580265,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004990297431441391,
      "loss": 6.2026,
      "mean_token_accuracy": 0.13681671246886254,
      "num_tokens": 8837482.0,
      "step": 4355
    },
    {
      "entropy": 6.328002500534057,
      "epoch": 0.3810189635585074,
      "grad_norm": 4.1875,
      "learning_rate": 0.000499026850241555,
      "loss": 6.2416,
      "mean_token_accuracy": 0.13217343688011168,
      "num_tokens": 8847041.0,
      "step": 4360
    },
    {
      "entropy": 6.330576324462891,
      "epoch": 0.3814559119112121,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004990239530420185,
      "loss": 6.2224,
      "mean_token_accuracy": 0.13127602115273476,
      "num_tokens": 8856963.0,
      "step": 4365
    },
    {
      "entropy": 6.333612012863159,
      "epoch": 0.3818928602639168,
      "grad_norm": 3.0,
      "learning_rate": 0.0004990210515455853,
      "loss": 6.1718,
      "mean_token_accuracy": 0.132768964022398,
      "num_tokens": 8867318.0,
      "step": 4370
    },
    {
      "entropy": 6.3557415962219235,
      "epoch": 0.3823298086166215,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004990181457523109,
      "loss": 6.2149,
      "mean_token_accuracy": 0.13497575670480727,
      "num_tokens": 8877634.0,
      "step": 4375
    },
    {
      "entropy": 6.368174457550049,
      "epoch": 0.3827667569693262,
      "grad_norm": 2.4375,
      "learning_rate": 0.000499015235662251,
      "loss": 6.2842,
      "mean_token_accuracy": 0.13122443556785585,
      "num_tokens": 8887708.0,
      "step": 4380
    },
    {
      "entropy": 6.3176483631134035,
      "epoch": 0.38320370532203096,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004990123212754616,
      "loss": 6.1709,
      "mean_token_accuracy": 0.1362153872847557,
      "num_tokens": 8897510.0,
      "step": 4385
    },
    {
      "entropy": 6.30413408279419,
      "epoch": 0.38364065367473565,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004990094025919985,
      "loss": 6.233,
      "mean_token_accuracy": 0.125152063369751,
      "num_tokens": 8908282.0,
      "step": 4390
    },
    {
      "entropy": 6.432730293273925,
      "epoch": 0.38407760202744035,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004990064796119175,
      "loss": 6.2594,
      "mean_token_accuracy": 0.12678856030106544,
      "num_tokens": 8918332.0,
      "step": 4395
    },
    {
      "entropy": 6.330333995819092,
      "epoch": 0.3845145503801451,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004990035523352751,
      "loss": 6.2167,
      "mean_token_accuracy": 0.13248514011502266,
      "num_tokens": 8929138.0,
      "step": 4400
    },
    {
      "entropy": 6.285112810134888,
      "epoch": 0.3849514987328498,
      "grad_norm": 3.046875,
      "learning_rate": 0.000499000620762127,
      "loss": 6.2683,
      "mean_token_accuracy": 0.1243371956050396,
      "num_tokens": 8940871.0,
      "step": 4405
    },
    {
      "entropy": 6.322092437744141,
      "epoch": 0.3853884470855545,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004989976848925297,
      "loss": 6.1851,
      "mean_token_accuracy": 0.13609126284718515,
      "num_tokens": 8951199.0,
      "step": 4410
    },
    {
      "entropy": 6.383452224731445,
      "epoch": 0.3858253954382592,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004989947447265394,
      "loss": 6.3375,
      "mean_token_accuracy": 0.1250775396823883,
      "num_tokens": 8961482.0,
      "step": 4415
    },
    {
      "entropy": 6.421683025360108,
      "epoch": 0.3862623437909639,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004989918002642126,
      "loss": 6.2318,
      "mean_token_accuracy": 0.13731383457779883,
      "num_tokens": 8971884.0,
      "step": 4420
    },
    {
      "entropy": 6.323845338821411,
      "epoch": 0.3866992921436686,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004989888515056057,
      "loss": 6.1909,
      "mean_token_accuracy": 0.12858997508883477,
      "num_tokens": 8982086.0,
      "step": 4425
    },
    {
      "entropy": 6.266624975204468,
      "epoch": 0.38713624049637335,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004989858984507752,
      "loss": 6.2337,
      "mean_token_accuracy": 0.13364778310060502,
      "num_tokens": 8991555.0,
      "step": 4430
    },
    {
      "entropy": 6.407159423828125,
      "epoch": 0.38757318884907804,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004989829410997778,
      "loss": 6.2281,
      "mean_token_accuracy": 0.12950120121240616,
      "num_tokens": 9002459.0,
      "step": 4435
    },
    {
      "entropy": 6.351750326156616,
      "epoch": 0.38801013720178273,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004989799794526702,
      "loss": 6.25,
      "mean_token_accuracy": 0.13045488223433493,
      "num_tokens": 9012514.0,
      "step": 4440
    },
    {
      "entropy": 6.2411726951599125,
      "epoch": 0.3884470855544875,
      "grad_norm": 2.875,
      "learning_rate": 0.0004989770135095093,
      "loss": 6.182,
      "mean_token_accuracy": 0.13259212225675582,
      "num_tokens": 9022636.0,
      "step": 4445
    },
    {
      "entropy": 6.271308183670044,
      "epoch": 0.3888840339071922,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004989740432703519,
      "loss": 6.1872,
      "mean_token_accuracy": 0.1361326076090336,
      "num_tokens": 9032868.0,
      "step": 4450
    },
    {
      "entropy": 6.3515726089477536,
      "epoch": 0.38932098225989686,
      "grad_norm": 2.25,
      "learning_rate": 0.000498971068735255,
      "loss": 6.2127,
      "mean_token_accuracy": 0.13220859691500664,
      "num_tokens": 9043057.0,
      "step": 4455
    },
    {
      "entropy": 6.357582473754883,
      "epoch": 0.3897579306126016,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004989680899042756,
      "loss": 6.2157,
      "mean_token_accuracy": 0.1331144779920578,
      "num_tokens": 9053179.0,
      "step": 4460
    },
    {
      "entropy": 6.333834505081176,
      "epoch": 0.3901948789653063,
      "grad_norm": 5.65625,
      "learning_rate": 0.0004989651067774708,
      "loss": 6.2166,
      "mean_token_accuracy": 0.1353854015469551,
      "num_tokens": 9063112.0,
      "step": 4465
    },
    {
      "entropy": 6.4165887355804445,
      "epoch": 0.390631827318011,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004989621193548981,
      "loss": 6.2326,
      "mean_token_accuracy": 0.13364190012216567,
      "num_tokens": 9073512.0,
      "step": 4470
    },
    {
      "entropy": 6.397773551940918,
      "epoch": 0.39106877567071574,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004989591276366145,
      "loss": 6.2467,
      "mean_token_accuracy": 0.1245313473045826,
      "num_tokens": 9083369.0,
      "step": 4475
    },
    {
      "entropy": 6.377582025527954,
      "epoch": 0.39150572402342043,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004989561316226774,
      "loss": 6.2289,
      "mean_token_accuracy": 0.12666978389024736,
      "num_tokens": 9092814.0,
      "step": 4480
    },
    {
      "entropy": 6.289837169647217,
      "epoch": 0.3919426723761251,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004989531313131444,
      "loss": 6.1567,
      "mean_token_accuracy": 0.14440342783927917,
      "num_tokens": 9102816.0,
      "step": 4485
    },
    {
      "entropy": 6.369090747833252,
      "epoch": 0.39237962072882987,
      "grad_norm": 2.890625,
      "learning_rate": 0.000498950126708073,
      "loss": 6.2747,
      "mean_token_accuracy": 0.12634778171777725,
      "num_tokens": 9114470.0,
      "step": 4490
    },
    {
      "entropy": 6.343052768707276,
      "epoch": 0.39281656908153456,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004989471178075208,
      "loss": 6.263,
      "mean_token_accuracy": 0.12880805879831314,
      "num_tokens": 9125604.0,
      "step": 4495
    },
    {
      "entropy": 6.344299697875977,
      "epoch": 0.39325351743423925,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004989441046115456,
      "loss": 6.2132,
      "mean_token_accuracy": 0.13917397931218148,
      "num_tokens": 9135088.0,
      "step": 4500
    },
    {
      "entropy": 6.366170072555542,
      "epoch": 0.393690465786944,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004989410871202051,
      "loss": 6.2753,
      "mean_token_accuracy": 0.1303461529314518,
      "num_tokens": 9145302.0,
      "step": 4505
    },
    {
      "entropy": 6.302958202362061,
      "epoch": 0.3941274141396487,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004989380653335571,
      "loss": 6.1889,
      "mean_token_accuracy": 0.13727670386433602,
      "num_tokens": 9155104.0,
      "step": 4510
    },
    {
      "entropy": 6.275707244873047,
      "epoch": 0.3945643624923534,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004989350392516598,
      "loss": 6.1058,
      "mean_token_accuracy": 0.14107887595891952,
      "num_tokens": 9165055.0,
      "step": 4515
    },
    {
      "entropy": 6.355810928344726,
      "epoch": 0.39500131084505813,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004989320088745711,
      "loss": 6.1412,
      "mean_token_accuracy": 0.13949491381645202,
      "num_tokens": 9175305.0,
      "step": 4520
    },
    {
      "entropy": 6.277345275878906,
      "epoch": 0.3954382591977628,
      "grad_norm": 2.1875,
      "learning_rate": 0.000498928974202349,
      "loss": 6.2243,
      "mean_token_accuracy": 0.1311698243021965,
      "num_tokens": 9185153.0,
      "step": 4525
    },
    {
      "entropy": 6.344827270507812,
      "epoch": 0.3958752075504675,
      "grad_norm": 2.0625,
      "learning_rate": 0.000498925935235052,
      "loss": 6.1768,
      "mean_token_accuracy": 0.13333819434046745,
      "num_tokens": 9195385.0,
      "step": 4530
    },
    {
      "entropy": 6.31925482749939,
      "epoch": 0.39631215590317226,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004989228919727382,
      "loss": 6.1545,
      "mean_token_accuracy": 0.13643745556473733,
      "num_tokens": 9205009.0,
      "step": 4535
    },
    {
      "entropy": 6.33775782585144,
      "epoch": 0.39674910425587695,
      "grad_norm": 3.40625,
      "learning_rate": 0.000498919844415466,
      "loss": 6.2189,
      "mean_token_accuracy": 0.1322916030883789,
      "num_tokens": 9214684.0,
      "step": 4540
    },
    {
      "entropy": 6.304624128341675,
      "epoch": 0.39718605260858164,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004989167925632937,
      "loss": 6.1618,
      "mean_token_accuracy": 0.13362237066030502,
      "num_tokens": 9225721.0,
      "step": 4545
    },
    {
      "entropy": 6.3220361232757565,
      "epoch": 0.3976230009612864,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004989137364162801,
      "loss": 6.2355,
      "mean_token_accuracy": 0.1321760594844818,
      "num_tokens": 9236167.0,
      "step": 4550
    },
    {
      "entropy": 6.333114337921143,
      "epoch": 0.3980599493139911,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004989106759744837,
      "loss": 6.2027,
      "mean_token_accuracy": 0.13957723453640938,
      "num_tokens": 9245716.0,
      "step": 4555
    },
    {
      "entropy": 6.3294679641723635,
      "epoch": 0.39849689766669577,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004989076112379633,
      "loss": 6.2993,
      "mean_token_accuracy": 0.1256670795381069,
      "num_tokens": 9256897.0,
      "step": 4560
    },
    {
      "entropy": 6.35600323677063,
      "epoch": 0.3989338460194005,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004989045422067775,
      "loss": 6.2766,
      "mean_token_accuracy": 0.12301432862877845,
      "num_tokens": 9266080.0,
      "step": 4565
    },
    {
      "entropy": 6.39034857749939,
      "epoch": 0.3993707943721052,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004989014688809853,
      "loss": 6.2087,
      "mean_token_accuracy": 0.13007153794169427,
      "num_tokens": 9276815.0,
      "step": 4570
    },
    {
      "entropy": 6.38375153541565,
      "epoch": 0.3998077427248099,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004988983912606454,
      "loss": 6.2623,
      "mean_token_accuracy": 0.1269598439335823,
      "num_tokens": 9287339.0,
      "step": 4575
    },
    {
      "entropy": 6.3262513160705565,
      "epoch": 0.40024469107751465,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004988953093458173,
      "loss": 6.1506,
      "mean_token_accuracy": 0.134234007447958,
      "num_tokens": 9298002.0,
      "step": 4580
    },
    {
      "entropy": 6.309459829330445,
      "epoch": 0.40068163943021934,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004988922231365597,
      "loss": 6.27,
      "mean_token_accuracy": 0.12791937291622163,
      "num_tokens": 9309008.0,
      "step": 4585
    },
    {
      "entropy": 6.359013319015503,
      "epoch": 0.4011185877829241,
      "grad_norm": 3.203125,
      "learning_rate": 0.000498889132632932,
      "loss": 6.2533,
      "mean_token_accuracy": 0.12615596279501914,
      "num_tokens": 9319391.0,
      "step": 4590
    },
    {
      "entropy": 6.307515335083008,
      "epoch": 0.4015555361356288,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004988860378349934,
      "loss": 6.1203,
      "mean_token_accuracy": 0.14449022710323334,
      "num_tokens": 9329808.0,
      "step": 4595
    },
    {
      "entropy": 6.343695974349975,
      "epoch": 0.40199248448833347,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004988829387428033,
      "loss": 6.1742,
      "mean_token_accuracy": 0.13350894600152968,
      "num_tokens": 9339580.0,
      "step": 4600
    },
    {
      "entropy": 6.302004814147949,
      "epoch": 0.4024294328410382,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004988798353564211,
      "loss": 6.1957,
      "mean_token_accuracy": 0.13340522944927216,
      "num_tokens": 9350008.0,
      "step": 4605
    },
    {
      "entropy": 6.3295636653900145,
      "epoch": 0.4028663811937429,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004988767276759064,
      "loss": 6.2057,
      "mean_token_accuracy": 0.1354494571685791,
      "num_tokens": 9359026.0,
      "step": 4610
    },
    {
      "entropy": 6.372815275192261,
      "epoch": 0.4033033295464476,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004988736157013186,
      "loss": 6.1657,
      "mean_token_accuracy": 0.13331000208854676,
      "num_tokens": 9369084.0,
      "step": 4615
    },
    {
      "entropy": 6.298906898498535,
      "epoch": 0.40374027789915234,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004988704994327178,
      "loss": 6.1527,
      "mean_token_accuracy": 0.14080883488059043,
      "num_tokens": 9378273.0,
      "step": 4620
    },
    {
      "entropy": 6.248374652862549,
      "epoch": 0.40417722625185704,
      "grad_norm": 3.625,
      "learning_rate": 0.0004988673788701633,
      "loss": 6.1726,
      "mean_token_accuracy": 0.12975530996918677,
      "num_tokens": 9388434.0,
      "step": 4625
    },
    {
      "entropy": 6.304504823684693,
      "epoch": 0.4046141746045617,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004988642540137152,
      "loss": 6.2795,
      "mean_token_accuracy": 0.129329115152359,
      "num_tokens": 9398484.0,
      "step": 4630
    },
    {
      "entropy": 6.35351710319519,
      "epoch": 0.4050511229572665,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004988611248634335,
      "loss": 6.1633,
      "mean_token_accuracy": 0.13218208849430085,
      "num_tokens": 9409544.0,
      "step": 4635
    },
    {
      "entropy": 6.400244426727295,
      "epoch": 0.40548807130997117,
      "grad_norm": 2.375,
      "learning_rate": 0.000498857991419378,
      "loss": 6.2198,
      "mean_token_accuracy": 0.1309620589017868,
      "num_tokens": 9419391.0,
      "step": 4640
    },
    {
      "entropy": 6.342773294448852,
      "epoch": 0.40592501966267586,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004988548536816091,
      "loss": 6.1654,
      "mean_token_accuracy": 0.14036228135228157,
      "num_tokens": 9429342.0,
      "step": 4645
    },
    {
      "entropy": 6.304135370254516,
      "epoch": 0.4063619680153806,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004988517116501867,
      "loss": 6.2819,
      "mean_token_accuracy": 0.12506008446216582,
      "num_tokens": 9440643.0,
      "step": 4650
    },
    {
      "entropy": 6.373707723617554,
      "epoch": 0.4067989163680853,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004988485653251711,
      "loss": 6.2326,
      "mean_token_accuracy": 0.13396082818508148,
      "num_tokens": 9450674.0,
      "step": 4655
    },
    {
      "entropy": 6.3689381122589115,
      "epoch": 0.40723586472079,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004988454147066228,
      "loss": 6.1979,
      "mean_token_accuracy": 0.13252903372049332,
      "num_tokens": 9460075.0,
      "step": 4660
    },
    {
      "entropy": 6.369057035446167,
      "epoch": 0.40767281307349473,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004988422597946022,
      "loss": 6.1971,
      "mean_token_accuracy": 0.1391727052628994,
      "num_tokens": 9469327.0,
      "step": 4665
    },
    {
      "entropy": 6.350367784500122,
      "epoch": 0.4081097614261994,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004988391005891698,
      "loss": 6.2211,
      "mean_token_accuracy": 0.12823874801397322,
      "num_tokens": 9478304.0,
      "step": 4670
    },
    {
      "entropy": 6.3489550113677975,
      "epoch": 0.4085467097789041,
      "grad_norm": 3.359375,
      "learning_rate": 0.000498835937090386,
      "loss": 6.1608,
      "mean_token_accuracy": 0.13767316490411757,
      "num_tokens": 9487453.0,
      "step": 4675
    },
    {
      "entropy": 6.3565105438232425,
      "epoch": 0.40898365813160886,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004988327692983116,
      "loss": 6.13,
      "mean_token_accuracy": 0.13946412429213523,
      "num_tokens": 9496623.0,
      "step": 4680
    },
    {
      "entropy": 6.271836757659912,
      "epoch": 0.40942060648431355,
      "grad_norm": 2.125,
      "learning_rate": 0.0004988295972130076,
      "loss": 6.2357,
      "mean_token_accuracy": 0.12985719367861748,
      "num_tokens": 9506293.0,
      "step": 4685
    },
    {
      "entropy": 6.262523937225342,
      "epoch": 0.40985755483701825,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004988264208345346,
      "loss": 6.1556,
      "mean_token_accuracy": 0.13543434739112853,
      "num_tokens": 9516127.0,
      "step": 4690
    },
    {
      "entropy": 6.22827033996582,
      "epoch": 0.410294503189723,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004988232401629535,
      "loss": 6.1498,
      "mean_token_accuracy": 0.13715630173683166,
      "num_tokens": 9525597.0,
      "step": 4695
    },
    {
      "entropy": 6.259063291549682,
      "epoch": 0.4107314515424277,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004988200551983254,
      "loss": 6.1272,
      "mean_token_accuracy": 0.13409832268953323,
      "num_tokens": 9535208.0,
      "step": 4700
    },
    {
      "entropy": 6.256589746475219,
      "epoch": 0.4111683998951324,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004988168659407113,
      "loss": 6.1913,
      "mean_token_accuracy": 0.14242529571056367,
      "num_tokens": 9545078.0,
      "step": 4705
    },
    {
      "entropy": 6.364013671875,
      "epoch": 0.4116053482478371,
      "grad_norm": 2.5,
      "learning_rate": 0.0004988136723901725,
      "loss": 6.2199,
      "mean_token_accuracy": 0.12954655215144156,
      "num_tokens": 9555074.0,
      "step": 4710
    },
    {
      "entropy": 6.331203413009644,
      "epoch": 0.4120422966005418,
      "grad_norm": 3.0,
      "learning_rate": 0.0004988104745467702,
      "loss": 6.2046,
      "mean_token_accuracy": 0.13258554860949517,
      "num_tokens": 9564777.0,
      "step": 4715
    },
    {
      "entropy": 6.3739911079406735,
      "epoch": 0.4124792449532465,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004988072724105657,
      "loss": 6.1463,
      "mean_token_accuracy": 0.12695277780294417,
      "num_tokens": 9574551.0,
      "step": 4720
    },
    {
      "entropy": 6.40868067741394,
      "epoch": 0.41291619330595125,
      "grad_norm": 3.5,
      "learning_rate": 0.0004988040659816203,
      "loss": 6.2926,
      "mean_token_accuracy": 0.12665969356894494,
      "num_tokens": 9585162.0,
      "step": 4725
    },
    {
      "entropy": 6.364977884292602,
      "epoch": 0.41335314165865594,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004988008552599958,
      "loss": 6.158,
      "mean_token_accuracy": 0.13905332908034324,
      "num_tokens": 9595422.0,
      "step": 4730
    },
    {
      "entropy": 6.346607732772827,
      "epoch": 0.41379009001136063,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004987976402457536,
      "loss": 6.1699,
      "mean_token_accuracy": 0.13569161742925645,
      "num_tokens": 9604315.0,
      "step": 4735
    },
    {
      "entropy": 6.317305278778076,
      "epoch": 0.4142270383640654,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004987944209389555,
      "loss": 6.2102,
      "mean_token_accuracy": 0.12762513458728791,
      "num_tokens": 9614369.0,
      "step": 4740
    },
    {
      "entropy": 6.371903419494629,
      "epoch": 0.41466398671677007,
      "grad_norm": 3.6875,
      "learning_rate": 0.000498791197339663,
      "loss": 6.2261,
      "mean_token_accuracy": 0.133137733489275,
      "num_tokens": 9624154.0,
      "step": 4745
    },
    {
      "entropy": 6.384650421142578,
      "epoch": 0.41510093506947476,
      "grad_norm": 1.78125,
      "learning_rate": 0.000498787969447938,
      "loss": 6.2193,
      "mean_token_accuracy": 0.1304043561220169,
      "num_tokens": 9634903.0,
      "step": 4750
    },
    {
      "entropy": 6.273547124862671,
      "epoch": 0.4155378834221795,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004987847372638427,
      "loss": 6.2591,
      "mean_token_accuracy": 0.13624944239854814,
      "num_tokens": 9644873.0,
      "step": 4755
    },
    {
      "entropy": 6.343345975875854,
      "epoch": 0.4159748317748842,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004987815007874388,
      "loss": 6.1598,
      "mean_token_accuracy": 0.13865959793329238,
      "num_tokens": 9654661.0,
      "step": 4760
    },
    {
      "entropy": 6.3764279842376705,
      "epoch": 0.4164117801275889,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004987782600187884,
      "loss": 6.186,
      "mean_token_accuracy": 0.1353089243173599,
      "num_tokens": 9664980.0,
      "step": 4765
    },
    {
      "entropy": 6.31857647895813,
      "epoch": 0.41684872848029364,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004987750149579537,
      "loss": 6.1903,
      "mean_token_accuracy": 0.13203165382146836,
      "num_tokens": 9675117.0,
      "step": 4770
    },
    {
      "entropy": 6.308708715438843,
      "epoch": 0.41728567683299833,
      "grad_norm": 3.1875,
      "learning_rate": 0.000498771765604997,
      "loss": 6.2248,
      "mean_token_accuracy": 0.133675067871809,
      "num_tokens": 9684900.0,
      "step": 4775
    },
    {
      "entropy": 6.302854871749878,
      "epoch": 0.417722625185703,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004987685119599806,
      "loss": 6.1933,
      "mean_token_accuracy": 0.13234477266669273,
      "num_tokens": 9694586.0,
      "step": 4780
    },
    {
      "entropy": 6.364311838150025,
      "epoch": 0.41815957353840777,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004987652540229668,
      "loss": 6.2179,
      "mean_token_accuracy": 0.13373128995299338,
      "num_tokens": 9705676.0,
      "step": 4785
    },
    {
      "entropy": 6.37935996055603,
      "epoch": 0.41859652189111246,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004987619917940183,
      "loss": 6.2444,
      "mean_token_accuracy": 0.13876397609710694,
      "num_tokens": 9715442.0,
      "step": 4790
    },
    {
      "entropy": 6.350925874710083,
      "epoch": 0.4190334702438172,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004987587252731975,
      "loss": 6.3194,
      "mean_token_accuracy": 0.128681218624115,
      "num_tokens": 9726621.0,
      "step": 4795
    },
    {
      "entropy": 6.380429744720459,
      "epoch": 0.4194704185965219,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004987554544605671,
      "loss": 6.283,
      "mean_token_accuracy": 0.12499735727906228,
      "num_tokens": 9736688.0,
      "step": 4800
    },
    {
      "entropy": 6.395989799499512,
      "epoch": 0.4199073669492266,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004987521793561898,
      "loss": 6.1869,
      "mean_token_accuracy": 0.13693563565611838,
      "num_tokens": 9746478.0,
      "step": 4805
    },
    {
      "entropy": 6.310982275009155,
      "epoch": 0.42034431530193134,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004987488999601285,
      "loss": 6.225,
      "mean_token_accuracy": 0.13415534943342208,
      "num_tokens": 9756370.0,
      "step": 4810
    },
    {
      "entropy": 6.357937002182007,
      "epoch": 0.42078126365463603,
      "grad_norm": 2.25,
      "learning_rate": 0.000498745616272446,
      "loss": 6.234,
      "mean_token_accuracy": 0.12906936183571815,
      "num_tokens": 9766293.0,
      "step": 4815
    },
    {
      "entropy": 6.360986566543579,
      "epoch": 0.4212182120073407,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004987423282932054,
      "loss": 6.1952,
      "mean_token_accuracy": 0.13485168293118477,
      "num_tokens": 9777172.0,
      "step": 4820
    },
    {
      "entropy": 6.26967487335205,
      "epoch": 0.42165516036004547,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004987390360224697,
      "loss": 6.183,
      "mean_token_accuracy": 0.13216223046183587,
      "num_tokens": 9787194.0,
      "step": 4825
    },
    {
      "entropy": 6.295327281951904,
      "epoch": 0.42209210871275016,
      "grad_norm": 1.90625,
      "learning_rate": 0.000498735739460302,
      "loss": 6.2207,
      "mean_token_accuracy": 0.13707452416419982,
      "num_tokens": 9797528.0,
      "step": 4830
    },
    {
      "entropy": 6.361196422576905,
      "epoch": 0.42252905706545485,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004987324386067656,
      "loss": 6.2418,
      "mean_token_accuracy": 0.1308910645544529,
      "num_tokens": 9808412.0,
      "step": 4835
    },
    {
      "entropy": 6.439935207366943,
      "epoch": 0.4229660054181596,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004987291334619239,
      "loss": 6.3085,
      "mean_token_accuracy": 0.1274378143250942,
      "num_tokens": 9820047.0,
      "step": 4840
    },
    {
      "entropy": 6.343143939971924,
      "epoch": 0.4234029537708643,
      "grad_norm": 2.484375,
      "learning_rate": 0.00049872582402584,
      "loss": 6.1404,
      "mean_token_accuracy": 0.13856913149356842,
      "num_tokens": 9828812.0,
      "step": 4845
    },
    {
      "entropy": 6.288135051727295,
      "epoch": 0.423839902123569,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004987225102985778,
      "loss": 6.1672,
      "mean_token_accuracy": 0.13984474390745164,
      "num_tokens": 9838288.0,
      "step": 4850
    },
    {
      "entropy": 6.376266527175903,
      "epoch": 0.4242768504762737,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004987191922802003,
      "loss": 6.2097,
      "mean_token_accuracy": 0.13068010956048964,
      "num_tokens": 9848536.0,
      "step": 4855
    },
    {
      "entropy": 6.359303569793701,
      "epoch": 0.4247137988289784,
      "grad_norm": 3.125,
      "learning_rate": 0.0004987158699707717,
      "loss": 6.2452,
      "mean_token_accuracy": 0.13113328889012338,
      "num_tokens": 9859741.0,
      "step": 4860
    },
    {
      "entropy": 6.354174041748047,
      "epoch": 0.4251507471816831,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004987125433703554,
      "loss": 6.2068,
      "mean_token_accuracy": 0.1325574554502964,
      "num_tokens": 9869173.0,
      "step": 4865
    },
    {
      "entropy": 6.3488525390625,
      "epoch": 0.42558769553438786,
      "grad_norm": 3.875,
      "learning_rate": 0.0004987092124790155,
      "loss": 6.2126,
      "mean_token_accuracy": 0.1348761223256588,
      "num_tokens": 9879619.0,
      "step": 4870
    },
    {
      "entropy": 6.317868375778199,
      "epoch": 0.42602464388709255,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004987058772968154,
      "loss": 6.1489,
      "mean_token_accuracy": 0.14312925338745117,
      "num_tokens": 9890214.0,
      "step": 4875
    },
    {
      "entropy": 6.235472631454468,
      "epoch": 0.42646159223979724,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004987025378238195,
      "loss": 6.1783,
      "mean_token_accuracy": 0.13242367133498192,
      "num_tokens": 9900266.0,
      "step": 4880
    },
    {
      "entropy": 6.318535995483399,
      "epoch": 0.426898540592502,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004986991940600917,
      "loss": 6.2121,
      "mean_token_accuracy": 0.1316740058362484,
      "num_tokens": 9910790.0,
      "step": 4885
    },
    {
      "entropy": 6.401111221313476,
      "epoch": 0.4273354889452067,
      "grad_norm": 2.5,
      "learning_rate": 0.0004986958460056961,
      "loss": 6.2292,
      "mean_token_accuracy": 0.12438440248370171,
      "num_tokens": 9921167.0,
      "step": 4890
    },
    {
      "entropy": 6.350015115737915,
      "epoch": 0.42777243729791137,
      "grad_norm": 2.828125,
      "learning_rate": 0.000498692493660697,
      "loss": 6.1613,
      "mean_token_accuracy": 0.1354078970849514,
      "num_tokens": 9931674.0,
      "step": 4895
    },
    {
      "entropy": 6.311030673980713,
      "epoch": 0.4282093856506161,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004986891370251587,
      "loss": 6.1063,
      "mean_token_accuracy": 0.13919934183359145,
      "num_tokens": 9941484.0,
      "step": 4900
    },
    {
      "entropy": 6.307171487808228,
      "epoch": 0.4286463340033208,
      "grad_norm": 13.1875,
      "learning_rate": 0.0004986857760991454,
      "loss": 6.2663,
      "mean_token_accuracy": 0.12672565057873725,
      "num_tokens": 9951885.0,
      "step": 4905
    },
    {
      "entropy": 6.349592924118042,
      "epoch": 0.4290832823560255,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004986824108827218,
      "loss": 6.1208,
      "mean_token_accuracy": 0.13942600041627884,
      "num_tokens": 9961658.0,
      "step": 4910
    },
    {
      "entropy": 6.295965528488159,
      "epoch": 0.42952023070873024,
      "grad_norm": 3.5,
      "learning_rate": 0.0004986790413759523,
      "loss": 6.1927,
      "mean_token_accuracy": 0.13160355910658836,
      "num_tokens": 9971971.0,
      "step": 4915
    },
    {
      "entropy": 6.230391168594361,
      "epoch": 0.42995717906143494,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004986756675789016,
      "loss": 6.1128,
      "mean_token_accuracy": 0.1330376461148262,
      "num_tokens": 9982211.0,
      "step": 4920
    },
    {
      "entropy": 6.269839906692505,
      "epoch": 0.4303941274141396,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004986722894916344,
      "loss": 6.1637,
      "mean_token_accuracy": 0.13699071183800698,
      "num_tokens": 9991935.0,
      "step": 4925
    },
    {
      "entropy": 6.339365673065186,
      "epoch": 0.4308310757668444,
      "grad_norm": 4.25,
      "learning_rate": 0.0004986689071142155,
      "loss": 6.2111,
      "mean_token_accuracy": 0.12881848737597465,
      "num_tokens": 10002014.0,
      "step": 4930
    },
    {
      "entropy": 6.390799903869629,
      "epoch": 0.43126802411954906,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004986655204467097,
      "loss": 6.2255,
      "mean_token_accuracy": 0.1304694876074791,
      "num_tokens": 10012030.0,
      "step": 4935
    },
    {
      "entropy": 6.374510955810547,
      "epoch": 0.43170497247225376,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004986621294891822,
      "loss": 6.1999,
      "mean_token_accuracy": 0.1363585837185383,
      "num_tokens": 10022188.0,
      "step": 4940
    },
    {
      "entropy": 6.369847154617309,
      "epoch": 0.4321419208249585,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004986587342416976,
      "loss": 6.11,
      "mean_token_accuracy": 0.13716743886470795,
      "num_tokens": 10032131.0,
      "step": 4945
    },
    {
      "entropy": 6.316085147857666,
      "epoch": 0.4325788691776632,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004986553347043214,
      "loss": 6.1581,
      "mean_token_accuracy": 0.13114331290125847,
      "num_tokens": 10043204.0,
      "step": 4950
    },
    {
      "entropy": 6.303411293029785,
      "epoch": 0.4330158175303679,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004986519308771188,
      "loss": 6.2368,
      "mean_token_accuracy": 0.12862323522567748,
      "num_tokens": 10052863.0,
      "step": 4955
    },
    {
      "entropy": 6.290483283996582,
      "epoch": 0.43345276588307263,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004986485227601548,
      "loss": 6.193,
      "mean_token_accuracy": 0.13968416228890418,
      "num_tokens": 10063212.0,
      "step": 4960
    },
    {
      "entropy": 6.378300428390503,
      "epoch": 0.4338897142357773,
      "grad_norm": 2.578125,
      "learning_rate": 0.000498645110353495,
      "loss": 6.2219,
      "mean_token_accuracy": 0.13019359335303307,
      "num_tokens": 10074180.0,
      "step": 4965
    },
    {
      "entropy": 6.3034398555755615,
      "epoch": 0.434326662588482,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004986416936572049,
      "loss": 6.1842,
      "mean_token_accuracy": 0.1369595482945442,
      "num_tokens": 10084513.0,
      "step": 4970
    },
    {
      "entropy": 6.340153932571411,
      "epoch": 0.43476361094118676,
      "grad_norm": 2.625,
      "learning_rate": 0.0004986382726713497,
      "loss": 6.199,
      "mean_token_accuracy": 0.1338914752006531,
      "num_tokens": 10094352.0,
      "step": 4975
    },
    {
      "entropy": 6.265435123443604,
      "epoch": 0.43520055929389145,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004986348473959955,
      "loss": 6.0588,
      "mean_token_accuracy": 0.14778584688901902,
      "num_tokens": 10103677.0,
      "step": 4980
    },
    {
      "entropy": 6.282804775238037,
      "epoch": 0.43563750764659614,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004986314178312076,
      "loss": 6.2775,
      "mean_token_accuracy": 0.12731671780347825,
      "num_tokens": 10113479.0,
      "step": 4985
    },
    {
      "entropy": 6.30030460357666,
      "epoch": 0.4360744559993009,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004986279839770519,
      "loss": 6.1793,
      "mean_token_accuracy": 0.13244004175066948,
      "num_tokens": 10123526.0,
      "step": 4990
    },
    {
      "entropy": 6.376892614364624,
      "epoch": 0.4365114043520056,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004986245458335944,
      "loss": 6.192,
      "mean_token_accuracy": 0.1345130018889904,
      "num_tokens": 10134176.0,
      "step": 4995
    },
    {
      "entropy": 6.4130466938018795,
      "epoch": 0.43694835270471033,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004986211034009008,
      "loss": 6.1506,
      "mean_token_accuracy": 0.14179377555847167,
      "num_tokens": 10144763.0,
      "step": 5000
    },
    {
      "entropy": 6.292379522323609,
      "epoch": 0.437385301057415,
      "grad_norm": 2.375,
      "learning_rate": 0.0004986176566790375,
      "loss": 6.1342,
      "mean_token_accuracy": 0.1395754799246788,
      "num_tokens": 10155011.0,
      "step": 5005
    },
    {
      "entropy": 6.2485833168029785,
      "epoch": 0.4378222494101197,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004986142056680701,
      "loss": 6.2109,
      "mean_token_accuracy": 0.1308772437274456,
      "num_tokens": 10165474.0,
      "step": 5010
    },
    {
      "entropy": 6.34594316482544,
      "epoch": 0.43825919776282446,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004986107503680652,
      "loss": 6.2349,
      "mean_token_accuracy": 0.13177637830376626,
      "num_tokens": 10175137.0,
      "step": 5015
    },
    {
      "entropy": 6.391843223571778,
      "epoch": 0.43869614611552915,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004986072907790889,
      "loss": 6.2546,
      "mean_token_accuracy": 0.12697769850492477,
      "num_tokens": 10185623.0,
      "step": 5020
    },
    {
      "entropy": 6.431057596206665,
      "epoch": 0.43913309446823384,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004986038269012076,
      "loss": 6.2052,
      "mean_token_accuracy": 0.13435077294707298,
      "num_tokens": 10195201.0,
      "step": 5025
    },
    {
      "entropy": 6.393267679214477,
      "epoch": 0.4395700428209386,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004986003587344877,
      "loss": 6.2374,
      "mean_token_accuracy": 0.13112057819962503,
      "num_tokens": 10205951.0,
      "step": 5030
    },
    {
      "entropy": 6.33199782371521,
      "epoch": 0.4400069911736433,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004985968862789957,
      "loss": 6.122,
      "mean_token_accuracy": 0.13634700551629067,
      "num_tokens": 10215665.0,
      "step": 5035
    },
    {
      "entropy": 6.311248540878296,
      "epoch": 0.44044393952634797,
      "grad_norm": 2.5,
      "learning_rate": 0.0004985934095347985,
      "loss": 6.1846,
      "mean_token_accuracy": 0.13659994825720786,
      "num_tokens": 10225037.0,
      "step": 5040
    },
    {
      "entropy": 6.28298134803772,
      "epoch": 0.4408808878790527,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004985899285019624,
      "loss": 6.1574,
      "mean_token_accuracy": 0.1360540434718132,
      "num_tokens": 10234581.0,
      "step": 5045
    },
    {
      "entropy": 6.256371593475341,
      "epoch": 0.4413178362317574,
      "grad_norm": 2.125,
      "learning_rate": 0.0004985864431805543,
      "loss": 6.082,
      "mean_token_accuracy": 0.1376258499920368,
      "num_tokens": 10243716.0,
      "step": 5050
    },
    {
      "entropy": 6.271577787399292,
      "epoch": 0.4417547845844621,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004985829535706411,
      "loss": 6.1668,
      "mean_token_accuracy": 0.13901599496603012,
      "num_tokens": 10253198.0,
      "step": 5055
    },
    {
      "entropy": 6.30764946937561,
      "epoch": 0.44219173293716685,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004985794596722897,
      "loss": 6.2539,
      "mean_token_accuracy": 0.12913227304816247,
      "num_tokens": 10262912.0,
      "step": 5060
    },
    {
      "entropy": 6.372983932495117,
      "epoch": 0.44262868128987154,
      "grad_norm": 2.234375,
      "learning_rate": 0.000498575961485567,
      "loss": 6.1868,
      "mean_token_accuracy": 0.13781487494707106,
      "num_tokens": 10272960.0,
      "step": 5065
    },
    {
      "entropy": 6.438020038604736,
      "epoch": 0.44306562964257623,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004985724590105403,
      "loss": 6.2518,
      "mean_token_accuracy": 0.1274810776114464,
      "num_tokens": 10283034.0,
      "step": 5070
    },
    {
      "entropy": 6.281031751632691,
      "epoch": 0.443502577995281,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004985689522472767,
      "loss": 6.2111,
      "mean_token_accuracy": 0.13171893507242202,
      "num_tokens": 10292553.0,
      "step": 5075
    },
    {
      "entropy": 6.276030969619751,
      "epoch": 0.44393952634798567,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004985654411958434,
      "loss": 6.1626,
      "mean_token_accuracy": 0.13330712243914605,
      "num_tokens": 10302475.0,
      "step": 5080
    },
    {
      "entropy": 6.3227907657623295,
      "epoch": 0.44437647470069036,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004985619258563078,
      "loss": 6.1629,
      "mean_token_accuracy": 0.13461405634880066,
      "num_tokens": 10311693.0,
      "step": 5085
    },
    {
      "entropy": 6.348954439163208,
      "epoch": 0.4448134230533951,
      "grad_norm": 2.5,
      "learning_rate": 0.0004985584062287372,
      "loss": 6.1709,
      "mean_token_accuracy": 0.137611423432827,
      "num_tokens": 10323052.0,
      "step": 5090
    },
    {
      "entropy": 6.2841047763824465,
      "epoch": 0.4452503714060998,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004985548823131994,
      "loss": 6.1634,
      "mean_token_accuracy": 0.13897646069526673,
      "num_tokens": 10332997.0,
      "step": 5095
    },
    {
      "entropy": 6.34373288154602,
      "epoch": 0.4456873197588045,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004985513541097618,
      "loss": 6.2678,
      "mean_token_accuracy": 0.13269016444683074,
      "num_tokens": 10343298.0,
      "step": 5100
    },
    {
      "entropy": 6.440412282943726,
      "epoch": 0.44612426811150924,
      "grad_norm": 6.15625,
      "learning_rate": 0.000498547821618492,
      "loss": 6.1717,
      "mean_token_accuracy": 0.1324860617518425,
      "num_tokens": 10352485.0,
      "step": 5105
    },
    {
      "entropy": 6.2851152420043945,
      "epoch": 0.44656121646421393,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004985442848394579,
      "loss": 6.2224,
      "mean_token_accuracy": 0.1322377398610115,
      "num_tokens": 10362810.0,
      "step": 5110
    },
    {
      "entropy": 6.300913715362549,
      "epoch": 0.4469981648169186,
      "grad_norm": 2.625,
      "learning_rate": 0.0004985407437727273,
      "loss": 6.1552,
      "mean_token_accuracy": 0.13889532908797264,
      "num_tokens": 10372817.0,
      "step": 5115
    },
    {
      "entropy": 6.364538240432739,
      "epoch": 0.44743511316962337,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004985371984183681,
      "loss": 6.2031,
      "mean_token_accuracy": 0.13284733667969703,
      "num_tokens": 10383046.0,
      "step": 5120
    },
    {
      "entropy": 6.3004148483276365,
      "epoch": 0.44787206152232806,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004985336487764483,
      "loss": 6.2286,
      "mean_token_accuracy": 0.12916007712483407,
      "num_tokens": 10393498.0,
      "step": 5125
    },
    {
      "entropy": 6.2739440441131595,
      "epoch": 0.44830900987503275,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004985300948470359,
      "loss": 6.2289,
      "mean_token_accuracy": 0.13032302632927895,
      "num_tokens": 10403595.0,
      "step": 5130
    },
    {
      "entropy": 6.306627941131592,
      "epoch": 0.4487459582277375,
      "grad_norm": 2.0,
      "learning_rate": 0.0004985265366301992,
      "loss": 6.1391,
      "mean_token_accuracy": 0.13291553109884263,
      "num_tokens": 10413475.0,
      "step": 5135
    },
    {
      "entropy": 6.338770961761474,
      "epoch": 0.4491829065804422,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004985229741260065,
      "loss": 6.1999,
      "mean_token_accuracy": 0.13361337631940842,
      "num_tokens": 10423512.0,
      "step": 5140
    },
    {
      "entropy": 6.328917694091797,
      "epoch": 0.4496198549331469,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004985194073345259,
      "loss": 6.2122,
      "mean_token_accuracy": 0.1277785651385784,
      "num_tokens": 10434198.0,
      "step": 5145
    },
    {
      "entropy": 6.311338138580322,
      "epoch": 0.4500568032858516,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004985158362558261,
      "loss": 6.1418,
      "mean_token_accuracy": 0.13038049191236495,
      "num_tokens": 10444077.0,
      "step": 5150
    },
    {
      "entropy": 6.301530551910401,
      "epoch": 0.4504937516385563,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004985122608899752,
      "loss": 6.1973,
      "mean_token_accuracy": 0.13268831595778466,
      "num_tokens": 10454481.0,
      "step": 5155
    },
    {
      "entropy": 6.357873582839966,
      "epoch": 0.450930699991261,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004985086812370422,
      "loss": 6.1964,
      "mean_token_accuracy": 0.13561240211129189,
      "num_tokens": 10464453.0,
      "step": 5160
    },
    {
      "entropy": 6.399531555175781,
      "epoch": 0.45136764834396575,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004985050972970956,
      "loss": 6.2809,
      "mean_token_accuracy": 0.1275574430823326,
      "num_tokens": 10474932.0,
      "step": 5165
    },
    {
      "entropy": 6.3716878414154055,
      "epoch": 0.45180459669667045,
      "grad_norm": 2.25,
      "learning_rate": 0.000498501509070204,
      "loss": 6.2712,
      "mean_token_accuracy": 0.1266408421099186,
      "num_tokens": 10484719.0,
      "step": 5170
    },
    {
      "entropy": 6.365037822723389,
      "epoch": 0.45224154504937514,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004984979165564364,
      "loss": 6.3363,
      "mean_token_accuracy": 0.12200213298201561,
      "num_tokens": 10495808.0,
      "step": 5175
    },
    {
      "entropy": 6.379946517944336,
      "epoch": 0.4526784934020799,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004984943197558616,
      "loss": 6.2041,
      "mean_token_accuracy": 0.13217753320932388,
      "num_tokens": 10506528.0,
      "step": 5180
    },
    {
      "entropy": 6.386626052856445,
      "epoch": 0.4531154417547846,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004984907186685487,
      "loss": 6.1934,
      "mean_token_accuracy": 0.13962555974721907,
      "num_tokens": 10517117.0,
      "step": 5185
    },
    {
      "entropy": 6.403228855133056,
      "epoch": 0.45355239010748927,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004984871132945667,
      "loss": 6.1919,
      "mean_token_accuracy": 0.1314509056508541,
      "num_tokens": 10527394.0,
      "step": 5190
    },
    {
      "entropy": 6.286072683334351,
      "epoch": 0.453989338460194,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004984835036339848,
      "loss": 6.1802,
      "mean_token_accuracy": 0.13925824984908103,
      "num_tokens": 10536779.0,
      "step": 5195
    },
    {
      "entropy": 6.300886631011963,
      "epoch": 0.4544262868128987,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004984798896868722,
      "loss": 6.2658,
      "mean_token_accuracy": 0.1270495541393757,
      "num_tokens": 10547116.0,
      "step": 5200
    },
    {
      "entropy": 6.323485946655273,
      "epoch": 0.45486323516560345,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004984762714532981,
      "loss": 6.1365,
      "mean_token_accuracy": 0.13762983307242393,
      "num_tokens": 10556396.0,
      "step": 5205
    },
    {
      "entropy": 6.350305223464966,
      "epoch": 0.45530018351830814,
      "grad_norm": 2.53125,
      "learning_rate": 0.000498472648933332,
      "loss": 6.3255,
      "mean_token_accuracy": 0.1301370792090893,
      "num_tokens": 10568127.0,
      "step": 5210
    },
    {
      "entropy": 6.379745674133301,
      "epoch": 0.45573713187101283,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004984690221270436,
      "loss": 6.1786,
      "mean_token_accuracy": 0.13736633732914924,
      "num_tokens": 10578565.0,
      "step": 5215
    },
    {
      "entropy": 6.385982942581177,
      "epoch": 0.4561740802237176,
      "grad_norm": 2.625,
      "learning_rate": 0.0004984653910345021,
      "loss": 6.2013,
      "mean_token_accuracy": 0.1263721540570259,
      "num_tokens": 10588633.0,
      "step": 5220
    },
    {
      "entropy": 6.334294176101684,
      "epoch": 0.4566110285764223,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004984617556557774,
      "loss": 6.1692,
      "mean_token_accuracy": 0.13824631050229072,
      "num_tokens": 10597917.0,
      "step": 5225
    },
    {
      "entropy": 6.390229654312134,
      "epoch": 0.45704797692912696,
      "grad_norm": 3.15625,
      "learning_rate": 0.000498458115990939,
      "loss": 6.2051,
      "mean_token_accuracy": 0.12898601666092874,
      "num_tokens": 10608396.0,
      "step": 5230
    },
    {
      "entropy": 6.364322996139526,
      "epoch": 0.4574849252818317,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004984544720400568,
      "loss": 6.2151,
      "mean_token_accuracy": 0.13011696860194205,
      "num_tokens": 10618191.0,
      "step": 5235
    },
    {
      "entropy": 6.351131725311279,
      "epoch": 0.4579218736345364,
      "grad_norm": 2.828125,
      "learning_rate": 0.000498450823803201,
      "loss": 6.1956,
      "mean_token_accuracy": 0.12906027510762214,
      "num_tokens": 10627594.0,
      "step": 5240
    },
    {
      "entropy": 6.312578296661377,
      "epoch": 0.4583588219872411,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004984471712804412,
      "loss": 6.1777,
      "mean_token_accuracy": 0.13861244320869445,
      "num_tokens": 10637260.0,
      "step": 5245
    },
    {
      "entropy": 6.344043016433716,
      "epoch": 0.45879577033994584,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004984435144718475,
      "loss": 6.2064,
      "mean_token_accuracy": 0.13199804052710534,
      "num_tokens": 10647968.0,
      "step": 5250
    },
    {
      "entropy": 6.339229726791382,
      "epoch": 0.45923271869265053,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004984398533774901,
      "loss": 6.1677,
      "mean_token_accuracy": 0.13596777692437173,
      "num_tokens": 10657760.0,
      "step": 5255
    },
    {
      "entropy": 6.365975046157837,
      "epoch": 0.4596696670453552,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004984361879974393,
      "loss": 6.1914,
      "mean_token_accuracy": 0.1301260285079479,
      "num_tokens": 10667577.0,
      "step": 5260
    },
    {
      "entropy": 6.365857696533203,
      "epoch": 0.46010661539805997,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004984325183317654,
      "loss": 6.1568,
      "mean_token_accuracy": 0.130755764991045,
      "num_tokens": 10677984.0,
      "step": 5265
    },
    {
      "entropy": 6.308723402023316,
      "epoch": 0.46054356375076466,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004984288443805386,
      "loss": 6.2005,
      "mean_token_accuracy": 0.1354597106575966,
      "num_tokens": 10687700.0,
      "step": 5270
    },
    {
      "entropy": 6.301192712783814,
      "epoch": 0.46098051210346935,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004984251661438295,
      "loss": 6.1775,
      "mean_token_accuracy": 0.13819514736533164,
      "num_tokens": 10697351.0,
      "step": 5275
    },
    {
      "entropy": 6.317161750793457,
      "epoch": 0.4614174604561741,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004984214836217087,
      "loss": 6.1829,
      "mean_token_accuracy": 0.13575254157185554,
      "num_tokens": 10707498.0,
      "step": 5280
    },
    {
      "entropy": 6.309888744354248,
      "epoch": 0.4618544088088788,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004984177968142467,
      "loss": 6.234,
      "mean_token_accuracy": 0.1295247919857502,
      "num_tokens": 10717724.0,
      "step": 5285
    },
    {
      "entropy": 6.365609979629516,
      "epoch": 0.4622913571615835,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004984141057215143,
      "loss": 6.186,
      "mean_token_accuracy": 0.13355940878391265,
      "num_tokens": 10727030.0,
      "step": 5290
    },
    {
      "entropy": 6.307609176635742,
      "epoch": 0.46272830551428823,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004984104103435824,
      "loss": 6.1141,
      "mean_token_accuracy": 0.13749504536390306,
      "num_tokens": 10736861.0,
      "step": 5295
    },
    {
      "entropy": 6.395886850357056,
      "epoch": 0.4631652538669929,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004984067106805217,
      "loss": 6.2836,
      "mean_token_accuracy": 0.1279846616089344,
      "num_tokens": 10748310.0,
      "step": 5300
    },
    {
      "entropy": 6.370503568649292,
      "epoch": 0.4636022022196976,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004984030067324032,
      "loss": 6.1226,
      "mean_token_accuracy": 0.13517538011074065,
      "num_tokens": 10758110.0,
      "step": 5305
    },
    {
      "entropy": 6.285466575622559,
      "epoch": 0.46403915057240236,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004983992984992979,
      "loss": 6.1879,
      "mean_token_accuracy": 0.13820246160030364,
      "num_tokens": 10767919.0,
      "step": 5310
    },
    {
      "entropy": 6.334744119644165,
      "epoch": 0.46447609892510705,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004983955859812771,
      "loss": 6.1145,
      "mean_token_accuracy": 0.13637230396270753,
      "num_tokens": 10776888.0,
      "step": 5315
    },
    {
      "entropy": 6.329569911956787,
      "epoch": 0.46491304727781174,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004983918691784119,
      "loss": 6.1812,
      "mean_token_accuracy": 0.13182662799954414,
      "num_tokens": 10786806.0,
      "step": 5320
    },
    {
      "entropy": 6.315571355819702,
      "epoch": 0.4653499956305165,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004983881480907736,
      "loss": 6.163,
      "mean_token_accuracy": 0.13472113385796547,
      "num_tokens": 10796452.0,
      "step": 5325
    },
    {
      "entropy": 6.3357062339782715,
      "epoch": 0.4657869439832212,
      "grad_norm": 4.125,
      "learning_rate": 0.0004983844227184334,
      "loss": 6.1453,
      "mean_token_accuracy": 0.1334402397274971,
      "num_tokens": 10806718.0,
      "step": 5330
    },
    {
      "entropy": 6.3808036804199215,
      "epoch": 0.46622389233592587,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004983806930614631,
      "loss": 6.253,
      "mean_token_accuracy": 0.12785169035196303,
      "num_tokens": 10817978.0,
      "step": 5335
    },
    {
      "entropy": 6.343908977508545,
      "epoch": 0.4666608406886306,
      "grad_norm": 2.515625,
      "learning_rate": 0.000498376959119934,
      "loss": 6.1145,
      "mean_token_accuracy": 0.14180885255336761,
      "num_tokens": 10827818.0,
      "step": 5340
    },
    {
      "entropy": 6.30909833908081,
      "epoch": 0.4670977890413353,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004983732208939178,
      "loss": 6.2746,
      "mean_token_accuracy": 0.1307710699737072,
      "num_tokens": 10838492.0,
      "step": 5345
    },
    {
      "entropy": 6.330020904541016,
      "epoch": 0.46753473739404,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004983694783834863,
      "loss": 6.1281,
      "mean_token_accuracy": 0.1398046277463436,
      "num_tokens": 10847807.0,
      "step": 5350
    },
    {
      "entropy": 6.315368604660034,
      "epoch": 0.46797168574674475,
      "grad_norm": 3.375,
      "learning_rate": 0.0004983657315887109,
      "loss": 6.2387,
      "mean_token_accuracy": 0.1277824804186821,
      "num_tokens": 10858637.0,
      "step": 5355
    },
    {
      "entropy": 6.3383056163787845,
      "epoch": 0.46840863409944944,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004983619805096639,
      "loss": 6.1821,
      "mean_token_accuracy": 0.13585800677537918,
      "num_tokens": 10868042.0,
      "step": 5360
    },
    {
      "entropy": 6.390500259399414,
      "epoch": 0.46884558245215413,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004983582251464171,
      "loss": 6.1779,
      "mean_token_accuracy": 0.135035040974617,
      "num_tokens": 10877171.0,
      "step": 5365
    },
    {
      "entropy": 6.274995994567871,
      "epoch": 0.4692825308048589,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004983544654990423,
      "loss": 6.1792,
      "mean_token_accuracy": 0.1332053028047085,
      "num_tokens": 10888434.0,
      "step": 5370
    },
    {
      "entropy": 6.36237645149231,
      "epoch": 0.46971947915756357,
      "grad_norm": 2.203125,
      "learning_rate": 0.000498350701567612,
      "loss": 6.2336,
      "mean_token_accuracy": 0.1295624278485775,
      "num_tokens": 10898467.0,
      "step": 5375
    },
    {
      "entropy": 6.360772275924683,
      "epoch": 0.47015642751026826,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004983469333521983,
      "loss": 6.1991,
      "mean_token_accuracy": 0.13911381512880325,
      "num_tokens": 10908178.0,
      "step": 5380
    },
    {
      "entropy": 6.218901824951172,
      "epoch": 0.470593375862973,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004983431608528732,
      "loss": 6.2692,
      "mean_token_accuracy": 0.12432105541229248,
      "num_tokens": 10919957.0,
      "step": 5385
    },
    {
      "entropy": 6.373246669769287,
      "epoch": 0.4710303242156777,
      "grad_norm": 3.125,
      "learning_rate": 0.0004983393840697092,
      "loss": 6.2107,
      "mean_token_accuracy": 0.13679133504629135,
      "num_tokens": 10929728.0,
      "step": 5390
    },
    {
      "entropy": 6.334309816360474,
      "epoch": 0.4714672725683824,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004983356030027789,
      "loss": 6.2371,
      "mean_token_accuracy": 0.13146899715065957,
      "num_tokens": 10940427.0,
      "step": 5395
    },
    {
      "entropy": 6.282358932495117,
      "epoch": 0.47190422092108714,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004983318176521546,
      "loss": 6.2177,
      "mean_token_accuracy": 0.13522961586713791,
      "num_tokens": 10950641.0,
      "step": 5400
    },
    {
      "entropy": 6.423878717422485,
      "epoch": 0.4723411692737918,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004983280280179092,
      "loss": 6.1448,
      "mean_token_accuracy": 0.1371025525033474,
      "num_tokens": 10959341.0,
      "step": 5405
    },
    {
      "entropy": 6.356134843826294,
      "epoch": 0.4727781176264966,
      "grad_norm": 2.15625,
      "learning_rate": 0.000498324234100115,
      "loss": 6.1632,
      "mean_token_accuracy": 0.13320271000266076,
      "num_tokens": 10967745.0,
      "step": 5410
    },
    {
      "entropy": 6.232340669631958,
      "epoch": 0.47321506597920127,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004983204358988451,
      "loss": 6.1552,
      "mean_token_accuracy": 0.1321762166917324,
      "num_tokens": 10978345.0,
      "step": 5415
    },
    {
      "entropy": 6.312339735031128,
      "epoch": 0.47365201433190596,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004983166334141722,
      "loss": 6.2139,
      "mean_token_accuracy": 0.1294349268078804,
      "num_tokens": 10988030.0,
      "step": 5420
    },
    {
      "entropy": 6.424719047546387,
      "epoch": 0.4740889626846107,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004983128266461693,
      "loss": 6.1958,
      "mean_token_accuracy": 0.12833590060472488,
      "num_tokens": 10999169.0,
      "step": 5425
    },
    {
      "entropy": 6.305271911621094,
      "epoch": 0.4745259110373154,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004983090155949093,
      "loss": 6.2676,
      "mean_token_accuracy": 0.12454533725976943,
      "num_tokens": 11009608.0,
      "step": 5430
    },
    {
      "entropy": 6.325997018814087,
      "epoch": 0.4749628593900201,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004983052002604653,
      "loss": 6.2095,
      "mean_token_accuracy": 0.129917711019516,
      "num_tokens": 11020187.0,
      "step": 5435
    },
    {
      "entropy": 6.3779693126678465,
      "epoch": 0.47539980774272483,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004983013806429107,
      "loss": 6.1842,
      "mean_token_accuracy": 0.1343964971601963,
      "num_tokens": 11030459.0,
      "step": 5440
    },
    {
      "entropy": 6.256487226486206,
      "epoch": 0.4758367560954295,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004982975567423185,
      "loss": 6.1222,
      "mean_token_accuracy": 0.13884098082780838,
      "num_tokens": 11040880.0,
      "step": 5445
    },
    {
      "entropy": 6.26084041595459,
      "epoch": 0.4762737044481342,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004982937285587623,
      "loss": 6.1563,
      "mean_token_accuracy": 0.14116177335381508,
      "num_tokens": 11050901.0,
      "step": 5450
    },
    {
      "entropy": 6.377487134933472,
      "epoch": 0.47671065280083896,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004982898960923152,
      "loss": 6.3116,
      "mean_token_accuracy": 0.12609442695975304,
      "num_tokens": 11061274.0,
      "step": 5455
    },
    {
      "entropy": 6.331740617752075,
      "epoch": 0.47714760115354365,
      "grad_norm": 2.125,
      "learning_rate": 0.0004982860593430509,
      "loss": 6.1176,
      "mean_token_accuracy": 0.129807560890913,
      "num_tokens": 11070644.0,
      "step": 5460
    },
    {
      "entropy": 6.303116273880005,
      "epoch": 0.47758454950624835,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004982822183110431,
      "loss": 6.1807,
      "mean_token_accuracy": 0.13428399190306664,
      "num_tokens": 11081154.0,
      "step": 5465
    },
    {
      "entropy": 6.374970769882202,
      "epoch": 0.4780214978589531,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004982783729963654,
      "loss": 6.2508,
      "mean_token_accuracy": 0.12831852883100509,
      "num_tokens": 11091929.0,
      "step": 5470
    },
    {
      "entropy": 6.3436531066894535,
      "epoch": 0.4784584462116578,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004982745233990914,
      "loss": 6.1183,
      "mean_token_accuracy": 0.14144821017980574,
      "num_tokens": 11102075.0,
      "step": 5475
    },
    {
      "entropy": 6.272120141983033,
      "epoch": 0.4788953945643625,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004982706695192949,
      "loss": 6.206,
      "mean_token_accuracy": 0.13271668031811715,
      "num_tokens": 11111886.0,
      "step": 5480
    },
    {
      "entropy": 6.376845359802246,
      "epoch": 0.4793323429170672,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004982668113570503,
      "loss": 6.1928,
      "mean_token_accuracy": 0.12939932346343994,
      "num_tokens": 11122329.0,
      "step": 5485
    },
    {
      "entropy": 6.308321857452393,
      "epoch": 0.4797692912697719,
      "grad_norm": 2.953125,
      "learning_rate": 0.000498262948912431,
      "loss": 6.1139,
      "mean_token_accuracy": 0.132107026129961,
      "num_tokens": 11132348.0,
      "step": 5490
    },
    {
      "entropy": 6.265446043014526,
      "epoch": 0.4802062396224766,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004982590821855115,
      "loss": 6.193,
      "mean_token_accuracy": 0.13845708966255188,
      "num_tokens": 11141975.0,
      "step": 5495
    },
    {
      "entropy": 6.322985982894897,
      "epoch": 0.48064318797518135,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004982552111763657,
      "loss": 6.2016,
      "mean_token_accuracy": 0.13201960399746895,
      "num_tokens": 11152174.0,
      "step": 5500
    },
    {
      "entropy": 6.361165475845337,
      "epoch": 0.48108013632788604,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004982513358850681,
      "loss": 6.2007,
      "mean_token_accuracy": 0.12985819950699806,
      "num_tokens": 11162563.0,
      "step": 5505
    },
    {
      "entropy": 6.306317663192749,
      "epoch": 0.48151708468059073,
      "grad_norm": 2.0,
      "learning_rate": 0.0004982474563116928,
      "loss": 6.1319,
      "mean_token_accuracy": 0.13499016240239142,
      "num_tokens": 11172572.0,
      "step": 5510
    },
    {
      "entropy": 6.278728675842285,
      "epoch": 0.4819540330332955,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004982435724563143,
      "loss": 6.1907,
      "mean_token_accuracy": 0.13621274530887603,
      "num_tokens": 11183404.0,
      "step": 5515
    },
    {
      "entropy": 6.291526937484742,
      "epoch": 0.4823909813860002,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004982396843190072,
      "loss": 6.1892,
      "mean_token_accuracy": 0.13114827796816825,
      "num_tokens": 11193905.0,
      "step": 5520
    },
    {
      "entropy": 6.37049241065979,
      "epoch": 0.48282792973870486,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004982357918998459,
      "loss": 6.1019,
      "mean_token_accuracy": 0.1481143280863762,
      "num_tokens": 11203127.0,
      "step": 5525
    },
    {
      "entropy": 6.353377819061279,
      "epoch": 0.4832648780914096,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004982318951989053,
      "loss": 6.139,
      "mean_token_accuracy": 0.13807440996170045,
      "num_tokens": 11213468.0,
      "step": 5530
    },
    {
      "entropy": 6.267552280426026,
      "epoch": 0.4837018264441143,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004982279942162597,
      "loss": 6.2151,
      "mean_token_accuracy": 0.1353520266711712,
      "num_tokens": 11224722.0,
      "step": 5535
    },
    {
      "entropy": 6.29327712059021,
      "epoch": 0.484138774796819,
      "grad_norm": 2.875,
      "learning_rate": 0.0004982240889519844,
      "loss": 6.1887,
      "mean_token_accuracy": 0.13441331386566163,
      "num_tokens": 11234565.0,
      "step": 5540
    },
    {
      "entropy": 6.319821119308472,
      "epoch": 0.48457572314952374,
      "grad_norm": 2.296875,
      "learning_rate": 0.000498220179406154,
      "loss": 6.2136,
      "mean_token_accuracy": 0.13793121948838233,
      "num_tokens": 11244496.0,
      "step": 5545
    },
    {
      "entropy": 6.424495363235474,
      "epoch": 0.48501267150222843,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004982162655788436,
      "loss": 6.1513,
      "mean_token_accuracy": 0.13436891064047812,
      "num_tokens": 11254187.0,
      "step": 5550
    },
    {
      "entropy": 6.298796510696411,
      "epoch": 0.4854496198549331,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004982123474701283,
      "loss": 6.1266,
      "mean_token_accuracy": 0.14183791875839233,
      "num_tokens": 11264580.0,
      "step": 5555
    },
    {
      "entropy": 6.247805261611939,
      "epoch": 0.48588656820763787,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004982084250800832,
      "loss": 6.2717,
      "mean_token_accuracy": 0.13148501366376877,
      "num_tokens": 11275104.0,
      "step": 5560
    },
    {
      "entropy": 6.427922487258911,
      "epoch": 0.48632351656034256,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004982044984087836,
      "loss": 6.2906,
      "mean_token_accuracy": 0.12793218344449997,
      "num_tokens": 11285983.0,
      "step": 5565
    },
    {
      "entropy": 6.38091344833374,
      "epoch": 0.48676046491304725,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004982005674563048,
      "loss": 6.1651,
      "mean_token_accuracy": 0.13897799775004388,
      "num_tokens": 11295378.0,
      "step": 5570
    },
    {
      "entropy": 6.317071628570557,
      "epoch": 0.487197413265752,
      "grad_norm": 3.09375,
      "learning_rate": 0.000498196632222722,
      "loss": 6.2856,
      "mean_token_accuracy": 0.12998063266277313,
      "num_tokens": 11305500.0,
      "step": 5575
    },
    {
      "entropy": 6.280082845687867,
      "epoch": 0.4876343616184567,
      "grad_norm": 6.375,
      "learning_rate": 0.0004981926927081109,
      "loss": 6.1958,
      "mean_token_accuracy": 0.130854519456625,
      "num_tokens": 11316455.0,
      "step": 5580
    },
    {
      "entropy": 6.307268714904785,
      "epoch": 0.4880713099711614,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004981887489125471,
      "loss": 6.1491,
      "mean_token_accuracy": 0.12794427946209908,
      "num_tokens": 11325865.0,
      "step": 5585
    },
    {
      "entropy": 6.273912572860718,
      "epoch": 0.48850825832386613,
      "grad_norm": 2.90625,
      "learning_rate": 0.000498184800836106,
      "loss": 6.1234,
      "mean_token_accuracy": 0.15140984952449799,
      "num_tokens": 11337368.0,
      "step": 5590
    },
    {
      "entropy": 6.320558404922485,
      "epoch": 0.4889452066765708,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004981808484788635,
      "loss": 6.1271,
      "mean_token_accuracy": 0.14346376284956933,
      "num_tokens": 11347129.0,
      "step": 5595
    },
    {
      "entropy": 6.300006437301636,
      "epoch": 0.48938215502927557,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004981768918408955,
      "loss": 6.1666,
      "mean_token_accuracy": 0.13408696353435517,
      "num_tokens": 11356641.0,
      "step": 5600
    },
    {
      "entropy": 6.256478977203369,
      "epoch": 0.48981910338198026,
      "grad_norm": 2.25,
      "learning_rate": 0.0004981729309222777,
      "loss": 6.0425,
      "mean_token_accuracy": 0.13872768878936767,
      "num_tokens": 11366485.0,
      "step": 5605
    },
    {
      "entropy": 6.2726397037506105,
      "epoch": 0.49025605173468495,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004981689657230861,
      "loss": 6.2155,
      "mean_token_accuracy": 0.13545479625463486,
      "num_tokens": 11376305.0,
      "step": 5610
    },
    {
      "entropy": 6.332991027832032,
      "epoch": 0.4906930000873897,
      "grad_norm": 2.8125,
      "learning_rate": 0.000498164996243397,
      "loss": 6.1836,
      "mean_token_accuracy": 0.13017890974879265,
      "num_tokens": 11385837.0,
      "step": 5615
    },
    {
      "entropy": 6.277083444595337,
      "epoch": 0.4911299484400944,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004981610224832861,
      "loss": 6.2329,
      "mean_token_accuracy": 0.13513108268380164,
      "num_tokens": 11396719.0,
      "step": 5620
    },
    {
      "entropy": 6.3372125148773195,
      "epoch": 0.4915668967927991,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004981570444428299,
      "loss": 6.2329,
      "mean_token_accuracy": 0.13215182572603226,
      "num_tokens": 11408107.0,
      "step": 5625
    },
    {
      "entropy": 6.320571231842041,
      "epoch": 0.4920038451455038,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004981530621221048,
      "loss": 6.1596,
      "mean_token_accuracy": 0.14050165489315986,
      "num_tokens": 11417547.0,
      "step": 5630
    },
    {
      "entropy": 6.34918212890625,
      "epoch": 0.4924407934982085,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004981490755211869,
      "loss": 6.2037,
      "mean_token_accuracy": 0.1393429346382618,
      "num_tokens": 11427660.0,
      "step": 5635
    },
    {
      "entropy": 6.321030855178833,
      "epoch": 0.4928777418509132,
      "grad_norm": 10.0625,
      "learning_rate": 0.0004981450846401528,
      "loss": 6.1013,
      "mean_token_accuracy": 0.13890570998191834,
      "num_tokens": 11438032.0,
      "step": 5640
    },
    {
      "entropy": 6.337495899200439,
      "epoch": 0.49331469020361796,
      "grad_norm": 2.46875,
      "learning_rate": 0.000498141089479079,
      "loss": 6.2008,
      "mean_token_accuracy": 0.13348471969366074,
      "num_tokens": 11448987.0,
      "step": 5645
    },
    {
      "entropy": 6.30088472366333,
      "epoch": 0.49375163855632265,
      "grad_norm": 8.0625,
      "learning_rate": 0.0004981370900380423,
      "loss": 6.1875,
      "mean_token_accuracy": 0.13803953304886818,
      "num_tokens": 11458659.0,
      "step": 5650
    },
    {
      "entropy": 6.353876447677612,
      "epoch": 0.49418858690902734,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004981330863171194,
      "loss": 6.1478,
      "mean_token_accuracy": 0.13499215990304947,
      "num_tokens": 11468104.0,
      "step": 5655
    },
    {
      "entropy": 6.306606817245483,
      "epoch": 0.4946255352617321,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004981290783163867,
      "loss": 6.1219,
      "mean_token_accuracy": 0.14236722737550736,
      "num_tokens": 11479579.0,
      "step": 5660
    },
    {
      "entropy": 6.322608280181885,
      "epoch": 0.4950624836144368,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004981250660359216,
      "loss": 6.1754,
      "mean_token_accuracy": 0.13996674194931985,
      "num_tokens": 11489202.0,
      "step": 5665
    },
    {
      "entropy": 6.355022048950195,
      "epoch": 0.49549943196714147,
      "grad_norm": 4.375,
      "learning_rate": 0.0004981210494758007,
      "loss": 6.213,
      "mean_token_accuracy": 0.12901572212576867,
      "num_tokens": 11499425.0,
      "step": 5670
    },
    {
      "entropy": 6.3729571342468265,
      "epoch": 0.4959363803198462,
      "grad_norm": 3.078125,
      "learning_rate": 0.000498117028636101,
      "loss": 6.1686,
      "mean_token_accuracy": 0.13260312229394913,
      "num_tokens": 11508983.0,
      "step": 5675
    },
    {
      "entropy": 6.37848482131958,
      "epoch": 0.4963733286725509,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004981130035169,
      "loss": 6.2376,
      "mean_token_accuracy": 0.13701794296503067,
      "num_tokens": 11519791.0,
      "step": 5680
    },
    {
      "entropy": 6.326758193969726,
      "epoch": 0.4968102770252556,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004981089741182747,
      "loss": 6.1668,
      "mean_token_accuracy": 0.1366004914045334,
      "num_tokens": 11530537.0,
      "step": 5685
    },
    {
      "entropy": 6.322508049011231,
      "epoch": 0.49724722537796034,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004981049404403023,
      "loss": 6.1955,
      "mean_token_accuracy": 0.13650267720222473,
      "num_tokens": 11540591.0,
      "step": 5690
    },
    {
      "entropy": 6.31669602394104,
      "epoch": 0.49768417373066504,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004981009024830602,
      "loss": 6.1336,
      "mean_token_accuracy": 0.13225832656025888,
      "num_tokens": 11550719.0,
      "step": 5695
    },
    {
      "entropy": 6.237813186645508,
      "epoch": 0.4981211220833697,
      "grad_norm": 2.671875,
      "learning_rate": 0.000498096860246626,
      "loss": 6.1833,
      "mean_token_accuracy": 0.13999730348587036,
      "num_tokens": 11561805.0,
      "step": 5700
    },
    {
      "entropy": 6.414583778381347,
      "epoch": 0.4985580704360745,
      "grad_norm": 2.734375,
      "learning_rate": 0.000498092813731077,
      "loss": 6.2152,
      "mean_token_accuracy": 0.1306938074529171,
      "num_tokens": 11571380.0,
      "step": 5705
    },
    {
      "entropy": 6.382923650741577,
      "epoch": 0.49899501878877917,
      "grad_norm": 3.3125,
      "learning_rate": 0.000498088762936491,
      "loss": 6.2112,
      "mean_token_accuracy": 0.13316197246313094,
      "num_tokens": 11580853.0,
      "step": 5710
    },
    {
      "entropy": 6.152510786056519,
      "epoch": 0.49943196714148386,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004980847078629457,
      "loss": 6.111,
      "mean_token_accuracy": 0.15095606520771981,
      "num_tokens": 11591966.0,
      "step": 5715
    },
    {
      "entropy": 6.2414463520050045,
      "epoch": 0.4998689154941886,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004980806485105186,
      "loss": 6.1539,
      "mean_token_accuracy": 0.13871203660964965,
      "num_tokens": 11601239.0,
      "step": 5720
    },
    {
      "entropy": 6.321985960006714,
      "epoch": 0.5003058638468934,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004980765848792879,
      "loss": 6.0866,
      "mean_token_accuracy": 0.1453379437327385,
      "num_tokens": 11609857.0,
      "step": 5725
    },
    {
      "entropy": 6.363383865356445,
      "epoch": 0.500742812199598,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004980725169693314,
      "loss": 6.1483,
      "mean_token_accuracy": 0.1401673175394535,
      "num_tokens": 11618952.0,
      "step": 5730
    },
    {
      "entropy": 6.3067420482635494,
      "epoch": 0.5011797605523027,
      "grad_norm": 5.53125,
      "learning_rate": 0.0004980684447807272,
      "loss": 6.0849,
      "mean_token_accuracy": 0.13827861100435257,
      "num_tokens": 11627638.0,
      "step": 5735
    },
    {
      "entropy": 6.26917839050293,
      "epoch": 0.5016167089050074,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004980643683135533,
      "loss": 6.2234,
      "mean_token_accuracy": 0.13652559220790864,
      "num_tokens": 11637842.0,
      "step": 5740
    },
    {
      "entropy": 6.344669914245605,
      "epoch": 0.5020536572577121,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004980602875678878,
      "loss": 6.0766,
      "mean_token_accuracy": 0.1468154013156891,
      "num_tokens": 11647217.0,
      "step": 5745
    },
    {
      "entropy": 6.2920269012451175,
      "epoch": 0.5024906056104168,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004980562025438093,
      "loss": 6.099,
      "mean_token_accuracy": 0.13941217660903932,
      "num_tokens": 11657558.0,
      "step": 5750
    },
    {
      "entropy": 6.24279375076294,
      "epoch": 0.5029275539631216,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004980521132413958,
      "loss": 6.1255,
      "mean_token_accuracy": 0.1355189189314842,
      "num_tokens": 11668054.0,
      "step": 5755
    },
    {
      "entropy": 6.309354162216186,
      "epoch": 0.5033645023158263,
      "grad_norm": 2.765625,
      "learning_rate": 0.000498048019660726,
      "loss": 6.1447,
      "mean_token_accuracy": 0.13731754645705224,
      "num_tokens": 11677779.0,
      "step": 5760
    },
    {
      "entropy": 6.3172831535339355,
      "epoch": 0.503801450668531,
      "grad_norm": 3.5,
      "learning_rate": 0.0004980439218018783,
      "loss": 6.1887,
      "mean_token_accuracy": 0.1342960625886917,
      "num_tokens": 11687577.0,
      "step": 5765
    },
    {
      "entropy": 6.359254550933838,
      "epoch": 0.5042383990212357,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004980398196649312,
      "loss": 6.2603,
      "mean_token_accuracy": 0.1296543449163437,
      "num_tokens": 11699630.0,
      "step": 5770
    },
    {
      "entropy": 6.343809318542481,
      "epoch": 0.5046753473739404,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004980357132499636,
      "loss": 6.2129,
      "mean_token_accuracy": 0.13326307982206345,
      "num_tokens": 11710139.0,
      "step": 5775
    },
    {
      "entropy": 6.323934555053711,
      "epoch": 0.5051122957266451,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004980316025570542,
      "loss": 6.163,
      "mean_token_accuracy": 0.13800614327192307,
      "num_tokens": 11719070.0,
      "step": 5780
    },
    {
      "entropy": 6.3309252738952635,
      "epoch": 0.5055492440793499,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004980274875862816,
      "loss": 6.1964,
      "mean_token_accuracy": 0.13725585490465164,
      "num_tokens": 11728770.0,
      "step": 5785
    },
    {
      "entropy": 6.378334474563599,
      "epoch": 0.5059861924320546,
      "grad_norm": 2.328125,
      "learning_rate": 0.000498023368337725,
      "loss": 6.1524,
      "mean_token_accuracy": 0.13897238597273825,
      "num_tokens": 11737753.0,
      "step": 5790
    },
    {
      "entropy": 6.228318881988526,
      "epoch": 0.5064231407847593,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004980192448114635,
      "loss": 6.1226,
      "mean_token_accuracy": 0.13694148138165474,
      "num_tokens": 11747837.0,
      "step": 5795
    },
    {
      "entropy": 6.225673532485962,
      "epoch": 0.5068600891374639,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004980151170075758,
      "loss": 6.261,
      "mean_token_accuracy": 0.13704407438635827,
      "num_tokens": 11757911.0,
      "step": 5800
    },
    {
      "entropy": 6.434248065948486,
      "epoch": 0.5072970374901686,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004980109849261414,
      "loss": 6.1192,
      "mean_token_accuracy": 0.13910652995109557,
      "num_tokens": 11766213.0,
      "step": 5805
    },
    {
      "entropy": 6.340929460525513,
      "epoch": 0.5077339858428733,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004980068485672394,
      "loss": 6.0813,
      "mean_token_accuracy": 0.14624812453985214,
      "num_tokens": 11775640.0,
      "step": 5810
    },
    {
      "entropy": 6.366253280639649,
      "epoch": 0.5081709341955781,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004980027079309493,
      "loss": 6.1474,
      "mean_token_accuracy": 0.14522576928138733,
      "num_tokens": 11785664.0,
      "step": 5815
    },
    {
      "entropy": 6.283198547363281,
      "epoch": 0.5086078825482828,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004979985630173503,
      "loss": 6.2335,
      "mean_token_accuracy": 0.12381275817751884,
      "num_tokens": 11795402.0,
      "step": 5820
    },
    {
      "entropy": 6.26009783744812,
      "epoch": 0.5090448309009875,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004979944138265221,
      "loss": 6.1075,
      "mean_token_accuracy": 0.13531444519758223,
      "num_tokens": 11805923.0,
      "step": 5825
    },
    {
      "entropy": 6.269751787185669,
      "epoch": 0.5094817792536922,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004979902603585442,
      "loss": 6.1816,
      "mean_token_accuracy": 0.13383990600705148,
      "num_tokens": 11816097.0,
      "step": 5830
    },
    {
      "entropy": 6.357937622070312,
      "epoch": 0.5099187276063969,
      "grad_norm": 8.5,
      "learning_rate": 0.0004979861026134961,
      "loss": 6.155,
      "mean_token_accuracy": 0.1380440667271614,
      "num_tokens": 11826889.0,
      "step": 5835
    },
    {
      "entropy": 6.396521854400635,
      "epoch": 0.5103556759591016,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004979819405914579,
      "loss": 6.1816,
      "mean_token_accuracy": 0.1380002774298191,
      "num_tokens": 11837935.0,
      "step": 5840
    },
    {
      "entropy": 6.3219574928283695,
      "epoch": 0.5107926243118064,
      "grad_norm": 5.15625,
      "learning_rate": 0.0004979777742925092,
      "loss": 6.1743,
      "mean_token_accuracy": 0.1355532541871071,
      "num_tokens": 11847715.0,
      "step": 5845
    },
    {
      "entropy": 6.372812747955322,
      "epoch": 0.5112295726645111,
      "grad_norm": 32.5,
      "learning_rate": 0.0004979736037167298,
      "loss": 6.1842,
      "mean_token_accuracy": 0.13102409169077872,
      "num_tokens": 11857841.0,
      "step": 5850
    },
    {
      "entropy": 6.357076358795166,
      "epoch": 0.5116665210172158,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004979694288642,
      "loss": 6.1865,
      "mean_token_accuracy": 0.13729090094566346,
      "num_tokens": 11868135.0,
      "step": 5855
    },
    {
      "entropy": 6.354056882858276,
      "epoch": 0.5121034693699205,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004979652497349997,
      "loss": 6.1154,
      "mean_token_accuracy": 0.13748626708984374,
      "num_tokens": 11877637.0,
      "step": 5860
    },
    {
      "entropy": 6.226434755325317,
      "epoch": 0.5125404177226252,
      "grad_norm": 2.375,
      "learning_rate": 0.000497961066329209,
      "loss": 6.1697,
      "mean_token_accuracy": 0.13338080421090126,
      "num_tokens": 11887128.0,
      "step": 5865
    },
    {
      "entropy": 6.329555034637451,
      "epoch": 0.5129773660753298,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004979568786469083,
      "loss": 6.1766,
      "mean_token_accuracy": 0.13152823224663734,
      "num_tokens": 11897885.0,
      "step": 5870
    },
    {
      "entropy": 6.3101404190063475,
      "epoch": 0.5134143144280346,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004979526866881777,
      "loss": 6.1393,
      "mean_token_accuracy": 0.13783316761255265,
      "num_tokens": 11907919.0,
      "step": 5875
    },
    {
      "entropy": 6.290446805953979,
      "epoch": 0.5138512627807393,
      "grad_norm": 2.03125,
      "learning_rate": 0.000497948490453098,
      "loss": 6.1277,
      "mean_token_accuracy": 0.13891322165727615,
      "num_tokens": 11918612.0,
      "step": 5880
    },
    {
      "entropy": 6.268454742431641,
      "epoch": 0.514288211133444,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004979442899417493,
      "loss": 6.2025,
      "mean_token_accuracy": 0.13276560604572296,
      "num_tokens": 11927759.0,
      "step": 5885
    },
    {
      "entropy": 6.337447023391723,
      "epoch": 0.5147251594861487,
      "grad_norm": 2.5,
      "learning_rate": 0.0004979400851542122,
      "loss": 6.1389,
      "mean_token_accuracy": 0.14539865031838417,
      "num_tokens": 11937522.0,
      "step": 5890
    },
    {
      "entropy": 6.264947843551636,
      "epoch": 0.5151621078388534,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004979358760905675,
      "loss": 6.1436,
      "mean_token_accuracy": 0.13401950821280478,
      "num_tokens": 11947787.0,
      "step": 5895
    },
    {
      "entropy": 6.198224306106567,
      "epoch": 0.5155990561915581,
      "grad_norm": 2.171875,
      "learning_rate": 0.000497931662750896,
      "loss": 6.1801,
      "mean_token_accuracy": 0.13323895558714866,
      "num_tokens": 11958929.0,
      "step": 5900
    },
    {
      "entropy": 6.329074192047119,
      "epoch": 0.5160360045442629,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004979274451352783,
      "loss": 6.1285,
      "mean_token_accuracy": 0.13562881723046302,
      "num_tokens": 11969242.0,
      "step": 5905
    },
    {
      "entropy": 6.324266195297241,
      "epoch": 0.5164729528969676,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004979232232437954,
      "loss": 6.1994,
      "mean_token_accuracy": 0.13172708749771117,
      "num_tokens": 11980149.0,
      "step": 5910
    },
    {
      "entropy": 6.341662359237671,
      "epoch": 0.5169099012496723,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004979189970765282,
      "loss": 6.1932,
      "mean_token_accuracy": 0.13332394734025002,
      "num_tokens": 11990636.0,
      "step": 5915
    },
    {
      "entropy": 6.332500743865967,
      "epoch": 0.517346849602377,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004979147666335579,
      "loss": 6.2118,
      "mean_token_accuracy": 0.13603553026914597,
      "num_tokens": 12000615.0,
      "step": 5920
    },
    {
      "entropy": 6.320348405838013,
      "epoch": 0.5177837979550817,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004979105319149656,
      "loss": 6.2275,
      "mean_token_accuracy": 0.1333947792649269,
      "num_tokens": 12011072.0,
      "step": 5925
    },
    {
      "entropy": 6.295902681350708,
      "epoch": 0.5182207463077865,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004979062929208324,
      "loss": 6.1213,
      "mean_token_accuracy": 0.13844589218497277,
      "num_tokens": 12021726.0,
      "step": 5930
    },
    {
      "entropy": 6.332152032852173,
      "epoch": 0.5186576946604912,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004979020496512398,
      "loss": 6.2387,
      "mean_token_accuracy": 0.1295247420668602,
      "num_tokens": 12031353.0,
      "step": 5935
    },
    {
      "entropy": 6.396020746231079,
      "epoch": 0.5190946430131959,
      "grad_norm": 2.546875,
      "learning_rate": 0.000497897802106269,
      "loss": 6.186,
      "mean_token_accuracy": 0.12760654240846633,
      "num_tokens": 12041234.0,
      "step": 5940
    },
    {
      "entropy": 6.275166034698486,
      "epoch": 0.5195315913659005,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004978935502860017,
      "loss": 6.1339,
      "mean_token_accuracy": 0.1391075909137726,
      "num_tokens": 12051100.0,
      "step": 5945
    },
    {
      "entropy": 6.28311676979065,
      "epoch": 0.5199685397186052,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004978892941905191,
      "loss": 6.2045,
      "mean_token_accuracy": 0.1361614905297756,
      "num_tokens": 12062012.0,
      "step": 5950
    },
    {
      "entropy": 6.364482879638672,
      "epoch": 0.5204054880713099,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004978850338199033,
      "loss": 6.2007,
      "mean_token_accuracy": 0.13510503694415094,
      "num_tokens": 12072763.0,
      "step": 5955
    },
    {
      "entropy": 6.309931421279908,
      "epoch": 0.5208424364240147,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004978807691742356,
      "loss": 6.0415,
      "mean_token_accuracy": 0.1455541208386421,
      "num_tokens": 12081892.0,
      "step": 5960
    },
    {
      "entropy": 6.252465724945068,
      "epoch": 0.5212793847767194,
      "grad_norm": 2.796875,
      "learning_rate": 0.000497876500253598,
      "loss": 6.1524,
      "mean_token_accuracy": 0.13437189161777496,
      "num_tokens": 12091428.0,
      "step": 5965
    },
    {
      "entropy": 6.259633779525757,
      "epoch": 0.5217163331294241,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004978722270580722,
      "loss": 6.1623,
      "mean_token_accuracy": 0.1332266591489315,
      "num_tokens": 12101687.0,
      "step": 5970
    },
    {
      "entropy": 6.221106910705567,
      "epoch": 0.5221532814821288,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004978679495877405,
      "loss": 6.0981,
      "mean_token_accuracy": 0.136045141518116,
      "num_tokens": 12111726.0,
      "step": 5975
    },
    {
      "entropy": 6.299628782272339,
      "epoch": 0.5225902298348335,
      "grad_norm": 3.875,
      "learning_rate": 0.0004978636678426847,
      "loss": 6.1376,
      "mean_token_accuracy": 0.142653451859951,
      "num_tokens": 12122171.0,
      "step": 5980
    },
    {
      "entropy": 6.462085723876953,
      "epoch": 0.5230271781875382,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004978593818229869,
      "loss": 6.155,
      "mean_token_accuracy": 0.1355630151927471,
      "num_tokens": 12131414.0,
      "step": 5985
    },
    {
      "entropy": 6.311098957061768,
      "epoch": 0.523464126540243,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004978550915287294,
      "loss": 6.1543,
      "mean_token_accuracy": 0.12896232530474663,
      "num_tokens": 12141665.0,
      "step": 5990
    },
    {
      "entropy": 6.279083251953125,
      "epoch": 0.5239010748929477,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004978507969599945,
      "loss": 6.2116,
      "mean_token_accuracy": 0.1313002362847328,
      "num_tokens": 12151263.0,
      "step": 5995
    },
    {
      "entropy": 6.320516920089721,
      "epoch": 0.5243380232456524,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004978464981168645,
      "loss": 6.1818,
      "mean_token_accuracy": 0.1371571436524391,
      "num_tokens": 12161070.0,
      "step": 6000
    },
    {
      "epoch": 0.5243380232456524,
      "eval_entropy": 6.134990820751764,
      "eval_loss": 6.199563980102539,
      "eval_mean_token_accuracy": 0.14097981776390808,
      "eval_num_tokens": 12161070.0,
      "eval_runtime": 27.8165,
      "eval_samples_per_second": 1330.647,
      "eval_steps_per_second": 166.34,
      "step": 6000
    },
    {
      "entropy": 6.305163955688476,
      "epoch": 0.5247749715983571,
      "grad_norm": 2.46875,
      "learning_rate": 0.000497842194999422,
      "loss": 6.1656,
      "mean_token_accuracy": 0.1394128642976284,
      "num_tokens": 12171197.0,
      "step": 6005
    },
    {
      "entropy": 6.254998970031738,
      "epoch": 0.5252119199510618,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004978378876077493,
      "loss": 6.1798,
      "mean_token_accuracy": 0.13538631796836853,
      "num_tokens": 12181462.0,
      "step": 6010
    },
    {
      "entropy": 6.333068084716797,
      "epoch": 0.5256488683037664,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004978335759419293,
      "loss": 6.1517,
      "mean_token_accuracy": 0.14131810739636422,
      "num_tokens": 12191473.0,
      "step": 6015
    },
    {
      "entropy": 6.368027305603027,
      "epoch": 0.5260858166564713,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004978292600020444,
      "loss": 6.1239,
      "mean_token_accuracy": 0.13592676743865012,
      "num_tokens": 12200920.0,
      "step": 6020
    },
    {
      "entropy": 6.296689748764038,
      "epoch": 0.5265227650091759,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004978249397881776,
      "loss": 6.1834,
      "mean_token_accuracy": 0.13516806066036224,
      "num_tokens": 12210823.0,
      "step": 6025
    },
    {
      "entropy": 6.338432645797729,
      "epoch": 0.5269597133618806,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004978206153004117,
      "loss": 6.2136,
      "mean_token_accuracy": 0.1305401362478733,
      "num_tokens": 12221967.0,
      "step": 6030
    },
    {
      "entropy": 6.325851964950561,
      "epoch": 0.5273966617145853,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004978162865388297,
      "loss": 6.1592,
      "mean_token_accuracy": 0.13554399758577346,
      "num_tokens": 12232519.0,
      "step": 6035
    },
    {
      "entropy": 6.2794294357299805,
      "epoch": 0.52783361006729,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004978119535035144,
      "loss": 6.1336,
      "mean_token_accuracy": 0.1330802820622921,
      "num_tokens": 12243193.0,
      "step": 6040
    },
    {
      "entropy": 6.3019904613494875,
      "epoch": 0.5282705584199947,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004978076161945492,
      "loss": 6.1589,
      "mean_token_accuracy": 0.13709976598620416,
      "num_tokens": 12253640.0,
      "step": 6045
    },
    {
      "entropy": 6.434941053390503,
      "epoch": 0.5287075067726995,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004978032746120173,
      "loss": 6.2615,
      "mean_token_accuracy": 0.1237427055835724,
      "num_tokens": 12265072.0,
      "step": 6050
    },
    {
      "entropy": 6.346615123748779,
      "epoch": 0.5291444551254042,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004977989287560016,
      "loss": 6.1989,
      "mean_token_accuracy": 0.13342683538794517,
      "num_tokens": 12274449.0,
      "step": 6055
    },
    {
      "entropy": 6.305394840240479,
      "epoch": 0.5295814034781089,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004977945786265857,
      "loss": 6.1983,
      "mean_token_accuracy": 0.1254110664129257,
      "num_tokens": 12284910.0,
      "step": 6060
    },
    {
      "entropy": 6.403089141845703,
      "epoch": 0.5300183518308136,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004977902242238533,
      "loss": 6.1901,
      "mean_token_accuracy": 0.13085756748914718,
      "num_tokens": 12294046.0,
      "step": 6065
    },
    {
      "entropy": 6.321627950668335,
      "epoch": 0.5304553001835183,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004977858655478874,
      "loss": 6.2061,
      "mean_token_accuracy": 0.1336338959634304,
      "num_tokens": 12303900.0,
      "step": 6070
    },
    {
      "entropy": 6.290190172195435,
      "epoch": 0.530892248536223,
      "grad_norm": 4.375,
      "learning_rate": 0.000497781502598772,
      "loss": 6.1539,
      "mean_token_accuracy": 0.1417690172791481,
      "num_tokens": 12313326.0,
      "step": 6075
    },
    {
      "entropy": 6.330353784561157,
      "epoch": 0.5313291968889278,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004977771353765906,
      "loss": 6.1407,
      "mean_token_accuracy": 0.1372034527361393,
      "num_tokens": 12322792.0,
      "step": 6080
    },
    {
      "entropy": 6.384373474121094,
      "epoch": 0.5317661452416325,
      "grad_norm": 2.59375,
      "learning_rate": 0.000497772763881427,
      "loss": 6.2596,
      "mean_token_accuracy": 0.12987482249736787,
      "num_tokens": 12333417.0,
      "step": 6085
    },
    {
      "entropy": 6.3237316608428955,
      "epoch": 0.5322030935943372,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004977683881133651,
      "loss": 6.1584,
      "mean_token_accuracy": 0.13096655830740928,
      "num_tokens": 12343944.0,
      "step": 6090
    },
    {
      "entropy": 6.307975625991821,
      "epoch": 0.5326400419470418,
      "grad_norm": 7.9375,
      "learning_rate": 0.0004977640080724888,
      "loss": 6.1371,
      "mean_token_accuracy": 0.13328821882605552,
      "num_tokens": 12353838.0,
      "step": 6095
    },
    {
      "entropy": 6.362212610244751,
      "epoch": 0.5330769902997465,
      "grad_norm": 2.875,
      "learning_rate": 0.0004977596237588821,
      "loss": 6.2411,
      "mean_token_accuracy": 0.12665953263640403,
      "num_tokens": 12364996.0,
      "step": 6100
    },
    {
      "entropy": 6.355288171768189,
      "epoch": 0.5335139386524512,
      "grad_norm": 18.125,
      "learning_rate": 0.000497755235172629,
      "loss": 6.2021,
      "mean_token_accuracy": 0.13149810060858727,
      "num_tokens": 12375111.0,
      "step": 6105
    },
    {
      "entropy": 6.315223169326782,
      "epoch": 0.533950887005156,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004977508423138139,
      "loss": 6.1704,
      "mean_token_accuracy": 0.13136148154735566,
      "num_tokens": 12384376.0,
      "step": 6110
    },
    {
      "entropy": 6.29996018409729,
      "epoch": 0.5343878353578607,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004977464451825208,
      "loss": 6.2157,
      "mean_token_accuracy": 0.1282039150595665,
      "num_tokens": 12395248.0,
      "step": 6115
    },
    {
      "entropy": 6.406261253356933,
      "epoch": 0.5348247837105654,
      "grad_norm": 5.625,
      "learning_rate": 0.0004977420437788341,
      "loss": 6.186,
      "mean_token_accuracy": 0.1329594500362873,
      "num_tokens": 12405083.0,
      "step": 6120
    },
    {
      "entropy": 6.325230550765991,
      "epoch": 0.5352617320632701,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004977376381028385,
      "loss": 6.1255,
      "mean_token_accuracy": 0.14437345489859582,
      "num_tokens": 12415787.0,
      "step": 6125
    },
    {
      "entropy": 6.349701833724976,
      "epoch": 0.5356986804159748,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004977332281546181,
      "loss": 6.1632,
      "mean_token_accuracy": 0.13416141122579575,
      "num_tokens": 12425911.0,
      "step": 6130
    },
    {
      "entropy": 6.279951524734497,
      "epoch": 0.5361356287686796,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004977288139342578,
      "loss": 6.1977,
      "mean_token_accuracy": 0.1325320601463318,
      "num_tokens": 12437270.0,
      "step": 6135
    },
    {
      "entropy": 6.293383312225342,
      "epoch": 0.5365725771213843,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004977243954418422,
      "loss": 6.1556,
      "mean_token_accuracy": 0.132023236900568,
      "num_tokens": 12446778.0,
      "step": 6140
    },
    {
      "entropy": 6.26518816947937,
      "epoch": 0.537009525474089,
      "grad_norm": 22.125,
      "learning_rate": 0.0004977199726774558,
      "loss": 6.1153,
      "mean_token_accuracy": 0.13906113430857658,
      "num_tokens": 12456550.0,
      "step": 6145
    },
    {
      "entropy": 6.263458347320556,
      "epoch": 0.5374464738267937,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004977155456411837,
      "loss": 6.1209,
      "mean_token_accuracy": 0.137327628582716,
      "num_tokens": 12465562.0,
      "step": 6150
    },
    {
      "entropy": 6.362095165252685,
      "epoch": 0.5378834221794984,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004977111143331108,
      "loss": 6.1818,
      "mean_token_accuracy": 0.1301529623568058,
      "num_tokens": 12475277.0,
      "step": 6155
    },
    {
      "entropy": 6.327201700210571,
      "epoch": 0.538320370532203,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004977066787533219,
      "loss": 6.1065,
      "mean_token_accuracy": 0.13954316452145576,
      "num_tokens": 12485478.0,
      "step": 6160
    },
    {
      "entropy": 6.304299020767212,
      "epoch": 0.5387573188849079,
      "grad_norm": 6.125,
      "learning_rate": 0.0004977022389019023,
      "loss": 6.2304,
      "mean_token_accuracy": 0.13455819711089134,
      "num_tokens": 12495714.0,
      "step": 6165
    },
    {
      "entropy": 6.388070440292358,
      "epoch": 0.5391942672376125,
      "grad_norm": 3.5,
      "learning_rate": 0.0004976977947789371,
      "loss": 6.1671,
      "mean_token_accuracy": 0.14347837418317794,
      "num_tokens": 12505125.0,
      "step": 6170
    },
    {
      "entropy": 6.331564044952392,
      "epoch": 0.5396312155903172,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004976933463845115,
      "loss": 6.1574,
      "mean_token_accuracy": 0.138451986014843,
      "num_tokens": 12515051.0,
      "step": 6175
    },
    {
      "entropy": 6.262215662002563,
      "epoch": 0.5400681639430219,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004976888937187107,
      "loss": 6.101,
      "mean_token_accuracy": 0.14152950122952462,
      "num_tokens": 12525389.0,
      "step": 6180
    },
    {
      "entropy": 6.31134524345398,
      "epoch": 0.5405051122957266,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004976844367816203,
      "loss": 6.2029,
      "mean_token_accuracy": 0.12673113346099854,
      "num_tokens": 12535454.0,
      "step": 6185
    },
    {
      "entropy": 6.360104990005493,
      "epoch": 0.5409420606484313,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004976799755733257,
      "loss": 6.0285,
      "mean_token_accuracy": 0.1447511374950409,
      "num_tokens": 12546092.0,
      "step": 6190
    },
    {
      "entropy": 6.318111419677734,
      "epoch": 0.5413790090011361,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004976755100939126,
      "loss": 6.1569,
      "mean_token_accuracy": 0.1301625706255436,
      "num_tokens": 12555462.0,
      "step": 6195
    },
    {
      "entropy": 6.2724151611328125,
      "epoch": 0.5418159573538408,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004976710403434663,
      "loss": 6.1901,
      "mean_token_accuracy": 0.13680475503206252,
      "num_tokens": 12564506.0,
      "step": 6200
    },
    {
      "entropy": 6.327898025512695,
      "epoch": 0.5422529057065455,
      "grad_norm": 3.0625,
      "learning_rate": 0.000497666566322073,
      "loss": 6.1626,
      "mean_token_accuracy": 0.13060543686151505,
      "num_tokens": 12574887.0,
      "step": 6205
    },
    {
      "entropy": 6.426037168502807,
      "epoch": 0.5426898540592502,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004976620880298182,
      "loss": 6.1813,
      "mean_token_accuracy": 0.1336768552660942,
      "num_tokens": 12584499.0,
      "step": 6210
    },
    {
      "entropy": 6.338817405700683,
      "epoch": 0.5431268024119549,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004976576054667878,
      "loss": 6.1303,
      "mean_token_accuracy": 0.1396690636873245,
      "num_tokens": 12593796.0,
      "step": 6215
    },
    {
      "entropy": 6.27109317779541,
      "epoch": 0.5435637507646596,
      "grad_norm": 2.578125,
      "learning_rate": 0.000497653118633068,
      "loss": 6.1706,
      "mean_token_accuracy": 0.13551239520311356,
      "num_tokens": 12603719.0,
      "step": 6220
    },
    {
      "entropy": 6.281965684890747,
      "epoch": 0.5440006991173644,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004976486275287446,
      "loss": 6.1864,
      "mean_token_accuracy": 0.13840994611382484,
      "num_tokens": 12614407.0,
      "step": 6225
    },
    {
      "entropy": 6.349825143814087,
      "epoch": 0.5444376474700691,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004976441321539038,
      "loss": 6.1791,
      "mean_token_accuracy": 0.13927434831857682,
      "num_tokens": 12624414.0,
      "step": 6230
    },
    {
      "entropy": 6.343572378158569,
      "epoch": 0.5448745958227738,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004976396325086321,
      "loss": 6.1968,
      "mean_token_accuracy": 0.13521063327789307,
      "num_tokens": 12635679.0,
      "step": 6235
    },
    {
      "entropy": 6.346814107894898,
      "epoch": 0.5453115441754784,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004976351285930154,
      "loss": 6.1113,
      "mean_token_accuracy": 0.1346683382987976,
      "num_tokens": 12645253.0,
      "step": 6240
    },
    {
      "entropy": 6.209513187408447,
      "epoch": 0.5457484925281831,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004976306204071404,
      "loss": 6.0974,
      "mean_token_accuracy": 0.14378805905580522,
      "num_tokens": 12655815.0,
      "step": 6245
    },
    {
      "entropy": 6.201349687576294,
      "epoch": 0.5461854408808878,
      "grad_norm": 2.25,
      "learning_rate": 0.0004976261079510934,
      "loss": 6.1627,
      "mean_token_accuracy": 0.1401552975177765,
      "num_tokens": 12666173.0,
      "step": 6250
    },
    {
      "entropy": 6.293998289108276,
      "epoch": 0.5466223892335926,
      "grad_norm": 3.0625,
      "learning_rate": 0.000497621591224961,
      "loss": 6.0965,
      "mean_token_accuracy": 0.14540037736296654,
      "num_tokens": 12676008.0,
      "step": 6255
    },
    {
      "entropy": 6.414225959777832,
      "epoch": 0.5470593375862973,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004976170702288298,
      "loss": 6.1891,
      "mean_token_accuracy": 0.12984596341848373,
      "num_tokens": 12686217.0,
      "step": 6260
    },
    {
      "entropy": 6.341739273071289,
      "epoch": 0.547496285939002,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004976125449627866,
      "loss": 6.1867,
      "mean_token_accuracy": 0.13160810321569444,
      "num_tokens": 12696809.0,
      "step": 6265
    },
    {
      "entropy": 6.351231813430786,
      "epoch": 0.5479332342917067,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004976080154269181,
      "loss": 6.2538,
      "mean_token_accuracy": 0.12306354716420173,
      "num_tokens": 12708328.0,
      "step": 6270
    },
    {
      "entropy": 6.400419807434082,
      "epoch": 0.5483701826444114,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004976034816213112,
      "loss": 6.1052,
      "mean_token_accuracy": 0.13385439068078994,
      "num_tokens": 12718478.0,
      "step": 6275
    },
    {
      "entropy": 6.249461078643799,
      "epoch": 0.5488071309971161,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004975989435460528,
      "loss": 6.1392,
      "mean_token_accuracy": 0.1379825122654438,
      "num_tokens": 12728853.0,
      "step": 6280
    },
    {
      "entropy": 6.255404615402222,
      "epoch": 0.5492440793498209,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004975944012012302,
      "loss": 6.2152,
      "mean_token_accuracy": 0.1395138293504715,
      "num_tokens": 12739036.0,
      "step": 6285
    },
    {
      "entropy": 6.373023986816406,
      "epoch": 0.5496810277025256,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004975898545869303,
      "loss": 6.1161,
      "mean_token_accuracy": 0.13932859525084496,
      "num_tokens": 12748908.0,
      "step": 6290
    },
    {
      "entropy": 6.278858661651611,
      "epoch": 0.5501179760552303,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004975853037032403,
      "loss": 6.1887,
      "mean_token_accuracy": 0.12895094454288483,
      "num_tokens": 12758671.0,
      "step": 6295
    },
    {
      "entropy": 6.26996431350708,
      "epoch": 0.550554924407935,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004975807485502476,
      "loss": 6.1657,
      "mean_token_accuracy": 0.13549817726016045,
      "num_tokens": 12769552.0,
      "step": 6300
    },
    {
      "entropy": 6.4022118091583256,
      "epoch": 0.5509918727606397,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004975761891280394,
      "loss": 6.3027,
      "mean_token_accuracy": 0.1262983836233616,
      "num_tokens": 12781032.0,
      "step": 6305
    },
    {
      "entropy": 6.450229167938232,
      "epoch": 0.5514288211133443,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004975716254367033,
      "loss": 6.2084,
      "mean_token_accuracy": 0.12668768540024758,
      "num_tokens": 12790818.0,
      "step": 6310
    },
    {
      "entropy": 6.247216510772705,
      "epoch": 0.5518657694660491,
      "grad_norm": 2.140625,
      "learning_rate": 0.000497567057476327,
      "loss": 6.1059,
      "mean_token_accuracy": 0.13741389140486718,
      "num_tokens": 12799701.0,
      "step": 6315
    },
    {
      "entropy": 6.2810930728912355,
      "epoch": 0.5523027178187538,
      "grad_norm": 1.75,
      "learning_rate": 0.0004975624852469976,
      "loss": 6.2224,
      "mean_token_accuracy": 0.14172836691141127,
      "num_tokens": 12809575.0,
      "step": 6320
    },
    {
      "entropy": 6.394656229019165,
      "epoch": 0.5527396661714585,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004975579087488033,
      "loss": 6.1582,
      "mean_token_accuracy": 0.1377180337905884,
      "num_tokens": 12820278.0,
      "step": 6325
    },
    {
      "entropy": 6.313861083984375,
      "epoch": 0.5531766145241632,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004975533279818317,
      "loss": 6.1605,
      "mean_token_accuracy": 0.132416795194149,
      "num_tokens": 12830774.0,
      "step": 6330
    },
    {
      "entropy": 6.284891986846924,
      "epoch": 0.5536135628768679,
      "grad_norm": 2.5,
      "learning_rate": 0.0004975487429461706,
      "loss": 6.1321,
      "mean_token_accuracy": 0.14171141535043716,
      "num_tokens": 12839487.0,
      "step": 6335
    },
    {
      "entropy": 6.276378107070923,
      "epoch": 0.5540505112295727,
      "grad_norm": 2.59375,
      "learning_rate": 0.000497544153641908,
      "loss": 6.1028,
      "mean_token_accuracy": 0.1363065131008625,
      "num_tokens": 12849614.0,
      "step": 6340
    },
    {
      "entropy": 6.313328075408935,
      "epoch": 0.5544874595822774,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004975395600691319,
      "loss": 6.2117,
      "mean_token_accuracy": 0.13812530860304834,
      "num_tokens": 12859428.0,
      "step": 6345
    },
    {
      "entropy": 6.3166584968566895,
      "epoch": 0.5549244079349821,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004975349622279305,
      "loss": 6.2019,
      "mean_token_accuracy": 0.13376846089959143,
      "num_tokens": 12869167.0,
      "step": 6350
    },
    {
      "entropy": 6.350392580032349,
      "epoch": 0.5553613562876868,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004975303601183919,
      "loss": 6.1315,
      "mean_token_accuracy": 0.139154614508152,
      "num_tokens": 12878410.0,
      "step": 6355
    },
    {
      "entropy": 6.400624752044678,
      "epoch": 0.5557983046403915,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004975257537406043,
      "loss": 6.149,
      "mean_token_accuracy": 0.13052380234003066,
      "num_tokens": 12888372.0,
      "step": 6360
    },
    {
      "entropy": 6.29941873550415,
      "epoch": 0.5562352529930962,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004975211430946563,
      "loss": 6.1864,
      "mean_token_accuracy": 0.13280265554785728,
      "num_tokens": 12897958.0,
      "step": 6365
    },
    {
      "entropy": 6.303383827209473,
      "epoch": 0.556672201345801,
      "grad_norm": 2.765625,
      "learning_rate": 0.000497516528180636,
      "loss": 6.1966,
      "mean_token_accuracy": 0.13260417133569719,
      "num_tokens": 12907890.0,
      "step": 6370
    },
    {
      "entropy": 6.299492692947387,
      "epoch": 0.5571091496985057,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004975119089986322,
      "loss": 6.1287,
      "mean_token_accuracy": 0.13472474217414857,
      "num_tokens": 12917914.0,
      "step": 6375
    },
    {
      "entropy": 6.248823356628418,
      "epoch": 0.5575460980512104,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004975072855487333,
      "loss": 6.1652,
      "mean_token_accuracy": 0.13773834183812142,
      "num_tokens": 12927090.0,
      "step": 6380
    },
    {
      "entropy": 6.315898132324219,
      "epoch": 0.557983046403915,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004975026578310282,
      "loss": 6.1601,
      "mean_token_accuracy": 0.13427378386259078,
      "num_tokens": 12937236.0,
      "step": 6385
    },
    {
      "entropy": 6.389422750473022,
      "epoch": 0.5584199947566197,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004974980258456055,
      "loss": 6.1126,
      "mean_token_accuracy": 0.13714217841625215,
      "num_tokens": 12947635.0,
      "step": 6390
    },
    {
      "entropy": 6.395745325088501,
      "epoch": 0.5588569431093244,
      "grad_norm": 2.296875,
      "learning_rate": 0.000497493389592554,
      "loss": 6.0669,
      "mean_token_accuracy": 0.1372316911816597,
      "num_tokens": 12957236.0,
      "step": 6395
    },
    {
      "entropy": 6.3218677043914795,
      "epoch": 0.5592938914620292,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004974887490719627,
      "loss": 6.2009,
      "mean_token_accuracy": 0.1317098468542099,
      "num_tokens": 12967497.0,
      "step": 6400
    },
    {
      "entropy": 6.258624696731568,
      "epoch": 0.5597308398147339,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004974841042839207,
      "loss": 6.1191,
      "mean_token_accuracy": 0.13637408539652823,
      "num_tokens": 12977467.0,
      "step": 6405
    },
    {
      "entropy": 6.297503423690796,
      "epoch": 0.5601677881674386,
      "grad_norm": 2.234375,
      "learning_rate": 0.000497479455228517,
      "loss": 6.1633,
      "mean_token_accuracy": 0.13608675226569175,
      "num_tokens": 12987867.0,
      "step": 6410
    },
    {
      "entropy": 6.310818338394165,
      "epoch": 0.5606047365201433,
      "grad_norm": 5.84375,
      "learning_rate": 0.0004974748019058407,
      "loss": 6.1838,
      "mean_token_accuracy": 0.12945412024855613,
      "num_tokens": 12998081.0,
      "step": 6415
    },
    {
      "entropy": 6.2934483051300045,
      "epoch": 0.561041684872848,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004974701443159812,
      "loss": 6.1889,
      "mean_token_accuracy": 0.13335443213582038,
      "num_tokens": 13008047.0,
      "step": 6420
    },
    {
      "entropy": 6.3565295219421385,
      "epoch": 0.5614786332255527,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004974654824590277,
      "loss": 6.1688,
      "mean_token_accuracy": 0.13209476619958876,
      "num_tokens": 13018048.0,
      "step": 6425
    },
    {
      "entropy": 6.303196668624878,
      "epoch": 0.5619155815782575,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004974608163350697,
      "loss": 6.1513,
      "mean_token_accuracy": 0.13664785400032997,
      "num_tokens": 13028156.0,
      "step": 6430
    },
    {
      "entropy": 6.329297637939453,
      "epoch": 0.5623525299309622,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004974561459441967,
      "loss": 6.1499,
      "mean_token_accuracy": 0.13396846279501914,
      "num_tokens": 13038191.0,
      "step": 6435
    },
    {
      "entropy": 6.337226486206054,
      "epoch": 0.5627894782836669,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004974514712864981,
      "loss": 6.1672,
      "mean_token_accuracy": 0.13834557086229324,
      "num_tokens": 13049085.0,
      "step": 6440
    },
    {
      "entropy": 6.28109884262085,
      "epoch": 0.5632264266363716,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004974467923620639,
      "loss": 6.1723,
      "mean_token_accuracy": 0.13435535803437232,
      "num_tokens": 13060010.0,
      "step": 6445
    },
    {
      "entropy": 6.370020151138306,
      "epoch": 0.5636633749890763,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004974421091709836,
      "loss": 6.1643,
      "mean_token_accuracy": 0.13094285652041435,
      "num_tokens": 13069768.0,
      "step": 6450
    },
    {
      "entropy": 6.357784605026245,
      "epoch": 0.564100323341781,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004974374217133471,
      "loss": 6.1731,
      "mean_token_accuracy": 0.1343522012233734,
      "num_tokens": 13079842.0,
      "step": 6455
    },
    {
      "entropy": 6.2995692729949955,
      "epoch": 0.5645372716944858,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004974327299892443,
      "loss": 6.1008,
      "mean_token_accuracy": 0.14095863923430443,
      "num_tokens": 13089235.0,
      "step": 6460
    },
    {
      "entropy": 6.322852849960327,
      "epoch": 0.5649742200471904,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004974280339987651,
      "loss": 6.1354,
      "mean_token_accuracy": 0.13147189393639563,
      "num_tokens": 13098364.0,
      "step": 6465
    },
    {
      "entropy": 6.28232889175415,
      "epoch": 0.5654111683998951,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004974233337419997,
      "loss": 6.258,
      "mean_token_accuracy": 0.13029298931360245,
      "num_tokens": 13108876.0,
      "step": 6470
    },
    {
      "entropy": 6.3876729011535645,
      "epoch": 0.5658481167525998,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004974186292190383,
      "loss": 6.2742,
      "mean_token_accuracy": 0.12826967537403106,
      "num_tokens": 13119591.0,
      "step": 6475
    },
    {
      "entropy": 6.396380424499512,
      "epoch": 0.5662850651053045,
      "grad_norm": 2.828125,
      "learning_rate": 0.000497413920429971,
      "loss": 6.1757,
      "mean_token_accuracy": 0.13720449060201645,
      "num_tokens": 13129796.0,
      "step": 6480
    },
    {
      "entropy": 6.355126810073853,
      "epoch": 0.5667220134580092,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004974092073748882,
      "loss": 6.1249,
      "mean_token_accuracy": 0.13661054372787476,
      "num_tokens": 13139302.0,
      "step": 6485
    },
    {
      "entropy": 6.311205148696899,
      "epoch": 0.567158961810714,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004974044900538802,
      "loss": 6.1476,
      "mean_token_accuracy": 0.1332809656858444,
      "num_tokens": 13148983.0,
      "step": 6490
    },
    {
      "entropy": 6.338843631744385,
      "epoch": 0.5675959101634187,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004973997684670375,
      "loss": 6.1631,
      "mean_token_accuracy": 0.1345617838203907,
      "num_tokens": 13159016.0,
      "step": 6495
    },
    {
      "entropy": 6.383847713470459,
      "epoch": 0.5680328585161234,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004973950426144508,
      "loss": 6.1803,
      "mean_token_accuracy": 0.13043221682310105,
      "num_tokens": 13169705.0,
      "step": 6500
    },
    {
      "entropy": 6.37493486404419,
      "epoch": 0.5684698068688281,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004973903124962106,
      "loss": 6.1592,
      "mean_token_accuracy": 0.1301020860671997,
      "num_tokens": 13181372.0,
      "step": 6505
    },
    {
      "entropy": 6.34631929397583,
      "epoch": 0.5689067552215328,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004973855781124077,
      "loss": 6.2214,
      "mean_token_accuracy": 0.12929339930415154,
      "num_tokens": 13192581.0,
      "step": 6510
    },
    {
      "entropy": 6.366477584838867,
      "epoch": 0.5693437035742375,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004973808394631329,
      "loss": 6.2553,
      "mean_token_accuracy": 0.12961444333195687,
      "num_tokens": 13203269.0,
      "step": 6515
    },
    {
      "entropy": 6.324162244796753,
      "epoch": 0.5697806519269423,
      "grad_norm": 2.25,
      "learning_rate": 0.000497376096548477,
      "loss": 6.1224,
      "mean_token_accuracy": 0.13464102745056153,
      "num_tokens": 13213183.0,
      "step": 6520
    },
    {
      "entropy": 6.287112998962402,
      "epoch": 0.570217600279647,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004973713493685311,
      "loss": 6.1911,
      "mean_token_accuracy": 0.12940949946641922,
      "num_tokens": 13223196.0,
      "step": 6525
    },
    {
      "entropy": 6.326357460021972,
      "epoch": 0.5706545486323517,
      "grad_norm": 2.25,
      "learning_rate": 0.0004973665979233861,
      "loss": 6.1225,
      "mean_token_accuracy": 0.13936646655201912,
      "num_tokens": 13233915.0,
      "step": 6530
    },
    {
      "entropy": 6.312110328674317,
      "epoch": 0.5710914969850563,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004973618422131334,
      "loss": 6.1647,
      "mean_token_accuracy": 0.13392532393336296,
      "num_tokens": 13245108.0,
      "step": 6535
    },
    {
      "entropy": 6.330479478836059,
      "epoch": 0.571528445337761,
      "grad_norm": 2.765625,
      "learning_rate": 0.000497357082237864,
      "loss": 6.1847,
      "mean_token_accuracy": 0.1351300448179245,
      "num_tokens": 13255508.0,
      "step": 6540
    },
    {
      "entropy": 6.371770572662354,
      "epoch": 0.5719653936904658,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004973523179976692,
      "loss": 6.195,
      "mean_token_accuracy": 0.13498456329107283,
      "num_tokens": 13265118.0,
      "step": 6545
    },
    {
      "entropy": 6.335427713394165,
      "epoch": 0.5724023420431705,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004973475494926404,
      "loss": 6.1106,
      "mean_token_accuracy": 0.13610025942325593,
      "num_tokens": 13275947.0,
      "step": 6550
    },
    {
      "entropy": 6.261903476715088,
      "epoch": 0.5728392903958752,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004973427767228692,
      "loss": 6.128,
      "mean_token_accuracy": 0.13728638961911202,
      "num_tokens": 13284662.0,
      "step": 6555
    },
    {
      "entropy": 6.243236875534057,
      "epoch": 0.5732762387485799,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000497337999688447,
      "loss": 6.2064,
      "mean_token_accuracy": 0.13668475821614265,
      "num_tokens": 13295627.0,
      "step": 6560
    },
    {
      "entropy": 6.3944309711456295,
      "epoch": 0.5737131871012846,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004973332183894653,
      "loss": 6.1103,
      "mean_token_accuracy": 0.14005890041589736,
      "num_tokens": 13305205.0,
      "step": 6565
    },
    {
      "entropy": 6.321352815628051,
      "epoch": 0.5741501354539893,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004973284328260162,
      "loss": 6.1527,
      "mean_token_accuracy": 0.13679827749729156,
      "num_tokens": 13315464.0,
      "step": 6570
    },
    {
      "entropy": 6.250048112869263,
      "epoch": 0.5745870838066941,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004973236429981912,
      "loss": 6.0602,
      "mean_token_accuracy": 0.14108327478170396,
      "num_tokens": 13325468.0,
      "step": 6575
    },
    {
      "entropy": 6.28951473236084,
      "epoch": 0.5750240321593988,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004973188489060821,
      "loss": 6.1686,
      "mean_token_accuracy": 0.1383883535861969,
      "num_tokens": 13334719.0,
      "step": 6580
    },
    {
      "entropy": 6.2976781845092775,
      "epoch": 0.5754609805121035,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004973140505497812,
      "loss": 6.1347,
      "mean_token_accuracy": 0.13694736659526824,
      "num_tokens": 13344545.0,
      "step": 6585
    },
    {
      "entropy": 6.351121807098389,
      "epoch": 0.5758979288648082,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004973092479293802,
      "loss": 6.2063,
      "mean_token_accuracy": 0.12937029898166658,
      "num_tokens": 13355212.0,
      "step": 6590
    },
    {
      "entropy": 6.2793004512786865,
      "epoch": 0.5763348772175129,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004973044410449713,
      "loss": 6.1015,
      "mean_token_accuracy": 0.13832251355051994,
      "num_tokens": 13364869.0,
      "step": 6595
    },
    {
      "entropy": 6.31839919090271,
      "epoch": 0.5767718255702176,
      "grad_norm": 4.25,
      "learning_rate": 0.0004972996298966468,
      "loss": 6.1357,
      "mean_token_accuracy": 0.1325461283326149,
      "num_tokens": 13376511.0,
      "step": 6600
    },
    {
      "entropy": 6.308404445648193,
      "epoch": 0.5772087739229224,
      "grad_norm": 2.90625,
      "learning_rate": 0.000497294814484499,
      "loss": 6.1693,
      "mean_token_accuracy": 0.13341064378619194,
      "num_tokens": 13385470.0,
      "step": 6605
    },
    {
      "entropy": 6.283018159866333,
      "epoch": 0.577645722275627,
      "grad_norm": 2.796875,
      "learning_rate": 0.00049728999480862,
      "loss": 6.2387,
      "mean_token_accuracy": 0.13731758594512938,
      "num_tokens": 13395328.0,
      "step": 6610
    },
    {
      "entropy": 6.377890968322754,
      "epoch": 0.5780826706283317,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004972851708691025,
      "loss": 6.1403,
      "mean_token_accuracy": 0.1410699650645256,
      "num_tokens": 13405319.0,
      "step": 6615
    },
    {
      "entropy": 6.3066065311431885,
      "epoch": 0.5785196189810364,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004972803426660389,
      "loss": 6.1418,
      "mean_token_accuracy": 0.13646004125475883,
      "num_tokens": 13416119.0,
      "step": 6620
    },
    {
      "entropy": 6.243694305419922,
      "epoch": 0.5789565673337411,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004972755101995219,
      "loss": 6.1544,
      "mean_token_accuracy": 0.12985534444451333,
      "num_tokens": 13425824.0,
      "step": 6625
    },
    {
      "entropy": 6.2633123874664305,
      "epoch": 0.5793935156864458,
      "grad_norm": 5.90625,
      "learning_rate": 0.0004972706734696442,
      "loss": 6.09,
      "mean_token_accuracy": 0.13971349224448204,
      "num_tokens": 13435506.0,
      "step": 6630
    },
    {
      "entropy": 6.343959188461303,
      "epoch": 0.5798304640391506,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004972658324764984,
      "loss": 6.1218,
      "mean_token_accuracy": 0.14166830033063887,
      "num_tokens": 13445313.0,
      "step": 6635
    },
    {
      "entropy": 6.2723253726959225,
      "epoch": 0.5802674123918553,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004972609872201775,
      "loss": 6.0481,
      "mean_token_accuracy": 0.1416056774556637,
      "num_tokens": 13455312.0,
      "step": 6640
    },
    {
      "entropy": 6.261868619918824,
      "epoch": 0.58070436074456,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004972561377007744,
      "loss": 6.1891,
      "mean_token_accuracy": 0.13037768751382828,
      "num_tokens": 13465721.0,
      "step": 6645
    },
    {
      "entropy": 6.351018190383911,
      "epoch": 0.5811413090972647,
      "grad_norm": 10.8125,
      "learning_rate": 0.0004972512839183822,
      "loss": 6.235,
      "mean_token_accuracy": 0.1339089147746563,
      "num_tokens": 13477185.0,
      "step": 6650
    },
    {
      "entropy": 6.357801485061645,
      "epoch": 0.5815782574499694,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004972464258730939,
      "loss": 6.1288,
      "mean_token_accuracy": 0.14132801145315171,
      "num_tokens": 13487846.0,
      "step": 6655
    },
    {
      "entropy": 6.2941771984100345,
      "epoch": 0.5820152058026741,
      "grad_norm": 2.125,
      "learning_rate": 0.0004972415635650027,
      "loss": 6.1946,
      "mean_token_accuracy": 0.14400067999958993,
      "num_tokens": 13498175.0,
      "step": 6660
    },
    {
      "entropy": 6.358014297485352,
      "epoch": 0.5824521541553789,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004972366969942017,
      "loss": 6.1362,
      "mean_token_accuracy": 0.1327988550066948,
      "num_tokens": 13506864.0,
      "step": 6665
    },
    {
      "entropy": 6.289809226989746,
      "epoch": 0.5828891025080836,
      "grad_norm": 5.84375,
      "learning_rate": 0.0004972318261607845,
      "loss": 6.1486,
      "mean_token_accuracy": 0.13644013032317162,
      "num_tokens": 13516208.0,
      "step": 6670
    },
    {
      "entropy": 6.331784009933472,
      "epoch": 0.5833260508607883,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004972269510648444,
      "loss": 6.2081,
      "mean_token_accuracy": 0.13210436776280404,
      "num_tokens": 13526902.0,
      "step": 6675
    },
    {
      "entropy": 6.355490064620971,
      "epoch": 0.583762999213493,
      "grad_norm": 3.703125,
      "learning_rate": 0.000497222071706475,
      "loss": 6.1278,
      "mean_token_accuracy": 0.138092490285635,
      "num_tokens": 13537001.0,
      "step": 6680
    },
    {
      "entropy": 6.282208824157715,
      "epoch": 0.5841999475661976,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004972171880857698,
      "loss": 6.1018,
      "mean_token_accuracy": 0.1446065366268158,
      "num_tokens": 13546971.0,
      "step": 6685
    },
    {
      "entropy": 6.310001945495605,
      "epoch": 0.5846368959189023,
      "grad_norm": 18.125,
      "learning_rate": 0.0004972123002028226,
      "loss": 6.1016,
      "mean_token_accuracy": 0.14149885922670363,
      "num_tokens": 13557202.0,
      "step": 6690
    },
    {
      "entropy": 6.229309225082398,
      "epoch": 0.5850738442716071,
      "grad_norm": 3.125,
      "learning_rate": 0.0004972074080577269,
      "loss": 6.0796,
      "mean_token_accuracy": 0.14349437952041627,
      "num_tokens": 13567291.0,
      "step": 6695
    },
    {
      "entropy": 6.310549211502075,
      "epoch": 0.5855107926243118,
      "grad_norm": 4.0,
      "learning_rate": 0.0004972025116505767,
      "loss": 6.1662,
      "mean_token_accuracy": 0.14298462718725205,
      "num_tokens": 13576864.0,
      "step": 6700
    },
    {
      "entropy": 6.333677148818969,
      "epoch": 0.5859477409770165,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004971976109814658,
      "loss": 6.1154,
      "mean_token_accuracy": 0.14293619692325593,
      "num_tokens": 13586728.0,
      "step": 6705
    },
    {
      "entropy": 6.335979557037353,
      "epoch": 0.5863846893297212,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004971927060504885,
      "loss": 6.1452,
      "mean_token_accuracy": 0.13798118457198144,
      "num_tokens": 13596875.0,
      "step": 6710
    },
    {
      "entropy": 6.360797977447509,
      "epoch": 0.5868216376824259,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004971877968577386,
      "loss": 6.187,
      "mean_token_accuracy": 0.12828269079327584,
      "num_tokens": 13606483.0,
      "step": 6715
    },
    {
      "entropy": 6.362807178497315,
      "epoch": 0.5872585860351307,
      "grad_norm": 2.5,
      "learning_rate": 0.0004971828834033102,
      "loss": 6.2249,
      "mean_token_accuracy": 0.13809165954589844,
      "num_tokens": 13616780.0,
      "step": 6720
    },
    {
      "entropy": 6.308199882507324,
      "epoch": 0.5876955343878354,
      "grad_norm": 4.5,
      "learning_rate": 0.0004971779656872978,
      "loss": 6.1758,
      "mean_token_accuracy": 0.1352609895169735,
      "num_tokens": 13626418.0,
      "step": 6725
    },
    {
      "entropy": 6.36847448348999,
      "epoch": 0.5881324827405401,
      "grad_norm": 4.375,
      "learning_rate": 0.0004971730437097955,
      "loss": 6.231,
      "mean_token_accuracy": 0.13466500863432884,
      "num_tokens": 13636956.0,
      "step": 6730
    },
    {
      "entropy": 6.362085008621216,
      "epoch": 0.5885694310932448,
      "grad_norm": 2.828125,
      "learning_rate": 0.000497168117470898,
      "loss": 6.1592,
      "mean_token_accuracy": 0.13687244802713394,
      "num_tokens": 13647458.0,
      "step": 6735
    },
    {
      "entropy": 6.269703054428101,
      "epoch": 0.5890063794459495,
      "grad_norm": 5.5625,
      "learning_rate": 0.0004971631869706995,
      "loss": 6.0995,
      "mean_token_accuracy": 0.1452358029782772,
      "num_tokens": 13657655.0,
      "step": 6740
    },
    {
      "entropy": 6.319853115081787,
      "epoch": 0.5894433277986542,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004971582522092946,
      "loss": 6.2271,
      "mean_token_accuracy": 0.12915387377142906,
      "num_tokens": 13668472.0,
      "step": 6745
    },
    {
      "entropy": 6.306661701202392,
      "epoch": 0.589880276151359,
      "grad_norm": 2.46875,
      "learning_rate": 0.000497153313186778,
      "loss": 6.0775,
      "mean_token_accuracy": 0.1425430193543434,
      "num_tokens": 13677319.0,
      "step": 6750
    },
    {
      "entropy": 6.3182823181152346,
      "epoch": 0.5903172245040637,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004971483699032445,
      "loss": 6.15,
      "mean_token_accuracy": 0.13728732392191886,
      "num_tokens": 13688037.0,
      "step": 6755
    },
    {
      "entropy": 6.380808782577515,
      "epoch": 0.5907541728567683,
      "grad_norm": 5.625,
      "learning_rate": 0.0004971434223587888,
      "loss": 6.2344,
      "mean_token_accuracy": 0.12719021663069724,
      "num_tokens": 13698441.0,
      "step": 6760
    },
    {
      "entropy": 6.311367702484131,
      "epoch": 0.591191121209473,
      "grad_norm": 3.953125,
      "learning_rate": 0.000497138470553506,
      "loss": 6.1575,
      "mean_token_accuracy": 0.13726550638675689,
      "num_tokens": 13708140.0,
      "step": 6765
    },
    {
      "entropy": 6.345147705078125,
      "epoch": 0.5916280695621777,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004971335144874909,
      "loss": 6.1682,
      "mean_token_accuracy": 0.1383343979716301,
      "num_tokens": 13718069.0,
      "step": 6770
    },
    {
      "entropy": 6.2756141185760494,
      "epoch": 0.5920650179148824,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004971285541608385,
      "loss": 6.1296,
      "mean_token_accuracy": 0.14841375946998597,
      "num_tokens": 13727932.0,
      "step": 6775
    },
    {
      "entropy": 6.4002752780914305,
      "epoch": 0.5925019662675872,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004971235895736441,
      "loss": 6.2114,
      "mean_token_accuracy": 0.12681224048137665,
      "num_tokens": 13738239.0,
      "step": 6780
    },
    {
      "entropy": 6.368058204650879,
      "epoch": 0.5929389146202919,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004971186207260029,
      "loss": 6.1611,
      "mean_token_accuracy": 0.13095037937164306,
      "num_tokens": 13748622.0,
      "step": 6785
    },
    {
      "entropy": 6.222450685501099,
      "epoch": 0.5933758629729966,
      "grad_norm": 9.625,
      "learning_rate": 0.00049711364761801,
      "loss": 6.0928,
      "mean_token_accuracy": 0.14112447202205658,
      "num_tokens": 13759221.0,
      "step": 6790
    },
    {
      "entropy": 6.279159927368164,
      "epoch": 0.5938128113257013,
      "grad_norm": 4.0625,
      "learning_rate": 0.000497108670249761,
      "loss": 6.1065,
      "mean_token_accuracy": 0.13884979709982873,
      "num_tokens": 13768343.0,
      "step": 6795
    },
    {
      "entropy": 6.275318622589111,
      "epoch": 0.594249759678406,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004971036886213515,
      "loss": 6.1151,
      "mean_token_accuracy": 0.13884662017226218,
      "num_tokens": 13778225.0,
      "step": 6800
    },
    {
      "entropy": 6.294787120819092,
      "epoch": 0.5946867080311107,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004970987027328767,
      "loss": 6.1887,
      "mean_token_accuracy": 0.1303704373538494,
      "num_tokens": 13788633.0,
      "step": 6805
    },
    {
      "entropy": 6.349935913085938,
      "epoch": 0.5951236563838155,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004970937125844325,
      "loss": 6.0731,
      "mean_token_accuracy": 0.14252061694860457,
      "num_tokens": 13798172.0,
      "step": 6810
    },
    {
      "entropy": 6.251861238479615,
      "epoch": 0.5955606047365202,
      "grad_norm": 5.1875,
      "learning_rate": 0.0004970887181761144,
      "loss": 6.0973,
      "mean_token_accuracy": 0.14462856948375702,
      "num_tokens": 13807284.0,
      "step": 6815
    },
    {
      "entropy": 6.379451656341553,
      "epoch": 0.5959975530892249,
      "grad_norm": 7.1875,
      "learning_rate": 0.0004970837195080184,
      "loss": 6.1545,
      "mean_token_accuracy": 0.13342333361506462,
      "num_tokens": 13816807.0,
      "step": 6820
    },
    {
      "entropy": 6.3940691471099855,
      "epoch": 0.5964345014419296,
      "grad_norm": 6.0625,
      "learning_rate": 0.0004970787165802403,
      "loss": 6.118,
      "mean_token_accuracy": 0.13524844348430634,
      "num_tokens": 13827635.0,
      "step": 6825
    },
    {
      "entropy": 6.31161675453186,
      "epoch": 0.5968714497946342,
      "grad_norm": 5.71875,
      "learning_rate": 0.000497073709392876,
      "loss": 6.1676,
      "mean_token_accuracy": 0.13472561687231063,
      "num_tokens": 13838060.0,
      "step": 6830
    },
    {
      "entropy": 6.362737655639648,
      "epoch": 0.5973083981473389,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004970686979460216,
      "loss": 6.177,
      "mean_token_accuracy": 0.13541199788451194,
      "num_tokens": 13848669.0,
      "step": 6835
    },
    {
      "entropy": 6.341948461532593,
      "epoch": 0.5977453465000437,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004970636822397733,
      "loss": 6.1378,
      "mean_token_accuracy": 0.14245806485414506,
      "num_tokens": 13857949.0,
      "step": 6840
    },
    {
      "entropy": 6.303737688064575,
      "epoch": 0.5981822948527484,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004970586622742272,
      "loss": 6.1558,
      "mean_token_accuracy": 0.1407779723405838,
      "num_tokens": 13867726.0,
      "step": 6845
    },
    {
      "entropy": 6.399082803726197,
      "epoch": 0.5986192432054531,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004970536380494795,
      "loss": 6.2199,
      "mean_token_accuracy": 0.1303854450583458,
      "num_tokens": 13878072.0,
      "step": 6850
    },
    {
      "entropy": 6.320159959793091,
      "epoch": 0.5990561915581578,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004970486095656268,
      "loss": 6.0914,
      "mean_token_accuracy": 0.14282031282782554,
      "num_tokens": 13887771.0,
      "step": 6855
    },
    {
      "entropy": 6.339229726791382,
      "epoch": 0.5994931399108625,
      "grad_norm": 8.1875,
      "learning_rate": 0.0004970435768227652,
      "loss": 6.1338,
      "mean_token_accuracy": 0.13607884123921393,
      "num_tokens": 13897867.0,
      "step": 6860
    },
    {
      "entropy": 6.341369199752807,
      "epoch": 0.5999300882635672,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004970385398209916,
      "loss": 6.2143,
      "mean_token_accuracy": 0.13161174058914185,
      "num_tokens": 13909171.0,
      "step": 6865
    },
    {
      "entropy": 6.393149471282959,
      "epoch": 0.600367036616272,
      "grad_norm": 4.9375,
      "learning_rate": 0.0004970334985604024,
      "loss": 6.2082,
      "mean_token_accuracy": 0.1333576537668705,
      "num_tokens": 13920374.0,
      "step": 6870
    },
    {
      "entropy": 6.360220670700073,
      "epoch": 0.6008039849689767,
      "grad_norm": 6.125,
      "learning_rate": 0.0004970284530410945,
      "loss": 6.2093,
      "mean_token_accuracy": 0.13334948271512986,
      "num_tokens": 13930150.0,
      "step": 6875
    },
    {
      "entropy": 6.3598161220550535,
      "epoch": 0.6012409333216814,
      "grad_norm": 4.125,
      "learning_rate": 0.0004970234032631644,
      "loss": 6.1472,
      "mean_token_accuracy": 0.1360732689499855,
      "num_tokens": 13940553.0,
      "step": 6880
    },
    {
      "entropy": 6.344177913665772,
      "epoch": 0.6016778816743861,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004970183492267091,
      "loss": 6.2598,
      "mean_token_accuracy": 0.1356998786330223,
      "num_tokens": 13951086.0,
      "step": 6885
    },
    {
      "entropy": 6.317593240737915,
      "epoch": 0.6021148300270908,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004970132909318254,
      "loss": 6.1697,
      "mean_token_accuracy": 0.1353231117129326,
      "num_tokens": 13960684.0,
      "step": 6890
    },
    {
      "entropy": 6.218622016906738,
      "epoch": 0.6025517783797955,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004970082283786107,
      "loss": 6.0058,
      "mean_token_accuracy": 0.15277925729751587,
      "num_tokens": 13971409.0,
      "step": 6895
    },
    {
      "entropy": 6.338640642166138,
      "epoch": 0.6029887267325003,
      "grad_norm": 55.75,
      "learning_rate": 0.0004970031615671617,
      "loss": 6.1865,
      "mean_token_accuracy": 0.1368209384381771,
      "num_tokens": 13981185.0,
      "step": 6900
    },
    {
      "entropy": 6.359845113754273,
      "epoch": 0.603425675085205,
      "grad_norm": 15.375,
      "learning_rate": 0.0004969980904975758,
      "loss": 6.2016,
      "mean_token_accuracy": 0.13212283998727797,
      "num_tokens": 13991670.0,
      "step": 6905
    },
    {
      "entropy": 6.350120592117309,
      "epoch": 0.6038626234379096,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004969930151699501,
      "loss": 6.16,
      "mean_token_accuracy": 0.1345134235918522,
      "num_tokens": 14002343.0,
      "step": 6910
    },
    {
      "entropy": 6.364055728912353,
      "epoch": 0.6042995717906143,
      "grad_norm": 4.625,
      "learning_rate": 0.000496987935584382,
      "loss": 6.2192,
      "mean_token_accuracy": 0.1319100923836231,
      "num_tokens": 14013373.0,
      "step": 6915
    },
    {
      "entropy": 6.372997045516968,
      "epoch": 0.604736520143319,
      "grad_norm": 5.625,
      "learning_rate": 0.0004969828517409693,
      "loss": 6.1214,
      "mean_token_accuracy": 0.1412570893764496,
      "num_tokens": 14023233.0,
      "step": 6920
    },
    {
      "entropy": 6.347461080551147,
      "epoch": 0.6051734684960238,
      "grad_norm": 3.96875,
      "learning_rate": 0.000496977763639809,
      "loss": 6.1658,
      "mean_token_accuracy": 0.1363045133650303,
      "num_tokens": 14033054.0,
      "step": 6925
    },
    {
      "entropy": 6.3707435607910154,
      "epoch": 0.6056104168487285,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004969726712809988,
      "loss": 6.1063,
      "mean_token_accuracy": 0.133543062210083,
      "num_tokens": 14042039.0,
      "step": 6930
    },
    {
      "entropy": 6.4379918575286865,
      "epoch": 0.6060473652014332,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004969675746646364,
      "loss": 6.2545,
      "mean_token_accuracy": 0.12576044723391533,
      "num_tokens": 14051957.0,
      "step": 6935
    },
    {
      "entropy": 6.353121519088745,
      "epoch": 0.6064843135541379,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004969624737908197,
      "loss": 6.1579,
      "mean_token_accuracy": 0.13342536017298698,
      "num_tokens": 14062438.0,
      "step": 6940
    },
    {
      "entropy": 6.308655166625977,
      "epoch": 0.6069212619068426,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004969573686596464,
      "loss": 6.1677,
      "mean_token_accuracy": 0.12700469344854354,
      "num_tokens": 14073196.0,
      "step": 6945
    },
    {
      "entropy": 6.410180759429932,
      "epoch": 0.6073582102595473,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004969522592712146,
      "loss": 6.1795,
      "mean_token_accuracy": 0.1318407915532589,
      "num_tokens": 14083462.0,
      "step": 6950
    },
    {
      "entropy": 6.324456691741943,
      "epoch": 0.6077951586122521,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004969471456256221,
      "loss": 6.2502,
      "mean_token_accuracy": 0.13041185066103936,
      "num_tokens": 14094194.0,
      "step": 6955
    },
    {
      "entropy": 6.330078363418579,
      "epoch": 0.6082321069649568,
      "grad_norm": 3.53125,
      "learning_rate": 0.000496942027722967,
      "loss": 6.1616,
      "mean_token_accuracy": 0.13390202224254608,
      "num_tokens": 14104553.0,
      "step": 6960
    },
    {
      "entropy": 6.425537824630737,
      "epoch": 0.6086690553176615,
      "grad_norm": 6.59375,
      "learning_rate": 0.0004969369055633476,
      "loss": 6.1179,
      "mean_token_accuracy": 0.13076433464884757,
      "num_tokens": 14114140.0,
      "step": 6965
    },
    {
      "entropy": 6.355055904388427,
      "epoch": 0.6091060036703662,
      "grad_norm": 4.71875,
      "learning_rate": 0.000496931779146862,
      "loss": 6.24,
      "mean_token_accuracy": 0.13108323216438295,
      "num_tokens": 14124723.0,
      "step": 6970
    },
    {
      "entropy": 6.322403049468994,
      "epoch": 0.6095429520230708,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004969266484736085,
      "loss": 6.0939,
      "mean_token_accuracy": 0.13703892305493354,
      "num_tokens": 14134030.0,
      "step": 6975
    },
    {
      "entropy": 6.276351737976074,
      "epoch": 0.6099799003757755,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004969215135436858,
      "loss": 6.077,
      "mean_token_accuracy": 0.1351178802549839,
      "num_tokens": 14144691.0,
      "step": 6980
    },
    {
      "entropy": 6.334858226776123,
      "epoch": 0.6104168487284803,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004969163743571921,
      "loss": 6.251,
      "mean_token_accuracy": 0.1316261798143387,
      "num_tokens": 14155114.0,
      "step": 6985
    },
    {
      "entropy": 6.434923791885376,
      "epoch": 0.610853797081185,
      "grad_norm": 3.9375,
      "learning_rate": 0.000496911230914226,
      "loss": 6.2458,
      "mean_token_accuracy": 0.12821090817451478,
      "num_tokens": 14166097.0,
      "step": 6990
    },
    {
      "entropy": 6.445264053344727,
      "epoch": 0.6112907454338897,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004969060832148862,
      "loss": 6.1467,
      "mean_token_accuracy": 0.12940879240632058,
      "num_tokens": 14176538.0,
      "step": 6995
    },
    {
      "entropy": 6.320191192626953,
      "epoch": 0.6117276937865944,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004969009312592715,
      "loss": 6.1497,
      "mean_token_accuracy": 0.13426011949777603,
      "num_tokens": 14187912.0,
      "step": 7000
    },
    {
      "entropy": 6.191708612442016,
      "epoch": 0.6121646421392991,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004968957750474806,
      "loss": 6.0494,
      "mean_token_accuracy": 0.14366793483495713,
      "num_tokens": 14198446.0,
      "step": 7005
    },
    {
      "entropy": 6.356047010421753,
      "epoch": 0.6126015904920038,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004968906145796126,
      "loss": 6.2574,
      "mean_token_accuracy": 0.13040417656302453,
      "num_tokens": 14207723.0,
      "step": 7010
    },
    {
      "entropy": 6.388284826278687,
      "epoch": 0.6130385388447086,
      "grad_norm": 7.8125,
      "learning_rate": 0.0004968854498557662,
      "loss": 6.1492,
      "mean_token_accuracy": 0.13261253163218498,
      "num_tokens": 14217590.0,
      "step": 7015
    },
    {
      "entropy": 6.34548020362854,
      "epoch": 0.6134754871974133,
      "grad_norm": 12.0,
      "learning_rate": 0.0004968802808760407,
      "loss": 6.2036,
      "mean_token_accuracy": 0.13060805723071098,
      "num_tokens": 14226629.0,
      "step": 7020
    },
    {
      "entropy": 6.326340198516846,
      "epoch": 0.613912435550118,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004968751076405351,
      "loss": 6.2563,
      "mean_token_accuracy": 0.12426506131887435,
      "num_tokens": 14235782.0,
      "step": 7025
    },
    {
      "entropy": 6.268276023864746,
      "epoch": 0.6143493839028227,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004968699301493488,
      "loss": 6.1368,
      "mean_token_accuracy": 0.1401347555220127,
      "num_tokens": 14246687.0,
      "step": 7030
    },
    {
      "entropy": 6.419150543212891,
      "epoch": 0.6147863322555274,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004968647484025808,
      "loss": 6.167,
      "mean_token_accuracy": 0.1361520580947399,
      "num_tokens": 14255961.0,
      "step": 7035
    },
    {
      "entropy": 6.390906524658203,
      "epoch": 0.6152232806082321,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004968595624003307,
      "loss": 6.1793,
      "mean_token_accuracy": 0.13171206116676332,
      "num_tokens": 14266024.0,
      "step": 7040
    },
    {
      "entropy": 6.316295480728149,
      "epoch": 0.6156602289609369,
      "grad_norm": 6.59375,
      "learning_rate": 0.000496854372142698,
      "loss": 6.2079,
      "mean_token_accuracy": 0.13027237132191657,
      "num_tokens": 14276590.0,
      "step": 7045
    },
    {
      "entropy": 6.286552572250367,
      "epoch": 0.6160971773136416,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004968491776297821,
      "loss": 6.2061,
      "mean_token_accuracy": 0.13636200949549676,
      "num_tokens": 14285992.0,
      "step": 7050
    },
    {
      "entropy": 6.310267114639283,
      "epoch": 0.6165341256663462,
      "grad_norm": 4.25,
      "learning_rate": 0.0004968439788616829,
      "loss": 6.099,
      "mean_token_accuracy": 0.13655925542116165,
      "num_tokens": 14295843.0,
      "step": 7055
    },
    {
      "entropy": 6.3398643970489506,
      "epoch": 0.6169710740190509,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004968387758384999,
      "loss": 6.128,
      "mean_token_accuracy": 0.1405101127922535,
      "num_tokens": 14305643.0,
      "step": 7060
    },
    {
      "entropy": 6.366613101959229,
      "epoch": 0.6174080223717556,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004968335685603328,
      "loss": 6.1262,
      "mean_token_accuracy": 0.13317034766077995,
      "num_tokens": 14315907.0,
      "step": 7065
    },
    {
      "entropy": 6.359871578216553,
      "epoch": 0.6178449707244603,
      "grad_norm": 5.40625,
      "learning_rate": 0.0004968283570272817,
      "loss": 6.1844,
      "mean_token_accuracy": 0.12782085537910462,
      "num_tokens": 14325893.0,
      "step": 7070
    },
    {
      "entropy": 6.358923864364624,
      "epoch": 0.6182819190771651,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004968231412394464,
      "loss": 6.2003,
      "mean_token_accuracy": 0.13409364894032477,
      "num_tokens": 14337510.0,
      "step": 7075
    },
    {
      "entropy": 6.273475790023804,
      "epoch": 0.6187188674298698,
      "grad_norm": 5.53125,
      "learning_rate": 0.0004968179211969272,
      "loss": 6.1212,
      "mean_token_accuracy": 0.13653568550944328,
      "num_tokens": 14346373.0,
      "step": 7080
    },
    {
      "entropy": 6.189327239990234,
      "epoch": 0.6191558157825745,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004968126968998239,
      "loss": 6.0052,
      "mean_token_accuracy": 0.1548612542450428,
      "num_tokens": 14356303.0,
      "step": 7085
    },
    {
      "entropy": 6.390761184692383,
      "epoch": 0.6195927641352792,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004968074683482369,
      "loss": 6.1739,
      "mean_token_accuracy": 0.1371545359492302,
      "num_tokens": 14366171.0,
      "step": 7090
    },
    {
      "entropy": 6.385332822799683,
      "epoch": 0.6200297124879839,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004968022355422663,
      "loss": 6.2293,
      "mean_token_accuracy": 0.13305337429046632,
      "num_tokens": 14376133.0,
      "step": 7095
    },
    {
      "entropy": 6.343425273895264,
      "epoch": 0.6204666608406886,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004967969984820128,
      "loss": 6.1668,
      "mean_token_accuracy": 0.13266517072916031,
      "num_tokens": 14385670.0,
      "step": 7100
    },
    {
      "entropy": 6.340363454818726,
      "epoch": 0.6209036091933934,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004967917571675765,
      "loss": 6.1517,
      "mean_token_accuracy": 0.1349860928952694,
      "num_tokens": 14396002.0,
      "step": 7105
    },
    {
      "entropy": 6.2696342945098875,
      "epoch": 0.6213405575460981,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004967865115990582,
      "loss": 6.2383,
      "mean_token_accuracy": 0.12619198262691497,
      "num_tokens": 14406718.0,
      "step": 7110
    },
    {
      "entropy": 6.357182788848877,
      "epoch": 0.6217775058988028,
      "grad_norm": 4.125,
      "learning_rate": 0.0004967812617765583,
      "loss": 6.1724,
      "mean_token_accuracy": 0.13269176408648492,
      "num_tokens": 14417027.0,
      "step": 7115
    },
    {
      "entropy": 6.34207239151001,
      "epoch": 0.6222144542515075,
      "grad_norm": 5.125,
      "learning_rate": 0.0004967760077001775,
      "loss": 6.117,
      "mean_token_accuracy": 0.14298692122101783,
      "num_tokens": 14426885.0,
      "step": 7120
    },
    {
      "entropy": 6.265546083450317,
      "epoch": 0.6226514026042121,
      "grad_norm": 3.5,
      "learning_rate": 0.0004967707493700169,
      "loss": 6.1257,
      "mean_token_accuracy": 0.13680213391780854,
      "num_tokens": 14437743.0,
      "step": 7125
    },
    {
      "entropy": 6.282887697219849,
      "epoch": 0.623088350956917,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004967654867861769,
      "loss": 6.16,
      "mean_token_accuracy": 0.1403965324163437,
      "num_tokens": 14448808.0,
      "step": 7130
    },
    {
      "entropy": 6.385322999954224,
      "epoch": 0.6235252993096216,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004967602199487588,
      "loss": 6.1935,
      "mean_token_accuracy": 0.13783178329467774,
      "num_tokens": 14458560.0,
      "step": 7135
    },
    {
      "entropy": 6.29409646987915,
      "epoch": 0.6239622476623263,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004967549488578635,
      "loss": 6.2364,
      "mean_token_accuracy": 0.1326675944030285,
      "num_tokens": 14469375.0,
      "step": 7140
    },
    {
      "entropy": 6.406088399887085,
      "epoch": 0.624399196015031,
      "grad_norm": 2.203125,
      "learning_rate": 0.000496749673513592,
      "loss": 6.1583,
      "mean_token_accuracy": 0.13255954682826995,
      "num_tokens": 14478444.0,
      "step": 7145
    },
    {
      "entropy": 6.283285903930664,
      "epoch": 0.6248361443677357,
      "grad_norm": 2.875,
      "learning_rate": 0.0004967443939160456,
      "loss": 6.0344,
      "mean_token_accuracy": 0.1515200987458229,
      "num_tokens": 14487776.0,
      "step": 7150
    },
    {
      "entropy": 6.315397882461548,
      "epoch": 0.6252730927204404,
      "grad_norm": 3.0,
      "learning_rate": 0.0004967391100653255,
      "loss": 6.1855,
      "mean_token_accuracy": 0.136892918497324,
      "num_tokens": 14497879.0,
      "step": 7155
    },
    {
      "entropy": 6.300903606414795,
      "epoch": 0.6257100410731452,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004967338219615331,
      "loss": 6.0914,
      "mean_token_accuracy": 0.14213423877954484,
      "num_tokens": 14507753.0,
      "step": 7160
    },
    {
      "entropy": 6.342875289916992,
      "epoch": 0.6261469894258499,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004967285296047699,
      "loss": 6.0735,
      "mean_token_accuracy": 0.1423469103872776,
      "num_tokens": 14517086.0,
      "step": 7165
    },
    {
      "entropy": 6.339991331100464,
      "epoch": 0.6265839377785546,
      "grad_norm": 9.5,
      "learning_rate": 0.0004967232329951372,
      "loss": 6.1504,
      "mean_token_accuracy": 0.1302580177783966,
      "num_tokens": 14527097.0,
      "step": 7170
    },
    {
      "entropy": 6.302025365829468,
      "epoch": 0.6270208861312593,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004967179321327367,
      "loss": 6.1759,
      "mean_token_accuracy": 0.13884464353322984,
      "num_tokens": 14537924.0,
      "step": 7175
    },
    {
      "entropy": 6.347482013702392,
      "epoch": 0.627457834483964,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004967126270176702,
      "loss": 6.1905,
      "mean_token_accuracy": 0.12967984080314637,
      "num_tokens": 14547542.0,
      "step": 7180
    },
    {
      "entropy": 6.271533727645874,
      "epoch": 0.6278947828366687,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004967073176500392,
      "loss": 6.1143,
      "mean_token_accuracy": 0.14234207719564437,
      "num_tokens": 14557150.0,
      "step": 7185
    },
    {
      "entropy": 6.360161018371582,
      "epoch": 0.6283317311893735,
      "grad_norm": 5.5,
      "learning_rate": 0.0004967020040299458,
      "loss": 6.1153,
      "mean_token_accuracy": 0.14050035178661346,
      "num_tokens": 14567242.0,
      "step": 7190
    },
    {
      "entropy": 6.289652585983276,
      "epoch": 0.6287686795420782,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004966966861574917,
      "loss": 6.1063,
      "mean_token_accuracy": 0.1372591659426689,
      "num_tokens": 14576631.0,
      "step": 7195
    },
    {
      "entropy": 6.303082895278931,
      "epoch": 0.6292056278947828,
      "grad_norm": 2.25,
      "learning_rate": 0.000496691364032779,
      "loss": 6.1854,
      "mean_token_accuracy": 0.13459793478250504,
      "num_tokens": 14587731.0,
      "step": 7200
    },
    {
      "entropy": 6.384452438354492,
      "epoch": 0.6296425762474875,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004966860376559097,
      "loss": 6.188,
      "mean_token_accuracy": 0.13302761688828468,
      "num_tokens": 14598136.0,
      "step": 7205
    },
    {
      "entropy": 6.293945074081421,
      "epoch": 0.6300795246001922,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004966807070269862,
      "loss": 6.157,
      "mean_token_accuracy": 0.13672245666384697,
      "num_tokens": 14607243.0,
      "step": 7210
    },
    {
      "entropy": 6.309554576873779,
      "epoch": 0.6305164729528969,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004966753721461105,
      "loss": 6.16,
      "mean_token_accuracy": 0.136593097448349,
      "num_tokens": 14617094.0,
      "step": 7215
    },
    {
      "entropy": 6.379037666320801,
      "epoch": 0.6309534213056017,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004966700330133849,
      "loss": 6.1939,
      "mean_token_accuracy": 0.13255172744393348,
      "num_tokens": 14626486.0,
      "step": 7220
    },
    {
      "entropy": 6.217536783218383,
      "epoch": 0.6313903696583064,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004966646896289121,
      "loss": 6.0402,
      "mean_token_accuracy": 0.1502256378531456,
      "num_tokens": 14637170.0,
      "step": 7225
    },
    {
      "entropy": 6.293237590789795,
      "epoch": 0.6318273180110111,
      "grad_norm": 2.0,
      "learning_rate": 0.0004966593419927944,
      "loss": 6.1139,
      "mean_token_accuracy": 0.13644453063607215,
      "num_tokens": 14646396.0,
      "step": 7230
    },
    {
      "entropy": 6.382175016403198,
      "epoch": 0.6322642663637158,
      "grad_norm": 2.75,
      "learning_rate": 0.0004966539901051343,
      "loss": 6.1744,
      "mean_token_accuracy": 0.12860749289393425,
      "num_tokens": 14656097.0,
      "step": 7235
    },
    {
      "entropy": 6.311509609222412,
      "epoch": 0.6327012147164205,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004966486339660345,
      "loss": 6.1473,
      "mean_token_accuracy": 0.13462281972169876,
      "num_tokens": 14667124.0,
      "step": 7240
    },
    {
      "entropy": 6.310620927810669,
      "epoch": 0.6331381630691252,
      "grad_norm": 3.25,
      "learning_rate": 0.0004966432735755978,
      "loss": 6.2222,
      "mean_token_accuracy": 0.13140853717923165,
      "num_tokens": 14677427.0,
      "step": 7245
    },
    {
      "entropy": 6.292732667922974,
      "epoch": 0.63357511142183,
      "grad_norm": 2.140625,
      "learning_rate": 0.000496637908933927,
      "loss": 6.0643,
      "mean_token_accuracy": 0.138239386677742,
      "num_tokens": 14686429.0,
      "step": 7250
    },
    {
      "entropy": 6.267800521850586,
      "epoch": 0.6340120597745347,
      "grad_norm": 2.640625,
      "learning_rate": 0.000496632540041125,
      "loss": 6.1266,
      "mean_token_accuracy": 0.14233689829707147,
      "num_tokens": 14696264.0,
      "step": 7255
    },
    {
      "entropy": 6.349011468887329,
      "epoch": 0.6344490081272394,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004966271668972946,
      "loss": 6.1533,
      "mean_token_accuracy": 0.13102434501051902,
      "num_tokens": 14705935.0,
      "step": 7260
    },
    {
      "entropy": 6.348269033432007,
      "epoch": 0.6348859564799441,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004966217895025391,
      "loss": 6.1238,
      "mean_token_accuracy": 0.13866823092103003,
      "num_tokens": 14715978.0,
      "step": 7265
    },
    {
      "entropy": 6.338259506225586,
      "epoch": 0.6353229048326487,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004966164078569616,
      "loss": 6.116,
      "mean_token_accuracy": 0.13448933362960816,
      "num_tokens": 14726170.0,
      "step": 7270
    },
    {
      "entropy": 6.323014926910401,
      "epoch": 0.6357598531853534,
      "grad_norm": 2.875,
      "learning_rate": 0.0004966110219606652,
      "loss": 6.1125,
      "mean_token_accuracy": 0.13669896498322487,
      "num_tokens": 14735775.0,
      "step": 7275
    },
    {
      "entropy": 6.28701605796814,
      "epoch": 0.6361968015380582,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004966056318137532,
      "loss": 6.1122,
      "mean_token_accuracy": 0.1354524552822113,
      "num_tokens": 14746143.0,
      "step": 7280
    },
    {
      "entropy": 6.267648029327392,
      "epoch": 0.6366337498907629,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004966002374163293,
      "loss": 6.1562,
      "mean_token_accuracy": 0.14047074541449547,
      "num_tokens": 14756707.0,
      "step": 7285
    },
    {
      "entropy": 6.325721645355225,
      "epoch": 0.6370706982434676,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004965948387684964,
      "loss": 6.1557,
      "mean_token_accuracy": 0.1365771099925041,
      "num_tokens": 14765778.0,
      "step": 7290
    },
    {
      "entropy": 6.360436630249024,
      "epoch": 0.6375076465961723,
      "grad_norm": 2.0,
      "learning_rate": 0.0004965894358703587,
      "loss": 6.1595,
      "mean_token_accuracy": 0.13039254695177077,
      "num_tokens": 14775264.0,
      "step": 7295
    },
    {
      "entropy": 6.343347072601318,
      "epoch": 0.637944594948877,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004965840287220193,
      "loss": 6.1235,
      "mean_token_accuracy": 0.1351149559020996,
      "num_tokens": 14785754.0,
      "step": 7300
    },
    {
      "entropy": 6.30988187789917,
      "epoch": 0.6383815433015817,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004965786173235821,
      "loss": 6.1935,
      "mean_token_accuracy": 0.13218039646744728,
      "num_tokens": 14795936.0,
      "step": 7305
    },
    {
      "entropy": 6.3907781600952145,
      "epoch": 0.6388184916542865,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004965732016751511,
      "loss": 6.1658,
      "mean_token_accuracy": 0.13335740938782692,
      "num_tokens": 14805721.0,
      "step": 7310
    },
    {
      "entropy": 6.380353546142578,
      "epoch": 0.6392554400069912,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004965677817768299,
      "loss": 6.197,
      "mean_token_accuracy": 0.13154036104679107,
      "num_tokens": 14816610.0,
      "step": 7315
    },
    {
      "entropy": 6.312654209136963,
      "epoch": 0.6396923883596959,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004965623576287225,
      "loss": 6.1507,
      "mean_token_accuracy": 0.13144177570939064,
      "num_tokens": 14826126.0,
      "step": 7320
    },
    {
      "entropy": 6.210688495635987,
      "epoch": 0.6401293367124006,
      "grad_norm": 5.84375,
      "learning_rate": 0.0004965569292309331,
      "loss": 6.0681,
      "mean_token_accuracy": 0.14340270310640335,
      "num_tokens": 14835657.0,
      "step": 7325
    },
    {
      "entropy": 6.249003791809082,
      "epoch": 0.6405662850651053,
      "grad_norm": 2.25,
      "learning_rate": 0.0004965514965835655,
      "loss": 6.1574,
      "mean_token_accuracy": 0.13696584478020668,
      "num_tokens": 14845430.0,
      "step": 7330
    },
    {
      "entropy": 6.291607141494751,
      "epoch": 0.6410032334178101,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004965460596867243,
      "loss": 6.0925,
      "mean_token_accuracy": 0.14165626242756843,
      "num_tokens": 14855459.0,
      "step": 7335
    },
    {
      "entropy": 6.377842092514038,
      "epoch": 0.6414401817705148,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004965406185405135,
      "loss": 6.1704,
      "mean_token_accuracy": 0.13619712814688684,
      "num_tokens": 14865375.0,
      "step": 7340
    },
    {
      "entropy": 6.382111024856568,
      "epoch": 0.6418771301232195,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004965351731450376,
      "loss": 6.1176,
      "mean_token_accuracy": 0.140096253156662,
      "num_tokens": 14875211.0,
      "step": 7345
    },
    {
      "entropy": 6.305885076522827,
      "epoch": 0.6423140784759241,
      "grad_norm": 2.140625,
      "learning_rate": 0.000496529723500401,
      "loss": 6.1578,
      "mean_token_accuracy": 0.13519529476761818,
      "num_tokens": 14886221.0,
      "step": 7350
    },
    {
      "entropy": 6.360910415649414,
      "epoch": 0.6427510268286288,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004965242696067081,
      "loss": 6.1194,
      "mean_token_accuracy": 0.1420270599424839,
      "num_tokens": 14896002.0,
      "step": 7355
    },
    {
      "entropy": 6.278655576705932,
      "epoch": 0.6431879751813335,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004965188114640637,
      "loss": 6.1754,
      "mean_token_accuracy": 0.1330043725669384,
      "num_tokens": 14906311.0,
      "step": 7360
    },
    {
      "entropy": 6.365869235992432,
      "epoch": 0.6436249235340383,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004965133490725725,
      "loss": 6.1453,
      "mean_token_accuracy": 0.1350757323205471,
      "num_tokens": 14915890.0,
      "step": 7365
    },
    {
      "entropy": 6.378992938995362,
      "epoch": 0.644061871886743,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004965078824323391,
      "loss": 6.1085,
      "mean_token_accuracy": 0.13755420818924904,
      "num_tokens": 14925891.0,
      "step": 7370
    },
    {
      "entropy": 6.283592891693115,
      "epoch": 0.6444988202394477,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004965024115434685,
      "loss": 6.1582,
      "mean_token_accuracy": 0.1345057025551796,
      "num_tokens": 14937272.0,
      "step": 7375
    },
    {
      "entropy": 6.2707963466644285,
      "epoch": 0.6449357685921524,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004964969364060656,
      "loss": 6.1874,
      "mean_token_accuracy": 0.13462205603718758,
      "num_tokens": 14948135.0,
      "step": 7380
    },
    {
      "entropy": 6.414066171646118,
      "epoch": 0.6453727169448571,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004964914570202353,
      "loss": 6.2413,
      "mean_token_accuracy": 0.12213373556733131,
      "num_tokens": 14959254.0,
      "step": 7385
    },
    {
      "entropy": 6.346686363220215,
      "epoch": 0.6458096652975618,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004964859733860829,
      "loss": 6.1624,
      "mean_token_accuracy": 0.1392471708357334,
      "num_tokens": 14969313.0,
      "step": 7390
    },
    {
      "entropy": 6.251855516433716,
      "epoch": 0.6462466136502666,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004964804855037134,
      "loss": 6.1414,
      "mean_token_accuracy": 0.13435712680220604,
      "num_tokens": 14978963.0,
      "step": 7395
    },
    {
      "entropy": 6.413202095031738,
      "epoch": 0.6466835620029713,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004964749933732322,
      "loss": 6.165,
      "mean_token_accuracy": 0.13349838182330132,
      "num_tokens": 14989355.0,
      "step": 7400
    },
    {
      "entropy": 6.319285202026367,
      "epoch": 0.647120510355676,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004964694969947445,
      "loss": 6.1172,
      "mean_token_accuracy": 0.1404731571674347,
      "num_tokens": 14998995.0,
      "step": 7405
    },
    {
      "entropy": 6.266659355163574,
      "epoch": 0.6475574587083807,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004964639963683558,
      "loss": 6.1815,
      "mean_token_accuracy": 0.13053784519433975,
      "num_tokens": 15008762.0,
      "step": 7410
    },
    {
      "entropy": 6.399480962753296,
      "epoch": 0.6479944070610854,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004964584914941716,
      "loss": 6.1732,
      "mean_token_accuracy": 0.13472879379987718,
      "num_tokens": 15020062.0,
      "step": 7415
    },
    {
      "entropy": 6.437808322906494,
      "epoch": 0.64843135541379,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004964529823722975,
      "loss": 6.3081,
      "mean_token_accuracy": 0.12480103820562363,
      "num_tokens": 15030440.0,
      "step": 7420
    },
    {
      "entropy": 6.357998037338257,
      "epoch": 0.6488683037664948,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004964474690028393,
      "loss": 6.1297,
      "mean_token_accuracy": 0.13256605863571166,
      "num_tokens": 15041709.0,
      "step": 7425
    },
    {
      "entropy": 6.3610413551330565,
      "epoch": 0.6493052521191995,
      "grad_norm": 6.90625,
      "learning_rate": 0.0004964419513859025,
      "loss": 6.2451,
      "mean_token_accuracy": 0.13353117629885675,
      "num_tokens": 15052277.0,
      "step": 7430
    },
    {
      "entropy": 6.323017311096192,
      "epoch": 0.6497422004719042,
      "grad_norm": 3.359375,
      "learning_rate": 0.000496436429521593,
      "loss": 6.1472,
      "mean_token_accuracy": 0.13462546169757844,
      "num_tokens": 15061870.0,
      "step": 7435
    },
    {
      "entropy": 6.278857231140137,
      "epoch": 0.6501791488246089,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004964309034100168,
      "loss": 6.0203,
      "mean_token_accuracy": 0.13972338140010834,
      "num_tokens": 15071403.0,
      "step": 7440
    },
    {
      "entropy": 6.290305805206299,
      "epoch": 0.6506160971773136,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004964253730512799,
      "loss": 6.1283,
      "mean_token_accuracy": 0.136251313239336,
      "num_tokens": 15081651.0,
      "step": 7445
    },
    {
      "entropy": 6.313625907897949,
      "epoch": 0.6510530455300183,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004964198384454882,
      "loss": 6.1868,
      "mean_token_accuracy": 0.133196110278368,
      "num_tokens": 15093403.0,
      "step": 7450
    },
    {
      "entropy": 6.381255149841309,
      "epoch": 0.6514899938827231,
      "grad_norm": 3.0625,
      "learning_rate": 0.000496414299592748,
      "loss": 6.2296,
      "mean_token_accuracy": 0.13075929880142212,
      "num_tokens": 15103618.0,
      "step": 7455
    },
    {
      "entropy": 6.315995502471924,
      "epoch": 0.6519269422354278,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004964087564931656,
      "loss": 6.1704,
      "mean_token_accuracy": 0.13908491432666778,
      "num_tokens": 15113926.0,
      "step": 7460
    },
    {
      "entropy": 6.4091630458831785,
      "epoch": 0.6523638905881325,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004964032091468472,
      "loss": 6.2177,
      "mean_token_accuracy": 0.13231534510850906,
      "num_tokens": 15125302.0,
      "step": 7465
    },
    {
      "entropy": 6.377340793609619,
      "epoch": 0.6528008389408372,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004963976575538991,
      "loss": 6.1148,
      "mean_token_accuracy": 0.14140793234109877,
      "num_tokens": 15135943.0,
      "step": 7470
    },
    {
      "entropy": 6.280066013336182,
      "epoch": 0.6532377872935419,
      "grad_norm": 4.28125,
      "learning_rate": 0.000496392101714428,
      "loss": 6.1643,
      "mean_token_accuracy": 0.13941967934370042,
      "num_tokens": 15145001.0,
      "step": 7475
    },
    {
      "entropy": 6.311770868301392,
      "epoch": 0.6536747356462466,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004963865416285405,
      "loss": 6.0664,
      "mean_token_accuracy": 0.1385023608803749,
      "num_tokens": 15154567.0,
      "step": 7480
    },
    {
      "entropy": 6.364741420745849,
      "epoch": 0.6541116839989514,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004963809772963431,
      "loss": 6.245,
      "mean_token_accuracy": 0.13161061927676201,
      "num_tokens": 15164311.0,
      "step": 7485
    },
    {
      "entropy": 6.328248071670532,
      "epoch": 0.6545486323516561,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004963754087179425,
      "loss": 6.1371,
      "mean_token_accuracy": 0.13413339108228683,
      "num_tokens": 15175217.0,
      "step": 7490
    },
    {
      "entropy": 6.350486087799072,
      "epoch": 0.6549855807043607,
      "grad_norm": 2.0,
      "learning_rate": 0.0004963698358934455,
      "loss": 6.1305,
      "mean_token_accuracy": 0.13944238424301147,
      "num_tokens": 15184895.0,
      "step": 7495
    },
    {
      "entropy": 6.254253005981445,
      "epoch": 0.6554225290570654,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004963642588229591,
      "loss": 6.1126,
      "mean_token_accuracy": 0.13657531812787055,
      "num_tokens": 15194380.0,
      "step": 7500
    },
    {
      "entropy": 6.356040525436401,
      "epoch": 0.6558594774097701,
      "grad_norm": 2.25,
      "learning_rate": 0.0004963586775065902,
      "loss": 6.1709,
      "mean_token_accuracy": 0.13733863532543183,
      "num_tokens": 15204364.0,
      "step": 7505
    },
    {
      "entropy": 6.364305400848389,
      "epoch": 0.6562964257624748,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004963530919444459,
      "loss": 6.1295,
      "mean_token_accuracy": 0.13430537655949593,
      "num_tokens": 15214485.0,
      "step": 7510
    },
    {
      "entropy": 6.371200847625732,
      "epoch": 0.6567333741151796,
      "grad_norm": 4.375,
      "learning_rate": 0.0004963475021366333,
      "loss": 6.2635,
      "mean_token_accuracy": 0.12953484579920768,
      "num_tokens": 15224764.0,
      "step": 7515
    },
    {
      "entropy": 6.394616222381591,
      "epoch": 0.6571703224678843,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004963419080832597,
      "loss": 6.2747,
      "mean_token_accuracy": 0.13087878227233887,
      "num_tokens": 15235481.0,
      "step": 7520
    },
    {
      "entropy": 6.347023391723633,
      "epoch": 0.657607270820589,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004963363097844323,
      "loss": 6.2057,
      "mean_token_accuracy": 0.1339855894446373,
      "num_tokens": 15245332.0,
      "step": 7525
    },
    {
      "entropy": 6.352317523956299,
      "epoch": 0.6580442191732937,
      "grad_norm": 6.21875,
      "learning_rate": 0.0004963307072402584,
      "loss": 6.2231,
      "mean_token_accuracy": 0.13197808638215064,
      "num_tokens": 15255562.0,
      "step": 7530
    },
    {
      "entropy": 6.400061321258545,
      "epoch": 0.6584811675259984,
      "grad_norm": 2.125,
      "learning_rate": 0.0004963251004508456,
      "loss": 6.1605,
      "mean_token_accuracy": 0.13426938354969026,
      "num_tokens": 15265655.0,
      "step": 7535
    },
    {
      "entropy": 6.340074062347412,
      "epoch": 0.6589181158787032,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004963194894163015,
      "loss": 6.0992,
      "mean_token_accuracy": 0.13870711848139763,
      "num_tokens": 15276311.0,
      "step": 7540
    },
    {
      "entropy": 6.337990760803223,
      "epoch": 0.6593550642314079,
      "grad_norm": 4.0,
      "learning_rate": 0.0004963138741367334,
      "loss": 6.2427,
      "mean_token_accuracy": 0.13313638120889665,
      "num_tokens": 15287142.0,
      "step": 7545
    },
    {
      "entropy": 6.342582845687867,
      "epoch": 0.6597920125841126,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004963082546122495,
      "loss": 6.2187,
      "mean_token_accuracy": 0.13226590305566788,
      "num_tokens": 15297484.0,
      "step": 7550
    },
    {
      "entropy": 6.374447011947632,
      "epoch": 0.6602289609368173,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004963026308429571,
      "loss": 6.1683,
      "mean_token_accuracy": 0.1329437769949436,
      "num_tokens": 15307468.0,
      "step": 7555
    },
    {
      "entropy": 6.275135040283203,
      "epoch": 0.660665909289522,
      "grad_norm": 5.375,
      "learning_rate": 0.0004962970028289644,
      "loss": 6.108,
      "mean_token_accuracy": 0.14276514574885368,
      "num_tokens": 15318487.0,
      "step": 7560
    },
    {
      "entropy": 6.222762203216552,
      "epoch": 0.6611028576422266,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004962913705703791,
      "loss": 6.1623,
      "mean_token_accuracy": 0.13728345781564713,
      "num_tokens": 15329193.0,
      "step": 7565
    },
    {
      "entropy": 6.339148664474488,
      "epoch": 0.6615398059949315,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004962857340673095,
      "loss": 6.1187,
      "mean_token_accuracy": 0.13656602278351784,
      "num_tokens": 15338886.0,
      "step": 7570
    },
    {
      "entropy": 6.299920988082886,
      "epoch": 0.6619767543476361,
      "grad_norm": 3.0,
      "learning_rate": 0.0004962800933198635,
      "loss": 6.0728,
      "mean_token_accuracy": 0.14619033560156822,
      "num_tokens": 15348629.0,
      "step": 7575
    },
    {
      "entropy": 6.242480754852295,
      "epoch": 0.6624137027003408,
      "grad_norm": 2.25,
      "learning_rate": 0.0004962744483281495,
      "loss": 6.1446,
      "mean_token_accuracy": 0.14091664478182792,
      "num_tokens": 15358004.0,
      "step": 7580
    },
    {
      "entropy": 6.253900623321533,
      "epoch": 0.6628506510530455,
      "grad_norm": 7.25,
      "learning_rate": 0.0004962687990922754,
      "loss": 6.1829,
      "mean_token_accuracy": 0.13517104610800743,
      "num_tokens": 15368018.0,
      "step": 7585
    },
    {
      "entropy": 6.411022138595581,
      "epoch": 0.6632875994057502,
      "grad_norm": 2.296875,
      "learning_rate": 0.00049626314561235,
      "loss": 6.1249,
      "mean_token_accuracy": 0.13804595544934273,
      "num_tokens": 15377443.0,
      "step": 7590
    },
    {
      "entropy": 6.347370195388794,
      "epoch": 0.6637245477584549,
      "grad_norm": 13.125,
      "learning_rate": 0.0004962574878884814,
      "loss": 6.1422,
      "mean_token_accuracy": 0.1377406157553196,
      "num_tokens": 15388278.0,
      "step": 7595
    },
    {
      "entropy": 6.294799661636352,
      "epoch": 0.6641614961111597,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004962518259207782,
      "loss": 6.1648,
      "mean_token_accuracy": 0.1370404079556465,
      "num_tokens": 15398619.0,
      "step": 7600
    },
    {
      "entropy": 6.282000255584717,
      "epoch": 0.6645984444638644,
      "grad_norm": 5.53125,
      "learning_rate": 0.0004962461597093491,
      "loss": 6.0699,
      "mean_token_accuracy": 0.14072233140468599,
      "num_tokens": 15408635.0,
      "step": 7605
    },
    {
      "entropy": 6.322996425628662,
      "epoch": 0.6650353928165691,
      "grad_norm": 2.625,
      "learning_rate": 0.0004962404892543028,
      "loss": 6.1707,
      "mean_token_accuracy": 0.13194511234760284,
      "num_tokens": 15419113.0,
      "step": 7610
    },
    {
      "entropy": 6.387387609481811,
      "epoch": 0.6654723411692738,
      "grad_norm": 6.125,
      "learning_rate": 0.000496234814555748,
      "loss": 6.1447,
      "mean_token_accuracy": 0.1328147642314434,
      "num_tokens": 15428959.0,
      "step": 7615
    },
    {
      "entropy": 6.29799222946167,
      "epoch": 0.6659092895219785,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004962291356137934,
      "loss": 6.2138,
      "mean_token_accuracy": 0.12744563370943068,
      "num_tokens": 15439116.0,
      "step": 7620
    },
    {
      "entropy": 6.332618618011475,
      "epoch": 0.6663462378746832,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004962234524285481,
      "loss": 6.1916,
      "mean_token_accuracy": 0.13926200196146965,
      "num_tokens": 15449715.0,
      "step": 7625
    },
    {
      "entropy": 6.311315870285034,
      "epoch": 0.666783186227388,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004962177650001212,
      "loss": 6.008,
      "mean_token_accuracy": 0.14588041976094246,
      "num_tokens": 15459170.0,
      "step": 7630
    },
    {
      "entropy": 6.2327296257019045,
      "epoch": 0.6672201345800927,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004962120733286215,
      "loss": 6.0843,
      "mean_token_accuracy": 0.1366695433855057,
      "num_tokens": 15469497.0,
      "step": 7635
    },
    {
      "entropy": 6.266078948974609,
      "epoch": 0.6676570829327974,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004962063774141583,
      "loss": 6.2066,
      "mean_token_accuracy": 0.13620064333081244,
      "num_tokens": 15481298.0,
      "step": 7640
    },
    {
      "entropy": 6.309061717987061,
      "epoch": 0.668094031285502,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004962006772568409,
      "loss": 6.1803,
      "mean_token_accuracy": 0.13686127960681915,
      "num_tokens": 15491696.0,
      "step": 7645
    },
    {
      "entropy": 6.4100768089294435,
      "epoch": 0.6685309796382067,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004961949728567787,
      "loss": 6.1975,
      "mean_token_accuracy": 0.13449274897575378,
      "num_tokens": 15502412.0,
      "step": 7650
    },
    {
      "entropy": 6.3524696826934814,
      "epoch": 0.6689679279909114,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004961892642140809,
      "loss": 6.1788,
      "mean_token_accuracy": 0.13505377173423766,
      "num_tokens": 15513436.0,
      "step": 7655
    },
    {
      "entropy": 6.204056167602539,
      "epoch": 0.6694048763436162,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004961835513288572,
      "loss": 6.1694,
      "mean_token_accuracy": 0.1393854446709156,
      "num_tokens": 15524564.0,
      "step": 7660
    },
    {
      "entropy": 6.329281282424927,
      "epoch": 0.6698418246963209,
      "grad_norm": 2.46875,
      "learning_rate": 0.000496177834201217,
      "loss": 6.1112,
      "mean_token_accuracy": 0.1332158237695694,
      "num_tokens": 15534539.0,
      "step": 7665
    },
    {
      "entropy": 6.389490270614624,
      "epoch": 0.6702787730490256,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004961721128312702,
      "loss": 6.193,
      "mean_token_accuracy": 0.12418471574783325,
      "num_tokens": 15544571.0,
      "step": 7670
    },
    {
      "entropy": 6.343524360656739,
      "epoch": 0.6707157214017303,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004961663872191263,
      "loss": 6.145,
      "mean_token_accuracy": 0.14224390238523482,
      "num_tokens": 15555066.0,
      "step": 7675
    },
    {
      "entropy": 6.301436519622802,
      "epoch": 0.671152669754435,
      "grad_norm": 5.96875,
      "learning_rate": 0.0004961606573648951,
      "loss": 6.1802,
      "mean_token_accuracy": 0.13548950403928756,
      "num_tokens": 15565128.0,
      "step": 7680
    },
    {
      "entropy": 6.266865921020508,
      "epoch": 0.6715896181071397,
      "grad_norm": 3.875,
      "learning_rate": 0.0004961549232686867,
      "loss": 6.2408,
      "mean_token_accuracy": 0.13169539347290993,
      "num_tokens": 15576394.0,
      "step": 7685
    },
    {
      "entropy": 6.377629899978638,
      "epoch": 0.6720265664598445,
      "grad_norm": 1.96875,
      "learning_rate": 0.000496149184930611,
      "loss": 6.14,
      "mean_token_accuracy": 0.1358833722770214,
      "num_tokens": 15585806.0,
      "step": 7690
    },
    {
      "entropy": 6.372461462020874,
      "epoch": 0.6724635148125492,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004961434423507779,
      "loss": 6.1438,
      "mean_token_accuracy": 0.13500701561570166,
      "num_tokens": 15596471.0,
      "step": 7695
    },
    {
      "entropy": 6.2446071147918705,
      "epoch": 0.6729004631652539,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004961376955292979,
      "loss": 6.0557,
      "mean_token_accuracy": 0.1475015513598919,
      "num_tokens": 15605720.0,
      "step": 7700
    },
    {
      "entropy": 6.303630399703979,
      "epoch": 0.6733374115179586,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004961319444662809,
      "loss": 6.1561,
      "mean_token_accuracy": 0.1391787923872471,
      "num_tokens": 15616167.0,
      "step": 7705
    },
    {
      "entropy": 6.423930263519287,
      "epoch": 0.6737743598706633,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004961261891618374,
      "loss": 6.1929,
      "mean_token_accuracy": 0.13162370771169662,
      "num_tokens": 15627033.0,
      "step": 7710
    },
    {
      "entropy": 6.342663526535034,
      "epoch": 0.6742113082233679,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004961204296160777,
      "loss": 6.1761,
      "mean_token_accuracy": 0.1345347762107849,
      "num_tokens": 15637460.0,
      "step": 7715
    },
    {
      "entropy": 6.301181793212891,
      "epoch": 0.6746482565760727,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004961146658291123,
      "loss": 6.0966,
      "mean_token_accuracy": 0.14560577273368835,
      "num_tokens": 15646027.0,
      "step": 7720
    },
    {
      "entropy": 6.333016061782837,
      "epoch": 0.6750852049287774,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004961088978010518,
      "loss": 6.1245,
      "mean_token_accuracy": 0.13643936365842818,
      "num_tokens": 15655241.0,
      "step": 7725
    },
    {
      "entropy": 6.243926048278809,
      "epoch": 0.6755221532814821,
      "grad_norm": 4.0,
      "learning_rate": 0.0004961031255320067,
      "loss": 6.0954,
      "mean_token_accuracy": 0.13370569348335265,
      "num_tokens": 15666213.0,
      "step": 7730
    },
    {
      "entropy": 6.346992158889771,
      "epoch": 0.6759591016341868,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004960973490220878,
      "loss": 6.1297,
      "mean_token_accuracy": 0.13411166220903398,
      "num_tokens": 15676985.0,
      "step": 7735
    },
    {
      "entropy": 6.3931060314178465,
      "epoch": 0.6763960499868915,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004960915682714059,
      "loss": 6.1974,
      "mean_token_accuracy": 0.13105277717113495,
      "num_tokens": 15687519.0,
      "step": 7740
    },
    {
      "entropy": 6.3519970893859865,
      "epoch": 0.6768329983395963,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004960857832800719,
      "loss": 6.1587,
      "mean_token_accuracy": 0.13897129520773888,
      "num_tokens": 15697401.0,
      "step": 7745
    },
    {
      "entropy": 6.314958763122559,
      "epoch": 0.677269946692301,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004960799940481968,
      "loss": 6.1909,
      "mean_token_accuracy": 0.12982374280691147,
      "num_tokens": 15708159.0,
      "step": 7750
    },
    {
      "entropy": 6.33341498374939,
      "epoch": 0.6777068950450057,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004960742005758914,
      "loss": 6.1565,
      "mean_token_accuracy": 0.12773403748869896,
      "num_tokens": 15717749.0,
      "step": 7755
    },
    {
      "entropy": 6.2849602699279785,
      "epoch": 0.6781438433977104,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004960684028632671,
      "loss": 6.1066,
      "mean_token_accuracy": 0.1378270260989666,
      "num_tokens": 15726868.0,
      "step": 7760
    },
    {
      "entropy": 6.2948707103729244,
      "epoch": 0.6785807917504151,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004960626009104349,
      "loss": 6.1069,
      "mean_token_accuracy": 0.13979058861732482,
      "num_tokens": 15736038.0,
      "step": 7765
    },
    {
      "entropy": 6.337671279907227,
      "epoch": 0.6790177401031198,
      "grad_norm": 4.25,
      "learning_rate": 0.0004960567947175064,
      "loss": 6.139,
      "mean_token_accuracy": 0.13427291736006736,
      "num_tokens": 15746676.0,
      "step": 7770
    },
    {
      "entropy": 6.325739860534668,
      "epoch": 0.6794546884558246,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004960509842845926,
      "loss": 6.1388,
      "mean_token_accuracy": 0.1389234334230423,
      "num_tokens": 15756064.0,
      "step": 7775
    },
    {
      "entropy": 6.295583915710449,
      "epoch": 0.6798916368085293,
      "grad_norm": 3.375,
      "learning_rate": 0.0004960451696118049,
      "loss": 6.1459,
      "mean_token_accuracy": 0.13471026122570037,
      "num_tokens": 15767414.0,
      "step": 7780
    },
    {
      "entropy": 6.335659980773926,
      "epoch": 0.680328585161234,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004960393506992553,
      "loss": 6.1252,
      "mean_token_accuracy": 0.13987841308116913,
      "num_tokens": 15778276.0,
      "step": 7785
    },
    {
      "entropy": 6.2856926918029785,
      "epoch": 0.6807655335139386,
      "grad_norm": 2.5,
      "learning_rate": 0.0004960335275470551,
      "loss": 6.0401,
      "mean_token_accuracy": 0.14274335131049157,
      "num_tokens": 15788365.0,
      "step": 7790
    },
    {
      "entropy": 6.282574462890625,
      "epoch": 0.6812024818666433,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004960277001553159,
      "loss": 6.066,
      "mean_token_accuracy": 0.14194058403372764,
      "num_tokens": 15799021.0,
      "step": 7795
    },
    {
      "entropy": 6.245684099197388,
      "epoch": 0.681639430219348,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004960218685241498,
      "loss": 6.1343,
      "mean_token_accuracy": 0.13751402646303176,
      "num_tokens": 15808779.0,
      "step": 7800
    },
    {
      "entropy": 6.313097286224365,
      "epoch": 0.6820763785720528,
      "grad_norm": 2.875,
      "learning_rate": 0.0004960160326536681,
      "loss": 6.1545,
      "mean_token_accuracy": 0.13918619751930236,
      "num_tokens": 15818515.0,
      "step": 7805
    },
    {
      "entropy": 6.32176833152771,
      "epoch": 0.6825133269247575,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004960101925439834,
      "loss": 6.1181,
      "mean_token_accuracy": 0.1356852985918522,
      "num_tokens": 15828843.0,
      "step": 7810
    },
    {
      "entropy": 6.349264335632324,
      "epoch": 0.6829502752774622,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004960043481952072,
      "loss": 6.1324,
      "mean_token_accuracy": 0.1336068369448185,
      "num_tokens": 15839323.0,
      "step": 7815
    },
    {
      "entropy": 6.274594020843506,
      "epoch": 0.6833872236301669,
      "grad_norm": 14.375,
      "learning_rate": 0.000495998499607452,
      "loss": 6.1532,
      "mean_token_accuracy": 0.1330889195203781,
      "num_tokens": 15848399.0,
      "step": 7820
    },
    {
      "entropy": 6.313720083236694,
      "epoch": 0.6838241719828716,
      "grad_norm": 2.125,
      "learning_rate": 0.0004959926467808296,
      "loss": 6.0393,
      "mean_token_accuracy": 0.139193619787693,
      "num_tokens": 15858137.0,
      "step": 7825
    },
    {
      "entropy": 6.257869625091553,
      "epoch": 0.6842611203355763,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004959867897154526,
      "loss": 6.1366,
      "mean_token_accuracy": 0.13022374883294105,
      "num_tokens": 15867643.0,
      "step": 7830
    },
    {
      "entropy": 6.3172338008880615,
      "epoch": 0.6846980686882811,
      "grad_norm": 2.9375,
      "learning_rate": 0.000495980928411433,
      "loss": 6.0268,
      "mean_token_accuracy": 0.14135612100362777,
      "num_tokens": 15877523.0,
      "step": 7835
    },
    {
      "entropy": 6.351158857345581,
      "epoch": 0.6851350170409858,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004959750628688834,
      "loss": 6.2195,
      "mean_token_accuracy": 0.13262092396616937,
      "num_tokens": 15886946.0,
      "step": 7840
    },
    {
      "entropy": 6.302695274353027,
      "epoch": 0.6855719653936905,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004959691930879164,
      "loss": 6.0644,
      "mean_token_accuracy": 0.13698486387729644,
      "num_tokens": 15896646.0,
      "step": 7845
    },
    {
      "entropy": 6.243091630935669,
      "epoch": 0.6860089137463952,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004959633190686445,
      "loss": 6.1075,
      "mean_token_accuracy": 0.13988875299692155,
      "num_tokens": 15907204.0,
      "step": 7850
    },
    {
      "entropy": 6.254873275756836,
      "epoch": 0.6864458620990999,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004959574408111804,
      "loss": 6.132,
      "mean_token_accuracy": 0.14005031287670136,
      "num_tokens": 15917634.0,
      "step": 7855
    },
    {
      "entropy": 6.316693544387817,
      "epoch": 0.6868828104518045,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004959515583156367,
      "loss": 6.1323,
      "mean_token_accuracy": 0.13513658791780472,
      "num_tokens": 15926584.0,
      "step": 7860
    },
    {
      "entropy": 6.303711652755737,
      "epoch": 0.6873197588045094,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004959456715821264,
      "loss": 6.0943,
      "mean_token_accuracy": 0.13764404207468034,
      "num_tokens": 15936636.0,
      "step": 7865
    },
    {
      "entropy": 6.327462816238404,
      "epoch": 0.687756707157214,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004959397806107624,
      "loss": 6.0702,
      "mean_token_accuracy": 0.14857126772403717,
      "num_tokens": 15946113.0,
      "step": 7870
    },
    {
      "entropy": 6.3101166725158695,
      "epoch": 0.6881936555099187,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004959338854016574,
      "loss": 6.1178,
      "mean_token_accuracy": 0.13507317006587982,
      "num_tokens": 15956630.0,
      "step": 7875
    },
    {
      "entropy": 6.247363138198852,
      "epoch": 0.6886306038626234,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004959279859549249,
      "loss": 6.0675,
      "mean_token_accuracy": 0.13953003510832787,
      "num_tokens": 15967129.0,
      "step": 7880
    },
    {
      "entropy": 6.263906526565552,
      "epoch": 0.6890675522153281,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004959220822706779,
      "loss": 6.0719,
      "mean_token_accuracy": 0.13742638230323792,
      "num_tokens": 15976663.0,
      "step": 7885
    },
    {
      "entropy": 6.329449319839478,
      "epoch": 0.6895045005680328,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004959161743490295,
      "loss": 6.0707,
      "mean_token_accuracy": 0.1389826789498329,
      "num_tokens": 15986666.0,
      "step": 7890
    },
    {
      "entropy": 6.214951705932617,
      "epoch": 0.6899414489207376,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004959102621900931,
      "loss": 6.0179,
      "mean_token_accuracy": 0.1482565775513649,
      "num_tokens": 15996233.0,
      "step": 7895
    },
    {
      "entropy": 6.308627080917359,
      "epoch": 0.6903783972734423,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004959043457939822,
      "loss": 6.2059,
      "mean_token_accuracy": 0.13284177035093309,
      "num_tokens": 16008188.0,
      "step": 7900
    },
    {
      "entropy": 6.321988201141357,
      "epoch": 0.690815345626147,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004958984251608101,
      "loss": 6.121,
      "mean_token_accuracy": 0.13530101999640465,
      "num_tokens": 16018494.0,
      "step": 7905
    },
    {
      "entropy": 6.261188268661499,
      "epoch": 0.6912522939788517,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004958925002906905,
      "loss": 6.1116,
      "mean_token_accuracy": 0.14374242946505547,
      "num_tokens": 16028531.0,
      "step": 7910
    },
    {
      "entropy": 6.316179656982422,
      "epoch": 0.6916892423315564,
      "grad_norm": 5.25,
      "learning_rate": 0.000495886571183737,
      "loss": 6.1173,
      "mean_token_accuracy": 0.1372798353433609,
      "num_tokens": 16039050.0,
      "step": 7915
    },
    {
      "entropy": 6.34809627532959,
      "epoch": 0.6921261906842611,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004958806378400634,
      "loss": 6.2007,
      "mean_token_accuracy": 0.13284706249833106,
      "num_tokens": 16050751.0,
      "step": 7920
    },
    {
      "entropy": 6.303271150588989,
      "epoch": 0.6925631390369659,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004958747002597834,
      "loss": 6.1386,
      "mean_token_accuracy": 0.13526258915662764,
      "num_tokens": 16060758.0,
      "step": 7925
    },
    {
      "entropy": 6.321645975112915,
      "epoch": 0.6930000873896706,
      "grad_norm": 2.625,
      "learning_rate": 0.0004958687584430109,
      "loss": 6.1467,
      "mean_token_accuracy": 0.13234784454107285,
      "num_tokens": 16070852.0,
      "step": 7930
    },
    {
      "entropy": 6.286686897277832,
      "epoch": 0.6934370357423753,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004958628123898597,
      "loss": 6.06,
      "mean_token_accuracy": 0.13403984606266023,
      "num_tokens": 16080081.0,
      "step": 7935
    },
    {
      "entropy": 6.269439697265625,
      "epoch": 0.6938739840950799,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004958568621004442,
      "loss": 6.1085,
      "mean_token_accuracy": 0.1386524945497513,
      "num_tokens": 16090483.0,
      "step": 7940
    },
    {
      "entropy": 6.320281839370727,
      "epoch": 0.6943109324477846,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004958509075748785,
      "loss": 6.173,
      "mean_token_accuracy": 0.12933071553707123,
      "num_tokens": 16101136.0,
      "step": 7945
    },
    {
      "entropy": 6.335004138946533,
      "epoch": 0.6947478808004894,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004958449488132764,
      "loss": 6.1083,
      "mean_token_accuracy": 0.14272436797618865,
      "num_tokens": 16110124.0,
      "step": 7950
    },
    {
      "entropy": 6.291327476501465,
      "epoch": 0.6951848291531941,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004958389858157526,
      "loss": 6.0254,
      "mean_token_accuracy": 0.146212537586689,
      "num_tokens": 16119581.0,
      "step": 7955
    },
    {
      "entropy": 6.268547868728637,
      "epoch": 0.6956217775058988,
      "grad_norm": 2.625,
      "learning_rate": 0.0004958330185824213,
      "loss": 6.1451,
      "mean_token_accuracy": 0.13911877274513246,
      "num_tokens": 16130119.0,
      "step": 7960
    },
    {
      "entropy": 6.285935497283935,
      "epoch": 0.6960587258586035,
      "grad_norm": 2.53125,
      "learning_rate": 0.000495827047113397,
      "loss": 6.15,
      "mean_token_accuracy": 0.13733095303177834,
      "num_tokens": 16140702.0,
      "step": 7965
    },
    {
      "entropy": 6.315868377685547,
      "epoch": 0.6964956742113082,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004958210714087942,
      "loss": 6.1311,
      "mean_token_accuracy": 0.13964533656835557,
      "num_tokens": 16151107.0,
      "step": 7970
    },
    {
      "entropy": 6.337750959396362,
      "epoch": 0.6969326225640129,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004958150914687276,
      "loss": 6.0678,
      "mean_token_accuracy": 0.1425984174013138,
      "num_tokens": 16160779.0,
      "step": 7975
    },
    {
      "entropy": 6.19505729675293,
      "epoch": 0.6973695709167177,
      "grad_norm": 1.75,
      "learning_rate": 0.0004958091072933118,
      "loss": 5.9934,
      "mean_token_accuracy": 0.14167692959308625,
      "num_tokens": 16169624.0,
      "step": 7980
    },
    {
      "entropy": 6.171296787261963,
      "epoch": 0.6978065192694224,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004958031188826616,
      "loss": 6.0127,
      "mean_token_accuracy": 0.15032820254564286,
      "num_tokens": 16179192.0,
      "step": 7985
    },
    {
      "entropy": 6.233458757400513,
      "epoch": 0.6982434676221271,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004957971262368919,
      "loss": 6.0979,
      "mean_token_accuracy": 0.1411059893667698,
      "num_tokens": 16188847.0,
      "step": 7990
    },
    {
      "entropy": 6.390803623199463,
      "epoch": 0.6986804159748318,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004957911293561176,
      "loss": 6.2124,
      "mean_token_accuracy": 0.13191223964095117,
      "num_tokens": 16199480.0,
      "step": 7995
    },
    {
      "entropy": 6.360818862915039,
      "epoch": 0.6991173643275365,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004957851282404538,
      "loss": 6.1234,
      "mean_token_accuracy": 0.13945151567459108,
      "num_tokens": 16208818.0,
      "step": 8000
    },
    {
      "entropy": 6.281424045562744,
      "epoch": 0.6995543126802412,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004957791228900154,
      "loss": 6.0735,
      "mean_token_accuracy": 0.13729138821363449,
      "num_tokens": 16219298.0,
      "step": 8005
    },
    {
      "entropy": 6.201127910614014,
      "epoch": 0.699991261032946,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004957731133049179,
      "loss": 6.0797,
      "mean_token_accuracy": 0.146750608086586,
      "num_tokens": 16228170.0,
      "step": 8010
    },
    {
      "entropy": 6.240867757797242,
      "epoch": 0.7004282093856506,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004957670994852764,
      "loss": 6.1453,
      "mean_token_accuracy": 0.1332260087132454,
      "num_tokens": 16237552.0,
      "step": 8015
    },
    {
      "entropy": 6.370295143127441,
      "epoch": 0.7008651577383553,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004957610814312062,
      "loss": 6.2017,
      "mean_token_accuracy": 0.1324914202094078,
      "num_tokens": 16248157.0,
      "step": 8020
    },
    {
      "entropy": 6.340698385238648,
      "epoch": 0.70130210609106,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004957550591428229,
      "loss": 6.0672,
      "mean_token_accuracy": 0.14167074039578437,
      "num_tokens": 16258266.0,
      "step": 8025
    },
    {
      "entropy": 6.2213664054870605,
      "epoch": 0.7017390544437647,
      "grad_norm": 2.75,
      "learning_rate": 0.0004957490326202418,
      "loss": 6.1828,
      "mean_token_accuracy": 0.1369075708091259,
      "num_tokens": 16268769.0,
      "step": 8030
    },
    {
      "entropy": 6.267453241348266,
      "epoch": 0.7021760027964694,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004957430018635785,
      "loss": 6.1522,
      "mean_token_accuracy": 0.137697421759367,
      "num_tokens": 16279688.0,
      "step": 8035
    },
    {
      "entropy": 6.33360276222229,
      "epoch": 0.7026129511491742,
      "grad_norm": 2.75,
      "learning_rate": 0.0004957369668729488,
      "loss": 6.0781,
      "mean_token_accuracy": 0.13956429958343505,
      "num_tokens": 16289441.0,
      "step": 8040
    },
    {
      "entropy": 6.3231260776519775,
      "epoch": 0.7030498995018789,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004957309276484685,
      "loss": 6.1013,
      "mean_token_accuracy": 0.13597154393792152,
      "num_tokens": 16299802.0,
      "step": 8045
    },
    {
      "entropy": 6.211606121063232,
      "epoch": 0.7034868478545836,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004957248841902533,
      "loss": 6.0496,
      "mean_token_accuracy": 0.1428252324461937,
      "num_tokens": 16309577.0,
      "step": 8050
    },
    {
      "entropy": 6.255250358581543,
      "epoch": 0.7039237962072883,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004957188364984191,
      "loss": 6.0264,
      "mean_token_accuracy": 0.14443229883909225,
      "num_tokens": 16319205.0,
      "step": 8055
    },
    {
      "entropy": 6.231384897232056,
      "epoch": 0.704360744559993,
      "grad_norm": 3.625,
      "learning_rate": 0.000495712784573082,
      "loss": 6.0883,
      "mean_token_accuracy": 0.14341194108128547,
      "num_tokens": 16329650.0,
      "step": 8060
    },
    {
      "entropy": 6.187993907928467,
      "epoch": 0.7047976929126977,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004957067284143581,
      "loss": 6.1087,
      "mean_token_accuracy": 0.14184321984648704,
      "num_tokens": 16339242.0,
      "step": 8065
    },
    {
      "entropy": 6.246601629257202,
      "epoch": 0.7052346412654025,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004957006680223634,
      "loss": 6.0691,
      "mean_token_accuracy": 0.14063308238983155,
      "num_tokens": 16348441.0,
      "step": 8070
    },
    {
      "entropy": 6.289533090591431,
      "epoch": 0.7056715896181072,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004956946033972143,
      "loss": 6.0731,
      "mean_token_accuracy": 0.139460801333189,
      "num_tokens": 16357425.0,
      "step": 8075
    },
    {
      "entropy": 6.268570327758789,
      "epoch": 0.7061085379708119,
      "grad_norm": 4.09375,
      "learning_rate": 0.000495688534539027,
      "loss": 6.0701,
      "mean_token_accuracy": 0.13391998857259751,
      "num_tokens": 16366180.0,
      "step": 8080
    },
    {
      "entropy": 6.207866430282593,
      "epoch": 0.7065454863235165,
      "grad_norm": 3.25,
      "learning_rate": 0.0004956824614479181,
      "loss": 6.1213,
      "mean_token_accuracy": 0.13996209427714348,
      "num_tokens": 16375714.0,
      "step": 8085
    },
    {
      "entropy": 6.310677623748779,
      "epoch": 0.7069824346762212,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004956763841240038,
      "loss": 6.1393,
      "mean_token_accuracy": 0.1337546318769455,
      "num_tokens": 16386051.0,
      "step": 8090
    },
    {
      "entropy": 6.395681858062744,
      "epoch": 0.7074193830289259,
      "grad_norm": 3.4375,
      "learning_rate": 0.000495670302567401,
      "loss": 6.1791,
      "mean_token_accuracy": 0.1311746656894684,
      "num_tokens": 16396616.0,
      "step": 8095
    },
    {
      "entropy": 6.33921709060669,
      "epoch": 0.7078563313816307,
      "grad_norm": 2.734375,
      "learning_rate": 0.000495664216778226,
      "loss": 6.0928,
      "mean_token_accuracy": 0.14873505532741546,
      "num_tokens": 16406301.0,
      "step": 8100
    },
    {
      "entropy": 6.3197471618652346,
      "epoch": 0.7082932797343354,
      "grad_norm": 12.0,
      "learning_rate": 0.0004956581267565957,
      "loss": 6.1369,
      "mean_token_accuracy": 0.13589697554707528,
      "num_tokens": 16417107.0,
      "step": 8105
    },
    {
      "entropy": 6.316809225082397,
      "epoch": 0.7087302280870401,
      "grad_norm": 3.734375,
      "learning_rate": 0.000495652032502627,
      "loss": 6.077,
      "mean_token_accuracy": 0.13454096093773843,
      "num_tokens": 16427292.0,
      "step": 8110
    },
    {
      "entropy": 6.202370500564575,
      "epoch": 0.7091671764397448,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004956459340164366,
      "loss": 6.0992,
      "mean_token_accuracy": 0.13893202617764472,
      "num_tokens": 16438191.0,
      "step": 8115
    },
    {
      "entropy": 6.286994743347168,
      "epoch": 0.7096041247924495,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004956398312981417,
      "loss": 6.0777,
      "mean_token_accuracy": 0.13912276029586793,
      "num_tokens": 16448237.0,
      "step": 8120
    },
    {
      "entropy": 6.300420188903809,
      "epoch": 0.7100410731451542,
      "grad_norm": 5.25,
      "learning_rate": 0.0004956337243478592,
      "loss": 6.1839,
      "mean_token_accuracy": 0.136966959387064,
      "num_tokens": 16458947.0,
      "step": 8125
    },
    {
      "entropy": 6.282527065277099,
      "epoch": 0.710478021497859,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004956276131657061,
      "loss": 6.0764,
      "mean_token_accuracy": 0.14468176513910294,
      "num_tokens": 16467464.0,
      "step": 8130
    },
    {
      "entropy": 6.2707196235656735,
      "epoch": 0.7109149698505637,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004956214977518,
      "loss": 6.0791,
      "mean_token_accuracy": 0.1483884185552597,
      "num_tokens": 16477747.0,
      "step": 8135
    },
    {
      "entropy": 6.3166053771972654,
      "epoch": 0.7113519182032684,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004956153781062578,
      "loss": 6.219,
      "mean_token_accuracy": 0.13383654430508612,
      "num_tokens": 16488111.0,
      "step": 8140
    },
    {
      "entropy": 6.316377210617065,
      "epoch": 0.7117888665559731,
      "grad_norm": 6.0,
      "learning_rate": 0.0004956092542291971,
      "loss": 6.088,
      "mean_token_accuracy": 0.1412655994296074,
      "num_tokens": 16497689.0,
      "step": 8145
    },
    {
      "entropy": 6.2846932888031,
      "epoch": 0.7122258149086778,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004956031261207353,
      "loss": 6.176,
      "mean_token_accuracy": 0.13831764832139015,
      "num_tokens": 16508515.0,
      "step": 8150
    },
    {
      "entropy": 6.228969860076904,
      "epoch": 0.7126627632613826,
      "grad_norm": 4.15625,
      "learning_rate": 0.00049559699378099,
      "loss": 6.1028,
      "mean_token_accuracy": 0.14722624123096467,
      "num_tokens": 16519184.0,
      "step": 8155
    },
    {
      "entropy": 6.301027107238769,
      "epoch": 0.7130997116140873,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004955908572100787,
      "loss": 6.1848,
      "mean_token_accuracy": 0.13245953619480133,
      "num_tokens": 16529917.0,
      "step": 8160
    },
    {
      "entropy": 6.322576093673706,
      "epoch": 0.7135366599667919,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004955847164081192,
      "loss": 6.0994,
      "mean_token_accuracy": 0.13878415450453757,
      "num_tokens": 16539687.0,
      "step": 8165
    },
    {
      "entropy": 6.339302635192871,
      "epoch": 0.7139736083194966,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004955785713752292,
      "loss": 6.1641,
      "mean_token_accuracy": 0.14130710139870645,
      "num_tokens": 16550273.0,
      "step": 8170
    },
    {
      "entropy": 6.335113000869751,
      "epoch": 0.7144105566722013,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004955724221115267,
      "loss": 6.2319,
      "mean_token_accuracy": 0.13179654628038406,
      "num_tokens": 16562215.0,
      "step": 8175
    },
    {
      "entropy": 6.374898672103882,
      "epoch": 0.714847505024906,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004955662686171295,
      "loss": 6.197,
      "mean_token_accuracy": 0.1347971700131893,
      "num_tokens": 16573218.0,
      "step": 8180
    },
    {
      "entropy": 6.405144929885864,
      "epoch": 0.7152844533776108,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004955601108921557,
      "loss": 6.1583,
      "mean_token_accuracy": 0.1329104147851467,
      "num_tokens": 16582740.0,
      "step": 8185
    },
    {
      "entropy": 6.250557088851929,
      "epoch": 0.7157214017303155,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004955539489367235,
      "loss": 6.1348,
      "mean_token_accuracy": 0.1445598915219307,
      "num_tokens": 16592169.0,
      "step": 8190
    },
    {
      "entropy": 6.30103120803833,
      "epoch": 0.7161583500830202,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004955477827509509,
      "loss": 6.0944,
      "mean_token_accuracy": 0.13646115437150003,
      "num_tokens": 16602706.0,
      "step": 8195
    },
    {
      "entropy": 6.323256969451904,
      "epoch": 0.7165952984357249,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004955416123349562,
      "loss": 6.0794,
      "mean_token_accuracy": 0.13657411709427833,
      "num_tokens": 16612012.0,
      "step": 8200
    },
    {
      "entropy": 6.28774585723877,
      "epoch": 0.7170322467884296,
      "grad_norm": 3.515625,
      "learning_rate": 0.000495535437688858,
      "loss": 6.1922,
      "mean_token_accuracy": 0.13566832467913628,
      "num_tokens": 16621798.0,
      "step": 8205
    },
    {
      "entropy": 6.334479427337646,
      "epoch": 0.7174691951411343,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004955292588127744,
      "loss": 6.1154,
      "mean_token_accuracy": 0.14004983603954316,
      "num_tokens": 16631792.0,
      "step": 8210
    },
    {
      "entropy": 6.364751482009888,
      "epoch": 0.7179061434938391,
      "grad_norm": 3.75,
      "learning_rate": 0.000495523075706824,
      "loss": 6.0632,
      "mean_token_accuracy": 0.14040024876594542,
      "num_tokens": 16641655.0,
      "step": 8215
    },
    {
      "entropy": 6.31078314781189,
      "epoch": 0.7183430918465438,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004955168883711257,
      "loss": 6.164,
      "mean_token_accuracy": 0.13185230493545533,
      "num_tokens": 16651210.0,
      "step": 8220
    },
    {
      "entropy": 6.317300939559937,
      "epoch": 0.7187800401992485,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004955106968057977,
      "loss": 6.117,
      "mean_token_accuracy": 0.1367509976029396,
      "num_tokens": 16660078.0,
      "step": 8225
    },
    {
      "entropy": 6.3061644554138185,
      "epoch": 0.7192169885519532,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004955045010109592,
      "loss": 6.1484,
      "mean_token_accuracy": 0.1368064872920513,
      "num_tokens": 16671211.0,
      "step": 8230
    },
    {
      "entropy": 6.266268539428711,
      "epoch": 0.7196539369046578,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004954983009867286,
      "loss": 6.0765,
      "mean_token_accuracy": 0.13917433172464372,
      "num_tokens": 16680850.0,
      "step": 8235
    },
    {
      "entropy": 6.295585680007934,
      "epoch": 0.7200908852573625,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004954920967332252,
      "loss": 6.0529,
      "mean_token_accuracy": 0.14017029255628585,
      "num_tokens": 16690447.0,
      "step": 8240
    },
    {
      "entropy": 6.318373441696167,
      "epoch": 0.7205278336100673,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004954858882505678,
      "loss": 6.1676,
      "mean_token_accuracy": 0.13783882185816765,
      "num_tokens": 16701061.0,
      "step": 8245
    },
    {
      "entropy": 6.2861353874206545,
      "epoch": 0.720964781962772,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004954796755388755,
      "loss": 6.0601,
      "mean_token_accuracy": 0.14330372959375381,
      "num_tokens": 16711139.0,
      "step": 8250
    },
    {
      "entropy": 6.326176309585572,
      "epoch": 0.7214017303154767,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004954734585982675,
      "loss": 6.1868,
      "mean_token_accuracy": 0.12853687554597854,
      "num_tokens": 16722343.0,
      "step": 8255
    },
    {
      "entropy": 6.357975625991822,
      "epoch": 0.7218386786681814,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004954672374288631,
      "loss": 6.1988,
      "mean_token_accuracy": 0.13234879523515702,
      "num_tokens": 16733475.0,
      "step": 8260
    },
    {
      "entropy": 6.306359148025512,
      "epoch": 0.7222756270208861,
      "grad_norm": 3.875,
      "learning_rate": 0.0004954610120307815,
      "loss": 6.1281,
      "mean_token_accuracy": 0.13623349815607072,
      "num_tokens": 16744356.0,
      "step": 8265
    },
    {
      "entropy": 6.292976713180542,
      "epoch": 0.7227125753735908,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004954547824041422,
      "loss": 6.1663,
      "mean_token_accuracy": 0.13137195706367494,
      "num_tokens": 16755275.0,
      "step": 8270
    },
    {
      "entropy": 6.269350862503051,
      "epoch": 0.7231495237262956,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004954485485490647,
      "loss": 6.0788,
      "mean_token_accuracy": 0.14117906391620635,
      "num_tokens": 16766114.0,
      "step": 8275
    },
    {
      "entropy": 6.260144662857056,
      "epoch": 0.7235864720790003,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004954423104656685,
      "loss": 6.0808,
      "mean_token_accuracy": 0.142653901129961,
      "num_tokens": 16776391.0,
      "step": 8280
    },
    {
      "entropy": 6.246277141571045,
      "epoch": 0.724023420431705,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004954360681540732,
      "loss": 6.0628,
      "mean_token_accuracy": 0.14554407298564911,
      "num_tokens": 16784845.0,
      "step": 8285
    },
    {
      "entropy": 6.258530139923096,
      "epoch": 0.7244603687844097,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004954298216143986,
      "loss": 6.0762,
      "mean_token_accuracy": 0.13803903311491011,
      "num_tokens": 16794735.0,
      "step": 8290
    },
    {
      "entropy": 6.375484323501587,
      "epoch": 0.7248973171371144,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004954235708467647,
      "loss": 6.1497,
      "mean_token_accuracy": 0.13852656334638597,
      "num_tokens": 16803643.0,
      "step": 8295
    },
    {
      "entropy": 6.379577684402466,
      "epoch": 0.725334265489819,
      "grad_norm": 3.578125,
      "learning_rate": 0.000495417315851291,
      "loss": 6.1614,
      "mean_token_accuracy": 0.1349959596991539,
      "num_tokens": 16813064.0,
      "step": 8300
    },
    {
      "entropy": 6.285477638244629,
      "epoch": 0.7257712138425239,
      "grad_norm": 3.25,
      "learning_rate": 0.0004954110566280978,
      "loss": 6.1739,
      "mean_token_accuracy": 0.13573629334568976,
      "num_tokens": 16823166.0,
      "step": 8305
    },
    {
      "entropy": 6.291155385971069,
      "epoch": 0.7262081621952285,
      "grad_norm": 2.6875,
      "learning_rate": 0.000495404793177305,
      "loss": 6.1867,
      "mean_token_accuracy": 0.13619570210576057,
      "num_tokens": 16833546.0,
      "step": 8310
    },
    {
      "entropy": 6.3617393493652346,
      "epoch": 0.7266451105479332,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004953985254990328,
      "loss": 6.0551,
      "mean_token_accuracy": 0.1392693541944027,
      "num_tokens": 16843794.0,
      "step": 8315
    },
    {
      "entropy": 6.364054155349732,
      "epoch": 0.7270820589006379,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004953922535934014,
      "loss": 6.1956,
      "mean_token_accuracy": 0.13290442973375322,
      "num_tokens": 16855677.0,
      "step": 8320
    },
    {
      "entropy": 6.336766624450684,
      "epoch": 0.7275190072533426,
      "grad_norm": 2.765625,
      "learning_rate": 0.000495385977460531,
      "loss": 6.1205,
      "mean_token_accuracy": 0.1345275528728962,
      "num_tokens": 16866137.0,
      "step": 8325
    },
    {
      "entropy": 6.314247465133667,
      "epoch": 0.7279559556060473,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004953796971005422,
      "loss": 6.0663,
      "mean_token_accuracy": 0.13793630823493003,
      "num_tokens": 16875997.0,
      "step": 8330
    },
    {
      "entropy": 6.249747657775879,
      "epoch": 0.7283929039587521,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004953734125135552,
      "loss": 6.0877,
      "mean_token_accuracy": 0.1465596079826355,
      "num_tokens": 16885838.0,
      "step": 8335
    },
    {
      "entropy": 6.228337526321411,
      "epoch": 0.7288298523114568,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004953671236996909,
      "loss": 6.043,
      "mean_token_accuracy": 0.14424114897847176,
      "num_tokens": 16895533.0,
      "step": 8340
    },
    {
      "entropy": 6.279504203796387,
      "epoch": 0.7292668006641615,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004953608306590696,
      "loss": 6.1744,
      "mean_token_accuracy": 0.13422726467251778,
      "num_tokens": 16906258.0,
      "step": 8345
    },
    {
      "entropy": 6.357880067825318,
      "epoch": 0.7297037490168662,
      "grad_norm": 4.90625,
      "learning_rate": 0.000495354533391812,
      "loss": 6.1178,
      "mean_token_accuracy": 0.1417343333363533,
      "num_tokens": 16917123.0,
      "step": 8350
    },
    {
      "entropy": 6.33694167137146,
      "epoch": 0.7301406973695709,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004953482318980389,
      "loss": 6.1319,
      "mean_token_accuracy": 0.14400000125169754,
      "num_tokens": 16927533.0,
      "step": 8355
    },
    {
      "entropy": 6.344516134262085,
      "epoch": 0.7305776457222757,
      "grad_norm": 5.4375,
      "learning_rate": 0.0004953419261778714,
      "loss": 6.1022,
      "mean_token_accuracy": 0.13950815945863723,
      "num_tokens": 16937763.0,
      "step": 8360
    },
    {
      "entropy": 6.216820955276489,
      "epoch": 0.7310145940749804,
      "grad_norm": 6.4375,
      "learning_rate": 0.0004953356162314304,
      "loss": 6.0495,
      "mean_token_accuracy": 0.14419286102056503,
      "num_tokens": 16947173.0,
      "step": 8365
    },
    {
      "entropy": 6.2570271492004395,
      "epoch": 0.7314515424276851,
      "grad_norm": 5.4375,
      "learning_rate": 0.0004953293020588368,
      "loss": 6.1177,
      "mean_token_accuracy": 0.14414651691913605,
      "num_tokens": 16956856.0,
      "step": 8370
    },
    {
      "entropy": 6.334005403518677,
      "epoch": 0.7318884907803898,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004953229836602115,
      "loss": 6.1027,
      "mean_token_accuracy": 0.1370401054620743,
      "num_tokens": 16966705.0,
      "step": 8375
    },
    {
      "entropy": 6.38493914604187,
      "epoch": 0.7323254391330944,
      "grad_norm": 2.890625,
      "learning_rate": 0.000495316661035676,
      "loss": 6.1537,
      "mean_token_accuracy": 0.1347280979156494,
      "num_tokens": 16977026.0,
      "step": 8380
    },
    {
      "entropy": 6.286597442626953,
      "epoch": 0.7327623874857991,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004953103341853517,
      "loss": 6.071,
      "mean_token_accuracy": 0.14289721995592117,
      "num_tokens": 16986800.0,
      "step": 8385
    },
    {
      "entropy": 6.217516231536865,
      "epoch": 0.7331993358385039,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004953040031093596,
      "loss": 6.0458,
      "mean_token_accuracy": 0.1443341962993145,
      "num_tokens": 16996392.0,
      "step": 8390
    },
    {
      "entropy": 6.263503932952881,
      "epoch": 0.7336362841912086,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004952976678078213,
      "loss": 6.085,
      "mean_token_accuracy": 0.135968130081892,
      "num_tokens": 17006430.0,
      "step": 8395
    },
    {
      "entropy": 6.363481616973877,
      "epoch": 0.7340732325439133,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004952913282808584,
      "loss": 6.1252,
      "mean_token_accuracy": 0.1351695641875267,
      "num_tokens": 17017168.0,
      "step": 8400
    },
    {
      "entropy": 6.250575017929077,
      "epoch": 0.734510180896618,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004952849845285922,
      "loss": 6.0264,
      "mean_token_accuracy": 0.14791935682296753,
      "num_tokens": 17027337.0,
      "step": 8405
    },
    {
      "entropy": 6.1948007106781,
      "epoch": 0.7349471292493227,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004952786365511447,
      "loss": 6.0079,
      "mean_token_accuracy": 0.14440675601363182,
      "num_tokens": 17036839.0,
      "step": 8410
    },
    {
      "entropy": 6.264123010635376,
      "epoch": 0.7353840776020274,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004952722843486375,
      "loss": 6.2198,
      "mean_token_accuracy": 0.13194009140133858,
      "num_tokens": 17048221.0,
      "step": 8415
    },
    {
      "entropy": 6.276167869567871,
      "epoch": 0.7358210259547322,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004952659279211924,
      "loss": 6.0503,
      "mean_token_accuracy": 0.13819127455353736,
      "num_tokens": 17057694.0,
      "step": 8420
    },
    {
      "entropy": 6.296573209762573,
      "epoch": 0.7362579743074369,
      "grad_norm": 4.0,
      "learning_rate": 0.0004952595672689315,
      "loss": 6.1391,
      "mean_token_accuracy": 0.1282465673983097,
      "num_tokens": 17066492.0,
      "step": 8425
    },
    {
      "entropy": 6.326595783233643,
      "epoch": 0.7366949226601416,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004952532023919766,
      "loss": 6.1506,
      "mean_token_accuracy": 0.13878269344568253,
      "num_tokens": 17077211.0,
      "step": 8430
    },
    {
      "entropy": 6.329667472839356,
      "epoch": 0.7371318710128463,
      "grad_norm": 2.625,
      "learning_rate": 0.0004952468332904498,
      "loss": 6.1227,
      "mean_token_accuracy": 0.1378251016139984,
      "num_tokens": 17088665.0,
      "step": 8435
    },
    {
      "entropy": 6.337694406509399,
      "epoch": 0.737568819365551,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004952404599644735,
      "loss": 6.1381,
      "mean_token_accuracy": 0.13944955617189408,
      "num_tokens": 17099767.0,
      "step": 8440
    },
    {
      "entropy": 6.316484117507935,
      "epoch": 0.7380057677182557,
      "grad_norm": 5.78125,
      "learning_rate": 0.0004952340824141698,
      "loss": 6.0289,
      "mean_token_accuracy": 0.14461853057146073,
      "num_tokens": 17109414.0,
      "step": 8445
    },
    {
      "entropy": 6.271963787078858,
      "epoch": 0.7384427160709605,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004952277006396609,
      "loss": 6.1026,
      "mean_token_accuracy": 0.13500657677650452,
      "num_tokens": 17119107.0,
      "step": 8450
    },
    {
      "entropy": 6.270470762252808,
      "epoch": 0.7388796644236652,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004952213146410693,
      "loss": 6.1529,
      "mean_token_accuracy": 0.13341975063085557,
      "num_tokens": 17129010.0,
      "step": 8455
    },
    {
      "entropy": 6.315335607528686,
      "epoch": 0.7393166127763698,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004952149244185174,
      "loss": 6.1553,
      "mean_token_accuracy": 0.1360753908753395,
      "num_tokens": 17138951.0,
      "step": 8460
    },
    {
      "entropy": 6.358991432189941,
      "epoch": 0.7397535611290745,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004952085299721279,
      "loss": 6.1062,
      "mean_token_accuracy": 0.14176160097122192,
      "num_tokens": 17149416.0,
      "step": 8465
    },
    {
      "entropy": 6.342560911178589,
      "epoch": 0.7401905094817792,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004952021313020235,
      "loss": 6.0654,
      "mean_token_accuracy": 0.14540593847632408,
      "num_tokens": 17160068.0,
      "step": 8470
    },
    {
      "entropy": 6.261391258239746,
      "epoch": 0.7406274578344839,
      "grad_norm": 4.75,
      "learning_rate": 0.0004951957284083269,
      "loss": 6.089,
      "mean_token_accuracy": 0.14151019155979155,
      "num_tokens": 17168972.0,
      "step": 8475
    },
    {
      "entropy": 6.229527616500855,
      "epoch": 0.7410644061871887,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004951893212911607,
      "loss": 6.091,
      "mean_token_accuracy": 0.13669056445360184,
      "num_tokens": 17178239.0,
      "step": 8480
    },
    {
      "entropy": 6.282192087173462,
      "epoch": 0.7415013545398934,
      "grad_norm": 6.40625,
      "learning_rate": 0.000495182909950648,
      "loss": 6.2385,
      "mean_token_accuracy": 0.134371055662632,
      "num_tokens": 17190619.0,
      "step": 8485
    },
    {
      "entropy": 6.2542366027832035,
      "epoch": 0.7419383028925981,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004951764943869117,
      "loss": 6.0278,
      "mean_token_accuracy": 0.14065780788660048,
      "num_tokens": 17201041.0,
      "step": 8490
    },
    {
      "entropy": 6.265298843383789,
      "epoch": 0.7423752512453028,
      "grad_norm": 6.75,
      "learning_rate": 0.0004951700746000749,
      "loss": 6.1246,
      "mean_token_accuracy": 0.1372768275439739,
      "num_tokens": 17211779.0,
      "step": 8495
    },
    {
      "entropy": 6.258416748046875,
      "epoch": 0.7428121995980075,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004951636505902606,
      "loss": 6.0938,
      "mean_token_accuracy": 0.14661502689123154,
      "num_tokens": 17221249.0,
      "step": 8500
    },
    {
      "entropy": 6.314711809158325,
      "epoch": 0.7432491479507122,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004951572223575922,
      "loss": 6.161,
      "mean_token_accuracy": 0.13593681082129477,
      "num_tokens": 17231634.0,
      "step": 8505
    },
    {
      "entropy": 6.360071992874145,
      "epoch": 0.743686096303417,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004951507899021929,
      "loss": 6.1459,
      "mean_token_accuracy": 0.13390619456768035,
      "num_tokens": 17241751.0,
      "step": 8510
    },
    {
      "entropy": 6.34587664604187,
      "epoch": 0.7441230446561217,
      "grad_norm": 4.3125,
      "learning_rate": 0.000495144353224186,
      "loss": 6.1577,
      "mean_token_accuracy": 0.1342741884291172,
      "num_tokens": 17251962.0,
      "step": 8515
    },
    {
      "entropy": 6.3369556903839115,
      "epoch": 0.7445599930088264,
      "grad_norm": 5.25,
      "learning_rate": 0.0004951379123236952,
      "loss": 6.0373,
      "mean_token_accuracy": 0.14202437177300453,
      "num_tokens": 17261895.0,
      "step": 8520
    },
    {
      "entropy": 6.3497782230377195,
      "epoch": 0.744996941361531,
      "grad_norm": 4.125,
      "learning_rate": 0.0004951314672008439,
      "loss": 6.1266,
      "mean_token_accuracy": 0.1377466045320034,
      "num_tokens": 17272025.0,
      "step": 8525
    },
    {
      "entropy": 6.339599895477295,
      "epoch": 0.7454338897142357,
      "grad_norm": 5.40625,
      "learning_rate": 0.0004951250178557556,
      "loss": 6.1156,
      "mean_token_accuracy": 0.13780690878629684,
      "num_tokens": 17282705.0,
      "step": 8530
    },
    {
      "entropy": 6.279929828643799,
      "epoch": 0.7458708380669404,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004951185642885542,
      "loss": 6.0547,
      "mean_token_accuracy": 0.1497568592429161,
      "num_tokens": 17293045.0,
      "step": 8535
    },
    {
      "entropy": 6.272597122192383,
      "epoch": 0.7463077864196452,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004951121064993633,
      "loss": 6.1658,
      "mean_token_accuracy": 0.1351180925965309,
      "num_tokens": 17303893.0,
      "step": 8540
    },
    {
      "entropy": 6.304271173477173,
      "epoch": 0.7467447347723499,
      "grad_norm": 5.125,
      "learning_rate": 0.0004951056444883069,
      "loss": 6.1126,
      "mean_token_accuracy": 0.135734910517931,
      "num_tokens": 17314250.0,
      "step": 8545
    },
    {
      "entropy": 6.325373554229737,
      "epoch": 0.7471816831250546,
      "grad_norm": 3.265625,
      "learning_rate": 0.000495099178255509,
      "loss": 6.0528,
      "mean_token_accuracy": 0.14083902388811112,
      "num_tokens": 17324268.0,
      "step": 8550
    },
    {
      "entropy": 6.330234527587891,
      "epoch": 0.7476186314777593,
      "grad_norm": 4.25,
      "learning_rate": 0.0004950927078010934,
      "loss": 6.1673,
      "mean_token_accuracy": 0.12885900139808654,
      "num_tokens": 17334512.0,
      "step": 8555
    },
    {
      "entropy": 6.318421220779419,
      "epoch": 0.748055579830464,
      "grad_norm": 4.625,
      "learning_rate": 0.0004950862331251845,
      "loss": 6.1285,
      "mean_token_accuracy": 0.14281737804412842,
      "num_tokens": 17346244.0,
      "step": 8560
    },
    {
      "entropy": 6.281004238128662,
      "epoch": 0.7484925281831688,
      "grad_norm": 8.75,
      "learning_rate": 0.0004950797542279061,
      "loss": 6.1456,
      "mean_token_accuracy": 0.14064847230911254,
      "num_tokens": 17355832.0,
      "step": 8565
    },
    {
      "entropy": 6.338721036911011,
      "epoch": 0.7489294765358735,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004950732711093828,
      "loss": 6.1597,
      "mean_token_accuracy": 0.13870706632733346,
      "num_tokens": 17366871.0,
      "step": 8570
    },
    {
      "entropy": 6.384263134002685,
      "epoch": 0.7493664248885782,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004950667837697388,
      "loss": 6.2001,
      "mean_token_accuracy": 0.1251978687942028,
      "num_tokens": 17377242.0,
      "step": 8575
    },
    {
      "entropy": 6.434356832504273,
      "epoch": 0.7498033732412829,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004950602922090988,
      "loss": 6.1597,
      "mean_token_accuracy": 0.13345890566706659,
      "num_tokens": 17387942.0,
      "step": 8580
    },
    {
      "entropy": 6.403023099899292,
      "epoch": 0.7502403215939876,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004950537964275868,
      "loss": 6.1425,
      "mean_token_accuracy": 0.12995332702994347,
      "num_tokens": 17397413.0,
      "step": 8585
    },
    {
      "entropy": 6.348374223709106,
      "epoch": 0.7506772699466923,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004950472964253278,
      "loss": 6.0938,
      "mean_token_accuracy": 0.13502024337649346,
      "num_tokens": 17408456.0,
      "step": 8590
    },
    {
      "entropy": 6.313529682159424,
      "epoch": 0.7511142182993971,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004950407922024463,
      "loss": 6.0879,
      "mean_token_accuracy": 0.14105957746505737,
      "num_tokens": 17417805.0,
      "step": 8595
    },
    {
      "entropy": 6.32263765335083,
      "epoch": 0.7515511666521018,
      "grad_norm": 7.5625,
      "learning_rate": 0.0004950342837590671,
      "loss": 6.0351,
      "mean_token_accuracy": 0.14731526821851731,
      "num_tokens": 17426932.0,
      "step": 8600
    },
    {
      "entropy": 6.283595085144043,
      "epoch": 0.7519881150048064,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004950277710953152,
      "loss": 6.1298,
      "mean_token_accuracy": 0.14013703614473344,
      "num_tokens": 17436770.0,
      "step": 8605
    },
    {
      "entropy": 6.256362676620483,
      "epoch": 0.7524250633575111,
      "grad_norm": 4.0,
      "learning_rate": 0.0004950212542113153,
      "loss": 6.0234,
      "mean_token_accuracy": 0.139969040453434,
      "num_tokens": 17447852.0,
      "step": 8610
    },
    {
      "entropy": 6.236438322067261,
      "epoch": 0.7528620117102158,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004950147331071924,
      "loss": 6.0464,
      "mean_token_accuracy": 0.1350158728659153,
      "num_tokens": 17458020.0,
      "step": 8615
    },
    {
      "entropy": 6.312313461303711,
      "epoch": 0.7532989600629205,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004950082077830717,
      "loss": 6.1468,
      "mean_token_accuracy": 0.13640186041593552,
      "num_tokens": 17468604.0,
      "step": 8620
    },
    {
      "entropy": 6.353255796432495,
      "epoch": 0.7537359084156253,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004950016782390782,
      "loss": 6.1128,
      "mean_token_accuracy": 0.1394171766936779,
      "num_tokens": 17478374.0,
      "step": 8625
    },
    {
      "entropy": 6.37905740737915,
      "epoch": 0.75417285676833,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004949951444753372,
      "loss": 6.1244,
      "mean_token_accuracy": 0.13772233128547667,
      "num_tokens": 17487894.0,
      "step": 8630
    },
    {
      "entropy": 6.373466348648071,
      "epoch": 0.7546098051210347,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004949886064919741,
      "loss": 6.0796,
      "mean_token_accuracy": 0.13704085126519203,
      "num_tokens": 17498473.0,
      "step": 8635
    },
    {
      "entropy": 6.3342742919921875,
      "epoch": 0.7550467534737394,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004949820642891142,
      "loss": 6.1111,
      "mean_token_accuracy": 0.13532850742340088,
      "num_tokens": 17507796.0,
      "step": 8640
    },
    {
      "entropy": 6.2665783882141115,
      "epoch": 0.7554837018264441,
      "grad_norm": 3.5,
      "learning_rate": 0.0004949755178668832,
      "loss": 6.1007,
      "mean_token_accuracy": 0.14242385923862458,
      "num_tokens": 17517629.0,
      "step": 8645
    },
    {
      "entropy": 6.346342897415161,
      "epoch": 0.7559206501791488,
      "grad_norm": 4.5,
      "learning_rate": 0.0004949689672254064,
      "loss": 6.1894,
      "mean_token_accuracy": 0.12464173063635826,
      "num_tokens": 17528726.0,
      "step": 8650
    },
    {
      "entropy": 6.348553800582886,
      "epoch": 0.7563575985318536,
      "grad_norm": 4.125,
      "learning_rate": 0.0004949624123648096,
      "loss": 6.0855,
      "mean_token_accuracy": 0.1432050459086895,
      "num_tokens": 17539022.0,
      "step": 8655
    },
    {
      "entropy": 6.336993885040283,
      "epoch": 0.7567945468845583,
      "grad_norm": 10.375,
      "learning_rate": 0.0004949558532852185,
      "loss": 6.1235,
      "mean_token_accuracy": 0.13665900975465775,
      "num_tokens": 17549293.0,
      "step": 8660
    },
    {
      "entropy": 6.378032541275024,
      "epoch": 0.757231495237263,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004949492899867587,
      "loss": 6.1028,
      "mean_token_accuracy": 0.1337118148803711,
      "num_tokens": 17559072.0,
      "step": 8665
    },
    {
      "entropy": 6.340274715423584,
      "epoch": 0.7576684435899677,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004949427224695564,
      "loss": 6.0833,
      "mean_token_accuracy": 0.13981281220912933,
      "num_tokens": 17568757.0,
      "step": 8670
    },
    {
      "entropy": 6.31947021484375,
      "epoch": 0.7581053919426723,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004949361507337374,
      "loss": 6.1053,
      "mean_token_accuracy": 0.13988024294376372,
      "num_tokens": 17578288.0,
      "step": 8675
    },
    {
      "entropy": 6.334840869903564,
      "epoch": 0.758542340295377,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004949295747794277,
      "loss": 6.1871,
      "mean_token_accuracy": 0.13526775017380716,
      "num_tokens": 17588864.0,
      "step": 8680
    },
    {
      "entropy": 6.370552396774292,
      "epoch": 0.7589792886480818,
      "grad_norm": 4.125,
      "learning_rate": 0.0004949229946067537,
      "loss": 6.1685,
      "mean_token_accuracy": 0.1344526492059231,
      "num_tokens": 17599103.0,
      "step": 8685
    },
    {
      "entropy": 6.310125112533569,
      "epoch": 0.7594162370007865,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004949164102158413,
      "loss": 6.0696,
      "mean_token_accuracy": 0.1386950768530369,
      "num_tokens": 17608608.0,
      "step": 8690
    },
    {
      "entropy": 6.324038076400757,
      "epoch": 0.7598531853534912,
      "grad_norm": 3.015625,
      "learning_rate": 0.000494909821606817,
      "loss": 6.1705,
      "mean_token_accuracy": 0.1365269348025322,
      "num_tokens": 17618604.0,
      "step": 8695
    },
    {
      "entropy": 6.3650219440460205,
      "epoch": 0.7602901337061959,
      "grad_norm": 6.96875,
      "learning_rate": 0.0004949032287798069,
      "loss": 6.1242,
      "mean_token_accuracy": 0.13540336787700652,
      "num_tokens": 17628479.0,
      "step": 8700
    },
    {
      "entropy": 6.354051876068115,
      "epoch": 0.7607270820589006,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004948966317349379,
      "loss": 6.1272,
      "mean_token_accuracy": 0.14550747722387314,
      "num_tokens": 17637519.0,
      "step": 8705
    },
    {
      "entropy": 6.318744087219239,
      "epoch": 0.7611640304116053,
      "grad_norm": 2.75,
      "learning_rate": 0.0004948900304723361,
      "loss": 6.1032,
      "mean_token_accuracy": 0.14002725481987,
      "num_tokens": 17647902.0,
      "step": 8710
    },
    {
      "entropy": 6.341483783721924,
      "epoch": 0.7616009787643101,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004948834249921285,
      "loss": 6.1177,
      "mean_token_accuracy": 0.13890863582491875,
      "num_tokens": 17658576.0,
      "step": 8715
    },
    {
      "entropy": 6.35099687576294,
      "epoch": 0.7620379271170148,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004948768152944414,
      "loss": 6.1433,
      "mean_token_accuracy": 0.13804591372609137,
      "num_tokens": 17667059.0,
      "step": 8720
    },
    {
      "entropy": 6.387050580978394,
      "epoch": 0.7624748754697195,
      "grad_norm": 4.125,
      "learning_rate": 0.0004948702013794018,
      "loss": 6.2326,
      "mean_token_accuracy": 0.1327890172600746,
      "num_tokens": 17677447.0,
      "step": 8725
    },
    {
      "entropy": 6.353907442092895,
      "epoch": 0.7629118238224242,
      "grad_norm": 3.375,
      "learning_rate": 0.0004948635832471366,
      "loss": 6.1093,
      "mean_token_accuracy": 0.14533415511250497,
      "num_tokens": 17687893.0,
      "step": 8730
    },
    {
      "entropy": 6.3049877166748045,
      "epoch": 0.7633487721751289,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004948569608977726,
      "loss": 6.1521,
      "mean_token_accuracy": 0.13721927627921104,
      "num_tokens": 17698599.0,
      "step": 8735
    },
    {
      "entropy": 6.289543485641479,
      "epoch": 0.7637857205278336,
      "grad_norm": 4.0,
      "learning_rate": 0.0004948503343314369,
      "loss": 6.1437,
      "mean_token_accuracy": 0.13201402947306634,
      "num_tokens": 17709711.0,
      "step": 8740
    },
    {
      "entropy": 6.304942846298218,
      "epoch": 0.7642226688805384,
      "grad_norm": 3.625,
      "learning_rate": 0.0004948437035482566,
      "loss": 6.1182,
      "mean_token_accuracy": 0.13628509640693665,
      "num_tokens": 17719613.0,
      "step": 8745
    },
    {
      "entropy": 6.348080444335937,
      "epoch": 0.764659617233243,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004948370685483588,
      "loss": 6.1001,
      "mean_token_accuracy": 0.1430316984653473,
      "num_tokens": 17729829.0,
      "step": 8750
    },
    {
      "entropy": 6.380725240707397,
      "epoch": 0.7650965655859477,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004948304293318709,
      "loss": 6.092,
      "mean_token_accuracy": 0.1387576162815094,
      "num_tokens": 17739254.0,
      "step": 8755
    },
    {
      "entropy": 6.347923707962036,
      "epoch": 0.7655335139386524,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004948237858989201,
      "loss": 6.1089,
      "mean_token_accuracy": 0.13667647540569305,
      "num_tokens": 17749294.0,
      "step": 8760
    },
    {
      "entropy": 6.283276987075806,
      "epoch": 0.7659704622913571,
      "grad_norm": 3.21875,
      "learning_rate": 0.000494817138249634,
      "loss": 6.0452,
      "mean_token_accuracy": 0.1471354141831398,
      "num_tokens": 17758866.0,
      "step": 8765
    },
    {
      "entropy": 6.398133420944214,
      "epoch": 0.7664074106440619,
      "grad_norm": 5.125,
      "learning_rate": 0.00049481048638414,
      "loss": 6.2241,
      "mean_token_accuracy": 0.12746726348996162,
      "num_tokens": 17769202.0,
      "step": 8770
    },
    {
      "entropy": 6.302561950683594,
      "epoch": 0.7668443589967666,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004948038303025657,
      "loss": 6.1776,
      "mean_token_accuracy": 0.13436749428510666,
      "num_tokens": 17779125.0,
      "step": 8775
    },
    {
      "entropy": 6.3434587001800535,
      "epoch": 0.7672813073494713,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004947971700050387,
      "loss": 6.1331,
      "mean_token_accuracy": 0.13466654866933822,
      "num_tokens": 17788696.0,
      "step": 8780
    },
    {
      "entropy": 6.323318910598755,
      "epoch": 0.767718255702176,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004947905054916869,
      "loss": 6.1342,
      "mean_token_accuracy": 0.13619852140545846,
      "num_tokens": 17799867.0,
      "step": 8785
    },
    {
      "entropy": 6.370768737792969,
      "epoch": 0.7681552040548807,
      "grad_norm": 5.875,
      "learning_rate": 0.0004947838367626379,
      "loss": 6.1065,
      "mean_token_accuracy": 0.13464439362287522,
      "num_tokens": 17809418.0,
      "step": 8790
    },
    {
      "entropy": 6.2917236328125,
      "epoch": 0.7685921524075854,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004947771638180198,
      "loss": 6.0498,
      "mean_token_accuracy": 0.14432308450341225,
      "num_tokens": 17818554.0,
      "step": 8795
    },
    {
      "entropy": 6.2488714218139645,
      "epoch": 0.7690291007602902,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004947704866579605,
      "loss": 6.0983,
      "mean_token_accuracy": 0.14032392278313638,
      "num_tokens": 17828918.0,
      "step": 8800
    },
    {
      "entropy": 6.367287635803223,
      "epoch": 0.7694660491129949,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004947638052825883,
      "loss": 6.2381,
      "mean_token_accuracy": 0.12880766093730928,
      "num_tokens": 17840791.0,
      "step": 8805
    },
    {
      "entropy": 6.384181356430053,
      "epoch": 0.7699029974656996,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004947571196920309,
      "loss": 6.1377,
      "mean_token_accuracy": 0.13416934832930566,
      "num_tokens": 17850675.0,
      "step": 8810
    },
    {
      "entropy": 6.342396354675293,
      "epoch": 0.7703399458184043,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004947504298864169,
      "loss": 6.1065,
      "mean_token_accuracy": 0.13892019391059876,
      "num_tokens": 17860684.0,
      "step": 8815
    },
    {
      "entropy": 6.309155607223511,
      "epoch": 0.770776894171109,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004947437358658746,
      "loss": 6.115,
      "mean_token_accuracy": 0.13639792203903198,
      "num_tokens": 17870394.0,
      "step": 8820
    },
    {
      "entropy": 6.354349803924561,
      "epoch": 0.7712138425238136,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004947370376305321,
      "loss": 6.1006,
      "mean_token_accuracy": 0.13592427223920822,
      "num_tokens": 17881256.0,
      "step": 8825
    },
    {
      "entropy": 6.288863563537598,
      "epoch": 0.7716507908765184,
      "grad_norm": 5.8125,
      "learning_rate": 0.0004947303351805183,
      "loss": 6.1309,
      "mean_token_accuracy": 0.1394450284540653,
      "num_tokens": 17890888.0,
      "step": 8830
    },
    {
      "entropy": 6.319786548614502,
      "epoch": 0.7720877392292231,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004947236285159615,
      "loss": 6.1032,
      "mean_token_accuracy": 0.13891692012548446,
      "num_tokens": 17900507.0,
      "step": 8835
    },
    {
      "entropy": 6.363798999786377,
      "epoch": 0.7725246875819278,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004947169176369902,
      "loss": 6.1674,
      "mean_token_accuracy": 0.13376756832003595,
      "num_tokens": 17911333.0,
      "step": 8840
    },
    {
      "entropy": 6.352057981491089,
      "epoch": 0.7729616359346325,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004947102025437334,
      "loss": 6.1297,
      "mean_token_accuracy": 0.134784222394228,
      "num_tokens": 17921540.0,
      "step": 8845
    },
    {
      "entropy": 6.3458559036254885,
      "epoch": 0.7733985842873372,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004947034832363199,
      "loss": 6.2215,
      "mean_token_accuracy": 0.1366083487868309,
      "num_tokens": 17931820.0,
      "step": 8850
    },
    {
      "entropy": 6.32614312171936,
      "epoch": 0.7738355326400419,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004946967597148782,
      "loss": 6.1282,
      "mean_token_accuracy": 0.1310492865741253,
      "num_tokens": 17942717.0,
      "step": 8855
    },
    {
      "entropy": 6.331774377822876,
      "epoch": 0.7742724809927467,
      "grad_norm": 5.0,
      "learning_rate": 0.0004946900319795377,
      "loss": 6.1184,
      "mean_token_accuracy": 0.13957384824752808,
      "num_tokens": 17953775.0,
      "step": 8860
    },
    {
      "entropy": 6.330040645599365,
      "epoch": 0.7747094293454514,
      "grad_norm": 12.25,
      "learning_rate": 0.0004946833000304274,
      "loss": 6.1199,
      "mean_token_accuracy": 0.13655293360352516,
      "num_tokens": 17963877.0,
      "step": 8865
    },
    {
      "entropy": 6.336996269226074,
      "epoch": 0.7751463776981561,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004946765638676761,
      "loss": 6.0998,
      "mean_token_accuracy": 0.1423903562128544,
      "num_tokens": 17974015.0,
      "step": 8870
    },
    {
      "entropy": 6.305118227005005,
      "epoch": 0.7755833260508608,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004946698234914132,
      "loss": 6.1433,
      "mean_token_accuracy": 0.13647328838706016,
      "num_tokens": 17984189.0,
      "step": 8875
    },
    {
      "entropy": 6.302814102172851,
      "epoch": 0.7760202744035655,
      "grad_norm": 3.703125,
      "learning_rate": 0.000494663078901768,
      "loss": 6.0892,
      "mean_token_accuracy": 0.14488258212804794,
      "num_tokens": 17994615.0,
      "step": 8880
    },
    {
      "entropy": 6.350739669799805,
      "epoch": 0.7764572227562702,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004946563300988699,
      "loss": 6.1016,
      "mean_token_accuracy": 0.13663954585790633,
      "num_tokens": 18004727.0,
      "step": 8885
    },
    {
      "entropy": 6.360508298873901,
      "epoch": 0.776894171108975,
      "grad_norm": 7.25,
      "learning_rate": 0.0004946495770828482,
      "loss": 6.1581,
      "mean_token_accuracy": 0.1373465873301029,
      "num_tokens": 18014576.0,
      "step": 8890
    },
    {
      "entropy": 6.363616561889648,
      "epoch": 0.7773311194616797,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004946428198538326,
      "loss": 6.125,
      "mean_token_accuracy": 0.14108447805047036,
      "num_tokens": 18024709.0,
      "step": 8895
    },
    {
      "entropy": 6.309369230270386,
      "epoch": 0.7777680678143843,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004946360584119526,
      "loss": 6.0874,
      "mean_token_accuracy": 0.13786243125796319,
      "num_tokens": 18034673.0,
      "step": 8900
    },
    {
      "entropy": 6.304690504074097,
      "epoch": 0.778205016167089,
      "grad_norm": 3.140625,
      "learning_rate": 0.000494629292757338,
      "loss": 6.1114,
      "mean_token_accuracy": 0.1372987873852253,
      "num_tokens": 18044774.0,
      "step": 8905
    },
    {
      "entropy": 6.304149580001831,
      "epoch": 0.7786419645197937,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004946225228901184,
      "loss": 6.0987,
      "mean_token_accuracy": 0.13560848236083983,
      "num_tokens": 18055244.0,
      "step": 8910
    },
    {
      "entropy": 6.366212272644043,
      "epoch": 0.7790789128724984,
      "grad_norm": 5.59375,
      "learning_rate": 0.0004946157488104236,
      "loss": 6.2037,
      "mean_token_accuracy": 0.13057628497481347,
      "num_tokens": 18066051.0,
      "step": 8915
    },
    {
      "entropy": 6.386141633987426,
      "epoch": 0.7795158612252032,
      "grad_norm": 3.125,
      "learning_rate": 0.0004946089705183839,
      "loss": 6.2288,
      "mean_token_accuracy": 0.12433409839868545,
      "num_tokens": 18077693.0,
      "step": 8920
    },
    {
      "entropy": 6.363635015487671,
      "epoch": 0.7799528095779079,
      "grad_norm": 5.875,
      "learning_rate": 0.0004946021880141289,
      "loss": 6.1658,
      "mean_token_accuracy": 0.13873492777347565,
      "num_tokens": 18088032.0,
      "step": 8925
    },
    {
      "entropy": 6.312368726730346,
      "epoch": 0.7803897579306126,
      "grad_norm": 2.546875,
      "learning_rate": 0.000494595401297789,
      "loss": 6.1198,
      "mean_token_accuracy": 0.13863285034894943,
      "num_tokens": 18097998.0,
      "step": 8930
    },
    {
      "entropy": 6.30918779373169,
      "epoch": 0.7808267062833173,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004945886103694941,
      "loss": 6.0723,
      "mean_token_accuracy": 0.14402743503451348,
      "num_tokens": 18107538.0,
      "step": 8935
    },
    {
      "entropy": 6.299681568145752,
      "epoch": 0.781263654636022,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004945818152293747,
      "loss": 6.1204,
      "mean_token_accuracy": 0.13765494748950005,
      "num_tokens": 18117792.0,
      "step": 8940
    },
    {
      "entropy": 6.335578441619873,
      "epoch": 0.7817006029887267,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004945750158775609,
      "loss": 6.1407,
      "mean_token_accuracy": 0.13752539902925492,
      "num_tokens": 18128005.0,
      "step": 8945
    },
    {
      "entropy": 6.291286945343018,
      "epoch": 0.7821375513414315,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004945682123141833,
      "loss": 6.1312,
      "mean_token_accuracy": 0.14197242707014085,
      "num_tokens": 18138398.0,
      "step": 8950
    },
    {
      "entropy": 6.384673357009888,
      "epoch": 0.7825744996941362,
      "grad_norm": 2.875,
      "learning_rate": 0.0004945614045393724,
      "loss": 6.103,
      "mean_token_accuracy": 0.13442164435982704,
      "num_tokens": 18147676.0,
      "step": 8955
    },
    {
      "entropy": 6.281360530853272,
      "epoch": 0.7830114480468409,
      "grad_norm": 5.53125,
      "learning_rate": 0.0004945545925532587,
      "loss": 6.011,
      "mean_token_accuracy": 0.14690378531813622,
      "num_tokens": 18156842.0,
      "step": 8960
    },
    {
      "entropy": 6.276273775100708,
      "epoch": 0.7834483963995456,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004945477763559728,
      "loss": 6.1268,
      "mean_token_accuracy": 0.1400029681622982,
      "num_tokens": 18166902.0,
      "step": 8965
    },
    {
      "entropy": 6.374386739730835,
      "epoch": 0.7838853447522502,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004945409559476455,
      "loss": 6.1192,
      "mean_token_accuracy": 0.13782411366701125,
      "num_tokens": 18177561.0,
      "step": 8970
    },
    {
      "entropy": 6.320522594451904,
      "epoch": 0.784322293104955,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004945341313284077,
      "loss": 6.1129,
      "mean_token_accuracy": 0.14229698330163956,
      "num_tokens": 18187739.0,
      "step": 8975
    },
    {
      "entropy": 6.288054132461548,
      "epoch": 0.7847592414576597,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004945273024983902,
      "loss": 6.0457,
      "mean_token_accuracy": 0.14413127526640893,
      "num_tokens": 18198435.0,
      "step": 8980
    },
    {
      "entropy": 6.314607286453247,
      "epoch": 0.7851961898103644,
      "grad_norm": 4.0,
      "learning_rate": 0.000494520469457724,
      "loss": 6.0615,
      "mean_token_accuracy": 0.14262245818972588,
      "num_tokens": 18208374.0,
      "step": 8985
    },
    {
      "entropy": 6.248941230773926,
      "epoch": 0.7856331381630691,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004945136322065402,
      "loss": 6.1369,
      "mean_token_accuracy": 0.14011382460594177,
      "num_tokens": 18219229.0,
      "step": 8990
    },
    {
      "entropy": 6.372687911987304,
      "epoch": 0.7860700865157738,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004945067907449699,
      "loss": 6.1213,
      "mean_token_accuracy": 0.1343177653849125,
      "num_tokens": 18229208.0,
      "step": 8995
    },
    {
      "entropy": 6.450346040725708,
      "epoch": 0.7865070348684785,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004944999450731442,
      "loss": 6.1983,
      "mean_token_accuracy": 0.1274542599916458,
      "num_tokens": 18238454.0,
      "step": 9000
    },
    {
      "epoch": 0.7865070348684785,
      "eval_entropy": 6.25084467162987,
      "eval_loss": 6.142418384552002,
      "eval_mean_token_accuracy": 0.14447606859963386,
      "eval_num_tokens": 18238454.0,
      "eval_runtime": 27.7943,
      "eval_samples_per_second": 1331.712,
      "eval_steps_per_second": 166.473,
      "step": 9000
    },
    {
      "entropy": 6.413679885864258,
      "epoch": 0.7869439832211833,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004944930951911947,
      "loss": 6.1216,
      "mean_token_accuracy": 0.138817098736763,
      "num_tokens": 18248845.0,
      "step": 9005
    },
    {
      "entropy": 6.320638275146484,
      "epoch": 0.787380931573888,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004944862410992526,
      "loss": 6.0939,
      "mean_token_accuracy": 0.14076521769165992,
      "num_tokens": 18258912.0,
      "step": 9010
    },
    {
      "entropy": 6.335213613510132,
      "epoch": 0.7878178799265927,
      "grad_norm": 8.5,
      "learning_rate": 0.0004944793827974494,
      "loss": 6.1492,
      "mean_token_accuracy": 0.13492099940776825,
      "num_tokens": 18269330.0,
      "step": 9015
    },
    {
      "entropy": 6.3489879131317135,
      "epoch": 0.7882548282792974,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004944725202859165,
      "loss": 6.1339,
      "mean_token_accuracy": 0.1364932857453823,
      "num_tokens": 18279721.0,
      "step": 9020
    },
    {
      "entropy": 6.262695550918579,
      "epoch": 0.7886917766320021,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004944656535647858,
      "loss": 6.0405,
      "mean_token_accuracy": 0.14772520437836648,
      "num_tokens": 18288581.0,
      "step": 9025
    },
    {
      "entropy": 6.233278226852417,
      "epoch": 0.7891287249847068,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004944587826341888,
      "loss": 6.0943,
      "mean_token_accuracy": 0.13744651898741722,
      "num_tokens": 18298179.0,
      "step": 9030
    },
    {
      "entropy": 6.355013132095337,
      "epoch": 0.7895656733374116,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004944519074942574,
      "loss": 6.0904,
      "mean_token_accuracy": 0.1357803873717785,
      "num_tokens": 18307957.0,
      "step": 9035
    },
    {
      "entropy": 6.334473943710327,
      "epoch": 0.7900026216901163,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004944450281451234,
      "loss": 6.0939,
      "mean_token_accuracy": 0.13603418916463852,
      "num_tokens": 18317799.0,
      "step": 9040
    },
    {
      "entropy": 6.3885739803314205,
      "epoch": 0.790439570042821,
      "grad_norm": 6.125,
      "learning_rate": 0.0004944381445869188,
      "loss": 6.162,
      "mean_token_accuracy": 0.1330812856554985,
      "num_tokens": 18327889.0,
      "step": 9045
    },
    {
      "entropy": 6.3655026912689205,
      "epoch": 0.7908765183955256,
      "grad_norm": 3.5,
      "learning_rate": 0.0004944312568197756,
      "loss": 6.139,
      "mean_token_accuracy": 0.13588144779205322,
      "num_tokens": 18338288.0,
      "step": 9050
    },
    {
      "entropy": 6.3286608219146725,
      "epoch": 0.7913134667482303,
      "grad_norm": 7.75,
      "learning_rate": 0.0004944243648438258,
      "loss": 6.1814,
      "mean_token_accuracy": 0.13608478307723998,
      "num_tokens": 18349070.0,
      "step": 9055
    },
    {
      "entropy": 6.394107818603516,
      "epoch": 0.791750415100935,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004944174686592019,
      "loss": 6.1024,
      "mean_token_accuracy": 0.1398714780807495,
      "num_tokens": 18359683.0,
      "step": 9060
    },
    {
      "entropy": 6.420367622375489,
      "epoch": 0.7921873634536398,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004944105682660358,
      "loss": 6.1219,
      "mean_token_accuracy": 0.1339648388326168,
      "num_tokens": 18371355.0,
      "step": 9065
    },
    {
      "entropy": 6.2922303676605225,
      "epoch": 0.7926243118063445,
      "grad_norm": 4.375,
      "learning_rate": 0.0004944036636644601,
      "loss": 6.1385,
      "mean_token_accuracy": 0.13980514258146287,
      "num_tokens": 18381591.0,
      "step": 9070
    },
    {
      "entropy": 6.3173857688903805,
      "epoch": 0.7930612601590492,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004943967548546071,
      "loss": 6.1027,
      "mean_token_accuracy": 0.13918379694223404,
      "num_tokens": 18390760.0,
      "step": 9075
    },
    {
      "entropy": 6.348713159561157,
      "epoch": 0.7934982085117539,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004943898418366096,
      "loss": 6.1532,
      "mean_token_accuracy": 0.13411854654550553,
      "num_tokens": 18400643.0,
      "step": 9080
    },
    {
      "entropy": 6.3693907260894775,
      "epoch": 0.7939351568644586,
      "grad_norm": 5.0,
      "learning_rate": 0.0004943829246105997,
      "loss": 6.2103,
      "mean_token_accuracy": 0.13443097919225694,
      "num_tokens": 18411559.0,
      "step": 9085
    },
    {
      "entropy": 6.409049987792969,
      "epoch": 0.7943721052171633,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004943760031767106,
      "loss": 6.1789,
      "mean_token_accuracy": 0.1374424912035465,
      "num_tokens": 18422645.0,
      "step": 9090
    },
    {
      "entropy": 6.381908941268921,
      "epoch": 0.7948090535698681,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004943690775350746,
      "loss": 6.1318,
      "mean_token_accuracy": 0.13344489857554437,
      "num_tokens": 18432369.0,
      "step": 9095
    },
    {
      "entropy": 6.322442674636841,
      "epoch": 0.7952460019225728,
      "grad_norm": 4.125,
      "learning_rate": 0.0004943621476858249,
      "loss": 6.0936,
      "mean_token_accuracy": 0.13766884431242943,
      "num_tokens": 18442143.0,
      "step": 9100
    },
    {
      "entropy": 6.268150806427002,
      "epoch": 0.7956829502752775,
      "grad_norm": 31.0,
      "learning_rate": 0.0004943552136290942,
      "loss": 6.1477,
      "mean_token_accuracy": 0.1359400451183319,
      "num_tokens": 18452161.0,
      "step": 9105
    },
    {
      "entropy": 6.334364414215088,
      "epoch": 0.7961198986279822,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004943482753650154,
      "loss": 6.1987,
      "mean_token_accuracy": 0.13220474123954773,
      "num_tokens": 18461873.0,
      "step": 9110
    },
    {
      "entropy": 6.343059635162353,
      "epoch": 0.7965568469806869,
      "grad_norm": 4.9375,
      "learning_rate": 0.0004943413328937221,
      "loss": 6.1203,
      "mean_token_accuracy": 0.13832321614027024,
      "num_tokens": 18472584.0,
      "step": 9115
    },
    {
      "entropy": 6.353497123718261,
      "epoch": 0.7969937953333915,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004943343862153467,
      "loss": 6.0802,
      "mean_token_accuracy": 0.13698331639170647,
      "num_tokens": 18482174.0,
      "step": 9120
    },
    {
      "entropy": 6.378763914108276,
      "epoch": 0.7974307436860963,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004943274353300231,
      "loss": 6.2269,
      "mean_token_accuracy": 0.13241873309016228,
      "num_tokens": 18492861.0,
      "step": 9125
    },
    {
      "entropy": 6.330440998077393,
      "epoch": 0.797867692038801,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004943204802378843,
      "loss": 6.133,
      "mean_token_accuracy": 0.14144847691059112,
      "num_tokens": 18502830.0,
      "step": 9130
    },
    {
      "entropy": 6.38723726272583,
      "epoch": 0.7983046403915057,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004943135209390638,
      "loss": 6.1146,
      "mean_token_accuracy": 0.13234059438109397,
      "num_tokens": 18513063.0,
      "step": 9135
    },
    {
      "entropy": 6.3327476501464846,
      "epoch": 0.7987415887442104,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004943065574336949,
      "loss": 6.1125,
      "mean_token_accuracy": 0.14246777221560478,
      "num_tokens": 18523773.0,
      "step": 9140
    },
    {
      "entropy": 6.325470209121704,
      "epoch": 0.7991785370969151,
      "grad_norm": 6.625,
      "learning_rate": 0.0004942995897219115,
      "loss": 6.1359,
      "mean_token_accuracy": 0.13829580172896386,
      "num_tokens": 18533101.0,
      "step": 9145
    },
    {
      "entropy": 6.334008979797363,
      "epoch": 0.7996154854496198,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004942926178038469,
      "loss": 6.1561,
      "mean_token_accuracy": 0.13714182898402213,
      "num_tokens": 18543863.0,
      "step": 9150
    },
    {
      "entropy": 6.262722206115723,
      "epoch": 0.8000524338023246,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004942856416796352,
      "loss": 6.1276,
      "mean_token_accuracy": 0.13816729709506034,
      "num_tokens": 18554249.0,
      "step": 9155
    },
    {
      "entropy": 6.319381046295166,
      "epoch": 0.8004893821550293,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004942786613494098,
      "loss": 6.1089,
      "mean_token_accuracy": 0.1372742734849453,
      "num_tokens": 18563525.0,
      "step": 9160
    },
    {
      "entropy": 6.423503780364991,
      "epoch": 0.800926330507734,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004942716768133048,
      "loss": 6.1072,
      "mean_token_accuracy": 0.13621978759765624,
      "num_tokens": 18572734.0,
      "step": 9165
    },
    {
      "entropy": 6.3984089374542235,
      "epoch": 0.8013632788604387,
      "grad_norm": 7.3125,
      "learning_rate": 0.0004942646880714543,
      "loss": 6.1617,
      "mean_token_accuracy": 0.13423240557312965,
      "num_tokens": 18581939.0,
      "step": 9170
    },
    {
      "entropy": 6.280920505523682,
      "epoch": 0.8018002272131434,
      "grad_norm": 2.953125,
      "learning_rate": 0.000494257695123992,
      "loss": 6.1416,
      "mean_token_accuracy": 0.13927967026829718,
      "num_tokens": 18591635.0,
      "step": 9175
    },
    {
      "entropy": 6.308006000518799,
      "epoch": 0.8022371755658482,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004942506979710524,
      "loss": 6.0661,
      "mean_token_accuracy": 0.13813204541802407,
      "num_tokens": 18602060.0,
      "step": 9180
    },
    {
      "entropy": 6.347301721572876,
      "epoch": 0.8026741239185529,
      "grad_norm": 4.125,
      "learning_rate": 0.0004942436966127693,
      "loss": 6.1116,
      "mean_token_accuracy": 0.13629916235804557,
      "num_tokens": 18611187.0,
      "step": 9185
    },
    {
      "entropy": 6.375120401382446,
      "epoch": 0.8031110722712576,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004942366910492775,
      "loss": 6.0458,
      "mean_token_accuracy": 0.14481936544179916,
      "num_tokens": 18620329.0,
      "step": 9190
    },
    {
      "entropy": 6.314260244369507,
      "epoch": 0.8035480206239622,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004942296812807109,
      "loss": 6.1084,
      "mean_token_accuracy": 0.1347486101090908,
      "num_tokens": 18629600.0,
      "step": 9195
    },
    {
      "entropy": 6.3188141822814945,
      "epoch": 0.8039849689766669,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004942226673072041,
      "loss": 6.1495,
      "mean_token_accuracy": 0.13466936498880386,
      "num_tokens": 18640771.0,
      "step": 9200
    },
    {
      "entropy": 6.263028621673584,
      "epoch": 0.8044219173293716,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004942156491288917,
      "loss": 6.0532,
      "mean_token_accuracy": 0.14267104715108872,
      "num_tokens": 18650606.0,
      "step": 9205
    },
    {
      "entropy": 6.314154767990113,
      "epoch": 0.8048588656820764,
      "grad_norm": 4.75,
      "learning_rate": 0.0004942086267459083,
      "loss": 6.1181,
      "mean_token_accuracy": 0.13790291026234627,
      "num_tokens": 18660487.0,
      "step": 9210
    },
    {
      "entropy": 6.327486562728882,
      "epoch": 0.8052958140347811,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004942016001583886,
      "loss": 6.1148,
      "mean_token_accuracy": 0.14065281972289084,
      "num_tokens": 18670987.0,
      "step": 9215
    },
    {
      "entropy": 6.318988943099976,
      "epoch": 0.8057327623874858,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004941945693664674,
      "loss": 6.1796,
      "mean_token_accuracy": 0.12995615229010582,
      "num_tokens": 18682560.0,
      "step": 9220
    },
    {
      "entropy": 6.313672065734863,
      "epoch": 0.8061697107401905,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004941875343702795,
      "loss": 6.1563,
      "mean_token_accuracy": 0.1352015405893326,
      "num_tokens": 18692743.0,
      "step": 9225
    },
    {
      "entropy": 6.305946493148804,
      "epoch": 0.8066066590928952,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004941804951699597,
      "loss": 6.0624,
      "mean_token_accuracy": 0.14648055136203766,
      "num_tokens": 18703039.0,
      "step": 9230
    },
    {
      "entropy": 6.270686101913452,
      "epoch": 0.8070436074455999,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004941734517656432,
      "loss": 6.1567,
      "mean_token_accuracy": 0.13572880551218985,
      "num_tokens": 18713777.0,
      "step": 9235
    },
    {
      "entropy": 6.3084405899047855,
      "epoch": 0.8074805557983047,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004941664041574651,
      "loss": 6.1439,
      "mean_token_accuracy": 0.1358615830540657,
      "num_tokens": 18723422.0,
      "step": 9240
    },
    {
      "entropy": 6.351822566986084,
      "epoch": 0.8079175041510094,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004941593523455605,
      "loss": 6.1011,
      "mean_token_accuracy": 0.13754042834043503,
      "num_tokens": 18734369.0,
      "step": 9245
    },
    {
      "entropy": 6.365796899795532,
      "epoch": 0.8083544525037141,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004941522963300647,
      "loss": 6.1569,
      "mean_token_accuracy": 0.13374169319868087,
      "num_tokens": 18744647.0,
      "step": 9250
    },
    {
      "entropy": 6.3727634906768795,
      "epoch": 0.8087914008564188,
      "grad_norm": 4.09375,
      "learning_rate": 0.000494145236111113,
      "loss": 6.1338,
      "mean_token_accuracy": 0.13805847242474556,
      "num_tokens": 18753929.0,
      "step": 9255
    },
    {
      "entropy": 6.33428201675415,
      "epoch": 0.8092283492091235,
      "grad_norm": 2.625,
      "learning_rate": 0.0004941381716888409,
      "loss": 6.1076,
      "mean_token_accuracy": 0.13491604775190352,
      "num_tokens": 18764103.0,
      "step": 9260
    },
    {
      "entropy": 6.2914961814880375,
      "epoch": 0.8096652975618281,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004941311030633838,
      "loss": 6.1432,
      "mean_token_accuracy": 0.13503317907452583,
      "num_tokens": 18773756.0,
      "step": 9265
    },
    {
      "entropy": 6.327725982666015,
      "epoch": 0.810102245914533,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004941240302348773,
      "loss": 6.0185,
      "mean_token_accuracy": 0.15296769067645072,
      "num_tokens": 18783778.0,
      "step": 9270
    },
    {
      "entropy": 6.3459676742553714,
      "epoch": 0.8105391942672376,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004941169532034571,
      "loss": 6.1588,
      "mean_token_accuracy": 0.1339312747120857,
      "num_tokens": 18794664.0,
      "step": 9275
    },
    {
      "entropy": 6.280435037612915,
      "epoch": 0.8109761426199423,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004941098719692589,
      "loss": 6.0885,
      "mean_token_accuracy": 0.1346994385123253,
      "num_tokens": 18803773.0,
      "step": 9280
    },
    {
      "entropy": 6.316810846328735,
      "epoch": 0.811413090972647,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004941027865324186,
      "loss": 6.148,
      "mean_token_accuracy": 0.13416729792952536,
      "num_tokens": 18814419.0,
      "step": 9285
    },
    {
      "entropy": 6.346670198440552,
      "epoch": 0.8118500393253517,
      "grad_norm": 2.671875,
      "learning_rate": 0.000494095696893072,
      "loss": 6.1047,
      "mean_token_accuracy": 0.13342042416334152,
      "num_tokens": 18825011.0,
      "step": 9290
    },
    {
      "entropy": 6.31347622871399,
      "epoch": 0.8122869876780564,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004940886030513551,
      "loss": 6.0216,
      "mean_token_accuracy": 0.1385837107896805,
      "num_tokens": 18835003.0,
      "step": 9295
    },
    {
      "entropy": 6.301917886734008,
      "epoch": 0.8127239360307612,
      "grad_norm": 6.96875,
      "learning_rate": 0.0004940815050074041,
      "loss": 6.135,
      "mean_token_accuracy": 0.13724869936704637,
      "num_tokens": 18845072.0,
      "step": 9300
    },
    {
      "entropy": 6.334423780441284,
      "epoch": 0.8131608843834659,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004940744027613549,
      "loss": 6.0877,
      "mean_token_accuracy": 0.14113619774580002,
      "num_tokens": 18854583.0,
      "step": 9305
    },
    {
      "entropy": 6.330839014053344,
      "epoch": 0.8135978327361706,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004940672963133439,
      "loss": 6.1441,
      "mean_token_accuracy": 0.1413642071187496,
      "num_tokens": 18865460.0,
      "step": 9310
    },
    {
      "entropy": 6.353619241714478,
      "epoch": 0.8140347810888753,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004940601856635074,
      "loss": 6.1188,
      "mean_token_accuracy": 0.1418474294245243,
      "num_tokens": 18875568.0,
      "step": 9315
    },
    {
      "entropy": 6.238967609405518,
      "epoch": 0.81447172944158,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004940530708119818,
      "loss": 6.0335,
      "mean_token_accuracy": 0.14070352241396905,
      "num_tokens": 18884894.0,
      "step": 9320
    },
    {
      "entropy": 6.258545827865601,
      "epoch": 0.8149086777942847,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004940459517589034,
      "loss": 6.0998,
      "mean_token_accuracy": 0.14078273847699166,
      "num_tokens": 18895353.0,
      "step": 9325
    },
    {
      "entropy": 6.340686845779419,
      "epoch": 0.8153456261469895,
      "grad_norm": 3.46875,
      "learning_rate": 0.000494038828504409,
      "loss": 6.1344,
      "mean_token_accuracy": 0.14093361496925355,
      "num_tokens": 18905474.0,
      "step": 9330
    },
    {
      "entropy": 6.3816570281982425,
      "epoch": 0.8157825744996942,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004940317010486351,
      "loss": 6.0889,
      "mean_token_accuracy": 0.139781591296196,
      "num_tokens": 18914797.0,
      "step": 9335
    },
    {
      "entropy": 6.372083902359009,
      "epoch": 0.8162195228523988,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004940245693917183,
      "loss": 6.1271,
      "mean_token_accuracy": 0.1346873514354229,
      "num_tokens": 18924642.0,
      "step": 9340
    },
    {
      "entropy": 6.262750339508057,
      "epoch": 0.8166564712051035,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004940174335337955,
      "loss": 6.0103,
      "mean_token_accuracy": 0.14784939587116241,
      "num_tokens": 18934026.0,
      "step": 9345
    },
    {
      "entropy": 6.272441053390503,
      "epoch": 0.8170934195578082,
      "grad_norm": 4.0,
      "learning_rate": 0.0004940102934750037,
      "loss": 6.0417,
      "mean_token_accuracy": 0.14669739156961442,
      "num_tokens": 18943262.0,
      "step": 9350
    },
    {
      "entropy": 6.313084936141967,
      "epoch": 0.8175303679105129,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004940031492154795,
      "loss": 6.0739,
      "mean_token_accuracy": 0.1375577449798584,
      "num_tokens": 18953082.0,
      "step": 9355
    },
    {
      "entropy": 6.272457933425903,
      "epoch": 0.8179673162632177,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004939960007553603,
      "loss": 6.1007,
      "mean_token_accuracy": 0.1415972113609314,
      "num_tokens": 18962408.0,
      "step": 9360
    },
    {
      "entropy": 6.322740650177002,
      "epoch": 0.8184042646159224,
      "grad_norm": 2.890625,
      "learning_rate": 0.000493988848094783,
      "loss": 6.1522,
      "mean_token_accuracy": 0.13074711784720422,
      "num_tokens": 18972918.0,
      "step": 9365
    },
    {
      "entropy": 6.3623453140258786,
      "epoch": 0.8188412129686271,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004939816912338848,
      "loss": 6.1139,
      "mean_token_accuracy": 0.13417126834392548,
      "num_tokens": 18983152.0,
      "step": 9370
    },
    {
      "entropy": 6.316799259185791,
      "epoch": 0.8192781613213318,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004939745301728031,
      "loss": 6.1076,
      "mean_token_accuracy": 0.13396622389554977,
      "num_tokens": 18993613.0,
      "step": 9375
    },
    {
      "entropy": 6.222745561599732,
      "epoch": 0.8197151096740365,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004939673649116752,
      "loss": 6.0453,
      "mean_token_accuracy": 0.1451654776930809,
      "num_tokens": 19004079.0,
      "step": 9380
    },
    {
      "entropy": 6.240976476669312,
      "epoch": 0.8201520580267413,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004939601954506383,
      "loss": 6.1819,
      "mean_token_accuracy": 0.13958189636468887,
      "num_tokens": 19013994.0,
      "step": 9385
    },
    {
      "entropy": 6.391075134277344,
      "epoch": 0.820589006379446,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004939530217898303,
      "loss": 6.1799,
      "mean_token_accuracy": 0.13269131407141685,
      "num_tokens": 19024674.0,
      "step": 9390
    },
    {
      "entropy": 6.347418975830078,
      "epoch": 0.8210259547321507,
      "grad_norm": 7.34375,
      "learning_rate": 0.0004939458439293886,
      "loss": 6.1374,
      "mean_token_accuracy": 0.1346823126077652,
      "num_tokens": 19035333.0,
      "step": 9395
    },
    {
      "entropy": 6.404700040817261,
      "epoch": 0.8214629030848554,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004939386618694508,
      "loss": 6.1754,
      "mean_token_accuracy": 0.13146359324455262,
      "num_tokens": 19046279.0,
      "step": 9400
    },
    {
      "entropy": 6.334047174453735,
      "epoch": 0.8218998514375601,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004939314756101547,
      "loss": 6.1274,
      "mean_token_accuracy": 0.13152439817786216,
      "num_tokens": 19056506.0,
      "step": 9405
    },
    {
      "entropy": 6.323962879180908,
      "epoch": 0.8223367997902647,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004939242851516384,
      "loss": 6.1197,
      "mean_token_accuracy": 0.13797605261206627,
      "num_tokens": 19067217.0,
      "step": 9410
    },
    {
      "entropy": 6.28490104675293,
      "epoch": 0.8227737481429696,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004939170904940395,
      "loss": 6.1885,
      "mean_token_accuracy": 0.13418862596154213,
      "num_tokens": 19079028.0,
      "step": 9415
    },
    {
      "entropy": 6.300374746322632,
      "epoch": 0.8232106964956742,
      "grad_norm": 3.0,
      "learning_rate": 0.000493909891637496,
      "loss": 6.1295,
      "mean_token_accuracy": 0.14405135363340377,
      "num_tokens": 19088236.0,
      "step": 9420
    },
    {
      "entropy": 6.301871204376221,
      "epoch": 0.8236476448483789,
      "grad_norm": 5.375,
      "learning_rate": 0.0004939026885821461,
      "loss": 6.1067,
      "mean_token_accuracy": 0.1437576323747635,
      "num_tokens": 19099598.0,
      "step": 9425
    },
    {
      "entropy": 6.305684185028076,
      "epoch": 0.8240845932010836,
      "grad_norm": 3.578125,
      "learning_rate": 0.000493895481328128,
      "loss": 6.1646,
      "mean_token_accuracy": 0.1411864496767521,
      "num_tokens": 19110347.0,
      "step": 9430
    },
    {
      "entropy": 6.3884978771209715,
      "epoch": 0.8245215415537883,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004938882698755798,
      "loss": 6.1218,
      "mean_token_accuracy": 0.13640627190470694,
      "num_tokens": 19120643.0,
      "step": 9435
    },
    {
      "entropy": 6.266851758956909,
      "epoch": 0.824958489906493,
      "grad_norm": 6.28125,
      "learning_rate": 0.00049388105422464,
      "loss": 6.098,
      "mean_token_accuracy": 0.13893204778432847,
      "num_tokens": 19130209.0,
      "step": 9440
    },
    {
      "entropy": 6.267712831497192,
      "epoch": 0.8253954382591978,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004938738343754467,
      "loss": 6.0495,
      "mean_token_accuracy": 0.14382429793477058,
      "num_tokens": 19139196.0,
      "step": 9445
    },
    {
      "entropy": 6.297996282577515,
      "epoch": 0.8258323866119025,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004938666103281389,
      "loss": 6.0796,
      "mean_token_accuracy": 0.14202444553375243,
      "num_tokens": 19149618.0,
      "step": 9450
    },
    {
      "entropy": 6.295094156265259,
      "epoch": 0.8262693349646072,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004938593820828545,
      "loss": 6.0824,
      "mean_token_accuracy": 0.13520338907837867,
      "num_tokens": 19159875.0,
      "step": 9455
    },
    {
      "entropy": 6.2894859790802,
      "epoch": 0.8267062833173119,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004938521496397326,
      "loss": 6.1067,
      "mean_token_accuracy": 0.13488565310835837,
      "num_tokens": 19170403.0,
      "step": 9460
    },
    {
      "entropy": 6.297162485122681,
      "epoch": 0.8271432316700166,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004938449129989117,
      "loss": 6.0912,
      "mean_token_accuracy": 0.14424026757478714,
      "num_tokens": 19181187.0,
      "step": 9465
    },
    {
      "entropy": 6.295498275756836,
      "epoch": 0.8275801800227213,
      "grad_norm": 3.5,
      "learning_rate": 0.0004938376721605308,
      "loss": 6.1288,
      "mean_token_accuracy": 0.1320436604321003,
      "num_tokens": 19191328.0,
      "step": 9470
    },
    {
      "entropy": 6.270652151107788,
      "epoch": 0.8280171283754261,
      "grad_norm": 6.875,
      "learning_rate": 0.0004938304271247286,
      "loss": 6.0833,
      "mean_token_accuracy": 0.1415044516324997,
      "num_tokens": 19202041.0,
      "step": 9475
    },
    {
      "entropy": 6.326199913024903,
      "epoch": 0.8284540767281308,
      "grad_norm": 25.125,
      "learning_rate": 0.0004938231778916442,
      "loss": 6.0803,
      "mean_token_accuracy": 0.13748242929577828,
      "num_tokens": 19212016.0,
      "step": 9480
    },
    {
      "entropy": 6.2696051597595215,
      "epoch": 0.8288910250808355,
      "grad_norm": 2.875,
      "learning_rate": 0.0004938159244614167,
      "loss": 6.089,
      "mean_token_accuracy": 0.1424597054719925,
      "num_tokens": 19222111.0,
      "step": 9485
    },
    {
      "entropy": 6.348085021972656,
      "epoch": 0.8293279734335401,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004938086668341849,
      "loss": 6.0969,
      "mean_token_accuracy": 0.14148656874895096,
      "num_tokens": 19231865.0,
      "step": 9490
    },
    {
      "entropy": 6.30455527305603,
      "epoch": 0.8297649217862448,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004938014050100884,
      "loss": 6.0753,
      "mean_token_accuracy": 0.142066989839077,
      "num_tokens": 19242993.0,
      "step": 9495
    },
    {
      "entropy": 6.306135511398315,
      "epoch": 0.8302018701389495,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004937941389892662,
      "loss": 6.1146,
      "mean_token_accuracy": 0.13835363686084748,
      "num_tokens": 19252776.0,
      "step": 9500
    },
    {
      "entropy": 6.273995351791382,
      "epoch": 0.8306388184916543,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004937868687718577,
      "loss": 6.0627,
      "mean_token_accuracy": 0.14160662591457368,
      "num_tokens": 19263022.0,
      "step": 9505
    },
    {
      "entropy": 6.304497241973877,
      "epoch": 0.831075766844359,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004937795943580026,
      "loss": 6.1185,
      "mean_token_accuracy": 0.14135142266750336,
      "num_tokens": 19271859.0,
      "step": 9510
    },
    {
      "entropy": 6.313362121582031,
      "epoch": 0.8315127151970637,
      "grad_norm": 3.703125,
      "learning_rate": 0.00049377231574784,
      "loss": 6.07,
      "mean_token_accuracy": 0.14312165528535842,
      "num_tokens": 19282771.0,
      "step": 9515
    },
    {
      "entropy": 6.341313982009888,
      "epoch": 0.8319496635497684,
      "grad_norm": 7.25,
      "learning_rate": 0.0004937650329415099,
      "loss": 6.1075,
      "mean_token_accuracy": 0.1366374149918556,
      "num_tokens": 19292985.0,
      "step": 9520
    },
    {
      "entropy": 6.274918031692505,
      "epoch": 0.8323866119024731,
      "grad_norm": 6.6875,
      "learning_rate": 0.0004937577459391519,
      "loss": 6.1169,
      "mean_token_accuracy": 0.1395903632044792,
      "num_tokens": 19303409.0,
      "step": 9525
    },
    {
      "entropy": 6.268351888656616,
      "epoch": 0.8328235602551778,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004937504547409056,
      "loss": 6.1088,
      "mean_token_accuracy": 0.13732085153460502,
      "num_tokens": 19312632.0,
      "step": 9530
    },
    {
      "entropy": 6.296071577072143,
      "epoch": 0.8332605086078826,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004937431593469109,
      "loss": 5.9949,
      "mean_token_accuracy": 0.14204887300729752,
      "num_tokens": 19322759.0,
      "step": 9535
    },
    {
      "entropy": 6.360665321350098,
      "epoch": 0.8336974569605873,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004937358597573079,
      "loss": 6.0404,
      "mean_token_accuracy": 0.1453598991036415,
      "num_tokens": 19333307.0,
      "step": 9540
    },
    {
      "entropy": 6.27874345779419,
      "epoch": 0.834134405313292,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004937285559722364,
      "loss": 6.1347,
      "mean_token_accuracy": 0.1412837713956833,
      "num_tokens": 19342528.0,
      "step": 9545
    },
    {
      "entropy": 6.307973384857178,
      "epoch": 0.8345713536659967,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004937212479918367,
      "loss": 6.0961,
      "mean_token_accuracy": 0.13428374230861664,
      "num_tokens": 19352640.0,
      "step": 9550
    },
    {
      "entropy": 6.32742166519165,
      "epoch": 0.8350083020187014,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004937139358162488,
      "loss": 6.0991,
      "mean_token_accuracy": 0.14064025580883027,
      "num_tokens": 19363285.0,
      "step": 9555
    },
    {
      "entropy": 6.371377468109131,
      "epoch": 0.835445250371406,
      "grad_norm": 6.1875,
      "learning_rate": 0.000493706619445613,
      "loss": 6.0973,
      "mean_token_accuracy": 0.14355772584676743,
      "num_tokens": 19373141.0,
      "step": 9560
    },
    {
      "entropy": 6.273716020584106,
      "epoch": 0.8358821987241108,
      "grad_norm": 3.625,
      "learning_rate": 0.0004936992988800697,
      "loss": 6.1365,
      "mean_token_accuracy": 0.14243221282958984,
      "num_tokens": 19383091.0,
      "step": 9565
    },
    {
      "entropy": 6.301488447189331,
      "epoch": 0.8363191470768155,
      "grad_norm": 4.375,
      "learning_rate": 0.0004936919741197591,
      "loss": 6.0643,
      "mean_token_accuracy": 0.14135712757706642,
      "num_tokens": 19393163.0,
      "step": 9570
    },
    {
      "entropy": 6.300067615509033,
      "epoch": 0.8367560954295202,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004936846451648219,
      "loss": 6.0408,
      "mean_token_accuracy": 0.14510954022407532,
      "num_tokens": 19403475.0,
      "step": 9575
    },
    {
      "entropy": 6.268555402755737,
      "epoch": 0.8371930437822249,
      "grad_norm": 7.40625,
      "learning_rate": 0.0004936773120153987,
      "loss": 6.0492,
      "mean_token_accuracy": 0.14586118161678313,
      "num_tokens": 19413302.0,
      "step": 9580
    },
    {
      "entropy": 6.324727725982666,
      "epoch": 0.8376299921349296,
      "grad_norm": 4.78125,
      "learning_rate": 0.00049366997467163,
      "loss": 6.0513,
      "mean_token_accuracy": 0.14475452303886413,
      "num_tokens": 19423473.0,
      "step": 9585
    },
    {
      "entropy": 6.323733758926392,
      "epoch": 0.8380669404876344,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004936626331336567,
      "loss": 6.1571,
      "mean_token_accuracy": 0.13587832003831862,
      "num_tokens": 19433289.0,
      "step": 9590
    },
    {
      "entropy": 6.319749736785889,
      "epoch": 0.8385038888403391,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004936552874016194,
      "loss": 6.1625,
      "mean_token_accuracy": 0.13446955978870392,
      "num_tokens": 19442188.0,
      "step": 9595
    },
    {
      "entropy": 6.2969639778137205,
      "epoch": 0.8389408371930438,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004936479374756592,
      "loss": 6.1421,
      "mean_token_accuracy": 0.13432279229164124,
      "num_tokens": 19452971.0,
      "step": 9600
    },
    {
      "entropy": 6.380914258956909,
      "epoch": 0.8393777855457485,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004936405833559169,
      "loss": 6.1469,
      "mean_token_accuracy": 0.13013540729880332,
      "num_tokens": 19463263.0,
      "step": 9605
    },
    {
      "entropy": 6.397865962982178,
      "epoch": 0.8398147338984532,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004936332250425337,
      "loss": 6.1079,
      "mean_token_accuracy": 0.13533571138978004,
      "num_tokens": 19473234.0,
      "step": 9610
    },
    {
      "entropy": 6.3180419445037845,
      "epoch": 0.8402516822511579,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004936258625356507,
      "loss": 6.1125,
      "mean_token_accuracy": 0.13625560626387595,
      "num_tokens": 19483714.0,
      "step": 9615
    },
    {
      "entropy": 6.233173131942749,
      "epoch": 0.8406886306038627,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004936184958354091,
      "loss": 6.0278,
      "mean_token_accuracy": 0.1502446174621582,
      "num_tokens": 19493128.0,
      "step": 9620
    },
    {
      "entropy": 6.36079969406128,
      "epoch": 0.8411255789565674,
      "grad_norm": 5.0,
      "learning_rate": 0.00049361112494195,
      "loss": 6.1451,
      "mean_token_accuracy": 0.13676347061991692,
      "num_tokens": 19503401.0,
      "step": 9625
    },
    {
      "entropy": 6.320564079284668,
      "epoch": 0.8415625273092721,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004936037498554153,
      "loss": 6.186,
      "mean_token_accuracy": 0.1418322131037712,
      "num_tokens": 19513050.0,
      "step": 9630
    },
    {
      "entropy": 6.27621545791626,
      "epoch": 0.8419994756619767,
      "grad_norm": 9.875,
      "learning_rate": 0.000493596370575946,
      "loss": 6.0756,
      "mean_token_accuracy": 0.14127040207386016,
      "num_tokens": 19524534.0,
      "step": 9635
    },
    {
      "entropy": 6.302616357803345,
      "epoch": 0.8424364240146814,
      "grad_norm": 3.125,
      "learning_rate": 0.0004935889871036837,
      "loss": 6.0589,
      "mean_token_accuracy": 0.14106576889753342,
      "num_tokens": 19534863.0,
      "step": 9640
    },
    {
      "entropy": 6.3598301887512205,
      "epoch": 0.8428733723673861,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004935815994387701,
      "loss": 6.1408,
      "mean_token_accuracy": 0.13566820770502092,
      "num_tokens": 19545037.0,
      "step": 9645
    },
    {
      "entropy": 6.284095525741577,
      "epoch": 0.8433103207200909,
      "grad_norm": 3.5,
      "learning_rate": 0.000493574207581347,
      "loss": 6.1192,
      "mean_token_accuracy": 0.12875717654824256,
      "num_tokens": 19555377.0,
      "step": 9650
    },
    {
      "entropy": 6.305630159378052,
      "epoch": 0.8437472690727956,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004935668115315559,
      "loss": 6.1259,
      "mean_token_accuracy": 0.1410840019583702,
      "num_tokens": 19565874.0,
      "step": 9655
    },
    {
      "entropy": 6.230661106109619,
      "epoch": 0.8441842174255003,
      "grad_norm": 2.671875,
      "learning_rate": 0.000493559411289539,
      "loss": 5.965,
      "mean_token_accuracy": 0.15507861822843552,
      "num_tokens": 19575002.0,
      "step": 9660
    },
    {
      "entropy": 6.2578362941741945,
      "epoch": 0.844621165778205,
      "grad_norm": 2.875,
      "learning_rate": 0.000493552006855438,
      "loss": 6.0928,
      "mean_token_accuracy": 0.14111133813858032,
      "num_tokens": 19585863.0,
      "step": 9665
    },
    {
      "entropy": 6.301001787185669,
      "epoch": 0.8450581141309097,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004935445982293949,
      "loss": 6.0434,
      "mean_token_accuracy": 0.14006902128458024,
      "num_tokens": 19596787.0,
      "step": 9670
    },
    {
      "entropy": 6.314519548416138,
      "epoch": 0.8454950624836144,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004935371854115519,
      "loss": 6.1413,
      "mean_token_accuracy": 0.13553963527083396,
      "num_tokens": 19607114.0,
      "step": 9675
    },
    {
      "entropy": 6.279674005508423,
      "epoch": 0.8459320108363192,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004935297684020513,
      "loss": 6.034,
      "mean_token_accuracy": 0.14807514995336532,
      "num_tokens": 19617334.0,
      "step": 9680
    },
    {
      "entropy": 6.323421478271484,
      "epoch": 0.8463689591890239,
      "grad_norm": 2.625,
      "learning_rate": 0.0004935223472010352,
      "loss": 6.1223,
      "mean_token_accuracy": 0.13846837654709815,
      "num_tokens": 19626844.0,
      "step": 9685
    },
    {
      "entropy": 6.232555246353149,
      "epoch": 0.8468059075417286,
      "grad_norm": 6.0,
      "learning_rate": 0.0004935149218086459,
      "loss": 6.1153,
      "mean_token_accuracy": 0.14279309585690497,
      "num_tokens": 19637897.0,
      "step": 9690
    },
    {
      "entropy": 6.265404844284058,
      "epoch": 0.8472428558944333,
      "grad_norm": 3.859375,
      "learning_rate": 0.000493507492225026,
      "loss": 6.1935,
      "mean_token_accuracy": 0.1338166780769825,
      "num_tokens": 19649161.0,
      "step": 9695
    },
    {
      "entropy": 6.344244861602784,
      "epoch": 0.847679804247138,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004935000584503178,
      "loss": 6.0973,
      "mean_token_accuracy": 0.14095838516950607,
      "num_tokens": 19659328.0,
      "step": 9700
    },
    {
      "entropy": 6.3858569145202635,
      "epoch": 0.8481167525998426,
      "grad_norm": 2.875,
      "learning_rate": 0.0004934926204846641,
      "loss": 6.194,
      "mean_token_accuracy": 0.13516309708356858,
      "num_tokens": 19669196.0,
      "step": 9705
    },
    {
      "entropy": 6.302838373184204,
      "epoch": 0.8485537009525475,
      "grad_norm": 2.25,
      "learning_rate": 0.0004934851783282073,
      "loss": 6.0512,
      "mean_token_accuracy": 0.13648490384221076,
      "num_tokens": 19680019.0,
      "step": 9710
    },
    {
      "entropy": 6.3036750793457035,
      "epoch": 0.8489906493052521,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004934777319810904,
      "loss": 6.155,
      "mean_token_accuracy": 0.1395260751247406,
      "num_tokens": 19689391.0,
      "step": 9715
    },
    {
      "entropy": 6.345256805419922,
      "epoch": 0.8494275976579568,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004934702814434561,
      "loss": 6.15,
      "mean_token_accuracy": 0.13347968012094497,
      "num_tokens": 19699683.0,
      "step": 9720
    },
    {
      "entropy": 6.342925643920898,
      "epoch": 0.8498645460106615,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004934628267154474,
      "loss": 6.2154,
      "mean_token_accuracy": 0.13180404156446457,
      "num_tokens": 19709844.0,
      "step": 9725
    },
    {
      "entropy": 6.361215353012085,
      "epoch": 0.8503014943633662,
      "grad_norm": 3.15625,
      "learning_rate": 0.000493455367797207,
      "loss": 6.1616,
      "mean_token_accuracy": 0.1310235984623432,
      "num_tokens": 19719926.0,
      "step": 9730
    },
    {
      "entropy": 6.313539028167725,
      "epoch": 0.8507384427160709,
      "grad_norm": 4.375,
      "learning_rate": 0.0004934479046888784,
      "loss": 6.0879,
      "mean_token_accuracy": 0.13864776939153672,
      "num_tokens": 19730548.0,
      "step": 9735
    },
    {
      "entropy": 6.284141159057617,
      "epoch": 0.8511753910687757,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004934404373906045,
      "loss": 6.147,
      "mean_token_accuracy": 0.14178260564804077,
      "num_tokens": 19740471.0,
      "step": 9740
    },
    {
      "entropy": 6.3286220073699955,
      "epoch": 0.8516123394214804,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004934329659025283,
      "loss": 6.0518,
      "mean_token_accuracy": 0.14398444071412086,
      "num_tokens": 19749395.0,
      "step": 9745
    },
    {
      "entropy": 6.318807744979859,
      "epoch": 0.8520492877741851,
      "grad_norm": 2.625,
      "learning_rate": 0.0004934254902247936,
      "loss": 6.1514,
      "mean_token_accuracy": 0.12988874614238738,
      "num_tokens": 19759203.0,
      "step": 9750
    },
    {
      "entropy": 6.328744316101075,
      "epoch": 0.8524862361268898,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004934180103575436,
      "loss": 6.0371,
      "mean_token_accuracy": 0.15000907182693482,
      "num_tokens": 19768977.0,
      "step": 9755
    },
    {
      "entropy": 6.301958274841309,
      "epoch": 0.8529231844795945,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004934105263009216,
      "loss": 6.0253,
      "mean_token_accuracy": 0.13983262777328492,
      "num_tokens": 19778569.0,
      "step": 9760
    },
    {
      "entropy": 6.268443489074707,
      "epoch": 0.8533601328322992,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004934030380550713,
      "loss": 6.1194,
      "mean_token_accuracy": 0.14030743166804313,
      "num_tokens": 19787939.0,
      "step": 9765
    },
    {
      "entropy": 6.324241256713867,
      "epoch": 0.853797081185004,
      "grad_norm": 3.0,
      "learning_rate": 0.0004933955456201362,
      "loss": 6.1122,
      "mean_token_accuracy": 0.13815711662173272,
      "num_tokens": 19798407.0,
      "step": 9770
    },
    {
      "entropy": 6.356347560882568,
      "epoch": 0.8542340295377087,
      "grad_norm": 3.625,
      "learning_rate": 0.0004933880489962602,
      "loss": 6.1319,
      "mean_token_accuracy": 0.143503650277853,
      "num_tokens": 19808495.0,
      "step": 9775
    },
    {
      "entropy": 6.261553382873535,
      "epoch": 0.8546709778904134,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004933805481835868,
      "loss": 6.0821,
      "mean_token_accuracy": 0.13822968304157257,
      "num_tokens": 19818709.0,
      "step": 9780
    },
    {
      "entropy": 6.3347124576568605,
      "epoch": 0.855107926243118,
      "grad_norm": 10.4375,
      "learning_rate": 0.0004933730431822603,
      "loss": 6.1439,
      "mean_token_accuracy": 0.14029505774378775,
      "num_tokens": 19828865.0,
      "step": 9785
    },
    {
      "entropy": 6.34427170753479,
      "epoch": 0.8555448745958227,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004933655339924243,
      "loss": 6.0448,
      "mean_token_accuracy": 0.138470046967268,
      "num_tokens": 19838757.0,
      "step": 9790
    },
    {
      "entropy": 6.243005180358887,
      "epoch": 0.8559818229485275,
      "grad_norm": 3.15625,
      "learning_rate": 0.000493358020614223,
      "loss": 6.0983,
      "mean_token_accuracy": 0.14339999556541444,
      "num_tokens": 19848967.0,
      "step": 9795
    },
    {
      "entropy": 6.302224016189575,
      "epoch": 0.8564187713012322,
      "grad_norm": 3.5,
      "learning_rate": 0.0004933505030478003,
      "loss": 6.1189,
      "mean_token_accuracy": 0.14107196927070617,
      "num_tokens": 19859487.0,
      "step": 9800
    },
    {
      "entropy": 6.411745691299439,
      "epoch": 0.8568557196539369,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004933429812933005,
      "loss": 6.1826,
      "mean_token_accuracy": 0.13540088906884193,
      "num_tokens": 19870170.0,
      "step": 9805
    },
    {
      "entropy": 6.327482128143311,
      "epoch": 0.8572926680066416,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004933354553508681,
      "loss": 6.0698,
      "mean_token_accuracy": 0.1402559757232666,
      "num_tokens": 19881031.0,
      "step": 9810
    },
    {
      "entropy": 6.370117712020874,
      "epoch": 0.8577296163593463,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004933279252206472,
      "loss": 6.1194,
      "mean_token_accuracy": 0.13485998436808586,
      "num_tokens": 19891082.0,
      "step": 9815
    },
    {
      "entropy": 6.3631432056427,
      "epoch": 0.858166564712051,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004933203909027823,
      "loss": 6.1159,
      "mean_token_accuracy": 0.13487059101462365,
      "num_tokens": 19900626.0,
      "step": 9820
    },
    {
      "entropy": 6.2999410152435305,
      "epoch": 0.8586035130647558,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004933128523974179,
      "loss": 6.0992,
      "mean_token_accuracy": 0.14569904431700706,
      "num_tokens": 19910660.0,
      "step": 9825
    },
    {
      "entropy": 6.302130174636841,
      "epoch": 0.8590404614174605,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004933053097046986,
      "loss": 6.0865,
      "mean_token_accuracy": 0.13779011592268944,
      "num_tokens": 19920476.0,
      "step": 9830
    },
    {
      "entropy": 6.423284816741943,
      "epoch": 0.8594774097701652,
      "grad_norm": 3.296875,
      "learning_rate": 0.000493297762824769,
      "loss": 6.196,
      "mean_token_accuracy": 0.12945295348763466,
      "num_tokens": 19931151.0,
      "step": 9835
    },
    {
      "entropy": 6.360500049591065,
      "epoch": 0.8599143581228699,
      "grad_norm": 3.296875,
      "learning_rate": 0.000493290211757774,
      "loss": 6.0928,
      "mean_token_accuracy": 0.13197456449270248,
      "num_tokens": 19942469.0,
      "step": 9840
    },
    {
      "entropy": 6.296321010589599,
      "epoch": 0.8603513064755746,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004932826565038584,
      "loss": 6.1329,
      "mean_token_accuracy": 0.1406553119421005,
      "num_tokens": 19953055.0,
      "step": 9845
    },
    {
      "entropy": 6.207673788070679,
      "epoch": 0.8607882548282793,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004932750970631669,
      "loss": 6.1178,
      "mean_token_accuracy": 0.13957864940166473,
      "num_tokens": 19963944.0,
      "step": 9850
    },
    {
      "entropy": 6.284408521652222,
      "epoch": 0.8612252031809841,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004932675334358448,
      "loss": 6.0237,
      "mean_token_accuracy": 0.143646302819252,
      "num_tokens": 19973913.0,
      "step": 9855
    },
    {
      "entropy": 6.301597929000854,
      "epoch": 0.8616621515336887,
      "grad_norm": 2.5,
      "learning_rate": 0.000493259965622037,
      "loss": 6.1027,
      "mean_token_accuracy": 0.1427308902144432,
      "num_tokens": 19984130.0,
      "step": 9860
    },
    {
      "entropy": 6.293789625167847,
      "epoch": 0.8620990998863934,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004932523936218887,
      "loss": 6.0197,
      "mean_token_accuracy": 0.14451179653406143,
      "num_tokens": 19993840.0,
      "step": 9865
    },
    {
      "entropy": 6.266719150543213,
      "epoch": 0.8625360482390981,
      "grad_norm": 3.0,
      "learning_rate": 0.000493244817435545,
      "loss": 6.123,
      "mean_token_accuracy": 0.13436217829585076,
      "num_tokens": 20004228.0,
      "step": 9870
    },
    {
      "entropy": 6.279605865478516,
      "epoch": 0.8629729965918028,
      "grad_norm": 2.875,
      "learning_rate": 0.0004932372370631515,
      "loss": 6.1029,
      "mean_token_accuracy": 0.14053030908107758,
      "num_tokens": 20012620.0,
      "step": 9875
    },
    {
      "entropy": 6.317138433456421,
      "epoch": 0.8634099449445075,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004932296525048535,
      "loss": 6.1793,
      "mean_token_accuracy": 0.13904112949967384,
      "num_tokens": 20022892.0,
      "step": 9880
    },
    {
      "entropy": 6.357314920425415,
      "epoch": 0.8638468932972123,
      "grad_norm": 4.9375,
      "learning_rate": 0.0004932220637607962,
      "loss": 6.0606,
      "mean_token_accuracy": 0.1402808427810669,
      "num_tokens": 20032788.0,
      "step": 9885
    },
    {
      "entropy": 6.356874799728393,
      "epoch": 0.864283841649917,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004932144708311256,
      "loss": 6.1024,
      "mean_token_accuracy": 0.13635034263134002,
      "num_tokens": 20043176.0,
      "step": 9890
    },
    {
      "entropy": 6.2990387916564945,
      "epoch": 0.8647207900026217,
      "grad_norm": 3.703125,
      "learning_rate": 0.000493206873715987,
      "loss": 6.1049,
      "mean_token_accuracy": 0.14306221008300782,
      "num_tokens": 20052298.0,
      "step": 9895
    },
    {
      "entropy": 6.271191215515136,
      "epoch": 0.8651577383553264,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004931992724155262,
      "loss": 6.0957,
      "mean_token_accuracy": 0.13465766236186028,
      "num_tokens": 20063163.0,
      "step": 9900
    },
    {
      "entropy": 6.348688316345215,
      "epoch": 0.8655946867080311,
      "grad_norm": 3.53125,
      "learning_rate": 0.000493191666929889,
      "loss": 6.1297,
      "mean_token_accuracy": 0.13673932179808618,
      "num_tokens": 20073644.0,
      "step": 9905
    },
    {
      "entropy": 6.362282562255859,
      "epoch": 0.8660316350607358,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004931840572592214,
      "loss": 6.0813,
      "mean_token_accuracy": 0.14217628836631774,
      "num_tokens": 20084457.0,
      "step": 9910
    },
    {
      "entropy": 6.237162303924561,
      "epoch": 0.8664685834134406,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004931764434036692,
      "loss": 6.0056,
      "mean_token_accuracy": 0.1441192664206028,
      "num_tokens": 20094153.0,
      "step": 9915
    },
    {
      "entropy": 6.242369174957275,
      "epoch": 0.8669055317661453,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004931688253633784,
      "loss": 6.185,
      "mean_token_accuracy": 0.13711377531290053,
      "num_tokens": 20104031.0,
      "step": 9920
    },
    {
      "entropy": 6.345116853713989,
      "epoch": 0.86734248011885,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004931612031384953,
      "loss": 6.1082,
      "mean_token_accuracy": 0.14258310720324516,
      "num_tokens": 20114431.0,
      "step": 9925
    },
    {
      "entropy": 6.282162666320801,
      "epoch": 0.8677794284715546,
      "grad_norm": 2.234375,
      "learning_rate": 0.000493153576729166,
      "loss": 6.0028,
      "mean_token_accuracy": 0.14464861676096916,
      "num_tokens": 20124161.0,
      "step": 9930
    },
    {
      "entropy": 6.291960573196411,
      "epoch": 0.8682163768242593,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004931459461355368,
      "loss": 6.0623,
      "mean_token_accuracy": 0.14231914654374123,
      "num_tokens": 20134227.0,
      "step": 9935
    },
    {
      "entropy": 6.334458684921264,
      "epoch": 0.868653325176964,
      "grad_norm": 3.625,
      "learning_rate": 0.0004931383113577541,
      "loss": 6.064,
      "mean_token_accuracy": 0.13991656079888343,
      "num_tokens": 20143912.0,
      "step": 9940
    },
    {
      "entropy": 6.317588996887207,
      "epoch": 0.8690902735296688,
      "grad_norm": 6.71875,
      "learning_rate": 0.0004931306723959641,
      "loss": 5.9923,
      "mean_token_accuracy": 0.14443909451365472,
      "num_tokens": 20153858.0,
      "step": 9945
    },
    {
      "entropy": 6.373877429962159,
      "epoch": 0.8695272218823735,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004931230292503137,
      "loss": 6.154,
      "mean_token_accuracy": 0.13323474749922753,
      "num_tokens": 20164309.0,
      "step": 9950
    },
    {
      "entropy": 6.278200578689575,
      "epoch": 0.8699641702350782,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004931153819209492,
      "loss": 6.0904,
      "mean_token_accuracy": 0.14033587723970414,
      "num_tokens": 20173949.0,
      "step": 9955
    },
    {
      "entropy": 6.3203877925872805,
      "epoch": 0.8704011185877829,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004931077304080175,
      "loss": 6.0877,
      "mean_token_accuracy": 0.14140881821513177,
      "num_tokens": 20184206.0,
      "step": 9960
    },
    {
      "entropy": 6.4086357116699215,
      "epoch": 0.8708380669404876,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004931000747116651,
      "loss": 6.2161,
      "mean_token_accuracy": 0.13129571452736855,
      "num_tokens": 20194430.0,
      "step": 9965
    },
    {
      "entropy": 6.414946460723877,
      "epoch": 0.8712750152931923,
      "grad_norm": 2.328125,
      "learning_rate": 0.000493092414832039,
      "loss": 6.0873,
      "mean_token_accuracy": 0.14281998723745346,
      "num_tokens": 20205004.0,
      "step": 9970
    },
    {
      "entropy": 6.317124557495117,
      "epoch": 0.8717119636458971,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004930847507692862,
      "loss": 6.0374,
      "mean_token_accuracy": 0.14227123335003852,
      "num_tokens": 20214698.0,
      "step": 9975
    },
    {
      "entropy": 6.279891586303711,
      "epoch": 0.8721489119986018,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004930770825235536,
      "loss": 6.1266,
      "mean_token_accuracy": 0.1352207213640213,
      "num_tokens": 20226534.0,
      "step": 9980
    },
    {
      "entropy": 6.357548475265503,
      "epoch": 0.8725858603513065,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004930694100949882,
      "loss": 6.1184,
      "mean_token_accuracy": 0.14219956696033478,
      "num_tokens": 20237414.0,
      "step": 9985
    },
    {
      "entropy": 6.333403825759888,
      "epoch": 0.8730228087040112,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004930617334837373,
      "loss": 5.9847,
      "mean_token_accuracy": 0.14630696922540665,
      "num_tokens": 20246948.0,
      "step": 9990
    },
    {
      "entropy": 6.325153827667236,
      "epoch": 0.8734597570567159,
      "grad_norm": 3.40625,
      "learning_rate": 0.000493054052689948,
      "loss": 6.1663,
      "mean_token_accuracy": 0.139209770411253,
      "num_tokens": 20257281.0,
      "step": 9995
    },
    {
      "entropy": 6.325224685668945,
      "epoch": 0.8738967054094207,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004930463677137679,
      "loss": 6.1342,
      "mean_token_accuracy": 0.13920582458376884,
      "num_tokens": 20269041.0,
      "step": 10000
    },
    {
      "entropy": 6.31692476272583,
      "epoch": 0.8743336537621254,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004930386785553441,
      "loss": 6.1732,
      "mean_token_accuracy": 0.13193241953849794,
      "num_tokens": 20279688.0,
      "step": 10005
    },
    {
      "entropy": 6.315338134765625,
      "epoch": 0.87477060211483,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004930309852148242,
      "loss": 6.071,
      "mean_token_accuracy": 0.143155225366354,
      "num_tokens": 20288913.0,
      "step": 10010
    },
    {
      "entropy": 6.342848062515259,
      "epoch": 0.8752075504675347,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004930232876923558,
      "loss": 6.1255,
      "mean_token_accuracy": 0.1398486912250519,
      "num_tokens": 20299009.0,
      "step": 10015
    },
    {
      "entropy": 6.342072010040283,
      "epoch": 0.8756444988202394,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004930155859880865,
      "loss": 6.1128,
      "mean_token_accuracy": 0.13403648734092713,
      "num_tokens": 20309478.0,
      "step": 10020
    },
    {
      "entropy": 6.324098825454712,
      "epoch": 0.8760814471729441,
      "grad_norm": 2.5,
      "learning_rate": 0.000493007880102164,
      "loss": 6.1403,
      "mean_token_accuracy": 0.14031611755490303,
      "num_tokens": 20319451.0,
      "step": 10025
    },
    {
      "entropy": 6.3427809715271,
      "epoch": 0.8765183955256489,
      "grad_norm": 5.15625,
      "learning_rate": 0.000493000170034736,
      "loss": 6.1007,
      "mean_token_accuracy": 0.14017148017883302,
      "num_tokens": 20328497.0,
      "step": 10030
    },
    {
      "entropy": 6.209623050689697,
      "epoch": 0.8769553438783536,
      "grad_norm": 6.625,
      "learning_rate": 0.0004929924557859507,
      "loss": 5.9686,
      "mean_token_accuracy": 0.15394297242164612,
      "num_tokens": 20338584.0,
      "step": 10035
    },
    {
      "entropy": 6.2693451881408695,
      "epoch": 0.8773922922310583,
      "grad_norm": 3.0,
      "learning_rate": 0.0004929847373559557,
      "loss": 6.0774,
      "mean_token_accuracy": 0.14089732617139816,
      "num_tokens": 20348808.0,
      "step": 10040
    },
    {
      "entropy": 6.2784830093383786,
      "epoch": 0.877829240583763,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004929770147448992,
      "loss": 6.0575,
      "mean_token_accuracy": 0.14349573254585266,
      "num_tokens": 20358301.0,
      "step": 10045
    },
    {
      "entropy": 6.287945747375488,
      "epoch": 0.8782661889364677,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004929692879529295,
      "loss": 6.059,
      "mean_token_accuracy": 0.14038483202457427,
      "num_tokens": 20369032.0,
      "step": 10050
    },
    {
      "entropy": 6.356151103973389,
      "epoch": 0.8787031372891724,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004929615569801946,
      "loss": 6.1092,
      "mean_token_accuracy": 0.131279506534338,
      "num_tokens": 20379101.0,
      "step": 10055
    },
    {
      "entropy": 6.2750293731689455,
      "epoch": 0.8791400856418772,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004929538218268428,
      "loss": 6.0417,
      "mean_token_accuracy": 0.14429329782724382,
      "num_tokens": 20388864.0,
      "step": 10060
    },
    {
      "entropy": 6.36595139503479,
      "epoch": 0.8795770339945819,
      "grad_norm": 6.53125,
      "learning_rate": 0.0004929460824930226,
      "loss": 6.2001,
      "mean_token_accuracy": 0.13594384714961052,
      "num_tokens": 20398402.0,
      "step": 10065
    },
    {
      "entropy": 6.295538330078125,
      "epoch": 0.8800139823472866,
      "grad_norm": 8.1875,
      "learning_rate": 0.0004929383389788821,
      "loss": 6.0823,
      "mean_token_accuracy": 0.14368000254034996,
      "num_tokens": 20408209.0,
      "step": 10070
    },
    {
      "entropy": 6.30260272026062,
      "epoch": 0.8804509306999913,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004929305912845703,
      "loss": 6.1142,
      "mean_token_accuracy": 0.13810660243034362,
      "num_tokens": 20418117.0,
      "step": 10075
    },
    {
      "entropy": 6.341778755187988,
      "epoch": 0.8808878790526959,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004929228394102355,
      "loss": 6.1374,
      "mean_token_accuracy": 0.13197655901312827,
      "num_tokens": 20428347.0,
      "step": 10080
    },
    {
      "entropy": 6.37437481880188,
      "epoch": 0.8813248274054006,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004929150833560263,
      "loss": 6.0798,
      "mean_token_accuracy": 0.1326317861676216,
      "num_tokens": 20437804.0,
      "step": 10085
    },
    {
      "entropy": 6.333003377914428,
      "epoch": 0.8817617757581054,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004929073231220917,
      "loss": 6.1193,
      "mean_token_accuracy": 0.1354121319949627,
      "num_tokens": 20447600.0,
      "step": 10090
    },
    {
      "entropy": 6.325081634521484,
      "epoch": 0.8821987241108101,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004928995587085806,
      "loss": 6.1407,
      "mean_token_accuracy": 0.13665893375873567,
      "num_tokens": 20458023.0,
      "step": 10095
    },
    {
      "entropy": 6.289272117614746,
      "epoch": 0.8826356724635148,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004928917901156416,
      "loss": 6.0633,
      "mean_token_accuracy": 0.15005782544612883,
      "num_tokens": 20468195.0,
      "step": 10100
    },
    {
      "entropy": 6.256479024887085,
      "epoch": 0.8830726208162195,
      "grad_norm": 3.5625,
      "learning_rate": 0.000492884017343424,
      "loss": 6.1912,
      "mean_token_accuracy": 0.13451133593916892,
      "num_tokens": 20479250.0,
      "step": 10105
    },
    {
      "entropy": 6.377841329574585,
      "epoch": 0.8835095691689242,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004928762403920767,
      "loss": 6.089,
      "mean_token_accuracy": 0.13967251107096673,
      "num_tokens": 20489587.0,
      "step": 10110
    },
    {
      "entropy": 6.338430309295655,
      "epoch": 0.8839465175216289,
      "grad_norm": 3.4375,
      "learning_rate": 0.000492868459261749,
      "loss": 6.105,
      "mean_token_accuracy": 0.14157025814056395,
      "num_tokens": 20499788.0,
      "step": 10115
    },
    {
      "entropy": 6.265410470962524,
      "epoch": 0.8843834658743337,
      "grad_norm": 4.9375,
      "learning_rate": 0.00049286067395259,
      "loss": 6.1095,
      "mean_token_accuracy": 0.14112475886940956,
      "num_tokens": 20509716.0,
      "step": 10120
    },
    {
      "entropy": 6.274189281463623,
      "epoch": 0.8848204142270384,
      "grad_norm": 7.84375,
      "learning_rate": 0.0004928528844647493,
      "loss": 6.0708,
      "mean_token_accuracy": 0.14273151904344558,
      "num_tokens": 20519020.0,
      "step": 10125
    },
    {
      "entropy": 6.295938444137573,
      "epoch": 0.8852573625797431,
      "grad_norm": 2.625,
      "learning_rate": 0.000492845090798376,
      "loss": 6.1068,
      "mean_token_accuracy": 0.1379813827574253,
      "num_tokens": 20528997.0,
      "step": 10130
    },
    {
      "entropy": 6.351982307434082,
      "epoch": 0.8856943109324478,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004928372929536197,
      "loss": 6.105,
      "mean_token_accuracy": 0.14334506988525392,
      "num_tokens": 20538599.0,
      "step": 10135
    },
    {
      "entropy": 6.3612597465515135,
      "epoch": 0.8861312592851525,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004928294909306301,
      "loss": 6.0386,
      "mean_token_accuracy": 0.1422891288995743,
      "num_tokens": 20547664.0,
      "step": 10140
    },
    {
      "entropy": 6.298583555221557,
      "epoch": 0.8865682076378572,
      "grad_norm": 25.75,
      "learning_rate": 0.0004928216847295565,
      "loss": 6.1693,
      "mean_token_accuracy": 0.13632721528410913,
      "num_tokens": 20557860.0,
      "step": 10145
    },
    {
      "entropy": 6.263783311843872,
      "epoch": 0.887005155990562,
      "grad_norm": 2.703125,
      "learning_rate": 0.000492813874350549,
      "loss": 6.1103,
      "mean_token_accuracy": 0.13911132216453553,
      "num_tokens": 20569003.0,
      "step": 10150
    },
    {
      "entropy": 6.330843448638916,
      "epoch": 0.8874421043432666,
      "grad_norm": 2.625,
      "learning_rate": 0.0004928060597937573,
      "loss": 6.0847,
      "mean_token_accuracy": 0.1446116164326668,
      "num_tokens": 20578402.0,
      "step": 10155
    },
    {
      "entropy": 6.357872486114502,
      "epoch": 0.8878790526959713,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004927982410593312,
      "loss": 6.027,
      "mean_token_accuracy": 0.141134175658226,
      "num_tokens": 20587874.0,
      "step": 10160
    },
    {
      "entropy": 6.328481674194336,
      "epoch": 0.888316001048676,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004927904181474207,
      "loss": 6.0413,
      "mean_token_accuracy": 0.13903062492609025,
      "num_tokens": 20597270.0,
      "step": 10165
    },
    {
      "entropy": 6.304629611968994,
      "epoch": 0.8887529494013807,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004927825910581758,
      "loss": 6.1328,
      "mean_token_accuracy": 0.13831514194607736,
      "num_tokens": 20607898.0,
      "step": 10170
    },
    {
      "entropy": 6.291359233856201,
      "epoch": 0.8891898977540854,
      "grad_norm": 6.9375,
      "learning_rate": 0.0004927747597917467,
      "loss": 6.0147,
      "mean_token_accuracy": 0.14925124496221542,
      "num_tokens": 20617300.0,
      "step": 10175
    },
    {
      "entropy": 6.340207481384278,
      "epoch": 0.8896268461067902,
      "grad_norm": 3.5,
      "learning_rate": 0.0004927669243482836,
      "loss": 6.1454,
      "mean_token_accuracy": 0.13517979606986047,
      "num_tokens": 20626582.0,
      "step": 10180
    },
    {
      "entropy": 6.212292861938477,
      "epoch": 0.8900637944594949,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004927590847279369,
      "loss": 6.091,
      "mean_token_accuracy": 0.14199552088975906,
      "num_tokens": 20635802.0,
      "step": 10185
    },
    {
      "entropy": 6.278824281692505,
      "epoch": 0.8905007428121996,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004927512409308568,
      "loss": 6.1102,
      "mean_token_accuracy": 0.13981541767716407,
      "num_tokens": 20646057.0,
      "step": 10190
    },
    {
      "entropy": 6.365256643295288,
      "epoch": 0.8909376911649043,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004927433929571937,
      "loss": 6.1906,
      "mean_token_accuracy": 0.12761444449424744,
      "num_tokens": 20656511.0,
      "step": 10195
    },
    {
      "entropy": 6.380337762832641,
      "epoch": 0.891374639517609,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004927355408070982,
      "loss": 6.0926,
      "mean_token_accuracy": 0.14216326475143432,
      "num_tokens": 20666479.0,
      "step": 10200
    },
    {
      "entropy": 6.316223239898681,
      "epoch": 0.8918115878703138,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004927276844807211,
      "loss": 6.0286,
      "mean_token_accuracy": 0.13918322920799256,
      "num_tokens": 20676830.0,
      "step": 10205
    },
    {
      "entropy": 6.281838321685791,
      "epoch": 0.8922485362230185,
      "grad_norm": 3.625,
      "learning_rate": 0.0004927198239782129,
      "loss": 6.0785,
      "mean_token_accuracy": 0.14153069108724595,
      "num_tokens": 20687316.0,
      "step": 10210
    },
    {
      "entropy": 6.262103319168091,
      "epoch": 0.8926854845757232,
      "grad_norm": 5.625,
      "learning_rate": 0.0004927119592997242,
      "loss": 6.15,
      "mean_token_accuracy": 0.13477864786982535,
      "num_tokens": 20697901.0,
      "step": 10215
    },
    {
      "entropy": 6.346318340301513,
      "epoch": 0.8931224329284279,
      "grad_norm": 5.1875,
      "learning_rate": 0.0004927040904454062,
      "loss": 6.0409,
      "mean_token_accuracy": 0.14855594411492348,
      "num_tokens": 20707468.0,
      "step": 10220
    },
    {
      "entropy": 6.377528524398803,
      "epoch": 0.8935593812811325,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004926962174154097,
      "loss": 6.101,
      "mean_token_accuracy": 0.14022160694003105,
      "num_tokens": 20717098.0,
      "step": 10225
    },
    {
      "entropy": 6.342899894714355,
      "epoch": 0.8939963296338372,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004926883402098857,
      "loss": 6.1874,
      "mean_token_accuracy": 0.13333363458514214,
      "num_tokens": 20727217.0,
      "step": 10230
    },
    {
      "entropy": 6.321839380264282,
      "epoch": 0.894433277986542,
      "grad_norm": 11.5625,
      "learning_rate": 0.0004926804588289852,
      "loss": 6.1541,
      "mean_token_accuracy": 0.13853437528014184,
      "num_tokens": 20737397.0,
      "step": 10235
    },
    {
      "entropy": 6.322508001327515,
      "epoch": 0.8948702263392467,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004926725732728595,
      "loss": 5.9801,
      "mean_token_accuracy": 0.14743001759052277,
      "num_tokens": 20746712.0,
      "step": 10240
    },
    {
      "entropy": 6.361848068237305,
      "epoch": 0.8953071746919514,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004926646835416597,
      "loss": 6.115,
      "mean_token_accuracy": 0.1382555492222309,
      "num_tokens": 20757161.0,
      "step": 10245
    },
    {
      "entropy": 6.401134395599366,
      "epoch": 0.8957441230446561,
      "grad_norm": 4.5,
      "learning_rate": 0.0004926567896355372,
      "loss": 6.0489,
      "mean_token_accuracy": 0.14929357692599296,
      "num_tokens": 20766478.0,
      "step": 10250
    },
    {
      "entropy": 6.293815612792969,
      "epoch": 0.8961810713973608,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004926488915546435,
      "loss": 6.0847,
      "mean_token_accuracy": 0.14050380364060402,
      "num_tokens": 20776317.0,
      "step": 10255
    },
    {
      "entropy": 6.302836275100708,
      "epoch": 0.8966180197500655,
      "grad_norm": 3.890625,
      "learning_rate": 0.00049264098929913,
      "loss": 6.0329,
      "mean_token_accuracy": 0.13966116085648536,
      "num_tokens": 20786534.0,
      "step": 10260
    },
    {
      "entropy": 6.311154413223266,
      "epoch": 0.8970549681027703,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004926330828691482,
      "loss": 6.0399,
      "mean_token_accuracy": 0.13973370790481568,
      "num_tokens": 20796791.0,
      "step": 10265
    },
    {
      "entropy": 6.291688585281372,
      "epoch": 0.897491916455475,
      "grad_norm": 4.625,
      "learning_rate": 0.0004926251722648499,
      "loss": 6.0601,
      "mean_token_accuracy": 0.14372975155711173,
      "num_tokens": 20806834.0,
      "step": 10270
    },
    {
      "entropy": 6.21226921081543,
      "epoch": 0.8979288648081797,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004926172574863868,
      "loss": 6.0763,
      "mean_token_accuracy": 0.14103258028626442,
      "num_tokens": 20816664.0,
      "step": 10275
    },
    {
      "entropy": 6.254770278930664,
      "epoch": 0.8983658131608844,
      "grad_norm": 2.625,
      "learning_rate": 0.0004926093385339105,
      "loss": 6.1785,
      "mean_token_accuracy": 0.13757620975375176,
      "num_tokens": 20826936.0,
      "step": 10280
    },
    {
      "entropy": 6.317480278015137,
      "epoch": 0.8988027615135891,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004926014154075732,
      "loss": 6.0667,
      "mean_token_accuracy": 0.14395977631211282,
      "num_tokens": 20836732.0,
      "step": 10285
    },
    {
      "entropy": 6.413108158111572,
      "epoch": 0.8992397098662938,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004925934881075268,
      "loss": 6.1372,
      "mean_token_accuracy": 0.1347360298037529,
      "num_tokens": 20848093.0,
      "step": 10290
    },
    {
      "entropy": 6.361913824081421,
      "epoch": 0.8996766582189986,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004925855566339232,
      "loss": 6.0651,
      "mean_token_accuracy": 0.14020102322101594,
      "num_tokens": 20858678.0,
      "step": 10295
    },
    {
      "entropy": 6.217088985443115,
      "epoch": 0.9001136065717033,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004925776209869146,
      "loss": 6.042,
      "mean_token_accuracy": 0.141775668412447,
      "num_tokens": 20869319.0,
      "step": 10300
    },
    {
      "entropy": 6.278215789794922,
      "epoch": 0.9005505549244079,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004925696811666532,
      "loss": 6.0589,
      "mean_token_accuracy": 0.14474918693304062,
      "num_tokens": 20879475.0,
      "step": 10305
    },
    {
      "entropy": 6.315779066085815,
      "epoch": 0.9009875032771126,
      "grad_norm": 4.0,
      "learning_rate": 0.0004925617371732914,
      "loss": 6.0631,
      "mean_token_accuracy": 0.14823911041021348,
      "num_tokens": 20889479.0,
      "step": 10310
    },
    {
      "entropy": 6.362603855133057,
      "epoch": 0.9014244516298173,
      "grad_norm": 5.9375,
      "learning_rate": 0.0004925537890069815,
      "loss": 6.1478,
      "mean_token_accuracy": 0.13896243870258332,
      "num_tokens": 20899864.0,
      "step": 10315
    },
    {
      "entropy": 6.294197940826416,
      "epoch": 0.901861399982522,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004925458366678758,
      "loss": 6.1362,
      "mean_token_accuracy": 0.1393444925546646,
      "num_tokens": 20909006.0,
      "step": 10320
    },
    {
      "entropy": 6.294833040237426,
      "epoch": 0.9022983483352268,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004925378801561269,
      "loss": 6.0644,
      "mean_token_accuracy": 0.14355043172836304,
      "num_tokens": 20917998.0,
      "step": 10325
    },
    {
      "entropy": 6.2318922996521,
      "epoch": 0.9027352966879315,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004925299194718876,
      "loss": 6.0737,
      "mean_token_accuracy": 0.1455825909972191,
      "num_tokens": 20928407.0,
      "step": 10330
    },
    {
      "entropy": 6.237778663635254,
      "epoch": 0.9031722450406362,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004925219546153105,
      "loss": 6.093,
      "mean_token_accuracy": 0.1376163676381111,
      "num_tokens": 20938016.0,
      "step": 10335
    },
    {
      "entropy": 6.355024242401123,
      "epoch": 0.9036091933933409,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004925139855865483,
      "loss": 6.1468,
      "mean_token_accuracy": 0.13310243636369706,
      "num_tokens": 20948578.0,
      "step": 10340
    },
    {
      "entropy": 6.347204971313476,
      "epoch": 0.9040461417460456,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004925060123857539,
      "loss": 6.003,
      "mean_token_accuracy": 0.13599753081798555,
      "num_tokens": 20958193.0,
      "step": 10345
    },
    {
      "entropy": 6.243391656875611,
      "epoch": 0.9044830900987503,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004924980350130802,
      "loss": 6.0608,
      "mean_token_accuracy": 0.14421631693840026,
      "num_tokens": 20968092.0,
      "step": 10350
    },
    {
      "entropy": 6.24665002822876,
      "epoch": 0.9049200384514551,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004924900534686801,
      "loss": 6.1071,
      "mean_token_accuracy": 0.13726885840296746,
      "num_tokens": 20978303.0,
      "step": 10355
    },
    {
      "entropy": 6.373175048828125,
      "epoch": 0.9053569868041598,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004924820677527069,
      "loss": 6.0884,
      "mean_token_accuracy": 0.14156295880675315,
      "num_tokens": 20988331.0,
      "step": 10360
    },
    {
      "entropy": 6.231040716171265,
      "epoch": 0.9057939351568645,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004924740778653137,
      "loss": 6.0152,
      "mean_token_accuracy": 0.14053283855319024,
      "num_tokens": 20997785.0,
      "step": 10365
    },
    {
      "entropy": 6.282933759689331,
      "epoch": 0.9062308835095692,
      "grad_norm": 3.25,
      "learning_rate": 0.0004924660838066538,
      "loss": 6.1388,
      "mean_token_accuracy": 0.1367447182536125,
      "num_tokens": 21008026.0,
      "step": 10370
    },
    {
      "entropy": 6.304414510726929,
      "epoch": 0.9066678318622738,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004924580855768803,
      "loss": 6.1078,
      "mean_token_accuracy": 0.1424880988895893,
      "num_tokens": 21017993.0,
      "step": 10375
    },
    {
      "entropy": 6.260972404479981,
      "epoch": 0.9071047802149785,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004924500831761468,
      "loss": 6.0432,
      "mean_token_accuracy": 0.14570880830287933,
      "num_tokens": 21028149.0,
      "step": 10380
    },
    {
      "entropy": 6.334520864486694,
      "epoch": 0.9075417285676833,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004924420766046069,
      "loss": 6.202,
      "mean_token_accuracy": 0.12750043347477913,
      "num_tokens": 21039229.0,
      "step": 10385
    },
    {
      "entropy": 6.348653364181518,
      "epoch": 0.907978676920388,
      "grad_norm": 3.5,
      "learning_rate": 0.0004924340658624139,
      "loss": 6.1021,
      "mean_token_accuracy": 0.14049511551856994,
      "num_tokens": 21049358.0,
      "step": 10390
    },
    {
      "entropy": 6.350093078613281,
      "epoch": 0.9084156252730927,
      "grad_norm": 2.75,
      "learning_rate": 0.0004924260509497217,
      "loss": 6.1523,
      "mean_token_accuracy": 0.14013315364718437,
      "num_tokens": 21060385.0,
      "step": 10395
    },
    {
      "entropy": 6.338852310180664,
      "epoch": 0.9088525736257974,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004924180318666838,
      "loss": 6.1561,
      "mean_token_accuracy": 0.135500368475914,
      "num_tokens": 21070423.0,
      "step": 10400
    },
    {
      "entropy": 6.458673000335693,
      "epoch": 0.9092895219785021,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004924100086134541,
      "loss": 6.1168,
      "mean_token_accuracy": 0.13829629346728325,
      "num_tokens": 21080505.0,
      "step": 10405
    },
    {
      "entropy": 6.314735269546508,
      "epoch": 0.9097264703312069,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004924019811901866,
      "loss": 6.1056,
      "mean_token_accuracy": 0.13905404806137084,
      "num_tokens": 21090875.0,
      "step": 10410
    },
    {
      "entropy": 6.256156015396118,
      "epoch": 0.9101634186839116,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004923939495970352,
      "loss": 6.0579,
      "mean_token_accuracy": 0.14047153294086456,
      "num_tokens": 21101427.0,
      "step": 10415
    },
    {
      "entropy": 6.377035236358642,
      "epoch": 0.9106003670366163,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004923859138341539,
      "loss": 6.1826,
      "mean_token_accuracy": 0.13201440423727034,
      "num_tokens": 21112691.0,
      "step": 10420
    },
    {
      "entropy": 6.369430875778198,
      "epoch": 0.911037315389321,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004923778739016969,
      "loss": 6.1788,
      "mean_token_accuracy": 0.13466673269867896,
      "num_tokens": 21123932.0,
      "step": 10425
    },
    {
      "entropy": 6.336857652664184,
      "epoch": 0.9114742637420257,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004923698297998183,
      "loss": 6.1634,
      "mean_token_accuracy": 0.13220114186406134,
      "num_tokens": 21135380.0,
      "step": 10430
    },
    {
      "entropy": 6.304929161071778,
      "epoch": 0.9119112120947304,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004923617815286725,
      "loss": 6.0673,
      "mean_token_accuracy": 0.1438829481601715,
      "num_tokens": 21144951.0,
      "step": 10435
    },
    {
      "entropy": 6.273499965667725,
      "epoch": 0.9123481604474352,
      "grad_norm": 4.25,
      "learning_rate": 0.0004923537290884138,
      "loss": 5.9916,
      "mean_token_accuracy": 0.14025868400931357,
      "num_tokens": 21155919.0,
      "step": 10440
    },
    {
      "entropy": 6.225626230239868,
      "epoch": 0.9127851088001399,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004923456724791967,
      "loss": 6.0579,
      "mean_token_accuracy": 0.14243191182613374,
      "num_tokens": 21166186.0,
      "step": 10445
    },
    {
      "entropy": 6.28761773109436,
      "epoch": 0.9132220571528445,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004923376117011758,
      "loss": 6.0437,
      "mean_token_accuracy": 0.1450629323720932,
      "num_tokens": 21176043.0,
      "step": 10450
    },
    {
      "entropy": 6.271642875671387,
      "epoch": 0.9136590055055492,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004923295467545055,
      "loss": 6.0481,
      "mean_token_accuracy": 0.14478113055229186,
      "num_tokens": 21185677.0,
      "step": 10455
    },
    {
      "entropy": 6.287078475952148,
      "epoch": 0.9140959538582539,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004923214776393407,
      "loss": 6.1479,
      "mean_token_accuracy": 0.1319975309073925,
      "num_tokens": 21195977.0,
      "step": 10460
    },
    {
      "entropy": 6.382370710372925,
      "epoch": 0.9145329022109586,
      "grad_norm": 2.6875,
      "learning_rate": 0.000492313404355836,
      "loss": 6.1256,
      "mean_token_accuracy": 0.13694120794534684,
      "num_tokens": 21205454.0,
      "step": 10465
    },
    {
      "entropy": 6.326042127609253,
      "epoch": 0.9149698505636634,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004923053269041464,
      "loss": 6.0306,
      "mean_token_accuracy": 0.14461090862751008,
      "num_tokens": 21215700.0,
      "step": 10470
    },
    {
      "entropy": 6.237873601913452,
      "epoch": 0.9154067989163681,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004922972452844267,
      "loss": 6.1332,
      "mean_token_accuracy": 0.13864517137408255,
      "num_tokens": 21226544.0,
      "step": 10475
    },
    {
      "entropy": 6.299554777145386,
      "epoch": 0.9158437472690728,
      "grad_norm": 5.46875,
      "learning_rate": 0.0004922891594968321,
      "loss": 6.1649,
      "mean_token_accuracy": 0.1335770383477211,
      "num_tokens": 21236792.0,
      "step": 10480
    },
    {
      "entropy": 6.3094775676727295,
      "epoch": 0.9162806956217775,
      "grad_norm": 2.75,
      "learning_rate": 0.0004922810695415175,
      "loss": 6.0623,
      "mean_token_accuracy": 0.1388197846710682,
      "num_tokens": 21247039.0,
      "step": 10485
    },
    {
      "entropy": 6.229126882553101,
      "epoch": 0.9167176439744822,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004922729754186381,
      "loss": 6.082,
      "mean_token_accuracy": 0.14173468947410583,
      "num_tokens": 21257008.0,
      "step": 10490
    },
    {
      "entropy": 6.325405645370483,
      "epoch": 0.9171545923271869,
      "grad_norm": 15.5625,
      "learning_rate": 0.0004922648771283492,
      "loss": 6.1898,
      "mean_token_accuracy": 0.13802886083722116,
      "num_tokens": 21267638.0,
      "step": 10495
    },
    {
      "entropy": 6.431231355667114,
      "epoch": 0.9175915406798917,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004922567746708062,
      "loss": 6.0726,
      "mean_token_accuracy": 0.14293336123228073,
      "num_tokens": 21277327.0,
      "step": 10500
    },
    {
      "entropy": 6.289720296859741,
      "epoch": 0.9180284890325964,
      "grad_norm": 3.0,
      "learning_rate": 0.0004922486680461642,
      "loss": 6.1585,
      "mean_token_accuracy": 0.13599686250090598,
      "num_tokens": 21287579.0,
      "step": 10505
    },
    {
      "entropy": 6.241109085083008,
      "epoch": 0.9184654373853011,
      "grad_norm": 4.5,
      "learning_rate": 0.0004922405572545792,
      "loss": 6.0908,
      "mean_token_accuracy": 0.13013199865818023,
      "num_tokens": 21297888.0,
      "step": 10510
    },
    {
      "entropy": 6.362125968933105,
      "epoch": 0.9189023857380058,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004922324422962063,
      "loss": 6.1187,
      "mean_token_accuracy": 0.1398759014904499,
      "num_tokens": 21308506.0,
      "step": 10515
    },
    {
      "entropy": 6.395807313919067,
      "epoch": 0.9193393340907104,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004922243231712013,
      "loss": 6.1643,
      "mean_token_accuracy": 0.13145793378353118,
      "num_tokens": 21319252.0,
      "step": 10520
    },
    {
      "entropy": 6.33790135383606,
      "epoch": 0.9197762824434151,
      "grad_norm": 3.765625,
      "learning_rate": 0.00049221619987972,
      "loss": 6.1234,
      "mean_token_accuracy": 0.1420030564069748,
      "num_tokens": 21328829.0,
      "step": 10525
    },
    {
      "entropy": 6.305783748626709,
      "epoch": 0.9202132307961199,
      "grad_norm": 3.375,
      "learning_rate": 0.0004922080724219182,
      "loss": 5.9619,
      "mean_token_accuracy": 0.14907341450452805,
      "num_tokens": 21338524.0,
      "step": 10530
    },
    {
      "entropy": 6.27631196975708,
      "epoch": 0.9206501791488246,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004921999407979518,
      "loss": 6.1077,
      "mean_token_accuracy": 0.13910981491208077,
      "num_tokens": 21348559.0,
      "step": 10535
    },
    {
      "entropy": 6.317100191116333,
      "epoch": 0.9210871275015293,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004921918050079766,
      "loss": 6.1276,
      "mean_token_accuracy": 0.14055971801280975,
      "num_tokens": 21360763.0,
      "step": 10540
    },
    {
      "entropy": 6.343057346343994,
      "epoch": 0.921524075854234,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004921836650521489,
      "loss": 6.1145,
      "mean_token_accuracy": 0.13225539699196814,
      "num_tokens": 21370048.0,
      "step": 10545
    },
    {
      "entropy": 6.257398796081543,
      "epoch": 0.9219610242069387,
      "grad_norm": 5.6875,
      "learning_rate": 0.0004921755209306247,
      "loss": 6.1007,
      "mean_token_accuracy": 0.1446698412299156,
      "num_tokens": 21379760.0,
      "step": 10550
    },
    {
      "entropy": 6.321170902252197,
      "epoch": 0.9223979725596434,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004921673726435601,
      "loss": 6.0849,
      "mean_token_accuracy": 0.14090224951505662,
      "num_tokens": 21389712.0,
      "step": 10555
    },
    {
      "entropy": 6.365037250518799,
      "epoch": 0.9228349209123482,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004921592201911115,
      "loss": 6.0533,
      "mean_token_accuracy": 0.13930047750473024,
      "num_tokens": 21399515.0,
      "step": 10560
    },
    {
      "entropy": 6.2545778274536135,
      "epoch": 0.9232718692650529,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004921510635734353,
      "loss": 6.0322,
      "mean_token_accuracy": 0.1435614198446274,
      "num_tokens": 21409704.0,
      "step": 10565
    },
    {
      "entropy": 6.227499389648438,
      "epoch": 0.9237088176177576,
      "grad_norm": 3.1875,
      "learning_rate": 0.000492142902790688,
      "loss": 6.0556,
      "mean_token_accuracy": 0.14314178228378296,
      "num_tokens": 21419182.0,
      "step": 10570
    },
    {
      "entropy": 6.235004997253418,
      "epoch": 0.9241457659704623,
      "grad_norm": 3.75,
      "learning_rate": 0.0004921347378430259,
      "loss": 5.9404,
      "mean_token_accuracy": 0.14756003767251968,
      "num_tokens": 21429749.0,
      "step": 10575
    },
    {
      "entropy": 6.407311201095581,
      "epoch": 0.924582714323167,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004921265687306059,
      "loss": 6.1985,
      "mean_token_accuracy": 0.13048061951994896,
      "num_tokens": 21440514.0,
      "step": 10580
    },
    {
      "entropy": 6.357501745223999,
      "epoch": 0.9250196626758717,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004921183954535844,
      "loss": 6.0656,
      "mean_token_accuracy": 0.13995085656642914,
      "num_tokens": 21450647.0,
      "step": 10585
    },
    {
      "entropy": 6.273309135437012,
      "epoch": 0.9254566110285765,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004921102180121184,
      "loss": 6.0829,
      "mean_token_accuracy": 0.13841803446412088,
      "num_tokens": 21460233.0,
      "step": 10590
    },
    {
      "entropy": 6.277070474624634,
      "epoch": 0.9258935593812812,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004921020364063646,
      "loss": 6.0859,
      "mean_token_accuracy": 0.14465343654155732,
      "num_tokens": 21470207.0,
      "step": 10595
    },
    {
      "entropy": 6.316789388656616,
      "epoch": 0.9263305077339858,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004920938506364801,
      "loss": 6.1191,
      "mean_token_accuracy": 0.14205100312829017,
      "num_tokens": 21480658.0,
      "step": 10600
    },
    {
      "entropy": 6.287032175064087,
      "epoch": 0.9267674560866905,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004920856607026216,
      "loss": 6.0094,
      "mean_token_accuracy": 0.15317313969135285,
      "num_tokens": 21490566.0,
      "step": 10605
    },
    {
      "entropy": 6.285973072052002,
      "epoch": 0.9272044044393952,
      "grad_norm": 8.1875,
      "learning_rate": 0.0004920774666049465,
      "loss": 6.0971,
      "mean_token_accuracy": 0.14561666771769524,
      "num_tokens": 21500752.0,
      "step": 10610
    },
    {
      "entropy": 6.282587003707886,
      "epoch": 0.9276413527921,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004920692683436118,
      "loss": 6.0986,
      "mean_token_accuracy": 0.13332514762878417,
      "num_tokens": 21509802.0,
      "step": 10615
    },
    {
      "entropy": 6.293485879898071,
      "epoch": 0.9280783011448047,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004920610659187748,
      "loss": 6.1688,
      "mean_token_accuracy": 0.1338999591767788,
      "num_tokens": 21520962.0,
      "step": 10620
    },
    {
      "entropy": 6.392551612854004,
      "epoch": 0.9285152494975094,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004920528593305929,
      "loss": 6.1216,
      "mean_token_accuracy": 0.1380319967865944,
      "num_tokens": 21530580.0,
      "step": 10625
    },
    {
      "entropy": 6.298496866226197,
      "epoch": 0.9289521978502141,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004920446485792233,
      "loss": 6.0647,
      "mean_token_accuracy": 0.1389675423502922,
      "num_tokens": 21541431.0,
      "step": 10630
    },
    {
      "entropy": 6.241003131866455,
      "epoch": 0.9293891462029188,
      "grad_norm": 4.375,
      "learning_rate": 0.0004920364336648236,
      "loss": 6.0969,
      "mean_token_accuracy": 0.1396551623940468,
      "num_tokens": 21551012.0,
      "step": 10635
    },
    {
      "entropy": 6.2428161144256595,
      "epoch": 0.9298260945556235,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004920282145875514,
      "loss": 6.038,
      "mean_token_accuracy": 0.14159249365329743,
      "num_tokens": 21560676.0,
      "step": 10640
    },
    {
      "entropy": 6.306736803054809,
      "epoch": 0.9302630429083283,
      "grad_norm": 6.03125,
      "learning_rate": 0.0004920199913475644,
      "loss": 6.0172,
      "mean_token_accuracy": 0.14645390063524247,
      "num_tokens": 21570848.0,
      "step": 10645
    },
    {
      "entropy": 6.320476007461548,
      "epoch": 0.930699991261033,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004920117639450202,
      "loss": 6.0957,
      "mean_token_accuracy": 0.13792168572545052,
      "num_tokens": 21582358.0,
      "step": 10650
    },
    {
      "entropy": 6.289608716964722,
      "epoch": 0.9311369396137377,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004920035323800767,
      "loss": 6.0569,
      "mean_token_accuracy": 0.14116547256708145,
      "num_tokens": 21591709.0,
      "step": 10655
    },
    {
      "entropy": 6.323925542831421,
      "epoch": 0.9315738879664424,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004919952966528917,
      "loss": 6.0029,
      "mean_token_accuracy": 0.14624925032258035,
      "num_tokens": 21601092.0,
      "step": 10660
    },
    {
      "entropy": 6.279575061798096,
      "epoch": 0.932010836319147,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004919870567636233,
      "loss": 6.0456,
      "mean_token_accuracy": 0.14417173564434052,
      "num_tokens": 21611003.0,
      "step": 10665
    },
    {
      "entropy": 6.296811199188232,
      "epoch": 0.9324477846718517,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004919788127124293,
      "loss": 6.1246,
      "mean_token_accuracy": 0.13917409628629684,
      "num_tokens": 21621528.0,
      "step": 10670
    },
    {
      "entropy": 6.285241174697876,
      "epoch": 0.9328847330245565,
      "grad_norm": 4.25,
      "learning_rate": 0.000491970564499468,
      "loss": 6.034,
      "mean_token_accuracy": 0.14918432906270027,
      "num_tokens": 21631431.0,
      "step": 10675
    },
    {
      "entropy": 6.301537704467774,
      "epoch": 0.9333216813772612,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004919623121248976,
      "loss": 6.064,
      "mean_token_accuracy": 0.14415694102644921,
      "num_tokens": 21640717.0,
      "step": 10680
    },
    {
      "entropy": 6.334484481811524,
      "epoch": 0.9337586297299659,
      "grad_norm": 4.0,
      "learning_rate": 0.0004919540555888765,
      "loss": 6.0607,
      "mean_token_accuracy": 0.14421268552541733,
      "num_tokens": 21650493.0,
      "step": 10685
    },
    {
      "entropy": 6.283117389678955,
      "epoch": 0.9341955780826706,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004919457948915628,
      "loss": 6.0727,
      "mean_token_accuracy": 0.14414482265710832,
      "num_tokens": 21660062.0,
      "step": 10690
    },
    {
      "entropy": 6.221496200561523,
      "epoch": 0.9346325264353753,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004919375300331151,
      "loss": 6.0408,
      "mean_token_accuracy": 0.13765192478895188,
      "num_tokens": 21669364.0,
      "step": 10695
    },
    {
      "entropy": 6.292454814910888,
      "epoch": 0.93506947478808,
      "grad_norm": 3.5,
      "learning_rate": 0.0004919292610136919,
      "loss": 6.1302,
      "mean_token_accuracy": 0.13302175104618072,
      "num_tokens": 21680793.0,
      "step": 10700
    },
    {
      "entropy": 6.331288528442383,
      "epoch": 0.9355064231407848,
      "grad_norm": 4.375,
      "learning_rate": 0.0004919209878334518,
      "loss": 6.1407,
      "mean_token_accuracy": 0.14002989903092383,
      "num_tokens": 21690862.0,
      "step": 10705
    },
    {
      "entropy": 6.304392957687378,
      "epoch": 0.9359433714934895,
      "grad_norm": 5.1875,
      "learning_rate": 0.0004919127104925535,
      "loss": 6.0302,
      "mean_token_accuracy": 0.1459456965327263,
      "num_tokens": 21700665.0,
      "step": 10710
    },
    {
      "entropy": 6.283321523666382,
      "epoch": 0.9363803198461942,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004919044289911557,
      "loss": 6.0833,
      "mean_token_accuracy": 0.14758456200361253,
      "num_tokens": 21711128.0,
      "step": 10715
    },
    {
      "entropy": 6.380154228210449,
      "epoch": 0.9368172681988989,
      "grad_norm": 3.0,
      "learning_rate": 0.0004918961433294173,
      "loss": 6.1456,
      "mean_token_accuracy": 0.13450410589575768,
      "num_tokens": 21720233.0,
      "step": 10720
    },
    {
      "entropy": 6.334622478485107,
      "epoch": 0.9372542165516036,
      "grad_norm": 3.25,
      "learning_rate": 0.0004918878535074973,
      "loss": 6.0885,
      "mean_token_accuracy": 0.14097136110067368,
      "num_tokens": 21730980.0,
      "step": 10725
    },
    {
      "entropy": 6.252231073379517,
      "epoch": 0.9376911649043083,
      "grad_norm": 3.125,
      "learning_rate": 0.0004918795595255545,
      "loss": 6.0933,
      "mean_token_accuracy": 0.14346909672021865,
      "num_tokens": 21742802.0,
      "step": 10730
    },
    {
      "entropy": 6.369295740127564,
      "epoch": 0.9381281132570131,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004918712613837483,
      "loss": 6.0981,
      "mean_token_accuracy": 0.1403065361082554,
      "num_tokens": 21752743.0,
      "step": 10735
    },
    {
      "entropy": 6.332711839675904,
      "epoch": 0.9385650616097178,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004918629590822375,
      "loss": 6.0773,
      "mean_token_accuracy": 0.14234824255108833,
      "num_tokens": 21763198.0,
      "step": 10740
    },
    {
      "entropy": 6.2796478271484375,
      "epoch": 0.9390020099624224,
      "grad_norm": 3.0,
      "learning_rate": 0.0004918546526211816,
      "loss": 6.0715,
      "mean_token_accuracy": 0.1428854875266552,
      "num_tokens": 21773030.0,
      "step": 10745
    },
    {
      "entropy": 6.264680433273315,
      "epoch": 0.9394389583151271,
      "grad_norm": 17.875,
      "learning_rate": 0.0004918463420007398,
      "loss": 6.0371,
      "mean_token_accuracy": 0.14570278525352479,
      "num_tokens": 21783796.0,
      "step": 10750
    },
    {
      "entropy": 6.296322631835937,
      "epoch": 0.9398759066678318,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004918380272210715,
      "loss": 6.0665,
      "mean_token_accuracy": 0.1399317555129528,
      "num_tokens": 21792504.0,
      "step": 10755
    },
    {
      "entropy": 6.349744749069214,
      "epoch": 0.9403128550205365,
      "grad_norm": 6.78125,
      "learning_rate": 0.0004918297082823362,
      "loss": 6.1434,
      "mean_token_accuracy": 0.13404348492622375,
      "num_tokens": 21802764.0,
      "step": 10760
    },
    {
      "entropy": 6.356295490264893,
      "epoch": 0.9407498033732413,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004918213851846935,
      "loss": 6.1048,
      "mean_token_accuracy": 0.13999746963381768,
      "num_tokens": 21813051.0,
      "step": 10765
    },
    {
      "entropy": 6.333794832229614,
      "epoch": 0.941186751725946,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004918130579283031,
      "loss": 6.1451,
      "mean_token_accuracy": 0.13952677547931672,
      "num_tokens": 21822941.0,
      "step": 10770
    },
    {
      "entropy": 6.293758821487427,
      "epoch": 0.9416237000786507,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004918047265133246,
      "loss": 6.0817,
      "mean_token_accuracy": 0.13951387107372284,
      "num_tokens": 21833225.0,
      "step": 10775
    },
    {
      "entropy": 6.291474723815918,
      "epoch": 0.9420606484313554,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004917963909399178,
      "loss": 6.0776,
      "mean_token_accuracy": 0.14075872376561166,
      "num_tokens": 21842360.0,
      "step": 10780
    },
    {
      "entropy": 6.352324056625366,
      "epoch": 0.9424975967840601,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004917880512082427,
      "loss": 6.0853,
      "mean_token_accuracy": 0.13701972514390945,
      "num_tokens": 21852515.0,
      "step": 10785
    },
    {
      "entropy": 6.3446691036224365,
      "epoch": 0.9429345451367648,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004917797073184591,
      "loss": 6.1883,
      "mean_token_accuracy": 0.13434592112898827,
      "num_tokens": 21863040.0,
      "step": 10790
    },
    {
      "entropy": 6.369406795501709,
      "epoch": 0.9433714934894696,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004917713592707271,
      "loss": 6.1001,
      "mean_token_accuracy": 0.14156363010406495,
      "num_tokens": 21872101.0,
      "step": 10795
    },
    {
      "entropy": 6.319452810287475,
      "epoch": 0.9438084418421743,
      "grad_norm": 2.75,
      "learning_rate": 0.0004917630070652069,
      "loss": 6.1005,
      "mean_token_accuracy": 0.14293818473815917,
      "num_tokens": 21882148.0,
      "step": 10800
    },
    {
      "entropy": 6.280195426940918,
      "epoch": 0.944245390194879,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004917546507020587,
      "loss": 6.0742,
      "mean_token_accuracy": 0.14549223482608795,
      "num_tokens": 21891832.0,
      "step": 10805
    },
    {
      "entropy": 6.267314434051514,
      "epoch": 0.9446823385475837,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004917462901814427,
      "loss": 6.0571,
      "mean_token_accuracy": 0.13927336037158966,
      "num_tokens": 21901544.0,
      "step": 10810
    },
    {
      "entropy": 6.322244358062744,
      "epoch": 0.9451192869002883,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004917379255035192,
      "loss": 6.1296,
      "mean_token_accuracy": 0.1371576137840748,
      "num_tokens": 21911690.0,
      "step": 10815
    },
    {
      "entropy": 6.390911483764649,
      "epoch": 0.9455562352529931,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004917295566684487,
      "loss": 6.1356,
      "mean_token_accuracy": 0.13117053657770156,
      "num_tokens": 21922968.0,
      "step": 10820
    },
    {
      "entropy": 6.26896710395813,
      "epoch": 0.9459931836056978,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004917211836763919,
      "loss": 6.0013,
      "mean_token_accuracy": 0.14541664198040963,
      "num_tokens": 21932904.0,
      "step": 10825
    },
    {
      "entropy": 6.221676397323608,
      "epoch": 0.9464301319584025,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004917128065275091,
      "loss": 5.9942,
      "mean_token_accuracy": 0.15279717221856118,
      "num_tokens": 21943065.0,
      "step": 10830
    },
    {
      "entropy": 6.300609445571899,
      "epoch": 0.9468670803111072,
      "grad_norm": 5.40625,
      "learning_rate": 0.0004917044252219612,
      "loss": 6.0731,
      "mean_token_accuracy": 0.1324281983077526,
      "num_tokens": 21953448.0,
      "step": 10835
    },
    {
      "entropy": 6.348029375076294,
      "epoch": 0.9473040286638119,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004916960397599088,
      "loss": 6.0763,
      "mean_token_accuracy": 0.1386580340564251,
      "num_tokens": 21963374.0,
      "step": 10840
    },
    {
      "entropy": 6.314178562164306,
      "epoch": 0.9477409770165166,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004916876501415129,
      "loss": 6.0817,
      "mean_token_accuracy": 0.13696496933698654,
      "num_tokens": 21973726.0,
      "step": 10845
    },
    {
      "entropy": 6.24705662727356,
      "epoch": 0.9481779253692214,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004916792563669342,
      "loss": 6.0021,
      "mean_token_accuracy": 0.14274803400039673,
      "num_tokens": 21983203.0,
      "step": 10850
    },
    {
      "entropy": 6.320058107376099,
      "epoch": 0.9486148737219261,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004916708584363339,
      "loss": 6.1273,
      "mean_token_accuracy": 0.14053093940019606,
      "num_tokens": 21993435.0,
      "step": 10855
    },
    {
      "entropy": 6.354315280914307,
      "epoch": 0.9490518220746308,
      "grad_norm": 2.828125,
      "learning_rate": 0.000491662456349873,
      "loss": 6.1251,
      "mean_token_accuracy": 0.13344834968447686,
      "num_tokens": 22003230.0,
      "step": 10860
    },
    {
      "entropy": 6.3009168148040775,
      "epoch": 0.9494887704273355,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004916540501077127,
      "loss": 6.0753,
      "mean_token_accuracy": 0.14576857537031174,
      "num_tokens": 22012910.0,
      "step": 10865
    },
    {
      "entropy": 6.323046398162842,
      "epoch": 0.9499257187800402,
      "grad_norm": 3.125,
      "learning_rate": 0.0004916456397100141,
      "loss": 6.0539,
      "mean_token_accuracy": 0.13978277817368506,
      "num_tokens": 22022802.0,
      "step": 10870
    },
    {
      "entropy": 6.338568210601807,
      "epoch": 0.9503626671327449,
      "grad_norm": 5.0625,
      "learning_rate": 0.0004916372251569386,
      "loss": 6.1412,
      "mean_token_accuracy": 0.13572175204753875,
      "num_tokens": 22032669.0,
      "step": 10875
    },
    {
      "entropy": 6.2824968814849855,
      "epoch": 0.9507996154854497,
      "grad_norm": 10.6875,
      "learning_rate": 0.0004916288064486477,
      "loss": 6.0518,
      "mean_token_accuracy": 0.14592863917350768,
      "num_tokens": 22042477.0,
      "step": 10880
    },
    {
      "entropy": 6.2819411277771,
      "epoch": 0.9512365638381544,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004916203835853027,
      "loss": 6.1136,
      "mean_token_accuracy": 0.13779654875397682,
      "num_tokens": 22052891.0,
      "step": 10885
    },
    {
      "entropy": 6.2667803287506105,
      "epoch": 0.951673512190859,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004916119565670652,
      "loss": 6.0415,
      "mean_token_accuracy": 0.15155872255563735,
      "num_tokens": 22062413.0,
      "step": 10890
    },
    {
      "entropy": 6.386011123657227,
      "epoch": 0.9521104605435637,
      "grad_norm": 2.921875,
      "learning_rate": 0.000491603525394097,
      "loss": 6.1115,
      "mean_token_accuracy": 0.13975054621696473,
      "num_tokens": 22072565.0,
      "step": 10895
    },
    {
      "entropy": 6.314344978332519,
      "epoch": 0.9525474088962684,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004915950900665596,
      "loss": 6.0734,
      "mean_token_accuracy": 0.13680414631962776,
      "num_tokens": 22082910.0,
      "step": 10900
    },
    {
      "entropy": 6.19198751449585,
      "epoch": 0.9529843572489731,
      "grad_norm": 2.515625,
      "learning_rate": 0.000491586650584615,
      "loss": 6.0102,
      "mean_token_accuracy": 0.15061622858047485,
      "num_tokens": 22092636.0,
      "step": 10905
    },
    {
      "entropy": 6.3510682582855225,
      "epoch": 0.9534213056016779,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004915782069484248,
      "loss": 6.073,
      "mean_token_accuracy": 0.13827937245368957,
      "num_tokens": 22102232.0,
      "step": 10910
    },
    {
      "entropy": 6.419935274124145,
      "epoch": 0.9538582539543826,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004915697591581511,
      "loss": 6.1568,
      "mean_token_accuracy": 0.1371769055724144,
      "num_tokens": 22113592.0,
      "step": 10915
    },
    {
      "entropy": 6.313469982147216,
      "epoch": 0.9542952023070873,
      "grad_norm": 2.640625,
      "learning_rate": 0.000491561307213956,
      "loss": 6.0889,
      "mean_token_accuracy": 0.14445227086544038,
      "num_tokens": 22123731.0,
      "step": 10920
    },
    {
      "entropy": 6.325977087020874,
      "epoch": 0.954732150659792,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004915528511160016,
      "loss": 6.1367,
      "mean_token_accuracy": 0.12917937487363815,
      "num_tokens": 22133801.0,
      "step": 10925
    },
    {
      "entropy": 6.37852931022644,
      "epoch": 0.9551690990124967,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004915443908644501,
      "loss": 6.0959,
      "mean_token_accuracy": 0.13814767599105834,
      "num_tokens": 22143939.0,
      "step": 10930
    },
    {
      "entropy": 6.281315279006958,
      "epoch": 0.9556060473652014,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004915359264594636,
      "loss": 6.139,
      "mean_token_accuracy": 0.1383027106523514,
      "num_tokens": 22154954.0,
      "step": 10935
    },
    {
      "entropy": 6.278833055496216,
      "epoch": 0.9560429957179062,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004915274579012045,
      "loss": 6.0902,
      "mean_token_accuracy": 0.13915328606963157,
      "num_tokens": 22165225.0,
      "step": 10940
    },
    {
      "entropy": 6.346120405197143,
      "epoch": 0.9564799440706109,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004915189851898355,
      "loss": 6.1253,
      "mean_token_accuracy": 0.13293414786458016,
      "num_tokens": 22174891.0,
      "step": 10945
    },
    {
      "entropy": 6.293230819702148,
      "epoch": 0.9569168924233156,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004915105083255187,
      "loss": 6.1112,
      "mean_token_accuracy": 0.14333042055368422,
      "num_tokens": 22185255.0,
      "step": 10950
    },
    {
      "entropy": 6.298870611190796,
      "epoch": 0.9573538407760203,
      "grad_norm": 3.921875,
      "learning_rate": 0.000491502027308417,
      "loss": 6.1624,
      "mean_token_accuracy": 0.1368987739086151,
      "num_tokens": 22196124.0,
      "step": 10955
    },
    {
      "entropy": 6.34798755645752,
      "epoch": 0.957790789128725,
      "grad_norm": 2.921875,
      "learning_rate": 0.000491493542138693,
      "loss": 6.0913,
      "mean_token_accuracy": 0.1389659345149994,
      "num_tokens": 22207563.0,
      "step": 10960
    },
    {
      "entropy": 6.326799154281616,
      "epoch": 0.9582277374814296,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004914850528165094,
      "loss": 6.126,
      "mean_token_accuracy": 0.14058008939027786,
      "num_tokens": 22218527.0,
      "step": 10965
    },
    {
      "entropy": 6.3176027774810795,
      "epoch": 0.9586646858341344,
      "grad_norm": 3.140625,
      "learning_rate": 0.000491476559342029,
      "loss": 6.0861,
      "mean_token_accuracy": 0.13868209570646287,
      "num_tokens": 22229538.0,
      "step": 10970
    },
    {
      "entropy": 6.321827125549317,
      "epoch": 0.9591016341868391,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004914680617154148,
      "loss": 5.9952,
      "mean_token_accuracy": 0.14945639073848724,
      "num_tokens": 22240047.0,
      "step": 10975
    },
    {
      "entropy": 6.216324377059936,
      "epoch": 0.9595385825395438,
      "grad_norm": 6.875,
      "learning_rate": 0.0004914595599368297,
      "loss": 6.0157,
      "mean_token_accuracy": 0.14426889941096305,
      "num_tokens": 22250440.0,
      "step": 10980
    },
    {
      "entropy": 6.242133235931396,
      "epoch": 0.9599755308922485,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004914510540064369,
      "loss": 6.0582,
      "mean_token_accuracy": 0.14533379822969436,
      "num_tokens": 22259909.0,
      "step": 10985
    },
    {
      "entropy": 6.309207630157471,
      "epoch": 0.9604124792449532,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004914425439243994,
      "loss": 6.0927,
      "mean_token_accuracy": 0.13932787701487542,
      "num_tokens": 22270358.0,
      "step": 10990
    },
    {
      "entropy": 6.293834257125854,
      "epoch": 0.9608494275976579,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004914340296908805,
      "loss": 6.1309,
      "mean_token_accuracy": 0.134774786233902,
      "num_tokens": 22280591.0,
      "step": 10995
    },
    {
      "entropy": 6.349157428741455,
      "epoch": 0.9612863759503627,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004914255113060435,
      "loss": 6.0999,
      "mean_token_accuracy": 0.1414651446044445,
      "num_tokens": 22291067.0,
      "step": 11000
    },
    {
      "entropy": 6.298183870315552,
      "epoch": 0.9617233243030674,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004914169887700519,
      "loss": 6.0417,
      "mean_token_accuracy": 0.14579998254776,
      "num_tokens": 22300919.0,
      "step": 11005
    },
    {
      "entropy": 6.314806699752808,
      "epoch": 0.9621602726557721,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004914084620830689,
      "loss": 6.1361,
      "mean_token_accuracy": 0.13179718255996703,
      "num_tokens": 22311562.0,
      "step": 11010
    },
    {
      "entropy": 6.32667818069458,
      "epoch": 0.9625972210084768,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004913999312452583,
      "loss": 6.0986,
      "mean_token_accuracy": 0.1386895827949047,
      "num_tokens": 22322980.0,
      "step": 11015
    },
    {
      "entropy": 6.301436138153076,
      "epoch": 0.9630341693611815,
      "grad_norm": 17.625,
      "learning_rate": 0.0004913913962567836,
      "loss": 6.0352,
      "mean_token_accuracy": 0.143113811314106,
      "num_tokens": 22333242.0,
      "step": 11020
    },
    {
      "entropy": 6.3390820026397705,
      "epoch": 0.9634711177138863,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004913828571178085,
      "loss": 6.1173,
      "mean_token_accuracy": 0.13968848809599876,
      "num_tokens": 22343618.0,
      "step": 11025
    },
    {
      "entropy": 6.277801036834717,
      "epoch": 0.963908066066591,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004913743138284969,
      "loss": 6.1047,
      "mean_token_accuracy": 0.13118595331907273,
      "num_tokens": 22354249.0,
      "step": 11030
    },
    {
      "entropy": 6.303753280639649,
      "epoch": 0.9643450144192957,
      "grad_norm": 7.5,
      "learning_rate": 0.0004913657663890124,
      "loss": 6.0655,
      "mean_token_accuracy": 0.13843395337462425,
      "num_tokens": 22363824.0,
      "step": 11035
    },
    {
      "entropy": 6.329318809509277,
      "epoch": 0.9647819627720003,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004913572147995193,
      "loss": 6.0313,
      "mean_token_accuracy": 0.14556610733270645,
      "num_tokens": 22373470.0,
      "step": 11040
    },
    {
      "entropy": 6.253305816650391,
      "epoch": 0.965218911124705,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004913486590601814,
      "loss": 6.0183,
      "mean_token_accuracy": 0.13501980900764465,
      "num_tokens": 22382999.0,
      "step": 11045
    },
    {
      "entropy": 6.25350546836853,
      "epoch": 0.9656558594774097,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004913400991711629,
      "loss": 6.0735,
      "mean_token_accuracy": 0.14586086347699165,
      "num_tokens": 22393415.0,
      "step": 11050
    },
    {
      "entropy": 6.290294170379639,
      "epoch": 0.9660928078301145,
      "grad_norm": 12.3125,
      "learning_rate": 0.0004913315351326279,
      "loss": 6.0095,
      "mean_token_accuracy": 0.14657122418284416,
      "num_tokens": 22403040.0,
      "step": 11055
    },
    {
      "entropy": 6.2926122665405275,
      "epoch": 0.9665297561828192,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004913229669447406,
      "loss": 6.0371,
      "mean_token_accuracy": 0.1438804402947426,
      "num_tokens": 22413434.0,
      "step": 11060
    },
    {
      "entropy": 6.36156210899353,
      "epoch": 0.9669667045355239,
      "grad_norm": 3.25,
      "learning_rate": 0.0004913143946076656,
      "loss": 6.1683,
      "mean_token_accuracy": 0.1354618638753891,
      "num_tokens": 22424446.0,
      "step": 11065
    },
    {
      "entropy": 6.285894060134888,
      "epoch": 0.9674036528882286,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004913058181215671,
      "loss": 6.1018,
      "mean_token_accuracy": 0.14309248700737953,
      "num_tokens": 22433928.0,
      "step": 11070
    },
    {
      "entropy": 6.309577941894531,
      "epoch": 0.9678406012409333,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004912972374866096,
      "loss": 6.0803,
      "mean_token_accuracy": 0.13912419006228446,
      "num_tokens": 22443778.0,
      "step": 11075
    },
    {
      "entropy": 6.375659799575805,
      "epoch": 0.968277549593638,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004912886527029579,
      "loss": 6.136,
      "mean_token_accuracy": 0.13936564698815346,
      "num_tokens": 22453423.0,
      "step": 11080
    },
    {
      "entropy": 6.421954202651977,
      "epoch": 0.9687144979463428,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004912800637707764,
      "loss": 6.0689,
      "mean_token_accuracy": 0.14664240404963494,
      "num_tokens": 22464067.0,
      "step": 11085
    },
    {
      "entropy": 6.229804182052613,
      "epoch": 0.9691514462990475,
      "grad_norm": 3.796875,
      "learning_rate": 0.00049127147069023,
      "loss": 6.0591,
      "mean_token_accuracy": 0.14649880304932594,
      "num_tokens": 22474332.0,
      "step": 11090
    },
    {
      "entropy": 6.322487878799438,
      "epoch": 0.9695883946517522,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004912628734614834,
      "loss": 6.0912,
      "mean_token_accuracy": 0.14226026684045792,
      "num_tokens": 22485304.0,
      "step": 11095
    },
    {
      "entropy": 6.319126844406128,
      "epoch": 0.9700253430044569,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004912542720847016,
      "loss": 6.007,
      "mean_token_accuracy": 0.1463754341006279,
      "num_tokens": 22494447.0,
      "step": 11100
    },
    {
      "entropy": 6.316814422607422,
      "epoch": 0.9704622913571616,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004912456665600495,
      "loss": 6.0467,
      "mean_token_accuracy": 0.14602853655815123,
      "num_tokens": 22503936.0,
      "step": 11105
    },
    {
      "entropy": 6.306980228424072,
      "epoch": 0.9708992397098662,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004912370568876923,
      "loss": 6.005,
      "mean_token_accuracy": 0.1438455268740654,
      "num_tokens": 22513960.0,
      "step": 11110
    },
    {
      "entropy": 6.299431705474854,
      "epoch": 0.971336188062571,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004912284430677949,
      "loss": 6.1,
      "mean_token_accuracy": 0.14221810847520827,
      "num_tokens": 22522309.0,
      "step": 11115
    },
    {
      "entropy": 6.284624433517456,
      "epoch": 0.9717731364152757,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004912198251005228,
      "loss": 6.0745,
      "mean_token_accuracy": 0.13813216611742973,
      "num_tokens": 22533510.0,
      "step": 11120
    },
    {
      "entropy": 6.245187091827392,
      "epoch": 0.9722100847679804,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004912112029860411,
      "loss": 6.0447,
      "mean_token_accuracy": 0.13898858428001404,
      "num_tokens": 22543644.0,
      "step": 11125
    },
    {
      "entropy": 6.372382164001465,
      "epoch": 0.9726470331206851,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004912025767245153,
      "loss": 6.1726,
      "mean_token_accuracy": 0.1314910337328911,
      "num_tokens": 22553475.0,
      "step": 11130
    },
    {
      "entropy": 6.328226852416992,
      "epoch": 0.9730839814733898,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004911939463161108,
      "loss": 6.0153,
      "mean_token_accuracy": 0.14131451025605202,
      "num_tokens": 22563190.0,
      "step": 11135
    },
    {
      "entropy": 6.291207790374756,
      "epoch": 0.9735209298260945,
      "grad_norm": 4.0,
      "learning_rate": 0.000491185311760993,
      "loss": 6.044,
      "mean_token_accuracy": 0.14799989089369775,
      "num_tokens": 22572069.0,
      "step": 11140
    },
    {
      "entropy": 6.3459868907928465,
      "epoch": 0.9739578781787993,
      "grad_norm": 4.75,
      "learning_rate": 0.0004911766730593276,
      "loss": 6.1419,
      "mean_token_accuracy": 0.13061940670013428,
      "num_tokens": 22581597.0,
      "step": 11145
    },
    {
      "entropy": 6.2907146453857425,
      "epoch": 0.974394826531504,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004911680302112806,
      "loss": 6.0522,
      "mean_token_accuracy": 0.1439785972237587,
      "num_tokens": 22591486.0,
      "step": 11150
    },
    {
      "entropy": 6.316544771194458,
      "epoch": 0.9748317748842087,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004911593832170173,
      "loss": 6.095,
      "mean_token_accuracy": 0.13291386589407922,
      "num_tokens": 22601615.0,
      "step": 11155
    },
    {
      "entropy": 6.339579391479492,
      "epoch": 0.9752687232369134,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004911507320767038,
      "loss": 6.0952,
      "mean_token_accuracy": 0.1390828087925911,
      "num_tokens": 22611397.0,
      "step": 11160
    },
    {
      "entropy": 6.275540828704834,
      "epoch": 0.9757056715896181,
      "grad_norm": 6.25,
      "learning_rate": 0.000491142076790506,
      "loss": 6.0567,
      "mean_token_accuracy": 0.1440751150250435,
      "num_tokens": 22621269.0,
      "step": 11165
    },
    {
      "entropy": 6.348693180084228,
      "epoch": 0.9761426199423228,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004911334173585899,
      "loss": 6.0713,
      "mean_token_accuracy": 0.14052906781435012,
      "num_tokens": 22631782.0,
      "step": 11170
    },
    {
      "entropy": 6.311023807525634,
      "epoch": 0.9765795682950276,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004911247537811216,
      "loss": 6.1343,
      "mean_token_accuracy": 0.13727387338876723,
      "num_tokens": 22642240.0,
      "step": 11175
    },
    {
      "entropy": 6.3354864597320555,
      "epoch": 0.9770165166477323,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004911160860582674,
      "loss": 6.0783,
      "mean_token_accuracy": 0.137482201308012,
      "num_tokens": 22651977.0,
      "step": 11180
    },
    {
      "entropy": 6.3712903499603275,
      "epoch": 0.977453465000437,
      "grad_norm": 5.40625,
      "learning_rate": 0.0004911074141901933,
      "loss": 6.1095,
      "mean_token_accuracy": 0.13988933116197586,
      "num_tokens": 22662871.0,
      "step": 11185
    },
    {
      "entropy": 6.295696210861206,
      "epoch": 0.9778904133531416,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004910987381770657,
      "loss": 6.058,
      "mean_token_accuracy": 0.14654023200273514,
      "num_tokens": 22672083.0,
      "step": 11190
    },
    {
      "entropy": 6.27540168762207,
      "epoch": 0.9783273617058463,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004910900580190512,
      "loss": 6.0738,
      "mean_token_accuracy": 0.14009279012680054,
      "num_tokens": 22681247.0,
      "step": 11195
    },
    {
      "entropy": 6.317129278182984,
      "epoch": 0.9787643100585511,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004910813737163161,
      "loss": 6.1223,
      "mean_token_accuracy": 0.13929112553596495,
      "num_tokens": 22691231.0,
      "step": 11200
    },
    {
      "entropy": 6.310454225540161,
      "epoch": 0.9792012584112558,
      "grad_norm": 2.40625,
      "learning_rate": 0.000491072685269027,
      "loss": 6.0961,
      "mean_token_accuracy": 0.13486940711736678,
      "num_tokens": 22700712.0,
      "step": 11205
    },
    {
      "entropy": 6.366319370269776,
      "epoch": 0.9796382067639605,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004910639926773506,
      "loss": 6.2025,
      "mean_token_accuracy": 0.13474841490387918,
      "num_tokens": 22710961.0,
      "step": 11210
    },
    {
      "entropy": 6.3684430599212645,
      "epoch": 0.9800751551166652,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004910552959414538,
      "loss": 6.0657,
      "mean_token_accuracy": 0.14733427613973618,
      "num_tokens": 22720450.0,
      "step": 11215
    },
    {
      "entropy": 6.25629506111145,
      "epoch": 0.9805121034693699,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004910465950615029,
      "loss": 6.0292,
      "mean_token_accuracy": 0.14489766582846642,
      "num_tokens": 22730337.0,
      "step": 11220
    },
    {
      "entropy": 6.290736865997315,
      "epoch": 0.9809490518220746,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004910378900376651,
      "loss": 6.122,
      "mean_token_accuracy": 0.13638080954551696,
      "num_tokens": 22742379.0,
      "step": 11225
    },
    {
      "entropy": 6.362155103683472,
      "epoch": 0.9813860001747794,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004910291808701077,
      "loss": 6.1557,
      "mean_token_accuracy": 0.13226249143481256,
      "num_tokens": 22753176.0,
      "step": 11230
    },
    {
      "entropy": 6.296923303604126,
      "epoch": 0.9818229485274841,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004910204675589972,
      "loss": 6.1057,
      "mean_token_accuracy": 0.1391991712152958,
      "num_tokens": 22763263.0,
      "step": 11235
    },
    {
      "entropy": 6.272580957412719,
      "epoch": 0.9822598968801888,
      "grad_norm": 5.125,
      "learning_rate": 0.0004910117501045009,
      "loss": 6.1003,
      "mean_token_accuracy": 0.14199652522802353,
      "num_tokens": 22774630.0,
      "step": 11240
    },
    {
      "entropy": 6.358653783798218,
      "epoch": 0.9826968452328935,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004910030285067861,
      "loss": 6.0989,
      "mean_token_accuracy": 0.13960972279310227,
      "num_tokens": 22785027.0,
      "step": 11245
    },
    {
      "entropy": 6.361986398696899,
      "epoch": 0.9831337935855982,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004909943027660201,
      "loss": 6.1269,
      "mean_token_accuracy": 0.13264269456267358,
      "num_tokens": 22796399.0,
      "step": 11250
    },
    {
      "entropy": 6.305996322631836,
      "epoch": 0.9835707419383029,
      "grad_norm": 2.640625,
      "learning_rate": 0.00049098557288237,
      "loss": 6.0767,
      "mean_token_accuracy": 0.13587312400341034,
      "num_tokens": 22806352.0,
      "step": 11255
    },
    {
      "entropy": 6.295252418518066,
      "epoch": 0.9840076902910077,
      "grad_norm": 7.59375,
      "learning_rate": 0.0004909768388560036,
      "loss": 6.1449,
      "mean_token_accuracy": 0.13461200669407844,
      "num_tokens": 22818369.0,
      "step": 11260
    },
    {
      "entropy": 6.293525075912475,
      "epoch": 0.9844446386437123,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004909681006870882,
      "loss": 6.0358,
      "mean_token_accuracy": 0.14125941097736358,
      "num_tokens": 22828063.0,
      "step": 11265
    },
    {
      "entropy": 6.312255620956421,
      "epoch": 0.984881586996417,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004909593583757915,
      "loss": 6.124,
      "mean_token_accuracy": 0.13259077668190003,
      "num_tokens": 22839265.0,
      "step": 11270
    },
    {
      "entropy": 6.278290557861328,
      "epoch": 0.9853185353491217,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004909506119222812,
      "loss": 6.0695,
      "mean_token_accuracy": 0.14639016538858413,
      "num_tokens": 22848501.0,
      "step": 11275
    },
    {
      "entropy": 6.242041540145874,
      "epoch": 0.9857554837018264,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004909418613267248,
      "loss": 6.0004,
      "mean_token_accuracy": 0.14201925843954086,
      "num_tokens": 22859176.0,
      "step": 11280
    },
    {
      "entropy": 6.261257934570312,
      "epoch": 0.9861924320545311,
      "grad_norm": 3.0,
      "learning_rate": 0.0004909331065892905,
      "loss": 6.0119,
      "mean_token_accuracy": 0.14795347377657891,
      "num_tokens": 22869312.0,
      "step": 11285
    },
    {
      "entropy": 6.349720048904419,
      "epoch": 0.9866293804072359,
      "grad_norm": 4.6875,
      "learning_rate": 0.000490924347710146,
      "loss": 6.1349,
      "mean_token_accuracy": 0.13793745189905166,
      "num_tokens": 22878699.0,
      "step": 11290
    },
    {
      "entropy": 6.333830451965332,
      "epoch": 0.9870663287599406,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004909155846894595,
      "loss": 6.1782,
      "mean_token_accuracy": 0.1386388584971428,
      "num_tokens": 22889147.0,
      "step": 11295
    },
    {
      "entropy": 6.3388330936431885,
      "epoch": 0.9875032771126453,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004909068175273989,
      "loss": 6.0628,
      "mean_token_accuracy": 0.14102457389235495,
      "num_tokens": 22899085.0,
      "step": 11300
    },
    {
      "entropy": 6.338387870788575,
      "epoch": 0.98794022546535,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004908980462241323,
      "loss": 6.1879,
      "mean_token_accuracy": 0.13312806710600852,
      "num_tokens": 22909924.0,
      "step": 11305
    },
    {
      "entropy": 6.392333030700684,
      "epoch": 0.9883771738180547,
      "grad_norm": 4.875,
      "learning_rate": 0.0004908892707798281,
      "loss": 6.1711,
      "mean_token_accuracy": 0.13665535673499107,
      "num_tokens": 22920426.0,
      "step": 11310
    },
    {
      "entropy": 6.336014938354492,
      "epoch": 0.9888141221707594,
      "grad_norm": 5.125,
      "learning_rate": 0.0004908804911946546,
      "loss": 6.0681,
      "mean_token_accuracy": 0.13778116554021835,
      "num_tokens": 22929929.0,
      "step": 11315
    },
    {
      "entropy": 6.2337281703948975,
      "epoch": 0.9892510705234642,
      "grad_norm": 6.4375,
      "learning_rate": 0.0004908717074687801,
      "loss": 6.0179,
      "mean_token_accuracy": 0.15140015557408332,
      "num_tokens": 22940102.0,
      "step": 11320
    },
    {
      "entropy": 6.245944690704346,
      "epoch": 0.9896880188761689,
      "grad_norm": 4.125,
      "learning_rate": 0.0004908629196023732,
      "loss": 6.0519,
      "mean_token_accuracy": 0.15104130730032922,
      "num_tokens": 22949490.0,
      "step": 11325
    },
    {
      "entropy": 6.323406028747558,
      "epoch": 0.9901249672288736,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004908541275956024,
      "loss": 6.0001,
      "mean_token_accuracy": 0.14379229098558427,
      "num_tokens": 22958119.0,
      "step": 11330
    },
    {
      "entropy": 6.360242700576782,
      "epoch": 0.9905619155815782,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004908453314486363,
      "loss": 6.0786,
      "mean_token_accuracy": 0.13511401116847993,
      "num_tokens": 22969447.0,
      "step": 11335
    },
    {
      "entropy": 6.274440288543701,
      "epoch": 0.9909988639342829,
      "grad_norm": 5.25,
      "learning_rate": 0.0004908365311616437,
      "loss": 5.9465,
      "mean_token_accuracy": 0.1467752128839493,
      "num_tokens": 22978925.0,
      "step": 11340
    },
    {
      "entropy": 6.299304103851318,
      "epoch": 0.9914358122869876,
      "grad_norm": 5.34375,
      "learning_rate": 0.0004908277267347931,
      "loss": 6.0968,
      "mean_token_accuracy": 0.144174288213253,
      "num_tokens": 22988861.0,
      "step": 11345
    },
    {
      "entropy": 6.317723608016967,
      "epoch": 0.9918727606396924,
      "grad_norm": 20.75,
      "learning_rate": 0.0004908189181682538,
      "loss": 6.102,
      "mean_token_accuracy": 0.13930935636162758,
      "num_tokens": 22998700.0,
      "step": 11350
    },
    {
      "entropy": 6.352739953994751,
      "epoch": 0.9923097089923971,
      "grad_norm": 5.78125,
      "learning_rate": 0.0004908101054621946,
      "loss": 6.037,
      "mean_token_accuracy": 0.1433849774301052,
      "num_tokens": 23009095.0,
      "step": 11355
    },
    {
      "entropy": 6.268067789077759,
      "epoch": 0.9927466573451018,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004908012886167845,
      "loss": 6.0544,
      "mean_token_accuracy": 0.1435396946966648,
      "num_tokens": 23019484.0,
      "step": 11360
    },
    {
      "entropy": 6.227122020721436,
      "epoch": 0.9931836056978065,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004907924676321924,
      "loss": 6.0402,
      "mean_token_accuracy": 0.1408992163836956,
      "num_tokens": 23029867.0,
      "step": 11365
    },
    {
      "entropy": 6.2543541431427006,
      "epoch": 0.9936205540505112,
      "grad_norm": 4.25,
      "learning_rate": 0.000490783642508588,
      "loss": 6.0106,
      "mean_token_accuracy": 0.15098831951618194,
      "num_tokens": 23040179.0,
      "step": 11370
    },
    {
      "entropy": 6.328528261184692,
      "epoch": 0.9940575024032159,
      "grad_norm": 5.09375,
      "learning_rate": 0.00049077481324614,
      "loss": 6.0086,
      "mean_token_accuracy": 0.1476857304573059,
      "num_tokens": 23050389.0,
      "step": 11375
    },
    {
      "entropy": 6.21253023147583,
      "epoch": 0.9944944507559207,
      "grad_norm": 6.78125,
      "learning_rate": 0.000490765979845018,
      "loss": 6.0424,
      "mean_token_accuracy": 0.14085501357913016,
      "num_tokens": 23059984.0,
      "step": 11380
    },
    {
      "entropy": 6.30299940109253,
      "epoch": 0.9949313991086254,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004907571423053917,
      "loss": 6.0592,
      "mean_token_accuracy": 0.14175608158111572,
      "num_tokens": 23069828.0,
      "step": 11385
    },
    {
      "entropy": 6.323674297332763,
      "epoch": 0.9953683474613301,
      "grad_norm": 6.0625,
      "learning_rate": 0.0004907483006274302,
      "loss": 6.0951,
      "mean_token_accuracy": 0.13630429282784462,
      "num_tokens": 23079223.0,
      "step": 11390
    },
    {
      "entropy": 6.330858898162842,
      "epoch": 0.9958052958140348,
      "grad_norm": 4.625,
      "learning_rate": 0.0004907394548113033,
      "loss": 6.1084,
      "mean_token_accuracy": 0.13805128633975983,
      "num_tokens": 23088540.0,
      "step": 11395
    },
    {
      "entropy": 6.26229248046875,
      "epoch": 0.9962422441667395,
      "grad_norm": 5.625,
      "learning_rate": 0.0004907306048571806,
      "loss": 6.1571,
      "mean_token_accuracy": 0.13580722212791443,
      "num_tokens": 23098617.0,
      "step": 11400
    },
    {
      "entropy": 6.290054988861084,
      "epoch": 0.9966791925194443,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004907217507652317,
      "loss": 6.0884,
      "mean_token_accuracy": 0.14138040095567703,
      "num_tokens": 23109702.0,
      "step": 11405
    },
    {
      "entropy": 6.351577568054199,
      "epoch": 0.997116140872149,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004907128925356268,
      "loss": 6.1072,
      "mean_token_accuracy": 0.1415163680911064,
      "num_tokens": 23120295.0,
      "step": 11410
    },
    {
      "entropy": 6.355620288848877,
      "epoch": 0.9975530892248536,
      "grad_norm": 7.4375,
      "learning_rate": 0.0004907040301685355,
      "loss": 6.1005,
      "mean_token_accuracy": 0.14062532410025597,
      "num_tokens": 23131999.0,
      "step": 11415
    },
    {
      "entropy": 6.332456493377686,
      "epoch": 0.9979900375775583,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004906951636641279,
      "loss": 6.1007,
      "mean_token_accuracy": 0.1434568651020527,
      "num_tokens": 23141060.0,
      "step": 11420
    },
    {
      "entropy": 6.272871828079223,
      "epoch": 0.998426985930263,
      "grad_norm": 4.25,
      "learning_rate": 0.000490686293022574,
      "loss": 6.0738,
      "mean_token_accuracy": 0.14589928239583969,
      "num_tokens": 23151607.0,
      "step": 11425
    },
    {
      "entropy": 6.310302829742431,
      "epoch": 0.9988639342829677,
      "grad_norm": 4.03125,
      "learning_rate": 0.000490677418244044,
      "loss": 6.1292,
      "mean_token_accuracy": 0.13410354033112526,
      "num_tokens": 23161892.0,
      "step": 11430
    },
    {
      "entropy": 6.352420711517334,
      "epoch": 0.9993008826356725,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004906685393287081,
      "loss": 6.1236,
      "mean_token_accuracy": 0.1383567027747631,
      "num_tokens": 23172112.0,
      "step": 11435
    },
    {
      "entropy": 6.3306403160095215,
      "epoch": 0.9997378309883772,
      "grad_norm": 7.90625,
      "learning_rate": 0.0004906596562767367,
      "loss": 6.012,
      "mean_token_accuracy": 0.1421718865633011,
      "num_tokens": 23181435.0,
      "step": 11440
    },
    {
      "entropy": 6.250259494781494,
      "epoch": 1.0001747793410818,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004906507690883,
      "loss": 5.9755,
      "mean_token_accuracy": 0.14648089706897735,
      "num_tokens": 23191084.0,
      "step": 11445
    },
    {
      "entropy": 6.271656703948975,
      "epoch": 1.0006117276937867,
      "grad_norm": 5.15625,
      "learning_rate": 0.0004906418777635685,
      "loss": 6.0847,
      "mean_token_accuracy": 0.1350153349339962,
      "num_tokens": 23202372.0,
      "step": 11450
    },
    {
      "entropy": 6.355922269821167,
      "epoch": 1.0010486760464914,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004906329823027129,
      "loss": 6.0728,
      "mean_token_accuracy": 0.13810350894927978,
      "num_tokens": 23214113.0,
      "step": 11455
    },
    {
      "entropy": 6.3330892562866214,
      "epoch": 1.001485624399196,
      "grad_norm": 6.0625,
      "learning_rate": 0.0004906240827059036,
      "loss": 5.9893,
      "mean_token_accuracy": 0.14305231273174285,
      "num_tokens": 23223562.0,
      "step": 11460
    },
    {
      "entropy": 6.338467073440552,
      "epoch": 1.0019225727519008,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004906151789733114,
      "loss": 6.0706,
      "mean_token_accuracy": 0.13875445052981378,
      "num_tokens": 23232901.0,
      "step": 11465
    },
    {
      "entropy": 6.365130805969239,
      "epoch": 1.0023595211046055,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004906062711051073,
      "loss": 6.0855,
      "mean_token_accuracy": 0.13602058216929436,
      "num_tokens": 23242869.0,
      "step": 11470
    },
    {
      "entropy": 6.325608825683593,
      "epoch": 1.0027964694573102,
      "grad_norm": 4.25,
      "learning_rate": 0.0004905973591014617,
      "loss": 5.9817,
      "mean_token_accuracy": 0.14721679762005807,
      "num_tokens": 23253653.0,
      "step": 11475
    },
    {
      "entropy": 6.274672746658325,
      "epoch": 1.0032334178100148,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004905884429625459,
      "loss": 6.0123,
      "mean_token_accuracy": 0.14675771445035934,
      "num_tokens": 23263530.0,
      "step": 11480
    },
    {
      "entropy": 6.257175397872925,
      "epoch": 1.0036703661627195,
      "grad_norm": 5.84375,
      "learning_rate": 0.0004905795226885308,
      "loss": 6.0476,
      "mean_token_accuracy": 0.14138330221176149,
      "num_tokens": 23274946.0,
      "step": 11485
    },
    {
      "entropy": 6.299904775619507,
      "epoch": 1.0041073145154242,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004905705982795874,
      "loss": 6.0505,
      "mean_token_accuracy": 0.1501994363963604,
      "num_tokens": 23283811.0,
      "step": 11490
    },
    {
      "entropy": 6.2783280372619625,
      "epoch": 1.004544262868129,
      "grad_norm": 4.03125,
      "learning_rate": 0.000490561669735887,
      "loss": 6.0775,
      "mean_token_accuracy": 0.13852036818861962,
      "num_tokens": 23294554.0,
      "step": 11495
    },
    {
      "entropy": 6.190992212295532,
      "epoch": 1.0049812112208336,
      "grad_norm": 6.1875,
      "learning_rate": 0.0004905527370576009,
      "loss": 5.9443,
      "mean_token_accuracy": 0.14840464368462564,
      "num_tokens": 23304245.0,
      "step": 11500
    },
    {
      "entropy": 6.382401371002198,
      "epoch": 1.0054181595735383,
      "grad_norm": 3.25,
      "learning_rate": 0.0004905438002449004,
      "loss": 6.0987,
      "mean_token_accuracy": 0.12784476280212403,
      "num_tokens": 23315169.0,
      "step": 11505
    },
    {
      "entropy": 6.356419610977173,
      "epoch": 1.0058551079262432,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004905348592979567,
      "loss": 6.0259,
      "mean_token_accuracy": 0.14320833757519721,
      "num_tokens": 23325528.0,
      "step": 11510
    },
    {
      "entropy": 6.280652570724487,
      "epoch": 1.006292056278948,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004905259142169415,
      "loss": 6.1178,
      "mean_token_accuracy": 0.13942288905382155,
      "num_tokens": 23336744.0,
      "step": 11515
    },
    {
      "entropy": 6.289211463928223,
      "epoch": 1.0067290046316526,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004905169650020264,
      "loss": 5.9894,
      "mean_token_accuracy": 0.1478257656097412,
      "num_tokens": 23346092.0,
      "step": 11520
    },
    {
      "entropy": 6.258255577087402,
      "epoch": 1.0071659529843573,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004905080116533831,
      "loss": 6.034,
      "mean_token_accuracy": 0.1397782027721405,
      "num_tokens": 23357326.0,
      "step": 11525
    },
    {
      "entropy": 6.243620443344116,
      "epoch": 1.007602901337062,
      "grad_norm": 2.875,
      "learning_rate": 0.000490499054171183,
      "loss": 6.0413,
      "mean_token_accuracy": 0.144442218542099,
      "num_tokens": 23367789.0,
      "step": 11530
    },
    {
      "entropy": 6.309927701950073,
      "epoch": 1.0080398496897667,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004904900925555982,
      "loss": 6.0172,
      "mean_token_accuracy": 0.1435712419450283,
      "num_tokens": 23377499.0,
      "step": 11535
    },
    {
      "entropy": 6.352451944351197,
      "epoch": 1.0084767980424714,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004904811268068005,
      "loss": 6.0157,
      "mean_token_accuracy": 0.14252193495631219,
      "num_tokens": 23387867.0,
      "step": 11540
    },
    {
      "entropy": 6.283497905731201,
      "epoch": 1.008913746395176,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004904721569249618,
      "loss": 6.0909,
      "mean_token_accuracy": 0.13998097479343413,
      "num_tokens": 23398179.0,
      "step": 11545
    },
    {
      "entropy": 6.3255054473876955,
      "epoch": 1.0093506947478808,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004904631829102543,
      "loss": 6.0889,
      "mean_token_accuracy": 0.138653901219368,
      "num_tokens": 23407407.0,
      "step": 11550
    },
    {
      "entropy": 6.2761577606201175,
      "epoch": 1.0097876431005854,
      "grad_norm": 3.5,
      "learning_rate": 0.0004904542047628501,
      "loss": 5.9389,
      "mean_token_accuracy": 0.146822252869606,
      "num_tokens": 23417324.0,
      "step": 11555
    },
    {
      "entropy": 6.311769151687622,
      "epoch": 1.0102245914532901,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004904452224829212,
      "loss": 6.0538,
      "mean_token_accuracy": 0.1456182509660721,
      "num_tokens": 23427157.0,
      "step": 11560
    },
    {
      "entropy": 6.299171495437622,
      "epoch": 1.0106615398059948,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004904362360706403,
      "loss": 6.0695,
      "mean_token_accuracy": 0.13748525828123093,
      "num_tokens": 23437712.0,
      "step": 11565
    },
    {
      "entropy": 6.331509113311768,
      "epoch": 1.0110984881586997,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004904272455261793,
      "loss": 5.9941,
      "mean_token_accuracy": 0.1442384146153927,
      "num_tokens": 23447949.0,
      "step": 11570
    },
    {
      "entropy": 6.263286304473877,
      "epoch": 1.0115354365114044,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004904182508497109,
      "loss": 5.974,
      "mean_token_accuracy": 0.14125097617506982,
      "num_tokens": 23457319.0,
      "step": 11575
    },
    {
      "entropy": 6.341108703613282,
      "epoch": 1.0119723848641091,
      "grad_norm": 4.5,
      "learning_rate": 0.0004904092520414075,
      "loss": 6.0982,
      "mean_token_accuracy": 0.13681202828884126,
      "num_tokens": 23468245.0,
      "step": 11580
    },
    {
      "entropy": 6.3376500606536865,
      "epoch": 1.0124093332168138,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004904002491014418,
      "loss": 6.0192,
      "mean_token_accuracy": 0.15169726610183715,
      "num_tokens": 23477049.0,
      "step": 11585
    },
    {
      "entropy": 6.263778305053711,
      "epoch": 1.0128462815695185,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004903912420299866,
      "loss": 6.1118,
      "mean_token_accuracy": 0.13571786433458327,
      "num_tokens": 23488574.0,
      "step": 11590
    },
    {
      "entropy": 6.319870948791504,
      "epoch": 1.0132832299222232,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004903822308272144,
      "loss": 6.0356,
      "mean_token_accuracy": 0.14312700927257538,
      "num_tokens": 23499282.0,
      "step": 11595
    },
    {
      "entropy": 6.300172519683838,
      "epoch": 1.0137201782749279,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004903732154932982,
      "loss": 6.0494,
      "mean_token_accuracy": 0.14974600151181222,
      "num_tokens": 23508777.0,
      "step": 11600
    },
    {
      "entropy": 6.234613084793091,
      "epoch": 1.0141571266276326,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004903641960284108,
      "loss": 6.0204,
      "mean_token_accuracy": 0.14372783079743384,
      "num_tokens": 23518818.0,
      "step": 11605
    },
    {
      "entropy": 6.233246326446533,
      "epoch": 1.0145940749803373,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004903551724327254,
      "loss": 6.0247,
      "mean_token_accuracy": 0.13878755420446395,
      "num_tokens": 23529668.0,
      "step": 11610
    },
    {
      "entropy": 6.33185887336731,
      "epoch": 1.015031023333042,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004903461447064147,
      "loss": 6.0552,
      "mean_token_accuracy": 0.1433694027364254,
      "num_tokens": 23540418.0,
      "step": 11615
    },
    {
      "entropy": 6.311904001235962,
      "epoch": 1.0154679716857467,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004903371128496523,
      "loss": 6.0257,
      "mean_token_accuracy": 0.14616795629262924,
      "num_tokens": 23550035.0,
      "step": 11620
    },
    {
      "entropy": 6.219232988357544,
      "epoch": 1.0159049200384516,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004903280768626113,
      "loss": 6.0474,
      "mean_token_accuracy": 0.13819248154759406,
      "num_tokens": 23559487.0,
      "step": 11625
    },
    {
      "entropy": 6.298178386688233,
      "epoch": 1.0163418683911563,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004903190367454648,
      "loss": 6.0004,
      "mean_token_accuracy": 0.14764697402715682,
      "num_tokens": 23569146.0,
      "step": 11630
    },
    {
      "entropy": 6.261106014251709,
      "epoch": 1.016778816743861,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004903099924983864,
      "loss": 5.9812,
      "mean_token_accuracy": 0.13940563425421715,
      "num_tokens": 23579458.0,
      "step": 11635
    },
    {
      "entropy": 6.285499858856201,
      "epoch": 1.0172157650965656,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004903009441215495,
      "loss": 5.9628,
      "mean_token_accuracy": 0.14242735803127288,
      "num_tokens": 23589416.0,
      "step": 11640
    },
    {
      "entropy": 6.237664890289307,
      "epoch": 1.0176527134492703,
      "grad_norm": 3.0,
      "learning_rate": 0.0004902918916151278,
      "loss": 6.0698,
      "mean_token_accuracy": 0.1439569339156151,
      "num_tokens": 23599588.0,
      "step": 11645
    },
    {
      "entropy": 6.305758762359619,
      "epoch": 1.018089661801975,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004902828349792947,
      "loss": 6.0619,
      "mean_token_accuracy": 0.1401212230324745,
      "num_tokens": 23610116.0,
      "step": 11650
    },
    {
      "entropy": 6.265839767456055,
      "epoch": 1.0185266101546797,
      "grad_norm": 2.859375,
      "learning_rate": 0.000490273774214224,
      "loss": 6.0059,
      "mean_token_accuracy": 0.14778365790843964,
      "num_tokens": 23619906.0,
      "step": 11655
    },
    {
      "entropy": 6.308337640762329,
      "epoch": 1.0189635585073844,
      "grad_norm": 20.125,
      "learning_rate": 0.0004902647093200896,
      "loss": 6.0291,
      "mean_token_accuracy": 0.1446967750787735,
      "num_tokens": 23630351.0,
      "step": 11660
    },
    {
      "entropy": 6.348279762268066,
      "epoch": 1.019400506860089,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004902556402970652,
      "loss": 6.0871,
      "mean_token_accuracy": 0.1381585605442524,
      "num_tokens": 23639696.0,
      "step": 11665
    },
    {
      "entropy": 6.295462226867675,
      "epoch": 1.0198374552127938,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004902465671453248,
      "loss": 6.0398,
      "mean_token_accuracy": 0.13871490880846976,
      "num_tokens": 23649388.0,
      "step": 11670
    },
    {
      "entropy": 6.297808980941772,
      "epoch": 1.0202744035654985,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004902374898650425,
      "loss": 6.0596,
      "mean_token_accuracy": 0.14254109635949136,
      "num_tokens": 23659125.0,
      "step": 11675
    },
    {
      "entropy": 6.409031629562378,
      "epoch": 1.0207113519182032,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004902284084563924,
      "loss": 6.0823,
      "mean_token_accuracy": 0.13974968418478967,
      "num_tokens": 23668817.0,
      "step": 11680
    },
    {
      "entropy": 6.2786108493804935,
      "epoch": 1.021148300270908,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004902193229195487,
      "loss": 6.029,
      "mean_token_accuracy": 0.1445097655057907,
      "num_tokens": 23679378.0,
      "step": 11685
    },
    {
      "entropy": 6.232928991317749,
      "epoch": 1.0215852486236128,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004902102332546855,
      "loss": 6.0172,
      "mean_token_accuracy": 0.138995860517025,
      "num_tokens": 23689400.0,
      "step": 11690
    },
    {
      "entropy": 6.374091577529907,
      "epoch": 1.0220221969763175,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004902011394619772,
      "loss": 6.1335,
      "mean_token_accuracy": 0.13373068049550058,
      "num_tokens": 23699503.0,
      "step": 11695
    },
    {
      "entropy": 6.358836698532104,
      "epoch": 1.0224591453290222,
      "grad_norm": 5.6875,
      "learning_rate": 0.0004901920415415985,
      "loss": 6.0792,
      "mean_token_accuracy": 0.13738290965557098,
      "num_tokens": 23710913.0,
      "step": 11700
    },
    {
      "entropy": 6.277243661880493,
      "epoch": 1.0228960936817268,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004901829394937234,
      "loss": 5.9836,
      "mean_token_accuracy": 0.14158999249339105,
      "num_tokens": 23720730.0,
      "step": 11705
    },
    {
      "entropy": 6.298077774047852,
      "epoch": 1.0233330420344315,
      "grad_norm": 3.1875,
      "learning_rate": 0.000490173833318527,
      "loss": 6.0495,
      "mean_token_accuracy": 0.13970689699053765,
      "num_tokens": 23730902.0,
      "step": 11710
    },
    {
      "entropy": 6.277667999267578,
      "epoch": 1.0237699903871362,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004901647230161837,
      "loss": 6.0802,
      "mean_token_accuracy": 0.1407393164932728,
      "num_tokens": 23741901.0,
      "step": 11715
    },
    {
      "entropy": 6.2825366973876955,
      "epoch": 1.024206938739841,
      "grad_norm": 15.9375,
      "learning_rate": 0.0004901556085868681,
      "loss": 5.9967,
      "mean_token_accuracy": 0.14154327660799026,
      "num_tokens": 23752029.0,
      "step": 11720
    },
    {
      "entropy": 6.228249406814575,
      "epoch": 1.0246438870925456,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004901464900307553,
      "loss": 5.9882,
      "mean_token_accuracy": 0.14955639690160752,
      "num_tokens": 23762125.0,
      "step": 11725
    },
    {
      "entropy": 6.267906188964844,
      "epoch": 1.0250808354452503,
      "grad_norm": 6.3125,
      "learning_rate": 0.0004901373673480201,
      "loss": 6.0444,
      "mean_token_accuracy": 0.1449929282069206,
      "num_tokens": 23772056.0,
      "step": 11730
    },
    {
      "entropy": 6.250489854812622,
      "epoch": 1.025517783797955,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004901282405388375,
      "loss": 5.9744,
      "mean_token_accuracy": 0.14939951449632644,
      "num_tokens": 23781441.0,
      "step": 11735
    },
    {
      "entropy": 6.264109039306641,
      "epoch": 1.0259547321506597,
      "grad_norm": 5.53125,
      "learning_rate": 0.0004901191096033825,
      "loss": 5.9757,
      "mean_token_accuracy": 0.14637915715575217,
      "num_tokens": 23791472.0,
      "step": 11740
    },
    {
      "entropy": 6.272996520996093,
      "epoch": 1.0263916805033646,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004901099745418302,
      "loss": 5.9915,
      "mean_token_accuracy": 0.14525725543498993,
      "num_tokens": 23801940.0,
      "step": 11745
    },
    {
      "entropy": 6.211066150665284,
      "epoch": 1.0268286288560693,
      "grad_norm": 23.25,
      "learning_rate": 0.0004901008353543559,
      "loss": 5.9397,
      "mean_token_accuracy": 0.14958102405071258,
      "num_tokens": 23811515.0,
      "step": 11750
    },
    {
      "entropy": 6.19355993270874,
      "epoch": 1.027265577208774,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004900916920411349,
      "loss": 5.9931,
      "mean_token_accuracy": 0.14347109645605088,
      "num_tokens": 23821871.0,
      "step": 11755
    },
    {
      "entropy": 6.217649698257446,
      "epoch": 1.0277025255614787,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004900825446023426,
      "loss": 5.9822,
      "mean_token_accuracy": 0.14646174460649491,
      "num_tokens": 23831677.0,
      "step": 11760
    },
    {
      "entropy": 6.317294502258301,
      "epoch": 1.0281394739141834,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004900733930381543,
      "loss": 6.0467,
      "mean_token_accuracy": 0.1394490122795105,
      "num_tokens": 23841415.0,
      "step": 11765
    },
    {
      "entropy": 6.325410652160644,
      "epoch": 1.028576422266888,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004900642373487455,
      "loss": 6.06,
      "mean_token_accuracy": 0.14292762130498887,
      "num_tokens": 23851809.0,
      "step": 11770
    },
    {
      "entropy": 6.312018632888794,
      "epoch": 1.0290133706195927,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004900550775342921,
      "loss": 5.9989,
      "mean_token_accuracy": 0.1379021294414997,
      "num_tokens": 23861981.0,
      "step": 11775
    },
    {
      "entropy": 6.278544235229492,
      "epoch": 1.0294503189722974,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004900459135949697,
      "loss": 5.9932,
      "mean_token_accuracy": 0.1436294950544834,
      "num_tokens": 23872437.0,
      "step": 11780
    },
    {
      "entropy": 6.277001905441284,
      "epoch": 1.0298872673250021,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004900367455309537,
      "loss": 6.0334,
      "mean_token_accuracy": 0.14337587356567383,
      "num_tokens": 23881650.0,
      "step": 11785
    },
    {
      "entropy": 6.259961318969727,
      "epoch": 1.0303242156777068,
      "grad_norm": 4.625,
      "learning_rate": 0.0004900275733424205,
      "loss": 6.0448,
      "mean_token_accuracy": 0.14538421779870986,
      "num_tokens": 23891631.0,
      "step": 11790
    },
    {
      "entropy": 6.266254615783692,
      "epoch": 1.0307611640304115,
      "grad_norm": 4.5,
      "learning_rate": 0.0004900183970295457,
      "loss": 5.9836,
      "mean_token_accuracy": 0.1478603810071945,
      "num_tokens": 23901725.0,
      "step": 11795
    },
    {
      "entropy": 6.286513996124268,
      "epoch": 1.0311981123831164,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004900092165925053,
      "loss": 6.0131,
      "mean_token_accuracy": 0.1434925228357315,
      "num_tokens": 23911850.0,
      "step": 11800
    },
    {
      "entropy": 6.299400663375854,
      "epoch": 1.0316350607358211,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004900000320314757,
      "loss": 6.1181,
      "mean_token_accuracy": 0.1349219873547554,
      "num_tokens": 23922571.0,
      "step": 11805
    },
    {
      "entropy": 6.252105474472046,
      "epoch": 1.0320720090885258,
      "grad_norm": 6.40625,
      "learning_rate": 0.0004899908433466326,
      "loss": 6.0048,
      "mean_token_accuracy": 0.14797542989253998,
      "num_tokens": 23932637.0,
      "step": 11810
    },
    {
      "entropy": 6.356252527236938,
      "epoch": 1.0325089574412305,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004899816505381525,
      "loss": 6.0987,
      "mean_token_accuracy": 0.13480637818574906,
      "num_tokens": 23943288.0,
      "step": 11815
    },
    {
      "entropy": 6.3145510196685795,
      "epoch": 1.0329459057939352,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004899724536062117,
      "loss": 5.9915,
      "mean_token_accuracy": 0.14358639791607858,
      "num_tokens": 23952853.0,
      "step": 11820
    },
    {
      "entropy": 6.1985148906707765,
      "epoch": 1.0333828541466399,
      "grad_norm": 3.375,
      "learning_rate": 0.0004899632525509867,
      "loss": 6.026,
      "mean_token_accuracy": 0.14297728165984153,
      "num_tokens": 23963111.0,
      "step": 11825
    },
    {
      "entropy": 6.261249446868897,
      "epoch": 1.0338198024993446,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004899540473726538,
      "loss": 5.9708,
      "mean_token_accuracy": 0.14381370022892953,
      "num_tokens": 23972468.0,
      "step": 11830
    },
    {
      "entropy": 6.327579736709595,
      "epoch": 1.0342567508520493,
      "grad_norm": 3.125,
      "learning_rate": 0.0004899448380713897,
      "loss": 5.9966,
      "mean_token_accuracy": 0.14456282705068588,
      "num_tokens": 23982567.0,
      "step": 11835
    },
    {
      "entropy": 6.231605577468872,
      "epoch": 1.034693699204754,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004899356246473709,
      "loss": 6.0116,
      "mean_token_accuracy": 0.14457231909036636,
      "num_tokens": 23993283.0,
      "step": 11840
    },
    {
      "entropy": 6.26971173286438,
      "epoch": 1.0351306475574586,
      "grad_norm": 2.75,
      "learning_rate": 0.0004899264071007741,
      "loss": 5.9759,
      "mean_token_accuracy": 0.14723847508430482,
      "num_tokens": 24004504.0,
      "step": 11845
    },
    {
      "entropy": 6.257011413574219,
      "epoch": 1.0355675959101633,
      "grad_norm": 5.78125,
      "learning_rate": 0.0004899171854317764,
      "loss": 6.0235,
      "mean_token_accuracy": 0.1439991980791092,
      "num_tokens": 24014557.0,
      "step": 11850
    },
    {
      "entropy": 6.232924365997315,
      "epoch": 1.036004544262868,
      "grad_norm": 5.125,
      "learning_rate": 0.0004899079596405543,
      "loss": 6.023,
      "mean_token_accuracy": 0.1417549394071102,
      "num_tokens": 24024495.0,
      "step": 11855
    },
    {
      "entropy": 6.264948940277099,
      "epoch": 1.036441492615573,
      "grad_norm": 3.234375,
      "learning_rate": 0.000489898729727285,
      "loss": 6.022,
      "mean_token_accuracy": 0.1407949671149254,
      "num_tokens": 24034984.0,
      "step": 11860
    },
    {
      "entropy": 6.318152666091919,
      "epoch": 1.0368784409682776,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004898894956921455,
      "loss": 5.9838,
      "mean_token_accuracy": 0.14549652710556985,
      "num_tokens": 24044379.0,
      "step": 11865
    },
    {
      "entropy": 6.320955896377564,
      "epoch": 1.0373153893209823,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004898802575353129,
      "loss": 6.0198,
      "mean_token_accuracy": 0.1421714425086975,
      "num_tokens": 24054133.0,
      "step": 11870
    },
    {
      "entropy": 6.194418573379517,
      "epoch": 1.037752337673687,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004898710152569642,
      "loss": 5.9525,
      "mean_token_accuracy": 0.14660960882902146,
      "num_tokens": 24064868.0,
      "step": 11875
    },
    {
      "entropy": 6.323087930679321,
      "epoch": 1.0381892860263917,
      "grad_norm": 4.9375,
      "learning_rate": 0.0004898617688572769,
      "loss": 6.0382,
      "mean_token_accuracy": 0.1431502014398575,
      "num_tokens": 24074977.0,
      "step": 11880
    },
    {
      "entropy": 6.276342248916626,
      "epoch": 1.0386262343790964,
      "grad_norm": 6.3125,
      "learning_rate": 0.0004898525183364284,
      "loss": 6.0807,
      "mean_token_accuracy": 0.14337649792432786,
      "num_tokens": 24085673.0,
      "step": 11885
    },
    {
      "entropy": 6.299204397201538,
      "epoch": 1.039063182731801,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004898432636945959,
      "loss": 6.0769,
      "mean_token_accuracy": 0.1408418744802475,
      "num_tokens": 24095465.0,
      "step": 11890
    },
    {
      "entropy": 6.248886346817017,
      "epoch": 1.0395001310845058,
      "grad_norm": 3.203125,
      "learning_rate": 0.000489834004931957,
      "loss": 6.0115,
      "mean_token_accuracy": 0.14345256835222245,
      "num_tokens": 24105576.0,
      "step": 11895
    },
    {
      "entropy": 6.271331453323365,
      "epoch": 1.0399370794372105,
      "grad_norm": 8.0,
      "learning_rate": 0.0004898247420486895,
      "loss": 5.9828,
      "mean_token_accuracy": 0.14389715492725372,
      "num_tokens": 24114998.0,
      "step": 11900
    },
    {
      "entropy": 6.326773548126221,
      "epoch": 1.0403740277899152,
      "grad_norm": 4.375,
      "learning_rate": 0.0004898154750449707,
      "loss": 6.0035,
      "mean_token_accuracy": 0.1425488144159317,
      "num_tokens": 24125059.0,
      "step": 11905
    },
    {
      "entropy": 6.30100450515747,
      "epoch": 1.0408109761426199,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004898062039209786,
      "loss": 6.0123,
      "mean_token_accuracy": 0.14239903390407563,
      "num_tokens": 24134062.0,
      "step": 11910
    },
    {
      "entropy": 6.296336030960083,
      "epoch": 1.0412479244953246,
      "grad_norm": 3.1875,
      "learning_rate": 0.000489796928676891,
      "loss": 6.0079,
      "mean_token_accuracy": 0.14153445586562158,
      "num_tokens": 24144016.0,
      "step": 11915
    },
    {
      "entropy": 6.315558624267578,
      "epoch": 1.0416848728480295,
      "grad_norm": 5.875,
      "learning_rate": 0.0004897876493128856,
      "loss": 6.0042,
      "mean_token_accuracy": 0.14538640677928924,
      "num_tokens": 24154337.0,
      "step": 11920
    },
    {
      "entropy": 6.237430667877197,
      "epoch": 1.0421218212007342,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004897783658291406,
      "loss": 5.9824,
      "mean_token_accuracy": 0.14965825974941255,
      "num_tokens": 24164440.0,
      "step": 11925
    },
    {
      "entropy": 6.262679862976074,
      "epoch": 1.0425587695534388,
      "grad_norm": 4.0,
      "learning_rate": 0.000489769078225834,
      "loss": 6.1171,
      "mean_token_accuracy": 0.13906511813402175,
      "num_tokens": 24174388.0,
      "step": 11930
    },
    {
      "entropy": 6.288280725479126,
      "epoch": 1.0429957179061435,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004897597865031441,
      "loss": 5.9834,
      "mean_token_accuracy": 0.14701861292123794,
      "num_tokens": 24183543.0,
      "step": 11935
    },
    {
      "entropy": 6.2797914981842045,
      "epoch": 1.0434326662588482,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004897504906612487,
      "loss": 6.0019,
      "mean_token_accuracy": 0.1461532950401306,
      "num_tokens": 24193847.0,
      "step": 11940
    },
    {
      "entropy": 6.240224933624267,
      "epoch": 1.043869614611553,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004897411907003265,
      "loss": 5.99,
      "mean_token_accuracy": 0.1445078432559967,
      "num_tokens": 24203072.0,
      "step": 11945
    },
    {
      "entropy": 6.27644829750061,
      "epoch": 1.0443065629642576,
      "grad_norm": 5.8125,
      "learning_rate": 0.0004897318866205558,
      "loss": 5.9479,
      "mean_token_accuracy": 0.1481520190834999,
      "num_tokens": 24213610.0,
      "step": 11950
    },
    {
      "entropy": 6.227467966079712,
      "epoch": 1.0447435113169623,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004897225784221149,
      "loss": 5.9893,
      "mean_token_accuracy": 0.14834156781435012,
      "num_tokens": 24223571.0,
      "step": 11955
    },
    {
      "entropy": 6.241715574264527,
      "epoch": 1.045180459669667,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004897132661051826,
      "loss": 6.0421,
      "mean_token_accuracy": 0.14348333775997163,
      "num_tokens": 24233867.0,
      "step": 11960
    },
    {
      "entropy": 6.219062471389771,
      "epoch": 1.0456174080223717,
      "grad_norm": 4.25,
      "learning_rate": 0.0004897039496699372,
      "loss": 6.0346,
      "mean_token_accuracy": 0.14338019862771034,
      "num_tokens": 24243853.0,
      "step": 11965
    },
    {
      "entropy": 6.267466068267822,
      "epoch": 1.0460543563750764,
      "grad_norm": 6.21875,
      "learning_rate": 0.0004896946291165575,
      "loss": 6.0126,
      "mean_token_accuracy": 0.1390203393995762,
      "num_tokens": 24254049.0,
      "step": 11970
    },
    {
      "entropy": 6.2667333602905275,
      "epoch": 1.0464913047277813,
      "grad_norm": 6.09375,
      "learning_rate": 0.0004896853044452224,
      "loss": 6.0034,
      "mean_token_accuracy": 0.141124863922596,
      "num_tokens": 24265166.0,
      "step": 11975
    },
    {
      "entropy": 6.295901870727539,
      "epoch": 1.046928253080486,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004896759756561106,
      "loss": 6.0016,
      "mean_token_accuracy": 0.14524534791707994,
      "num_tokens": 24274981.0,
      "step": 11980
    },
    {
      "entropy": 6.3488493919372555,
      "epoch": 1.0473652014331907,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004896666427494012,
      "loss": 6.0332,
      "mean_token_accuracy": 0.14379826486110686,
      "num_tokens": 24285009.0,
      "step": 11985
    },
    {
      "entropy": 6.339479970932007,
      "epoch": 1.0478021497858954,
      "grad_norm": 4.0,
      "learning_rate": 0.000489657305725273,
      "loss": 6.0859,
      "mean_token_accuracy": 0.13717985674738883,
      "num_tokens": 24294669.0,
      "step": 11990
    },
    {
      "entropy": 6.295821762084961,
      "epoch": 1.0482390981386,
      "grad_norm": 7.34375,
      "learning_rate": 0.0004896479645839052,
      "loss": 6.0375,
      "mean_token_accuracy": 0.14141014218330383,
      "num_tokens": 24304647.0,
      "step": 11995
    },
    {
      "entropy": 6.217085504531861,
      "epoch": 1.0486760464913047,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004896386193254771,
      "loss": 5.9264,
      "mean_token_accuracy": 0.15297415032982825,
      "num_tokens": 24314266.0,
      "step": 12000
    },
    {
      "epoch": 1.0486760464913047,
      "eval_entropy": 5.999692401490254,
      "eval_loss": 6.076780319213867,
      "eval_mean_token_accuracy": 0.1489904253108148,
      "eval_num_tokens": 24314266.0,
      "eval_runtime": 27.7839,
      "eval_samples_per_second": 1332.209,
      "eval_steps_per_second": 166.535,
      "step": 12000
    },
    {
      "entropy": 6.210977554321289,
      "epoch": 1.0491129948440094,
      "grad_norm": 4.9375,
      "learning_rate": 0.0004896292699501677,
      "loss": 6.0005,
      "mean_token_accuracy": 0.14765506982803345,
      "num_tokens": 24323770.0,
      "step": 12005
    },
    {
      "entropy": 6.238453388214111,
      "epoch": 1.0495499431967141,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004896199164581564,
      "loss": 5.9493,
      "mean_token_accuracy": 0.14840982407331466,
      "num_tokens": 24333438.0,
      "step": 12010
    },
    {
      "entropy": 6.2071136951446535,
      "epoch": 1.0499868915494188,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004896105588496227,
      "loss": 6.0238,
      "mean_token_accuracy": 0.14103174358606338,
      "num_tokens": 24342929.0,
      "step": 12015
    },
    {
      "entropy": 6.268410778045654,
      "epoch": 1.0504238399021235,
      "grad_norm": 7.53125,
      "learning_rate": 0.0004896011971247461,
      "loss": 6.0486,
      "mean_token_accuracy": 0.14271478950977326,
      "num_tokens": 24352617.0,
      "step": 12020
    },
    {
      "entropy": 6.272074270248413,
      "epoch": 1.0508607882548282,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004895918312837061,
      "loss": 5.9981,
      "mean_token_accuracy": 0.14984055310487748,
      "num_tokens": 24363940.0,
      "step": 12025
    },
    {
      "entropy": 6.296630907058716,
      "epoch": 1.051297736607533,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004895824613266822,
      "loss": 6.0505,
      "mean_token_accuracy": 0.1388652168214321,
      "num_tokens": 24373168.0,
      "step": 12030
    },
    {
      "entropy": 6.3406775951385494,
      "epoch": 1.0517346849602376,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004895730872538543,
      "loss": 6.0809,
      "mean_token_accuracy": 0.14456688985228539,
      "num_tokens": 24383067.0,
      "step": 12035
    },
    {
      "entropy": 6.31103835105896,
      "epoch": 1.0521716333129425,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004895637090654021,
      "loss": 5.976,
      "mean_token_accuracy": 0.150381837785244,
      "num_tokens": 24392734.0,
      "step": 12040
    },
    {
      "entropy": 6.292594623565674,
      "epoch": 1.0526085816656472,
      "grad_norm": 3.625,
      "learning_rate": 0.0004895543267615057,
      "loss": 6.0124,
      "mean_token_accuracy": 0.14562387019395828,
      "num_tokens": 24403970.0,
      "step": 12045
    },
    {
      "entropy": 6.282194471359253,
      "epoch": 1.0530455300183519,
      "grad_norm": 3.625,
      "learning_rate": 0.0004895449403423447,
      "loss": 6.0188,
      "mean_token_accuracy": 0.14213434532284736,
      "num_tokens": 24413614.0,
      "step": 12050
    },
    {
      "entropy": 6.2534795761108395,
      "epoch": 1.0534824783710566,
      "grad_norm": 5.5,
      "learning_rate": 0.0004895355498080994,
      "loss": 5.9637,
      "mean_token_accuracy": 0.15000785738229752,
      "num_tokens": 24424237.0,
      "step": 12055
    },
    {
      "entropy": 6.235092496871948,
      "epoch": 1.0539194267237613,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004895261551589498,
      "loss": 5.969,
      "mean_token_accuracy": 0.14795639961957932,
      "num_tokens": 24433220.0,
      "step": 12060
    },
    {
      "entropy": 6.35542402267456,
      "epoch": 1.054356375076466,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004895167563950763,
      "loss": 6.1012,
      "mean_token_accuracy": 0.1393493801355362,
      "num_tokens": 24444224.0,
      "step": 12065
    },
    {
      "entropy": 6.340588569641113,
      "epoch": 1.0547933234291706,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004895073535166588,
      "loss": 6.1103,
      "mean_token_accuracy": 0.13900990039110184,
      "num_tokens": 24454431.0,
      "step": 12070
    },
    {
      "entropy": 6.310595512390137,
      "epoch": 1.0552302717818753,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004894979465238779,
      "loss": 6.0024,
      "mean_token_accuracy": 0.14073515981435775,
      "num_tokens": 24464006.0,
      "step": 12075
    },
    {
      "entropy": 6.33417239189148,
      "epoch": 1.05566722013458,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004894885354169141,
      "loss": 5.9533,
      "mean_token_accuracy": 0.15179412961006164,
      "num_tokens": 24472706.0,
      "step": 12080
    },
    {
      "entropy": 6.246649599075317,
      "epoch": 1.0561041684872847,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004894791201959477,
      "loss": 5.986,
      "mean_token_accuracy": 0.15158465579152108,
      "num_tokens": 24482785.0,
      "step": 12085
    },
    {
      "entropy": 6.251727056503296,
      "epoch": 1.0565411168399894,
      "grad_norm": 4.25,
      "learning_rate": 0.0004894697008611594,
      "loss": 6.0668,
      "mean_token_accuracy": 0.1382058635354042,
      "num_tokens": 24493318.0,
      "step": 12090
    },
    {
      "entropy": 6.273067665100098,
      "epoch": 1.0569780651926943,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004894602774127299,
      "loss": 6.052,
      "mean_token_accuracy": 0.14220259413123132,
      "num_tokens": 24503780.0,
      "step": 12095
    },
    {
      "entropy": 6.359951162338257,
      "epoch": 1.057415013545399,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004894508498508399,
      "loss": 6.0769,
      "mean_token_accuracy": 0.13981142789125442,
      "num_tokens": 24513905.0,
      "step": 12100
    },
    {
      "entropy": 6.379530334472657,
      "epoch": 1.0578519618981037,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004894414181756702,
      "loss": 6.0501,
      "mean_token_accuracy": 0.14481110051274299,
      "num_tokens": 24524194.0,
      "step": 12105
    },
    {
      "entropy": 6.348235750198365,
      "epoch": 1.0582889102508084,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004894319823874017,
      "loss": 6.0467,
      "mean_token_accuracy": 0.14205388352274895,
      "num_tokens": 24533456.0,
      "step": 12110
    },
    {
      "entropy": 6.323317146301269,
      "epoch": 1.058725858603513,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004894225424862154,
      "loss": 6.039,
      "mean_token_accuracy": 0.1416662335395813,
      "num_tokens": 24544191.0,
      "step": 12115
    },
    {
      "entropy": 6.2541454315185545,
      "epoch": 1.0591628069562178,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004894130984722926,
      "loss": 5.987,
      "mean_token_accuracy": 0.1505594313144684,
      "num_tokens": 24554353.0,
      "step": 12120
    },
    {
      "entropy": 6.2783770084381105,
      "epoch": 1.0595997553089225,
      "grad_norm": 4.46875,
      "learning_rate": 0.000489403650345814,
      "loss": 6.0682,
      "mean_token_accuracy": 0.1421813227236271,
      "num_tokens": 24564313.0,
      "step": 12125
    },
    {
      "entropy": 6.305402183532715,
      "epoch": 1.0600367036616272,
      "grad_norm": 5.0625,
      "learning_rate": 0.0004893941981069612,
      "loss": 6.0775,
      "mean_token_accuracy": 0.14055947437882424,
      "num_tokens": 24573129.0,
      "step": 12130
    },
    {
      "entropy": 6.276437282562256,
      "epoch": 1.0604736520143319,
      "grad_norm": 5.53125,
      "learning_rate": 0.0004893847417559152,
      "loss": 5.946,
      "mean_token_accuracy": 0.15007732063531876,
      "num_tokens": 24582324.0,
      "step": 12135
    },
    {
      "entropy": 6.293022966384887,
      "epoch": 1.0609106003670365,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004893752812928576,
      "loss": 6.0881,
      "mean_token_accuracy": 0.13785800263285636,
      "num_tokens": 24593360.0,
      "step": 12140
    },
    {
      "entropy": 6.314513540267944,
      "epoch": 1.0613475487197412,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004893658167179697,
      "loss": 6.0156,
      "mean_token_accuracy": 0.1466860368847847,
      "num_tokens": 24603597.0,
      "step": 12145
    },
    {
      "entropy": 6.3471923828125,
      "epoch": 1.061784497072446,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004893563480314331,
      "loss": 5.9609,
      "mean_token_accuracy": 0.1479771092534065,
      "num_tokens": 24612426.0,
      "step": 12150
    },
    {
      "entropy": 6.296992683410645,
      "epoch": 1.0622214454251508,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004893468752334294,
      "loss": 5.9623,
      "mean_token_accuracy": 0.14350661113858224,
      "num_tokens": 24623031.0,
      "step": 12155
    },
    {
      "entropy": 6.21689624786377,
      "epoch": 1.0626583937778555,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004893373983241404,
      "loss": 6.0319,
      "mean_token_accuracy": 0.14374882876873016,
      "num_tokens": 24633763.0,
      "step": 12160
    },
    {
      "entropy": 6.262351369857788,
      "epoch": 1.0630953421305602,
      "grad_norm": 2.625,
      "learning_rate": 0.0004893279173037478,
      "loss": 6.0619,
      "mean_token_accuracy": 0.14036355316638946,
      "num_tokens": 24644340.0,
      "step": 12165
    },
    {
      "entropy": 6.351654386520385,
      "epoch": 1.063532290483265,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004893184321724335,
      "loss": 6.0598,
      "mean_token_accuracy": 0.14955881237983704,
      "num_tokens": 24654751.0,
      "step": 12170
    },
    {
      "entropy": 6.300664854049683,
      "epoch": 1.0639692388359696,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004893089429303792,
      "loss": 6.0079,
      "mean_token_accuracy": 0.1465013802051544,
      "num_tokens": 24665134.0,
      "step": 12175
    },
    {
      "entropy": 6.263440275192261,
      "epoch": 1.0644061871886743,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004892994495777672,
      "loss": 5.9862,
      "mean_token_accuracy": 0.1457593783736229,
      "num_tokens": 24673796.0,
      "step": 12180
    },
    {
      "entropy": 6.228956460952759,
      "epoch": 1.064843135541379,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004892899521147794,
      "loss": 5.9512,
      "mean_token_accuracy": 0.15418416857719422,
      "num_tokens": 24684104.0,
      "step": 12185
    },
    {
      "entropy": 6.20899510383606,
      "epoch": 1.0652800838940837,
      "grad_norm": 2.890625,
      "learning_rate": 0.000489280450541598,
      "loss": 5.9826,
      "mean_token_accuracy": 0.14698734059929847,
      "num_tokens": 24693994.0,
      "step": 12190
    },
    {
      "entropy": 6.271563529968262,
      "epoch": 1.0657170322467884,
      "grad_norm": 3.25,
      "learning_rate": 0.0004892709448584054,
      "loss": 6.0184,
      "mean_token_accuracy": 0.14393462836742402,
      "num_tokens": 24703815.0,
      "step": 12195
    },
    {
      "entropy": 6.330807304382324,
      "epoch": 1.066153980599493,
      "grad_norm": 5.25,
      "learning_rate": 0.0004892614350653837,
      "loss": 5.9606,
      "mean_token_accuracy": 0.1477448657155037,
      "num_tokens": 24713631.0,
      "step": 12200
    },
    {
      "entropy": 6.315654134750366,
      "epoch": 1.0665909289521978,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004892519211627156,
      "loss": 6.0183,
      "mean_token_accuracy": 0.14530732482671738,
      "num_tokens": 24723353.0,
      "step": 12205
    },
    {
      "entropy": 6.30722975730896,
      "epoch": 1.0670278773049025,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004892424031505832,
      "loss": 6.0002,
      "mean_token_accuracy": 0.14479120671749116,
      "num_tokens": 24733246.0,
      "step": 12210
    },
    {
      "entropy": 6.268536329269409,
      "epoch": 1.0674648256576074,
      "grad_norm": 4.375,
      "learning_rate": 0.0004892328810291693,
      "loss": 5.9778,
      "mean_token_accuracy": 0.15238874703645705,
      "num_tokens": 24743097.0,
      "step": 12215
    },
    {
      "entropy": 6.180049657821655,
      "epoch": 1.067901774010312,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004892233547986564,
      "loss": 5.9734,
      "mean_token_accuracy": 0.1458934023976326,
      "num_tokens": 24752618.0,
      "step": 12220
    },
    {
      "entropy": 6.237815809249878,
      "epoch": 1.0683387223630167,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004892138244592274,
      "loss": 6.034,
      "mean_token_accuracy": 0.1438584841787815,
      "num_tokens": 24763960.0,
      "step": 12225
    },
    {
      "entropy": 6.329831504821778,
      "epoch": 1.0687756707157214,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004892042900110649,
      "loss": 5.9239,
      "mean_token_accuracy": 0.1431748539209366,
      "num_tokens": 24773040.0,
      "step": 12230
    },
    {
      "entropy": 6.256713056564331,
      "epoch": 1.0692126190684261,
      "grad_norm": 4.6875,
      "learning_rate": 0.000489194751454352,
      "loss": 6.0426,
      "mean_token_accuracy": 0.14289514124393463,
      "num_tokens": 24783049.0,
      "step": 12235
    },
    {
      "entropy": 6.323098802566529,
      "epoch": 1.0696495674211308,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004891852087892714,
      "loss": 6.0221,
      "mean_token_accuracy": 0.14387214854359626,
      "num_tokens": 24794590.0,
      "step": 12240
    },
    {
      "entropy": 6.261320924758911,
      "epoch": 1.0700865157738355,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004891756620160064,
      "loss": 5.9544,
      "mean_token_accuracy": 0.14590869545936586,
      "num_tokens": 24805146.0,
      "step": 12245
    },
    {
      "entropy": 6.274057769775391,
      "epoch": 1.0705234641265402,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004891661111347399,
      "loss": 6.0418,
      "mean_token_accuracy": 0.14030179157853126,
      "num_tokens": 24815573.0,
      "step": 12250
    },
    {
      "entropy": 6.302560615539551,
      "epoch": 1.070960412479245,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004891565561456552,
      "loss": 5.963,
      "mean_token_accuracy": 0.14889843091368676,
      "num_tokens": 24825417.0,
      "step": 12255
    },
    {
      "entropy": 6.1854896068573,
      "epoch": 1.0713973608319496,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004891469970489355,
      "loss": 5.9585,
      "mean_token_accuracy": 0.15139433294534682,
      "num_tokens": 24835589.0,
      "step": 12260
    },
    {
      "entropy": 6.203954744338989,
      "epoch": 1.0718343091846543,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004891374338447642,
      "loss": 6.0801,
      "mean_token_accuracy": 0.1411905348300934,
      "num_tokens": 24846603.0,
      "step": 12265
    },
    {
      "entropy": 6.357881736755371,
      "epoch": 1.0722712575373592,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004891278665333247,
      "loss": 5.9925,
      "mean_token_accuracy": 0.14592426121234894,
      "num_tokens": 24856368.0,
      "step": 12270
    },
    {
      "entropy": 6.202588081359863,
      "epoch": 1.0727082058900639,
      "grad_norm": 9.0625,
      "learning_rate": 0.0004891182951148005,
      "loss": 5.9202,
      "mean_token_accuracy": 0.1481042355298996,
      "num_tokens": 24866322.0,
      "step": 12275
    },
    {
      "entropy": 6.169778442382812,
      "epoch": 1.0731451542427686,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004891087195893753,
      "loss": 6.0177,
      "mean_token_accuracy": 0.14288844019174576,
      "num_tokens": 24876686.0,
      "step": 12280
    },
    {
      "entropy": 6.2514019966125485,
      "epoch": 1.0735821025954733,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004890991399572325,
      "loss": 5.9559,
      "mean_token_accuracy": 0.1452287882566452,
      "num_tokens": 24886561.0,
      "step": 12285
    },
    {
      "entropy": 6.337167501449585,
      "epoch": 1.074019050948178,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004890895562185562,
      "loss": 5.9993,
      "mean_token_accuracy": 0.14349799528717994,
      "num_tokens": 24896010.0,
      "step": 12290
    },
    {
      "entropy": 6.338354682922363,
      "epoch": 1.0744559993008826,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004890799683735299,
      "loss": 6.039,
      "mean_token_accuracy": 0.14137075543403627,
      "num_tokens": 24904852.0,
      "step": 12295
    },
    {
      "entropy": 6.248547792434692,
      "epoch": 1.0748929476535873,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004890703764223377,
      "loss": 6.0124,
      "mean_token_accuracy": 0.1425559177994728,
      "num_tokens": 24914828.0,
      "step": 12300
    },
    {
      "entropy": 6.290818166732788,
      "epoch": 1.075329896006292,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004890607803651635,
      "loss": 6.061,
      "mean_token_accuracy": 0.13714486807584764,
      "num_tokens": 24925602.0,
      "step": 12305
    },
    {
      "entropy": 6.311762380599975,
      "epoch": 1.0757668443589967,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004890511802021914,
      "loss": 6.0463,
      "mean_token_accuracy": 0.1492523729801178,
      "num_tokens": 24935175.0,
      "step": 12310
    },
    {
      "entropy": 6.287031078338623,
      "epoch": 1.0762037927117014,
      "grad_norm": 5.375,
      "learning_rate": 0.0004890415759336053,
      "loss": 6.0453,
      "mean_token_accuracy": 0.13570917546749114,
      "num_tokens": 24945828.0,
      "step": 12315
    },
    {
      "entropy": 6.277022218704223,
      "epoch": 1.076640741064406,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004890319675595899,
      "loss": 6.0031,
      "mean_token_accuracy": 0.14471784234046936,
      "num_tokens": 24955400.0,
      "step": 12320
    },
    {
      "entropy": 6.27427339553833,
      "epoch": 1.0770776894171108,
      "grad_norm": 3.140625,
      "learning_rate": 0.000489022355080329,
      "loss": 6.0685,
      "mean_token_accuracy": 0.1393893390893936,
      "num_tokens": 24966079.0,
      "step": 12325
    },
    {
      "entropy": 6.235183048248291,
      "epoch": 1.0775146377698157,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004890127384960073,
      "loss": 5.9501,
      "mean_token_accuracy": 0.15111135840415954,
      "num_tokens": 24975090.0,
      "step": 12330
    },
    {
      "entropy": 6.36564564704895,
      "epoch": 1.0779515861225204,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004890031178068091,
      "loss": 6.0913,
      "mean_token_accuracy": 0.13623978868126868,
      "num_tokens": 24986068.0,
      "step": 12335
    },
    {
      "entropy": 6.296871709823608,
      "epoch": 1.078388534475225,
      "grad_norm": 3.0,
      "learning_rate": 0.000488993493012919,
      "loss": 6.0718,
      "mean_token_accuracy": 0.14114599376916886,
      "num_tokens": 24995697.0,
      "step": 12340
    },
    {
      "entropy": 6.271590995788574,
      "epoch": 1.0788254828279298,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004889838641145215,
      "loss": 6.0194,
      "mean_token_accuracy": 0.13509748503565788,
      "num_tokens": 25005602.0,
      "step": 12345
    },
    {
      "entropy": 6.264921474456787,
      "epoch": 1.0792624311806345,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004889742311118014,
      "loss": 6.0302,
      "mean_token_accuracy": 0.15182781219482422,
      "num_tokens": 25016407.0,
      "step": 12350
    },
    {
      "entropy": 6.330304479598999,
      "epoch": 1.0796993795333392,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004889645940049434,
      "loss": 6.0557,
      "mean_token_accuracy": 0.1422361195087433,
      "num_tokens": 25025889.0,
      "step": 12355
    },
    {
      "entropy": 6.277034854888916,
      "epoch": 1.0801363278860439,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004889549527941324,
      "loss": 5.9989,
      "mean_token_accuracy": 0.14634465128183366,
      "num_tokens": 25036383.0,
      "step": 12360
    },
    {
      "entropy": 6.300783681869507,
      "epoch": 1.0805732762387485,
      "grad_norm": 6.34375,
      "learning_rate": 0.0004889453074795534,
      "loss": 6.1046,
      "mean_token_accuracy": 0.1315123327076435,
      "num_tokens": 25045718.0,
      "step": 12365
    },
    {
      "entropy": 6.276469469070435,
      "epoch": 1.0810102245914532,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004889356580613911,
      "loss": 6.0889,
      "mean_token_accuracy": 0.14077974781394004,
      "num_tokens": 25056244.0,
      "step": 12370
    },
    {
      "entropy": 6.33950138092041,
      "epoch": 1.081447172944158,
      "grad_norm": 5.75,
      "learning_rate": 0.000488926004539831,
      "loss": 6.0175,
      "mean_token_accuracy": 0.14324121847748755,
      "num_tokens": 25065954.0,
      "step": 12375
    },
    {
      "entropy": 6.298659086227417,
      "epoch": 1.0818841212968626,
      "grad_norm": 4.125,
      "learning_rate": 0.0004889163469150578,
      "loss": 6.0513,
      "mean_token_accuracy": 0.14505369290709497,
      "num_tokens": 25075851.0,
      "step": 12380
    },
    {
      "entropy": 6.249526596069336,
      "epoch": 1.0823210696495673,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004889066851872571,
      "loss": 5.9975,
      "mean_token_accuracy": 0.14572824388742447,
      "num_tokens": 25085324.0,
      "step": 12385
    },
    {
      "entropy": 6.287007856369018,
      "epoch": 1.0827580180022722,
      "grad_norm": 2.890625,
      "learning_rate": 0.000488897019356614,
      "loss": 6.0652,
      "mean_token_accuracy": 0.14163421392440795,
      "num_tokens": 25096566.0,
      "step": 12390
    },
    {
      "entropy": 6.359086227416992,
      "epoch": 1.083194966354977,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004888873494233141,
      "loss": 6.1332,
      "mean_token_accuracy": 0.13606862500309944,
      "num_tokens": 25107642.0,
      "step": 12395
    },
    {
      "entropy": 6.32436900138855,
      "epoch": 1.0836319147076816,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004888776753875427,
      "loss": 6.0224,
      "mean_token_accuracy": 0.14621842056512832,
      "num_tokens": 25117488.0,
      "step": 12400
    },
    {
      "entropy": 6.269311714172363,
      "epoch": 1.0840688630603863,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004888679972494854,
      "loss": 5.9512,
      "mean_token_accuracy": 0.14657400250434877,
      "num_tokens": 25127107.0,
      "step": 12405
    },
    {
      "entropy": 6.260431861877441,
      "epoch": 1.084505811413091,
      "grad_norm": 3.828125,
      "learning_rate": 0.000488858315009328,
      "loss": 5.9993,
      "mean_token_accuracy": 0.14757117852568627,
      "num_tokens": 25138272.0,
      "step": 12410
    },
    {
      "entropy": 6.265703630447388,
      "epoch": 1.0849427597657957,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004888486286672559,
      "loss": 6.0406,
      "mean_token_accuracy": 0.13752536550164224,
      "num_tokens": 25147128.0,
      "step": 12415
    },
    {
      "entropy": 6.329404640197754,
      "epoch": 1.0853797081185004,
      "grad_norm": 2.625,
      "learning_rate": 0.0004888389382234551,
      "loss": 5.9447,
      "mean_token_accuracy": 0.152037513256073,
      "num_tokens": 25157418.0,
      "step": 12420
    },
    {
      "entropy": 6.299213075637818,
      "epoch": 1.085816656471205,
      "grad_norm": 2.75,
      "learning_rate": 0.0004888292436781114,
      "loss": 6.0101,
      "mean_token_accuracy": 0.14738563299179078,
      "num_tokens": 25168123.0,
      "step": 12425
    },
    {
      "entropy": 6.216758966445923,
      "epoch": 1.0862536048239098,
      "grad_norm": 5.34375,
      "learning_rate": 0.0004888195450314108,
      "loss": 6.034,
      "mean_token_accuracy": 0.14486671537160872,
      "num_tokens": 25178445.0,
      "step": 12430
    },
    {
      "entropy": 6.238691902160644,
      "epoch": 1.0866905531766144,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004888098422835392,
      "loss": 6.0529,
      "mean_token_accuracy": 0.13712958320975305,
      "num_tokens": 25189246.0,
      "step": 12435
    },
    {
      "entropy": 6.260590982437134,
      "epoch": 1.0871275015293191,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004888001354346828,
      "loss": 6.038,
      "mean_token_accuracy": 0.14465494304895402,
      "num_tokens": 25201277.0,
      "step": 12440
    },
    {
      "entropy": 6.318410015106201,
      "epoch": 1.087564449882024,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004887904244850279,
      "loss": 6.0309,
      "mean_token_accuracy": 0.14625977948307992,
      "num_tokens": 25211353.0,
      "step": 12445
    },
    {
      "entropy": 6.292759275436401,
      "epoch": 1.0880013982347287,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004887807094347606,
      "loss": 6.0094,
      "mean_token_accuracy": 0.14752870947122573,
      "num_tokens": 25222236.0,
      "step": 12450
    },
    {
      "entropy": 6.205887794494629,
      "epoch": 1.0884383465874334,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004887709902840672,
      "loss": 5.9499,
      "mean_token_accuracy": 0.14649073630571366,
      "num_tokens": 25232464.0,
      "step": 12455
    },
    {
      "entropy": 6.282048988342285,
      "epoch": 1.0888752949401381,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004887612670331342,
      "loss": 6.0175,
      "mean_token_accuracy": 0.14160351380705832,
      "num_tokens": 25243518.0,
      "step": 12460
    },
    {
      "entropy": 6.296008968353272,
      "epoch": 1.0893122432928428,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004887515396821481,
      "loss": 6.0411,
      "mean_token_accuracy": 0.13612902611494065,
      "num_tokens": 25253168.0,
      "step": 12465
    },
    {
      "entropy": 6.261118650436401,
      "epoch": 1.0897491916455475,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004887418082312955,
      "loss": 5.9893,
      "mean_token_accuracy": 0.1448962867259979,
      "num_tokens": 25262750.0,
      "step": 12470
    },
    {
      "entropy": 6.304372787475586,
      "epoch": 1.0901861399982522,
      "grad_norm": 2.59375,
      "learning_rate": 0.000488732072680763,
      "loss": 6.0142,
      "mean_token_accuracy": 0.1454774484038353,
      "num_tokens": 25273520.0,
      "step": 12475
    },
    {
      "entropy": 6.259451627731323,
      "epoch": 1.090623088350957,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004887223330307373,
      "loss": 6.0564,
      "mean_token_accuracy": 0.14338118880987166,
      "num_tokens": 25284136.0,
      "step": 12480
    },
    {
      "entropy": 6.227500677108765,
      "epoch": 1.0910600367036616,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004887125892814053,
      "loss": 5.9588,
      "mean_token_accuracy": 0.14480671733617784,
      "num_tokens": 25294523.0,
      "step": 12485
    },
    {
      "entropy": 6.231381225585937,
      "epoch": 1.0914969850563663,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004887028414329538,
      "loss": 6.0097,
      "mean_token_accuracy": 0.14204122349619866,
      "num_tokens": 25305050.0,
      "step": 12490
    },
    {
      "entropy": 6.292223262786865,
      "epoch": 1.091933933409071,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048869308948557,
      "loss": 6.0158,
      "mean_token_accuracy": 0.14135024100542068,
      "num_tokens": 25314445.0,
      "step": 12495
    },
    {
      "entropy": 6.262028169631958,
      "epoch": 1.0923708817617757,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004886833334394406,
      "loss": 6.0077,
      "mean_token_accuracy": 0.14218891113996507,
      "num_tokens": 25324426.0,
      "step": 12500
    },
    {
      "entropy": 6.281785202026367,
      "epoch": 1.0928078301144806,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004886735732947529,
      "loss": 6.0065,
      "mean_token_accuracy": 0.15100001990795137,
      "num_tokens": 25335051.0,
      "step": 12505
    },
    {
      "entropy": 6.261873006820679,
      "epoch": 1.0932447784671853,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004886638090516942,
      "loss": 5.9377,
      "mean_token_accuracy": 0.14764679074287415,
      "num_tokens": 25343955.0,
      "step": 12510
    },
    {
      "entropy": 6.229827117919922,
      "epoch": 1.09368172681989,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004886540407104517,
      "loss": 6.0472,
      "mean_token_accuracy": 0.13872085958719255,
      "num_tokens": 25353817.0,
      "step": 12515
    },
    {
      "entropy": 6.347247838973999,
      "epoch": 1.0941186751725946,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004886442682712126,
      "loss": 5.9479,
      "mean_token_accuracy": 0.14330507963895797,
      "num_tokens": 25363882.0,
      "step": 12520
    },
    {
      "entropy": 6.300452804565429,
      "epoch": 1.0945556235252993,
      "grad_norm": 2.5,
      "learning_rate": 0.0004886344917341646,
      "loss": 5.961,
      "mean_token_accuracy": 0.1459484241902828,
      "num_tokens": 25373721.0,
      "step": 12525
    },
    {
      "entropy": 6.288984489440918,
      "epoch": 1.094992571878004,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004886247110994952,
      "loss": 6.0003,
      "mean_token_accuracy": 0.14331747740507125,
      "num_tokens": 25383520.0,
      "step": 12530
    },
    {
      "entropy": 6.2675549507141115,
      "epoch": 1.0954295202307087,
      "grad_norm": 5.53125,
      "learning_rate": 0.0004886149263673916,
      "loss": 5.982,
      "mean_token_accuracy": 0.14852372407913209,
      "num_tokens": 25394039.0,
      "step": 12535
    },
    {
      "entropy": 6.267333364486694,
      "epoch": 1.0958664685834134,
      "grad_norm": 2.625,
      "learning_rate": 0.000488605137538042,
      "loss": 6.0134,
      "mean_token_accuracy": 0.14209258705377578,
      "num_tokens": 25403404.0,
      "step": 12540
    },
    {
      "entropy": 6.24105806350708,
      "epoch": 1.096303416936118,
      "grad_norm": 2.625,
      "learning_rate": 0.0004885953446116339,
      "loss": 5.9979,
      "mean_token_accuracy": 0.14517151862382888,
      "num_tokens": 25413800.0,
      "step": 12545
    },
    {
      "entropy": 6.24944658279419,
      "epoch": 1.0967403652888228,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004885855475883551,
      "loss": 5.9609,
      "mean_token_accuracy": 0.15251148939132692,
      "num_tokens": 25424252.0,
      "step": 12550
    },
    {
      "entropy": 6.330325603485107,
      "epoch": 1.0971773136415275,
      "grad_norm": 2.625,
      "learning_rate": 0.0004885757464683937,
      "loss": 5.9651,
      "mean_token_accuracy": 0.1476839765906334,
      "num_tokens": 25433381.0,
      "step": 12555
    },
    {
      "entropy": 6.200826406478882,
      "epoch": 1.0976142619942322,
      "grad_norm": 2.125,
      "learning_rate": 0.0004885659412519376,
      "loss": 5.9445,
      "mean_token_accuracy": 0.15219245702028275,
      "num_tokens": 25443253.0,
      "step": 12560
    },
    {
      "entropy": 6.18723692893982,
      "epoch": 1.098051210346937,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004885561319391747,
      "loss": 6.0593,
      "mean_token_accuracy": 0.1378729447722435,
      "num_tokens": 25453590.0,
      "step": 12565
    },
    {
      "entropy": 6.297606468200684,
      "epoch": 1.0984881586996418,
      "grad_norm": 2.625,
      "learning_rate": 0.0004885463185302934,
      "loss": 5.997,
      "mean_token_accuracy": 0.1515969380736351,
      "num_tokens": 25464065.0,
      "step": 12570
    },
    {
      "entropy": 6.293933391571045,
      "epoch": 1.0989251070523465,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004885365010254817,
      "loss": 6.0986,
      "mean_token_accuracy": 0.13591940328478813,
      "num_tokens": 25475236.0,
      "step": 12575
    },
    {
      "entropy": 6.316383028030396,
      "epoch": 1.0993620554050512,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004885266794249281,
      "loss": 6.0628,
      "mean_token_accuracy": 0.14632176458835602,
      "num_tokens": 25485200.0,
      "step": 12580
    },
    {
      "entropy": 6.305580902099609,
      "epoch": 1.0997990037577559,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004885168537288209,
      "loss": 6.0284,
      "mean_token_accuracy": 0.1469673179090023,
      "num_tokens": 25494775.0,
      "step": 12585
    },
    {
      "entropy": 6.282433986663818,
      "epoch": 1.1002359521104605,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004885070239373484,
      "loss": 6.0228,
      "mean_token_accuracy": 0.1425797313451767,
      "num_tokens": 25504435.0,
      "step": 12590
    },
    {
      "entropy": 6.241544914245606,
      "epoch": 1.1006729004631652,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004884971900506995,
      "loss": 6.0312,
      "mean_token_accuracy": 0.14245644956827164,
      "num_tokens": 25514085.0,
      "step": 12595
    },
    {
      "entropy": 6.297436761856079,
      "epoch": 1.10110984881587,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004884873520690626,
      "loss": 6.061,
      "mean_token_accuracy": 0.13925779461860657,
      "num_tokens": 25523988.0,
      "step": 12600
    },
    {
      "entropy": 6.292375135421753,
      "epoch": 1.1015467971685746,
      "grad_norm": 4.875,
      "learning_rate": 0.0004884775099926264,
      "loss": 6.0316,
      "mean_token_accuracy": 0.14245699793100358,
      "num_tokens": 25534469.0,
      "step": 12605
    },
    {
      "entropy": 6.30132827758789,
      "epoch": 1.1019837455212793,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004884676638215797,
      "loss": 6.0863,
      "mean_token_accuracy": 0.13840550258755685,
      "num_tokens": 25544854.0,
      "step": 12610
    },
    {
      "entropy": 6.317642307281494,
      "epoch": 1.102420693873984,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004884578135561113,
      "loss": 5.9804,
      "mean_token_accuracy": 0.13972052186727524,
      "num_tokens": 25555465.0,
      "step": 12615
    },
    {
      "entropy": 6.2557484149932865,
      "epoch": 1.102857642226689,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004884479591964103,
      "loss": 6.0391,
      "mean_token_accuracy": 0.14196875393390657,
      "num_tokens": 25565949.0,
      "step": 12620
    },
    {
      "entropy": 6.255318737030029,
      "epoch": 1.1032945905793936,
      "grad_norm": 4.75,
      "learning_rate": 0.0004884381007426654,
      "loss": 6.0279,
      "mean_token_accuracy": 0.1437459796667099,
      "num_tokens": 25576511.0,
      "step": 12625
    },
    {
      "entropy": 6.255827283859253,
      "epoch": 1.1037315389320983,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004884282381950661,
      "loss": 5.9518,
      "mean_token_accuracy": 0.1462185025215149,
      "num_tokens": 25587224.0,
      "step": 12630
    },
    {
      "entropy": 6.29834566116333,
      "epoch": 1.104168487284803,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004884183715538012,
      "loss": 5.9732,
      "mean_token_accuracy": 0.14674068689346315,
      "num_tokens": 25596696.0,
      "step": 12635
    },
    {
      "entropy": 6.250792598724365,
      "epoch": 1.1046054356375077,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004884085008190602,
      "loss": 5.9295,
      "mean_token_accuracy": 0.1517595276236534,
      "num_tokens": 25606829.0,
      "step": 12640
    },
    {
      "entropy": 6.2257819175720215,
      "epoch": 1.1050423839902124,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004883986259910323,
      "loss": 6.016,
      "mean_token_accuracy": 0.14758857935667039,
      "num_tokens": 25617122.0,
      "step": 12645
    },
    {
      "entropy": 6.27727255821228,
      "epoch": 1.105479332342917,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004883887470699069,
      "loss": 5.9842,
      "mean_token_accuracy": 0.13973188251256943,
      "num_tokens": 25626485.0,
      "step": 12650
    },
    {
      "entropy": 6.3061079502105715,
      "epoch": 1.1059162806956218,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004883788640558735,
      "loss": 5.9807,
      "mean_token_accuracy": 0.14790340811014174,
      "num_tokens": 25636695.0,
      "step": 12655
    },
    {
      "entropy": 6.16483211517334,
      "epoch": 1.1063532290483264,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004883689769491218,
      "loss": 5.8654,
      "mean_token_accuracy": 0.15435231029987334,
      "num_tokens": 25646169.0,
      "step": 12660
    },
    {
      "entropy": 6.246884393692016,
      "epoch": 1.1067901774010311,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004883590857498413,
      "loss": 6.0436,
      "mean_token_accuracy": 0.13989266082644464,
      "num_tokens": 25656728.0,
      "step": 12665
    },
    {
      "entropy": 6.309694576263428,
      "epoch": 1.1072271257537358,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004883491904582217,
      "loss": 5.9681,
      "mean_token_accuracy": 0.14795630127191545,
      "num_tokens": 25666184.0,
      "step": 12670
    },
    {
      "entropy": 6.330066347122193,
      "epoch": 1.1076640741064405,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004883392910744528,
      "loss": 5.9874,
      "mean_token_accuracy": 0.14565075933933258,
      "num_tokens": 25675571.0,
      "step": 12675
    },
    {
      "entropy": 6.2592628479003904,
      "epoch": 1.1081010224591452,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004883293875987245,
      "loss": 5.9974,
      "mean_token_accuracy": 0.14445026591420174,
      "num_tokens": 25685485.0,
      "step": 12680
    },
    {
      "entropy": 6.2111633777618405,
      "epoch": 1.1085379708118501,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004883194800312268,
      "loss": 5.9431,
      "mean_token_accuracy": 0.14787764996290206,
      "num_tokens": 25695278.0,
      "step": 12685
    },
    {
      "entropy": 6.271595191955567,
      "epoch": 1.1089749191645548,
      "grad_norm": 5.625,
      "learning_rate": 0.0004883095683721497,
      "loss": 6.0198,
      "mean_token_accuracy": 0.14672136753797532,
      "num_tokens": 25704410.0,
      "step": 12690
    },
    {
      "entropy": 6.3308408737182615,
      "epoch": 1.1094118675172595,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004882996526216833,
      "loss": 6.0047,
      "mean_token_accuracy": 0.14059572890400887,
      "num_tokens": 25715027.0,
      "step": 12695
    },
    {
      "entropy": 6.330204200744629,
      "epoch": 1.1098488158699642,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004882897327800178,
      "loss": 6.0365,
      "mean_token_accuracy": 0.14217487275600432,
      "num_tokens": 25723992.0,
      "step": 12700
    },
    {
      "entropy": 6.2542520523071286,
      "epoch": 1.110285764222669,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004882798088473435,
      "loss": 6.0085,
      "mean_token_accuracy": 0.14430973008275033,
      "num_tokens": 25733791.0,
      "step": 12705
    },
    {
      "entropy": 6.277851009368897,
      "epoch": 1.1107227125753736,
      "grad_norm": 3.625,
      "learning_rate": 0.00048826988082385065,
      "loss": 6.026,
      "mean_token_accuracy": 0.14024776890873908,
      "num_tokens": 25743258.0,
      "step": 12710
    },
    {
      "entropy": 6.293457126617431,
      "epoch": 1.1111596609280783,
      "grad_norm": 3.28125,
      "learning_rate": 0.00048825994870972976,
      "loss": 6.0361,
      "mean_token_accuracy": 0.14702748656272888,
      "num_tokens": 25753183.0,
      "step": 12715
    },
    {
      "entropy": 6.335572147369385,
      "epoch": 1.111596609280783,
      "grad_norm": 3.53125,
      "learning_rate": 0.00048825001250517124,
      "loss": 6.0557,
      "mean_token_accuracy": 0.14060214534401894,
      "num_tokens": 25763308.0,
      "step": 12720
    },
    {
      "entropy": 6.285267114639282,
      "epoch": 1.1120335576334877,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004882400722103658,
      "loss": 5.9493,
      "mean_token_accuracy": 0.15074362009763717,
      "num_tokens": 25774269.0,
      "step": 12725
    },
    {
      "entropy": 6.241608190536499,
      "epoch": 1.1124705059861923,
      "grad_norm": 15.4375,
      "learning_rate": 0.000488230127825504,
      "loss": 5.9541,
      "mean_token_accuracy": 0.1465928629040718,
      "num_tokens": 25785194.0,
      "step": 12730
    },
    {
      "entropy": 6.27231912612915,
      "epoch": 1.112907454338897,
      "grad_norm": 4.0625,
      "learning_rate": 0.00048822017935077673,
      "loss": 5.98,
      "mean_token_accuracy": 0.1514858901500702,
      "num_tokens": 25793927.0,
      "step": 12735
    },
    {
      "entropy": 6.266482543945313,
      "epoch": 1.113344402691602,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004882102267863746,
      "loss": 6.0373,
      "mean_token_accuracy": 0.143649859726429,
      "num_tokens": 25805395.0,
      "step": 12740
    },
    {
      "entropy": 6.307115459442139,
      "epoch": 1.1137813510443066,
      "grad_norm": 4.375,
      "learning_rate": 0.00048820027013248864,
      "loss": 6.0817,
      "mean_token_accuracy": 0.1413551889359951,
      "num_tokens": 25816594.0,
      "step": 12745
    },
    {
      "entropy": 6.303380870819092,
      "epoch": 1.1142182993970113,
      "grad_norm": 3.984375,
      "learning_rate": 0.00048819030938930974,
      "loss": 5.9828,
      "mean_token_accuracy": 0.13826431334018707,
      "num_tokens": 25825943.0,
      "step": 12750
    },
    {
      "entropy": 6.181104755401611,
      "epoch": 1.114655247749716,
      "grad_norm": 4.53125,
      "learning_rate": 0.000488180344557029,
      "loss": 5.9761,
      "mean_token_accuracy": 0.1497185170650482,
      "num_tokens": 25835928.0,
      "step": 12755
    },
    {
      "entropy": 6.195069313049316,
      "epoch": 1.1150921961024207,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004881703756358376,
      "loss": 6.0168,
      "mean_token_accuracy": 0.1420762784779072,
      "num_tokens": 25847363.0,
      "step": 12760
    },
    {
      "entropy": 6.2775938510894775,
      "epoch": 1.1155291444551254,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004881604026259266,
      "loss": 6.0248,
      "mean_token_accuracy": 0.14353586882352828,
      "num_tokens": 25857246.0,
      "step": 12765
    },
    {
      "entropy": 6.391989850997925,
      "epoch": 1.11596609280783,
      "grad_norm": 3.375,
      "learning_rate": 0.00048815042552748745,
      "loss": 6.0404,
      "mean_token_accuracy": 0.1365166611969471,
      "num_tokens": 25866615.0,
      "step": 12770
    },
    {
      "entropy": 6.31094651222229,
      "epoch": 1.1164030411605348,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004881404443407115,
      "loss": 6.11,
      "mean_token_accuracy": 0.13781185671687127,
      "num_tokens": 25876698.0,
      "step": 12775
    },
    {
      "entropy": 6.3393003940582275,
      "epoch": 1.1168399895132395,
      "grad_norm": 3.5625,
      "learning_rate": 0.00048813045906579,
      "loss": 6.0572,
      "mean_token_accuracy": 0.13542587608098983,
      "num_tokens": 25888029.0,
      "step": 12780
    },
    {
      "entropy": 6.331125688552857,
      "epoch": 1.1172769378659442,
      "grad_norm": 5.21875,
      "learning_rate": 0.00048812046970291473,
      "loss": 6.0842,
      "mean_token_accuracy": 0.14129285514354706,
      "num_tokens": 25897472.0,
      "step": 12785
    },
    {
      "entropy": 6.3522954940795895,
      "epoch": 1.1177138862186489,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004881104762522772,
      "loss": 6.0497,
      "mean_token_accuracy": 0.13818502575159072,
      "num_tokens": 25908025.0,
      "step": 12790
    },
    {
      "entropy": 6.269056558609009,
      "epoch": 1.1181508345713538,
      "grad_norm": 5.1875,
      "learning_rate": 0.000488100478714069,
      "loss": 6.0402,
      "mean_token_accuracy": 0.13990265652537345,
      "num_tokens": 25919110.0,
      "step": 12795
    },
    {
      "entropy": 6.256549501419068,
      "epoch": 1.1185877829240585,
      "grad_norm": 3.046875,
      "learning_rate": 0.00048809047708848194,
      "loss": 5.9435,
      "mean_token_accuracy": 0.14816736429929733,
      "num_tokens": 25928322.0,
      "step": 12800
    },
    {
      "entropy": 6.289228248596191,
      "epoch": 1.1190247312767632,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048808047137570795,
      "loss": 5.9277,
      "mean_token_accuracy": 0.1469327464699745,
      "num_tokens": 25937347.0,
      "step": 12805
    },
    {
      "entropy": 6.204435729980469,
      "epoch": 1.1194616796294679,
      "grad_norm": 7.21875,
      "learning_rate": 0.0004880704615759388,
      "loss": 5.9503,
      "mean_token_accuracy": 0.1444361537694931,
      "num_tokens": 25948408.0,
      "step": 12810
    },
    {
      "entropy": 6.182691431045532,
      "epoch": 1.1198986279821725,
      "grad_norm": 4.1875,
      "learning_rate": 0.00048806044768936655,
      "loss": 6.0402,
      "mean_token_accuracy": 0.14018608182668685,
      "num_tokens": 25958249.0,
      "step": 12815
    },
    {
      "entropy": 6.273450565338135,
      "epoch": 1.1203355763348772,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004880504297161833,
      "loss": 6.0781,
      "mean_token_accuracy": 0.13792913407087326,
      "num_tokens": 25968849.0,
      "step": 12820
    },
    {
      "entropy": 6.310846328735352,
      "epoch": 1.120772524687582,
      "grad_norm": 3.28125,
      "learning_rate": 0.00048804040765658115,
      "loss": 6.0661,
      "mean_token_accuracy": 0.1436302088201046,
      "num_tokens": 25978787.0,
      "step": 12825
    },
    {
      "entropy": 6.24982213973999,
      "epoch": 1.1212094730402866,
      "grad_norm": 3.0,
      "learning_rate": 0.00048803038151075227,
      "loss": 6.0001,
      "mean_token_accuracy": 0.14503416568040847,
      "num_tokens": 25988619.0,
      "step": 12830
    },
    {
      "entropy": 6.330221843719483,
      "epoch": 1.1216464213929913,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004880203512788892,
      "loss": 6.0335,
      "mean_token_accuracy": 0.14514435678720475,
      "num_tokens": 25999080.0,
      "step": 12835
    },
    {
      "entropy": 6.2447222709655765,
      "epoch": 1.122083369745696,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048801031696118406,
      "loss": 5.9669,
      "mean_token_accuracy": 0.14782094433903695,
      "num_tokens": 26010329.0,
      "step": 12840
    },
    {
      "entropy": 6.2392191886901855,
      "epoch": 1.1225203180984007,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004880002785578295,
      "loss": 5.9403,
      "mean_token_accuracy": 0.14422629550099372,
      "num_tokens": 26019147.0,
      "step": 12845
    },
    {
      "entropy": 6.227623224258423,
      "epoch": 1.1229572664511054,
      "grad_norm": 2.640625,
      "learning_rate": 0.000487990236069018,
      "loss": 6.0137,
      "mean_token_accuracy": 0.14336463958024978,
      "num_tokens": 26029212.0,
      "step": 12850
    },
    {
      "entropy": 6.274884366989136,
      "epoch": 1.12339421480381,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048798018949494214,
      "loss": 5.8701,
      "mean_token_accuracy": 0.1475730463862419,
      "num_tokens": 26038059.0,
      "step": 12855
    },
    {
      "entropy": 6.264238500595093,
      "epoch": 1.123831163156515,
      "grad_norm": 3.0,
      "learning_rate": 0.0004879701388357946,
      "loss": 6.0934,
      "mean_token_accuracy": 0.14062134847044944,
      "num_tokens": 26048076.0,
      "step": 12860
    },
    {
      "entropy": 6.286392688751221,
      "epoch": 1.1242681115092197,
      "grad_norm": 3.125,
      "learning_rate": 0.0004879600840917683,
      "loss": 6.0094,
      "mean_token_accuracy": 0.14666071832180022,
      "num_tokens": 26059006.0,
      "step": 12865
    },
    {
      "entropy": 6.23958387374878,
      "epoch": 1.1247050598619244,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004879500252630559,
      "loss": 5.9605,
      "mean_token_accuracy": 0.14930840581655502,
      "num_tokens": 26069275.0,
      "step": 12870
    },
    {
      "entropy": 6.2704907894134525,
      "epoch": 1.125142008214629,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004879399623498504,
      "loss": 5.9388,
      "mean_token_accuracy": 0.14753743410110473,
      "num_tokens": 26078869.0,
      "step": 12875
    },
    {
      "entropy": 6.268105745315552,
      "epoch": 1.1255789565673338,
      "grad_norm": 5.4375,
      "learning_rate": 0.000487929895352345,
      "loss": 5.9527,
      "mean_token_accuracy": 0.14460204392671586,
      "num_tokens": 26088544.0,
      "step": 12880
    },
    {
      "entropy": 6.1638294696807865,
      "epoch": 1.1260159049200384,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004879198242707325,
      "loss": 5.9457,
      "mean_token_accuracy": 0.14733243063092233,
      "num_tokens": 26097503.0,
      "step": 12885
    },
    {
      "entropy": 6.320313882827759,
      "epoch": 1.1264528532727431,
      "grad_norm": 3.390625,
      "learning_rate": 0.00048790974910520627,
      "loss": 6.0551,
      "mean_token_accuracy": 0.1374025024473667,
      "num_tokens": 26107228.0,
      "step": 12890
    },
    {
      "entropy": 6.285621881484985,
      "epoch": 1.1268898016254478,
      "grad_norm": 4.03125,
      "learning_rate": 0.00048789966985595946,
      "loss": 5.982,
      "mean_token_accuracy": 0.15374408960342406,
      "num_tokens": 26117633.0,
      "step": 12895
    },
    {
      "entropy": 6.281871652603149,
      "epoch": 1.1273267499781525,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004878895865231854,
      "loss": 6.0385,
      "mean_token_accuracy": 0.14128054603934287,
      "num_tokens": 26127781.0,
      "step": 12900
    },
    {
      "entropy": 6.363052558898926,
      "epoch": 1.1277636983308572,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004878794991070776,
      "loss": 5.966,
      "mean_token_accuracy": 0.14915400743484497,
      "num_tokens": 26137675.0,
      "step": 12905
    },
    {
      "entropy": 6.3668440818786625,
      "epoch": 1.128200646683562,
      "grad_norm": 5.46875,
      "learning_rate": 0.00048786940760782937,
      "loss": 6.0646,
      "mean_token_accuracy": 0.14725981429219245,
      "num_tokens": 26148572.0,
      "step": 12910
    },
    {
      "entropy": 6.2178435802459715,
      "epoch": 1.1286375950362668,
      "grad_norm": 3.859375,
      "learning_rate": 0.00048785931202563443,
      "loss": 6.0546,
      "mean_token_accuracy": 0.1485311321914196,
      "num_tokens": 26159261.0,
      "step": 12915
    },
    {
      "entropy": 6.207454586029053,
      "epoch": 1.1290745433889715,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004878492123606863,
      "loss": 6.019,
      "mean_token_accuracy": 0.14837898835539817,
      "num_tokens": 26170072.0,
      "step": 12920
    },
    {
      "entropy": 6.25847053527832,
      "epoch": 1.1295114917416762,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048783910861317873,
      "loss": 5.9567,
      "mean_token_accuracy": 0.14195345491170883,
      "num_tokens": 26179947.0,
      "step": 12925
    },
    {
      "entropy": 6.317234182357788,
      "epoch": 1.129948440094381,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048782900078330556,
      "loss": 6.0138,
      "mean_token_accuracy": 0.14335481077432632,
      "num_tokens": 26189982.0,
      "step": 12930
    },
    {
      "entropy": 6.349755859375,
      "epoch": 1.1303853884470856,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004878188888712606,
      "loss": 6.1107,
      "mean_token_accuracy": 0.13310218304395677,
      "num_tokens": 26200832.0,
      "step": 12935
    },
    {
      "entropy": 6.310823440551758,
      "epoch": 1.1308223367997903,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004878087728772379,
      "loss": 6.0449,
      "mean_token_accuracy": 0.14565031081438065,
      "num_tokens": 26211506.0,
      "step": 12940
    },
    {
      "entropy": 6.201258134841919,
      "epoch": 1.131259285152495,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004877986528014314,
      "loss": 5.9672,
      "mean_token_accuracy": 0.14644421339035035,
      "num_tokens": 26222045.0,
      "step": 12945
    },
    {
      "entropy": 6.221653747558594,
      "epoch": 1.1316962335051997,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004877885286440353,
      "loss": 5.9258,
      "mean_token_accuracy": 0.14523024410009383,
      "num_tokens": 26231411.0,
      "step": 12950
    },
    {
      "entropy": 6.292685699462891,
      "epoch": 1.1321331818579043,
      "grad_norm": 3.46875,
      "learning_rate": 0.00048777840040524365,
      "loss": 6.0173,
      "mean_token_accuracy": 0.14554496705532075,
      "num_tokens": 26241491.0,
      "step": 12955
    },
    {
      "entropy": 6.266798114776611,
      "epoch": 1.132570130210609,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048776826808525074,
      "loss": 6.1156,
      "mean_token_accuracy": 0.14168209806084633,
      "num_tokens": 26252037.0,
      "step": 12960
    },
    {
      "entropy": 6.317578077316284,
      "epoch": 1.1330070785633137,
      "grad_norm": 6.84375,
      "learning_rate": 0.00048775813168425104,
      "loss": 6.0966,
      "mean_token_accuracy": 0.13841143026947975,
      "num_tokens": 26262310.0,
      "step": 12965
    },
    {
      "entropy": 6.333992290496826,
      "epoch": 1.1334440269160186,
      "grad_norm": 3.5625,
      "learning_rate": 0.00048774799120243883,
      "loss": 6.0055,
      "mean_token_accuracy": 0.1453311748802662,
      "num_tokens": 26273001.0,
      "step": 12970
    },
    {
      "entropy": 6.3125639915466305,
      "epoch": 1.1338809752687231,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004877378466400087,
      "loss": 6.0447,
      "mean_token_accuracy": 0.14394302815198898,
      "num_tokens": 26283500.0,
      "step": 12975
    },
    {
      "entropy": 6.260612201690674,
      "epoch": 1.134317923621428,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004877276979971552,
      "loss": 6.0073,
      "mean_token_accuracy": 0.1413673885166645,
      "num_tokens": 26294118.0,
      "step": 12980
    },
    {
      "entropy": 6.294612836837769,
      "epoch": 1.1347548719741327,
      "grad_norm": 2.4375,
      "learning_rate": 0.000487717545274073,
      "loss": 6.1127,
      "mean_token_accuracy": 0.13886088877916336,
      "num_tokens": 26303833.0,
      "step": 12985
    },
    {
      "entropy": 6.287276792526245,
      "epoch": 1.1351918203268374,
      "grad_norm": 3.171875,
      "learning_rate": 0.00048770738847095683,
      "loss": 6.0497,
      "mean_token_accuracy": 0.1447442039847374,
      "num_tokens": 26313542.0,
      "step": 12990
    },
    {
      "entropy": 6.299585342407227,
      "epoch": 1.135628768679542,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048769722758800144,
      "loss": 5.9947,
      "mean_token_accuracy": 0.15115102753043175,
      "num_tokens": 26322783.0,
      "step": 12995
    },
    {
      "entropy": 6.3491730213165285,
      "epoch": 1.1360657170322468,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004876870626254018,
      "loss": 6.0571,
      "mean_token_accuracy": 0.13855614587664605,
      "num_tokens": 26333770.0,
      "step": 13000
    },
    {
      "entropy": 6.316054248809815,
      "epoch": 1.1365026653849515,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048767689358335286,
      "loss": 6.0086,
      "mean_token_accuracy": 0.14718972966074945,
      "num_tokens": 26344435.0,
      "step": 13005
    },
    {
      "entropy": 6.221671199798584,
      "epoch": 1.1369396137376562,
      "grad_norm": 3.625,
      "learning_rate": 0.0004876667204620497,
      "loss": 5.9748,
      "mean_token_accuracy": 0.15438514724373817,
      "num_tokens": 26355410.0,
      "step": 13010
    },
    {
      "entropy": 6.243892955780029,
      "epoch": 1.1373765620903609,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004876565432616873,
      "loss": 6.026,
      "mean_token_accuracy": 0.14636286944150925,
      "num_tokens": 26365785.0,
      "step": 13015
    },
    {
      "entropy": 6.294614648818969,
      "epoch": 1.1378135104430656,
      "grad_norm": 5.0,
      "learning_rate": 0.00048764636198246104,
      "loss": 5.983,
      "mean_token_accuracy": 0.150182543694973,
      "num_tokens": 26375489.0,
      "step": 13020
    },
    {
      "entropy": 6.252637958526611,
      "epoch": 1.1382504587957702,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004876361766245661,
      "loss": 5.9695,
      "mean_token_accuracy": 0.14584982842206956,
      "num_tokens": 26385126.0,
      "step": 13025
    },
    {
      "entropy": 6.254229021072388,
      "epoch": 1.138687407148475,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004876259871881979,
      "loss": 5.97,
      "mean_token_accuracy": 0.14037542045116425,
      "num_tokens": 26395314.0,
      "step": 13030
    },
    {
      "entropy": 6.313602542877197,
      "epoch": 1.1391243555011799,
      "grad_norm": 3.25,
      "learning_rate": 0.0004876157936735519,
      "loss": 6.0248,
      "mean_token_accuracy": 0.1422035202383995,
      "num_tokens": 26404660.0,
      "step": 13035
    },
    {
      "entropy": 6.324435663223267,
      "epoch": 1.1395613038538845,
      "grad_norm": 4.03125,
      "learning_rate": 0.00048760559608082343,
      "loss": 6.0242,
      "mean_token_accuracy": 0.1459619387984276,
      "num_tokens": 26415371.0,
      "step": 13040
    },
    {
      "entropy": 6.279032611846924,
      "epoch": 1.1399982522065892,
      "grad_norm": 7.96875,
      "learning_rate": 0.0004875953944102084,
      "loss": 6.0506,
      "mean_token_accuracy": 0.14388590902090073,
      "num_tokens": 26424896.0,
      "step": 13045
    },
    {
      "entropy": 6.281288957595825,
      "epoch": 1.140435200559294,
      "grad_norm": 2.96875,
      "learning_rate": 0.00048758518866190214,
      "loss": 6.0061,
      "mean_token_accuracy": 0.14302221313118935,
      "num_tokens": 26434847.0,
      "step": 13050
    },
    {
      "entropy": 6.26182713508606,
      "epoch": 1.1408721489119986,
      "grad_norm": 5.3125,
      "learning_rate": 0.0004875749788361006,
      "loss": 5.9869,
      "mean_token_accuracy": 0.14150801301002502,
      "num_tokens": 26444138.0,
      "step": 13055
    },
    {
      "entropy": 6.284028339385986,
      "epoch": 1.1413090972647033,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004875647649329997,
      "loss": 5.9977,
      "mean_token_accuracy": 0.13945851475000381,
      "num_tokens": 26453549.0,
      "step": 13060
    },
    {
      "entropy": 6.300138711929321,
      "epoch": 1.141746045617408,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004875545469527952,
      "loss": 6.017,
      "mean_token_accuracy": 0.14119101613759993,
      "num_tokens": 26464386.0,
      "step": 13065
    },
    {
      "entropy": 6.234641408920288,
      "epoch": 1.1421829939701127,
      "grad_norm": 7.59375,
      "learning_rate": 0.000487544324895683,
      "loss": 5.9827,
      "mean_token_accuracy": 0.14655765295028686,
      "num_tokens": 26474146.0,
      "step": 13070
    },
    {
      "entropy": 6.199444150924682,
      "epoch": 1.1426199423228174,
      "grad_norm": 4.53125,
      "learning_rate": 0.00048753409876185934,
      "loss": 5.9552,
      "mean_token_accuracy": 0.1495281718671322,
      "num_tokens": 26484408.0,
      "step": 13075
    },
    {
      "entropy": 6.331406497955323,
      "epoch": 1.143056890675522,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004875238685515203,
      "loss": 5.9545,
      "mean_token_accuracy": 0.14020450711250304,
      "num_tokens": 26494769.0,
      "step": 13080
    },
    {
      "entropy": 6.214972686767578,
      "epoch": 1.1434938390282268,
      "grad_norm": 3.640625,
      "learning_rate": 0.00048751363426486206,
      "loss": 6.0178,
      "mean_token_accuracy": 0.1430754393339157,
      "num_tokens": 26505917.0,
      "step": 13085
    },
    {
      "entropy": 6.2634142398834225,
      "epoch": 1.1439307873809317,
      "grad_norm": 2.140625,
      "learning_rate": 0.000487503395902081,
      "loss": 6.0696,
      "mean_token_accuracy": 0.13239913880825044,
      "num_tokens": 26516388.0,
      "step": 13090
    },
    {
      "entropy": 6.40624418258667,
      "epoch": 1.1443677357336364,
      "grad_norm": 4.5,
      "learning_rate": 0.0004874931534633735,
      "loss": 6.0311,
      "mean_token_accuracy": 0.1352994807064533,
      "num_tokens": 26527180.0,
      "step": 13095
    },
    {
      "entropy": 6.302253484725952,
      "epoch": 1.144804684086341,
      "grad_norm": 4.125,
      "learning_rate": 0.0004874829069489359,
      "loss": 5.9961,
      "mean_token_accuracy": 0.14411711692810059,
      "num_tokens": 26537705.0,
      "step": 13100
    },
    {
      "entropy": 6.24558277130127,
      "epoch": 1.1452416324390458,
      "grad_norm": 4.5625,
      "learning_rate": 0.00048747265635896487,
      "loss": 6.1568,
      "mean_token_accuracy": 0.13134488761425017,
      "num_tokens": 26548862.0,
      "step": 13105
    },
    {
      "entropy": 6.307693958282471,
      "epoch": 1.1456785807917504,
      "grad_norm": 3.75,
      "learning_rate": 0.00048746240169365683,
      "loss": 6.0235,
      "mean_token_accuracy": 0.14433484971523286,
      "num_tokens": 26559344.0,
      "step": 13110
    },
    {
      "entropy": 6.311009740829467,
      "epoch": 1.1461155291444551,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004874521429532087,
      "loss": 5.9611,
      "mean_token_accuracy": 0.14302032440900803,
      "num_tokens": 26568439.0,
      "step": 13115
    },
    {
      "entropy": 6.310689687728882,
      "epoch": 1.1465524774971598,
      "grad_norm": 5.46875,
      "learning_rate": 0.00048744188013781714,
      "loss": 5.9735,
      "mean_token_accuracy": 0.14713335782289505,
      "num_tokens": 26578588.0,
      "step": 13120
    },
    {
      "entropy": 6.280047512054443,
      "epoch": 1.1469894258498645,
      "grad_norm": 3.46875,
      "learning_rate": 0.000487431613247679,
      "loss": 6.0333,
      "mean_token_accuracy": 0.14709607809782027,
      "num_tokens": 26588491.0,
      "step": 13125
    },
    {
      "entropy": 6.270607328414917,
      "epoch": 1.1474263742025692,
      "grad_norm": 5.1875,
      "learning_rate": 0.00048742134228299116,
      "loss": 6.0573,
      "mean_token_accuracy": 0.14218661859631537,
      "num_tokens": 26598945.0,
      "step": 13130
    },
    {
      "entropy": 6.228167343139648,
      "epoch": 1.147863322555274,
      "grad_norm": 7.3125,
      "learning_rate": 0.0004874110672439508,
      "loss": 5.9594,
      "mean_token_accuracy": 0.1498783528804779,
      "num_tokens": 26609842.0,
      "step": 13135
    },
    {
      "entropy": 6.332759761810303,
      "epoch": 1.1483002709079786,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004874007881307547,
      "loss": 6.0857,
      "mean_token_accuracy": 0.14003798812627793,
      "num_tokens": 26620836.0,
      "step": 13140
    },
    {
      "entropy": 6.304780006408691,
      "epoch": 1.1487372192606835,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048739050494360026,
      "loss": 6.0105,
      "mean_token_accuracy": 0.14240805357694625,
      "num_tokens": 26630148.0,
      "step": 13145
    },
    {
      "entropy": 6.2891865253448485,
      "epoch": 1.149174167613388,
      "grad_norm": 3.46875,
      "learning_rate": 0.00048738021768268463,
      "loss": 6.0383,
      "mean_token_accuracy": 0.14067976847290992,
      "num_tokens": 26640518.0,
      "step": 13150
    },
    {
      "entropy": 6.227099084854126,
      "epoch": 1.149611115966093,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004873699263482052,
      "loss": 5.9461,
      "mean_token_accuracy": 0.1525614231824875,
      "num_tokens": 26650156.0,
      "step": 13155
    },
    {
      "entropy": 6.265003728866577,
      "epoch": 1.1500480643187976,
      "grad_norm": 2.75,
      "learning_rate": 0.00048735963094035924,
      "loss": 6.0221,
      "mean_token_accuracy": 0.1425687700510025,
      "num_tokens": 26661251.0,
      "step": 13160
    },
    {
      "entropy": 6.304848146438599,
      "epoch": 1.1504850126715023,
      "grad_norm": 6.75,
      "learning_rate": 0.0004873493314593443,
      "loss": 6.0364,
      "mean_token_accuracy": 0.14249658733606338,
      "num_tokens": 26671851.0,
      "step": 13165
    },
    {
      "entropy": 6.303148460388184,
      "epoch": 1.150921961024207,
      "grad_norm": 6.1875,
      "learning_rate": 0.0004873390279053579,
      "loss": 6.0565,
      "mean_token_accuracy": 0.1360493965446949,
      "num_tokens": 26682984.0,
      "step": 13170
    },
    {
      "entropy": 6.299660682678223,
      "epoch": 1.1513589093769117,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004873287202785977,
      "loss": 5.9701,
      "mean_token_accuracy": 0.14819006770849227,
      "num_tokens": 26692706.0,
      "step": 13175
    },
    {
      "entropy": 6.259296035766601,
      "epoch": 1.1517958577296163,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004873184085792614,
      "loss": 5.9895,
      "mean_token_accuracy": 0.14518386423587798,
      "num_tokens": 26702152.0,
      "step": 13180
    },
    {
      "entropy": 6.20159764289856,
      "epoch": 1.152232806082321,
      "grad_norm": 4.5625,
      "learning_rate": 0.00048730809280754674,
      "loss": 5.9843,
      "mean_token_accuracy": 0.15009746626019477,
      "num_tokens": 26711404.0,
      "step": 13185
    },
    {
      "entropy": 6.279506492614746,
      "epoch": 1.1526697544350257,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004872977729636516,
      "loss": 5.9655,
      "mean_token_accuracy": 0.14627522900700568,
      "num_tokens": 26720776.0,
      "step": 13190
    },
    {
      "entropy": 6.337982177734375,
      "epoch": 1.1531067027877304,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048728744904777394,
      "loss": 6.0238,
      "mean_token_accuracy": 0.1452839970588684,
      "num_tokens": 26730700.0,
      "step": 13195
    },
    {
      "entropy": 6.290684652328491,
      "epoch": 1.1535436511404351,
      "grad_norm": 5.96875,
      "learning_rate": 0.00048727712106011185,
      "loss": 6.0782,
      "mean_token_accuracy": 0.13946773186326028,
      "num_tokens": 26741453.0,
      "step": 13200
    },
    {
      "entropy": 6.230269908905029,
      "epoch": 1.1539805994931398,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048726678900086327,
      "loss": 6.0146,
      "mean_token_accuracy": 0.14337396919727324,
      "num_tokens": 26750627.0,
      "step": 13205
    },
    {
      "entropy": 6.31253137588501,
      "epoch": 1.1544175478458447,
      "grad_norm": 2.75,
      "learning_rate": 0.00048725645287022644,
      "loss": 6.0175,
      "mean_token_accuracy": 0.14359994158148764,
      "num_tokens": 26760820.0,
      "step": 13210
    },
    {
      "entropy": 6.336253976821899,
      "epoch": 1.1548544961985494,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004872461126683996,
      "loss": 5.9694,
      "mean_token_accuracy": 0.14529047757387162,
      "num_tokens": 26771535.0,
      "step": 13215
    },
    {
      "entropy": 6.299289083480835,
      "epoch": 1.155291444551254,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004872357683955812,
      "loss": 6.0566,
      "mean_token_accuracy": 0.14061030894517898,
      "num_tokens": 26782755.0,
      "step": 13220
    },
    {
      "entropy": 6.294253492355347,
      "epoch": 1.1557283929039588,
      "grad_norm": 3.34375,
      "learning_rate": 0.00048722542005196954,
      "loss": 6.0123,
      "mean_token_accuracy": 0.13957512974739075,
      "num_tokens": 26792834.0,
      "step": 13225
    },
    {
      "entropy": 6.245408391952514,
      "epoch": 1.1561653412566635,
      "grad_norm": 9.125,
      "learning_rate": 0.00048721506763776306,
      "loss": 6.0118,
      "mean_token_accuracy": 0.14240480288863183,
      "num_tokens": 26802865.0,
      "step": 13230
    },
    {
      "entropy": 6.264209747314453,
      "epoch": 1.1566022896093682,
      "grad_norm": 5.59375,
      "learning_rate": 0.0004872047111531604,
      "loss": 5.9922,
      "mean_token_accuracy": 0.14144984185695647,
      "num_tokens": 26813566.0,
      "step": 13235
    },
    {
      "entropy": 6.312805795669556,
      "epoch": 1.1570392379620729,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004871943505983602,
      "loss": 5.9713,
      "mean_token_accuracy": 0.1530612587928772,
      "num_tokens": 26823885.0,
      "step": 13240
    },
    {
      "entropy": 6.165761995315552,
      "epoch": 1.1574761863147776,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004871839859735612,
      "loss": 5.9213,
      "mean_token_accuracy": 0.15802264362573623,
      "num_tokens": 26833236.0,
      "step": 13245
    },
    {
      "entropy": 6.231919240951538,
      "epoch": 1.1579131346674822,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048717361727896203,
      "loss": 6.0544,
      "mean_token_accuracy": 0.13938154727220536,
      "num_tokens": 26842651.0,
      "step": 13250
    },
    {
      "entropy": 6.288454389572143,
      "epoch": 1.158350083020187,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048716324451476183,
      "loss": 5.967,
      "mean_token_accuracy": 0.13859532848000528,
      "num_tokens": 26852198.0,
      "step": 13255
    },
    {
      "entropy": 6.372859621047974,
      "epoch": 1.1587870313728916,
      "grad_norm": 4.125,
      "learning_rate": 0.0004871528676811594,
      "loss": 6.0928,
      "mean_token_accuracy": 0.1340640239417553,
      "num_tokens": 26862807.0,
      "step": 13260
    },
    {
      "entropy": 6.354675531387329,
      "epoch": 1.1592239797255965,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004871424867783538,
      "loss": 6.0496,
      "mean_token_accuracy": 0.1338210716843605,
      "num_tokens": 26871992.0,
      "step": 13265
    },
    {
      "entropy": 6.2716117858886715,
      "epoch": 1.1596609280783012,
      "grad_norm": 3.78125,
      "learning_rate": 0.00048713210180654414,
      "loss": 5.9761,
      "mean_token_accuracy": 0.14388959258794784,
      "num_tokens": 26881417.0,
      "step": 13270
    },
    {
      "entropy": 6.227509355545044,
      "epoch": 1.160097876431006,
      "grad_norm": 3.375,
      "learning_rate": 0.00048712171276592964,
      "loss": 5.9698,
      "mean_token_accuracy": 0.1551525726914406,
      "num_tokens": 26891879.0,
      "step": 13275
    },
    {
      "entropy": 6.23972601890564,
      "epoch": 1.1605348247837106,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004871113196567095,
      "loss": 6.1103,
      "mean_token_accuracy": 0.14009310603141784,
      "num_tokens": 26902729.0,
      "step": 13280
    },
    {
      "entropy": 6.263691377639771,
      "epoch": 1.1609717731364153,
      "grad_norm": 3.90625,
      "learning_rate": 0.00048710092247908316,
      "loss": 6.005,
      "mean_token_accuracy": 0.1456309214234352,
      "num_tokens": 26914415.0,
      "step": 13285
    },
    {
      "entropy": 6.354875755310059,
      "epoch": 1.16140872148912,
      "grad_norm": 2.875,
      "learning_rate": 0.00048709052123324995,
      "loss": 6.0355,
      "mean_token_accuracy": 0.1448797345161438,
      "num_tokens": 26924513.0,
      "step": 13290
    },
    {
      "entropy": 6.280339193344116,
      "epoch": 1.1618456698418247,
      "grad_norm": 3.15625,
      "learning_rate": 0.00048708011591940936,
      "loss": 5.9286,
      "mean_token_accuracy": 0.1544247642159462,
      "num_tokens": 26935039.0,
      "step": 13295
    },
    {
      "entropy": 6.257584571838379,
      "epoch": 1.1622826181945294,
      "grad_norm": 4.21875,
      "learning_rate": 0.000487069706537761,
      "loss": 5.9302,
      "mean_token_accuracy": 0.15333217680454253,
      "num_tokens": 26945148.0,
      "step": 13300
    },
    {
      "entropy": 6.246370649337768,
      "epoch": 1.162719566547234,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004870592930885046,
      "loss": 5.978,
      "mean_token_accuracy": 0.1435015916824341,
      "num_tokens": 26955227.0,
      "step": 13305
    },
    {
      "entropy": 6.227784299850464,
      "epoch": 1.1631565148999388,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048704887557183993,
      "loss": 5.9599,
      "mean_token_accuracy": 0.14295344203710555,
      "num_tokens": 26964389.0,
      "step": 13310
    },
    {
      "entropy": 6.300914096832275,
      "epoch": 1.1635934632526435,
      "grad_norm": 2.859375,
      "learning_rate": 0.00048703845398796665,
      "loss": 5.9608,
      "mean_token_accuracy": 0.1500362515449524,
      "num_tokens": 26974695.0,
      "step": 13315
    },
    {
      "entropy": 6.287800073623657,
      "epoch": 1.1640304116053484,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048702802833708466,
      "loss": 5.953,
      "mean_token_accuracy": 0.15088015496730806,
      "num_tokens": 26984851.0,
      "step": 13320
    },
    {
      "entropy": 6.363280868530273,
      "epoch": 1.1644673599580528,
      "grad_norm": 5.78125,
      "learning_rate": 0.00048701759861939403,
      "loss": 6.0362,
      "mean_token_accuracy": 0.1460250124335289,
      "num_tokens": 26995967.0,
      "step": 13325
    },
    {
      "entropy": 6.264028310775757,
      "epoch": 1.1649043083107578,
      "grad_norm": 3.859375,
      "learning_rate": 0.00048700716483509476,
      "loss": 5.95,
      "mean_token_accuracy": 0.15107825696468352,
      "num_tokens": 27006469.0,
      "step": 13330
    },
    {
      "entropy": 6.268509340286255,
      "epoch": 1.1653412566634624,
      "grad_norm": 2.765625,
      "learning_rate": 0.000486996726984387,
      "loss": 6.0601,
      "mean_token_accuracy": 0.1393354967236519,
      "num_tokens": 27016419.0,
      "step": 13335
    },
    {
      "entropy": 6.297667551040649,
      "epoch": 1.1657782050161671,
      "grad_norm": 4.375,
      "learning_rate": 0.0004869862850674709,
      "loss": 6.0262,
      "mean_token_accuracy": 0.1397349491715431,
      "num_tokens": 27026322.0,
      "step": 13340
    },
    {
      "entropy": 6.241804456710815,
      "epoch": 1.1662151533688718,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048697583908454675,
      "loss": 5.963,
      "mean_token_accuracy": 0.14314978793263436,
      "num_tokens": 27036518.0,
      "step": 13345
    },
    {
      "entropy": 6.2728273391723635,
      "epoch": 1.1666521017215765,
      "grad_norm": 3.0,
      "learning_rate": 0.000486965389035815,
      "loss": 6.0312,
      "mean_token_accuracy": 0.14233418181538582,
      "num_tokens": 27048035.0,
      "step": 13350
    },
    {
      "entropy": 6.254635620117187,
      "epoch": 1.1670890500742812,
      "grad_norm": 2.953125,
      "learning_rate": 0.00048695493492147596,
      "loss": 6.0215,
      "mean_token_accuracy": 0.14137572199106216,
      "num_tokens": 27058507.0,
      "step": 13355
    },
    {
      "entropy": 6.242220830917359,
      "epoch": 1.167525998426986,
      "grad_norm": 3.46875,
      "learning_rate": 0.00048694447674173025,
      "loss": 5.9773,
      "mean_token_accuracy": 0.14808228686451913,
      "num_tokens": 27068350.0,
      "step": 13360
    },
    {
      "entropy": 6.263838005065918,
      "epoch": 1.1679629467796906,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004869340144967784,
      "loss": 5.9807,
      "mean_token_accuracy": 0.1443588390946388,
      "num_tokens": 27077840.0,
      "step": 13365
    },
    {
      "entropy": 6.212014150619507,
      "epoch": 1.1683998951323953,
      "grad_norm": 3.5,
      "learning_rate": 0.00048692354818682113,
      "loss": 5.9559,
      "mean_token_accuracy": 0.14607370272278786,
      "num_tokens": 27087687.0,
      "step": 13370
    },
    {
      "entropy": 6.303565502166748,
      "epoch": 1.1688368434851,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004869130778120591,
      "loss": 6.1158,
      "mean_token_accuracy": 0.13887638077139855,
      "num_tokens": 27099057.0,
      "step": 13375
    },
    {
      "entropy": 6.326699161529541,
      "epoch": 1.1692737918378047,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004869026033726933,
      "loss": 6.0626,
      "mean_token_accuracy": 0.14528990760445595,
      "num_tokens": 27109617.0,
      "step": 13380
    },
    {
      "entropy": 6.307464742660523,
      "epoch": 1.1697107401905096,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004868921248689244,
      "loss": 6.0395,
      "mean_token_accuracy": 0.14096909761428833,
      "num_tokens": 27119456.0,
      "step": 13385
    },
    {
      "entropy": 6.3064933776855465,
      "epoch": 1.1701476885432143,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004868816423009535,
      "loss": 5.9778,
      "mean_token_accuracy": 0.14718716070055962,
      "num_tokens": 27129333.0,
      "step": 13390
    },
    {
      "entropy": 6.2719793796539305,
      "epoch": 1.170584636895919,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004868711556689818,
      "loss": 5.9857,
      "mean_token_accuracy": 0.141240593791008,
      "num_tokens": 27139205.0,
      "step": 13395
    },
    {
      "entropy": 6.289956092834473,
      "epoch": 1.1710215852486237,
      "grad_norm": 4.09375,
      "learning_rate": 0.00048686066497321027,
      "loss": 6.0154,
      "mean_token_accuracy": 0.15101325511932373,
      "num_tokens": 27148530.0,
      "step": 13400
    },
    {
      "entropy": 6.350889539718628,
      "epoch": 1.1714585336013283,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048685017021384014,
      "loss": 6.0889,
      "mean_token_accuracy": 0.14009909927845002,
      "num_tokens": 27159000.0,
      "step": 13405
    },
    {
      "entropy": 6.3304972648620605,
      "epoch": 1.171895481954033,
      "grad_norm": 2.828125,
      "learning_rate": 0.00048683967139107276,
      "loss": 6.0124,
      "mean_token_accuracy": 0.14393264874815942,
      "num_tokens": 27169035.0,
      "step": 13410
    },
    {
      "entropy": 6.232841682434082,
      "epoch": 1.1723324303067377,
      "grad_norm": 10.125,
      "learning_rate": 0.0004868291685051095,
      "loss": 6.0192,
      "mean_token_accuracy": 0.1406059630215168,
      "num_tokens": 27179809.0,
      "step": 13415
    },
    {
      "entropy": 6.247319555282592,
      "epoch": 1.1727693786594424,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004868186615561518,
      "loss": 6.0313,
      "mean_token_accuracy": 0.14061970859766007,
      "num_tokens": 27190445.0,
      "step": 13420
    },
    {
      "entropy": 6.346269369125366,
      "epoch": 1.173206327012147,
      "grad_norm": 3.609375,
      "learning_rate": 0.00048680815054440114,
      "loss": 5.9632,
      "mean_token_accuracy": 0.14478692188858985,
      "num_tokens": 27201097.0,
      "step": 13425
    },
    {
      "entropy": 6.2391016483306885,
      "epoch": 1.1736432753648518,
      "grad_norm": 5.25,
      "learning_rate": 0.00048679763547005925,
      "loss": 5.9564,
      "mean_token_accuracy": 0.14840019047260283,
      "num_tokens": 27210181.0,
      "step": 13430
    },
    {
      "entropy": 6.21347599029541,
      "epoch": 1.1740802237175565,
      "grad_norm": 4.375,
      "learning_rate": 0.0004867871163333276,
      "loss": 6.0129,
      "mean_token_accuracy": 0.14009163677692413,
      "num_tokens": 27219869.0,
      "step": 13435
    },
    {
      "entropy": 6.3036645412445065,
      "epoch": 1.1745171720702614,
      "grad_norm": 2.75,
      "learning_rate": 0.0004867765931344082,
      "loss": 6.0236,
      "mean_token_accuracy": 0.14137037470936775,
      "num_tokens": 27229730.0,
      "step": 13440
    },
    {
      "entropy": 6.287320518493653,
      "epoch": 1.174954120422966,
      "grad_norm": 3.5,
      "learning_rate": 0.0004867660658735027,
      "loss": 6.0092,
      "mean_token_accuracy": 0.150214571505785,
      "num_tokens": 27239534.0,
      "step": 13445
    },
    {
      "entropy": 6.216949796676635,
      "epoch": 1.1753910687756708,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004867555345508131,
      "loss": 6.0643,
      "mean_token_accuracy": 0.13782035931944847,
      "num_tokens": 27249017.0,
      "step": 13450
    },
    {
      "entropy": 6.323127746582031,
      "epoch": 1.1758280171283755,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004867449991665414,
      "loss": 6.1122,
      "mean_token_accuracy": 0.1400067076086998,
      "num_tokens": 27261135.0,
      "step": 13455
    },
    {
      "entropy": 6.373724412918091,
      "epoch": 1.1762649654810802,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004867344597208897,
      "loss": 6.0068,
      "mean_token_accuracy": 0.13740583807229995,
      "num_tokens": 27270763.0,
      "step": 13460
    },
    {
      "entropy": 6.293536710739136,
      "epoch": 1.1767019138337849,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004867239162140601,
      "loss": 6.0326,
      "mean_token_accuracy": 0.14375947415828705,
      "num_tokens": 27280115.0,
      "step": 13465
    },
    {
      "entropy": 6.2515425205230715,
      "epoch": 1.1771388621864896,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004867133686462548,
      "loss": 5.9915,
      "mean_token_accuracy": 0.14531879276037216,
      "num_tokens": 27290162.0,
      "step": 13470
    },
    {
      "entropy": 6.257787084579467,
      "epoch": 1.1775758105391942,
      "grad_norm": 4.125,
      "learning_rate": 0.00048670281701767616,
      "loss": 5.9466,
      "mean_token_accuracy": 0.15601752400398256,
      "num_tokens": 27300467.0,
      "step": 13475
    },
    {
      "entropy": 6.248975038528442,
      "epoch": 1.178012758891899,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004866922613285266,
      "loss": 6.0126,
      "mean_token_accuracy": 0.14729927927255632,
      "num_tokens": 27310524.0,
      "step": 13480
    },
    {
      "entropy": 6.286457586288452,
      "epoch": 1.1784497072446036,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004866817015790085,
      "loss": 6.0346,
      "mean_token_accuracy": 0.14553877711296082,
      "num_tokens": 27320861.0,
      "step": 13485
    },
    {
      "entropy": 6.377088594436645,
      "epoch": 1.1788866555973083,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048667113776932453,
      "loss": 6.0479,
      "mean_token_accuracy": 0.14129312187433243,
      "num_tokens": 27331771.0,
      "step": 13490
    },
    {
      "entropy": 6.296218729019165,
      "epoch": 1.1793236039500132,
      "grad_norm": 5.1875,
      "learning_rate": 0.0004866605698996772,
      "loss": 6.0308,
      "mean_token_accuracy": 0.14158814772963524,
      "num_tokens": 27341449.0,
      "step": 13495
    },
    {
      "entropy": 6.285170793533325,
      "epoch": 1.1797605523027177,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004866499979702691,
      "loss": 6.1006,
      "mean_token_accuracy": 0.13451351150870322,
      "num_tokens": 27353127.0,
      "step": 13500
    },
    {
      "entropy": 6.330408048629761,
      "epoch": 1.1801975006554226,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004866394219813032,
      "loss": 5.9778,
      "mean_token_accuracy": 0.14553027898073195,
      "num_tokens": 27362116.0,
      "step": 13505
    },
    {
      "entropy": 6.271252250671386,
      "epoch": 1.1806344490081273,
      "grad_norm": 4.03125,
      "learning_rate": 0.00048662884193298233,
      "loss": 5.9919,
      "mean_token_accuracy": 0.15086607560515403,
      "num_tokens": 27372204.0,
      "step": 13510
    },
    {
      "entropy": 6.293806028366089,
      "epoch": 1.181071397360832,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004866182578255094,
      "loss": 6.0443,
      "mean_token_accuracy": 0.14480771049857138,
      "num_tokens": 27382963.0,
      "step": 13515
    },
    {
      "entropy": 6.27120156288147,
      "epoch": 1.1815083457135367,
      "grad_norm": 3.734375,
      "learning_rate": 0.00048660766965908736,
      "loss": 5.9905,
      "mean_token_accuracy": 0.14402315989136696,
      "num_tokens": 27392426.0,
      "step": 13520
    },
    {
      "entropy": 6.238733291625977,
      "epoch": 1.1819452940662414,
      "grad_norm": 4.0,
      "learning_rate": 0.0004865970774339194,
      "loss": 5.9907,
      "mean_token_accuracy": 0.14505817294120787,
      "num_tokens": 27402679.0,
      "step": 13525
    },
    {
      "entropy": 6.26131272315979,
      "epoch": 1.182382242418946,
      "grad_norm": 8.3125,
      "learning_rate": 0.00048658648115020855,
      "loss": 6.0456,
      "mean_token_accuracy": 0.14628567844629287,
      "num_tokens": 27412911.0,
      "step": 13530
    },
    {
      "entropy": 6.297979688644409,
      "epoch": 1.1828191907716508,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004865758808081581,
      "loss": 5.9777,
      "mean_token_accuracy": 0.14572669565677643,
      "num_tokens": 27423394.0,
      "step": 13535
    },
    {
      "entropy": 6.304268312454224,
      "epoch": 1.1832561391243555,
      "grad_norm": 3.296875,
      "learning_rate": 0.00048656527640797144,
      "loss": 5.9633,
      "mean_token_accuracy": 0.14959025979042054,
      "num_tokens": 27433016.0,
      "step": 13540
    },
    {
      "entropy": 6.3181616306304935,
      "epoch": 1.1836930874770601,
      "grad_norm": 3.65625,
      "learning_rate": 0.00048655466794985194,
      "loss": 6.0131,
      "mean_token_accuracy": 0.13816590160131453,
      "num_tokens": 27442414.0,
      "step": 13545
    },
    {
      "entropy": 6.203185319900513,
      "epoch": 1.1841300358297648,
      "grad_norm": 3.140625,
      "learning_rate": 0.000486544055434003,
      "loss": 6.0144,
      "mean_token_accuracy": 0.1459573432803154,
      "num_tokens": 27452410.0,
      "step": 13550
    },
    {
      "entropy": 6.315072107315063,
      "epoch": 1.1845669841824695,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004865334388606282,
      "loss": 5.9659,
      "mean_token_accuracy": 0.15402016043663025,
      "num_tokens": 27462196.0,
      "step": 13555
    },
    {
      "entropy": 6.261396741867065,
      "epoch": 1.1850039325351744,
      "grad_norm": 3.34375,
      "learning_rate": 0.00048652281822993127,
      "loss": 6.0718,
      "mean_token_accuracy": 0.1359599158167839,
      "num_tokens": 27473934.0,
      "step": 13560
    },
    {
      "entropy": 6.314400053024292,
      "epoch": 1.1854408808878791,
      "grad_norm": 5.5,
      "learning_rate": 0.00048651219354211585,
      "loss": 6.1121,
      "mean_token_accuracy": 0.13928996548056602,
      "num_tokens": 27484218.0,
      "step": 13565
    },
    {
      "entropy": 6.33268780708313,
      "epoch": 1.1858778292405838,
      "grad_norm": 8.9375,
      "learning_rate": 0.00048650156479738575,
      "loss": 6.1073,
      "mean_token_accuracy": 0.14311506375670432,
      "num_tokens": 27494663.0,
      "step": 13570
    },
    {
      "entropy": 6.306205224990845,
      "epoch": 1.1863147775932885,
      "grad_norm": 2.984375,
      "learning_rate": 0.00048649093199594473,
      "loss": 5.9415,
      "mean_token_accuracy": 0.14839862585067748,
      "num_tokens": 27504233.0,
      "step": 13575
    },
    {
      "entropy": 6.248500919342041,
      "epoch": 1.1867517259459932,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004864802951379969,
      "loss": 6.0096,
      "mean_token_accuracy": 0.14615406468510628,
      "num_tokens": 27514032.0,
      "step": 13580
    },
    {
      "entropy": 6.271261930465698,
      "epoch": 1.187188674298698,
      "grad_norm": 4.28125,
      "learning_rate": 0.00048646965422374615,
      "loss": 6.0909,
      "mean_token_accuracy": 0.13907042592763902,
      "num_tokens": 27524659.0,
      "step": 13585
    },
    {
      "entropy": 6.289261388778686,
      "epoch": 1.1876256226514026,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004864590092533966,
      "loss": 5.9568,
      "mean_token_accuracy": 0.14670784249901772,
      "num_tokens": 27534003.0,
      "step": 13590
    },
    {
      "entropy": 6.26062126159668,
      "epoch": 1.1880625710041073,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004864483602271524,
      "loss": 5.9739,
      "mean_token_accuracy": 0.13964569121599196,
      "num_tokens": 27543586.0,
      "step": 13595
    },
    {
      "entropy": 6.215987014770508,
      "epoch": 1.188499519356812,
      "grad_norm": 3.21875,
      "learning_rate": 0.000486437707145218,
      "loss": 5.9521,
      "mean_token_accuracy": 0.15114705488085747,
      "num_tokens": 27554210.0,
      "step": 13600
    },
    {
      "entropy": 6.3459361553192135,
      "epoch": 1.1889364677095167,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048642705000779754,
      "loss": 6.0676,
      "mean_token_accuracy": 0.14533160403370857,
      "num_tokens": 27565324.0,
      "step": 13605
    },
    {
      "entropy": 6.2655524730682375,
      "epoch": 1.1893734160622214,
      "grad_norm": 4.25,
      "learning_rate": 0.0004864163888150955,
      "loss": 5.995,
      "mean_token_accuracy": 0.1483528234064579,
      "num_tokens": 27575192.0,
      "step": 13610
    },
    {
      "entropy": 6.237028408050537,
      "epoch": 1.1898103644149263,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048640572356731623,
      "loss": 5.994,
      "mean_token_accuracy": 0.1405591756105423,
      "num_tokens": 27585294.0,
      "step": 13615
    },
    {
      "entropy": 6.29896879196167,
      "epoch": 1.190247312767631,
      "grad_norm": 4.25,
      "learning_rate": 0.00048639505426466447,
      "loss": 5.9996,
      "mean_token_accuracy": 0.1507054641842842,
      "num_tokens": 27595667.0,
      "step": 13620
    },
    {
      "entropy": 6.30621747970581,
      "epoch": 1.1906842611203357,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004863843809073448,
      "loss": 5.9813,
      "mean_token_accuracy": 0.14936920553445815,
      "num_tokens": 27605816.0,
      "step": 13625
    },
    {
      "entropy": 6.260953187942505,
      "epoch": 1.1911212094730403,
      "grad_norm": 3.921875,
      "learning_rate": 0.000486373703495562,
      "loss": 5.9993,
      "mean_token_accuracy": 0.14436556920409202,
      "num_tokens": 27614874.0,
      "step": 13630
    },
    {
      "entropy": 6.293641901016235,
      "epoch": 1.191558157825745,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004863630220295207,
      "loss": 5.9828,
      "mean_token_accuracy": 0.14630975127220153,
      "num_tokens": 27625610.0,
      "step": 13635
    },
    {
      "entropy": 6.23807373046875,
      "epoch": 1.1919951061784497,
      "grad_norm": 4.375,
      "learning_rate": 0.00048635233650942596,
      "loss": 6.0188,
      "mean_token_accuracy": 0.1418534517288208,
      "num_tokens": 27636646.0,
      "step": 13640
    },
    {
      "entropy": 6.261193895339966,
      "epoch": 1.1924320545311544,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048634164693548267,
      "loss": 6.0208,
      "mean_token_accuracy": 0.1407660536468029,
      "num_tokens": 27647311.0,
      "step": 13645
    },
    {
      "entropy": 6.265842485427856,
      "epoch": 1.192869002883859,
      "grad_norm": 3.125,
      "learning_rate": 0.0004863309533078957,
      "loss": 6.0144,
      "mean_token_accuracy": 0.1414756953716278,
      "num_tokens": 27658347.0,
      "step": 13650
    },
    {
      "entropy": 6.2654420852661135,
      "epoch": 1.1933059512365638,
      "grad_norm": 4.03125,
      "learning_rate": 0.00048632025562687047,
      "loss": 5.9692,
      "mean_token_accuracy": 0.15064362138509751,
      "num_tokens": 27669211.0,
      "step": 13655
    },
    {
      "entropy": 6.277724552154541,
      "epoch": 1.1937428995892685,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048630955389261193,
      "loss": 5.9855,
      "mean_token_accuracy": 0.14413262605667115,
      "num_tokens": 27678904.0,
      "step": 13660
    },
    {
      "entropy": 6.271093654632568,
      "epoch": 1.1941798479419732,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048629884810532547,
      "loss": 5.9619,
      "mean_token_accuracy": 0.14453675001859664,
      "num_tokens": 27688928.0,
      "step": 13665
    },
    {
      "entropy": 6.2728564739227295,
      "epoch": 1.1946167962946779,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004862881382652164,
      "loss": 6.0518,
      "mean_token_accuracy": 0.142328742146492,
      "num_tokens": 27699855.0,
      "step": 13670
    },
    {
      "entropy": 6.311590051651001,
      "epoch": 1.1950537446473826,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004862774243724901,
      "loss": 6.0073,
      "mean_token_accuracy": 0.15124157220125198,
      "num_tokens": 27709857.0,
      "step": 13675
    },
    {
      "entropy": 6.304900026321411,
      "epoch": 1.1954906930000875,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004862667064273521,
      "loss": 6.0575,
      "mean_token_accuracy": 0.13922526836395263,
      "num_tokens": 27719612.0,
      "step": 13680
    },
    {
      "entropy": 6.206588649749756,
      "epoch": 1.1959276413527922,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004862559844300079,
      "loss": 6.0379,
      "mean_token_accuracy": 0.14596628844738008,
      "num_tokens": 27730052.0,
      "step": 13685
    },
    {
      "entropy": 6.275610542297363,
      "epoch": 1.1963645897054969,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004862452583806633,
      "loss": 5.9022,
      "mean_token_accuracy": 0.15784385651350022,
      "num_tokens": 27739980.0,
      "step": 13690
    },
    {
      "entropy": 6.2699493885040285,
      "epoch": 1.1968015380582016,
      "grad_norm": 3.421875,
      "learning_rate": 0.00048623452827952385,
      "loss": 5.9984,
      "mean_token_accuracy": 0.14317866563796997,
      "num_tokens": 27748863.0,
      "step": 13695
    },
    {
      "entropy": 6.20341911315918,
      "epoch": 1.1972384864109062,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004862237941267956,
      "loss": 5.9607,
      "mean_token_accuracy": 0.15146528631448747,
      "num_tokens": 27758672.0,
      "step": 13700
    },
    {
      "entropy": 6.287423086166382,
      "epoch": 1.197675434763611,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004862130559226842,
      "loss": 5.979,
      "mean_token_accuracy": 0.15019524842500687,
      "num_tokens": 27768195.0,
      "step": 13705
    },
    {
      "entropy": 6.235621118545533,
      "epoch": 1.1981123831163156,
      "grad_norm": 3.75,
      "learning_rate": 0.0004862023136673958,
      "loss": 5.9412,
      "mean_token_accuracy": 0.14754905104637145,
      "num_tokens": 27778135.0,
      "step": 13710
    },
    {
      "entropy": 6.32276496887207,
      "epoch": 1.1985493314690203,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048619156736113634,
      "loss": 6.0331,
      "mean_token_accuracy": 0.14190757200121878,
      "num_tokens": 27788723.0,
      "step": 13715
    },
    {
      "entropy": 6.322002983093261,
      "epoch": 1.198986279821725,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004861808170041119,
      "loss": 6.0343,
      "mean_token_accuracy": 0.13997166603803635,
      "num_tokens": 27799060.0,
      "step": 13720
    },
    {
      "entropy": 6.332482624053955,
      "epoch": 1.1994232281744297,
      "grad_norm": 4.5,
      "learning_rate": 0.0004861700625965287,
      "loss": 6.0311,
      "mean_token_accuracy": 0.13932159468531607,
      "num_tokens": 27809542.0,
      "step": 13725
    },
    {
      "entropy": 6.2244843482971195,
      "epoch": 1.1998601765271344,
      "grad_norm": 3.515625,
      "learning_rate": 0.00048615930413859317,
      "loss": 6.0251,
      "mean_token_accuracy": 0.14485367015004158,
      "num_tokens": 27819738.0,
      "step": 13730
    },
    {
      "entropy": 6.308111000061035,
      "epoch": 1.2002971248798393,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048614854163051143,
      "loss": 5.9693,
      "mean_token_accuracy": 0.14438681155443192,
      "num_tokens": 27829052.0,
      "step": 13735
    },
    {
      "entropy": 6.268362998962402,
      "epoch": 1.200734073232544,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004861377750724901,
      "loss": 5.977,
      "mean_token_accuracy": 0.14190709069371224,
      "num_tokens": 27839277.0,
      "step": 13740
    },
    {
      "entropy": 6.259753799438476,
      "epoch": 1.2011710215852487,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004861270044647356,
      "loss": 6.0049,
      "mean_token_accuracy": 0.1425544522702694,
      "num_tokens": 27848550.0,
      "step": 13745
    },
    {
      "entropy": 6.276605463027954,
      "epoch": 1.2016079699379534,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048611622980745446,
      "loss": 6.037,
      "mean_token_accuracy": 0.13692583218216897,
      "num_tokens": 27858546.0,
      "step": 13750
    },
    {
      "entropy": 6.322558927536011,
      "epoch": 1.202044918290658,
      "grad_norm": 3.375,
      "learning_rate": 0.00048610545110085347,
      "loss": 6.0501,
      "mean_token_accuracy": 0.14019499868154525,
      "num_tokens": 27869192.0,
      "step": 13755
    },
    {
      "entropy": 6.313315296173096,
      "epoch": 1.2024818666433628,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004860946683451393,
      "loss": 6.009,
      "mean_token_accuracy": 0.14771492630243302,
      "num_tokens": 27878534.0,
      "step": 13760
    },
    {
      "entropy": 6.257963943481445,
      "epoch": 1.2029188149960675,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048608388154051875,
      "loss": 5.9354,
      "mean_token_accuracy": 0.1470448538661003,
      "num_tokens": 27888074.0,
      "step": 13765
    },
    {
      "entropy": 6.2740191459655765,
      "epoch": 1.2033557633487721,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048607309068719875,
      "loss": 5.9794,
      "mean_token_accuracy": 0.14136605858802795,
      "num_tokens": 27898210.0,
      "step": 13770
    },
    {
      "entropy": 6.21459550857544,
      "epoch": 1.2037927117014768,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004860622957853863,
      "loss": 5.9183,
      "mean_token_accuracy": 0.1530967101454735,
      "num_tokens": 27907472.0,
      "step": 13775
    },
    {
      "entropy": 6.197533369064331,
      "epoch": 1.2042296600541815,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004860514968352885,
      "loss": 6.0,
      "mean_token_accuracy": 0.14631540924310685,
      "num_tokens": 27917265.0,
      "step": 13780
    },
    {
      "entropy": 6.2461585521698,
      "epoch": 1.2046666084068862,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048604069383711225,
      "loss": 5.9617,
      "mean_token_accuracy": 0.15335527211427688,
      "num_tokens": 27926883.0,
      "step": 13785
    },
    {
      "entropy": 6.243220090866089,
      "epoch": 1.2051035567595911,
      "grad_norm": 3.0625,
      "learning_rate": 0.00048602988679106497,
      "loss": 5.9859,
      "mean_token_accuracy": 0.14618429765105248,
      "num_tokens": 27937167.0,
      "step": 13790
    },
    {
      "entropy": 6.245713472366333,
      "epoch": 1.2055405051122958,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004860190756973539,
      "loss": 5.9663,
      "mean_token_accuracy": 0.152706079185009,
      "num_tokens": 27947422.0,
      "step": 13795
    },
    {
      "entropy": 6.2771812915802006,
      "epoch": 1.2059774534650005,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048600826055618634,
      "loss": 6.042,
      "mean_token_accuracy": 0.14702495634555818,
      "num_tokens": 27958407.0,
      "step": 13800
    },
    {
      "entropy": 6.262581014633179,
      "epoch": 1.2064144018177052,
      "grad_norm": 2.953125,
      "learning_rate": 0.00048599744136776986,
      "loss": 5.8939,
      "mean_token_accuracy": 0.15216957926750183,
      "num_tokens": 27968277.0,
      "step": 13805
    },
    {
      "entropy": 6.207670783996582,
      "epoch": 1.20685135017041,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048598661813231177,
      "loss": 5.901,
      "mean_token_accuracy": 0.15533718913793565,
      "num_tokens": 27978116.0,
      "step": 13810
    },
    {
      "entropy": 6.216767740249634,
      "epoch": 1.2072882985231146,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048597579085001984,
      "loss": 5.9353,
      "mean_token_accuracy": 0.14771691113710403,
      "num_tokens": 27988383.0,
      "step": 13815
    },
    {
      "entropy": 6.220984077453613,
      "epoch": 1.2077252468758193,
      "grad_norm": 3.15625,
      "learning_rate": 0.00048596495952110176,
      "loss": 5.9375,
      "mean_token_accuracy": 0.15286461487412453,
      "num_tokens": 27999118.0,
      "step": 13820
    },
    {
      "entropy": 6.245439100265503,
      "epoch": 1.208162195228524,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004859541241457651,
      "loss": 5.9488,
      "mean_token_accuracy": 0.15157197043299675,
      "num_tokens": 28008849.0,
      "step": 13825
    },
    {
      "entropy": 6.231414890289306,
      "epoch": 1.2085991435812287,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004859432847242179,
      "loss": 6.0176,
      "mean_token_accuracy": 0.14566791355609893,
      "num_tokens": 28019063.0,
      "step": 13830
    },
    {
      "entropy": 6.259872150421143,
      "epoch": 1.2090360919339334,
      "grad_norm": 3.84375,
      "learning_rate": 0.00048593244125666793,
      "loss": 5.9167,
      "mean_token_accuracy": 0.14634425491094588,
      "num_tokens": 28028333.0,
      "step": 13835
    },
    {
      "entropy": 6.267088890075684,
      "epoch": 1.209473040286638,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004859215937433232,
      "loss": 5.9876,
      "mean_token_accuracy": 0.1460411384701729,
      "num_tokens": 28037704.0,
      "step": 13840
    },
    {
      "entropy": 6.301755428314209,
      "epoch": 1.2099099886393427,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004859107421843918,
      "loss": 6.0234,
      "mean_token_accuracy": 0.1450834810733795,
      "num_tokens": 28047157.0,
      "step": 13845
    },
    {
      "entropy": 6.252170610427856,
      "epoch": 1.2103469369920474,
      "grad_norm": 3.65625,
      "learning_rate": 0.00048589988658008184,
      "loss": 5.9508,
      "mean_token_accuracy": 0.14288986399769782,
      "num_tokens": 28057582.0,
      "step": 13850
    },
    {
      "entropy": 6.213747930526734,
      "epoch": 1.2107838853447523,
      "grad_norm": 2.984375,
      "learning_rate": 0.00048588902693060156,
      "loss": 6.0016,
      "mean_token_accuracy": 0.13579162806272507,
      "num_tokens": 28067458.0,
      "step": 13855
    },
    {
      "entropy": 6.3345701694488525,
      "epoch": 1.211220833697457,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004858781632361592,
      "loss": 5.999,
      "mean_token_accuracy": 0.14632351770997049,
      "num_tokens": 28078967.0,
      "step": 13860
    },
    {
      "entropy": 6.239714097976685,
      "epoch": 1.2116577820501617,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004858672954969632,
      "loss": 5.9278,
      "mean_token_accuracy": 0.15170783996582032,
      "num_tokens": 28088498.0,
      "step": 13865
    },
    {
      "entropy": 6.130061054229737,
      "epoch": 1.2120947304028664,
      "grad_norm": 2.25,
      "learning_rate": 0.00048585642371322196,
      "loss": 6.0295,
      "mean_token_accuracy": 0.14916756898164749,
      "num_tokens": 28098884.0,
      "step": 13870
    },
    {
      "entropy": 6.272214603424072,
      "epoch": 1.212531678755571,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004858455478851441,
      "loss": 5.9465,
      "mean_token_accuracy": 0.1504145957529545,
      "num_tokens": 28109629.0,
      "step": 13875
    },
    {
      "entropy": 6.247396087646484,
      "epoch": 1.2129686271082758,
      "grad_norm": 2.515625,
      "learning_rate": 0.000485834668012938,
      "loss": 5.9359,
      "mean_token_accuracy": 0.1393904335796833,
      "num_tokens": 28118555.0,
      "step": 13880
    },
    {
      "entropy": 6.303026342391968,
      "epoch": 1.2134055754609805,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004858237840968127,
      "loss": 6.055,
      "mean_token_accuracy": 0.14219084233045579,
      "num_tokens": 28129028.0,
      "step": 13885
    },
    {
      "entropy": 6.221658945083618,
      "epoch": 1.2138425238136852,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048581289613697654,
      "loss": 6.0146,
      "mean_token_accuracy": 0.14174478203058244,
      "num_tokens": 28139188.0,
      "step": 13890
    },
    {
      "entropy": 6.289360189437867,
      "epoch": 1.2142794721663899,
      "grad_norm": 4.375,
      "learning_rate": 0.00048580200413363866,
      "loss": 5.9601,
      "mean_token_accuracy": 0.14892472326755524,
      "num_tokens": 28148199.0,
      "step": 13895
    },
    {
      "entropy": 6.253590059280396,
      "epoch": 1.2147164205190946,
      "grad_norm": 5.0,
      "learning_rate": 0.0004857911080870079,
      "loss": 5.9498,
      "mean_token_accuracy": 0.1458425983786583,
      "num_tokens": 28157811.0,
      "step": 13900
    },
    {
      "entropy": 6.145600414276123,
      "epoch": 1.2151533688717993,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048578020799729327,
      "loss": 5.9401,
      "mean_token_accuracy": 0.15043851882219314,
      "num_tokens": 28168644.0,
      "step": 13905
    },
    {
      "entropy": 6.220997905731201,
      "epoch": 1.2155903172245042,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004857693038647037,
      "loss": 6.0229,
      "mean_token_accuracy": 0.14640347808599471,
      "num_tokens": 28179846.0,
      "step": 13910
    },
    {
      "entropy": 6.252190732955933,
      "epoch": 1.2160272655772089,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004857583956894485,
      "loss": 5.9523,
      "mean_token_accuracy": 0.1502534717321396,
      "num_tokens": 28189400.0,
      "step": 13915
    },
    {
      "entropy": 6.268828344345093,
      "epoch": 1.2164642139299136,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004857474834717368,
      "loss": 6.0064,
      "mean_token_accuracy": 0.14846311509609222,
      "num_tokens": 28199695.0,
      "step": 13920
    },
    {
      "entropy": 6.211232995986938,
      "epoch": 1.2169011622826182,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048573656721177805,
      "loss": 5.9545,
      "mean_token_accuracy": 0.15119586437940596,
      "num_tokens": 28209977.0,
      "step": 13925
    },
    {
      "entropy": 6.30388822555542,
      "epoch": 1.217338110635323,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004857256469097814,
      "loss": 6.0191,
      "mean_token_accuracy": 0.143757101893425,
      "num_tokens": 28220619.0,
      "step": 13930
    },
    {
      "entropy": 6.291887712478638,
      "epoch": 1.2177750589880276,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004857147225659564,
      "loss": 5.9715,
      "mean_token_accuracy": 0.1483130320906639,
      "num_tokens": 28230065.0,
      "step": 13935
    },
    {
      "entropy": 6.159342908859253,
      "epoch": 1.2182120073407323,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048570379418051273,
      "loss": 5.9687,
      "mean_token_accuracy": 0.14998046159744263,
      "num_tokens": 28239307.0,
      "step": 13940
    },
    {
      "entropy": 6.245069789886474,
      "epoch": 1.218648955693437,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004856928617536598,
      "loss": 5.9366,
      "mean_token_accuracy": 0.14555153250694275,
      "num_tokens": 28249421.0,
      "step": 13945
    },
    {
      "entropy": 6.180024719238281,
      "epoch": 1.2190859040461417,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004856819252856073,
      "loss": 5.9111,
      "mean_token_accuracy": 0.14603402614593505,
      "num_tokens": 28258125.0,
      "step": 13950
    },
    {
      "entropy": 6.136387062072754,
      "epoch": 1.2195228523988464,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048567098477656526,
      "loss": 5.9369,
      "mean_token_accuracy": 0.1469384476542473,
      "num_tokens": 28268897.0,
      "step": 13955
    },
    {
      "entropy": 6.269877147674561,
      "epoch": 1.219959800751551,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004856600402267432,
      "loss": 5.9812,
      "mean_token_accuracy": 0.1492268569767475,
      "num_tokens": 28278506.0,
      "step": 13960
    },
    {
      "entropy": 6.281660652160644,
      "epoch": 1.220396749104256,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004856490916363513,
      "loss": 5.9333,
      "mean_token_accuracy": 0.14631841480731964,
      "num_tokens": 28287905.0,
      "step": 13965
    },
    {
      "entropy": 6.196569585800171,
      "epoch": 1.2208336974569605,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004856381390055994,
      "loss": 6.0299,
      "mean_token_accuracy": 0.14677602499723436,
      "num_tokens": 28297123.0,
      "step": 13970
    },
    {
      "entropy": 6.218492126464843,
      "epoch": 1.2212706458096654,
      "grad_norm": 4.96875,
      "learning_rate": 0.00048562718233469756,
      "loss": 6.0277,
      "mean_token_accuracy": 0.13925908803939818,
      "num_tokens": 28306621.0,
      "step": 13975
    },
    {
      "entropy": 6.336264991760254,
      "epoch": 1.22170759416237,
      "grad_norm": 5.1875,
      "learning_rate": 0.00048561622162385606,
      "loss": 6.1039,
      "mean_token_accuracy": 0.1366852879524231,
      "num_tokens": 28318369.0,
      "step": 13980
    },
    {
      "entropy": 6.292627191543579,
      "epoch": 1.2221445425150748,
      "grad_norm": 3.5,
      "learning_rate": 0.00048560525687328497,
      "loss": 6.0296,
      "mean_token_accuracy": 0.14266280382871627,
      "num_tokens": 28328817.0,
      "step": 13985
    },
    {
      "entropy": 6.265448141098022,
      "epoch": 1.2225814908677795,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048559428808319483,
      "loss": 6.0478,
      "mean_token_accuracy": 0.14210547879338264,
      "num_tokens": 28339448.0,
      "step": 13990
    },
    {
      "entropy": 6.237249135971069,
      "epoch": 1.2230184392204841,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004855833152537958,
      "loss": 5.9595,
      "mean_token_accuracy": 0.14626851826906204,
      "num_tokens": 28349243.0,
      "step": 13995
    },
    {
      "entropy": 6.225575399398804,
      "epoch": 1.2234553875731888,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004855723383852985,
      "loss": 6.0108,
      "mean_token_accuracy": 0.1486746296286583,
      "num_tokens": 28359392.0,
      "step": 14000
    },
    {
      "entropy": 6.262087154388428,
      "epoch": 1.2238923359258935,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004855613574779134,
      "loss": 5.9497,
      "mean_token_accuracy": 0.1495212972164154,
      "num_tokens": 28369160.0,
      "step": 14005
    },
    {
      "entropy": 6.234268236160278,
      "epoch": 1.2243292842785982,
      "grad_norm": 3.546875,
      "learning_rate": 0.00048555037253185114,
      "loss": 6.01,
      "mean_token_accuracy": 0.14652542844414712,
      "num_tokens": 28379107.0,
      "step": 14010
    },
    {
      "entropy": 6.32548713684082,
      "epoch": 1.224766232631303,
      "grad_norm": 3.921875,
      "learning_rate": 0.00048553938354732243,
      "loss": 6.0027,
      "mean_token_accuracy": 0.14862387031316757,
      "num_tokens": 28388956.0,
      "step": 14015
    },
    {
      "entropy": 6.2433404445648195,
      "epoch": 1.2252031809840076,
      "grad_norm": 3.9375,
      "learning_rate": 0.00048552839052453794,
      "loss": 5.9727,
      "mean_token_accuracy": 0.14156188070774078,
      "num_tokens": 28398925.0,
      "step": 14020
    },
    {
      "entropy": 6.243257999420166,
      "epoch": 1.2256401293367123,
      "grad_norm": 4.40625,
      "learning_rate": 0.00048551739346370864,
      "loss": 6.0594,
      "mean_token_accuracy": 0.13862217515707015,
      "num_tokens": 28408714.0,
      "step": 14025
    },
    {
      "entropy": 6.315703535079956,
      "epoch": 1.2260770776894172,
      "grad_norm": 5.46875,
      "learning_rate": 0.0004855063923650455,
      "loss": 6.0237,
      "mean_token_accuracy": 0.1488606944680214,
      "num_tokens": 28419408.0,
      "step": 14030
    },
    {
      "entropy": 6.274506664276123,
      "epoch": 1.226514026042122,
      "grad_norm": 3.25,
      "learning_rate": 0.0004854953872287593,
      "loss": 5.9683,
      "mean_token_accuracy": 0.15097141414880752,
      "num_tokens": 28430044.0,
      "step": 14035
    },
    {
      "entropy": 6.247838115692138,
      "epoch": 1.2269509743948266,
      "grad_norm": 2.5,
      "learning_rate": 0.0004854843780550614,
      "loss": 5.8997,
      "mean_token_accuracy": 0.1495814338326454,
      "num_tokens": 28440100.0,
      "step": 14040
    },
    {
      "entropy": 6.211548757553101,
      "epoch": 1.2273879227475313,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004854733648441628,
      "loss": 5.9249,
      "mean_token_accuracy": 0.15269957780838012,
      "num_tokens": 28449497.0,
      "step": 14045
    },
    {
      "entropy": 6.228044652938843,
      "epoch": 1.227824871100236,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048546234759627477,
      "loss": 5.9676,
      "mean_token_accuracy": 0.15399148613214492,
      "num_tokens": 28459789.0,
      "step": 14050
    },
    {
      "entropy": 6.331049013137817,
      "epoch": 1.2282618194529407,
      "grad_norm": 3.09375,
      "learning_rate": 0.00048545132631160856,
      "loss": 6.0365,
      "mean_token_accuracy": 0.13970703333616258,
      "num_tokens": 28471728.0,
      "step": 14055
    },
    {
      "entropy": 6.297502517700195,
      "epoch": 1.2286987678056454,
      "grad_norm": 2.875,
      "learning_rate": 0.00048544030099037576,
      "loss": 6.0329,
      "mean_token_accuracy": 0.15137740969657898,
      "num_tokens": 28481876.0,
      "step": 14060
    },
    {
      "entropy": 6.234708404541015,
      "epoch": 1.22913571615835,
      "grad_norm": 3.46875,
      "learning_rate": 0.00048542927163278766,
      "loss": 5.9927,
      "mean_token_accuracy": 0.1487890213727951,
      "num_tokens": 28491800.0,
      "step": 14065
    },
    {
      "entropy": 6.215711784362793,
      "epoch": 1.2295726645110547,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004854182382390559,
      "loss": 5.9442,
      "mean_token_accuracy": 0.14995943903923034,
      "num_tokens": 28502033.0,
      "step": 14070
    },
    {
      "entropy": 6.272277212142944,
      "epoch": 1.2300096128637594,
      "grad_norm": 3.1875,
      "learning_rate": 0.000485407200809392,
      "loss": 6.0103,
      "mean_token_accuracy": 0.14350430592894553,
      "num_tokens": 28511921.0,
      "step": 14075
    },
    {
      "entropy": 6.315694379806518,
      "epoch": 1.2304465612164641,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004853961593440078,
      "loss": 5.9685,
      "mean_token_accuracy": 0.15184133797883986,
      "num_tokens": 28521449.0,
      "step": 14080
    },
    {
      "entropy": 6.205685472488403,
      "epoch": 1.230883509569169,
      "grad_norm": 2.78125,
      "learning_rate": 0.000485385113843115,
      "loss": 5.9616,
      "mean_token_accuracy": 0.1476466104388237,
      "num_tokens": 28532088.0,
      "step": 14085
    },
    {
      "entropy": 6.252324914932251,
      "epoch": 1.2313204579218737,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004853740643069255,
      "loss": 6.0081,
      "mean_token_accuracy": 0.1474382147192955,
      "num_tokens": 28541835.0,
      "step": 14090
    },
    {
      "entropy": 6.2742835521698,
      "epoch": 1.2317574062745784,
      "grad_norm": 3.8125,
      "learning_rate": 0.00048536301073565113,
      "loss": 5.9888,
      "mean_token_accuracy": 0.13718758299946784,
      "num_tokens": 28551601.0,
      "step": 14095
    },
    {
      "entropy": 6.213563537597656,
      "epoch": 1.232194354627283,
      "grad_norm": 3.875,
      "learning_rate": 0.0004853519531295041,
      "loss": 5.9643,
      "mean_token_accuracy": 0.15064739435911179,
      "num_tokens": 28561012.0,
      "step": 14100
    },
    {
      "entropy": 6.194061613082885,
      "epoch": 1.2326313029799878,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004853408914886963,
      "loss": 6.0296,
      "mean_token_accuracy": 0.14634308367967605,
      "num_tokens": 28571431.0,
      "step": 14105
    },
    {
      "entropy": 6.195002841949463,
      "epoch": 1.2330682513326925,
      "grad_norm": 4.375,
      "learning_rate": 0.00048532982581344004,
      "loss": 5.9169,
      "mean_token_accuracy": 0.15620483830571175,
      "num_tokens": 28581185.0,
      "step": 14110
    },
    {
      "entropy": 6.296385383605957,
      "epoch": 1.2335051996853972,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004853187561039475,
      "loss": 5.9469,
      "mean_token_accuracy": 0.1503499612212181,
      "num_tokens": 28591164.0,
      "step": 14115
    },
    {
      "entropy": 6.19132981300354,
      "epoch": 1.2339421480381019,
      "grad_norm": 3.84375,
      "learning_rate": 0.00048530768236043094,
      "loss": 5.8716,
      "mean_token_accuracy": 0.15093281716108323,
      "num_tokens": 28600931.0,
      "step": 14120
    },
    {
      "entropy": 6.176803350448608,
      "epoch": 1.2343790963908066,
      "grad_norm": 3.25,
      "learning_rate": 0.0004852966045831029,
      "loss": 6.0221,
      "mean_token_accuracy": 0.14687056690454484,
      "num_tokens": 28611908.0,
      "step": 14125
    },
    {
      "entropy": 6.257868337631225,
      "epoch": 1.2348160447435113,
      "grad_norm": 4.5,
      "learning_rate": 0.0004852855227721758,
      "loss": 6.0337,
      "mean_token_accuracy": 0.14042493999004363,
      "num_tokens": 28622106.0,
      "step": 14130
    },
    {
      "entropy": 6.327805852890014,
      "epoch": 1.235252993096216,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048527443692786217,
      "loss": 6.0467,
      "mean_token_accuracy": 0.14018301665782928,
      "num_tokens": 28632634.0,
      "step": 14135
    },
    {
      "entropy": 6.300795602798462,
      "epoch": 1.2356899414489209,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004852633470503746,
      "loss": 6.003,
      "mean_token_accuracy": 0.1457148589193821,
      "num_tokens": 28642936.0,
      "step": 14140
    },
    {
      "entropy": 6.272733592987061,
      "epoch": 1.2361268898016253,
      "grad_norm": 4.53125,
      "learning_rate": 0.00048525225313992594,
      "loss": 5.9709,
      "mean_token_accuracy": 0.14041142985224725,
      "num_tokens": 28653539.0,
      "step": 14145
    },
    {
      "entropy": 6.2503563404083256,
      "epoch": 1.2365638381543302,
      "grad_norm": 3.75,
      "learning_rate": 0.0004852411551967289,
      "loss": 6.0371,
      "mean_token_accuracy": 0.1395442582666874,
      "num_tokens": 28664212.0,
      "step": 14150
    },
    {
      "entropy": 6.335703754425049,
      "epoch": 1.237000786507035,
      "grad_norm": 6.75,
      "learning_rate": 0.00048523005322099626,
      "loss": 6.0284,
      "mean_token_accuracy": 0.14107294082641603,
      "num_tokens": 28676028.0,
      "step": 14155
    },
    {
      "entropy": 6.316219425201416,
      "epoch": 1.2374377348597396,
      "grad_norm": 4.21875,
      "learning_rate": 0.00048521894721294115,
      "loss": 5.9601,
      "mean_token_accuracy": 0.14468733817338944,
      "num_tokens": 28685370.0,
      "step": 14160
    },
    {
      "entropy": 6.19027099609375,
      "epoch": 1.2378746832124443,
      "grad_norm": 3.234375,
      "learning_rate": 0.00048520783717277645,
      "loss": 5.9345,
      "mean_token_accuracy": 0.1511252775788307,
      "num_tokens": 28695300.0,
      "step": 14165
    },
    {
      "entropy": 6.2153918743133545,
      "epoch": 1.238311631565149,
      "grad_norm": 4.375,
      "learning_rate": 0.00048519672310071523,
      "loss": 5.9861,
      "mean_token_accuracy": 0.14822299033403397,
      "num_tokens": 28705485.0,
      "step": 14170
    },
    {
      "entropy": 6.273805952072143,
      "epoch": 1.2387485799178537,
      "grad_norm": 4.6875,
      "learning_rate": 0.00048518560499697066,
      "loss": 6.0062,
      "mean_token_accuracy": 0.1421967104077339,
      "num_tokens": 28715407.0,
      "step": 14175
    },
    {
      "entropy": 6.293102216720581,
      "epoch": 1.2391855282705584,
      "grad_norm": 2.96875,
      "learning_rate": 0.00048517448286175617,
      "loss": 6.0128,
      "mean_token_accuracy": 0.14855145290493965,
      "num_tokens": 28725283.0,
      "step": 14180
    },
    {
      "entropy": 6.225545024871826,
      "epoch": 1.239622476623263,
      "grad_norm": 2.828125,
      "learning_rate": 0.00048516335669528495,
      "loss": 5.9288,
      "mean_token_accuracy": 0.15194211155176163,
      "num_tokens": 28736131.0,
      "step": 14185
    },
    {
      "entropy": 6.184223604202271,
      "epoch": 1.2400594249759678,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004851522264977704,
      "loss": 5.9566,
      "mean_token_accuracy": 0.1550428330898285,
      "num_tokens": 28745446.0,
      "step": 14190
    },
    {
      "entropy": 6.213590908050537,
      "epoch": 1.2404963733286725,
      "grad_norm": 5.0625,
      "learning_rate": 0.000485141092269426,
      "loss": 5.8697,
      "mean_token_accuracy": 0.15078481361269952,
      "num_tokens": 28754639.0,
      "step": 14195
    },
    {
      "entropy": 6.2128705978393555,
      "epoch": 1.2409333216813772,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004851299540104654,
      "loss": 5.9778,
      "mean_token_accuracy": 0.14906685054302216,
      "num_tokens": 28764397.0,
      "step": 14200
    },
    {
      "entropy": 6.253481054306031,
      "epoch": 1.241370270034082,
      "grad_norm": 3.921875,
      "learning_rate": 0.00048511881172110216,
      "loss": 5.9337,
      "mean_token_accuracy": 0.15423438400030137,
      "num_tokens": 28773680.0,
      "step": 14205
    },
    {
      "entropy": 6.27211332321167,
      "epoch": 1.2418072183867868,
      "grad_norm": 3.875,
      "learning_rate": 0.00048510766540154997,
      "loss": 6.0175,
      "mean_token_accuracy": 0.14772587716579438,
      "num_tokens": 28784034.0,
      "step": 14210
    },
    {
      "entropy": 6.1807698726654055,
      "epoch": 1.2422441667394915,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048509651505202275,
      "loss": 6.0289,
      "mean_token_accuracy": 0.14591992646455765,
      "num_tokens": 28794487.0,
      "step": 14215
    },
    {
      "entropy": 6.30363826751709,
      "epoch": 1.2426811150921961,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004850853606727342,
      "loss": 5.9354,
      "mean_token_accuracy": 0.1513354428112507,
      "num_tokens": 28803942.0,
      "step": 14220
    },
    {
      "entropy": 6.303828716278076,
      "epoch": 1.2431180634449008,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004850742022638984,
      "loss": 5.9627,
      "mean_token_accuracy": 0.1479203909635544,
      "num_tokens": 28814498.0,
      "step": 14225
    },
    {
      "entropy": 6.140864276885987,
      "epoch": 1.2435550117976055,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048506303982572924,
      "loss": 5.9188,
      "mean_token_accuracy": 0.15195949524641036,
      "num_tokens": 28824513.0,
      "step": 14230
    },
    {
      "entropy": 6.238526058197022,
      "epoch": 1.2439919601503102,
      "grad_norm": 3.3125,
      "learning_rate": 0.000485051873358441,
      "loss": 6.067,
      "mean_token_accuracy": 0.13562662824988364,
      "num_tokens": 28835643.0,
      "step": 14235
    },
    {
      "entropy": 6.368589925765991,
      "epoch": 1.244428908503015,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004850407028622477,
      "loss": 6.0901,
      "mean_token_accuracy": 0.1448112078011036,
      "num_tokens": 28845615.0,
      "step": 14240
    },
    {
      "entropy": 6.3137365818023685,
      "epoch": 1.2448658568557196,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048502952833736365,
      "loss": 5.9599,
      "mean_token_accuracy": 0.15123083740472792,
      "num_tokens": 28856152.0,
      "step": 14245
    },
    {
      "entropy": 6.204146671295166,
      "epoch": 1.2453028052084243,
      "grad_norm": 4.71875,
      "learning_rate": 0.00048501834978400323,
      "loss": 5.8941,
      "mean_token_accuracy": 0.15473318323493004,
      "num_tokens": 28866926.0,
      "step": 14250
    },
    {
      "entropy": 6.174571514129639,
      "epoch": 1.245739753561129,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004850071672023809,
      "loss": 5.9667,
      "mean_token_accuracy": 0.14428670853376388,
      "num_tokens": 28877188.0,
      "step": 14255
    },
    {
      "entropy": 6.227392482757568,
      "epoch": 1.246176701913834,
      "grad_norm": 3.71875,
      "learning_rate": 0.00048499598059271087,
      "loss": 5.9897,
      "mean_token_accuracy": 0.1426721140742302,
      "num_tokens": 28887710.0,
      "step": 14260
    },
    {
      "entropy": 6.32489652633667,
      "epoch": 1.2466136502665386,
      "grad_norm": 3.140625,
      "learning_rate": 0.000484984789955208,
      "loss": 6.0368,
      "mean_token_accuracy": 0.14417362809181214,
      "num_tokens": 28897064.0,
      "step": 14265
    },
    {
      "entropy": 6.259010601043701,
      "epoch": 1.2470505986192433,
      "grad_norm": 3.34375,
      "learning_rate": 0.00048497359529008684,
      "loss": 5.9984,
      "mean_token_accuracy": 0.1414656661450863,
      "num_tokens": 28907103.0,
      "step": 14270
    },
    {
      "entropy": 6.283895826339721,
      "epoch": 1.247487546971948,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004849623965975621,
      "loss": 5.9785,
      "mean_token_accuracy": 0.1467199519276619,
      "num_tokens": 28916851.0,
      "step": 14275
    },
    {
      "entropy": 6.253943061828613,
      "epoch": 1.2479244953246527,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004849511938778485,
      "loss": 6.0139,
      "mean_token_accuracy": 0.14170382767915726,
      "num_tokens": 28926638.0,
      "step": 14280
    },
    {
      "entropy": 6.287337923049927,
      "epoch": 1.2483614436773574,
      "grad_norm": 2.90625,
      "learning_rate": 0.00048493998713116103,
      "loss": 5.9725,
      "mean_token_accuracy": 0.14236486032605172,
      "num_tokens": 28936764.0,
      "step": 14285
    },
    {
      "entropy": 6.2817353248596195,
      "epoch": 1.248798392030062,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048492877635771464,
      "loss": 5.9777,
      "mean_token_accuracy": 0.14260494858026504,
      "num_tokens": 28946494.0,
      "step": 14290
    },
    {
      "entropy": 6.240438985824585,
      "epoch": 1.2492353403827667,
      "grad_norm": 2.25,
      "learning_rate": 0.00048491756155772433,
      "loss": 5.9221,
      "mean_token_accuracy": 0.15410663038492203,
      "num_tokens": 28955626.0,
      "step": 14295
    },
    {
      "entropy": 6.245014715194702,
      "epoch": 1.2496722887354714,
      "grad_norm": 3.078125,
      "learning_rate": 0.00048490634273140517,
      "loss": 5.9837,
      "mean_token_accuracy": 0.14560407251119614,
      "num_tokens": 28966317.0,
      "step": 14300
    },
    {
      "entropy": 6.232129001617432,
      "epoch": 1.2501092370881761,
      "grad_norm": 3.203125,
      "learning_rate": 0.00048489511987897236,
      "loss": 6.0096,
      "mean_token_accuracy": 0.14319005459547043,
      "num_tokens": 28976063.0,
      "step": 14305
    },
    {
      "entropy": 6.29182505607605,
      "epoch": 1.2505461854408808,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004848838930006411,
      "loss": 6.0033,
      "mean_token_accuracy": 0.14613813310861587,
      "num_tokens": 28985789.0,
      "step": 14310
    },
    {
      "entropy": 6.2645612239837645,
      "epoch": 1.2509831337935857,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004848726620966269,
      "loss": 6.0423,
      "mean_token_accuracy": 0.14048921316862106,
      "num_tokens": 28996458.0,
      "step": 14315
    },
    {
      "entropy": 6.241276597976684,
      "epoch": 1.2514200821462902,
      "grad_norm": 2.78125,
      "learning_rate": 0.000484861427167145,
      "loss": 5.963,
      "mean_token_accuracy": 0.14824941456317903,
      "num_tokens": 29007641.0,
      "step": 14320
    },
    {
      "entropy": 6.328365182876587,
      "epoch": 1.251857030498995,
      "grad_norm": 3.03125,
      "learning_rate": 0.00048485018821241104,
      "loss": 5.9867,
      "mean_token_accuracy": 0.13739373162388802,
      "num_tokens": 29017351.0,
      "step": 14325
    },
    {
      "entropy": 6.271762847900391,
      "epoch": 1.2522939788516998,
      "grad_norm": 2.484375,
      "learning_rate": 0.00048483894523264045,
      "loss": 6.0167,
      "mean_token_accuracy": 0.14268668591976166,
      "num_tokens": 29026951.0,
      "step": 14330
    },
    {
      "entropy": 6.186942291259766,
      "epoch": 1.2527309272044045,
      "grad_norm": 2.75,
      "learning_rate": 0.0004848276982280489,
      "loss": 5.9927,
      "mean_token_accuracy": 0.14950913786888123,
      "num_tokens": 29037596.0,
      "step": 14335
    },
    {
      "entropy": 6.268832731246948,
      "epoch": 1.2531678755571092,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004848164471988522,
      "loss": 5.9771,
      "mean_token_accuracy": 0.14903145432472228,
      "num_tokens": 29049010.0,
      "step": 14340
    },
    {
      "entropy": 6.319626998901367,
      "epoch": 1.2536048239098139,
      "grad_norm": 3.703125,
      "learning_rate": 0.00048480519214526616,
      "loss": 5.9922,
      "mean_token_accuracy": 0.14408280551433564,
      "num_tokens": 29059028.0,
      "step": 14345
    },
    {
      "entropy": 6.224538421630859,
      "epoch": 1.2540417722625186,
      "grad_norm": 2.984375,
      "learning_rate": 0.00048479393306750647,
      "loss": 5.9942,
      "mean_token_accuracy": 0.14553653225302696,
      "num_tokens": 29069661.0,
      "step": 14350
    },
    {
      "entropy": 6.277307605743408,
      "epoch": 1.2544787206152233,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004847826699657893,
      "loss": 5.9739,
      "mean_token_accuracy": 0.14941245764493943,
      "num_tokens": 29079971.0,
      "step": 14355
    },
    {
      "entropy": 6.280166673660278,
      "epoch": 1.254915668967928,
      "grad_norm": 3.625,
      "learning_rate": 0.00048477140284033054,
      "loss": 6.0248,
      "mean_token_accuracy": 0.14228300079703332,
      "num_tokens": 29091378.0,
      "step": 14360
    },
    {
      "entropy": 6.206805086135864,
      "epoch": 1.2553526173206326,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004847601316913464,
      "loss": 5.9426,
      "mean_token_accuracy": 0.1490841969847679,
      "num_tokens": 29101516.0,
      "step": 14365
    },
    {
      "entropy": 6.22272834777832,
      "epoch": 1.2557895656733375,
      "grad_norm": 3.3125,
      "learning_rate": 0.000484748856519053,
      "loss": 6.0284,
      "mean_token_accuracy": 0.14168782532215118,
      "num_tokens": 29112147.0,
      "step": 14370
    },
    {
      "entropy": 6.219483184814453,
      "epoch": 1.256226514026042,
      "grad_norm": 2.875,
      "learning_rate": 0.0004847375773236667,
      "loss": 5.9122,
      "mean_token_accuracy": 0.1536203183233738,
      "num_tokens": 29122084.0,
      "step": 14375
    },
    {
      "entropy": 6.290182256698609,
      "epoch": 1.256663462378747,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004847262941054037,
      "loss": 6.0026,
      "mean_token_accuracy": 0.14353920072317122,
      "num_tokens": 29131988.0,
      "step": 14380
    },
    {
      "entropy": 6.2619038105010985,
      "epoch": 1.2571004107314516,
      "grad_norm": 2.625,
      "learning_rate": 0.0004847150068644805,
      "loss": 6.0226,
      "mean_token_accuracy": 0.14589735716581345,
      "num_tokens": 29141731.0,
      "step": 14385
    },
    {
      "entropy": 6.2805346012115475,
      "epoch": 1.2575373590841563,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048470371560111353,
      "loss": 6.0463,
      "mean_token_accuracy": 0.14312550500035287,
      "num_tokens": 29152185.0,
      "step": 14390
    },
    {
      "entropy": 6.286951684951783,
      "epoch": 1.257974307436861,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004846924203155194,
      "loss": 5.8929,
      "mean_token_accuracy": 0.15279867202043534,
      "num_tokens": 29162337.0,
      "step": 14395
    },
    {
      "entropy": 6.274919080734253,
      "epoch": 1.2584112557895657,
      "grad_norm": 5.5,
      "learning_rate": 0.00048468112100791483,
      "loss": 6.0244,
      "mean_token_accuracy": 0.14480667635798455,
      "num_tokens": 29173246.0,
      "step": 14400
    },
    {
      "entropy": 6.2412937641143795,
      "epoch": 1.2588482041422704,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004846698176785165,
      "loss": 5.9831,
      "mean_token_accuracy": 0.13923313170671464,
      "num_tokens": 29183700.0,
      "step": 14405
    },
    {
      "entropy": 6.310512638092041,
      "epoch": 1.259285152494975,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004846585103275412,
      "loss": 5.9583,
      "mean_token_accuracy": 0.14318146780133248,
      "num_tokens": 29193686.0,
      "step": 14410
    },
    {
      "entropy": 6.238186264038086,
      "epoch": 1.2597221008476798,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048464719895520577,
      "loss": 6.0146,
      "mean_token_accuracy": 0.1423543483018875,
      "num_tokens": 29203575.0,
      "step": 14415
    },
    {
      "entropy": 6.19022946357727,
      "epoch": 1.2601590492003845,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004846358835617273,
      "loss": 5.9969,
      "mean_token_accuracy": 0.14594607800245285,
      "num_tokens": 29213591.0,
      "step": 14420
    },
    {
      "entropy": 6.282374143600464,
      "epoch": 1.2605959975530892,
      "grad_norm": 5.0,
      "learning_rate": 0.00048462456414732257,
      "loss": 5.9494,
      "mean_token_accuracy": 0.15209436565637588,
      "num_tokens": 29222799.0,
      "step": 14425
    },
    {
      "entropy": 6.2585005283355715,
      "epoch": 1.2610329459057938,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048461324071220896,
      "loss": 5.9613,
      "mean_token_accuracy": 0.14751915261149406,
      "num_tokens": 29232477.0,
      "step": 14430
    },
    {
      "entropy": 6.243257331848144,
      "epoch": 1.2614698942584988,
      "grad_norm": 4.0625,
      "learning_rate": 0.00048460191325660356,
      "loss": 5.956,
      "mean_token_accuracy": 0.14712598621845246,
      "num_tokens": 29243477.0,
      "step": 14435
    },
    {
      "entropy": 6.210073041915893,
      "epoch": 1.2619068426112032,
      "grad_norm": 2.859375,
      "learning_rate": 0.00048459058178072366,
      "loss": 6.023,
      "mean_token_accuracy": 0.13975027278065683,
      "num_tokens": 29253426.0,
      "step": 14440
    },
    {
      "entropy": 6.302518224716186,
      "epoch": 1.2623437909639081,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004845792462847866,
      "loss": 5.9593,
      "mean_token_accuracy": 0.15465997606515886,
      "num_tokens": 29263639.0,
      "step": 14445
    },
    {
      "entropy": 6.277014875411988,
      "epoch": 1.2627807393166128,
      "grad_norm": 7.3125,
      "learning_rate": 0.00048456790676900963,
      "loss": 6.0065,
      "mean_token_accuracy": 0.14326827749609947,
      "num_tokens": 29273980.0,
      "step": 14450
    },
    {
      "entropy": 6.179599237442017,
      "epoch": 1.2632176876693175,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048455656323361055,
      "loss": 5.9384,
      "mean_token_accuracy": 0.15367192924022674,
      "num_tokens": 29283983.0,
      "step": 14455
    },
    {
      "entropy": 6.328272867202759,
      "epoch": 1.2636546360220222,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004845452156788067,
      "loss": 6.0521,
      "mean_token_accuracy": 0.14356720745563506,
      "num_tokens": 29294460.0,
      "step": 14460
    },
    {
      "entropy": 6.259442234039307,
      "epoch": 1.264091584374727,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004845338641048158,
      "loss": 6.0412,
      "mean_token_accuracy": 0.14344794377684594,
      "num_tokens": 29305788.0,
      "step": 14465
    },
    {
      "entropy": 6.272274494171143,
      "epoch": 1.2645285327274316,
      "grad_norm": 3.1875,
      "learning_rate": 0.00048452250851185565,
      "loss": 6.024,
      "mean_token_accuracy": 0.1406084604561329,
      "num_tokens": 29315457.0,
      "step": 14470
    },
    {
      "entropy": 6.2957679271698,
      "epoch": 1.2649654810801363,
      "grad_norm": 3.0625,
      "learning_rate": 0.000484511148900144,
      "loss": 5.9501,
      "mean_token_accuracy": 0.14590116888284682,
      "num_tokens": 29325713.0,
      "step": 14475
    },
    {
      "entropy": 6.254780197143555,
      "epoch": 1.265402429432841,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004844997852698987,
      "loss": 5.9908,
      "mean_token_accuracy": 0.1452813357114792,
      "num_tokens": 29336489.0,
      "step": 14480
    },
    {
      "entropy": 6.265702533721924,
      "epoch": 1.2658393777855457,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004844884176213377,
      "loss": 5.9797,
      "mean_token_accuracy": 0.14464275687932968,
      "num_tokens": 29345567.0,
      "step": 14485
    },
    {
      "entropy": 6.277397966384887,
      "epoch": 1.2662763261382506,
      "grad_norm": 4.6875,
      "learning_rate": 0.00048447704595467917,
      "loss": 5.9466,
      "mean_token_accuracy": 0.14992308914661406,
      "num_tokens": 29355890.0,
      "step": 14490
    },
    {
      "entropy": 6.288480663299561,
      "epoch": 1.266713274490955,
      "grad_norm": 4.25,
      "learning_rate": 0.00048446567027014104,
      "loss": 6.0145,
      "mean_token_accuracy": 0.14984484016895294,
      "num_tokens": 29365948.0,
      "step": 14495
    },
    {
      "entropy": 6.2530731678009035,
      "epoch": 1.26715022284366,
      "grad_norm": 4.25,
      "learning_rate": 0.0004844542905679417,
      "loss": 5.9432,
      "mean_token_accuracy": 0.14695904403924942,
      "num_tokens": 29375673.0,
      "step": 14500
    },
    {
      "entropy": 6.297976064682007,
      "epoch": 1.2675871711963647,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004844429068482993,
      "loss": 5.9176,
      "mean_token_accuracy": 0.14199995696544648,
      "num_tokens": 29384567.0,
      "step": 14505
    },
    {
      "entropy": 6.229419422149658,
      "epoch": 1.2680241195490694,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048443151911143213,
      "loss": 5.8734,
      "mean_token_accuracy": 0.15397279262542723,
      "num_tokens": 29393747.0,
      "step": 14510
    },
    {
      "entropy": 6.241477584838867,
      "epoch": 1.268461067901774,
      "grad_norm": 4.5,
      "learning_rate": 0.0004844201273575587,
      "loss": 5.9909,
      "mean_token_accuracy": 0.14389295428991317,
      "num_tokens": 29404095.0,
      "step": 14515
    },
    {
      "entropy": 6.215361452102661,
      "epoch": 1.2688980162544787,
      "grad_norm": 3.046875,
      "learning_rate": 0.00048440873158689754,
      "loss": 5.9688,
      "mean_token_accuracy": 0.15107387006282808,
      "num_tokens": 29414450.0,
      "step": 14520
    },
    {
      "entropy": 6.221559000015259,
      "epoch": 1.2693349646071834,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004843973317996672,
      "loss": 5.9573,
      "mean_token_accuracy": 0.14573283046483992,
      "num_tokens": 29424054.0,
      "step": 14525
    },
    {
      "entropy": 6.259769582748413,
      "epoch": 1.2697719129598881,
      "grad_norm": 5.90625,
      "learning_rate": 0.0004843859279960862,
      "loss": 6.0278,
      "mean_token_accuracy": 0.14982415586709977,
      "num_tokens": 29434585.0,
      "step": 14530
    },
    {
      "entropy": 6.21767749786377,
      "epoch": 1.2702088613125928,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004843745201763735,
      "loss": 5.9828,
      "mean_token_accuracy": 0.1544716998934746,
      "num_tokens": 29443863.0,
      "step": 14535
    },
    {
      "entropy": 6.2638946056365965,
      "epoch": 1.2706458096652975,
      "grad_norm": 3.203125,
      "learning_rate": 0.00048436310834074764,
      "loss": 5.9119,
      "mean_token_accuracy": 0.14526710361242295,
      "num_tokens": 29453851.0,
      "step": 14540
    },
    {
      "entropy": 6.22825608253479,
      "epoch": 1.2710827580180022,
      "grad_norm": 6.53125,
      "learning_rate": 0.0004843516924894278,
      "loss": 5.8794,
      "mean_token_accuracy": 0.15693634003400803,
      "num_tokens": 29465152.0,
      "step": 14545
    },
    {
      "entropy": 6.214817380905151,
      "epoch": 1.2715197063707069,
      "grad_norm": 3.78125,
      "learning_rate": 0.00048434027262263274,
      "loss": 5.9058,
      "mean_token_accuracy": 0.15876191705465317,
      "num_tokens": 29474749.0,
      "step": 14550
    },
    {
      "entropy": 6.257167863845825,
      "epoch": 1.2719566547234118,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004843288487405816,
      "loss": 5.9475,
      "mean_token_accuracy": 0.1492299661040306,
      "num_tokens": 29483554.0,
      "step": 14555
    },
    {
      "entropy": 6.236217784881592,
      "epoch": 1.2723936030761165,
      "grad_norm": 4.25,
      "learning_rate": 0.00048431742084349347,
      "loss": 5.9328,
      "mean_token_accuracy": 0.1514594793319702,
      "num_tokens": 29493718.0,
      "step": 14560
    },
    {
      "entropy": 6.239266777038575,
      "epoch": 1.2728305514288212,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004843059889315874,
      "loss": 6.0334,
      "mean_token_accuracy": 0.1416141800582409,
      "num_tokens": 29504407.0,
      "step": 14565
    },
    {
      "entropy": 6.314605808258056,
      "epoch": 1.2732674997815259,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004842945530050829,
      "loss": 6.0173,
      "mean_token_accuracy": 0.13862134143710136,
      "num_tokens": 29514992.0,
      "step": 14570
    },
    {
      "entropy": 6.31018476486206,
      "epoch": 1.2737044481342306,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048428311306419913,
      "loss": 6.0256,
      "mean_token_accuracy": 0.14080001562833785,
      "num_tokens": 29525109.0,
      "step": 14575
    },
    {
      "entropy": 6.259932136535644,
      "epoch": 1.2741413964869353,
      "grad_norm": 5.34375,
      "learning_rate": 0.00048427166910915565,
      "loss": 6.0036,
      "mean_token_accuracy": 0.14481640979647636,
      "num_tokens": 29536131.0,
      "step": 14580
    },
    {
      "entropy": 6.246349477767945,
      "epoch": 1.27457834483964,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004842602211401719,
      "loss": 5.9359,
      "mean_token_accuracy": 0.15553129464387894,
      "num_tokens": 29547614.0,
      "step": 14585
    },
    {
      "entropy": 6.236628818511963,
      "epoch": 1.2750152931923446,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004842487691574675,
      "loss": 5.9811,
      "mean_token_accuracy": 0.1469951570034027,
      "num_tokens": 29557499.0,
      "step": 14590
    },
    {
      "entropy": 6.262730455398559,
      "epoch": 1.2754522415450493,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048423731316126194,
      "loss": 6.0074,
      "mean_token_accuracy": 0.1388388492166996,
      "num_tokens": 29569909.0,
      "step": 14595
    },
    {
      "entropy": 6.334180021286011,
      "epoch": 1.275889189897754,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004842258531517752,
      "loss": 6.1003,
      "mean_token_accuracy": 0.13757336586713792,
      "num_tokens": 29581261.0,
      "step": 14600
    },
    {
      "entropy": 6.309432935714722,
      "epoch": 1.2763261382504587,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004842143891292269,
      "loss": 5.9443,
      "mean_token_accuracy": 0.1518367797136307,
      "num_tokens": 29590604.0,
      "step": 14605
    },
    {
      "entropy": 6.263484382629395,
      "epoch": 1.2767630866031636,
      "grad_norm": 3.84375,
      "learning_rate": 0.00048420292109383704,
      "loss": 6.0239,
      "mean_token_accuracy": 0.14430704414844514,
      "num_tokens": 29600129.0,
      "step": 14610
    },
    {
      "entropy": 6.201832437515259,
      "epoch": 1.277200034955868,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048419144904582545,
      "loss": 5.9019,
      "mean_token_accuracy": 0.15638792365789414,
      "num_tokens": 29609608.0,
      "step": 14615
    },
    {
      "entropy": 6.23317518234253,
      "epoch": 1.277636983308573,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048417997298541235,
      "loss": 5.9012,
      "mean_token_accuracy": 0.14937984347343444,
      "num_tokens": 29618764.0,
      "step": 14620
    },
    {
      "entropy": 6.265982627868652,
      "epoch": 1.2780739316612777,
      "grad_norm": 5.65625,
      "learning_rate": 0.00048416849291281765,
      "loss": 6.0729,
      "mean_token_accuracy": 0.14258292689919472,
      "num_tokens": 29628703.0,
      "step": 14625
    },
    {
      "entropy": 6.290494823455811,
      "epoch": 1.2785108800139824,
      "grad_norm": 3.265625,
      "learning_rate": 0.00048415700882826167,
      "loss": 5.9921,
      "mean_token_accuracy": 0.14364574551582338,
      "num_tokens": 29639291.0,
      "step": 14630
    },
    {
      "entropy": 6.238933372497558,
      "epoch": 1.278947828366687,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048414552073196477,
      "loss": 5.9636,
      "mean_token_accuracy": 0.14596461802721022,
      "num_tokens": 29649058.0,
      "step": 14635
    },
    {
      "entropy": 6.272031354904175,
      "epoch": 1.2793847767193918,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004841340286241471,
      "loss": 6.0009,
      "mean_token_accuracy": 0.14845532402396203,
      "num_tokens": 29658629.0,
      "step": 14640
    },
    {
      "entropy": 6.270854377746582,
      "epoch": 1.2798217250720965,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004841225325050292,
      "loss": 6.057,
      "mean_token_accuracy": 0.14087523594498635,
      "num_tokens": 29668875.0,
      "step": 14645
    },
    {
      "entropy": 6.236531591415405,
      "epoch": 1.2802586734248012,
      "grad_norm": 2.5,
      "learning_rate": 0.00048411103237483156,
      "loss": 6.0217,
      "mean_token_accuracy": 0.14274818822741508,
      "num_tokens": 29678427.0,
      "step": 14650
    },
    {
      "entropy": 6.217003250122071,
      "epoch": 1.2806956217775058,
      "grad_norm": 3.5625,
      "learning_rate": 0.00048409952823377474,
      "loss": 5.9156,
      "mean_token_accuracy": 0.15165874361991882,
      "num_tokens": 29688975.0,
      "step": 14655
    },
    {
      "entropy": 6.30330376625061,
      "epoch": 1.2811325701302105,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048408802008207937,
      "loss": 6.0234,
      "mean_token_accuracy": 0.14318510740995408,
      "num_tokens": 29699550.0,
      "step": 14660
    },
    {
      "entropy": 6.242589092254638,
      "epoch": 1.2815695184829154,
      "grad_norm": 4.375,
      "learning_rate": 0.00048407650791996624,
      "loss": 5.956,
      "mean_token_accuracy": 0.14609276428818702,
      "num_tokens": 29709597.0,
      "step": 14665
    },
    {
      "entropy": 6.281386137008667,
      "epoch": 1.28200646683562,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004840649917476561,
      "loss": 6.0177,
      "mean_token_accuracy": 0.14150036722421647,
      "num_tokens": 29719250.0,
      "step": 14670
    },
    {
      "entropy": 6.334931230545044,
      "epoch": 1.2824434151883248,
      "grad_norm": 2.953125,
      "learning_rate": 0.00048405347156536995,
      "loss": 6.0316,
      "mean_token_accuracy": 0.13955721035599708,
      "num_tokens": 29728809.0,
      "step": 14675
    },
    {
      "entropy": 6.251283979415893,
      "epoch": 1.2828803635410295,
      "grad_norm": 4.5,
      "learning_rate": 0.0004840419473733287,
      "loss": 5.9453,
      "mean_token_accuracy": 0.15335193425416946,
      "num_tokens": 29739240.0,
      "step": 14680
    },
    {
      "entropy": 6.302812910079956,
      "epoch": 1.2833173118937342,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048403041917175323,
      "loss": 5.9929,
      "mean_token_accuracy": 0.1529959723353386,
      "num_tokens": 29749284.0,
      "step": 14685
    },
    {
      "entropy": 6.275013113021851,
      "epoch": 1.283754260246439,
      "grad_norm": 5.3125,
      "learning_rate": 0.00048401888696086485,
      "loss": 6.0365,
      "mean_token_accuracy": 0.14224981814622878,
      "num_tokens": 29758770.0,
      "step": 14690
    },
    {
      "entropy": 6.328972101211548,
      "epoch": 1.2841912085991436,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004840073507408847,
      "loss": 6.0053,
      "mean_token_accuracy": 0.14419203251600266,
      "num_tokens": 29770111.0,
      "step": 14695
    },
    {
      "entropy": 6.222262287139893,
      "epoch": 1.2846281569518483,
      "grad_norm": 3.328125,
      "learning_rate": 0.000483995810512034,
      "loss": 5.9462,
      "mean_token_accuracy": 0.1462593838572502,
      "num_tokens": 29780210.0,
      "step": 14700
    },
    {
      "entropy": 6.244054126739502,
      "epoch": 1.285065105304553,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048398426627453423,
      "loss": 5.9935,
      "mean_token_accuracy": 0.14562774151563646,
      "num_tokens": 29789903.0,
      "step": 14705
    },
    {
      "entropy": 6.343907117843628,
      "epoch": 1.2855020536572577,
      "grad_norm": 3.1875,
      "learning_rate": 0.00048397271802860667,
      "loss": 6.0191,
      "mean_token_accuracy": 0.14820703417062758,
      "num_tokens": 29799377.0,
      "step": 14710
    },
    {
      "entropy": 6.24927248954773,
      "epoch": 1.2859390020099624,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004839611657744729,
      "loss": 5.9501,
      "mean_token_accuracy": 0.1455349564552307,
      "num_tokens": 29809173.0,
      "step": 14715
    },
    {
      "entropy": 6.2046795845031735,
      "epoch": 1.286375950362667,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048394960951235444,
      "loss": 6.0516,
      "mean_token_accuracy": 0.14450864642858505,
      "num_tokens": 29819475.0,
      "step": 14720
    },
    {
      "entropy": 6.249078321456909,
      "epoch": 1.2868128987153717,
      "grad_norm": 3.109375,
      "learning_rate": 0.000483938049242473,
      "loss": 5.9359,
      "mean_token_accuracy": 0.1440932035446167,
      "num_tokens": 29829109.0,
      "step": 14725
    },
    {
      "entropy": 6.3110435009002686,
      "epoch": 1.2872498470680767,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048392648496505024,
      "loss": 5.9469,
      "mean_token_accuracy": 0.14318586885929108,
      "num_tokens": 29839735.0,
      "step": 14730
    },
    {
      "entropy": 6.211352443695068,
      "epoch": 1.2876867954207811,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004839149166803081,
      "loss": 5.9487,
      "mean_token_accuracy": 0.14468227773904802,
      "num_tokens": 29850039.0,
      "step": 14735
    },
    {
      "entropy": 6.21194167137146,
      "epoch": 1.288123743773486,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004839033443884684,
      "loss": 5.9433,
      "mean_token_accuracy": 0.14647406339645386,
      "num_tokens": 29859411.0,
      "step": 14740
    },
    {
      "entropy": 6.283131837844849,
      "epoch": 1.2885606921261907,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004838917680897531,
      "loss": 6.0516,
      "mean_token_accuracy": 0.1403127759695053,
      "num_tokens": 29870426.0,
      "step": 14745
    },
    {
      "entropy": 6.25393271446228,
      "epoch": 1.2889976404788954,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048388018778438413,
      "loss": 5.9858,
      "mean_token_accuracy": 0.14603035598993303,
      "num_tokens": 29881992.0,
      "step": 14750
    },
    {
      "entropy": 6.22723069190979,
      "epoch": 1.2894345888316001,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004838686034725838,
      "loss": 5.9706,
      "mean_token_accuracy": 0.15158261060714723,
      "num_tokens": 29891496.0,
      "step": 14755
    },
    {
      "entropy": 6.264457130432129,
      "epoch": 1.2898715371843048,
      "grad_norm": 2.78125,
      "learning_rate": 0.00048385701515457413,
      "loss": 5.976,
      "mean_token_accuracy": 0.14276368468999862,
      "num_tokens": 29901529.0,
      "step": 14760
    },
    {
      "entropy": 6.26305046081543,
      "epoch": 1.2903084855370095,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004838454228305775,
      "loss": 5.996,
      "mean_token_accuracy": 0.14828526377677917,
      "num_tokens": 29910773.0,
      "step": 14765
    },
    {
      "entropy": 6.1990419864654545,
      "epoch": 1.2907454338897142,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048383382650081633,
      "loss": 5.9945,
      "mean_token_accuracy": 0.14805665761232376,
      "num_tokens": 29919381.0,
      "step": 14770
    },
    {
      "entropy": 6.268475580215454,
      "epoch": 1.2911823822424189,
      "grad_norm": 4.125,
      "learning_rate": 0.0004838222261655129,
      "loss": 6.0073,
      "mean_token_accuracy": 0.1425060138106346,
      "num_tokens": 29929179.0,
      "step": 14775
    },
    {
      "entropy": 6.265458250045777,
      "epoch": 1.2916193305951236,
      "grad_norm": 5.53125,
      "learning_rate": 0.00048381062182488964,
      "loss": 5.959,
      "mean_token_accuracy": 0.1483205407857895,
      "num_tokens": 29938252.0,
      "step": 14780
    },
    {
      "entropy": 6.254838132858277,
      "epoch": 1.2920562789478285,
      "grad_norm": 3.265625,
      "learning_rate": 0.00048379901347916926,
      "loss": 5.9898,
      "mean_token_accuracy": 0.14832838103175164,
      "num_tokens": 29947489.0,
      "step": 14785
    },
    {
      "entropy": 6.223235607147217,
      "epoch": 1.292493227300533,
      "grad_norm": 3.03125,
      "learning_rate": 0.00048378740112857443,
      "loss": 5.9127,
      "mean_token_accuracy": 0.15767193734645843,
      "num_tokens": 29956788.0,
      "step": 14790
    },
    {
      "entropy": 6.254933500289917,
      "epoch": 1.2929301756532379,
      "grad_norm": 2.828125,
      "learning_rate": 0.00048377578477332785,
      "loss": 6.0263,
      "mean_token_accuracy": 0.13800561279058457,
      "num_tokens": 29967161.0,
      "step": 14795
    },
    {
      "entropy": 6.229949617385865,
      "epoch": 1.2933671240059426,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004837641644136523,
      "loss": 5.8578,
      "mean_token_accuracy": 0.16033025830984116,
      "num_tokens": 29976739.0,
      "step": 14800
    },
    {
      "entropy": 6.218859910964966,
      "epoch": 1.2938040723586473,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048375254004977074,
      "loss": 5.9476,
      "mean_token_accuracy": 0.14382242411375046,
      "num_tokens": 29986739.0,
      "step": 14805
    },
    {
      "entropy": 6.2075385570526125,
      "epoch": 1.294241020711352,
      "grad_norm": 4.625,
      "learning_rate": 0.0004837409116819061,
      "loss": 5.9515,
      "mean_token_accuracy": 0.15024821907281877,
      "num_tokens": 29996687.0,
      "step": 14810
    },
    {
      "entropy": 6.196431112289429,
      "epoch": 1.2946779690640566,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004837292793102813,
      "loss": 5.9849,
      "mean_token_accuracy": 0.14989448338747025,
      "num_tokens": 30007757.0,
      "step": 14815
    },
    {
      "entropy": 6.194946432113648,
      "epoch": 1.2951149174167613,
      "grad_norm": 3.390625,
      "learning_rate": 0.00048371764293511963,
      "loss": 5.9379,
      "mean_token_accuracy": 0.16399513632059098,
      "num_tokens": 30018129.0,
      "step": 14820
    },
    {
      "entropy": 6.298979902267456,
      "epoch": 1.295551865769466,
      "grad_norm": 5.59375,
      "learning_rate": 0.0004837060025566441,
      "loss": 5.9489,
      "mean_token_accuracy": 0.14364599138498307,
      "num_tokens": 30028194.0,
      "step": 14825
    },
    {
      "entropy": 6.286139631271363,
      "epoch": 1.2959888141221707,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004836943581750782,
      "loss": 6.0749,
      "mean_token_accuracy": 0.13306139782071114,
      "num_tokens": 30038182.0,
      "step": 14830
    },
    {
      "entropy": 6.270407867431641,
      "epoch": 1.2964257624748754,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004836827097906451,
      "loss": 5.9636,
      "mean_token_accuracy": 0.1489607535302639,
      "num_tokens": 30047848.0,
      "step": 14835
    },
    {
      "entropy": 6.229393672943115,
      "epoch": 1.2968627108275803,
      "grad_norm": 4.28125,
      "learning_rate": 0.00048367105740356825,
      "loss": 6.0707,
      "mean_token_accuracy": 0.14378170743584634,
      "num_tokens": 30058697.0,
      "step": 14840
    },
    {
      "entropy": 6.288531017303467,
      "epoch": 1.2972996591802848,
      "grad_norm": 3.09375,
      "learning_rate": 0.00048365940101407123,
      "loss": 5.8921,
      "mean_token_accuracy": 0.1516134425997734,
      "num_tokens": 30068611.0,
      "step": 14845
    },
    {
      "entropy": 6.242170524597168,
      "epoch": 1.2977366075329897,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004836477406223775,
      "loss": 5.9565,
      "mean_token_accuracy": 0.15131818801164626,
      "num_tokens": 30077992.0,
      "step": 14850
    },
    {
      "entropy": 6.365827894210815,
      "epoch": 1.2981735558856944,
      "grad_norm": 3.53125,
      "learning_rate": 0.00048363607622871086,
      "loss": 6.1115,
      "mean_token_accuracy": 0.14142927080392836,
      "num_tokens": 30088940.0,
      "step": 14855
    },
    {
      "entropy": 6.310481071472168,
      "epoch": 1.298610504238399,
      "grad_norm": 4.78125,
      "learning_rate": 0.00048362440783329485,
      "loss": 6.0119,
      "mean_token_accuracy": 0.1429837465286255,
      "num_tokens": 30099249.0,
      "step": 14860
    },
    {
      "entropy": 6.248052930831909,
      "epoch": 1.2990474525911038,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004836127354363535,
      "loss": 6.0069,
      "mean_token_accuracy": 0.1489975020289421,
      "num_tokens": 30110548.0,
      "step": 14865
    },
    {
      "entropy": 6.200257301330566,
      "epoch": 1.2994844009438085,
      "grad_norm": 5.8125,
      "learning_rate": 0.00048360105903811045,
      "loss": 5.9651,
      "mean_token_accuracy": 0.15626610219478607,
      "num_tokens": 30120842.0,
      "step": 14870
    },
    {
      "entropy": 6.28047924041748,
      "epoch": 1.2999213492965132,
      "grad_norm": 3.046875,
      "learning_rate": 0.00048358937863878984,
      "loss": 6.0462,
      "mean_token_accuracy": 0.1448766276240349,
      "num_tokens": 30130439.0,
      "step": 14875
    },
    {
      "entropy": 6.3083813190460205,
      "epoch": 1.3003582976492178,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004835776942386157,
      "loss": 6.0154,
      "mean_token_accuracy": 0.14484531804919243,
      "num_tokens": 30140976.0,
      "step": 14880
    },
    {
      "entropy": 6.2252037525177,
      "epoch": 1.3007952460019225,
      "grad_norm": 4.3125,
      "learning_rate": 0.000483566005837812,
      "loss": 5.9802,
      "mean_token_accuracy": 0.14780116975307464,
      "num_tokens": 30150989.0,
      "step": 14885
    },
    {
      "entropy": 6.240964269638061,
      "epoch": 1.3012321943546272,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048355431343660303,
      "loss": 5.9662,
      "mean_token_accuracy": 0.14413918629288675,
      "num_tokens": 30161605.0,
      "step": 14890
    },
    {
      "entropy": 6.303191995620727,
      "epoch": 1.301669142707332,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004835426170352131,
      "loss": 5.9646,
      "mean_token_accuracy": 0.14672381281852723,
      "num_tokens": 30171361.0,
      "step": 14895
    },
    {
      "entropy": 6.2487592697143555,
      "epoch": 1.3021060910600366,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048353091663386655,
      "loss": 5.9108,
      "mean_token_accuracy": 0.14946982339024545,
      "num_tokens": 30180352.0,
      "step": 14900
    },
    {
      "entropy": 6.25297212600708,
      "epoch": 1.3025430394127415,
      "grad_norm": 6.1875,
      "learning_rate": 0.0004835192122327877,
      "loss": 5.9519,
      "mean_token_accuracy": 0.14838891625404357,
      "num_tokens": 30190624.0,
      "step": 14905
    },
    {
      "entropy": 6.201724815368652,
      "epoch": 1.302979987765446,
      "grad_norm": 5.46875,
      "learning_rate": 0.00048350750383220105,
      "loss": 5.9684,
      "mean_token_accuracy": 0.14954227805137635,
      "num_tokens": 30200203.0,
      "step": 14910
    },
    {
      "entropy": 6.238126802444458,
      "epoch": 1.303416936118151,
      "grad_norm": 6.90625,
      "learning_rate": 0.00048349579143233126,
      "loss": 5.967,
      "mean_token_accuracy": 0.14993017613887788,
      "num_tokens": 30210546.0,
      "step": 14915
    },
    {
      "entropy": 6.32914457321167,
      "epoch": 1.3038538844708556,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004834840750334029,
      "loss": 6.077,
      "mean_token_accuracy": 0.1384377270936966,
      "num_tokens": 30221316.0,
      "step": 14920
    },
    {
      "entropy": 6.291773796081543,
      "epoch": 1.3042908328235603,
      "grad_norm": 5.0,
      "learning_rate": 0.0004834723546356408,
      "loss": 6.0368,
      "mean_token_accuracy": 0.14191703945398332,
      "num_tokens": 30232580.0,
      "step": 14925
    },
    {
      "entropy": 6.246729040145874,
      "epoch": 1.304727781176265,
      "grad_norm": 4.09375,
      "learning_rate": 0.00048346063023926964,
      "loss": 5.9478,
      "mean_token_accuracy": 0.15073101371526718,
      "num_tokens": 30241976.0,
      "step": 14930
    },
    {
      "entropy": 6.238508319854736,
      "epoch": 1.3051647295289697,
      "grad_norm": 3.265625,
      "learning_rate": 0.00048344890184451436,
      "loss": 5.9731,
      "mean_token_accuracy": 0.14270402118563652,
      "num_tokens": 30252723.0,
      "step": 14935
    },
    {
      "entropy": 6.2823339939117435,
      "epoch": 1.3056016778816744,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048343716945159997,
      "loss": 6.0141,
      "mean_token_accuracy": 0.14994250386953353,
      "num_tokens": 30263125.0,
      "step": 14940
    },
    {
      "entropy": 6.234030294418335,
      "epoch": 1.306038626234379,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004834254330607514,
      "loss": 5.9152,
      "mean_token_accuracy": 0.15902869403362274,
      "num_tokens": 30273113.0,
      "step": 14945
    },
    {
      "entropy": 6.240786600112915,
      "epoch": 1.3064755745870837,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004834136926721939,
      "loss": 5.8923,
      "mean_token_accuracy": 0.1545419067144394,
      "num_tokens": 30282550.0,
      "step": 14950
    },
    {
      "entropy": 6.179037570953369,
      "epoch": 1.3069125229397884,
      "grad_norm": 2.8125,
      "learning_rate": 0.00048340194828615246,
      "loss": 5.9506,
      "mean_token_accuracy": 0.14596819430589675,
      "num_tokens": 30293377.0,
      "step": 14955
    },
    {
      "entropy": 6.193124008178711,
      "epoch": 1.3073494712924933,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004833901999028525,
      "loss": 5.9396,
      "mean_token_accuracy": 0.15183319970965387,
      "num_tokens": 30303147.0,
      "step": 14960
    },
    {
      "entropy": 6.230455017089843,
      "epoch": 1.3077864196451978,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004833784475225192,
      "loss": 6.0304,
      "mean_token_accuracy": 0.14234444126486778,
      "num_tokens": 30313426.0,
      "step": 14965
    },
    {
      "entropy": 6.258969306945801,
      "epoch": 1.3082233679979027,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004833666911453783,
      "loss": 6.0058,
      "mean_token_accuracy": 0.13854765594005586,
      "num_tokens": 30323896.0,
      "step": 14970
    },
    {
      "entropy": 6.264124011993408,
      "epoch": 1.3086603163506074,
      "grad_norm": 2.734375,
      "learning_rate": 0.000483354930771655,
      "loss": 5.9806,
      "mean_token_accuracy": 0.14259461909532548,
      "num_tokens": 30333523.0,
      "step": 14975
    },
    {
      "entropy": 6.250613451004028,
      "epoch": 1.3090972647033121,
      "grad_norm": 3.0625,
      "learning_rate": 0.00048334316640157493,
      "loss": 5.9725,
      "mean_token_accuracy": 0.1500299885869026,
      "num_tokens": 30344035.0,
      "step": 14980
    },
    {
      "entropy": 6.216826677322388,
      "epoch": 1.3095342130560168,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004833313980353637,
      "loss": 6.0068,
      "mean_token_accuracy": 0.1451183430850506,
      "num_tokens": 30354385.0,
      "step": 14985
    },
    {
      "entropy": 6.239265632629395,
      "epoch": 1.3099711614087215,
      "grad_norm": 3.59375,
      "learning_rate": 0.00048331962567324717,
      "loss": 6.0778,
      "mean_token_accuracy": 0.1342620961368084,
      "num_tokens": 30364941.0,
      "step": 14990
    },
    {
      "entropy": 6.288142585754395,
      "epoch": 1.3104081097614262,
      "grad_norm": 2.625,
      "learning_rate": 0.000483307849315451,
      "loss": 5.9266,
      "mean_token_accuracy": 0.1477757290005684,
      "num_tokens": 30376160.0,
      "step": 14995
    },
    {
      "entropy": 6.3015539169311525,
      "epoch": 1.3108450581141309,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048329606896220116,
      "loss": 6.0127,
      "mean_token_accuracy": 0.14319907128810883,
      "num_tokens": 30385018.0,
      "step": 15000
    },
    {
      "epoch": 1.3108450581141309,
      "eval_entropy": 6.000400699960096,
      "eval_loss": 6.0264177322387695,
      "eval_mean_token_accuracy": 0.1532489470147226,
      "eval_num_tokens": 30385018.0,
      "eval_runtime": 27.8427,
      "eval_samples_per_second": 1329.396,
      "eval_steps_per_second": 166.183,
      "step": 15000
    },
    {
      "entropy": 6.218473434448242,
      "epoch": 1.3112820064668356,
      "grad_norm": 4.875,
      "learning_rate": 0.0004832842846137236,
      "loss": 5.9754,
      "mean_token_accuracy": 0.14659590721130372,
      "num_tokens": 30394618.0,
      "step": 15005
    },
    {
      "entropy": 6.222129583358765,
      "epoch": 1.3117189548195403,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048327249627024433,
      "loss": 5.954,
      "mean_token_accuracy": 0.14474146366119384,
      "num_tokens": 30404081.0,
      "step": 15010
    },
    {
      "entropy": 6.278634834289551,
      "epoch": 1.3121559031722452,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004832607039319894,
      "loss": 5.9443,
      "mean_token_accuracy": 0.14185508489608764,
      "num_tokens": 30413892.0,
      "step": 15015
    },
    {
      "entropy": 6.286070537567139,
      "epoch": 1.3125928515249496,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048324890759918513,
      "loss": 6.021,
      "mean_token_accuracy": 0.1428314745426178,
      "num_tokens": 30424523.0,
      "step": 15020
    },
    {
      "entropy": 6.216691398620606,
      "epoch": 1.3130297998776546,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004832371072720576,
      "loss": 5.9491,
      "mean_token_accuracy": 0.1486659437417984,
      "num_tokens": 30435092.0,
      "step": 15025
    },
    {
      "entropy": 6.165086841583252,
      "epoch": 1.3134667482303592,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004832253029508332,
      "loss": 5.9065,
      "mean_token_accuracy": 0.14866694062948227,
      "num_tokens": 30445628.0,
      "step": 15030
    },
    {
      "entropy": 6.236626815795899,
      "epoch": 1.313903696583064,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004832134946357384,
      "loss": 5.9832,
      "mean_token_accuracy": 0.14596127569675446,
      "num_tokens": 30457433.0,
      "step": 15035
    },
    {
      "entropy": 6.236681222915649,
      "epoch": 1.3143406449357686,
      "grad_norm": 12.5625,
      "learning_rate": 0.0004832016823269998,
      "loss": 5.9602,
      "mean_token_accuracy": 0.14650988280773164,
      "num_tokens": 30467987.0,
      "step": 15040
    },
    {
      "entropy": 6.228724575042724,
      "epoch": 1.3147775932884733,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004831898660248437,
      "loss": 5.9735,
      "mean_token_accuracy": 0.14209896326065063,
      "num_tokens": 30476577.0,
      "step": 15045
    },
    {
      "entropy": 6.245405483245849,
      "epoch": 1.315214541641178,
      "grad_norm": 2.96875,
      "learning_rate": 0.00048317804572949697,
      "loss": 5.9296,
      "mean_token_accuracy": 0.14847110211849213,
      "num_tokens": 30487410.0,
      "step": 15050
    },
    {
      "entropy": 6.31135745048523,
      "epoch": 1.3156514899938827,
      "grad_norm": 4.125,
      "learning_rate": 0.0004831662214411862,
      "loss": 6.0178,
      "mean_token_accuracy": 0.14240812584757806,
      "num_tokens": 30497778.0,
      "step": 15055
    },
    {
      "entropy": 6.227396440505982,
      "epoch": 1.3160884383465874,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004831543931601382,
      "loss": 5.956,
      "mean_token_accuracy": 0.14949349462985992,
      "num_tokens": 30507369.0,
      "step": 15060
    },
    {
      "entropy": 6.257428121566773,
      "epoch": 1.316525386699292,
      "grad_norm": 5.125,
      "learning_rate": 0.00048314256088657985,
      "loss": 6.0042,
      "mean_token_accuracy": 0.1459598571062088,
      "num_tokens": 30518126.0,
      "step": 15065
    },
    {
      "entropy": 6.2488030910491945,
      "epoch": 1.3169623350519968,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004831307246207382,
      "loss": 5.9644,
      "mean_token_accuracy": 0.15078468173742293,
      "num_tokens": 30528167.0,
      "step": 15070
    },
    {
      "entropy": 6.228219795227051,
      "epoch": 1.3173992834047015,
      "grad_norm": 3.390625,
      "learning_rate": 0.00048311888436284017,
      "loss": 5.9201,
      "mean_token_accuracy": 0.14862440004944802,
      "num_tokens": 30538111.0,
      "step": 15075
    },
    {
      "entropy": 6.240427112579345,
      "epoch": 1.3178362317574064,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004831070401131129,
      "loss": 6.0109,
      "mean_token_accuracy": 0.1512704536318779,
      "num_tokens": 30547773.0,
      "step": 15080
    },
    {
      "entropy": 6.2413389682769775,
      "epoch": 1.3182731801101109,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004830951918717834,
      "loss": 5.9621,
      "mean_token_accuracy": 0.1470685675740242,
      "num_tokens": 30558058.0,
      "step": 15085
    },
    {
      "entropy": 6.259869861602783,
      "epoch": 1.3187101284628158,
      "grad_norm": 3.46875,
      "learning_rate": 0.00048308333963907927,
      "loss": 5.9023,
      "mean_token_accuracy": 0.15156628638505937,
      "num_tokens": 30567297.0,
      "step": 15090
    },
    {
      "entropy": 6.273061943054199,
      "epoch": 1.3191470768155205,
      "grad_norm": 3.046875,
      "learning_rate": 0.00048307148341522756,
      "loss": 6.0258,
      "mean_token_accuracy": 0.1514521285891533,
      "num_tokens": 30578159.0,
      "step": 15095
    },
    {
      "entropy": 6.2799467086792,
      "epoch": 1.3195840251682251,
      "grad_norm": 2.875,
      "learning_rate": 0.00048305962320045574,
      "loss": 5.9904,
      "mean_token_accuracy": 0.1471402406692505,
      "num_tokens": 30588734.0,
      "step": 15100
    },
    {
      "entropy": 6.26523871421814,
      "epoch": 1.3200209735209298,
      "grad_norm": 2.90625,
      "learning_rate": 0.00048304775899499143,
      "loss": 5.8949,
      "mean_token_accuracy": 0.1520664744079113,
      "num_tokens": 30598496.0,
      "step": 15105
    },
    {
      "entropy": 6.228305673599243,
      "epoch": 1.3204579218736345,
      "grad_norm": 2.875,
      "learning_rate": 0.000483035890799062,
      "loss": 5.9626,
      "mean_token_accuracy": 0.1482691526412964,
      "num_tokens": 30608330.0,
      "step": 15110
    },
    {
      "entropy": 6.250581455230713,
      "epoch": 1.3208948702263392,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004830240186128952,
      "loss": 5.9459,
      "mean_token_accuracy": 0.15207852721214293,
      "num_tokens": 30618887.0,
      "step": 15115
    },
    {
      "entropy": 6.27552375793457,
      "epoch": 1.321331818579044,
      "grad_norm": 6.28125,
      "learning_rate": 0.00048301214243671884,
      "loss": 5.9366,
      "mean_token_accuracy": 0.14410503804683686,
      "num_tokens": 30629479.0,
      "step": 15120
    },
    {
      "entropy": 6.193431663513183,
      "epoch": 1.3217687669317486,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004830002622707604,
      "loss": 5.8976,
      "mean_token_accuracy": 0.14893739223480223,
      "num_tokens": 30639270.0,
      "step": 15125
    },
    {
      "entropy": 6.262009811401367,
      "epoch": 1.3222057152844533,
      "grad_norm": 3.5,
      "learning_rate": 0.0004829883781152481,
      "loss": 6.0214,
      "mean_token_accuracy": 0.14712801426649094,
      "num_tokens": 30649156.0,
      "step": 15130
    },
    {
      "entropy": 6.2490087985992435,
      "epoch": 1.3226426636371582,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048297648997040966,
      "loss": 6.0336,
      "mean_token_accuracy": 0.1460738144814968,
      "num_tokens": 30659919.0,
      "step": 15135
    },
    {
      "entropy": 6.2757744789123535,
      "epoch": 1.3230796119898627,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048296459783647304,
      "loss": 6.0009,
      "mean_token_accuracy": 0.1477440968155861,
      "num_tokens": 30670523.0,
      "step": 15140
    },
    {
      "entropy": 6.269595098495484,
      "epoch": 1.3235165603425676,
      "grad_norm": 3.828125,
      "learning_rate": 0.00048295270171366665,
      "loss": 6.0027,
      "mean_token_accuracy": 0.15111492276191713,
      "num_tokens": 30680803.0,
      "step": 15145
    },
    {
      "entropy": 6.263330936431885,
      "epoch": 1.3239535086952723,
      "grad_norm": 2.953125,
      "learning_rate": 0.00048294080160221837,
      "loss": 5.8875,
      "mean_token_accuracy": 0.1508491188287735,
      "num_tokens": 30690401.0,
      "step": 15150
    },
    {
      "entropy": 6.2723243713378904,
      "epoch": 1.324390457047977,
      "grad_norm": 3.421875,
      "learning_rate": 0.00048292889750235665,
      "loss": 6.0266,
      "mean_token_accuracy": 0.14290419220924377,
      "num_tokens": 30700779.0,
      "step": 15155
    },
    {
      "entropy": 6.303020429611206,
      "epoch": 1.3248274054006817,
      "grad_norm": 3.796875,
      "learning_rate": 0.00048291698941430955,
      "loss": 5.9778,
      "mean_token_accuracy": 0.1420856975018978,
      "num_tokens": 30711556.0,
      "step": 15160
    },
    {
      "entropy": 6.195237255096435,
      "epoch": 1.3252643537533864,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004829050773383058,
      "loss": 5.8715,
      "mean_token_accuracy": 0.15389792025089263,
      "num_tokens": 30721929.0,
      "step": 15165
    },
    {
      "entropy": 6.22774748802185,
      "epoch": 1.325701302106091,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004828931612745737,
      "loss": 6.0301,
      "mean_token_accuracy": 0.14290671795606613,
      "num_tokens": 30732848.0,
      "step": 15170
    },
    {
      "entropy": 6.267693281173706,
      "epoch": 1.3261382504587957,
      "grad_norm": 3.140625,
      "learning_rate": 0.00048288124122334176,
      "loss": 5.9717,
      "mean_token_accuracy": 0.1397448644042015,
      "num_tokens": 30742628.0,
      "step": 15175
    },
    {
      "entropy": 6.300339555740356,
      "epoch": 1.3265751988115004,
      "grad_norm": 5.3125,
      "learning_rate": 0.0004828693171848388,
      "loss": 6.0059,
      "mean_token_accuracy": 0.14635556042194367,
      "num_tokens": 30752866.0,
      "step": 15180
    },
    {
      "entropy": 6.22663106918335,
      "epoch": 1.3270121471642051,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004828573891592933,
      "loss": 5.9391,
      "mean_token_accuracy": 0.15515438914299012,
      "num_tokens": 30762861.0,
      "step": 15185
    },
    {
      "entropy": 6.229796600341797,
      "epoch": 1.32744909551691,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004828454571469342,
      "loss": 6.0018,
      "mean_token_accuracy": 0.15001337081193925,
      "num_tokens": 30772555.0,
      "step": 15190
    },
    {
      "entropy": 6.340445566177368,
      "epoch": 1.3278860438696145,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004828335211479903,
      "loss": 5.9592,
      "mean_token_accuracy": 0.1485216036438942,
      "num_tokens": 30782739.0,
      "step": 15195
    },
    {
      "entropy": 6.298048114776611,
      "epoch": 1.3283229922223194,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048282158116269065,
      "loss": 5.9929,
      "mean_token_accuracy": 0.14441455602645875,
      "num_tokens": 30792600.0,
      "step": 15200
    },
    {
      "entropy": 6.2524982452392575,
      "epoch": 1.3287599405750241,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004828096371912641,
      "loss": 5.9688,
      "mean_token_accuracy": 0.14459720999002457,
      "num_tokens": 30801433.0,
      "step": 15205
    },
    {
      "entropy": 6.273639869689942,
      "epoch": 1.3291968889277288,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048279768923393987,
      "loss": 6.056,
      "mean_token_accuracy": 0.13867955654859543,
      "num_tokens": 30811277.0,
      "step": 15210
    },
    {
      "entropy": 6.256853818893433,
      "epoch": 1.3296338372804335,
      "grad_norm": 3.265625,
      "learning_rate": 0.00048278573729094705,
      "loss": 5.9781,
      "mean_token_accuracy": 0.14866037368774415,
      "num_tokens": 30821120.0,
      "step": 15215
    },
    {
      "entropy": 6.27207317352295,
      "epoch": 1.3300707856331382,
      "grad_norm": 4.8125,
      "learning_rate": 0.00048277378136251487,
      "loss": 5.9231,
      "mean_token_accuracy": 0.15306596606969833,
      "num_tokens": 30831660.0,
      "step": 15220
    },
    {
      "entropy": 6.320482349395752,
      "epoch": 1.3305077339858429,
      "grad_norm": 3.328125,
      "learning_rate": 0.00048276182144887274,
      "loss": 5.9866,
      "mean_token_accuracy": 0.14095881581306458,
      "num_tokens": 30842530.0,
      "step": 15225
    },
    {
      "entropy": 6.302779769897461,
      "epoch": 1.3309446823385476,
      "grad_norm": 3.34375,
      "learning_rate": 0.00048274985755025003,
      "loss": 5.971,
      "mean_token_accuracy": 0.14355332404375076,
      "num_tokens": 30852954.0,
      "step": 15230
    },
    {
      "entropy": 6.216593980789185,
      "epoch": 1.3313816306912523,
      "grad_norm": 2.953125,
      "learning_rate": 0.00048273788966687615,
      "loss": 5.9235,
      "mean_token_accuracy": 0.14845923483371734,
      "num_tokens": 30862864.0,
      "step": 15235
    },
    {
      "entropy": 6.283112239837647,
      "epoch": 1.331818579043957,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004827259177989807,
      "loss": 5.9693,
      "mean_token_accuracy": 0.14308378770947455,
      "num_tokens": 30873533.0,
      "step": 15240
    },
    {
      "entropy": 6.29207935333252,
      "epoch": 1.3322555273966616,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048271394194679336,
      "loss": 5.9257,
      "mean_token_accuracy": 0.15183504968881606,
      "num_tokens": 30883564.0,
      "step": 15245
    },
    {
      "entropy": 6.24968843460083,
      "epoch": 1.3326924757493663,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004827019621105436,
      "loss": 5.9414,
      "mean_token_accuracy": 0.15080741196870803,
      "num_tokens": 30893516.0,
      "step": 15250
    },
    {
      "entropy": 6.265071725845337,
      "epoch": 1.3331294241020712,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048268997829046156,
      "loss": 5.9367,
      "mean_token_accuracy": 0.149213358014822,
      "num_tokens": 30903104.0,
      "step": 15255
    },
    {
      "entropy": 6.196661853790284,
      "epoch": 1.3335663724547757,
      "grad_norm": 3.484375,
      "learning_rate": 0.00048267799048677684,
      "loss": 5.9471,
      "mean_token_accuracy": 0.14992839992046356,
      "num_tokens": 30912759.0,
      "step": 15260
    },
    {
      "entropy": 6.2081873416900635,
      "epoch": 1.3340033208074806,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004826659986997194,
      "loss": 5.9644,
      "mean_token_accuracy": 0.14695684164762496,
      "num_tokens": 30921361.0,
      "step": 15265
    },
    {
      "entropy": 6.278805208206177,
      "epoch": 1.3344402691601853,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004826540029295193,
      "loss": 5.963,
      "mean_token_accuracy": 0.1438293643295765,
      "num_tokens": 30931006.0,
      "step": 15270
    },
    {
      "entropy": 6.280364274978638,
      "epoch": 1.33487721751289,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048264200317640665,
      "loss": 5.9776,
      "mean_token_accuracy": 0.14997672736644746,
      "num_tokens": 30940647.0,
      "step": 15275
    },
    {
      "entropy": 6.293081569671631,
      "epoch": 1.3353141658655947,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004826299994406116,
      "loss": 6.009,
      "mean_token_accuracy": 0.14322509691119195,
      "num_tokens": 30951028.0,
      "step": 15280
    },
    {
      "entropy": 6.235379266738891,
      "epoch": 1.3357511142182994,
      "grad_norm": 3.640625,
      "learning_rate": 0.00048261799172236435,
      "loss": 5.9667,
      "mean_token_accuracy": 0.1456873372197151,
      "num_tokens": 30960726.0,
      "step": 15285
    },
    {
      "entropy": 6.201272010803223,
      "epoch": 1.336188062571004,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004826059800218952,
      "loss": 6.028,
      "mean_token_accuracy": 0.14561986029148102,
      "num_tokens": 30971061.0,
      "step": 15290
    },
    {
      "entropy": 6.204651546478272,
      "epoch": 1.3366250109237088,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004825939643394346,
      "loss": 5.8416,
      "mean_token_accuracy": 0.1586669147014618,
      "num_tokens": 30980518.0,
      "step": 15295
    },
    {
      "entropy": 6.28301887512207,
      "epoch": 1.3370619592764135,
      "grad_norm": 4.53125,
      "learning_rate": 0.00048258194467521294,
      "loss": 6.0306,
      "mean_token_accuracy": 0.1419808343052864,
      "num_tokens": 30990759.0,
      "step": 15300
    },
    {
      "entropy": 6.378645133972168,
      "epoch": 1.3374989076291182,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048256992102946084,
      "loss": 6.0398,
      "mean_token_accuracy": 0.13811585754156114,
      "num_tokens": 31002218.0,
      "step": 15305
    },
    {
      "entropy": 6.216667127609253,
      "epoch": 1.337935855981823,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004825578934024089,
      "loss": 5.9651,
      "mean_token_accuracy": 0.1483850136399269,
      "num_tokens": 31012434.0,
      "step": 15310
    },
    {
      "entropy": 6.267811822891235,
      "epoch": 1.3383728043345275,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048254586179428776,
      "loss": 6.0498,
      "mean_token_accuracy": 0.14165866672992705,
      "num_tokens": 31022805.0,
      "step": 15315
    },
    {
      "entropy": 6.223868656158447,
      "epoch": 1.3388097526872325,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004825338262053283,
      "loss": 5.9571,
      "mean_token_accuracy": 0.1522163435816765,
      "num_tokens": 31032563.0,
      "step": 15320
    },
    {
      "entropy": 6.204619026184082,
      "epoch": 1.3392467010399371,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004825217866357613,
      "loss": 5.9032,
      "mean_token_accuracy": 0.15391222685575484,
      "num_tokens": 31041445.0,
      "step": 15325
    },
    {
      "entropy": 6.285332441329956,
      "epoch": 1.3396836493926418,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048250974308581765,
      "loss": 5.9552,
      "mean_token_accuracy": 0.15195013731718063,
      "num_tokens": 31051413.0,
      "step": 15330
    },
    {
      "entropy": 6.252322196960449,
      "epoch": 1.3401205977453465,
      "grad_norm": 4.8125,
      "learning_rate": 0.00048249769555572844,
      "loss": 5.9367,
      "mean_token_accuracy": 0.1474837228655815,
      "num_tokens": 31061822.0,
      "step": 15335
    },
    {
      "entropy": 6.174934911727905,
      "epoch": 1.3405575460980512,
      "grad_norm": 3.140625,
      "learning_rate": 0.00048248564404572463,
      "loss": 5.9495,
      "mean_token_accuracy": 0.14608196765184403,
      "num_tokens": 31071889.0,
      "step": 15340
    },
    {
      "entropy": 6.191765356063843,
      "epoch": 1.340994494450756,
      "grad_norm": 2.625,
      "learning_rate": 0.00048247358855603754,
      "loss": 5.8871,
      "mean_token_accuracy": 0.15071368217468262,
      "num_tokens": 31080933.0,
      "step": 15345
    },
    {
      "entropy": 6.259008836746216,
      "epoch": 1.3414314428034606,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004824615290868982,
      "loss": 6.0299,
      "mean_token_accuracy": 0.14466264843940735,
      "num_tokens": 31091349.0,
      "step": 15350
    },
    {
      "entropy": 6.356918096542358,
      "epoch": 1.3418683911561653,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004824494656385381,
      "loss": 5.9894,
      "mean_token_accuracy": 0.14460562765598298,
      "num_tokens": 31101485.0,
      "step": 15355
    },
    {
      "entropy": 6.195051527023315,
      "epoch": 1.34230533950887,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004824373982111886,
      "loss": 5.9618,
      "mean_token_accuracy": 0.14617480635643004,
      "num_tokens": 31111965.0,
      "step": 15360
    },
    {
      "entropy": 6.27819299697876,
      "epoch": 1.3427422878615747,
      "grad_norm": 2.59375,
      "learning_rate": 0.000482425326805081,
      "loss": 5.949,
      "mean_token_accuracy": 0.14827857688069343,
      "num_tokens": 31122284.0,
      "step": 15365
    },
    {
      "entropy": 6.286847400665283,
      "epoch": 1.3431792362142794,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048241325142044697,
      "loss": 5.9498,
      "mean_token_accuracy": 0.14226059392094612,
      "num_tokens": 31132156.0,
      "step": 15370
    },
    {
      "entropy": 6.222988843917847,
      "epoch": 1.3436161845669843,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048240117205751817,
      "loss": 5.8894,
      "mean_token_accuracy": 0.15056962445378302,
      "num_tokens": 31141609.0,
      "step": 15375
    },
    {
      "entropy": 6.157810544967651,
      "epoch": 1.344053132919689,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004823890887165261,
      "loss": 5.9967,
      "mean_token_accuracy": 0.14168056100606918,
      "num_tokens": 31153159.0,
      "step": 15380
    },
    {
      "entropy": 6.2409721374511715,
      "epoch": 1.3444900812723937,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048237700139770273,
      "loss": 5.999,
      "mean_token_accuracy": 0.1512889213860035,
      "num_tokens": 31163330.0,
      "step": 15385
    },
    {
      "entropy": 6.274799537658692,
      "epoch": 1.3449270296250984,
      "grad_norm": 2.75,
      "learning_rate": 0.0004823649101012797,
      "loss": 6.0209,
      "mean_token_accuracy": 0.14578405469655992,
      "num_tokens": 31173874.0,
      "step": 15390
    },
    {
      "entropy": 6.21087007522583,
      "epoch": 1.345363977977803,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004823528148274891,
      "loss": 5.9133,
      "mean_token_accuracy": 0.15412042811512947,
      "num_tokens": 31183933.0,
      "step": 15395
    },
    {
      "entropy": 6.225914144515992,
      "epoch": 1.3458009263305077,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048234071557656295,
      "loss": 5.9056,
      "mean_token_accuracy": 0.15846829861402512,
      "num_tokens": 31194308.0,
      "step": 15400
    },
    {
      "entropy": 6.260078239440918,
      "epoch": 1.3462378746832124,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004823286123487331,
      "loss": 5.9495,
      "mean_token_accuracy": 0.1534321576356888,
      "num_tokens": 31203757.0,
      "step": 15405
    },
    {
      "entropy": 6.2870128631591795,
      "epoch": 1.3466748230359171,
      "grad_norm": 2.984375,
      "learning_rate": 0.00048231650514423194,
      "loss": 6.0368,
      "mean_token_accuracy": 0.14239249601960183,
      "num_tokens": 31213980.0,
      "step": 15410
    },
    {
      "entropy": 6.33568000793457,
      "epoch": 1.3471117713886218,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048230439396329143,
      "loss": 6.0077,
      "mean_token_accuracy": 0.14486486315727234,
      "num_tokens": 31225034.0,
      "step": 15415
    },
    {
      "entropy": 6.238612365722656,
      "epoch": 1.3475487197413265,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048229227880614413,
      "loss": 5.9723,
      "mean_token_accuracy": 0.14105203822255136,
      "num_tokens": 31236393.0,
      "step": 15420
    },
    {
      "entropy": 6.294801378250122,
      "epoch": 1.3479856680940312,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004822801596730222,
      "loss": 5.9531,
      "mean_token_accuracy": 0.14863016754388808,
      "num_tokens": 31246577.0,
      "step": 15425
    },
    {
      "entropy": 6.225813293457032,
      "epoch": 1.348422616446736,
      "grad_norm": 6.0,
      "learning_rate": 0.00048226803656415833,
      "loss": 5.9204,
      "mean_token_accuracy": 0.14502865821123123,
      "num_tokens": 31256004.0,
      "step": 15430
    },
    {
      "entropy": 6.281371068954468,
      "epoch": 1.3488595647994406,
      "grad_norm": 2.0,
      "learning_rate": 0.0004822559094797849,
      "loss": 6.0224,
      "mean_token_accuracy": 0.14738284572958946,
      "num_tokens": 31267294.0,
      "step": 15435
    },
    {
      "entropy": 6.209185266494751,
      "epoch": 1.3492965131521455,
      "grad_norm": 2.25,
      "learning_rate": 0.00048224377842013443,
      "loss": 5.9354,
      "mean_token_accuracy": 0.15742374062538148,
      "num_tokens": 31276896.0,
      "step": 15440
    },
    {
      "entropy": 6.176313686370849,
      "epoch": 1.3497334615048502,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004822316433854397,
      "loss": 5.9776,
      "mean_token_accuracy": 0.14837896153330804,
      "num_tokens": 31287248.0,
      "step": 15445
    },
    {
      "entropy": 6.280122423171997,
      "epoch": 1.3501704098575549,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048221950437593353,
      "loss": 5.9812,
      "mean_token_accuracy": 0.14564943835139274,
      "num_tokens": 31297422.0,
      "step": 15450
    },
    {
      "entropy": 6.2574286460876465,
      "epoch": 1.3506073582102596,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048220736139184864,
      "loss": 5.9852,
      "mean_token_accuracy": 0.151013645529747,
      "num_tokens": 31307182.0,
      "step": 15455
    },
    {
      "entropy": 6.256010818481445,
      "epoch": 1.3510443065629643,
      "grad_norm": 3.28125,
      "learning_rate": 0.000482195214433418,
      "loss": 6.0692,
      "mean_token_accuracy": 0.14402149766683578,
      "num_tokens": 31317234.0,
      "step": 15460
    },
    {
      "entropy": 6.281969690322876,
      "epoch": 1.351481254915669,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004821830635008746,
      "loss": 6.0555,
      "mean_token_accuracy": 0.14271388128399848,
      "num_tokens": 31328294.0,
      "step": 15465
    },
    {
      "entropy": 6.2763152599334715,
      "epoch": 1.3519182032683736,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004821709085944515,
      "loss": 5.9229,
      "mean_token_accuracy": 0.14714621901512145,
      "num_tokens": 31337250.0,
      "step": 15470
    },
    {
      "entropy": 6.241772890090942,
      "epoch": 1.3523551516210783,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004821587497143817,
      "loss": 6.0127,
      "mean_token_accuracy": 0.1422139473259449,
      "num_tokens": 31346900.0,
      "step": 15475
    },
    {
      "entropy": 6.20913610458374,
      "epoch": 1.352792099973783,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004821465868608986,
      "loss": 5.9734,
      "mean_token_accuracy": 0.1391966961324215,
      "num_tokens": 31357233.0,
      "step": 15480
    },
    {
      "entropy": 6.267159032821655,
      "epoch": 1.353229048326488,
      "grad_norm": 3.625,
      "learning_rate": 0.0004821344200342355,
      "loss": 5.8944,
      "mean_token_accuracy": 0.14981924295425414,
      "num_tokens": 31367043.0,
      "step": 15485
    },
    {
      "entropy": 6.184830188751221,
      "epoch": 1.3536659966791924,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004821222492346256,
      "loss": 5.9282,
      "mean_token_accuracy": 0.14255743250250816,
      "num_tokens": 31376770.0,
      "step": 15490
    },
    {
      "entropy": 6.269280529022216,
      "epoch": 1.3541029450318973,
      "grad_norm": 2.625,
      "learning_rate": 0.00048211007446230243,
      "loss": 5.9613,
      "mean_token_accuracy": 0.14973199516534805,
      "num_tokens": 31386056.0,
      "step": 15495
    },
    {
      "entropy": 6.2864586353302006,
      "epoch": 1.354539893384602,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048209789571749954,
      "loss": 6.0203,
      "mean_token_accuracy": 0.143307613581419,
      "num_tokens": 31395129.0,
      "step": 15500
    },
    {
      "entropy": 6.312516546249389,
      "epoch": 1.3549768417373067,
      "grad_norm": 2.78125,
      "learning_rate": 0.00048208571300045045,
      "loss": 6.0291,
      "mean_token_accuracy": 0.1441151849925518,
      "num_tokens": 31404720.0,
      "step": 15505
    },
    {
      "entropy": 6.288467884063721,
      "epoch": 1.3554137900900114,
      "grad_norm": 4.6875,
      "learning_rate": 0.00048207352631138897,
      "loss": 5.9609,
      "mean_token_accuracy": 0.1494484454393387,
      "num_tokens": 31415588.0,
      "step": 15510
    },
    {
      "entropy": 6.289114618301392,
      "epoch": 1.355850738442716,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048206133565054863,
      "loss": 6.0702,
      "mean_token_accuracy": 0.1390301175415516,
      "num_tokens": 31425904.0,
      "step": 15515
    },
    {
      "entropy": 6.2553729057312015,
      "epoch": 1.3562876867954208,
      "grad_norm": 2.75,
      "learning_rate": 0.0004820491410181634,
      "loss": 5.9987,
      "mean_token_accuracy": 0.14046882316470147,
      "num_tokens": 31435282.0,
      "step": 15520
    },
    {
      "entropy": 6.337297582626343,
      "epoch": 1.3567246351481255,
      "grad_norm": 2.25,
      "learning_rate": 0.00048203694241446724,
      "loss": 6.0931,
      "mean_token_accuracy": 0.1424522206187248,
      "num_tokens": 31445681.0,
      "step": 15525
    },
    {
      "entropy": 6.2453587532043455,
      "epoch": 1.3571615835008302,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048202473983969396,
      "loss": 5.7845,
      "mean_token_accuracy": 0.1558709517121315,
      "num_tokens": 31456934.0,
      "step": 15530
    },
    {
      "entropy": 6.180977821350098,
      "epoch": 1.3575985318535349,
      "grad_norm": 3.421875,
      "learning_rate": 0.00048201253329407766,
      "loss": 5.9414,
      "mean_token_accuracy": 0.1518250048160553,
      "num_tokens": 31466692.0,
      "step": 15535
    },
    {
      "entropy": 6.2484055519104,
      "epoch": 1.3580354802062395,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004820003227778526,
      "loss": 5.9734,
      "mean_token_accuracy": 0.14486174583435057,
      "num_tokens": 31477166.0,
      "step": 15540
    },
    {
      "entropy": 6.282145071029663,
      "epoch": 1.3584724285589442,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004819881082912528,
      "loss": 5.9758,
      "mean_token_accuracy": 0.1447513684630394,
      "num_tokens": 31486998.0,
      "step": 15545
    },
    {
      "entropy": 6.218525123596192,
      "epoch": 1.3589093769116491,
      "grad_norm": 3.359375,
      "learning_rate": 0.00048197588983451265,
      "loss": 5.9897,
      "mean_token_accuracy": 0.1436692327260971,
      "num_tokens": 31496674.0,
      "step": 15550
    },
    {
      "entropy": 6.233652400970459,
      "epoch": 1.3593463252643536,
      "grad_norm": 3.265625,
      "learning_rate": 0.00048196366740786646,
      "loss": 5.9994,
      "mean_token_accuracy": 0.14509041905403136,
      "num_tokens": 31507412.0,
      "step": 15555
    },
    {
      "entropy": 6.25165810585022,
      "epoch": 1.3597832736170585,
      "grad_norm": 3.78125,
      "learning_rate": 0.00048195144101154867,
      "loss": 5.9699,
      "mean_token_accuracy": 0.14685448110103608,
      "num_tokens": 31517048.0,
      "step": 15560
    },
    {
      "entropy": 6.262503814697266,
      "epoch": 1.3602202219697632,
      "grad_norm": 2.625,
      "learning_rate": 0.00048193921064579385,
      "loss": 6.0417,
      "mean_token_accuracy": 0.14332766234874725,
      "num_tokens": 31526353.0,
      "step": 15565
    },
    {
      "entropy": 6.33519229888916,
      "epoch": 1.360657170322468,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004819269763108365,
      "loss": 6.0155,
      "mean_token_accuracy": 0.14124735891819,
      "num_tokens": 31537194.0,
      "step": 15570
    },
    {
      "entropy": 6.325431442260742,
      "epoch": 1.3610941186751726,
      "grad_norm": 4.09375,
      "learning_rate": 0.00048191473800691145,
      "loss": 5.9443,
      "mean_token_accuracy": 0.14575248658657075,
      "num_tokens": 31547896.0,
      "step": 15575
    },
    {
      "entropy": 6.2080686569213865,
      "epoch": 1.3615310670278773,
      "grad_norm": 6.375,
      "learning_rate": 0.0004819024957342531,
      "loss": 5.9665,
      "mean_token_accuracy": 0.14373627305030823,
      "num_tokens": 31558750.0,
      "step": 15580
    },
    {
      "entropy": 6.132077884674072,
      "epoch": 1.361968015380582,
      "grad_norm": 2.5,
      "learning_rate": 0.00048189024949309655,
      "loss": 5.9474,
      "mean_token_accuracy": 0.14752767980098724,
      "num_tokens": 31568721.0,
      "step": 15585
    },
    {
      "entropy": 6.1165117740631105,
      "epoch": 1.3624049637332867,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004818779992836766,
      "loss": 5.9126,
      "mean_token_accuracy": 0.14777434840798379,
      "num_tokens": 31579907.0,
      "step": 15590
    },
    {
      "entropy": 6.316179418563843,
      "epoch": 1.3628419120859914,
      "grad_norm": 4.21875,
      "learning_rate": 0.00048186574510622823,
      "loss": 5.9792,
      "mean_token_accuracy": 0.1450642928481102,
      "num_tokens": 31590137.0,
      "step": 15595
    },
    {
      "entropy": 6.31730899810791,
      "epoch": 1.363278860438696,
      "grad_norm": 3.921875,
      "learning_rate": 0.00048185348696098653,
      "loss": 6.0355,
      "mean_token_accuracy": 0.14310587346553802,
      "num_tokens": 31599969.0,
      "step": 15600
    },
    {
      "entropy": 6.236898899078369,
      "epoch": 1.363715808791401,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004818412248481865,
      "loss": 5.9817,
      "mean_token_accuracy": 0.14693498462438584,
      "num_tokens": 31609951.0,
      "step": 15605
    },
    {
      "entropy": 6.2361897945404055,
      "epoch": 1.3641527571441054,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048182895876806343,
      "loss": 5.947,
      "mean_token_accuracy": 0.15435246229171753,
      "num_tokens": 31619214.0,
      "step": 15610
    },
    {
      "entropy": 6.272160863876342,
      "epoch": 1.3645897054968104,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004818166887208524,
      "loss": 5.9496,
      "mean_token_accuracy": 0.1493683025240898,
      "num_tokens": 31629562.0,
      "step": 15615
    },
    {
      "entropy": 6.22668080329895,
      "epoch": 1.365026653849515,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004818044147067891,
      "loss": 5.9347,
      "mean_token_accuracy": 0.15137291252613067,
      "num_tokens": 31639801.0,
      "step": 15620
    },
    {
      "entropy": 6.25584397315979,
      "epoch": 1.3654636022022197,
      "grad_norm": 3.53125,
      "learning_rate": 0.00048179213672610866,
      "loss": 6.0042,
      "mean_token_accuracy": 0.14098496362566948,
      "num_tokens": 31650593.0,
      "step": 15625
    },
    {
      "entropy": 6.252009868621826,
      "epoch": 1.3659005505549244,
      "grad_norm": 3.0625,
      "learning_rate": 0.00048177985477904675,
      "loss": 5.9625,
      "mean_token_accuracy": 0.14374727457761766,
      "num_tokens": 31660942.0,
      "step": 15630
    },
    {
      "entropy": 6.286769437789917,
      "epoch": 1.3663374989076291,
      "grad_norm": 5.03125,
      "learning_rate": 0.00048176756886583884,
      "loss": 6.0305,
      "mean_token_accuracy": 0.14215213656425477,
      "num_tokens": 31671105.0,
      "step": 15635
    },
    {
      "entropy": 6.273901176452637,
      "epoch": 1.3667744472603338,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004817552789867206,
      "loss": 5.9552,
      "mean_token_accuracy": 0.1493411496281624,
      "num_tokens": 31680696.0,
      "step": 15640
    },
    {
      "entropy": 6.1789556503295895,
      "epoch": 1.3672113956130385,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048174298514192784,
      "loss": 5.8649,
      "mean_token_accuracy": 0.15480989068746567,
      "num_tokens": 31689755.0,
      "step": 15645
    },
    {
      "entropy": 6.160148668289184,
      "epoch": 1.3676483439657432,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048173068733169615,
      "loss": 5.9375,
      "mean_token_accuracy": 0.15040013939142227,
      "num_tokens": 31699614.0,
      "step": 15650
    },
    {
      "entropy": 6.266936492919922,
      "epoch": 1.3680852923184479,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004817183855562617,
      "loss": 5.9487,
      "mean_token_accuracy": 0.1471369281411171,
      "num_tokens": 31709106.0,
      "step": 15655
    },
    {
      "entropy": 6.326410579681396,
      "epoch": 1.3685222406711528,
      "grad_norm": 3.5625,
      "learning_rate": 0.00048170607981586023,
      "loss": 6.0587,
      "mean_token_accuracy": 0.14615088403224946,
      "num_tokens": 31719542.0,
      "step": 15660
    },
    {
      "entropy": 6.28284478187561,
      "epoch": 1.3689591890238573,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048169377011072784,
      "loss": 5.9653,
      "mean_token_accuracy": 0.14869323894381523,
      "num_tokens": 31730477.0,
      "step": 15665
    },
    {
      "entropy": 6.2312400341033936,
      "epoch": 1.3693961373765622,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004816814564411006,
      "loss": 5.9656,
      "mean_token_accuracy": 0.15014294236898423,
      "num_tokens": 31740269.0,
      "step": 15670
    },
    {
      "entropy": 6.246416854858398,
      "epoch": 1.3698330857292669,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004816691388072148,
      "loss": 5.944,
      "mean_token_accuracy": 0.15320937782526017,
      "num_tokens": 31750153.0,
      "step": 15675
    },
    {
      "entropy": 6.263941383361816,
      "epoch": 1.3702700340819716,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048165681720930653,
      "loss": 6.0218,
      "mean_token_accuracy": 0.1425149455666542,
      "num_tokens": 31760171.0,
      "step": 15680
    },
    {
      "entropy": 6.248102045059204,
      "epoch": 1.3707069824346763,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004816444916476123,
      "loss": 5.9958,
      "mean_token_accuracy": 0.14162324219942093,
      "num_tokens": 31771601.0,
      "step": 15685
    },
    {
      "entropy": 6.254783058166504,
      "epoch": 1.371143930787381,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048163216212236844,
      "loss": 5.9406,
      "mean_token_accuracy": 0.1484237790107727,
      "num_tokens": 31781423.0,
      "step": 15690
    },
    {
      "entropy": 6.182925510406494,
      "epoch": 1.3715808791400856,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004816198286338115,
      "loss": 5.865,
      "mean_token_accuracy": 0.15567789375782012,
      "num_tokens": 31791025.0,
      "step": 15695
    },
    {
      "entropy": 6.152139186859131,
      "epoch": 1.3720178274927903,
      "grad_norm": 3.140625,
      "learning_rate": 0.00048160749118217783,
      "loss": 5.9646,
      "mean_token_accuracy": 0.14640328288078308,
      "num_tokens": 31801125.0,
      "step": 15700
    },
    {
      "entropy": 6.246828508377075,
      "epoch": 1.372454775845495,
      "grad_norm": 2.375,
      "learning_rate": 0.0004815951497677043,
      "loss": 5.9715,
      "mean_token_accuracy": 0.1400390774011612,
      "num_tokens": 31811100.0,
      "step": 15705
    },
    {
      "entropy": 6.308065605163574,
      "epoch": 1.3728917241981997,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004815828043906275,
      "loss": 6.0044,
      "mean_token_accuracy": 0.1401717968285084,
      "num_tokens": 31821667.0,
      "step": 15710
    },
    {
      "entropy": 6.234771585464477,
      "epoch": 1.3733286725509044,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004815704550511843,
      "loss": 5.9383,
      "mean_token_accuracy": 0.14841840416193008,
      "num_tokens": 31831859.0,
      "step": 15715
    },
    {
      "entropy": 6.2490123271942135,
      "epoch": 1.373765620903609,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004815581017496115,
      "loss": 5.929,
      "mean_token_accuracy": 0.1484658844769001,
      "num_tokens": 31842952.0,
      "step": 15720
    },
    {
      "entropy": 6.215463924407959,
      "epoch": 1.374202569256314,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048154574448614607,
      "loss": 6.0308,
      "mean_token_accuracy": 0.1482709489762783,
      "num_tokens": 31852869.0,
      "step": 15725
    },
    {
      "entropy": 6.2763574600219725,
      "epoch": 1.3746395176090185,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004815333832610251,
      "loss": 5.9297,
      "mean_token_accuracy": 0.15339216515421866,
      "num_tokens": 31862590.0,
      "step": 15730
    },
    {
      "entropy": 6.241496467590332,
      "epoch": 1.3750764659617234,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048152101807448554,
      "loss": 5.9766,
      "mean_token_accuracy": 0.14535472989082338,
      "num_tokens": 31873071.0,
      "step": 15735
    },
    {
      "entropy": 6.256893682479858,
      "epoch": 1.375513414314428,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004815086489267647,
      "loss": 5.9862,
      "mean_token_accuracy": 0.15268728882074356,
      "num_tokens": 31883325.0,
      "step": 15740
    },
    {
      "entropy": 6.236064100265503,
      "epoch": 1.3759503626671328,
      "grad_norm": 2.953125,
      "learning_rate": 0.00048149627581809974,
      "loss": 5.9919,
      "mean_token_accuracy": 0.14895645081996917,
      "num_tokens": 31893071.0,
      "step": 15745
    },
    {
      "entropy": 6.238200283050537,
      "epoch": 1.3763873110198375,
      "grad_norm": 3.078125,
      "learning_rate": 0.000481483898748728,
      "loss": 5.9357,
      "mean_token_accuracy": 0.1547151267528534,
      "num_tokens": 31902948.0,
      "step": 15750
    },
    {
      "entropy": 6.3012641906738285,
      "epoch": 1.3768242593725422,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004814715177188869,
      "loss": 5.961,
      "mean_token_accuracy": 0.14558223485946656,
      "num_tokens": 31913425.0,
      "step": 15755
    },
    {
      "entropy": 6.169911432266235,
      "epoch": 1.3772612077252468,
      "grad_norm": 2.5,
      "learning_rate": 0.0004814591327288139,
      "loss": 5.9606,
      "mean_token_accuracy": 0.14975444450974465,
      "num_tokens": 31923608.0,
      "step": 15760
    },
    {
      "entropy": 6.149846839904785,
      "epoch": 1.3776981560779515,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048144674377874656,
      "loss": 5.915,
      "mean_token_accuracy": 0.14974069893360137,
      "num_tokens": 31932863.0,
      "step": 15765
    },
    {
      "entropy": 6.243475389480591,
      "epoch": 1.3781351044306562,
      "grad_norm": 17.875,
      "learning_rate": 0.00048143435086892244,
      "loss": 5.9794,
      "mean_token_accuracy": 0.1448725529015064,
      "num_tokens": 31942425.0,
      "step": 15770
    },
    {
      "entropy": 6.283931875228882,
      "epoch": 1.378572052783361,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004814219539995793,
      "loss": 5.9817,
      "mean_token_accuracy": 0.14573511630296707,
      "num_tokens": 31952152.0,
      "step": 15775
    },
    {
      "entropy": 6.204387044906616,
      "epoch": 1.3790090011360658,
      "grad_norm": 4.125,
      "learning_rate": 0.00048140955317095503,
      "loss": 6.0061,
      "mean_token_accuracy": 0.144778510928154,
      "num_tokens": 31962224.0,
      "step": 15780
    },
    {
      "entropy": 6.209999370574951,
      "epoch": 1.3794459494887703,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004813971483832873,
      "loss": 5.9351,
      "mean_token_accuracy": 0.1574947401881218,
      "num_tokens": 31971517.0,
      "step": 15785
    },
    {
      "entropy": 6.252928066253662,
      "epoch": 1.3798828978414752,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004813847396368141,
      "loss": 6.0084,
      "mean_token_accuracy": 0.14856935143470765,
      "num_tokens": 31982461.0,
      "step": 15790
    },
    {
      "entropy": 6.233519077301025,
      "epoch": 1.38031984619418,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004813723269317734,
      "loss": 5.9341,
      "mean_token_accuracy": 0.15391729176044464,
      "num_tokens": 31992202.0,
      "step": 15795
    },
    {
      "entropy": 6.277452802658081,
      "epoch": 1.3807567945468846,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004813599102684034,
      "loss": 5.9764,
      "mean_token_accuracy": 0.14731006175279618,
      "num_tokens": 32002283.0,
      "step": 15800
    },
    {
      "entropy": 6.265666437149048,
      "epoch": 1.3811937428995893,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048134748964694227,
      "loss": 5.956,
      "mean_token_accuracy": 0.14496510326862336,
      "num_tokens": 32013135.0,
      "step": 15805
    },
    {
      "entropy": 6.208707857131958,
      "epoch": 1.381630691252294,
      "grad_norm": 2.375,
      "learning_rate": 0.0004813350650676281,
      "loss": 5.9504,
      "mean_token_accuracy": 0.14549911320209502,
      "num_tokens": 32022631.0,
      "step": 15810
    },
    {
      "entropy": 6.243393182754517,
      "epoch": 1.3820676396049987,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004813226365306992,
      "loss": 5.9918,
      "mean_token_accuracy": 0.14390812665224076,
      "num_tokens": 32032216.0,
      "step": 15815
    },
    {
      "entropy": 6.227290058135987,
      "epoch": 1.3825045879577034,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048131020403639405,
      "loss": 5.9744,
      "mean_token_accuracy": 0.15362895429134368,
      "num_tokens": 32042382.0,
      "step": 15820
    },
    {
      "entropy": 6.227143812179565,
      "epoch": 1.382941536310408,
      "grad_norm": 4.125,
      "learning_rate": 0.00048129776758495116,
      "loss": 5.9382,
      "mean_token_accuracy": 0.15065433606505393,
      "num_tokens": 32052592.0,
      "step": 15825
    },
    {
      "entropy": 6.168285608291626,
      "epoch": 1.3833784846631128,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004812853271766089,
      "loss": 5.8886,
      "mean_token_accuracy": 0.1516219601035118,
      "num_tokens": 32062260.0,
      "step": 15830
    },
    {
      "entropy": 6.305352306365966,
      "epoch": 1.3838154330158177,
      "grad_norm": 2.828125,
      "learning_rate": 0.000481272882811606,
      "loss": 5.9814,
      "mean_token_accuracy": 0.14260451272130012,
      "num_tokens": 32072079.0,
      "step": 15835
    },
    {
      "entropy": 6.231011438369751,
      "epoch": 1.3842523813685221,
      "grad_norm": 2.25,
      "learning_rate": 0.00048126043449018115,
      "loss": 5.9765,
      "mean_token_accuracy": 0.14767121225595475,
      "num_tokens": 32082151.0,
      "step": 15840
    },
    {
      "entropy": 6.291854047775269,
      "epoch": 1.384689329721227,
      "grad_norm": 3.09375,
      "learning_rate": 0.000481247982212573,
      "loss": 6.0029,
      "mean_token_accuracy": 0.14576784819364547,
      "num_tokens": 32091893.0,
      "step": 15845
    },
    {
      "entropy": 6.229527711868286,
      "epoch": 1.3851262780739317,
      "grad_norm": 5.46875,
      "learning_rate": 0.00048123552597902053,
      "loss": 5.9215,
      "mean_token_accuracy": 0.15141658782958983,
      "num_tokens": 32101550.0,
      "step": 15850
    },
    {
      "entropy": 6.1221448421478275,
      "epoch": 1.3855632264266364,
      "grad_norm": 3.28125,
      "learning_rate": 0.00048122306578976266,
      "loss": 5.8501,
      "mean_token_accuracy": 0.15647443383932114,
      "num_tokens": 32112278.0,
      "step": 15855
    },
    {
      "entropy": 6.258195877075195,
      "epoch": 1.3860001747793411,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048121060164503835,
      "loss": 6.0045,
      "mean_token_accuracy": 0.14343008771538734,
      "num_tokens": 32122284.0,
      "step": 15860
    },
    {
      "entropy": 6.311136054992676,
      "epoch": 1.3864371231320458,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004811981335450865,
      "loss": 6.0168,
      "mean_token_accuracy": 0.142793457955122,
      "num_tokens": 32131552.0,
      "step": 15865
    },
    {
      "entropy": 6.298375511169434,
      "epoch": 1.3868740714847505,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048118566149014653,
      "loss": 6.042,
      "mean_token_accuracy": 0.14170629233121873,
      "num_tokens": 32141758.0,
      "step": 15870
    },
    {
      "entropy": 6.22963137626648,
      "epoch": 1.3873110198374552,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004811731854804575,
      "loss": 5.9324,
      "mean_token_accuracy": 0.14714763239026069,
      "num_tokens": 32150679.0,
      "step": 15875
    },
    {
      "entropy": 6.178016233444214,
      "epoch": 1.3877479681901599,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004811607055162587,
      "loss": 5.9203,
      "mean_token_accuracy": 0.14703680723905563,
      "num_tokens": 32160967.0,
      "step": 15880
    },
    {
      "entropy": 6.241407060623169,
      "epoch": 1.3881849165428646,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004811482215977895,
      "loss": 5.9781,
      "mean_token_accuracy": 0.1442311927676201,
      "num_tokens": 32170516.0,
      "step": 15885
    },
    {
      "entropy": 6.269442319869995,
      "epoch": 1.3886218648955693,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048113573372528956,
      "loss": 5.9932,
      "mean_token_accuracy": 0.14519371688365937,
      "num_tokens": 32180741.0,
      "step": 15890
    },
    {
      "entropy": 6.3387792110443115,
      "epoch": 1.389058813248274,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004811232418989981,
      "loss": 6.031,
      "mean_token_accuracy": 0.14694385826587678,
      "num_tokens": 32190646.0,
      "step": 15895
    },
    {
      "entropy": 6.240294885635376,
      "epoch": 1.3894957616009789,
      "grad_norm": 3.4375,
      "learning_rate": 0.00048111074611915483,
      "loss": 5.9514,
      "mean_token_accuracy": 0.15070675313472748,
      "num_tokens": 32201361.0,
      "step": 15900
    },
    {
      "entropy": 6.220612335205078,
      "epoch": 1.3899327099536833,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004810982463859996,
      "loss": 6.0711,
      "mean_token_accuracy": 0.1355058655142784,
      "num_tokens": 32211759.0,
      "step": 15905
    },
    {
      "entropy": 6.31219072341919,
      "epoch": 1.3903696583063883,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004810857426997719,
      "loss": 6.0138,
      "mean_token_accuracy": 0.1459475874900818,
      "num_tokens": 32222582.0,
      "step": 15910
    },
    {
      "entropy": 6.251216745376587,
      "epoch": 1.390806606659093,
      "grad_norm": 3.46875,
      "learning_rate": 0.00048107323506071167,
      "loss": 5.9948,
      "mean_token_accuracy": 0.15561108738183976,
      "num_tokens": 32232936.0,
      "step": 15915
    },
    {
      "entropy": 6.254213762283325,
      "epoch": 1.3912435550117976,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048106072346905875,
      "loss": 5.9959,
      "mean_token_accuracy": 0.14801934957504273,
      "num_tokens": 32242392.0,
      "step": 15920
    },
    {
      "entropy": 6.291469240188599,
      "epoch": 1.3916805033645023,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048104820792505335,
      "loss": 6.0479,
      "mean_token_accuracy": 0.1409144066274166,
      "num_tokens": 32252385.0,
      "step": 15925
    },
    {
      "entropy": 6.3295526027679445,
      "epoch": 1.392117451717207,
      "grad_norm": 3.125,
      "learning_rate": 0.00048103568842893527,
      "loss": 5.9929,
      "mean_token_accuracy": 0.14248449653387069,
      "num_tokens": 32263537.0,
      "step": 15930
    },
    {
      "entropy": 6.2674966812133786,
      "epoch": 1.3925544000699117,
      "grad_norm": 4.65625,
      "learning_rate": 0.00048102316498094467,
      "loss": 6.0087,
      "mean_token_accuracy": 0.14334188252687455,
      "num_tokens": 32273450.0,
      "step": 15935
    },
    {
      "entropy": 6.194003438949585,
      "epoch": 1.3929913484226164,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048101063758132186,
      "loss": 5.9699,
      "mean_token_accuracy": 0.14743351712822914,
      "num_tokens": 32283243.0,
      "step": 15940
    },
    {
      "entropy": 6.204927587509156,
      "epoch": 1.393428296775321,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004809981062303071,
      "loss": 5.9531,
      "mean_token_accuracy": 0.1456786423921585,
      "num_tokens": 32292458.0,
      "step": 15945
    },
    {
      "entropy": 6.289124345779419,
      "epoch": 1.3938652451280258,
      "grad_norm": 3.6875,
      "learning_rate": 0.00048098557092814066,
      "loss": 5.9587,
      "mean_token_accuracy": 0.14371125847101213,
      "num_tokens": 32302574.0,
      "step": 15950
    },
    {
      "entropy": 6.1906860828399655,
      "epoch": 1.3943021934807307,
      "grad_norm": 1.96875,
      "learning_rate": 0.000480973031675063,
      "loss": 5.9438,
      "mean_token_accuracy": 0.15153744518756868,
      "num_tokens": 32312658.0,
      "step": 15955
    },
    {
      "entropy": 6.253198671340942,
      "epoch": 1.3947391418334352,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004809604884713148,
      "loss": 6.0428,
      "mean_token_accuracy": 0.14454116076231002,
      "num_tokens": 32322717.0,
      "step": 15960
    },
    {
      "entropy": 6.203462743759156,
      "epoch": 1.39517609018614,
      "grad_norm": 4.125,
      "learning_rate": 0.00048094794131713645,
      "loss": 5.9429,
      "mean_token_accuracy": 0.15363920629024505,
      "num_tokens": 32333316.0,
      "step": 15965
    },
    {
      "entropy": 6.263971614837646,
      "epoch": 1.3956130385388448,
      "grad_norm": 4.625,
      "learning_rate": 0.0004809353902127686,
      "loss": 6.0032,
      "mean_token_accuracy": 0.1425010897219181,
      "num_tokens": 32343857.0,
      "step": 15970
    },
    {
      "entropy": 6.248172187805176,
      "epoch": 1.3960499868915495,
      "grad_norm": 3.125,
      "learning_rate": 0.00048092283515845216,
      "loss": 5.8899,
      "mean_token_accuracy": 0.15302196145057678,
      "num_tokens": 32353883.0,
      "step": 15975
    },
    {
      "entropy": 6.100135421752929,
      "epoch": 1.3964869352442542,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004809102761544278,
      "loss": 5.8726,
      "mean_token_accuracy": 0.14972448199987412,
      "num_tokens": 32363355.0,
      "step": 15980
    },
    {
      "entropy": 6.227034044265747,
      "epoch": 1.3969238835969588,
      "grad_norm": 8.1875,
      "learning_rate": 0.0004808977132009364,
      "loss": 5.9493,
      "mean_token_accuracy": 0.14863646626472474,
      "num_tokens": 32373066.0,
      "step": 15985
    },
    {
      "entropy": 6.226633739471436,
      "epoch": 1.3973608319496635,
      "grad_norm": 3.390625,
      "learning_rate": 0.000480885146298219,
      "loss": 5.9343,
      "mean_token_accuracy": 0.14998981952667237,
      "num_tokens": 32383720.0,
      "step": 15990
    },
    {
      "entropy": 6.294152545928955,
      "epoch": 1.3977977803023682,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004808725754465167,
      "loss": 5.954,
      "mean_token_accuracy": 0.14437422156333923,
      "num_tokens": 32393823.0,
      "step": 15995
    },
    {
      "entropy": 6.130046463012695,
      "epoch": 1.398234728655073,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048086000064607044,
      "loss": 5.8729,
      "mean_token_accuracy": 0.15461479648947715,
      "num_tokens": 32404745.0,
      "step": 16000
    },
    {
      "entropy": 6.208862257003784,
      "epoch": 1.3986716770077776,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048084742189712147,
      "loss": 5.9444,
      "mean_token_accuracy": 0.15324360728263856,
      "num_tokens": 32415325.0,
      "step": 16005
    },
    {
      "entropy": 6.287497520446777,
      "epoch": 1.3991086253604825,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004808348391999112,
      "loss": 6.038,
      "mean_token_accuracy": 0.14458170384168625,
      "num_tokens": 32425810.0,
      "step": 16010
    },
    {
      "entropy": 6.268242692947387,
      "epoch": 1.399545573713187,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004808222525546808,
      "loss": 6.0036,
      "mean_token_accuracy": 0.14741258919239045,
      "num_tokens": 32436799.0,
      "step": 16015
    },
    {
      "entropy": 6.233947896957398,
      "epoch": 1.399982522065892,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048080966196167163,
      "loss": 5.9093,
      "mean_token_accuracy": 0.1477771669626236,
      "num_tokens": 32446409.0,
      "step": 16020
    },
    {
      "entropy": 6.216257381439209,
      "epoch": 1.4004194704185966,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004807970674211255,
      "loss": 5.9616,
      "mean_token_accuracy": 0.14758329689502717,
      "num_tokens": 32456268.0,
      "step": 16025
    },
    {
      "entropy": 6.176546096801758,
      "epoch": 1.4008564187713013,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048078446893328364,
      "loss": 6.0608,
      "mean_token_accuracy": 0.143796955794096,
      "num_tokens": 32467622.0,
      "step": 16030
    },
    {
      "entropy": 6.287654733657837,
      "epoch": 1.401293367124006,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004807718664983879,
      "loss": 5.9976,
      "mean_token_accuracy": 0.1457716517150402,
      "num_tokens": 32477753.0,
      "step": 16035
    },
    {
      "entropy": 6.251500749588013,
      "epoch": 1.4017303154767107,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048075926011667997,
      "loss": 5.8939,
      "mean_token_accuracy": 0.15105220526456833,
      "num_tokens": 32487758.0,
      "step": 16040
    },
    {
      "entropy": 6.192537832260132,
      "epoch": 1.4021672638294154,
      "grad_norm": 4.34375,
      "learning_rate": 0.00048074664978840147,
      "loss": 5.9506,
      "mean_token_accuracy": 0.14708915501832961,
      "num_tokens": 32497725.0,
      "step": 16045
    },
    {
      "entropy": 6.19289083480835,
      "epoch": 1.40260421218212,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004807340355137945,
      "loss": 5.8568,
      "mean_token_accuracy": 0.1595625549554825,
      "num_tokens": 32507515.0,
      "step": 16050
    },
    {
      "entropy": 6.245632648468018,
      "epoch": 1.4030411605348247,
      "grad_norm": 2.578125,
      "learning_rate": 0.000480721417293101,
      "loss": 5.918,
      "mean_token_accuracy": 0.1559121146798134,
      "num_tokens": 32516943.0,
      "step": 16055
    },
    {
      "entropy": 6.152014446258545,
      "epoch": 1.4034781088875294,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004807087951265628,
      "loss": 5.8264,
      "mean_token_accuracy": 0.1638006165623665,
      "num_tokens": 32527223.0,
      "step": 16060
    },
    {
      "entropy": 6.196645927429199,
      "epoch": 1.4039150572402341,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004806961690144222,
      "loss": 5.9395,
      "mean_token_accuracy": 0.15159657821059228,
      "num_tokens": 32536668.0,
      "step": 16065
    },
    {
      "entropy": 6.261557388305664,
      "epoch": 1.4043520055929388,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004806835389569213,
      "loss": 5.9679,
      "mean_token_accuracy": 0.14944460839033127,
      "num_tokens": 32546566.0,
      "step": 16070
    },
    {
      "entropy": 6.160093688964844,
      "epoch": 1.4047889539456437,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048067090495430235,
      "loss": 5.8752,
      "mean_token_accuracy": 0.15293431654572487,
      "num_tokens": 32555927.0,
      "step": 16075
    },
    {
      "entropy": 6.252180910110473,
      "epoch": 1.4052259022983482,
      "grad_norm": 2.75,
      "learning_rate": 0.00048065826700680763,
      "loss": 5.925,
      "mean_token_accuracy": 0.14771692901849748,
      "num_tokens": 32566569.0,
      "step": 16080
    },
    {
      "entropy": 6.180606985092163,
      "epoch": 1.4056628506510531,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048064562511467956,
      "loss": 5.8899,
      "mean_token_accuracy": 0.15672991424798965,
      "num_tokens": 32577076.0,
      "step": 16085
    },
    {
      "entropy": 6.2085634708404545,
      "epoch": 1.4060997990037578,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048063297927816067,
      "loss": 5.9808,
      "mean_token_accuracy": 0.14610521197319032,
      "num_tokens": 32587926.0,
      "step": 16090
    },
    {
      "entropy": 6.284604263305664,
      "epoch": 1.4065367473564625,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004806203294974935,
      "loss": 6.0464,
      "mean_token_accuracy": 0.1469750262796879,
      "num_tokens": 32599012.0,
      "step": 16095
    },
    {
      "entropy": 6.311593627929687,
      "epoch": 1.4069736957091672,
      "grad_norm": 5.09375,
      "learning_rate": 0.00048060767577292067,
      "loss": 5.9275,
      "mean_token_accuracy": 0.14515378475189208,
      "num_tokens": 32609094.0,
      "step": 16100
    },
    {
      "entropy": 6.229230117797852,
      "epoch": 1.4074106440618719,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004805950181046849,
      "loss": 5.9931,
      "mean_token_accuracy": 0.148991596698761,
      "num_tokens": 32620356.0,
      "step": 16105
    },
    {
      "entropy": 6.20060362815857,
      "epoch": 1.4078475924145766,
      "grad_norm": 2.6875,
      "learning_rate": 0.000480582356493029,
      "loss": 5.9579,
      "mean_token_accuracy": 0.14758061915636062,
      "num_tokens": 32629841.0,
      "step": 16110
    },
    {
      "entropy": 6.246219730377197,
      "epoch": 1.4082845407672813,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004805696909381957,
      "loss": 5.9906,
      "mean_token_accuracy": 0.14725253432989122,
      "num_tokens": 32639006.0,
      "step": 16115
    },
    {
      "entropy": 6.291961288452148,
      "epoch": 1.408721489119986,
      "grad_norm": 2.625,
      "learning_rate": 0.00048055702144042794,
      "loss": 5.9856,
      "mean_token_accuracy": 0.14433166533708572,
      "num_tokens": 32648419.0,
      "step": 16120
    },
    {
      "entropy": 6.2535301685333256,
      "epoch": 1.4091584374726907,
      "grad_norm": 4.625,
      "learning_rate": 0.0004805443479999689,
      "loss": 5.9961,
      "mean_token_accuracy": 0.15461921244859694,
      "num_tokens": 32658050.0,
      "step": 16125
    },
    {
      "entropy": 6.226941776275635,
      "epoch": 1.4095953858253956,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004805316706170615,
      "loss": 5.9672,
      "mean_token_accuracy": 0.1499362602829933,
      "num_tokens": 32668176.0,
      "step": 16130
    },
    {
      "entropy": 6.156014490127563,
      "epoch": 1.4100323341781,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048051898929194907,
      "loss": 5.8945,
      "mean_token_accuracy": 0.1544634535908699,
      "num_tokens": 32678517.0,
      "step": 16135
    },
    {
      "entropy": 6.272788143157959,
      "epoch": 1.410469282530805,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004805063040248747,
      "loss": 5.9321,
      "mean_token_accuracy": 0.1507180415093899,
      "num_tokens": 32688459.0,
      "step": 16140
    },
    {
      "entropy": 6.26885666847229,
      "epoch": 1.4109062308835096,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004804936148160817,
      "loss": 5.9662,
      "mean_token_accuracy": 0.15213036984205247,
      "num_tokens": 32698481.0,
      "step": 16145
    },
    {
      "entropy": 6.185599374771118,
      "epoch": 1.4113431792362143,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004804809216658135,
      "loss": 5.9871,
      "mean_token_accuracy": 0.14561740532517434,
      "num_tokens": 32709031.0,
      "step": 16150
    },
    {
      "entropy": 6.21746072769165,
      "epoch": 1.411780127588919,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048046822457431347,
      "loss": 5.9743,
      "mean_token_accuracy": 0.15277617871761323,
      "num_tokens": 32720293.0,
      "step": 16155
    },
    {
      "entropy": 6.237871217727661,
      "epoch": 1.4122170759416237,
      "grad_norm": 2.125,
      "learning_rate": 0.00048045552354182536,
      "loss": 5.9849,
      "mean_token_accuracy": 0.15517314076423644,
      "num_tokens": 32730108.0,
      "step": 16160
    },
    {
      "entropy": 6.253078651428223,
      "epoch": 1.4126540242943284,
      "grad_norm": 2.859375,
      "learning_rate": 0.00048044281856859255,
      "loss": 5.9707,
      "mean_token_accuracy": 0.148306205868721,
      "num_tokens": 32740614.0,
      "step": 16165
    },
    {
      "entropy": 6.246525812149048,
      "epoch": 1.413090972647033,
      "grad_norm": 3.625,
      "learning_rate": 0.0004804301096548589,
      "loss": 5.9228,
      "mean_token_accuracy": 0.14760697782039642,
      "num_tokens": 32751155.0,
      "step": 16170
    },
    {
      "entropy": 6.147954845428467,
      "epoch": 1.4135279209997378,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004804173968008681,
      "loss": 5.8566,
      "mean_token_accuracy": 0.15081977248191833,
      "num_tokens": 32761067.0,
      "step": 16175
    },
    {
      "entropy": 6.103954601287842,
      "epoch": 1.4139648693524425,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048040468000686395,
      "loss": 5.8428,
      "mean_token_accuracy": 0.1576648086309433,
      "num_tokens": 32771357.0,
      "step": 16180
    },
    {
      "entropy": 6.2469274520874025,
      "epoch": 1.4144018177051472,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004803919592730904,
      "loss": 6.0369,
      "mean_token_accuracy": 0.14376489520072938,
      "num_tokens": 32781116.0,
      "step": 16185
    },
    {
      "entropy": 6.297688388824463,
      "epoch": 1.4148387660578519,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004803792345997915,
      "loss": 6.002,
      "mean_token_accuracy": 0.1477140575647354,
      "num_tokens": 32791660.0,
      "step": 16190
    },
    {
      "entropy": 6.229465579986572,
      "epoch": 1.4152757144105568,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048036650598721125,
      "loss": 5.9622,
      "mean_token_accuracy": 0.14267654791474343,
      "num_tokens": 32801413.0,
      "step": 16195
    },
    {
      "entropy": 6.2252421379089355,
      "epoch": 1.4157126627632615,
      "grad_norm": 3.75,
      "learning_rate": 0.00048035377343559377,
      "loss": 6.0036,
      "mean_token_accuracy": 0.14771608263254166,
      "num_tokens": 32811293.0,
      "step": 16200
    },
    {
      "entropy": 6.290157461166382,
      "epoch": 1.4161496111159662,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048034103694518336,
      "loss": 6.007,
      "mean_token_accuracy": 0.14904130399227142,
      "num_tokens": 32821609.0,
      "step": 16205
    },
    {
      "entropy": 6.268910312652588,
      "epoch": 1.4165865594686708,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048032829651622425,
      "loss": 5.9587,
      "mean_token_accuracy": 0.14592699110507965,
      "num_tokens": 32831219.0,
      "step": 16210
    },
    {
      "entropy": 6.196556138992309,
      "epoch": 1.4170235078213755,
      "grad_norm": 3.828125,
      "learning_rate": 0.00048031555214896076,
      "loss": 5.9749,
      "mean_token_accuracy": 0.1409635901451111,
      "num_tokens": 32841085.0,
      "step": 16215
    },
    {
      "entropy": 6.2680871963500975,
      "epoch": 1.4174604561740802,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004803028038436375,
      "loss": 5.9351,
      "mean_token_accuracy": 0.15112167000770568,
      "num_tokens": 32851592.0,
      "step": 16220
    },
    {
      "entropy": 6.3080836772918705,
      "epoch": 1.417897404526785,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048029005160049886,
      "loss": 5.929,
      "mean_token_accuracy": 0.1462145194411278,
      "num_tokens": 32862124.0,
      "step": 16225
    },
    {
      "entropy": 6.191161155700684,
      "epoch": 1.4183343528794896,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004802772954197894,
      "loss": 5.9157,
      "mean_token_accuracy": 0.15759778320789336,
      "num_tokens": 32872760.0,
      "step": 16230
    },
    {
      "entropy": 6.200194454193115,
      "epoch": 1.4187713012321943,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048026453530175393,
      "loss": 6.0333,
      "mean_token_accuracy": 0.14538905397057533,
      "num_tokens": 32883542.0,
      "step": 16235
    },
    {
      "entropy": 6.252602386474609,
      "epoch": 1.419208249584899,
      "grad_norm": 4.21875,
      "learning_rate": 0.00048025177124663704,
      "loss": 5.9225,
      "mean_token_accuracy": 0.1476988300681114,
      "num_tokens": 32893922.0,
      "step": 16240
    },
    {
      "entropy": 6.253632831573486,
      "epoch": 1.4196451979376037,
      "grad_norm": 3.625,
      "learning_rate": 0.00048023900325468367,
      "loss": 5.9802,
      "mean_token_accuracy": 0.15559055060148239,
      "num_tokens": 32903907.0,
      "step": 16245
    },
    {
      "entropy": 6.238755750656128,
      "epoch": 1.4200821462903086,
      "grad_norm": 4.28125,
      "learning_rate": 0.00048022623132613863,
      "loss": 5.9752,
      "mean_token_accuracy": 0.1458921805024147,
      "num_tokens": 32914656.0,
      "step": 16250
    },
    {
      "entropy": 6.210338830947876,
      "epoch": 1.420519094643013,
      "grad_norm": 2.984375,
      "learning_rate": 0.000480213455461247,
      "loss": 5.9927,
      "mean_token_accuracy": 0.13971116095781327,
      "num_tokens": 32925250.0,
      "step": 16255
    },
    {
      "entropy": 6.226225662231445,
      "epoch": 1.420956042995718,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048020067566025373,
      "loss": 5.9688,
      "mean_token_accuracy": 0.1488567814230919,
      "num_tokens": 32935435.0,
      "step": 16260
    },
    {
      "entropy": 6.286094331741333,
      "epoch": 1.4213929913484227,
      "grad_norm": 2.671875,
      "learning_rate": 0.000480187891923404,
      "loss": 5.9603,
      "mean_token_accuracy": 0.1469062991440296,
      "num_tokens": 32946166.0,
      "step": 16265
    },
    {
      "entropy": 6.203932094573974,
      "epoch": 1.4218299397011274,
      "grad_norm": 9.3125,
      "learning_rate": 0.0004801751042509429,
      "loss": 5.918,
      "mean_token_accuracy": 0.15250276327133178,
      "num_tokens": 32955751.0,
      "step": 16270
    },
    {
      "entropy": 6.209169530868531,
      "epoch": 1.422266888053832,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004801623126431159,
      "loss": 5.9973,
      "mean_token_accuracy": 0.14256021231412888,
      "num_tokens": 32966121.0,
      "step": 16275
    },
    {
      "entropy": 6.239215230941772,
      "epoch": 1.4227038364065367,
      "grad_norm": 5.65625,
      "learning_rate": 0.0004801495171001681,
      "loss": 5.9388,
      "mean_token_accuracy": 0.15983382761478424,
      "num_tokens": 32976950.0,
      "step": 16280
    },
    {
      "entropy": 6.246411323547363,
      "epoch": 1.4231407847592414,
      "grad_norm": 2.875,
      "learning_rate": 0.0004801367176223452,
      "loss": 5.9227,
      "mean_token_accuracy": 0.1513407863676548,
      "num_tokens": 32987278.0,
      "step": 16285
    },
    {
      "entropy": 6.235836505889893,
      "epoch": 1.4235777331119461,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004801239142098926,
      "loss": 6.0203,
      "mean_token_accuracy": 0.1415890708565712,
      "num_tokens": 32997050.0,
      "step": 16290
    },
    {
      "entropy": 6.282905387878418,
      "epoch": 1.4240146814646508,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004801111068630557,
      "loss": 5.9499,
      "mean_token_accuracy": 0.15070197880268096,
      "num_tokens": 33007799.0,
      "step": 16295
    },
    {
      "entropy": 6.29005069732666,
      "epoch": 1.4244516298173555,
      "grad_norm": 2.90625,
      "learning_rate": 0.00048009829558208044,
      "loss": 5.9297,
      "mean_token_accuracy": 0.15235686749219896,
      "num_tokens": 33017476.0,
      "step": 16300
    },
    {
      "entropy": 6.185603666305542,
      "epoch": 1.4248885781700604,
      "grad_norm": 3.328125,
      "learning_rate": 0.00048008548036721226,
      "loss": 5.9166,
      "mean_token_accuracy": 0.15080729573965074,
      "num_tokens": 33027617.0,
      "step": 16305
    },
    {
      "entropy": 6.220965242385864,
      "epoch": 1.425325526522765,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004800726612186972,
      "loss": 5.9833,
      "mean_token_accuracy": 0.1481410399079323,
      "num_tokens": 33037257.0,
      "step": 16310
    },
    {
      "entropy": 6.146099042892456,
      "epoch": 1.4257624748754698,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004800598381367811,
      "loss": 5.9176,
      "mean_token_accuracy": 0.1531670033931732,
      "num_tokens": 33047406.0,
      "step": 16315
    },
    {
      "entropy": 6.22388653755188,
      "epoch": 1.4261994232281745,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004800470111217096,
      "loss": 5.9874,
      "mean_token_accuracy": 0.1541450649499893,
      "num_tokens": 33057975.0,
      "step": 16320
    },
    {
      "entropy": 6.344366407394409,
      "epoch": 1.4266363715808792,
      "grad_norm": 3.140625,
      "learning_rate": 0.00048003418017372924,
      "loss": 6.0149,
      "mean_token_accuracy": 0.14468791708350182,
      "num_tokens": 33068645.0,
      "step": 16325
    },
    {
      "entropy": 6.256756734848023,
      "epoch": 1.4270733199335839,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004800213452930857,
      "loss": 5.9809,
      "mean_token_accuracy": 0.15164679139852524,
      "num_tokens": 33079147.0,
      "step": 16330
    },
    {
      "entropy": 6.251343250274658,
      "epoch": 1.4275102682862886,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004800085064800255,
      "loss": 5.9389,
      "mean_token_accuracy": 0.14512138664722443,
      "num_tokens": 33089229.0,
      "step": 16335
    },
    {
      "entropy": 6.285108661651611,
      "epoch": 1.4279472166389933,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004799956637347946,
      "loss": 5.9461,
      "mean_token_accuracy": 0.15022461414337157,
      "num_tokens": 33099688.0,
      "step": 16340
    },
    {
      "entropy": 6.261924695968628,
      "epoch": 1.428384164991698,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004799828170576395,
      "loss": 5.9638,
      "mean_token_accuracy": 0.15029398798942567,
      "num_tokens": 33110019.0,
      "step": 16345
    },
    {
      "entropy": 6.1305605411529545,
      "epoch": 1.4288211133444026,
      "grad_norm": 2.375,
      "learning_rate": 0.0004799699664488065,
      "loss": 5.8526,
      "mean_token_accuracy": 0.15944435745477675,
      "num_tokens": 33120141.0,
      "step": 16350
    },
    {
      "entropy": 6.216370153427124,
      "epoch": 1.4292580616971073,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004799571119085422,
      "loss": 5.9147,
      "mean_token_accuracy": 0.14900848418474197,
      "num_tokens": 33130047.0,
      "step": 16355
    },
    {
      "entropy": 6.296254873275757,
      "epoch": 1.429695010049812,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004799442534370931,
      "loss": 6.0239,
      "mean_token_accuracy": 0.1426090642809868,
      "num_tokens": 33141234.0,
      "step": 16360
    },
    {
      "entropy": 6.25029730796814,
      "epoch": 1.4301319584025167,
      "grad_norm": 2.5,
      "learning_rate": 0.0004799313910347057,
      "loss": 5.9558,
      "mean_token_accuracy": 0.15441144555807113,
      "num_tokens": 33150788.0,
      "step": 16365
    },
    {
      "entropy": 6.15256290435791,
      "epoch": 1.4305689067552216,
      "grad_norm": 2.84375,
      "learning_rate": 0.000479918524701627,
      "loss": 5.9597,
      "mean_token_accuracy": 0.1508673384785652,
      "num_tokens": 33160928.0,
      "step": 16370
    },
    {
      "entropy": 6.297958469390869,
      "epoch": 1.431005855107926,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047990565443810357,
      "loss": 6.0534,
      "mean_token_accuracy": 0.13855497986078263,
      "num_tokens": 33171333.0,
      "step": 16375
    },
    {
      "entropy": 6.172018098831177,
      "epoch": 1.431442803460631,
      "grad_norm": 3.15625,
      "learning_rate": 0.00047989278024438233,
      "loss": 5.8407,
      "mean_token_accuracy": 0.15293041467666627,
      "num_tokens": 33180396.0,
      "step": 16380
    },
    {
      "entropy": 6.177785587310791,
      "epoch": 1.4318797518133357,
      "grad_norm": 3.625,
      "learning_rate": 0.0004798799021207102,
      "loss": 5.926,
      "mean_token_accuracy": 0.1482950896024704,
      "num_tokens": 33189571.0,
      "step": 16385
    },
    {
      "entropy": 6.167497014999389,
      "epoch": 1.4323167001660404,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047986702006733417,
      "loss": 5.9063,
      "mean_token_accuracy": 0.14920893386006356,
      "num_tokens": 33198979.0,
      "step": 16390
    },
    {
      "entropy": 6.279789972305298,
      "epoch": 1.432753648518745,
      "grad_norm": 2.75,
      "learning_rate": 0.00047985413408450136,
      "loss": 5.954,
      "mean_token_accuracy": 0.15486048609018327,
      "num_tokens": 33208862.0,
      "step": 16395
    },
    {
      "entropy": 6.238927507400513,
      "epoch": 1.4331905968714498,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004798412441724589,
      "loss": 5.9824,
      "mean_token_accuracy": 0.14477131739258767,
      "num_tokens": 33218804.0,
      "step": 16400
    },
    {
      "entropy": 6.233418607711792,
      "epoch": 1.4336275452241545,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047982835033145413,
      "loss": 5.9923,
      "mean_token_accuracy": 0.14491497799754144,
      "num_tokens": 33228783.0,
      "step": 16405
    },
    {
      "entropy": 6.329058074951172,
      "epoch": 1.4340644935768592,
      "grad_norm": 3.484375,
      "learning_rate": 0.00047981545256173415,
      "loss": 5.9993,
      "mean_token_accuracy": 0.14347606301307678,
      "num_tokens": 33241029.0,
      "step": 16410
    },
    {
      "entropy": 6.335453653335572,
      "epoch": 1.4345014419295639,
      "grad_norm": 3.375,
      "learning_rate": 0.0004798025508635465,
      "loss": 6.0168,
      "mean_token_accuracy": 0.14243985190987588,
      "num_tokens": 33250383.0,
      "step": 16415
    },
    {
      "entropy": 6.185868406295777,
      "epoch": 1.4349383902822685,
      "grad_norm": 2.625,
      "learning_rate": 0.0004797896452371386,
      "loss": 5.8104,
      "mean_token_accuracy": 0.16262106895446776,
      "num_tokens": 33260424.0,
      "step": 16420
    },
    {
      "entropy": 6.212612247467041,
      "epoch": 1.4353753386349735,
      "grad_norm": 3.53125,
      "learning_rate": 0.00047977673568275805,
      "loss": 5.9076,
      "mean_token_accuracy": 0.1447680950164795,
      "num_tokens": 33270200.0,
      "step": 16425
    },
    {
      "entropy": 6.269780111312866,
      "epoch": 1.435812286987678,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004797638222006524,
      "loss": 6.0587,
      "mean_token_accuracy": 0.13845586776733398,
      "num_tokens": 33279850.0,
      "step": 16430
    },
    {
      "entropy": 6.211544752120972,
      "epoch": 1.4362492353403828,
      "grad_norm": 2.875,
      "learning_rate": 0.0004797509047910693,
      "loss": 5.9226,
      "mean_token_accuracy": 0.15329899787902831,
      "num_tokens": 33289053.0,
      "step": 16435
    },
    {
      "entropy": 6.275228357315063,
      "epoch": 1.4366861836930875,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047973798345425655,
      "loss": 5.9317,
      "mean_token_accuracy": 0.15224664211273192,
      "num_tokens": 33298988.0,
      "step": 16440
    },
    {
      "entropy": 6.269676065444946,
      "epoch": 1.4371231320457922,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047972505819046207,
      "loss": 5.9562,
      "mean_token_accuracy": 0.15344640463590623,
      "num_tokens": 33308683.0,
      "step": 16445
    },
    {
      "entropy": 6.1021791934967045,
      "epoch": 1.437560080398497,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004797121289999336,
      "loss": 5.9471,
      "mean_token_accuracy": 0.14735542088747025,
      "num_tokens": 33319230.0,
      "step": 16450
    },
    {
      "entropy": 6.245175218582153,
      "epoch": 1.4379970287512016,
      "grad_norm": 3.625,
      "learning_rate": 0.0004796991958829193,
      "loss": 6.0021,
      "mean_token_accuracy": 0.14225824028253556,
      "num_tokens": 33329277.0,
      "step": 16455
    },
    {
      "entropy": 6.295068168640137,
      "epoch": 1.4384339771039063,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004796862588396672,
      "loss": 5.9334,
      "mean_token_accuracy": 0.14979828745126725,
      "num_tokens": 33338644.0,
      "step": 16460
    },
    {
      "entropy": 6.234681892395019,
      "epoch": 1.438870925456611,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047967331787042536,
      "loss": 5.8603,
      "mean_token_accuracy": 0.16096805930137634,
      "num_tokens": 33348996.0,
      "step": 16465
    },
    {
      "entropy": 6.177588558197021,
      "epoch": 1.4393078738093157,
      "grad_norm": 6.15625,
      "learning_rate": 0.00047966037297544195,
      "loss": 5.9489,
      "mean_token_accuracy": 0.1462039642035961,
      "num_tokens": 33358581.0,
      "step": 16470
    },
    {
      "entropy": 6.212748241424561,
      "epoch": 1.4397448221620204,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047964742415496553,
      "loss": 5.9963,
      "mean_token_accuracy": 0.14993032515048982,
      "num_tokens": 33368925.0,
      "step": 16475
    },
    {
      "entropy": 6.268782901763916,
      "epoch": 1.4401817705147253,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004796344714092441,
      "loss": 5.9837,
      "mean_token_accuracy": 0.15035089626908302,
      "num_tokens": 33378918.0,
      "step": 16480
    },
    {
      "entropy": 6.232730865478516,
      "epoch": 1.4406187188674298,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004796215147385264,
      "loss": 5.9661,
      "mean_token_accuracy": 0.14385876506567002,
      "num_tokens": 33389039.0,
      "step": 16485
    },
    {
      "entropy": 6.291960000991821,
      "epoch": 1.4410556672201347,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004796085541430608,
      "loss": 5.8937,
      "mean_token_accuracy": 0.15657427310943603,
      "num_tokens": 33398893.0,
      "step": 16490
    },
    {
      "entropy": 6.309097766876221,
      "epoch": 1.4414926155728394,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004795955896230959,
      "loss": 5.9893,
      "mean_token_accuracy": 0.14650255292654038,
      "num_tokens": 33408188.0,
      "step": 16495
    },
    {
      "entropy": 6.215656566619873,
      "epoch": 1.441929563925544,
      "grad_norm": 3.484375,
      "learning_rate": 0.00047958262117888035,
      "loss": 5.9414,
      "mean_token_accuracy": 0.15506115108728408,
      "num_tokens": 33418926.0,
      "step": 16500
    },
    {
      "entropy": 6.224849224090576,
      "epoch": 1.4423665122782487,
      "grad_norm": 4.09375,
      "learning_rate": 0.00047956964881066297,
      "loss": 5.9826,
      "mean_token_accuracy": 0.15005790963768958,
      "num_tokens": 33429790.0,
      "step": 16505
    },
    {
      "entropy": 6.259890031814575,
      "epoch": 1.4428034606309534,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004795566725186925,
      "loss": 5.9668,
      "mean_token_accuracy": 0.14999188482761383,
      "num_tokens": 33440470.0,
      "step": 16510
    },
    {
      "entropy": 6.2552460670471195,
      "epoch": 1.4432404089836581,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004795436923032179,
      "loss": 5.9382,
      "mean_token_accuracy": 0.1503315970301628,
      "num_tokens": 33449875.0,
      "step": 16515
    },
    {
      "entropy": 6.21501579284668,
      "epoch": 1.4436773573363628,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004795307081644881,
      "loss": 5.9298,
      "mean_token_accuracy": 0.1515965387225151,
      "num_tokens": 33458841.0,
      "step": 16520
    },
    {
      "entropy": 6.173931312561035,
      "epoch": 1.4441143056890675,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004795177201027521,
      "loss": 5.8809,
      "mean_token_accuracy": 0.15887403786182402,
      "num_tokens": 33468656.0,
      "step": 16525
    },
    {
      "entropy": 6.228793048858643,
      "epoch": 1.4445512540417722,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047950472811825916,
      "loss": 5.9495,
      "mean_token_accuracy": 0.1519930198788643,
      "num_tokens": 33478977.0,
      "step": 16530
    },
    {
      "entropy": 6.294742107391357,
      "epoch": 1.444988202394477,
      "grad_norm": 2.0,
      "learning_rate": 0.0004794917322112583,
      "loss": 5.9219,
      "mean_token_accuracy": 0.14499156922101974,
      "num_tokens": 33488418.0,
      "step": 16535
    },
    {
      "entropy": 6.174397611618042,
      "epoch": 1.4454251507471816,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004794787323819989,
      "loss": 5.9208,
      "mean_token_accuracy": 0.15553233921527862,
      "num_tokens": 33499125.0,
      "step": 16540
    },
    {
      "entropy": 6.20950665473938,
      "epoch": 1.4458620990998865,
      "grad_norm": 2.6875,
      "learning_rate": 0.00047946572863073023,
      "loss": 5.935,
      "mean_token_accuracy": 0.14736442118883133,
      "num_tokens": 33508826.0,
      "step": 16545
    },
    {
      "entropy": 6.202508068084716,
      "epoch": 1.446299047452591,
      "grad_norm": 2.6875,
      "learning_rate": 0.00047945272095770174,
      "loss": 5.855,
      "mean_token_accuracy": 0.16542330980300904,
      "num_tokens": 33518398.0,
      "step": 16550
    },
    {
      "entropy": 6.202480411529541,
      "epoch": 1.4467359958052959,
      "grad_norm": 3.203125,
      "learning_rate": 0.000479439709363163,
      "loss": 5.9611,
      "mean_token_accuracy": 0.15026971250772475,
      "num_tokens": 33528289.0,
      "step": 16555
    },
    {
      "entropy": 6.2738584041595455,
      "epoch": 1.4471729441580006,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004794266938473635,
      "loss": 6.0323,
      "mean_token_accuracy": 0.1416584938764572,
      "num_tokens": 33539228.0,
      "step": 16560
    },
    {
      "entropy": 6.2578041553497314,
      "epoch": 1.4476098925107053,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047941367441055283,
      "loss": 5.9975,
      "mean_token_accuracy": 0.14788940101861953,
      "num_tokens": 33551015.0,
      "step": 16565
    },
    {
      "entropy": 6.2101719856262205,
      "epoch": 1.44804684086341,
      "grad_norm": 2.625,
      "learning_rate": 0.00047940065105298073,
      "loss": 5.9342,
      "mean_token_accuracy": 0.1546376571059227,
      "num_tokens": 33560374.0,
      "step": 16570
    },
    {
      "entropy": 6.143131303787231,
      "epoch": 1.4484837892161146,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004793876237748972,
      "loss": 5.7894,
      "mean_token_accuracy": 0.16446930915117264,
      "num_tokens": 33569916.0,
      "step": 16575
    },
    {
      "entropy": 6.15258936882019,
      "epoch": 1.4489207375688193,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004793745925765518,
      "loss": 5.8646,
      "mean_token_accuracy": 0.15508723855018616,
      "num_tokens": 33579930.0,
      "step": 16580
    },
    {
      "entropy": 6.160069322586059,
      "epoch": 1.449357685921524,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004793615574581947,
      "loss": 5.9577,
      "mean_token_accuracy": 0.1475573569536209,
      "num_tokens": 33588612.0,
      "step": 16585
    },
    {
      "entropy": 6.286575269699097,
      "epoch": 1.4497946342742287,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047934851842007585,
      "loss": 5.961,
      "mean_token_accuracy": 0.14599469006061555,
      "num_tokens": 33599253.0,
      "step": 16590
    },
    {
      "entropy": 6.267388343811035,
      "epoch": 1.4502315826269334,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004793354754624453,
      "loss": 5.8945,
      "mean_token_accuracy": 0.15684460401535033,
      "num_tokens": 33609484.0,
      "step": 16595
    },
    {
      "entropy": 6.225119876861572,
      "epoch": 1.4506685309796383,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004793224285855533,
      "loss": 5.8902,
      "mean_token_accuracy": 0.15363344550132751,
      "num_tokens": 33619660.0,
      "step": 16600
    },
    {
      "entropy": 6.199349784851075,
      "epoch": 1.4511054793323428,
      "grad_norm": 3.4375,
      "learning_rate": 0.00047930937778965,
      "loss": 5.9837,
      "mean_token_accuracy": 0.14849578142166137,
      "num_tokens": 33629505.0,
      "step": 16605
    },
    {
      "entropy": 6.243946933746338,
      "epoch": 1.4515424276850477,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004792963230749858,
      "loss": 5.9954,
      "mean_token_accuracy": 0.14741640016436577,
      "num_tokens": 33639680.0,
      "step": 16610
    },
    {
      "entropy": 6.228905582427979,
      "epoch": 1.4519793760377524,
      "grad_norm": 3.234375,
      "learning_rate": 0.00047928326444181117,
      "loss": 5.9771,
      "mean_token_accuracy": 0.14521362632513046,
      "num_tokens": 33649200.0,
      "step": 16615
    },
    {
      "entropy": 6.242034912109375,
      "epoch": 1.452416324390457,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047927020189037634,
      "loss": 5.922,
      "mean_token_accuracy": 0.15206531062722206,
      "num_tokens": 33658386.0,
      "step": 16620
    },
    {
      "entropy": 6.213899326324463,
      "epoch": 1.4528532727431618,
      "grad_norm": 7.5,
      "learning_rate": 0.0004792571354209321,
      "loss": 5.886,
      "mean_token_accuracy": 0.1557728350162506,
      "num_tokens": 33669139.0,
      "step": 16625
    },
    {
      "entropy": 6.234268236160278,
      "epoch": 1.4532902210958665,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004792440650337289,
      "loss": 6.0119,
      "mean_token_accuracy": 0.15172863528132438,
      "num_tokens": 33679629.0,
      "step": 16630
    },
    {
      "entropy": 6.260464715957641,
      "epoch": 1.4537271694485712,
      "grad_norm": 3.5,
      "learning_rate": 0.0004792309907290175,
      "loss": 5.9782,
      "mean_token_accuracy": 0.14303461462259293,
      "num_tokens": 33689633.0,
      "step": 16635
    },
    {
      "entropy": 6.261146926879883,
      "epoch": 1.4541641178012759,
      "grad_norm": 2.53125,
      "learning_rate": 0.00047921791250704875,
      "loss": 5.9424,
      "mean_token_accuracy": 0.14458367973566055,
      "num_tokens": 33699719.0,
      "step": 16640
    },
    {
      "entropy": 6.288723087310791,
      "epoch": 1.4546010661539805,
      "grad_norm": 4.28125,
      "learning_rate": 0.00047920483036807346,
      "loss": 5.9851,
      "mean_token_accuracy": 0.1452558621764183,
      "num_tokens": 33709389.0,
      "step": 16645
    },
    {
      "entropy": 6.256852006912231,
      "epoch": 1.4550380145066852,
      "grad_norm": 3.453125,
      "learning_rate": 0.00047919174431234246,
      "loss": 5.9666,
      "mean_token_accuracy": 0.1453314781188965,
      "num_tokens": 33719063.0,
      "step": 16650
    },
    {
      "entropy": 6.311029148101807,
      "epoch": 1.4554749628593902,
      "grad_norm": 5.59375,
      "learning_rate": 0.0004791786543401068,
      "loss": 6.0478,
      "mean_token_accuracy": 0.1352648541331291,
      "num_tokens": 33729798.0,
      "step": 16655
    },
    {
      "entropy": 6.245541906356811,
      "epoch": 1.4559119112120946,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004791655604516176,
      "loss": 5.976,
      "mean_token_accuracy": 0.13919324576854705,
      "num_tokens": 33740012.0,
      "step": 16660
    },
    {
      "entropy": 6.2229071140289305,
      "epoch": 1.4563488595647995,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047915246264712595,
      "loss": 5.9078,
      "mean_token_accuracy": 0.15275118127465248,
      "num_tokens": 33749478.0,
      "step": 16665
    },
    {
      "entropy": 6.194706010818481,
      "epoch": 1.4567858079175042,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004791393609268831,
      "loss": 5.7893,
      "mean_token_accuracy": 0.16538517773151398,
      "num_tokens": 33758935.0,
      "step": 16670
    },
    {
      "entropy": 6.205597591400147,
      "epoch": 1.457222756270209,
      "grad_norm": 9.875,
      "learning_rate": 0.0004791262552911404,
      "loss": 5.9077,
      "mean_token_accuracy": 0.15170277208089827,
      "num_tokens": 33769035.0,
      "step": 16675
    },
    {
      "entropy": 6.198661756515503,
      "epoch": 1.4576597046229136,
      "grad_norm": 2.984375,
      "learning_rate": 0.00047911314574014904,
      "loss": 5.9264,
      "mean_token_accuracy": 0.15082931518554688,
      "num_tokens": 33779288.0,
      "step": 16680
    },
    {
      "entropy": 6.22617015838623,
      "epoch": 1.4580966529756183,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004791000322741606,
      "loss": 5.9423,
      "mean_token_accuracy": 0.149807932972908,
      "num_tokens": 33789138.0,
      "step": 16685
    },
    {
      "entropy": 6.259691619873047,
      "epoch": 1.458533601328323,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004790869148934267,
      "loss": 5.9541,
      "mean_token_accuracy": 0.14378884434700012,
      "num_tokens": 33798649.0,
      "step": 16690
    },
    {
      "entropy": 6.193570232391357,
      "epoch": 1.4589705496810277,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047907379359819877,
      "loss": 5.8597,
      "mean_token_accuracy": 0.16417092382907866,
      "num_tokens": 33808742.0,
      "step": 16695
    },
    {
      "entropy": 6.149473094940186,
      "epoch": 1.4594074980337324,
      "grad_norm": 2.5,
      "learning_rate": 0.0004790606683887286,
      "loss": 5.8953,
      "mean_token_accuracy": 0.1553276002407074,
      "num_tokens": 33819507.0,
      "step": 16700
    },
    {
      "entropy": 6.252441453933716,
      "epoch": 1.459844446386437,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047904753926526777,
      "loss": 5.9655,
      "mean_token_accuracy": 0.144800066947937,
      "num_tokens": 33829722.0,
      "step": 16705
    },
    {
      "entropy": 6.235661315917969,
      "epoch": 1.4602813947391418,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004790344062280683,
      "loss": 6.0028,
      "mean_token_accuracy": 0.14807802587747573,
      "num_tokens": 33839350.0,
      "step": 16710
    },
    {
      "entropy": 6.180997467041015,
      "epoch": 1.4607183430918464,
      "grad_norm": 1.953125,
      "learning_rate": 0.00047902126927738196,
      "loss": 5.9311,
      "mean_token_accuracy": 0.16105913892388343,
      "num_tokens": 33849233.0,
      "step": 16715
    },
    {
      "entropy": 6.279181861877442,
      "epoch": 1.4611552914445514,
      "grad_norm": 3.65625,
      "learning_rate": 0.00047900812841346086,
      "loss": 6.003,
      "mean_token_accuracy": 0.1435595341026783,
      "num_tokens": 33859996.0,
      "step": 16720
    },
    {
      "entropy": 6.307798576354981,
      "epoch": 1.4615922397972558,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004789949836365569,
      "loss": 5.9861,
      "mean_token_accuracy": 0.1471328020095825,
      "num_tokens": 33870780.0,
      "step": 16725
    },
    {
      "entropy": 6.260661029815674,
      "epoch": 1.4620291881499607,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004789818349469222,
      "loss": 5.9623,
      "mean_token_accuracy": 0.15092600062489508,
      "num_tokens": 33880182.0,
      "step": 16730
    },
    {
      "entropy": 6.221206331253052,
      "epoch": 1.4624661365026654,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004789686823448091,
      "loss": 5.9711,
      "mean_token_accuracy": 0.14411755055189132,
      "num_tokens": 33889464.0,
      "step": 16735
    },
    {
      "entropy": 6.25152997970581,
      "epoch": 1.4629030848553701,
      "grad_norm": 2.0,
      "learning_rate": 0.0004789555258304698,
      "loss": 5.9189,
      "mean_token_accuracy": 0.14680106341838836,
      "num_tokens": 33899530.0,
      "step": 16740
    },
    {
      "entropy": 6.243572330474853,
      "epoch": 1.4633400332080748,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004789423654041566,
      "loss": 5.9918,
      "mean_token_accuracy": 0.1477881044149399,
      "num_tokens": 33910094.0,
      "step": 16745
    },
    {
      "entropy": 6.259621620178223,
      "epoch": 1.4637769815607795,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047892920106612204,
      "loss": 5.9304,
      "mean_token_accuracy": 0.15102397203445433,
      "num_tokens": 33921181.0,
      "step": 16750
    },
    {
      "entropy": 6.23029637336731,
      "epoch": 1.4642139299134842,
      "grad_norm": 3.0,
      "learning_rate": 0.0004789160328166184,
      "loss": 6.03,
      "mean_token_accuracy": 0.14533622562885284,
      "num_tokens": 33931567.0,
      "step": 16755
    },
    {
      "entropy": 6.259760808944702,
      "epoch": 1.464650878266189,
      "grad_norm": 3.125,
      "learning_rate": 0.00047890286065589856,
      "loss": 5.9884,
      "mean_token_accuracy": 0.14412049651145936,
      "num_tokens": 33941513.0,
      "step": 16760
    },
    {
      "entropy": 6.212833452224731,
      "epoch": 1.4650878266188936,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047888968458421487,
      "loss": 5.9255,
      "mean_token_accuracy": 0.14918189346790314,
      "num_tokens": 33953400.0,
      "step": 16765
    },
    {
      "entropy": 6.294416999816894,
      "epoch": 1.4655247749715983,
      "grad_norm": 2.375,
      "learning_rate": 0.0004788765046018204,
      "loss": 5.9837,
      "mean_token_accuracy": 0.1473192796111107,
      "num_tokens": 33964394.0,
      "step": 16770
    },
    {
      "entropy": 6.2879266262054445,
      "epoch": 1.4659617233243032,
      "grad_norm": 2.375,
      "learning_rate": 0.00047886332070896747,
      "loss": 6.0665,
      "mean_token_accuracy": 0.1400330424308777,
      "num_tokens": 33975266.0,
      "step": 16775
    },
    {
      "entropy": 6.271649026870728,
      "epoch": 1.4663986716770077,
      "grad_norm": 5.5,
      "learning_rate": 0.0004788501329059094,
      "loss": 6.0073,
      "mean_token_accuracy": 0.14281395226716995,
      "num_tokens": 33986159.0,
      "step": 16780
    },
    {
      "entropy": 6.3435163497924805,
      "epoch": 1.4668356200297126,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047883694119289896,
      "loss": 5.9965,
      "mean_token_accuracy": 0.14631564766168595,
      "num_tokens": 33997792.0,
      "step": 16785
    },
    {
      "entropy": 6.303979158401489,
      "epoch": 1.4672725683824173,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047882374557018916,
      "loss": 5.9683,
      "mean_token_accuracy": 0.15069078356027604,
      "num_tokens": 34007945.0,
      "step": 16790
    },
    {
      "entropy": 6.2502799987792965,
      "epoch": 1.467709516735122,
      "grad_norm": 4.59375,
      "learning_rate": 0.00047881054603803307,
      "loss": 6.0201,
      "mean_token_accuracy": 0.150532416254282,
      "num_tokens": 34018399.0,
      "step": 16795
    },
    {
      "entropy": 6.24795446395874,
      "epoch": 1.4681464650878266,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004787973425966839,
      "loss": 6.064,
      "mean_token_accuracy": 0.13784808963537215,
      "num_tokens": 34028267.0,
      "step": 16800
    },
    {
      "entropy": 6.289545679092408,
      "epoch": 1.4685834134405313,
      "grad_norm": 4.0625,
      "learning_rate": 0.000478784135246395,
      "loss": 6.0757,
      "mean_token_accuracy": 0.13724390491843225,
      "num_tokens": 34037904.0,
      "step": 16805
    },
    {
      "entropy": 6.277569150924682,
      "epoch": 1.469020361793236,
      "grad_norm": 6.21875,
      "learning_rate": 0.0004787709239874196,
      "loss": 5.9133,
      "mean_token_accuracy": 0.1552688866853714,
      "num_tokens": 34049203.0,
      "step": 16810
    },
    {
      "entropy": 6.32661805152893,
      "epoch": 1.4694573101459407,
      "grad_norm": 2.59375,
      "learning_rate": 0.000478757708820011,
      "loss": 6.0142,
      "mean_token_accuracy": 0.1511332094669342,
      "num_tokens": 34059670.0,
      "step": 16815
    },
    {
      "entropy": 6.220118999481201,
      "epoch": 1.4698942584986454,
      "grad_norm": 2.875,
      "learning_rate": 0.0004787444897444229,
      "loss": 5.9211,
      "mean_token_accuracy": 0.15061343759298323,
      "num_tokens": 34069730.0,
      "step": 16820
    },
    {
      "entropy": 6.262408876419068,
      "epoch": 1.47033120685135,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004787312667609086,
      "loss": 5.9269,
      "mean_token_accuracy": 0.14924972876906395,
      "num_tokens": 34079315.0,
      "step": 16825
    },
    {
      "entropy": 6.215321016311646,
      "epoch": 1.470768155204055,
      "grad_norm": 7.0,
      "learning_rate": 0.000478718039869722,
      "loss": 5.94,
      "mean_token_accuracy": 0.15296324640512465,
      "num_tokens": 34089431.0,
      "step": 16830
    },
    {
      "entropy": 6.231591653823853,
      "epoch": 1.4712051035567595,
      "grad_norm": 3.734375,
      "learning_rate": 0.00047870480907111657,
      "loss": 5.9413,
      "mean_token_accuracy": 0.1480317495763302,
      "num_tokens": 34100120.0,
      "step": 16835
    },
    {
      "entropy": 6.258674478530883,
      "epoch": 1.4716420519094644,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004786915743653461,
      "loss": 5.9557,
      "mean_token_accuracy": 0.14890800639986992,
      "num_tokens": 34110271.0,
      "step": 16840
    },
    {
      "entropy": 6.274297571182251,
      "epoch": 1.472079000262169,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047867833575266456,
      "loss": 6.0178,
      "mean_token_accuracy": 0.1457945540547371,
      "num_tokens": 34120240.0,
      "step": 16845
    },
    {
      "entropy": 6.283495140075684,
      "epoch": 1.4725159486148738,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004786650932333259,
      "loss": 5.9737,
      "mean_token_accuracy": 0.14640648812055587,
      "num_tokens": 34130813.0,
      "step": 16850
    },
    {
      "entropy": 6.273850679397583,
      "epoch": 1.4729528969675785,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004786518468075839,
      "loss": 5.9689,
      "mean_token_accuracy": 0.15442369654774665,
      "num_tokens": 34141169.0,
      "step": 16855
    },
    {
      "entropy": 6.2057372570037845,
      "epoch": 1.4733898453202832,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004786385964756928,
      "loss": 5.9696,
      "mean_token_accuracy": 0.14820237457752228,
      "num_tokens": 34151869.0,
      "step": 16860
    },
    {
      "entropy": 6.252609062194824,
      "epoch": 1.4738267936729879,
      "grad_norm": 2.75,
      "learning_rate": 0.0004786253422379068,
      "loss": 6.0501,
      "mean_token_accuracy": 0.14339821115136148,
      "num_tokens": 34162608.0,
      "step": 16865
    },
    {
      "entropy": 6.24493637084961,
      "epoch": 1.4742637420256925,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047861208409448,
      "loss": 5.946,
      "mean_token_accuracy": 0.14927000254392625,
      "num_tokens": 34172648.0,
      "step": 16870
    },
    {
      "entropy": 6.28076434135437,
      "epoch": 1.4747006903783972,
      "grad_norm": 3.25,
      "learning_rate": 0.0004785988220456666,
      "loss": 5.9769,
      "mean_token_accuracy": 0.14764530807733536,
      "num_tokens": 34183603.0,
      "step": 16875
    },
    {
      "entropy": 6.319261932373047,
      "epoch": 1.475137638731102,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004785855560917213,
      "loss": 5.9859,
      "mean_token_accuracy": 0.1437634766101837,
      "num_tokens": 34193830.0,
      "step": 16880
    },
    {
      "entropy": 6.2071678161621096,
      "epoch": 1.4755745870838066,
      "grad_norm": 3.703125,
      "learning_rate": 0.00047857228623289825,
      "loss": 5.8161,
      "mean_token_accuracy": 0.163958440721035,
      "num_tokens": 34203074.0,
      "step": 16885
    },
    {
      "entropy": 6.215431451797485,
      "epoch": 1.4760115354365113,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047855901246945205,
      "loss": 6.0321,
      "mean_token_accuracy": 0.14457381814718245,
      "num_tokens": 34212958.0,
      "step": 16890
    },
    {
      "entropy": 6.202045154571533,
      "epoch": 1.4764484837892162,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004785457348016374,
      "loss": 5.9811,
      "mean_token_accuracy": 0.14334235489368438,
      "num_tokens": 34223077.0,
      "step": 16895
    },
    {
      "entropy": 6.278396368026733,
      "epoch": 1.4768854321419207,
      "grad_norm": 6.84375,
      "learning_rate": 0.00047853245322970885,
      "loss": 5.9789,
      "mean_token_accuracy": 0.14898384660482406,
      "num_tokens": 34233386.0,
      "step": 16900
    },
    {
      "entropy": 6.3232142448425295,
      "epoch": 1.4773223804946256,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004785191677539212,
      "loss": 6.0059,
      "mean_token_accuracy": 0.1514424830675125,
      "num_tokens": 34244488.0,
      "step": 16905
    },
    {
      "entropy": 6.272506904602051,
      "epoch": 1.4777593288473303,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047850587837452936,
      "loss": 5.9561,
      "mean_token_accuracy": 0.14731967747211455,
      "num_tokens": 34254047.0,
      "step": 16910
    },
    {
      "entropy": 6.2387628078460695,
      "epoch": 1.478196277200035,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004784925850917881,
      "loss": 5.9481,
      "mean_token_accuracy": 0.14490032121539115,
      "num_tokens": 34264297.0,
      "step": 16915
    },
    {
      "entropy": 6.324604177474976,
      "epoch": 1.4786332255527397,
      "grad_norm": 3.25,
      "learning_rate": 0.0004784792879059524,
      "loss": 6.0254,
      "mean_token_accuracy": 0.14552840888500213,
      "num_tokens": 34274945.0,
      "step": 16920
    },
    {
      "entropy": 6.222376108169556,
      "epoch": 1.4790701739054444,
      "grad_norm": 2.5,
      "learning_rate": 0.00047846598681727733,
      "loss": 5.895,
      "mean_token_accuracy": 0.1482096143066883,
      "num_tokens": 34284914.0,
      "step": 16925
    },
    {
      "entropy": 6.1585673809051515,
      "epoch": 1.479507122258149,
      "grad_norm": 3.65625,
      "learning_rate": 0.000478452681826018,
      "loss": 5.952,
      "mean_token_accuracy": 0.15234848558902742,
      "num_tokens": 34294277.0,
      "step": 16930
    },
    {
      "entropy": 6.262396001815796,
      "epoch": 1.4799440706108538,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047843937293242967,
      "loss": 6.0354,
      "mean_token_accuracy": 0.14331667497754097,
      "num_tokens": 34304865.0,
      "step": 16935
    },
    {
      "entropy": 6.245778512954712,
      "epoch": 1.4803810189635584,
      "grad_norm": 3.21875,
      "learning_rate": 0.00047842606013676753,
      "loss": 5.8579,
      "mean_token_accuracy": 0.16440227776765823,
      "num_tokens": 34315404.0,
      "step": 16940
    },
    {
      "entropy": 6.326802349090576,
      "epoch": 1.4808179673162631,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047841274343928697,
      "loss": 5.9581,
      "mean_token_accuracy": 0.14401475340127945,
      "num_tokens": 34325448.0,
      "step": 16945
    },
    {
      "entropy": 6.149845218658447,
      "epoch": 1.481254915668968,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004783994228402434,
      "loss": 5.9473,
      "mean_token_accuracy": 0.15434338822960852,
      "num_tokens": 34335266.0,
      "step": 16950
    },
    {
      "entropy": 6.251708555221557,
      "epoch": 1.4816918640216725,
      "grad_norm": 2.875,
      "learning_rate": 0.0004783860983398924,
      "loss": 5.9336,
      "mean_token_accuracy": 0.15073499828577042,
      "num_tokens": 34345847.0,
      "step": 16955
    },
    {
      "entropy": 6.260920476913452,
      "epoch": 1.4821288123743774,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004783727699384894,
      "loss": 5.8987,
      "mean_token_accuracy": 0.1502774640917778,
      "num_tokens": 34355718.0,
      "step": 16960
    },
    {
      "entropy": 6.17747917175293,
      "epoch": 1.4825657607270821,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004783594376362901,
      "loss": 5.9719,
      "mean_token_accuracy": 0.15183843970298766,
      "num_tokens": 34366100.0,
      "step": 16965
    },
    {
      "entropy": 6.330912446975708,
      "epoch": 1.4830027090797868,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004783461014335502,
      "loss": 6.0468,
      "mean_token_accuracy": 0.1447932943701744,
      "num_tokens": 34376977.0,
      "step": 16970
    },
    {
      "entropy": 6.291689538955689,
      "epoch": 1.4834396574324915,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004783327613305256,
      "loss": 5.9837,
      "mean_token_accuracy": 0.14748658388853073,
      "num_tokens": 34387782.0,
      "step": 16975
    },
    {
      "entropy": 6.244571399688721,
      "epoch": 1.4838766057851962,
      "grad_norm": 2.09375,
      "learning_rate": 0.000478319417327472,
      "loss": 5.9121,
      "mean_token_accuracy": 0.15530938357114793,
      "num_tokens": 34397321.0,
      "step": 16980
    },
    {
      "entropy": 6.157162189483643,
      "epoch": 1.484313554137901,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004783060694246455,
      "loss": 5.9858,
      "mean_token_accuracy": 0.14787645637989044,
      "num_tokens": 34407594.0,
      "step": 16985
    },
    {
      "entropy": 6.297847175598145,
      "epoch": 1.4847505024906056,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004782927176223021,
      "loss": 5.9687,
      "mean_token_accuracy": 0.14593676775693892,
      "num_tokens": 34417337.0,
      "step": 16990
    },
    {
      "entropy": 6.149673557281494,
      "epoch": 1.4851874508433103,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004782793619206977,
      "loss": 5.8706,
      "mean_token_accuracy": 0.15154280066490172,
      "num_tokens": 34427643.0,
      "step": 16995
    },
    {
      "entropy": 6.2428642272949215,
      "epoch": 1.485624399196015,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004782660023200887,
      "loss": 5.9784,
      "mean_token_accuracy": 0.14189119935035704,
      "num_tokens": 34437224.0,
      "step": 17000
    },
    {
      "entropy": 6.319919586181641,
      "epoch": 1.4860613475487199,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004782526388207313,
      "loss": 5.9722,
      "mean_token_accuracy": 0.14042489230632782,
      "num_tokens": 34447376.0,
      "step": 17005
    },
    {
      "entropy": 6.185873126983642,
      "epoch": 1.4864982959014243,
      "grad_norm": 1.75,
      "learning_rate": 0.0004782392714228818,
      "loss": 6.0063,
      "mean_token_accuracy": 0.14482646584510803,
      "num_tokens": 34457649.0,
      "step": 17010
    },
    {
      "entropy": 6.290383338928223,
      "epoch": 1.4869352442541293,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047822590012679654,
      "loss": 5.9882,
      "mean_token_accuracy": 0.15194835215806962,
      "num_tokens": 34467437.0,
      "step": 17015
    },
    {
      "entropy": 6.273633861541748,
      "epoch": 1.487372192606834,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047821252493273204,
      "loss": 5.9098,
      "mean_token_accuracy": 0.14530453979969024,
      "num_tokens": 34477615.0,
      "step": 17020
    },
    {
      "entropy": 6.219990634918213,
      "epoch": 1.4878091409595386,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004781991458409448,
      "loss": 5.9387,
      "mean_token_accuracy": 0.14800671488046646,
      "num_tokens": 34487619.0,
      "step": 17025
    },
    {
      "entropy": 6.2041374206542965,
      "epoch": 1.4882460893122433,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047818576285169154,
      "loss": 5.9507,
      "mean_token_accuracy": 0.14818509593605994,
      "num_tokens": 34498837.0,
      "step": 17030
    },
    {
      "entropy": 6.193039226531982,
      "epoch": 1.488683037664948,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047817237596522885,
      "loss": 5.9528,
      "mean_token_accuracy": 0.14786773696541786,
      "num_tokens": 34508972.0,
      "step": 17035
    },
    {
      "entropy": 6.262591075897217,
      "epoch": 1.4891199860176527,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004781589851818135,
      "loss": 5.9382,
      "mean_token_accuracy": 0.15185491740703583,
      "num_tokens": 34519289.0,
      "step": 17040
    },
    {
      "entropy": 6.182735109329224,
      "epoch": 1.4895569343703574,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047814559050170236,
      "loss": 5.8515,
      "mean_token_accuracy": 0.15829512476921082,
      "num_tokens": 34528384.0,
      "step": 17045
    },
    {
      "entropy": 6.159409856796264,
      "epoch": 1.489993882723062,
      "grad_norm": 3.25,
      "learning_rate": 0.00047813219192515235,
      "loss": 5.9124,
      "mean_token_accuracy": 0.14457112848758696,
      "num_tokens": 34537442.0,
      "step": 17050
    },
    {
      "entropy": 6.245537233352661,
      "epoch": 1.4904308310757668,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004781187894524205,
      "loss": 5.9806,
      "mean_token_accuracy": 0.14736601784825326,
      "num_tokens": 34547960.0,
      "step": 17055
    },
    {
      "entropy": 6.2450079917907715,
      "epoch": 1.4908677794284715,
      "grad_norm": 3.375,
      "learning_rate": 0.00047810538308376376,
      "loss": 5.9061,
      "mean_token_accuracy": 0.14565444514155387,
      "num_tokens": 34556730.0,
      "step": 17060
    },
    {
      "entropy": 6.198627233505249,
      "epoch": 1.4913047277811762,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004780919728194394,
      "loss": 5.9529,
      "mean_token_accuracy": 0.14339002519845961,
      "num_tokens": 34566751.0,
      "step": 17065
    },
    {
      "entropy": 6.294309949874878,
      "epoch": 1.491741676133881,
      "grad_norm": 3.09375,
      "learning_rate": 0.00047807855865970454,
      "loss": 5.9455,
      "mean_token_accuracy": 0.14714618772268295,
      "num_tokens": 34576417.0,
      "step": 17070
    },
    {
      "entropy": 6.283307886123657,
      "epoch": 1.4921786244865856,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004780651406048166,
      "loss": 6.0361,
      "mean_token_accuracy": 0.14421438351273536,
      "num_tokens": 34586976.0,
      "step": 17075
    },
    {
      "entropy": 6.214851665496826,
      "epoch": 1.4926155728392905,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004780517186550328,
      "loss": 5.9282,
      "mean_token_accuracy": 0.15258136987686158,
      "num_tokens": 34597374.0,
      "step": 17080
    },
    {
      "entropy": 6.192280340194702,
      "epoch": 1.4930525211919952,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004780382928106107,
      "loss": 5.8963,
      "mean_token_accuracy": 0.1574055388569832,
      "num_tokens": 34607586.0,
      "step": 17085
    },
    {
      "entropy": 6.223878049850464,
      "epoch": 1.4934894695446999,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047802486307180767,
      "loss": 5.9742,
      "mean_token_accuracy": 0.15420927703380585,
      "num_tokens": 34617152.0,
      "step": 17090
    },
    {
      "entropy": 6.224582290649414,
      "epoch": 1.4939264178974045,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047801142943888135,
      "loss": 5.9058,
      "mean_token_accuracy": 0.15473876744508744,
      "num_tokens": 34627421.0,
      "step": 17095
    },
    {
      "entropy": 6.226395750045777,
      "epoch": 1.4943633662501092,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047799799191208955,
      "loss": 5.9105,
      "mean_token_accuracy": 0.15156465172767639,
      "num_tokens": 34637687.0,
      "step": 17100
    },
    {
      "entropy": 6.238168811798095,
      "epoch": 1.494800314602814,
      "grad_norm": 2.375,
      "learning_rate": 0.0004779845504916898,
      "loss": 5.985,
      "mean_token_accuracy": 0.14514171257615088,
      "num_tokens": 34647884.0,
      "step": 17105
    },
    {
      "entropy": 6.240828180313111,
      "epoch": 1.4952372629555186,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004779711051779401,
      "loss": 5.9654,
      "mean_token_accuracy": 0.14965278804302215,
      "num_tokens": 34658658.0,
      "step": 17110
    },
    {
      "entropy": 6.267143678665161,
      "epoch": 1.4956742113082233,
      "grad_norm": 4.625,
      "learning_rate": 0.0004779576559710982,
      "loss": 5.9766,
      "mean_token_accuracy": 0.15137157142162322,
      "num_tokens": 34668728.0,
      "step": 17115
    },
    {
      "entropy": 6.265066051483155,
      "epoch": 1.496111159660928,
      "grad_norm": 3.0625,
      "learning_rate": 0.00047794420287142203,
      "loss": 5.974,
      "mean_token_accuracy": 0.14480200856924058,
      "num_tokens": 34678576.0,
      "step": 17120
    },
    {
      "entropy": 6.3054274559021,
      "epoch": 1.496548108013633,
      "grad_norm": 2.0,
      "learning_rate": 0.00047793074587916984,
      "loss": 6.0231,
      "mean_token_accuracy": 0.14375167563557625,
      "num_tokens": 34688920.0,
      "step": 17125
    },
    {
      "entropy": 6.2138927459716795,
      "epoch": 1.4969850563663374,
      "grad_norm": 2.90625,
      "learning_rate": 0.00047791728499459955,
      "loss": 5.9666,
      "mean_token_accuracy": 0.14805330485105514,
      "num_tokens": 34699348.0,
      "step": 17130
    },
    {
      "entropy": 6.183752107620239,
      "epoch": 1.4974220047190423,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004779038202179694,
      "loss": 5.8572,
      "mean_token_accuracy": 0.15840872898697853,
      "num_tokens": 34708991.0,
      "step": 17135
    },
    {
      "entropy": 6.219084882736206,
      "epoch": 1.497858953071747,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047789035154953776,
      "loss": 5.9745,
      "mean_token_accuracy": 0.14171133562922478,
      "num_tokens": 34720869.0,
      "step": 17140
    },
    {
      "entropy": 6.19674334526062,
      "epoch": 1.4982959014244517,
      "grad_norm": 5.34375,
      "learning_rate": 0.0004778768789895627,
      "loss": 5.9038,
      "mean_token_accuracy": 0.15377528220415115,
      "num_tokens": 34731025.0,
      "step": 17145
    },
    {
      "entropy": 6.263921165466309,
      "epoch": 1.4987328497771564,
      "grad_norm": 3.5,
      "learning_rate": 0.0004778634025383029,
      "loss": 5.9726,
      "mean_token_accuracy": 0.14678846597671508,
      "num_tokens": 34741846.0,
      "step": 17150
    },
    {
      "entropy": 6.175172805786133,
      "epoch": 1.499169798129861,
      "grad_norm": 3.625,
      "learning_rate": 0.00047784992219601673,
      "loss": 5.8308,
      "mean_token_accuracy": 0.1679583564400673,
      "num_tokens": 34753294.0,
      "step": 17155
    },
    {
      "entropy": 6.274811315536499,
      "epoch": 1.4996067464825658,
      "grad_norm": 2.25,
      "learning_rate": 0.0004778364379629627,
      "loss": 5.9516,
      "mean_token_accuracy": 0.1464087523519993,
      "num_tokens": 34763997.0,
      "step": 17160
    },
    {
      "entropy": 6.195326805114746,
      "epoch": 1.5000436948352704,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047782294983939955,
      "loss": 5.8529,
      "mean_token_accuracy": 0.155233496427536,
      "num_tokens": 34772925.0,
      "step": 17165
    },
    {
      "entropy": 6.176755809783936,
      "epoch": 1.5004806431879751,
      "grad_norm": 2.125,
      "learning_rate": 0.0004778094578255859,
      "loss": 5.976,
      "mean_token_accuracy": 0.14671573787927628,
      "num_tokens": 34782608.0,
      "step": 17170
    },
    {
      "entropy": 6.214697027206421,
      "epoch": 1.5009175915406798,
      "grad_norm": 2.125,
      "learning_rate": 0.0004777959619217807,
      "loss": 5.9201,
      "mean_token_accuracy": 0.15435755401849746,
      "num_tokens": 34792533.0,
      "step": 17175
    },
    {
      "entropy": 6.280243158340454,
      "epoch": 1.5013545398933847,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047778246212824263,
      "loss": 6.0248,
      "mean_token_accuracy": 0.14021167904138565,
      "num_tokens": 34803069.0,
      "step": 17180
    },
    {
      "entropy": 6.220024108886719,
      "epoch": 1.5017914882460892,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004777689584452306,
      "loss": 5.9714,
      "mean_token_accuracy": 0.15224123001098633,
      "num_tokens": 34811925.0,
      "step": 17185
    },
    {
      "entropy": 6.236560964584351,
      "epoch": 1.5022284365987941,
      "grad_norm": 3.625,
      "learning_rate": 0.0004777554508730038,
      "loss": 5.8967,
      "mean_token_accuracy": 0.15015177577733993,
      "num_tokens": 34821511.0,
      "step": 17190
    },
    {
      "entropy": 6.258098602294922,
      "epoch": 1.5026653849514986,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004777419394118212,
      "loss": 6.0047,
      "mean_token_accuracy": 0.14640511125326156,
      "num_tokens": 34831683.0,
      "step": 17195
    },
    {
      "entropy": 6.248505687713623,
      "epoch": 1.5031023333042035,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004777284240619419,
      "loss": 5.9888,
      "mean_token_accuracy": 0.14582510143518448,
      "num_tokens": 34841578.0,
      "step": 17200
    },
    {
      "entropy": 6.255217123031616,
      "epoch": 1.5035392816569082,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004777149048236252,
      "loss": 5.9483,
      "mean_token_accuracy": 0.1492939591407776,
      "num_tokens": 34851959.0,
      "step": 17205
    },
    {
      "entropy": 6.2801214218139645,
      "epoch": 1.503976230009613,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047770138169713047,
      "loss": 5.9483,
      "mean_token_accuracy": 0.14534687325358392,
      "num_tokens": 34862557.0,
      "step": 17210
    },
    {
      "entropy": 6.248829793930054,
      "epoch": 1.5044131783623176,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047768785468271704,
      "loss": 5.9777,
      "mean_token_accuracy": 0.15007241666316987,
      "num_tokens": 34872394.0,
      "step": 17215
    },
    {
      "entropy": 6.23218674659729,
      "epoch": 1.5048501267150223,
      "grad_norm": 6.75,
      "learning_rate": 0.00047767432378064426,
      "loss": 5.9438,
      "mean_token_accuracy": 0.14820103943347931,
      "num_tokens": 34881938.0,
      "step": 17220
    },
    {
      "entropy": 6.324338769912719,
      "epoch": 1.505287075067727,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047766078899117174,
      "loss": 5.9327,
      "mean_token_accuracy": 0.1492886498570442,
      "num_tokens": 34892769.0,
      "step": 17225
    },
    {
      "entropy": 6.318613195419312,
      "epoch": 1.5057240234204317,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047764725031455927,
      "loss": 6.0558,
      "mean_token_accuracy": 0.13886793851852416,
      "num_tokens": 34903643.0,
      "step": 17230
    },
    {
      "entropy": 6.223487329483032,
      "epoch": 1.5061609717731366,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004776337077510662,
      "loss": 5.9416,
      "mean_token_accuracy": 0.14086148738861085,
      "num_tokens": 34913558.0,
      "step": 17235
    },
    {
      "entropy": 6.269551563262939,
      "epoch": 1.506597920125841,
      "grad_norm": 3.6875,
      "learning_rate": 0.00047762016130095246,
      "loss": 5.9215,
      "mean_token_accuracy": 0.14697461426258088,
      "num_tokens": 34923653.0,
      "step": 17240
    },
    {
      "entropy": 6.237241744995117,
      "epoch": 1.507034868478546,
      "grad_norm": 7.59375,
      "learning_rate": 0.00047760661096447784,
      "loss": 5.9669,
      "mean_token_accuracy": 0.15096956193447114,
      "num_tokens": 34933381.0,
      "step": 17245
    },
    {
      "entropy": 6.22724027633667,
      "epoch": 1.5074718168312504,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004775930567419023,
      "loss": 5.8665,
      "mean_token_accuracy": 0.15437265634536743,
      "num_tokens": 34942486.0,
      "step": 17250
    },
    {
      "entropy": 6.218029975891113,
      "epoch": 1.5079087651839553,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004775794986334857,
      "loss": 6.0137,
      "mean_token_accuracy": 0.1425306484103203,
      "num_tokens": 34952291.0,
      "step": 17255
    },
    {
      "entropy": 6.272081089019776,
      "epoch": 1.5083457135366598,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004775659366394883,
      "loss": 6.012,
      "mean_token_accuracy": 0.14193162992596625,
      "num_tokens": 34963842.0,
      "step": 17260
    },
    {
      "entropy": 6.221922922134399,
      "epoch": 1.5087826618893647,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047755237076016997,
      "loss": 5.909,
      "mean_token_accuracy": 0.15447845160961152,
      "num_tokens": 34973057.0,
      "step": 17265
    },
    {
      "entropy": 6.316521692276001,
      "epoch": 1.5092196102420694,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004775388009957911,
      "loss": 6.0656,
      "mean_token_accuracy": 0.14109252244234086,
      "num_tokens": 34983257.0,
      "step": 17270
    },
    {
      "entropy": 6.305471706390381,
      "epoch": 1.509656558594774,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047752522734661193,
      "loss": 5.9021,
      "mean_token_accuracy": 0.15028548687696458,
      "num_tokens": 34993066.0,
      "step": 17275
    },
    {
      "entropy": 6.2378569602966305,
      "epoch": 1.5100935069474788,
      "grad_norm": 4.0,
      "learning_rate": 0.00047751164981289264,
      "loss": 5.9079,
      "mean_token_accuracy": 0.14954061955213546,
      "num_tokens": 35002830.0,
      "step": 17280
    },
    {
      "entropy": 6.181393337249756,
      "epoch": 1.5105304553001835,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004774980683948939,
      "loss": 5.9312,
      "mean_token_accuracy": 0.14612747132778167,
      "num_tokens": 35011837.0,
      "step": 17285
    },
    {
      "entropy": 6.284710645675659,
      "epoch": 1.5109674036528882,
      "grad_norm": 3.234375,
      "learning_rate": 0.00047748448309287606,
      "loss": 5.9123,
      "mean_token_accuracy": 0.149068284034729,
      "num_tokens": 35021589.0,
      "step": 17290
    },
    {
      "entropy": 6.279403924942017,
      "epoch": 1.5114043520055929,
      "grad_norm": 3.75,
      "learning_rate": 0.00047747089390709973,
      "loss": 5.9843,
      "mean_token_accuracy": 0.1536121651530266,
      "num_tokens": 35032207.0,
      "step": 17295
    },
    {
      "entropy": 6.272064733505249,
      "epoch": 1.5118413003582978,
      "grad_norm": 2.375,
      "learning_rate": 0.0004774573008378256,
      "loss": 6.033,
      "mean_token_accuracy": 0.14229321628808975,
      "num_tokens": 35042661.0,
      "step": 17300
    },
    {
      "entropy": 6.254840850830078,
      "epoch": 1.5122782487110022,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004774437038853143,
      "loss": 6.009,
      "mean_token_accuracy": 0.14273399412631987,
      "num_tokens": 35053050.0,
      "step": 17305
    },
    {
      "entropy": 6.344922351837158,
      "epoch": 1.5127151970637072,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047743010304982674,
      "loss": 6.0014,
      "mean_token_accuracy": 0.1476335756480694,
      "num_tokens": 35064381.0,
      "step": 17310
    },
    {
      "entropy": 6.258383703231812,
      "epoch": 1.5131521454164116,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004774164983316237,
      "loss": 5.9025,
      "mean_token_accuracy": 0.15046497210860252,
      "num_tokens": 35074787.0,
      "step": 17315
    },
    {
      "entropy": 6.125835514068603,
      "epoch": 1.5135890937691165,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004774028897309662,
      "loss": 5.8688,
      "mean_token_accuracy": 0.1570577710866928,
      "num_tokens": 35085801.0,
      "step": 17320
    },
    {
      "entropy": 6.230676698684692,
      "epoch": 1.5140260421218212,
      "grad_norm": 2.5,
      "learning_rate": 0.00047738927724811514,
      "loss": 5.9673,
      "mean_token_accuracy": 0.15187445431947708,
      "num_tokens": 35096800.0,
      "step": 17325
    },
    {
      "entropy": 6.185333728790283,
      "epoch": 1.514462990474526,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004773756608833318,
      "loss": 5.7965,
      "mean_token_accuracy": 0.17201578468084336,
      "num_tokens": 35106735.0,
      "step": 17330
    },
    {
      "entropy": 6.159661483764649,
      "epoch": 1.5148999388272306,
      "grad_norm": 2.125,
      "learning_rate": 0.00047736204063687717,
      "loss": 5.9031,
      "mean_token_accuracy": 0.15116535723209382,
      "num_tokens": 35116221.0,
      "step": 17335
    },
    {
      "entropy": 6.229542207717896,
      "epoch": 1.5153368871799353,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047734841650901255,
      "loss": 5.8604,
      "mean_token_accuracy": 0.1552572578191757,
      "num_tokens": 35124896.0,
      "step": 17340
    },
    {
      "entropy": 6.220260286331177,
      "epoch": 1.51577383553264,
      "grad_norm": 2.75,
      "learning_rate": 0.0004773347884999993,
      "loss": 5.9509,
      "mean_token_accuracy": 0.14804214164614676,
      "num_tokens": 35134576.0,
      "step": 17345
    },
    {
      "entropy": 6.132141637802124,
      "epoch": 1.5162107838853447,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004773211566100988,
      "loss": 5.8621,
      "mean_token_accuracy": 0.16016653180122375,
      "num_tokens": 35144210.0,
      "step": 17350
    },
    {
      "entropy": 6.181662225723267,
      "epoch": 1.5166477322380496,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004773075208395725,
      "loss": 5.8446,
      "mean_token_accuracy": 0.15509474128484727,
      "num_tokens": 35153669.0,
      "step": 17355
    },
    {
      "entropy": 6.220783948898315,
      "epoch": 1.517084680590754,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004772938811886818,
      "loss": 5.9696,
      "mean_token_accuracy": 0.14793613627552987,
      "num_tokens": 35163469.0,
      "step": 17360
    },
    {
      "entropy": 6.268355178833008,
      "epoch": 1.517521628943459,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004772802376576887,
      "loss": 5.9884,
      "mean_token_accuracy": 0.1507078304886818,
      "num_tokens": 35173479.0,
      "step": 17365
    },
    {
      "entropy": 6.2484746932983395,
      "epoch": 1.5179585772961635,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004772665902468545,
      "loss": 5.9414,
      "mean_token_accuracy": 0.14933900237083436,
      "num_tokens": 35184570.0,
      "step": 17370
    },
    {
      "entropy": 6.2089080810546875,
      "epoch": 1.5183955256488684,
      "grad_norm": 2.5,
      "learning_rate": 0.0004772529389564412,
      "loss": 5.9632,
      "mean_token_accuracy": 0.14686267375946044,
      "num_tokens": 35193714.0,
      "step": 17375
    },
    {
      "entropy": 6.194644355773926,
      "epoch": 1.518832474001573,
      "grad_norm": 2.25,
      "learning_rate": 0.0004772392837867104,
      "loss": 5.965,
      "mean_token_accuracy": 0.149080727994442,
      "num_tokens": 35203806.0,
      "step": 17380
    },
    {
      "entropy": 6.332392692565918,
      "epoch": 1.5192694223542778,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047722562473792433,
      "loss": 5.9842,
      "mean_token_accuracy": 0.1451850175857544,
      "num_tokens": 35213677.0,
      "step": 17385
    },
    {
      "entropy": 6.248661661148072,
      "epoch": 1.5197063707069824,
      "grad_norm": 2.78125,
      "learning_rate": 0.00047721196181034465,
      "loss": 5.8491,
      "mean_token_accuracy": 0.1508323296904564,
      "num_tokens": 35223663.0,
      "step": 17390
    },
    {
      "entropy": 6.218564176559449,
      "epoch": 1.5201433190596871,
      "grad_norm": 3.0625,
      "learning_rate": 0.00047719829500423374,
      "loss": 5.9337,
      "mean_token_accuracy": 0.14723650366067886,
      "num_tokens": 35234551.0,
      "step": 17395
    },
    {
      "entropy": 6.220542001724243,
      "epoch": 1.5205802674123918,
      "grad_norm": 2.984375,
      "learning_rate": 0.00047718462431985354,
      "loss": 5.916,
      "mean_token_accuracy": 0.15115563124418258,
      "num_tokens": 35244350.0,
      "step": 17400
    },
    {
      "entropy": 6.218191146850586,
      "epoch": 1.5210172157650965,
      "grad_norm": 2.125,
      "learning_rate": 0.0004771709497574662,
      "loss": 5.8655,
      "mean_token_accuracy": 0.15681400001049042,
      "num_tokens": 35253970.0,
      "step": 17405
    },
    {
      "entropy": 6.200595808029175,
      "epoch": 1.5214541641178014,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047715727131733424,
      "loss": 5.8939,
      "mean_token_accuracy": 0.15645352154970169,
      "num_tokens": 35263316.0,
      "step": 17410
    },
    {
      "entropy": 6.167643928527832,
      "epoch": 1.521891112470506,
      "grad_norm": 3.375,
      "learning_rate": 0.00047714358899971984,
      "loss": 5.8828,
      "mean_token_accuracy": 0.1551271803677082,
      "num_tokens": 35273407.0,
      "step": 17415
    },
    {
      "entropy": 6.19045205116272,
      "epoch": 1.5223280608232108,
      "grad_norm": 2.625,
      "learning_rate": 0.0004771299028048855,
      "loss": 5.9621,
      "mean_token_accuracy": 0.14813019707798958,
      "num_tokens": 35284225.0,
      "step": 17420
    },
    {
      "entropy": 6.194149303436279,
      "epoch": 1.5227650091759153,
      "grad_norm": 2.8125,
      "learning_rate": 0.00047711621273309365,
      "loss": 5.849,
      "mean_token_accuracy": 0.1584537737071514,
      "num_tokens": 35294747.0,
      "step": 17425
    },
    {
      "entropy": 6.24236216545105,
      "epoch": 1.5232019575286202,
      "grad_norm": 4.09375,
      "learning_rate": 0.000477102518784607,
      "loss": 5.9346,
      "mean_token_accuracy": 0.1583493173122406,
      "num_tokens": 35304856.0,
      "step": 17430
    },
    {
      "entropy": 6.212068033218384,
      "epoch": 1.5236389058813247,
      "grad_norm": 4.4375,
      "learning_rate": 0.00047708882095968817,
      "loss": 5.9929,
      "mean_token_accuracy": 0.15340657085180281,
      "num_tokens": 35315550.0,
      "step": 17435
    },
    {
      "entropy": 6.283275747299195,
      "epoch": 1.5240758542340296,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004770751192585997,
      "loss": 6.0384,
      "mean_token_accuracy": 0.14197527319192887,
      "num_tokens": 35325970.0,
      "step": 17440
    },
    {
      "entropy": 6.29061107635498,
      "epoch": 1.5245128025867343,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004770614136816047,
      "loss": 5.9047,
      "mean_token_accuracy": 0.15022479593753815,
      "num_tokens": 35337277.0,
      "step": 17445
    },
    {
      "entropy": 6.239332103729248,
      "epoch": 1.524949750939439,
      "grad_norm": 2.890625,
      "learning_rate": 0.00047704770422896586,
      "loss": 5.9784,
      "mean_token_accuracy": 0.14786791950464248,
      "num_tokens": 35347686.0,
      "step": 17450
    },
    {
      "entropy": 6.258710527420044,
      "epoch": 1.5253866992921437,
      "grad_norm": 2.734375,
      "learning_rate": 0.00047703399090094623,
      "loss": 5.9254,
      "mean_token_accuracy": 0.15024724304676057,
      "num_tokens": 35358727.0,
      "step": 17455
    },
    {
      "entropy": 6.245534896850586,
      "epoch": 1.5258236476448483,
      "grad_norm": 6.53125,
      "learning_rate": 0.0004770202736978087,
      "loss": 5.9677,
      "mean_token_accuracy": 0.14978095665574073,
      "num_tokens": 35369741.0,
      "step": 17460
    },
    {
      "entropy": 6.2743419170379635,
      "epoch": 1.526260595997553,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004770065526198166,
      "loss": 6.0051,
      "mean_token_accuracy": 0.14772217124700546,
      "num_tokens": 35380213.0,
      "step": 17465
    },
    {
      "entropy": 6.239881181716919,
      "epoch": 1.5266975443502577,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004769928276672328,
      "loss": 5.9362,
      "mean_token_accuracy": 0.15523528829216957,
      "num_tokens": 35389937.0,
      "step": 17470
    },
    {
      "entropy": 6.215254831314087,
      "epoch": 1.5271344927029626,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004769790988403208,
      "loss": 5.9749,
      "mean_token_accuracy": 0.14452139884233475,
      "num_tokens": 35400534.0,
      "step": 17475
    },
    {
      "entropy": 6.213343381881714,
      "epoch": 1.5275714410556671,
      "grad_norm": 4.375,
      "learning_rate": 0.0004769653661393438,
      "loss": 5.9304,
      "mean_token_accuracy": 0.1539505884051323,
      "num_tokens": 35410607.0,
      "step": 17480
    },
    {
      "entropy": 6.2787621974945065,
      "epoch": 1.528008389408372,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004769516295645654,
      "loss": 5.9495,
      "mean_token_accuracy": 0.1477511152625084,
      "num_tokens": 35420242.0,
      "step": 17485
    },
    {
      "entropy": 6.235200881958008,
      "epoch": 1.5284453377610765,
      "grad_norm": 3.0625,
      "learning_rate": 0.00047693788911624876,
      "loss": 5.9011,
      "mean_token_accuracy": 0.15175773650407792,
      "num_tokens": 35429469.0,
      "step": 17490
    },
    {
      "entropy": 6.122993278503418,
      "epoch": 1.5288822861137814,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004769241447946577,
      "loss": 5.9416,
      "mean_token_accuracy": 0.15509707033634185,
      "num_tokens": 35439656.0,
      "step": 17495
    },
    {
      "entropy": 6.188720226287842,
      "epoch": 1.529319234466486,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004769103966000557,
      "loss": 5.8252,
      "mean_token_accuracy": 0.15157355815172197,
      "num_tokens": 35448600.0,
      "step": 17500
    },
    {
      "entropy": 6.176481771469116,
      "epoch": 1.5297561828191908,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004768966445327065,
      "loss": 5.9152,
      "mean_token_accuracy": 0.15760843306779862,
      "num_tokens": 35458124.0,
      "step": 17505
    },
    {
      "entropy": 6.234511137008667,
      "epoch": 1.5301931311718955,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004768828885928738,
      "loss": 5.8953,
      "mean_token_accuracy": 0.15073286294937133,
      "num_tokens": 35467616.0,
      "step": 17510
    },
    {
      "entropy": 6.188838958740234,
      "epoch": 1.5306300795246002,
      "grad_norm": 3.875,
      "learning_rate": 0.0004768691287808216,
      "loss": 5.9594,
      "mean_token_accuracy": 0.15026275292038918,
      "num_tokens": 35478356.0,
      "step": 17515
    },
    {
      "entropy": 6.212839651107788,
      "epoch": 1.5310670278773049,
      "grad_norm": 6.25,
      "learning_rate": 0.0004768553650968137,
      "loss": 5.9819,
      "mean_token_accuracy": 0.15059585720300675,
      "num_tokens": 35490213.0,
      "step": 17520
    },
    {
      "entropy": 6.194197082519532,
      "epoch": 1.5315039762300096,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004768415975411142,
      "loss": 5.8867,
      "mean_token_accuracy": 0.1565632089972496,
      "num_tokens": 35499982.0,
      "step": 17525
    },
    {
      "entropy": 6.26047191619873,
      "epoch": 1.5319409245827145,
      "grad_norm": 3.640625,
      "learning_rate": 0.00047682782611398693,
      "loss": 5.9656,
      "mean_token_accuracy": 0.14701954573392867,
      "num_tokens": 35510572.0,
      "step": 17530
    },
    {
      "entropy": 6.339392614364624,
      "epoch": 1.532377872935419,
      "grad_norm": 3.109375,
      "learning_rate": 0.00047681405081569637,
      "loss": 6.0251,
      "mean_token_accuracy": 0.143794996291399,
      "num_tokens": 35521560.0,
      "step": 17535
    },
    {
      "entropy": 6.299805212020874,
      "epoch": 1.5328148212881239,
      "grad_norm": 2.953125,
      "learning_rate": 0.00047680027164650644,
      "loss": 5.9984,
      "mean_token_accuracy": 0.14626324623823167,
      "num_tokens": 35530879.0,
      "step": 17540
    },
    {
      "entropy": 6.218653774261474,
      "epoch": 1.5332517696408283,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047678648860668163,
      "loss": 5.9852,
      "mean_token_accuracy": 0.15112681835889816,
      "num_tokens": 35541210.0,
      "step": 17545
    },
    {
      "entropy": 6.230462741851807,
      "epoch": 1.5336887179935332,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047677270169648616,
      "loss": 5.9487,
      "mean_token_accuracy": 0.15130465775728225,
      "num_tokens": 35550149.0,
      "step": 17550
    },
    {
      "entropy": 6.295032596588134,
      "epoch": 1.534125666346238,
      "grad_norm": 6.59375,
      "learning_rate": 0.00047675891091618464,
      "loss": 5.9157,
      "mean_token_accuracy": 0.15003982782363892,
      "num_tokens": 35561191.0,
      "step": 17555
    },
    {
      "entropy": 6.250424432754516,
      "epoch": 1.5345626146989426,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047674511626604144,
      "loss": 5.9195,
      "mean_token_accuracy": 0.15270240604877472,
      "num_tokens": 35571636.0,
      "step": 17560
    },
    {
      "entropy": 6.245758008956909,
      "epoch": 1.5349995630516473,
      "grad_norm": 5.0,
      "learning_rate": 0.0004767313177463212,
      "loss": 5.895,
      "mean_token_accuracy": 0.1456146001815796,
      "num_tokens": 35583191.0,
      "step": 17565
    },
    {
      "entropy": 6.158121013641358,
      "epoch": 1.535436511404352,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004767175153572885,
      "loss": 5.8845,
      "mean_token_accuracy": 0.15343640595674515,
      "num_tokens": 35592875.0,
      "step": 17570
    },
    {
      "entropy": 6.20794792175293,
      "epoch": 1.5358734597570567,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004767037090992083,
      "loss": 5.8757,
      "mean_token_accuracy": 0.15350353866815566,
      "num_tokens": 35602158.0,
      "step": 17575
    },
    {
      "entropy": 6.174899101257324,
      "epoch": 1.5363104081097614,
      "grad_norm": 6.1875,
      "learning_rate": 0.00047668989897234516,
      "loss": 5.8879,
      "mean_token_accuracy": 0.1542150244116783,
      "num_tokens": 35612077.0,
      "step": 17580
    },
    {
      "entropy": 6.176411867141724,
      "epoch": 1.5367473564624663,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047667608497696415,
      "loss": 5.8295,
      "mean_token_accuracy": 0.16061218082904816,
      "num_tokens": 35621856.0,
      "step": 17585
    },
    {
      "entropy": 6.192641258239746,
      "epoch": 1.5371843048151708,
      "grad_norm": 3.453125,
      "learning_rate": 0.00047666226711333014,
      "loss": 5.9673,
      "mean_token_accuracy": 0.15001493841409683,
      "num_tokens": 35633160.0,
      "step": 17590
    },
    {
      "entropy": 6.294309425354004,
      "epoch": 1.5376212531678757,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004766484453817081,
      "loss": 5.962,
      "mean_token_accuracy": 0.1477863609790802,
      "num_tokens": 35642477.0,
      "step": 17595
    },
    {
      "entropy": 6.246233940124512,
      "epoch": 1.5380582015205801,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004766346197823633,
      "loss": 5.885,
      "mean_token_accuracy": 0.14953289479017257,
      "num_tokens": 35651938.0,
      "step": 17600
    },
    {
      "entropy": 6.209073734283447,
      "epoch": 1.538495149873285,
      "grad_norm": 3.25,
      "learning_rate": 0.00047662079031556085,
      "loss": 5.938,
      "mean_token_accuracy": 0.1467727467417717,
      "num_tokens": 35662881.0,
      "step": 17605
    },
    {
      "entropy": 6.21721453666687,
      "epoch": 1.5389320982259895,
      "grad_norm": 2.25,
      "learning_rate": 0.0004766069569815659,
      "loss": 5.9052,
      "mean_token_accuracy": 0.14813161194324492,
      "num_tokens": 35673436.0,
      "step": 17610
    },
    {
      "entropy": 6.202668952941894,
      "epoch": 1.5393690465786944,
      "grad_norm": 6.375,
      "learning_rate": 0.000476593119780644,
      "loss": 5.8893,
      "mean_token_accuracy": 0.15059420764446257,
      "num_tokens": 35683411.0,
      "step": 17615
    },
    {
      "entropy": 6.175342655181884,
      "epoch": 1.5398059949313991,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047657927871306037,
      "loss": 5.8885,
      "mean_token_accuracy": 0.1543242372572422,
      "num_tokens": 35693806.0,
      "step": 17620
    },
    {
      "entropy": 6.250338077545166,
      "epoch": 1.5402429432841038,
      "grad_norm": 5.4375,
      "learning_rate": 0.0004765654337790805,
      "loss": 5.9268,
      "mean_token_accuracy": 0.15295714884996414,
      "num_tokens": 35705015.0,
      "step": 17625
    },
    {
      "entropy": 6.28689284324646,
      "epoch": 1.5406798916368085,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004765515849789701,
      "loss": 5.9944,
      "mean_token_accuracy": 0.15033574849367143,
      "num_tokens": 35715668.0,
      "step": 17630
    },
    {
      "entropy": 6.2792713165283205,
      "epoch": 1.5411168399895132,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047653773231299455,
      "loss": 5.9462,
      "mean_token_accuracy": 0.15053634122014045,
      "num_tokens": 35726381.0,
      "step": 17635
    },
    {
      "entropy": 6.203917646408081,
      "epoch": 1.541553788342218,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047652387578141973,
      "loss": 5.9593,
      "mean_token_accuracy": 0.15043184161186218,
      "num_tokens": 35736223.0,
      "step": 17640
    },
    {
      "entropy": 6.222280979156494,
      "epoch": 1.5419907366949226,
      "grad_norm": 3.3125,
      "learning_rate": 0.00047651001538451146,
      "loss": 5.9284,
      "mean_token_accuracy": 0.1501957282423973,
      "num_tokens": 35746959.0,
      "step": 17645
    },
    {
      "entropy": 6.229144668579101,
      "epoch": 1.5424276850476275,
      "grad_norm": 3.140625,
      "learning_rate": 0.00047649615112253547,
      "loss": 5.9138,
      "mean_token_accuracy": 0.1556854635477066,
      "num_tokens": 35756975.0,
      "step": 17650
    },
    {
      "entropy": 6.21838870048523,
      "epoch": 1.542864633400332,
      "grad_norm": 3.671875,
      "learning_rate": 0.00047648228299575766,
      "loss": 5.8966,
      "mean_token_accuracy": 0.15500865131616592,
      "num_tokens": 35765890.0,
      "step": 17655
    },
    {
      "entropy": 6.184146881103516,
      "epoch": 1.5433015817530369,
      "grad_norm": 3.578125,
      "learning_rate": 0.00047646841100444415,
      "loss": 5.8415,
      "mean_token_accuracy": 0.1615268275141716,
      "num_tokens": 35775630.0,
      "step": 17660
    },
    {
      "entropy": 6.18750033378601,
      "epoch": 1.5437385301057414,
      "grad_norm": 4.0,
      "learning_rate": 0.00047645453514886095,
      "loss": 5.9612,
      "mean_token_accuracy": 0.14893762320280074,
      "num_tokens": 35786053.0,
      "step": 17665
    },
    {
      "entropy": 6.2326250076293945,
      "epoch": 1.5441754784584463,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004764406554292742,
      "loss": 5.9599,
      "mean_token_accuracy": 0.14848692938685418,
      "num_tokens": 35795736.0,
      "step": 17670
    },
    {
      "entropy": 6.220575761795044,
      "epoch": 1.544612426811151,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004764267718459501,
      "loss": 5.878,
      "mean_token_accuracy": 0.1579812452197075,
      "num_tokens": 35805887.0,
      "step": 17675
    },
    {
      "entropy": 6.173495101928711,
      "epoch": 1.5450493751638557,
      "grad_norm": 4.15625,
      "learning_rate": 0.000476412884399155,
      "loss": 5.9451,
      "mean_token_accuracy": 0.15108732581138612,
      "num_tokens": 35815625.0,
      "step": 17680
    },
    {
      "entropy": 6.249575662612915,
      "epoch": 1.5454863235165603,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004763989930891552,
      "loss": 5.8975,
      "mean_token_accuracy": 0.14672638475894928,
      "num_tokens": 35825617.0,
      "step": 17685
    },
    {
      "entropy": 6.165812397003174,
      "epoch": 1.545923271869265,
      "grad_norm": 3.046875,
      "learning_rate": 0.00047638509791621714,
      "loss": 5.9216,
      "mean_token_accuracy": 0.15208155512809754,
      "num_tokens": 35835538.0,
      "step": 17690
    },
    {
      "entropy": 6.233427286148071,
      "epoch": 1.5463602202219697,
      "grad_norm": 3.296875,
      "learning_rate": 0.00047637119888060747,
      "loss": 5.8299,
      "mean_token_accuracy": 0.1588997021317482,
      "num_tokens": 35846302.0,
      "step": 17695
    },
    {
      "entropy": 6.267821550369263,
      "epoch": 1.5467971685746744,
      "grad_norm": 4.53125,
      "learning_rate": 0.00047635729598259266,
      "loss": 6.0099,
      "mean_token_accuracy": 0.14656817987561227,
      "num_tokens": 35857384.0,
      "step": 17700
    },
    {
      "entropy": 6.301066493988037,
      "epoch": 1.5472341169273793,
      "grad_norm": 6.875,
      "learning_rate": 0.0004763433892224394,
      "loss": 5.9376,
      "mean_token_accuracy": 0.13934424743056298,
      "num_tokens": 35867496.0,
      "step": 17705
    },
    {
      "entropy": 6.325211048126221,
      "epoch": 1.5476710652800838,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004763294786004144,
      "loss": 5.9577,
      "mean_token_accuracy": 0.15125152319669724,
      "num_tokens": 35878051.0,
      "step": 17710
    },
    {
      "entropy": 6.178198719024659,
      "epoch": 1.5481080136327887,
      "grad_norm": 6.03125,
      "learning_rate": 0.0004763155641167846,
      "loss": 5.8913,
      "mean_token_accuracy": 0.15541921705007553,
      "num_tokens": 35887926.0,
      "step": 17715
    },
    {
      "entropy": 6.18252911567688,
      "epoch": 1.5485449619854932,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047630164577181666,
      "loss": 5.9093,
      "mean_token_accuracy": 0.14828483685851096,
      "num_tokens": 35897982.0,
      "step": 17720
    },
    {
      "entropy": 6.195494222640991,
      "epoch": 1.548981910338198,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004762877235657777,
      "loss": 5.956,
      "mean_token_accuracy": 0.15171163007616997,
      "num_tokens": 35908805.0,
      "step": 17725
    },
    {
      "entropy": 6.243772697448731,
      "epoch": 1.5494188586909028,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004762737974989348,
      "loss": 5.9065,
      "mean_token_accuracy": 0.155516217648983,
      "num_tokens": 35918739.0,
      "step": 17730
    },
    {
      "entropy": 6.21890344619751,
      "epoch": 1.5498558070436075,
      "grad_norm": 3.828125,
      "learning_rate": 0.000476259867571555,
      "loss": 5.9704,
      "mean_token_accuracy": 0.1530129611492157,
      "num_tokens": 35928592.0,
      "step": 17735
    },
    {
      "entropy": 6.1204094886779785,
      "epoch": 1.5502927553963122,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004762459337839055,
      "loss": 5.9059,
      "mean_token_accuracy": 0.15198520869016646,
      "num_tokens": 35938287.0,
      "step": 17740
    },
    {
      "entropy": 6.286616945266724,
      "epoch": 1.5507297037490169,
      "grad_norm": 4.15625,
      "learning_rate": 0.00047623199613625343,
      "loss": 5.909,
      "mean_token_accuracy": 0.15354057997465134,
      "num_tokens": 35947890.0,
      "step": 17745
    },
    {
      "entropy": 6.2315479755401615,
      "epoch": 1.5511666521017216,
      "grad_norm": 2.875,
      "learning_rate": 0.0004762180546288663,
      "loss": 5.8976,
      "mean_token_accuracy": 0.15247896015644075,
      "num_tokens": 35956513.0,
      "step": 17750
    },
    {
      "entropy": 6.173036766052246,
      "epoch": 1.5516036004544262,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004762041092620114,
      "loss": 5.9267,
      "mean_token_accuracy": 0.15907410830259322,
      "num_tokens": 35966837.0,
      "step": 17755
    },
    {
      "entropy": 6.229518461227417,
      "epoch": 1.5520405488071312,
      "grad_norm": 3.484375,
      "learning_rate": 0.00047619016003595625,
      "loss": 6.0287,
      "mean_token_accuracy": 0.1436112068593502,
      "num_tokens": 35977076.0,
      "step": 17760
    },
    {
      "entropy": 6.351197957992554,
      "epoch": 1.5524774971598356,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047617620695096854,
      "loss": 5.9048,
      "mean_token_accuracy": 0.14976735562086105,
      "num_tokens": 35986862.0,
      "step": 17765
    },
    {
      "entropy": 6.244406604766846,
      "epoch": 1.5529144455125405,
      "grad_norm": 3.375,
      "learning_rate": 0.00047616225000731566,
      "loss": 5.9719,
      "mean_token_accuracy": 0.1438064031302929,
      "num_tokens": 35997488.0,
      "step": 17770
    },
    {
      "entropy": 6.210807657241821,
      "epoch": 1.553351393865245,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004761482892052655,
      "loss": 5.9351,
      "mean_token_accuracy": 0.1482605904340744,
      "num_tokens": 36008045.0,
      "step": 17775
    },
    {
      "entropy": 6.269315528869629,
      "epoch": 1.55378834221795,
      "grad_norm": 2.5,
      "learning_rate": 0.00047613432454508565,
      "loss": 5.9049,
      "mean_token_accuracy": 0.14770170971751212,
      "num_tokens": 36017864.0,
      "step": 17780
    },
    {
      "entropy": 6.255488729476928,
      "epoch": 1.5542252905706544,
      "grad_norm": 2.78125,
      "learning_rate": 0.00047612035602704407,
      "loss": 5.9449,
      "mean_token_accuracy": 0.14761437624692916,
      "num_tokens": 36027797.0,
      "step": 17785
    },
    {
      "entropy": 6.227194023132324,
      "epoch": 1.5546622389233593,
      "grad_norm": 4.375,
      "learning_rate": 0.00047610638365140867,
      "loss": 5.9427,
      "mean_token_accuracy": 0.14669045358896254,
      "num_tokens": 36037627.0,
      "step": 17790
    },
    {
      "entropy": 6.233815765380859,
      "epoch": 1.555099187276064,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047609240741844757,
      "loss": 5.8936,
      "mean_token_accuracy": 0.1456313759088516,
      "num_tokens": 36046324.0,
      "step": 17795
    },
    {
      "entropy": 6.208380317687988,
      "epoch": 1.5555361356287687,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004760784273284286,
      "loss": 5.8802,
      "mean_token_accuracy": 0.1562500014901161,
      "num_tokens": 36056212.0,
      "step": 17800
    },
    {
      "entropy": 6.24139666557312,
      "epoch": 1.5559730839814734,
      "grad_norm": 7.78125,
      "learning_rate": 0.0004760644433816201,
      "loss": 6.0328,
      "mean_token_accuracy": 0.14496950060129166,
      "num_tokens": 36066745.0,
      "step": 17805
    },
    {
      "entropy": 6.245906162261963,
      "epoch": 1.556410032334178,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004760504555782902,
      "loss": 5.9322,
      "mean_token_accuracy": 0.148170168697834,
      "num_tokens": 36077415.0,
      "step": 17810
    },
    {
      "entropy": 6.201148319244385,
      "epoch": 1.5568469806868828,
      "grad_norm": 5.0,
      "learning_rate": 0.00047603646391870715,
      "loss": 5.8933,
      "mean_token_accuracy": 0.15658055543899535,
      "num_tokens": 36088381.0,
      "step": 17815
    },
    {
      "entropy": 6.2176713943481445,
      "epoch": 1.5572839290395875,
      "grad_norm": 3.1875,
      "learning_rate": 0.00047602246840313945,
      "loss": 5.88,
      "mean_token_accuracy": 0.16125360429286956,
      "num_tokens": 36098333.0,
      "step": 17820
    },
    {
      "entropy": 6.242371225357056,
      "epoch": 1.5577208773922924,
      "grad_norm": 3.796875,
      "learning_rate": 0.00047600846903185544,
      "loss": 5.9425,
      "mean_token_accuracy": 0.14858102351427077,
      "num_tokens": 36107843.0,
      "step": 17825
    },
    {
      "entropy": 6.151204538345337,
      "epoch": 1.5581578257449968,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004759944658051237,
      "loss": 5.9281,
      "mean_token_accuracy": 0.15094674825668336,
      "num_tokens": 36116948.0,
      "step": 17830
    },
    {
      "entropy": 6.214093780517578,
      "epoch": 1.5585947740977018,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004759804587232127,
      "loss": 5.9305,
      "mean_token_accuracy": 0.15319917798042298,
      "num_tokens": 36126808.0,
      "step": 17835
    },
    {
      "entropy": 6.2757665634155275,
      "epoch": 1.5590317224504062,
      "grad_norm": 2.875,
      "learning_rate": 0.00047596644778639133,
      "loss": 6.0196,
      "mean_token_accuracy": 0.14472701177001,
      "num_tokens": 36136630.0,
      "step": 17840
    },
    {
      "entropy": 6.3242490768432615,
      "epoch": 1.5594686708031111,
      "grad_norm": 6.5,
      "learning_rate": 0.0004759524329949281,
      "loss": 5.9484,
      "mean_token_accuracy": 0.14195767939090728,
      "num_tokens": 36146422.0,
      "step": 17845
    },
    {
      "entropy": 6.237388324737549,
      "epoch": 1.5599056191558158,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004759384143490919,
      "loss": 5.9158,
      "mean_token_accuracy": 0.15070482045412065,
      "num_tokens": 36156169.0,
      "step": 17850
    },
    {
      "entropy": 6.19486403465271,
      "epoch": 1.5603425675085205,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004759243918491517,
      "loss": 5.9371,
      "mean_token_accuracy": 0.1564283102750778,
      "num_tokens": 36165173.0,
      "step": 17855
    },
    {
      "entropy": 6.295756006240845,
      "epoch": 1.5607795158612252,
      "grad_norm": 4.125,
      "learning_rate": 0.00047591036549537633,
      "loss": 5.9642,
      "mean_token_accuracy": 0.14893368631601334,
      "num_tokens": 36175852.0,
      "step": 17860
    },
    {
      "entropy": 6.247058343887329,
      "epoch": 1.56121646421393,
      "grad_norm": 5.125,
      "learning_rate": 0.0004758963352880348,
      "loss": 5.9209,
      "mean_token_accuracy": 0.1540413409471512,
      "num_tokens": 36185288.0,
      "step": 17865
    },
    {
      "entropy": 6.241706418991089,
      "epoch": 1.5616534125666346,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004758823012273964,
      "loss": 5.9414,
      "mean_token_accuracy": 0.14750997871160507,
      "num_tokens": 36196374.0,
      "step": 17870
    },
    {
      "entropy": 6.216095685958862,
      "epoch": 1.5620903609193393,
      "grad_norm": 2.5,
      "learning_rate": 0.00047586826331373025,
      "loss": 5.9794,
      "mean_token_accuracy": 0.1513463482260704,
      "num_tokens": 36206714.0,
      "step": 17875
    },
    {
      "entropy": 6.272429609298706,
      "epoch": 1.5625273092720442,
      "grad_norm": 3.5,
      "learning_rate": 0.0004758542215473054,
      "loss": 6.019,
      "mean_token_accuracy": 0.14360092505812644,
      "num_tokens": 36217891.0,
      "step": 17880
    },
    {
      "entropy": 6.264567184448242,
      "epoch": 1.5629642576247487,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004758401759283915,
      "loss": 5.8964,
      "mean_token_accuracy": 0.15006252974271775,
      "num_tokens": 36226776.0,
      "step": 17885
    },
    {
      "entropy": 6.2406933307647705,
      "epoch": 1.5634012059774536,
      "grad_norm": 5.5625,
      "learning_rate": 0.00047582612645725764,
      "loss": 5.9658,
      "mean_token_accuracy": 0.1482481986284256,
      "num_tokens": 36237362.0,
      "step": 17890
    },
    {
      "entropy": 6.181648778915405,
      "epoch": 1.563838154330158,
      "grad_norm": 3.640625,
      "learning_rate": 0.00047581207313417346,
      "loss": 5.8739,
      "mean_token_accuracy": 0.15206621289253236,
      "num_tokens": 36246264.0,
      "step": 17895
    },
    {
      "entropy": 6.307673549652099,
      "epoch": 1.564275102682863,
      "grad_norm": 3.859375,
      "learning_rate": 0.00047579801595940853,
      "loss": 5.9662,
      "mean_token_accuracy": 0.1449158102273941,
      "num_tokens": 36255676.0,
      "step": 17900
    },
    {
      "entropy": 6.224099826812744,
      "epoch": 1.5647120510355674,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047578395493323245,
      "loss": 5.8266,
      "mean_token_accuracy": 0.16165769696235657,
      "num_tokens": 36264346.0,
      "step": 17905
    },
    {
      "entropy": 6.16739501953125,
      "epoch": 1.5651489993882723,
      "grad_norm": 2.96875,
      "learning_rate": 0.00047576989005591495,
      "loss": 5.9177,
      "mean_token_accuracy": 0.1501021295785904,
      "num_tokens": 36273569.0,
      "step": 17910
    },
    {
      "entropy": 6.171949958801269,
      "epoch": 1.565585947740977,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004757558213277257,
      "loss": 5.8469,
      "mean_token_accuracy": 0.15618493258953095,
      "num_tokens": 36283708.0,
      "step": 17915
    },
    {
      "entropy": 6.254583883285522,
      "epoch": 1.5660228960936817,
      "grad_norm": 2.984375,
      "learning_rate": 0.00047574174874893456,
      "loss": 5.889,
      "mean_token_accuracy": 0.15610748827457427,
      "num_tokens": 36293184.0,
      "step": 17920
    },
    {
      "entropy": 6.22615008354187,
      "epoch": 1.5664598444463864,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004757276723198115,
      "loss": 5.9511,
      "mean_token_accuracy": 0.14798485934734346,
      "num_tokens": 36303071.0,
      "step": 17925
    },
    {
      "entropy": 6.218035411834717,
      "epoch": 1.566896792799091,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004757135920406266,
      "loss": 5.8738,
      "mean_token_accuracy": 0.15527887195348739,
      "num_tokens": 36313245.0,
      "step": 17930
    },
    {
      "entropy": 6.177819442749024,
      "epoch": 1.567333741151796,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004756995079116498,
      "loss": 5.8786,
      "mean_token_accuracy": 0.14913109838962554,
      "num_tokens": 36322711.0,
      "step": 17935
    },
    {
      "entropy": 6.167275667190552,
      "epoch": 1.5677706895045005,
      "grad_norm": 2.859375,
      "learning_rate": 0.00047568541993315124,
      "loss": 5.9194,
      "mean_token_accuracy": 0.14878253340721131,
      "num_tokens": 36333667.0,
      "step": 17940
    },
    {
      "entropy": 6.307104587554932,
      "epoch": 1.5682076378572054,
      "grad_norm": 4.21875,
      "learning_rate": 0.00047567132810540113,
      "loss": 5.9972,
      "mean_token_accuracy": 0.15099407583475113,
      "num_tokens": 36343660.0,
      "step": 17945
    },
    {
      "entropy": 6.212422800064087,
      "epoch": 1.5686445862099099,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047565723242866983,
      "loss": 5.9802,
      "mean_token_accuracy": 0.14920135587453842,
      "num_tokens": 36353581.0,
      "step": 17950
    },
    {
      "entropy": 6.232319927215576,
      "epoch": 1.5690815345626148,
      "grad_norm": 4.0625,
      "learning_rate": 0.00047564313290322774,
      "loss": 5.8864,
      "mean_token_accuracy": 0.14945686161518096,
      "num_tokens": 36363818.0,
      "step": 17955
    },
    {
      "entropy": 6.2147026538848875,
      "epoch": 1.5695184829153193,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004756290295293451,
      "loss": 5.9192,
      "mean_token_accuracy": 0.15518101900815964,
      "num_tokens": 36373590.0,
      "step": 17960
    },
    {
      "entropy": 6.2191849708557125,
      "epoch": 1.5699554312680242,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004756149223072926,
      "loss": 5.9703,
      "mean_token_accuracy": 0.1491720922291279,
      "num_tokens": 36383556.0,
      "step": 17965
    },
    {
      "entropy": 6.220646953582763,
      "epoch": 1.5703923796207289,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004756008112373408,
      "loss": 5.9014,
      "mean_token_accuracy": 0.14560514986515044,
      "num_tokens": 36393194.0,
      "step": 17970
    },
    {
      "entropy": 6.197233963012695,
      "epoch": 1.5708293279734336,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004755866963197603,
      "loss": 5.902,
      "mean_token_accuracy": 0.15024841725826263,
      "num_tokens": 36402962.0,
      "step": 17975
    },
    {
      "entropy": 6.210555219650269,
      "epoch": 1.5712662763261382,
      "grad_norm": 5.0,
      "learning_rate": 0.0004755725775548219,
      "loss": 5.8946,
      "mean_token_accuracy": 0.15133988708257676,
      "num_tokens": 36411932.0,
      "step": 17980
    },
    {
      "entropy": 6.249889612197876,
      "epoch": 1.571703224678843,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047555845494279633,
      "loss": 5.9126,
      "mean_token_accuracy": 0.15433190241456032,
      "num_tokens": 36421363.0,
      "step": 17985
    },
    {
      "entropy": 6.203109788894653,
      "epoch": 1.5721401730315476,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004755443284839545,
      "loss": 5.915,
      "mean_token_accuracy": 0.15464872270822524,
      "num_tokens": 36431422.0,
      "step": 17990
    },
    {
      "entropy": 6.202460193634034,
      "epoch": 1.5725771213842523,
      "grad_norm": 4.625,
      "learning_rate": 0.00047553019817856736,
      "loss": 5.9205,
      "mean_token_accuracy": 0.15240448266267775,
      "num_tokens": 36441004.0,
      "step": 17995
    },
    {
      "entropy": 6.298379230499267,
      "epoch": 1.5730140697369572,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047551606402690595,
      "loss": 6.0038,
      "mean_token_accuracy": 0.14423602148890496,
      "num_tokens": 36451474.0,
      "step": 18000
    },
    {
      "epoch": 1.5730140697369572,
      "eval_entropy": 6.042135502366131,
      "eval_loss": 5.976198673248291,
      "eval_mean_token_accuracy": 0.1565810376917128,
      "eval_num_tokens": 36451474.0,
      "eval_runtime": 27.8005,
      "eval_samples_per_second": 1331.417,
      "eval_steps_per_second": 166.436,
      "step": 18000
    },
    {
      "entropy": 6.21783766746521,
      "epoch": 1.5734510180896617,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004755019260292414,
      "loss": 5.9002,
      "mean_token_accuracy": 0.15177303105592727,
      "num_tokens": 36461341.0,
      "step": 18005
    },
    {
      "entropy": 6.19820384979248,
      "epoch": 1.5738879664423666,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004754877841858448,
      "loss": 5.9981,
      "mean_token_accuracy": 0.1488171100616455,
      "num_tokens": 36471679.0,
      "step": 18010
    },
    {
      "entropy": 6.269651985168457,
      "epoch": 1.574324914795071,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004754736384969875,
      "loss": 5.9459,
      "mean_token_accuracy": 0.14664804339408874,
      "num_tokens": 36481457.0,
      "step": 18015
    },
    {
      "entropy": 6.241263914108276,
      "epoch": 1.574761863147776,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004754594889629407,
      "loss": 5.8868,
      "mean_token_accuracy": 0.15130692943930626,
      "num_tokens": 36490596.0,
      "step": 18020
    },
    {
      "entropy": 6.261288022994995,
      "epoch": 1.5751988115004807,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004754453355839759,
      "loss": 5.9871,
      "mean_token_accuracy": 0.1465463861823082,
      "num_tokens": 36501027.0,
      "step": 18025
    },
    {
      "entropy": 6.216840410232544,
      "epoch": 1.5756357598531854,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047543117836036446,
      "loss": 5.8891,
      "mean_token_accuracy": 0.15028577595949172,
      "num_tokens": 36512089.0,
      "step": 18030
    },
    {
      "entropy": 6.1761674880981445,
      "epoch": 1.57607270820589,
      "grad_norm": 2.78125,
      "learning_rate": 0.00047541701729237815,
      "loss": 5.903,
      "mean_token_accuracy": 0.1536923184990883,
      "num_tokens": 36522521.0,
      "step": 18035
    },
    {
      "entropy": 6.2883178234100345,
      "epoch": 1.5765096565585948,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004754028523802883,
      "loss": 5.9842,
      "mean_token_accuracy": 0.14350027963519096,
      "num_tokens": 36533111.0,
      "step": 18040
    },
    {
      "entropy": 6.189882850646972,
      "epoch": 1.5769466049112995,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004753886836243668,
      "loss": 5.9042,
      "mean_token_accuracy": 0.1594758525490761,
      "num_tokens": 36542477.0,
      "step": 18045
    },
    {
      "entropy": 6.2232921600341795,
      "epoch": 1.5773835532640041,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047537451102488526,
      "loss": 5.9432,
      "mean_token_accuracy": 0.15247599482536317,
      "num_tokens": 36552137.0,
      "step": 18050
    },
    {
      "entropy": 6.20857343673706,
      "epoch": 1.577820501616709,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004753603345821157,
      "loss": 5.9497,
      "mean_token_accuracy": 0.1479859843850136,
      "num_tokens": 36562440.0,
      "step": 18055
    },
    {
      "entropy": 6.190438175201416,
      "epoch": 1.5782574499694135,
      "grad_norm": 3.734375,
      "learning_rate": 0.00047534615429632986,
      "loss": 5.8931,
      "mean_token_accuracy": 0.1514140784740448,
      "num_tokens": 36571863.0,
      "step": 18060
    },
    {
      "entropy": 6.240919351577759,
      "epoch": 1.5786943983221184,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004753319701677999,
      "loss": 5.8978,
      "mean_token_accuracy": 0.14914495646953582,
      "num_tokens": 36581409.0,
      "step": 18065
    },
    {
      "entropy": 6.272856760025024,
      "epoch": 1.579131346674823,
      "grad_norm": 4.78125,
      "learning_rate": 0.00047531778219679773,
      "loss": 5.9303,
      "mean_token_accuracy": 0.14515915811061858,
      "num_tokens": 36592779.0,
      "step": 18070
    },
    {
      "entropy": 6.2016143798828125,
      "epoch": 1.5795682950275278,
      "grad_norm": 4.625,
      "learning_rate": 0.00047530359038359547,
      "loss": 5.9008,
      "mean_token_accuracy": 0.1595821976661682,
      "num_tokens": 36601173.0,
      "step": 18075
    },
    {
      "entropy": 6.237958335876465,
      "epoch": 1.5800052433802323,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004752893947284654,
      "loss": 5.9503,
      "mean_token_accuracy": 0.1566769778728485,
      "num_tokens": 36610147.0,
      "step": 18080
    },
    {
      "entropy": 6.297000360488892,
      "epoch": 1.5804421917329372,
      "grad_norm": 2.875,
      "learning_rate": 0.0004752751952316799,
      "loss": 5.9276,
      "mean_token_accuracy": 0.15173804610967637,
      "num_tokens": 36620093.0,
      "step": 18085
    },
    {
      "entropy": 6.303887557983399,
      "epoch": 1.580879140085642,
      "grad_norm": 2.890625,
      "learning_rate": 0.00047526099189351115,
      "loss": 5.963,
      "mean_token_accuracy": 0.149624502658844,
      "num_tokens": 36631833.0,
      "step": 18090
    },
    {
      "entropy": 6.2172788143157955,
      "epoch": 1.5813160884383466,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004752467847142316,
      "loss": 5.8926,
      "mean_token_accuracy": 0.1482769101858139,
      "num_tokens": 36641123.0,
      "step": 18095
    },
    {
      "entropy": 6.2535796642303465,
      "epoch": 1.5817530367910513,
      "grad_norm": 3.203125,
      "learning_rate": 0.00047523257369411383,
      "loss": 5.9827,
      "mean_token_accuracy": 0.14856084287166596,
      "num_tokens": 36652310.0,
      "step": 18100
    },
    {
      "entropy": 6.294920110702515,
      "epoch": 1.582189985143756,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004752183588334303,
      "loss": 5.9967,
      "mean_token_accuracy": 0.14423601999878882,
      "num_tokens": 36662783.0,
      "step": 18105
    },
    {
      "entropy": 6.227855443954468,
      "epoch": 1.5826269334964607,
      "grad_norm": 3.28125,
      "learning_rate": 0.00047520414013245383,
      "loss": 5.8958,
      "mean_token_accuracy": 0.15466513484716415,
      "num_tokens": 36672933.0,
      "step": 18110
    },
    {
      "entropy": 6.275967597961426,
      "epoch": 1.5830638818491654,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004751899175914571,
      "loss": 5.99,
      "mean_token_accuracy": 0.14720086008310318,
      "num_tokens": 36682570.0,
      "step": 18115
    },
    {
      "entropy": 6.249949312210083,
      "epoch": 1.5835008302018703,
      "grad_norm": 5.375,
      "learning_rate": 0.0004751756912107127,
      "loss": 5.8849,
      "mean_token_accuracy": 0.1564955398440361,
      "num_tokens": 36693008.0,
      "step": 18120
    },
    {
      "entropy": 6.172162103652954,
      "epoch": 1.5839377785545747,
      "grad_norm": 5.125,
      "learning_rate": 0.00047516146099049376,
      "loss": 5.8092,
      "mean_token_accuracy": 0.15729403197765351,
      "num_tokens": 36702611.0,
      "step": 18125
    },
    {
      "entropy": 6.215594816207886,
      "epoch": 1.5843747269072797,
      "grad_norm": 3.953125,
      "learning_rate": 0.00047514722693107304,
      "loss": 5.9598,
      "mean_token_accuracy": 0.15555752366781234,
      "num_tokens": 36712470.0,
      "step": 18130
    },
    {
      "entropy": 6.23865122795105,
      "epoch": 1.5848116752599841,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004751329890327237,
      "loss": 5.9791,
      "mean_token_accuracy": 0.1464655317366123,
      "num_tokens": 36722578.0,
      "step": 18135
    },
    {
      "entropy": 6.290515422821045,
      "epoch": 1.585248623612689,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004751187472957187,
      "loss": 5.9417,
      "mean_token_accuracy": 0.1517208844423294,
      "num_tokens": 36732334.0,
      "step": 18140
    },
    {
      "entropy": 6.229752063751221,
      "epoch": 1.5856855719653937,
      "grad_norm": 2.75,
      "learning_rate": 0.0004751045017203313,
      "loss": 5.975,
      "mean_token_accuracy": 0.14879060685634612,
      "num_tokens": 36742328.0,
      "step": 18145
    },
    {
      "entropy": 6.216316890716553,
      "epoch": 1.5861225203180984,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004750902523068347,
      "loss": 5.988,
      "mean_token_accuracy": 0.1511794865131378,
      "num_tokens": 36754363.0,
      "step": 18150
    },
    {
      "entropy": 6.329362535476685,
      "epoch": 1.586559468670803,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004750759990555022,
      "loss": 5.9956,
      "mean_token_accuracy": 0.14169976115226746,
      "num_tokens": 36765588.0,
      "step": 18155
    },
    {
      "entropy": 6.172846508026123,
      "epoch": 1.5869964170235078,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004750617419666072,
      "loss": 5.8873,
      "mean_token_accuracy": 0.1488058939576149,
      "num_tokens": 36774851.0,
      "step": 18160
    },
    {
      "entropy": 6.194058132171631,
      "epoch": 1.5874333653762125,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047504748104042324,
      "loss": 5.9646,
      "mean_token_accuracy": 0.14732776135206221,
      "num_tokens": 36785836.0,
      "step": 18165
    },
    {
      "entropy": 6.243482065200806,
      "epoch": 1.5878703137289172,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047503321627722375,
      "loss": 5.8922,
      "mean_token_accuracy": 0.1582065239548683,
      "num_tokens": 36795367.0,
      "step": 18170
    },
    {
      "entropy": 6.26365294456482,
      "epoch": 1.588307262081622,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047501894767728225,
      "loss": 5.8609,
      "mean_token_accuracy": 0.15100716874003411,
      "num_tokens": 36804294.0,
      "step": 18175
    },
    {
      "entropy": 6.212438011169434,
      "epoch": 1.5887442104343266,
      "grad_norm": 3.453125,
      "learning_rate": 0.00047500467524087266,
      "loss": 5.9333,
      "mean_token_accuracy": 0.14897346645593643,
      "num_tokens": 36813711.0,
      "step": 18180
    },
    {
      "entropy": 6.186293601989746,
      "epoch": 1.5891811587870315,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004749903989682685,
      "loss": 5.9353,
      "mean_token_accuracy": 0.15037843734025955,
      "num_tokens": 36824498.0,
      "step": 18185
    },
    {
      "entropy": 6.243250513076783,
      "epoch": 1.589618107139736,
      "grad_norm": 3.34375,
      "learning_rate": 0.00047497611885974366,
      "loss": 5.8646,
      "mean_token_accuracy": 0.1559558629989624,
      "num_tokens": 36833329.0,
      "step": 18190
    },
    {
      "entropy": 6.210446166992187,
      "epoch": 1.5900550554924409,
      "grad_norm": 3.21875,
      "learning_rate": 0.00047496183491557215,
      "loss": 5.9446,
      "mean_token_accuracy": 0.14847447276115416,
      "num_tokens": 36842783.0,
      "step": 18195
    },
    {
      "entropy": 6.2099462985992435,
      "epoch": 1.5904920038451456,
      "grad_norm": 3.34375,
      "learning_rate": 0.00047494754713602787,
      "loss": 5.9369,
      "mean_token_accuracy": 0.15034055933356286,
      "num_tokens": 36852785.0,
      "step": 18200
    },
    {
      "entropy": 6.269167232513428,
      "epoch": 1.5909289521978502,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047493325552138484,
      "loss": 5.9395,
      "mean_token_accuracy": 0.14771086871623992,
      "num_tokens": 36862337.0,
      "step": 18205
    },
    {
      "entropy": 6.237546777725219,
      "epoch": 1.591365900550555,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004749189600719172,
      "loss": 5.9671,
      "mean_token_accuracy": 0.15017909407615662,
      "num_tokens": 36872026.0,
      "step": 18210
    },
    {
      "entropy": 6.205109262466431,
      "epoch": 1.5918028489032596,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004749046607878992,
      "loss": 5.9589,
      "mean_token_accuracy": 0.1508518174290657,
      "num_tokens": 36882520.0,
      "step": 18215
    },
    {
      "entropy": 6.193803405761718,
      "epoch": 1.5922397972559643,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047489035766960504,
      "loss": 5.845,
      "mean_token_accuracy": 0.16192845851182938,
      "num_tokens": 36892630.0,
      "step": 18220
    },
    {
      "entropy": 6.2536238670349125,
      "epoch": 1.592676745608669,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047487605071730904,
      "loss": 5.9485,
      "mean_token_accuracy": 0.14706256240606308,
      "num_tokens": 36902692.0,
      "step": 18225
    },
    {
      "entropy": 6.270265054702759,
      "epoch": 1.593113693961374,
      "grad_norm": 3.703125,
      "learning_rate": 0.00047486173993128566,
      "loss": 6.0644,
      "mean_token_accuracy": 0.14336446821689605,
      "num_tokens": 36913581.0,
      "step": 18230
    },
    {
      "entropy": 6.268398237228394,
      "epoch": 1.5935506423140784,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004748474253118094,
      "loss": 5.8976,
      "mean_token_accuracy": 0.15564172863960266,
      "num_tokens": 36924031.0,
      "step": 18235
    },
    {
      "entropy": 6.254928302764893,
      "epoch": 1.5939875906667833,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004748331068591548,
      "loss": 5.9326,
      "mean_token_accuracy": 0.14835700392723083,
      "num_tokens": 36934622.0,
      "step": 18240
    },
    {
      "entropy": 6.243341588973999,
      "epoch": 1.5944245390194878,
      "grad_norm": 2.875,
      "learning_rate": 0.0004748187845735965,
      "loss": 5.9412,
      "mean_token_accuracy": 0.15215089321136474,
      "num_tokens": 36944374.0,
      "step": 18245
    },
    {
      "entropy": 6.241748523712158,
      "epoch": 1.5948614873721927,
      "grad_norm": 3.046875,
      "learning_rate": 0.00047480445845540927,
      "loss": 5.9063,
      "mean_token_accuracy": 0.15598199516534805,
      "num_tokens": 36954371.0,
      "step": 18250
    },
    {
      "entropy": 6.2301403999328615,
      "epoch": 1.5952984357248972,
      "grad_norm": 2.75,
      "learning_rate": 0.00047479012850486777,
      "loss": 5.9104,
      "mean_token_accuracy": 0.15469518750905992,
      "num_tokens": 36964217.0,
      "step": 18255
    },
    {
      "entropy": 6.31103401184082,
      "epoch": 1.595735384077602,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047477579472224687,
      "loss": 5.9891,
      "mean_token_accuracy": 0.14359126910567283,
      "num_tokens": 36974580.0,
      "step": 18260
    },
    {
      "entropy": 6.148231887817383,
      "epoch": 1.5961723324303068,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004747614571078216,
      "loss": 5.7571,
      "mean_token_accuracy": 0.16748557835817338,
      "num_tokens": 36983434.0,
      "step": 18265
    },
    {
      "entropy": 6.214754629135132,
      "epoch": 1.5966092807830115,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004747471156618669,
      "loss": 5.9914,
      "mean_token_accuracy": 0.14866308718919755,
      "num_tokens": 36993813.0,
      "step": 18270
    },
    {
      "entropy": 6.261786699295044,
      "epoch": 1.5970462291357161,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004747327703846579,
      "loss": 5.9411,
      "mean_token_accuracy": 0.15035979300737382,
      "num_tokens": 37003855.0,
      "step": 18275
    },
    {
      "entropy": 6.195237684249878,
      "epoch": 1.5974831774884208,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047471842127646966,
      "loss": 5.8835,
      "mean_token_accuracy": 0.15649521797895433,
      "num_tokens": 37013660.0,
      "step": 18280
    },
    {
      "entropy": 6.151340007781982,
      "epoch": 1.5979201258411255,
      "grad_norm": 5.6875,
      "learning_rate": 0.00047470406833757746,
      "loss": 5.8488,
      "mean_token_accuracy": 0.1508024200797081,
      "num_tokens": 37023888.0,
      "step": 18285
    },
    {
      "entropy": 6.235114240646363,
      "epoch": 1.5983570741938302,
      "grad_norm": 3.75,
      "learning_rate": 0.00047468971156825656,
      "loss": 6.001,
      "mean_token_accuracy": 0.138613910228014,
      "num_tokens": 37034260.0,
      "step": 18290
    },
    {
      "entropy": 6.232266044616699,
      "epoch": 1.5987940225465351,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004746753509687824,
      "loss": 5.9657,
      "mean_token_accuracy": 0.1497529551386833,
      "num_tokens": 37045300.0,
      "step": 18295
    },
    {
      "entropy": 6.257679510116577,
      "epoch": 1.5992309708992396,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004746609865394303,
      "loss": 5.8877,
      "mean_token_accuracy": 0.15394216999411584,
      "num_tokens": 37054898.0,
      "step": 18300
    },
    {
      "entropy": 6.110081815719605,
      "epoch": 1.5996679192519445,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004746466182804759,
      "loss": 5.8494,
      "mean_token_accuracy": 0.15276305228471757,
      "num_tokens": 37064519.0,
      "step": 18305
    },
    {
      "entropy": 6.160031080245972,
      "epoch": 1.600104867604649,
      "grad_norm": 4.125,
      "learning_rate": 0.0004746322461921948,
      "loss": 5.9285,
      "mean_token_accuracy": 0.15049879178404807,
      "num_tokens": 37074967.0,
      "step": 18310
    },
    {
      "entropy": 6.244515705108642,
      "epoch": 1.600541815957354,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004746178702748626,
      "loss": 5.9338,
      "mean_token_accuracy": 0.15154035538434982,
      "num_tokens": 37084736.0,
      "step": 18315
    },
    {
      "entropy": 6.282709693908691,
      "epoch": 1.6009787643100586,
      "grad_norm": 2.90625,
      "learning_rate": 0.000474603490528755,
      "loss": 5.8919,
      "mean_token_accuracy": 0.15445174723863603,
      "num_tokens": 37094327.0,
      "step": 18320
    },
    {
      "entropy": 6.2026252269744875,
      "epoch": 1.6014157126627633,
      "grad_norm": 3.375,
      "learning_rate": 0.0004745891069541479,
      "loss": 5.8962,
      "mean_token_accuracy": 0.15504606813192368,
      "num_tokens": 37104099.0,
      "step": 18325
    },
    {
      "entropy": 6.199679660797119,
      "epoch": 1.601852661015468,
      "grad_norm": 3.109375,
      "learning_rate": 0.00047457471955131714,
      "loss": 5.9893,
      "mean_token_accuracy": 0.14886707067489624,
      "num_tokens": 37115337.0,
      "step": 18330
    },
    {
      "entropy": 6.290006303787232,
      "epoch": 1.6022896093681727,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004745603283205387,
      "loss": 5.9019,
      "mean_token_accuracy": 0.15265172719955444,
      "num_tokens": 37125522.0,
      "step": 18335
    },
    {
      "entropy": 6.285940361022949,
      "epoch": 1.6027265577208774,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004745459332620885,
      "loss": 5.9998,
      "mean_token_accuracy": 0.14351874142885207,
      "num_tokens": 37135847.0,
      "step": 18340
    },
    {
      "entropy": 6.1861279010772705,
      "epoch": 1.603163506073582,
      "grad_norm": 2.25,
      "learning_rate": 0.0004745315343762428,
      "loss": 5.8742,
      "mean_token_accuracy": 0.15652220398187638,
      "num_tokens": 37145714.0,
      "step": 18345
    },
    {
      "entropy": 6.204250001907349,
      "epoch": 1.603600454426287,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047451713166327776,
      "loss": 5.9056,
      "mean_token_accuracy": 0.15491210520267487,
      "num_tokens": 37154921.0,
      "step": 18350
    },
    {
      "entropy": 6.255270910263062,
      "epoch": 1.6040374027789914,
      "grad_norm": 2.421875,
      "learning_rate": 0.00047450272512346953,
      "loss": 5.9001,
      "mean_token_accuracy": 0.1575327679514885,
      "num_tokens": 37165303.0,
      "step": 18355
    },
    {
      "entropy": 6.192402076721192,
      "epoch": 1.6044743511316963,
      "grad_norm": 2.34375,
      "learning_rate": 0.00047448831475709454,
      "loss": 5.8563,
      "mean_token_accuracy": 0.15700275897979737,
      "num_tokens": 37175014.0,
      "step": 18360
    },
    {
      "entropy": 6.157375049591065,
      "epoch": 1.6049112994844008,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004744739005644291,
      "loss": 5.8991,
      "mean_token_accuracy": 0.15646206885576247,
      "num_tokens": 37185830.0,
      "step": 18365
    },
    {
      "entropy": 6.250677061080933,
      "epoch": 1.6053482478371057,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004744594825457498,
      "loss": 5.8333,
      "mean_token_accuracy": 0.15919142067432404,
      "num_tokens": 37196264.0,
      "step": 18370
    },
    {
      "entropy": 6.265071964263916,
      "epoch": 1.6057851961898104,
      "grad_norm": 2.578125,
      "learning_rate": 0.00047444506070133305,
      "loss": 5.96,
      "mean_token_accuracy": 0.14792527556419371,
      "num_tokens": 37206338.0,
      "step": 18375
    },
    {
      "entropy": 6.211864423751831,
      "epoch": 1.606222144542515,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004744306350314555,
      "loss": 5.9099,
      "mean_token_accuracy": 0.154256484657526,
      "num_tokens": 37216231.0,
      "step": 18380
    },
    {
      "entropy": 6.343891334533692,
      "epoch": 1.6066590928952198,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004744162055363939,
      "loss": 6.0281,
      "mean_token_accuracy": 0.14524662643671035,
      "num_tokens": 37227209.0,
      "step": 18385
    },
    {
      "entropy": 6.320376014709472,
      "epoch": 1.6070960412479245,
      "grad_norm": 3.296875,
      "learning_rate": 0.000474401772216425,
      "loss": 6.0285,
      "mean_token_accuracy": 0.15066935271024703,
      "num_tokens": 37238056.0,
      "step": 18390
    },
    {
      "entropy": 6.21752438545227,
      "epoch": 1.6075329896006292,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004743873350718257,
      "loss": 5.9057,
      "mean_token_accuracy": 0.1557021416723728,
      "num_tokens": 37247864.0,
      "step": 18395
    },
    {
      "entropy": 6.187521886825562,
      "epoch": 1.6079699379533339,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004743728941028727,
      "loss": 5.8467,
      "mean_token_accuracy": 0.15635615587234497,
      "num_tokens": 37256418.0,
      "step": 18400
    },
    {
      "entropy": 6.252369117736817,
      "epoch": 1.6084068863060388,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004743584493098431,
      "loss": 5.9736,
      "mean_token_accuracy": 0.142327731102705,
      "num_tokens": 37266540.0,
      "step": 18405
    },
    {
      "entropy": 6.289048290252685,
      "epoch": 1.6088438346587433,
      "grad_norm": 2.421875,
      "learning_rate": 0.00047434400069301414,
      "loss": 5.9407,
      "mean_token_accuracy": 0.1469181016087532,
      "num_tokens": 37277011.0,
      "step": 18410
    },
    {
      "entropy": 6.247420883178711,
      "epoch": 1.6092807830114482,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004743295482526627,
      "loss": 5.9279,
      "mean_token_accuracy": 0.15198372974991797,
      "num_tokens": 37287203.0,
      "step": 18415
    },
    {
      "entropy": 6.1986054420471195,
      "epoch": 1.6097177313641526,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004743150919890661,
      "loss": 5.8852,
      "mean_token_accuracy": 0.15781996846199037,
      "num_tokens": 37297293.0,
      "step": 18420
    },
    {
      "entropy": 6.212550497055053,
      "epoch": 1.6101546797168576,
      "grad_norm": 2.25,
      "learning_rate": 0.00047430063190250155,
      "loss": 5.9877,
      "mean_token_accuracy": 0.14676765277981757,
      "num_tokens": 37307882.0,
      "step": 18425
    },
    {
      "entropy": 6.179150724411011,
      "epoch": 1.610591628069562,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004742861679932464,
      "loss": 5.8549,
      "mean_token_accuracy": 0.15690417289733888,
      "num_tokens": 37317466.0,
      "step": 18430
    },
    {
      "entropy": 6.215849876403809,
      "epoch": 1.611028576422267,
      "grad_norm": 2.578125,
      "learning_rate": 0.00047427170026157823,
      "loss": 5.9655,
      "mean_token_accuracy": 0.14813189953565598,
      "num_tokens": 37327544.0,
      "step": 18435
    },
    {
      "entropy": 6.27412052154541,
      "epoch": 1.6114655247749716,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004742572287077744,
      "loss": 5.9464,
      "mean_token_accuracy": 0.150690658390522,
      "num_tokens": 37338082.0,
      "step": 18440
    },
    {
      "entropy": 6.270237112045288,
      "epoch": 1.6119024731276763,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047424275333211256,
      "loss": 5.9131,
      "mean_token_accuracy": 0.1540774703025818,
      "num_tokens": 37348127.0,
      "step": 18445
    },
    {
      "entropy": 6.272872257232666,
      "epoch": 1.612339421480381,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004742282741348702,
      "loss": 5.9958,
      "mean_token_accuracy": 0.1445587046444416,
      "num_tokens": 37359547.0,
      "step": 18450
    },
    {
      "entropy": 6.280548238754273,
      "epoch": 1.6127763698330857,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004742137911163253,
      "loss": 6.0188,
      "mean_token_accuracy": 0.15324459075927735,
      "num_tokens": 37369547.0,
      "step": 18455
    },
    {
      "entropy": 6.235742378234863,
      "epoch": 1.6132133181857904,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004741993042767554,
      "loss": 6.0105,
      "mean_token_accuracy": 0.15405722707509995,
      "num_tokens": 37380055.0,
      "step": 18460
    },
    {
      "entropy": 6.312399625778198,
      "epoch": 1.613650266538495,
      "grad_norm": 2.6875,
      "learning_rate": 0.00047418481361643844,
      "loss": 6.0083,
      "mean_token_accuracy": 0.14640052020549774,
      "num_tokens": 37390179.0,
      "step": 18465
    },
    {
      "entropy": 6.25750322341919,
      "epoch": 1.6140872148912,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047417031913565247,
      "loss": 5.9677,
      "mean_token_accuracy": 0.14947554618120193,
      "num_tokens": 37400663.0,
      "step": 18470
    },
    {
      "entropy": 6.2742884159088135,
      "epoch": 1.6145241632439045,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004741558208346753,
      "loss": 5.9481,
      "mean_token_accuracy": 0.15702445954084396,
      "num_tokens": 37410988.0,
      "step": 18475
    },
    {
      "entropy": 6.335418462753296,
      "epoch": 1.6149611115966094,
      "grad_norm": 3.09375,
      "learning_rate": 0.00047414131871378526,
      "loss": 6.0166,
      "mean_token_accuracy": 0.14472706541419028,
      "num_tokens": 37422092.0,
      "step": 18480
    },
    {
      "entropy": 6.2400980472564695,
      "epoch": 1.6153980599493138,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004741268127732603,
      "loss": 5.9335,
      "mean_token_accuracy": 0.1514393962919712,
      "num_tokens": 37431637.0,
      "step": 18485
    },
    {
      "entropy": 6.249762153625488,
      "epoch": 1.6158350083020188,
      "grad_norm": 3.375,
      "learning_rate": 0.00047411230301337877,
      "loss": 5.989,
      "mean_token_accuracy": 0.13977956473827363,
      "num_tokens": 37442678.0,
      "step": 18490
    },
    {
      "entropy": 6.24508957862854,
      "epoch": 1.6162719566547235,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004740977894344189,
      "loss": 5.9453,
      "mean_token_accuracy": 0.15044984892010688,
      "num_tokens": 37452324.0,
      "step": 18495
    },
    {
      "entropy": 6.250298118591308,
      "epoch": 1.6167089050074281,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004740832720366591,
      "loss": 5.9076,
      "mean_token_accuracy": 0.14865177720785142,
      "num_tokens": 37462940.0,
      "step": 18500
    },
    {
      "entropy": 6.247258186340332,
      "epoch": 1.6171458533601328,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004740687508203779,
      "loss": 5.9859,
      "mean_token_accuracy": 0.15113524198532105,
      "num_tokens": 37474121.0,
      "step": 18505
    },
    {
      "entropy": 6.198953580856323,
      "epoch": 1.6175828017128375,
      "grad_norm": 2.625,
      "learning_rate": 0.0004740542257858536,
      "loss": 5.95,
      "mean_token_accuracy": 0.15114662572741508,
      "num_tokens": 37483439.0,
      "step": 18510
    },
    {
      "entropy": 6.290823602676392,
      "epoch": 1.6180197500655422,
      "grad_norm": 3.625,
      "learning_rate": 0.00047403969693336507,
      "loss": 5.9722,
      "mean_token_accuracy": 0.15026990622282027,
      "num_tokens": 37493804.0,
      "step": 18515
    },
    {
      "entropy": 6.253290700912475,
      "epoch": 1.618456698418247,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004740251642631909,
      "loss": 5.8508,
      "mean_token_accuracy": 0.15106042772531508,
      "num_tokens": 37504158.0,
      "step": 18520
    },
    {
      "entropy": 6.1457220077514645,
      "epoch": 1.6188936467709518,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004740106277756097,
      "loss": 5.9223,
      "mean_token_accuracy": 0.1444391742348671,
      "num_tokens": 37515069.0,
      "step": 18525
    },
    {
      "entropy": 6.117239236831665,
      "epoch": 1.6193305951236563,
      "grad_norm": 3.3125,
      "learning_rate": 0.00047399608747090033,
      "loss": 5.8145,
      "mean_token_accuracy": 0.1537598118185997,
      "num_tokens": 37525486.0,
      "step": 18530
    },
    {
      "entropy": 6.30638165473938,
      "epoch": 1.6197675434763612,
      "grad_norm": 2.6875,
      "learning_rate": 0.00047398154334934187,
      "loss": 5.9645,
      "mean_token_accuracy": 0.1483389414846897,
      "num_tokens": 37535616.0,
      "step": 18535
    },
    {
      "entropy": 6.222192811965942,
      "epoch": 1.6202044918290657,
      "grad_norm": 2.984375,
      "learning_rate": 0.00047396699541121296,
      "loss": 5.9021,
      "mean_token_accuracy": 0.1531231388449669,
      "num_tokens": 37546084.0,
      "step": 18540
    },
    {
      "entropy": 6.159199810028076,
      "epoch": 1.6206414401817706,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004739524436567929,
      "loss": 5.9298,
      "mean_token_accuracy": 0.1584751546382904,
      "num_tokens": 37555865.0,
      "step": 18545
    },
    {
      "entropy": 6.130236530303955,
      "epoch": 1.6210783885344753,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004739378880863608,
      "loss": 5.9234,
      "mean_token_accuracy": 0.15189338624477386,
      "num_tokens": 37565985.0,
      "step": 18550
    },
    {
      "entropy": 6.295703172683716,
      "epoch": 1.62151533688718,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004739233287001957,
      "loss": 6.0075,
      "mean_token_accuracy": 0.14808081239461898,
      "num_tokens": 37575887.0,
      "step": 18555
    },
    {
      "entropy": 6.286135864257813,
      "epoch": 1.6219522852398847,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047390876549857695,
      "loss": 5.8467,
      "mean_token_accuracy": 0.15901709496974945,
      "num_tokens": 37585346.0,
      "step": 18560
    },
    {
      "entropy": 6.185558414459228,
      "epoch": 1.6223892335925894,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047389419848178384,
      "loss": 5.9342,
      "mean_token_accuracy": 0.14581326320767402,
      "num_tokens": 37595674.0,
      "step": 18565
    },
    {
      "entropy": 6.180164432525634,
      "epoch": 1.622826181945294,
      "grad_norm": 3.0,
      "learning_rate": 0.00047387962765009576,
      "loss": 5.911,
      "mean_token_accuracy": 0.1451365754008293,
      "num_tokens": 37605537.0,
      "step": 18570
    },
    {
      "entropy": 6.271160507202149,
      "epoch": 1.6232631302979987,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047386505300379224,
      "loss": 5.9645,
      "mean_token_accuracy": 0.14797806590795518,
      "num_tokens": 37615801.0,
      "step": 18575
    },
    {
      "entropy": 6.251804256439209,
      "epoch": 1.6237000786507036,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004738504745431528,
      "loss": 5.8505,
      "mean_token_accuracy": 0.15825356245040895,
      "num_tokens": 37625582.0,
      "step": 18580
    },
    {
      "entropy": 6.160108613967895,
      "epoch": 1.6241370270034081,
      "grad_norm": 2.53125,
      "learning_rate": 0.00047383589226845707,
      "loss": 5.9838,
      "mean_token_accuracy": 0.14603446722030639,
      "num_tokens": 37636075.0,
      "step": 18585
    },
    {
      "entropy": 6.108518171310425,
      "epoch": 1.624573975356113,
      "grad_norm": 3.375,
      "learning_rate": 0.00047382130617998477,
      "loss": 5.8246,
      "mean_token_accuracy": 0.1643337368965149,
      "num_tokens": 37645692.0,
      "step": 18590
    },
    {
      "entropy": 6.21349720954895,
      "epoch": 1.6250109237088175,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004738067162780156,
      "loss": 5.9082,
      "mean_token_accuracy": 0.15017776265740396,
      "num_tokens": 37655035.0,
      "step": 18595
    },
    {
      "entropy": 6.248477506637573,
      "epoch": 1.6254478720615224,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004737921225628294,
      "loss": 5.9058,
      "mean_token_accuracy": 0.14826771914958953,
      "num_tokens": 37665829.0,
      "step": 18600
    },
    {
      "entropy": 6.19279203414917,
      "epoch": 1.6258848204142269,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047377752503470613,
      "loss": 5.8943,
      "mean_token_accuracy": 0.16164066269993782,
      "num_tokens": 37676318.0,
      "step": 18605
    },
    {
      "entropy": 6.185751676559448,
      "epoch": 1.6263217687669318,
      "grad_norm": 2.859375,
      "learning_rate": 0.00047376292369392584,
      "loss": 5.9421,
      "mean_token_accuracy": 0.1491665408015251,
      "num_tokens": 37686977.0,
      "step": 18610
    },
    {
      "entropy": 6.230441617965698,
      "epoch": 1.6267587171196365,
      "grad_norm": 2.578125,
      "learning_rate": 0.00047374831854076843,
      "loss": 5.9024,
      "mean_token_accuracy": 0.1542597621679306,
      "num_tokens": 37697004.0,
      "step": 18615
    },
    {
      "entropy": 6.1912487030029295,
      "epoch": 1.6271956654723412,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047373370957551423,
      "loss": 5.9104,
      "mean_token_accuracy": 0.1541975036263466,
      "num_tokens": 37707613.0,
      "step": 18620
    },
    {
      "entropy": 6.206807947158813,
      "epoch": 1.6276326138250459,
      "grad_norm": 5.125,
      "learning_rate": 0.0004737190967984433,
      "loss": 5.8964,
      "mean_token_accuracy": 0.1490690991282463,
      "num_tokens": 37717167.0,
      "step": 18625
    },
    {
      "entropy": 6.230147409439087,
      "epoch": 1.6280695621777506,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004737044802098359,
      "loss": 5.9988,
      "mean_token_accuracy": 0.1447828784584999,
      "num_tokens": 37727496.0,
      "step": 18630
    },
    {
      "entropy": 6.230030679702759,
      "epoch": 1.6285065105304553,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004736898598099725,
      "loss": 5.9076,
      "mean_token_accuracy": 0.1508959114551544,
      "num_tokens": 37736844.0,
      "step": 18635
    },
    {
      "entropy": 6.2379639625549315,
      "epoch": 1.62894345888316,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047367523559913347,
      "loss": 5.937,
      "mean_token_accuracy": 0.15028771236538888,
      "num_tokens": 37746027.0,
      "step": 18640
    },
    {
      "entropy": 6.252640008926392,
      "epoch": 1.6293804072358649,
      "grad_norm": 2.5,
      "learning_rate": 0.00047366060757759933,
      "loss": 5.9079,
      "mean_token_accuracy": 0.1535932183265686,
      "num_tokens": 37756499.0,
      "step": 18645
    },
    {
      "entropy": 6.175581789016723,
      "epoch": 1.6298173555885693,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047364597574565064,
      "loss": 5.8836,
      "mean_token_accuracy": 0.1547905221581459,
      "num_tokens": 37766533.0,
      "step": 18650
    },
    {
      "entropy": 6.155520868301392,
      "epoch": 1.6302543039412742,
      "grad_norm": 2.578125,
      "learning_rate": 0.000473631340103568,
      "loss": 5.9105,
      "mean_token_accuracy": 0.15490448027849196,
      "num_tokens": 37775937.0,
      "step": 18655
    },
    {
      "entropy": 6.20485520362854,
      "epoch": 1.6306912522939787,
      "grad_norm": 2.625,
      "learning_rate": 0.00047361670065163225,
      "loss": 5.9103,
      "mean_token_accuracy": 0.1586038127541542,
      "num_tokens": 37786587.0,
      "step": 18660
    },
    {
      "entropy": 6.271780776977539,
      "epoch": 1.6311282006466836,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047360205739012406,
      "loss": 5.9299,
      "mean_token_accuracy": 0.15248845294117927,
      "num_tokens": 37797200.0,
      "step": 18665
    },
    {
      "entropy": 6.127733564376831,
      "epoch": 1.6315651489993883,
      "grad_norm": 4.0,
      "learning_rate": 0.0004735874103193244,
      "loss": 5.9025,
      "mean_token_accuracy": 0.15707122832536696,
      "num_tokens": 37806652.0,
      "step": 18670
    },
    {
      "entropy": 6.196038389205933,
      "epoch": 1.632002097352093,
      "grad_norm": 3.625,
      "learning_rate": 0.0004735727594395141,
      "loss": 5.968,
      "mean_token_accuracy": 0.14650043472647667,
      "num_tokens": 37817057.0,
      "step": 18675
    },
    {
      "entropy": 6.282775592803955,
      "epoch": 1.6324390457047977,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047355810475097423,
      "loss": 5.9076,
      "mean_token_accuracy": 0.14455318376421927,
      "num_tokens": 37826216.0,
      "step": 18680
    },
    {
      "entropy": 6.258368825912475,
      "epoch": 1.6328759940575024,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047354344625398596,
      "loss": 5.9692,
      "mean_token_accuracy": 0.14794138073921204,
      "num_tokens": 37835834.0,
      "step": 18685
    },
    {
      "entropy": 6.144515800476074,
      "epoch": 1.633312942410207,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004735287839488302,
      "loss": 5.8832,
      "mean_token_accuracy": 0.15540692955255508,
      "num_tokens": 37846252.0,
      "step": 18690
    },
    {
      "entropy": 6.171580457687378,
      "epoch": 1.6337498907629118,
      "grad_norm": 3.21875,
      "learning_rate": 0.00047351411783578846,
      "loss": 5.8946,
      "mean_token_accuracy": 0.1523268163204193,
      "num_tokens": 37856020.0,
      "step": 18695
    },
    {
      "entropy": 6.2836205005645756,
      "epoch": 1.6341868391156167,
      "grad_norm": 3.125,
      "learning_rate": 0.0004734994479151418,
      "loss": 5.9303,
      "mean_token_accuracy": 0.14986779391765595,
      "num_tokens": 37866132.0,
      "step": 18700
    },
    {
      "entropy": 6.211834049224853,
      "epoch": 1.6346237874683212,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004734847741871718,
      "loss": 5.9763,
      "mean_token_accuracy": 0.14670439213514327,
      "num_tokens": 37876779.0,
      "step": 18705
    },
    {
      "entropy": 6.223350524902344,
      "epoch": 1.635060735821026,
      "grad_norm": 3.765625,
      "learning_rate": 0.00047347009665215983,
      "loss": 5.9187,
      "mean_token_accuracy": 0.1540893495082855,
      "num_tokens": 37886981.0,
      "step": 18710
    },
    {
      "entropy": 6.157265710830688,
      "epoch": 1.6354976841737305,
      "grad_norm": 3.0,
      "learning_rate": 0.0004734554153103874,
      "loss": 5.8772,
      "mean_token_accuracy": 0.1621463492512703,
      "num_tokens": 37896974.0,
      "step": 18715
    },
    {
      "entropy": 6.268452739715576,
      "epoch": 1.6359346325264355,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047344073016213615,
      "loss": 6.0352,
      "mean_token_accuracy": 0.14203107878565788,
      "num_tokens": 37909020.0,
      "step": 18720
    },
    {
      "entropy": 6.314188671112061,
      "epoch": 1.63637158087914,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047342604120768764,
      "loss": 5.882,
      "mean_token_accuracy": 0.15404159873723983,
      "num_tokens": 37919417.0,
      "step": 18725
    },
    {
      "entropy": 6.259980964660644,
      "epoch": 1.6368085292318448,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004734113484473237,
      "loss": 5.9973,
      "mean_token_accuracy": 0.14409162402153014,
      "num_tokens": 37929030.0,
      "step": 18730
    },
    {
      "entropy": 6.235384941101074,
      "epoch": 1.6372454775845495,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004733966518813262,
      "loss": 5.9024,
      "mean_token_accuracy": 0.15571803748607635,
      "num_tokens": 37939157.0,
      "step": 18735
    },
    {
      "entropy": 6.242619609832763,
      "epoch": 1.6376824259372542,
      "grad_norm": 3.09375,
      "learning_rate": 0.00047338195150997686,
      "loss": 5.9092,
      "mean_token_accuracy": 0.14661599397659303,
      "num_tokens": 37950544.0,
      "step": 18740
    },
    {
      "entropy": 6.127318429946899,
      "epoch": 1.638119374289959,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047336724733355786,
      "loss": 5.8665,
      "mean_token_accuracy": 0.16251022070646287,
      "num_tokens": 37960059.0,
      "step": 18745
    },
    {
      "entropy": 6.213432168960571,
      "epoch": 1.6385563226426636,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047335253935235094,
      "loss": 5.8808,
      "mean_token_accuracy": 0.15522278100252151,
      "num_tokens": 37970669.0,
      "step": 18750
    },
    {
      "entropy": 6.200917530059814,
      "epoch": 1.6389932709953685,
      "grad_norm": 2.34375,
      "learning_rate": 0.00047333782756663853,
      "loss": 5.9267,
      "mean_token_accuracy": 0.15461772531270981,
      "num_tokens": 37980088.0,
      "step": 18755
    },
    {
      "entropy": 6.214099931716919,
      "epoch": 1.639430219348073,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047332311197670264,
      "loss": 5.9514,
      "mean_token_accuracy": 0.15639277249574662,
      "num_tokens": 37989433.0,
      "step": 18760
    },
    {
      "entropy": 6.274089097976685,
      "epoch": 1.639867167700778,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047330839258282544,
      "loss": 5.934,
      "mean_token_accuracy": 0.15188869461417198,
      "num_tokens": 37999735.0,
      "step": 18765
    },
    {
      "entropy": 6.242506694793701,
      "epoch": 1.6403041160534824,
      "grad_norm": 2.5,
      "learning_rate": 0.0004732936693852895,
      "loss": 5.8021,
      "mean_token_accuracy": 0.1634713053703308,
      "num_tokens": 38008515.0,
      "step": 18770
    },
    {
      "entropy": 6.193327808380127,
      "epoch": 1.6407410644061873,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047327894238437696,
      "loss": 5.9538,
      "mean_token_accuracy": 0.1491445094347,
      "num_tokens": 38018778.0,
      "step": 18775
    },
    {
      "entropy": 6.232254362106323,
      "epoch": 1.6411780127588917,
      "grad_norm": 2.65625,
      "learning_rate": 0.00047326421158037045,
      "loss": 5.9632,
      "mean_token_accuracy": 0.15069210529327393,
      "num_tokens": 38029062.0,
      "step": 18780
    },
    {
      "entropy": 6.2531843185424805,
      "epoch": 1.6416149611115967,
      "grad_norm": 3.015625,
      "learning_rate": 0.00047324947697355253,
      "loss": 5.8513,
      "mean_token_accuracy": 0.1561152920126915,
      "num_tokens": 38039945.0,
      "step": 18785
    },
    {
      "entropy": 6.249436140060425,
      "epoch": 1.6420519094643014,
      "grad_norm": 2.921875,
      "learning_rate": 0.00047323473856420573,
      "loss": 5.933,
      "mean_token_accuracy": 0.14410339593887328,
      "num_tokens": 38050213.0,
      "step": 18790
    },
    {
      "entropy": 6.171185159683228,
      "epoch": 1.642488857817006,
      "grad_norm": 4.9375,
      "learning_rate": 0.00047321999635261276,
      "loss": 5.8655,
      "mean_token_accuracy": 0.1509905532002449,
      "num_tokens": 38060147.0,
      "step": 18795
    },
    {
      "entropy": 6.245804405212402,
      "epoch": 1.6429258061697107,
      "grad_norm": 4.25,
      "learning_rate": 0.0004732052503390563,
      "loss": 5.939,
      "mean_token_accuracy": 0.15094834417104722,
      "num_tokens": 38070854.0,
      "step": 18800
    },
    {
      "entropy": 6.234937858581543,
      "epoch": 1.6433627545224154,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004731905005238194,
      "loss": 5.9783,
      "mean_token_accuracy": 0.14647122323513032,
      "num_tokens": 38080998.0,
      "step": 18805
    },
    {
      "entropy": 6.239195966720581,
      "epoch": 1.6437997028751201,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047317574690718486,
      "loss": 5.9258,
      "mean_token_accuracy": 0.14921616911888122,
      "num_tokens": 38091421.0,
      "step": 18810
    },
    {
      "entropy": 6.250820541381836,
      "epoch": 1.6442366512278248,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047316098948943557,
      "loss": 5.8871,
      "mean_token_accuracy": 0.1554766356945038,
      "num_tokens": 38101106.0,
      "step": 18815
    },
    {
      "entropy": 6.219632577896118,
      "epoch": 1.6446735995805297,
      "grad_norm": 3.109375,
      "learning_rate": 0.00047314622827085466,
      "loss": 5.917,
      "mean_token_accuracy": 0.14630459547042846,
      "num_tokens": 38111346.0,
      "step": 18820
    },
    {
      "entropy": 6.297738027572632,
      "epoch": 1.6451105479332342,
      "grad_norm": 3.375,
      "learning_rate": 0.00047313146325172534,
      "loss": 6.0309,
      "mean_token_accuracy": 0.14015766531229018,
      "num_tokens": 38121798.0,
      "step": 18825
    },
    {
      "entropy": 6.266808557510376,
      "epoch": 1.645547496285939,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004731166944323307,
      "loss": 5.9622,
      "mean_token_accuracy": 0.14645101130008698,
      "num_tokens": 38134145.0,
      "step": 18830
    },
    {
      "entropy": 6.229228591918945,
      "epoch": 1.6459844446386436,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047310192181295406,
      "loss": 5.897,
      "mean_token_accuracy": 0.1517496019601822,
      "num_tokens": 38144213.0,
      "step": 18835
    },
    {
      "entropy": 6.216114139556884,
      "epoch": 1.6464213929913485,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004730871453938788,
      "loss": 5.9117,
      "mean_token_accuracy": 0.15678840726614,
      "num_tokens": 38154245.0,
      "step": 18840
    },
    {
      "entropy": 6.2352433681488035,
      "epoch": 1.6468583413440532,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004730723651753882,
      "loss": 5.9087,
      "mean_token_accuracy": 0.15543807446956634,
      "num_tokens": 38163969.0,
      "step": 18845
    },
    {
      "entropy": 6.194263076782226,
      "epoch": 1.6472952896967579,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004730575811577659,
      "loss": 5.8453,
      "mean_token_accuracy": 0.1548292800784111,
      "num_tokens": 38173586.0,
      "step": 18850
    },
    {
      "entropy": 6.162303495407104,
      "epoch": 1.6477322380494626,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004730427933412954,
      "loss": 5.855,
      "mean_token_accuracy": 0.15056260526180268,
      "num_tokens": 38182702.0,
      "step": 18855
    },
    {
      "entropy": 6.173098087310791,
      "epoch": 1.6481691864021673,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004730280017262604,
      "loss": 5.8908,
      "mean_token_accuracy": 0.15521991550922393,
      "num_tokens": 38191937.0,
      "step": 18860
    },
    {
      "entropy": 6.192303466796875,
      "epoch": 1.648606134754872,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004730132063129445,
      "loss": 5.8991,
      "mean_token_accuracy": 0.1582408756017685,
      "num_tokens": 38201234.0,
      "step": 18865
    },
    {
      "entropy": 6.250973653793335,
      "epoch": 1.6490430831075766,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047299840710163157,
      "loss": 5.9173,
      "mean_token_accuracy": 0.1511777251958847,
      "num_tokens": 38211790.0,
      "step": 18870
    },
    {
      "entropy": 6.261939191818238,
      "epoch": 1.6494800314602815,
      "grad_norm": 2.78125,
      "learning_rate": 0.00047298360409260543,
      "loss": 5.9106,
      "mean_token_accuracy": 0.1530947729945183,
      "num_tokens": 38221946.0,
      "step": 18875
    },
    {
      "entropy": 6.170714235305786,
      "epoch": 1.649916979812986,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047296879728615,
      "loss": 5.9474,
      "mean_token_accuracy": 0.1479149267077446,
      "num_tokens": 38232366.0,
      "step": 18880
    },
    {
      "entropy": 6.207660865783692,
      "epoch": 1.650353928165691,
      "grad_norm": 3.671875,
      "learning_rate": 0.00047295398668254933,
      "loss": 5.9641,
      "mean_token_accuracy": 0.14790149182081222,
      "num_tokens": 38242504.0,
      "step": 18885
    },
    {
      "entropy": 6.342064952850341,
      "epoch": 1.6507908765183954,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004729391722820875,
      "loss": 5.9522,
      "mean_token_accuracy": 0.14760083258152007,
      "num_tokens": 38251916.0,
      "step": 18890
    },
    {
      "entropy": 6.2223903179168705,
      "epoch": 1.6512278248711003,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00047292435408504855,
      "loss": 5.8882,
      "mean_token_accuracy": 0.15217742174863816,
      "num_tokens": 38262438.0,
      "step": 18895
    },
    {
      "entropy": 6.101144456863404,
      "epoch": 1.6516647732238048,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004729095320917169,
      "loss": 5.8577,
      "mean_token_accuracy": 0.16664392948150636,
      "num_tokens": 38273095.0,
      "step": 18900
    },
    {
      "entropy": 6.250442886352539,
      "epoch": 1.6521017215765097,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004728947063023766,
      "loss": 5.8921,
      "mean_token_accuracy": 0.1484962970018387,
      "num_tokens": 38282827.0,
      "step": 18905
    },
    {
      "entropy": 6.266478490829468,
      "epoch": 1.6525386699292144,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004728798767173122,
      "loss": 5.8881,
      "mean_token_accuracy": 0.14773721992969513,
      "num_tokens": 38292827.0,
      "step": 18910
    },
    {
      "entropy": 6.24560136795044,
      "epoch": 1.652975618281919,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047286504333680804,
      "loss": 5.9473,
      "mean_token_accuracy": 0.14692333340644836,
      "num_tokens": 38302658.0,
      "step": 18915
    },
    {
      "entropy": 6.235063314437866,
      "epoch": 1.6534125666346238,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004728502061611488,
      "loss": 5.9094,
      "mean_token_accuracy": 0.1494828313589096,
      "num_tokens": 38312154.0,
      "step": 18920
    },
    {
      "entropy": 6.209419202804566,
      "epoch": 1.6538495149873285,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047283536519061875,
      "loss": 5.9101,
      "mean_token_accuracy": 0.15123223513364792,
      "num_tokens": 38321317.0,
      "step": 18925
    },
    {
      "entropy": 6.2039882183074955,
      "epoch": 1.6542864633400332,
      "grad_norm": 2.734375,
      "learning_rate": 0.00047282052042550276,
      "loss": 5.8933,
      "mean_token_accuracy": 0.14553627967834473,
      "num_tokens": 38331632.0,
      "step": 18930
    },
    {
      "entropy": 6.266951036453247,
      "epoch": 1.6547234116927378,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047280567186608563,
      "loss": 5.9712,
      "mean_token_accuracy": 0.14536414965987204,
      "num_tokens": 38342072.0,
      "step": 18935
    },
    {
      "entropy": 6.294876194000244,
      "epoch": 1.6551603600454428,
      "grad_norm": 2.203125,
      "learning_rate": 0.000472790819512652,
      "loss": 5.9084,
      "mean_token_accuracy": 0.14771281629800798,
      "num_tokens": 38351105.0,
      "step": 18940
    },
    {
      "entropy": 6.208601284027099,
      "epoch": 1.6555973083981472,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004727759633654868,
      "loss": 5.9749,
      "mean_token_accuracy": 0.14320026859641075,
      "num_tokens": 38361521.0,
      "step": 18945
    },
    {
      "entropy": 6.152694892883301,
      "epoch": 1.6560342567508521,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004727611034248751,
      "loss": 5.872,
      "mean_token_accuracy": 0.1490395784378052,
      "num_tokens": 38370720.0,
      "step": 18950
    },
    {
      "entropy": 6.219479608535766,
      "epoch": 1.6564712051035566,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004727462396911017,
      "loss": 5.9342,
      "mean_token_accuracy": 0.1528848811984062,
      "num_tokens": 38380434.0,
      "step": 18955
    },
    {
      "entropy": 6.1476343154907225,
      "epoch": 1.6569081534562615,
      "grad_norm": 3.78125,
      "learning_rate": 0.00047273137216445176,
      "loss": 5.8242,
      "mean_token_accuracy": 0.1722382739186287,
      "num_tokens": 38390503.0,
      "step": 18960
    },
    {
      "entropy": 6.2750279903411865,
      "epoch": 1.6573451018089662,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004727165008452106,
      "loss": 6.0159,
      "mean_token_accuracy": 0.14436568096280097,
      "num_tokens": 38401757.0,
      "step": 18965
    },
    {
      "entropy": 6.1856043338775635,
      "epoch": 1.657782050161671,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004727016257336633,
      "loss": 5.9285,
      "mean_token_accuracy": 0.15506427437067033,
      "num_tokens": 38412130.0,
      "step": 18970
    },
    {
      "entropy": 6.159600019454956,
      "epoch": 1.6582189985143756,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047268674683009526,
      "loss": 5.9158,
      "mean_token_accuracy": 0.15142978578805924,
      "num_tokens": 38422155.0,
      "step": 18975
    },
    {
      "entropy": 6.165815925598144,
      "epoch": 1.6586559468670803,
      "grad_norm": 3.828125,
      "learning_rate": 0.00047267186413479185,
      "loss": 5.833,
      "mean_token_accuracy": 0.160987389087677,
      "num_tokens": 38431621.0,
      "step": 18980
    },
    {
      "entropy": 6.309680461883545,
      "epoch": 1.659092895219785,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047265697764803843,
      "loss": 5.9857,
      "mean_token_accuracy": 0.14812910109758376,
      "num_tokens": 38442354.0,
      "step": 18985
    },
    {
      "entropy": 6.23242073059082,
      "epoch": 1.6595298435724897,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047264208737012064,
      "loss": 5.9199,
      "mean_token_accuracy": 0.14672013372182846,
      "num_tokens": 38452932.0,
      "step": 18990
    },
    {
      "entropy": 6.2252781867980955,
      "epoch": 1.6599667919251946,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004726271933013241,
      "loss": 5.9364,
      "mean_token_accuracy": 0.15081965178251266,
      "num_tokens": 38464442.0,
      "step": 18995
    },
    {
      "entropy": 6.252434301376343,
      "epoch": 1.660403740277899,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047261229544193434,
      "loss": 5.9318,
      "mean_token_accuracy": 0.14702566415071489,
      "num_tokens": 38475718.0,
      "step": 19000
    },
    {
      "entropy": 6.229882764816284,
      "epoch": 1.660840688630604,
      "grad_norm": 2.78125,
      "learning_rate": 0.00047259739379223726,
      "loss": 5.9103,
      "mean_token_accuracy": 0.14558252096176147,
      "num_tokens": 38485091.0,
      "step": 19005
    },
    {
      "entropy": 6.229349756240845,
      "epoch": 1.6612776369833084,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047258248835251867,
      "loss": 5.9178,
      "mean_token_accuracy": 0.15026048421859742,
      "num_tokens": 38495748.0,
      "step": 19010
    },
    {
      "entropy": 6.240655565261841,
      "epoch": 1.6617145853360134,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047256757912306435,
      "loss": 5.9683,
      "mean_token_accuracy": 0.15028906017541885,
      "num_tokens": 38506152.0,
      "step": 19015
    },
    {
      "entropy": 6.234006023406982,
      "epoch": 1.662151533688718,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004725526661041604,
      "loss": 5.8984,
      "mean_token_accuracy": 0.15526505708694457,
      "num_tokens": 38515722.0,
      "step": 19020
    },
    {
      "entropy": 6.185821151733398,
      "epoch": 1.6625884820414227,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004725377492960927,
      "loss": 5.9038,
      "mean_token_accuracy": 0.15831172913312913,
      "num_tokens": 38526270.0,
      "step": 19025
    },
    {
      "entropy": 6.2360594272613525,
      "epoch": 1.6630254303941274,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004725228286991475,
      "loss": 5.8824,
      "mean_token_accuracy": 0.14904798865318297,
      "num_tokens": 38537436.0,
      "step": 19030
    },
    {
      "entropy": 6.240790176391601,
      "epoch": 1.6634623787468321,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004725079043136111,
      "loss": 5.8921,
      "mean_token_accuracy": 0.153378726541996,
      "num_tokens": 38547298.0,
      "step": 19035
    },
    {
      "entropy": 6.176030015945434,
      "epoch": 1.6638993270995368,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004724929761397694,
      "loss": 5.9017,
      "mean_token_accuracy": 0.15330791026353835,
      "num_tokens": 38556465.0,
      "step": 19040
    },
    {
      "entropy": 6.194144010543823,
      "epoch": 1.6643362754522415,
      "grad_norm": 2.6875,
      "learning_rate": 0.000472478044177909,
      "loss": 5.9368,
      "mean_token_accuracy": 0.15573522299528123,
      "num_tokens": 38568009.0,
      "step": 19045
    },
    {
      "entropy": 6.196718454360962,
      "epoch": 1.6647732238049464,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004724631084283163,
      "loss": 5.9595,
      "mean_token_accuracy": 0.14904247969388962,
      "num_tokens": 38578778.0,
      "step": 19050
    },
    {
      "entropy": 6.199713850021363,
      "epoch": 1.6652101721576509,
      "grad_norm": 3.375,
      "learning_rate": 0.00047244816889127763,
      "loss": 5.8489,
      "mean_token_accuracy": 0.15676990896463394,
      "num_tokens": 38588991.0,
      "step": 19055
    },
    {
      "entropy": 6.279504251480103,
      "epoch": 1.6656471205103558,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004724332255670797,
      "loss": 5.8703,
      "mean_token_accuracy": 0.16127374172210693,
      "num_tokens": 38598800.0,
      "step": 19060
    },
    {
      "entropy": 6.166695690155029,
      "epoch": 1.6660840688630603,
      "grad_norm": 3.59375,
      "learning_rate": 0.00047241827845600896,
      "loss": 5.8932,
      "mean_token_accuracy": 0.15378281474113464,
      "num_tokens": 38608680.0,
      "step": 19065
    },
    {
      "entropy": 6.163301038742065,
      "epoch": 1.6665210172157652,
      "grad_norm": 4.3125,
      "learning_rate": 0.00047240332755835217,
      "loss": 5.9121,
      "mean_token_accuracy": 0.15003986060619354,
      "num_tokens": 38618402.0,
      "step": 19070
    },
    {
      "entropy": 6.273298740386963,
      "epoch": 1.6669579655684696,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047238837287439625,
      "loss": 5.9496,
      "mean_token_accuracy": 0.1489235371351242,
      "num_tokens": 38628370.0,
      "step": 19075
    },
    {
      "entropy": 6.196091890335083,
      "epoch": 1.6673949139211746,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004723734144044278,
      "loss": 5.8704,
      "mean_token_accuracy": 0.157895989716053,
      "num_tokens": 38638138.0,
      "step": 19080
    },
    {
      "entropy": 6.238557147979736,
      "epoch": 1.6678318622738793,
      "grad_norm": 3.046875,
      "learning_rate": 0.00047235845214873385,
      "loss": 5.9498,
      "mean_token_accuracy": 0.1485954999923706,
      "num_tokens": 38649412.0,
      "step": 19085
    },
    {
      "entropy": 6.219301795959472,
      "epoch": 1.668268810626584,
      "grad_norm": 3.375,
      "learning_rate": 0.00047234348610760145,
      "loss": 5.8679,
      "mean_token_accuracy": 0.15244266390800476,
      "num_tokens": 38660048.0,
      "step": 19090
    },
    {
      "entropy": 6.221240377426147,
      "epoch": 1.6687057589792886,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004723285162813175,
      "loss": 5.8742,
      "mean_token_accuracy": 0.16022849828004837,
      "num_tokens": 38671246.0,
      "step": 19095
    },
    {
      "entropy": 6.20300178527832,
      "epoch": 1.6691427073319933,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004723135426701693,
      "loss": 5.9148,
      "mean_token_accuracy": 0.15124292969703673,
      "num_tokens": 38680880.0,
      "step": 19100
    },
    {
      "entropy": 6.149772644042969,
      "epoch": 1.669579655684698,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004722985652744438,
      "loss": 5.9364,
      "mean_token_accuracy": 0.15482670515775682,
      "num_tokens": 38691440.0,
      "step": 19105
    },
    {
      "entropy": 6.17259464263916,
      "epoch": 1.6700166040374027,
      "grad_norm": 3.984375,
      "learning_rate": 0.00047228358409442863,
      "loss": 5.8996,
      "mean_token_accuracy": 0.15238867998123168,
      "num_tokens": 38700846.0,
      "step": 19110
    },
    {
      "entropy": 6.328890752792359,
      "epoch": 1.6704535523901076,
      "grad_norm": 2.890625,
      "learning_rate": 0.00047226859913041084,
      "loss": 5.9491,
      "mean_token_accuracy": 0.151666322350502,
      "num_tokens": 38712130.0,
      "step": 19115
    },
    {
      "entropy": 6.346985340118408,
      "epoch": 1.670890500742812,
      "grad_norm": 6.03125,
      "learning_rate": 0.00047225361038267795,
      "loss": 5.9904,
      "mean_token_accuracy": 0.15224644020199776,
      "num_tokens": 38722482.0,
      "step": 19120
    },
    {
      "entropy": 6.138849115371704,
      "epoch": 1.671327449095517,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047223861785151744,
      "loss": 5.8636,
      "mean_token_accuracy": 0.15834202468395234,
      "num_tokens": 38731936.0,
      "step": 19125
    },
    {
      "entropy": 6.146460103988647,
      "epoch": 1.6717643974482215,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004722236215372169,
      "loss": 5.8221,
      "mean_token_accuracy": 0.15554715991020202,
      "num_tokens": 38741444.0,
      "step": 19130
    },
    {
      "entropy": 6.227097606658935,
      "epoch": 1.6722013458009264,
      "grad_norm": 6.0,
      "learning_rate": 0.000472208621440064,
      "loss": 5.9639,
      "mean_token_accuracy": 0.1432076446712017,
      "num_tokens": 38751454.0,
      "step": 19135
    },
    {
      "entropy": 6.236496114730835,
      "epoch": 1.672638294153631,
      "grad_norm": 2.5,
      "learning_rate": 0.0004721936175603463,
      "loss": 5.9162,
      "mean_token_accuracy": 0.14393841549754144,
      "num_tokens": 38761373.0,
      "step": 19140
    },
    {
      "entropy": 6.249639701843262,
      "epoch": 1.6730752425063358,
      "grad_norm": 4.1875,
      "learning_rate": 0.00047217860989835184,
      "loss": 5.9786,
      "mean_token_accuracy": 0.15881305187940598,
      "num_tokens": 38772047.0,
      "step": 19145
    },
    {
      "entropy": 6.217469644546509,
      "epoch": 1.6735121908590405,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004721635984543682,
      "loss": 5.8626,
      "mean_token_accuracy": 0.15922709554433823,
      "num_tokens": 38781393.0,
      "step": 19150
    },
    {
      "entropy": 6.236492681503296,
      "epoch": 1.6739491392117452,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004721485832286835,
      "loss": 5.886,
      "mean_token_accuracy": 0.153799869120121,
      "num_tokens": 38791104.0,
      "step": 19155
    },
    {
      "entropy": 6.147913837432862,
      "epoch": 1.6743860875644498,
      "grad_norm": 5.625,
      "learning_rate": 0.0004721335642215856,
      "loss": 5.9423,
      "mean_token_accuracy": 0.15055005103349686,
      "num_tokens": 38801855.0,
      "step": 19160
    },
    {
      "entropy": 6.112828063964844,
      "epoch": 1.6748230359171545,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004721185414333627,
      "loss": 5.8435,
      "mean_token_accuracy": 0.15661778897047043,
      "num_tokens": 38811983.0,
      "step": 19165
    },
    {
      "entropy": 6.19040207862854,
      "epoch": 1.6752599842698594,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004721035148643029,
      "loss": 5.8687,
      "mean_token_accuracy": 0.1550484538078308,
      "num_tokens": 38821762.0,
      "step": 19170
    },
    {
      "entropy": 6.151162338256836,
      "epoch": 1.675696932622564,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004720884845146944,
      "loss": 5.9008,
      "mean_token_accuracy": 0.1557896539568901,
      "num_tokens": 38831840.0,
      "step": 19175
    },
    {
      "entropy": 6.174148225784302,
      "epoch": 1.6761338809752688,
      "grad_norm": 4.0625,
      "learning_rate": 0.00047207345038482545,
      "loss": 5.8214,
      "mean_token_accuracy": 0.1637551724910736,
      "num_tokens": 38841506.0,
      "step": 19180
    },
    {
      "entropy": 6.208927917480469,
      "epoch": 1.6765708293279733,
      "grad_norm": 4.0,
      "learning_rate": 0.0004720584124749844,
      "loss": 5.9534,
      "mean_token_accuracy": 0.15282663479447364,
      "num_tokens": 38851793.0,
      "step": 19185
    },
    {
      "entropy": 6.109137010574341,
      "epoch": 1.6770077776806782,
      "grad_norm": 3.484375,
      "learning_rate": 0.00047204337078545985,
      "loss": 5.8609,
      "mean_token_accuracy": 0.1592277929186821,
      "num_tokens": 38861958.0,
      "step": 19190
    },
    {
      "entropy": 6.174116897583008,
      "epoch": 1.677444726033383,
      "grad_norm": 5.1875,
      "learning_rate": 0.0004720283253165402,
      "loss": 5.8827,
      "mean_token_accuracy": 0.15537990480661393,
      "num_tokens": 38872181.0,
      "step": 19195
    },
    {
      "entropy": 6.22848162651062,
      "epoch": 1.6778816743860876,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047201327606851394,
      "loss": 5.9217,
      "mean_token_accuracy": 0.15566233098506926,
      "num_tokens": 38882423.0,
      "step": 19200
    },
    {
      "entropy": 6.283469343185425,
      "epoch": 1.6783186227387923,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004719982230416699,
      "loss": 5.9176,
      "mean_token_accuracy": 0.14823465496301652,
      "num_tokens": 38892833.0,
      "step": 19205
    },
    {
      "entropy": 6.2687585830688475,
      "epoch": 1.678755571091497,
      "grad_norm": 3.875,
      "learning_rate": 0.0004719831662362967,
      "loss": 5.9647,
      "mean_token_accuracy": 0.15153702050447465,
      "num_tokens": 38904037.0,
      "step": 19210
    },
    {
      "entropy": 6.206808376312256,
      "epoch": 1.6791925194442017,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004719681056526831,
      "loss": 5.8893,
      "mean_token_accuracy": 0.15857305526733398,
      "num_tokens": 38914194.0,
      "step": 19215
    },
    {
      "entropy": 6.170632934570312,
      "epoch": 1.6796294677969064,
      "grad_norm": 2.5,
      "learning_rate": 0.00047195304129111804,
      "loss": 5.9254,
      "mean_token_accuracy": 0.1507832795381546,
      "num_tokens": 38925354.0,
      "step": 19220
    },
    {
      "entropy": 6.283861589431763,
      "epoch": 1.6800664161496113,
      "grad_norm": 2.96875,
      "learning_rate": 0.00047193797315189047,
      "loss": 5.9479,
      "mean_token_accuracy": 0.1498409166932106,
      "num_tokens": 38936170.0,
      "step": 19225
    },
    {
      "entropy": 6.275005102157593,
      "epoch": 1.6805033645023157,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004719229012352894,
      "loss": 5.9315,
      "mean_token_accuracy": 0.15298874974250792,
      "num_tokens": 38945316.0,
      "step": 19230
    },
    {
      "entropy": 6.259566783905029,
      "epoch": 1.6809403128550207,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047190782554160386,
      "loss": 5.9574,
      "mean_token_accuracy": 0.14914653450250626,
      "num_tokens": 38955444.0,
      "step": 19235
    },
    {
      "entropy": 6.190140676498413,
      "epoch": 1.6813772612077251,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004718927460711231,
      "loss": 5.8488,
      "mean_token_accuracy": 0.15731645226478577,
      "num_tokens": 38965202.0,
      "step": 19240
    },
    {
      "entropy": 6.162310838699341,
      "epoch": 1.68181420956043,
      "grad_norm": 2.890625,
      "learning_rate": 0.00047187766282413623,
      "loss": 5.9343,
      "mean_token_accuracy": 0.15514635667204857,
      "num_tokens": 38975822.0,
      "step": 19245
    },
    {
      "entropy": 6.238598489761353,
      "epoch": 1.6822511579131345,
      "grad_norm": 3.25,
      "learning_rate": 0.0004718625758009327,
      "loss": 5.908,
      "mean_token_accuracy": 0.1520871937274933,
      "num_tokens": 38985890.0,
      "step": 19250
    },
    {
      "entropy": 6.2575523376464846,
      "epoch": 1.6826881062658394,
      "grad_norm": 3.234375,
      "learning_rate": 0.00047184748500180175,
      "loss": 5.9016,
      "mean_token_accuracy": 0.15715654343366622,
      "num_tokens": 38996374.0,
      "step": 19255
    },
    {
      "entropy": 6.216891860961914,
      "epoch": 1.6831250546185441,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004718323904270329,
      "loss": 5.9001,
      "mean_token_accuracy": 0.15249834656715394,
      "num_tokens": 39006371.0,
      "step": 19260
    },
    {
      "entropy": 6.192630767822266,
      "epoch": 1.6835620029712488,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004718172920769157,
      "loss": 5.8621,
      "mean_token_accuracy": 0.1579944372177124,
      "num_tokens": 39015514.0,
      "step": 19265
    },
    {
      "entropy": 6.165468311309814,
      "epoch": 1.6839989513239535,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004718021899517397,
      "loss": 5.887,
      "mean_token_accuracy": 0.15349025130271912,
      "num_tokens": 39025700.0,
      "step": 19270
    },
    {
      "entropy": 6.167158317565918,
      "epoch": 1.6844358996766582,
      "grad_norm": 3.359375,
      "learning_rate": 0.00047178708405179454,
      "loss": 5.9048,
      "mean_token_accuracy": 0.15944857150316238,
      "num_tokens": 39036157.0,
      "step": 19275
    },
    {
      "entropy": 6.29277229309082,
      "epoch": 1.6848728480293629,
      "grad_norm": 4.34375,
      "learning_rate": 0.00047177197437737003,
      "loss": 6.0618,
      "mean_token_accuracy": 0.14412114769220352,
      "num_tokens": 39046044.0,
      "step": 19280
    },
    {
      "entropy": 6.185094928741455,
      "epoch": 1.6853097963820676,
      "grad_norm": 3.5,
      "learning_rate": 0.0004717568609287559,
      "loss": 5.9415,
      "mean_token_accuracy": 0.1455281265079975,
      "num_tokens": 39056034.0,
      "step": 19285
    },
    {
      "entropy": 6.231845664978027,
      "epoch": 1.6857467447347725,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004717417437062421,
      "loss": 5.9157,
      "mean_token_accuracy": 0.1515277624130249,
      "num_tokens": 39065687.0,
      "step": 19290
    },
    {
      "entropy": 6.290487957000733,
      "epoch": 1.686183693087477,
      "grad_norm": 3.796875,
      "learning_rate": 0.00047172662271011855,
      "loss": 5.9159,
      "mean_token_accuracy": 0.15213676691055297,
      "num_tokens": 39077107.0,
      "step": 19295
    },
    {
      "entropy": 6.206365251541138,
      "epoch": 1.6866206414401819,
      "grad_norm": 2.921875,
      "learning_rate": 0.00047171149794067533,
      "loss": 5.9149,
      "mean_token_accuracy": 0.15343448221683503,
      "num_tokens": 39087220.0,
      "step": 19300
    },
    {
      "entropy": 6.224979257583618,
      "epoch": 1.6870575897928863,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004716963693982025,
      "loss": 5.8683,
      "mean_token_accuracy": 0.15687804371118547,
      "num_tokens": 39097706.0,
      "step": 19305
    },
    {
      "entropy": 6.200500583648681,
      "epoch": 1.6874945381455912,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004716812370829902,
      "loss": 5.9199,
      "mean_token_accuracy": 0.14626403898000717,
      "num_tokens": 39108044.0,
      "step": 19310
    },
    {
      "entropy": 6.207551050186157,
      "epoch": 1.687931486498296,
      "grad_norm": 4.5,
      "learning_rate": 0.0004716661009953287,
      "loss": 5.9009,
      "mean_token_accuracy": 0.15139940977096558,
      "num_tokens": 39116917.0,
      "step": 19315
    },
    {
      "entropy": 6.19309368133545,
      "epoch": 1.6883684348510006,
      "grad_norm": 2.125,
      "learning_rate": 0.0004716509611355084,
      "loss": 5.8951,
      "mean_token_accuracy": 0.1565606936812401,
      "num_tokens": 39126770.0,
      "step": 19320
    },
    {
      "entropy": 6.2025164604187015,
      "epoch": 1.6888053832037053,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047163581750381956,
      "loss": 5.9345,
      "mean_token_accuracy": 0.14297117590904235,
      "num_tokens": 39137406.0,
      "step": 19325
    },
    {
      "entropy": 6.207695245742798,
      "epoch": 1.68924233155641,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047162067010055275,
      "loss": 5.9008,
      "mean_token_accuracy": 0.15250989347696303,
      "num_tokens": 39146950.0,
      "step": 19330
    },
    {
      "entropy": 6.157593059539795,
      "epoch": 1.6896792799091147,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004716055189259984,
      "loss": 5.8535,
      "mean_token_accuracy": 0.1541092336177826,
      "num_tokens": 39157382.0,
      "step": 19335
    },
    {
      "entropy": 6.1588804721832275,
      "epoch": 1.6901162282618194,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047159036398044727,
      "loss": 5.9042,
      "mean_token_accuracy": 0.15435107350349425,
      "num_tokens": 39166641.0,
      "step": 19340
    },
    {
      "entropy": 6.272203350067139,
      "epoch": 1.6905531766145243,
      "grad_norm": 2.25,
      "learning_rate": 0.0004715752052641899,
      "loss": 5.9512,
      "mean_token_accuracy": 0.15413000881671907,
      "num_tokens": 39176731.0,
      "step": 19345
    },
    {
      "entropy": 6.2381476879119875,
      "epoch": 1.6909901249672288,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004715600427775171,
      "loss": 5.9548,
      "mean_token_accuracy": 0.15126673877239227,
      "num_tokens": 39187202.0,
      "step": 19350
    },
    {
      "entropy": 6.151942873001099,
      "epoch": 1.6914270733199337,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004715448765207196,
      "loss": 5.9301,
      "mean_token_accuracy": 0.14793859496712686,
      "num_tokens": 39197302.0,
      "step": 19355
    },
    {
      "entropy": 6.2880247116088865,
      "epoch": 1.6918640216726382,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004715297064940885,
      "loss": 5.9485,
      "mean_token_accuracy": 0.15435196310281754,
      "num_tokens": 39208382.0,
      "step": 19360
    },
    {
      "entropy": 6.280681991577149,
      "epoch": 1.692300970025343,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004715145326979146,
      "loss": 5.8938,
      "mean_token_accuracy": 0.15162560492753982,
      "num_tokens": 39218529.0,
      "step": 19365
    },
    {
      "entropy": 6.156153011322021,
      "epoch": 1.6927379183780478,
      "grad_norm": 2.625,
      "learning_rate": 0.00047149935513248896,
      "loss": 5.9481,
      "mean_token_accuracy": 0.14894766211509705,
      "num_tokens": 39228976.0,
      "step": 19370
    },
    {
      "entropy": 6.221373653411865,
      "epoch": 1.6931748667307525,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004714841737981028,
      "loss": 5.8352,
      "mean_token_accuracy": 0.15608796179294587,
      "num_tokens": 39238921.0,
      "step": 19375
    },
    {
      "entropy": 6.176568031311035,
      "epoch": 1.6936118150834572,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047146898869504717,
      "loss": 5.8654,
      "mean_token_accuracy": 0.1521398529410362,
      "num_tokens": 39247978.0,
      "step": 19380
    },
    {
      "entropy": 6.172025394439697,
      "epoch": 1.6940487634361618,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047145379982361336,
      "loss": 5.8994,
      "mean_token_accuracy": 0.15236786752939224,
      "num_tokens": 39257487.0,
      "step": 19385
    },
    {
      "entropy": 6.239286422729492,
      "epoch": 1.6944857117888665,
      "grad_norm": 3.09375,
      "learning_rate": 0.00047143860718409285,
      "loss": 5.8814,
      "mean_token_accuracy": 0.1526914969086647,
      "num_tokens": 39267819.0,
      "step": 19390
    },
    {
      "entropy": 6.235224819183349,
      "epoch": 1.6949226601415712,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047142341077677685,
      "loss": 5.9737,
      "mean_token_accuracy": 0.1467155821621418,
      "num_tokens": 39277673.0,
      "step": 19395
    },
    {
      "entropy": 6.211605215072632,
      "epoch": 1.6953596084942761,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004714082106019569,
      "loss": 5.838,
      "mean_token_accuracy": 0.1537025600671768,
      "num_tokens": 39287340.0,
      "step": 19400
    },
    {
      "entropy": 6.240170478820801,
      "epoch": 1.6957965568469806,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004713930066599246,
      "loss": 5.9608,
      "mean_token_accuracy": 0.1521234393119812,
      "num_tokens": 39298277.0,
      "step": 19405
    },
    {
      "entropy": 6.254997777938843,
      "epoch": 1.6962335051996855,
      "grad_norm": 3.4375,
      "learning_rate": 0.00047137779895097145,
      "loss": 5.9229,
      "mean_token_accuracy": 0.1478773459792137,
      "num_tokens": 39307996.0,
      "step": 19410
    },
    {
      "entropy": 6.19044246673584,
      "epoch": 1.69667045355239,
      "grad_norm": 3.609375,
      "learning_rate": 0.00047136258747538926,
      "loss": 5.9012,
      "mean_token_accuracy": 0.1545254871249199,
      "num_tokens": 39317327.0,
      "step": 19415
    },
    {
      "entropy": 6.165656089782715,
      "epoch": 1.697107401905095,
      "grad_norm": 2.375,
      "learning_rate": 0.0004713473722334697,
      "loss": 5.8381,
      "mean_token_accuracy": 0.16236181408166886,
      "num_tokens": 39326728.0,
      "step": 19420
    },
    {
      "entropy": 6.206114959716797,
      "epoch": 1.6975443502577994,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047133215322550464,
      "loss": 6.0188,
      "mean_token_accuracy": 0.14539464712142944,
      "num_tokens": 39336606.0,
      "step": 19425
    },
    {
      "entropy": 6.247616910934449,
      "epoch": 1.6979812986105043,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004713169304517861,
      "loss": 5.9294,
      "mean_token_accuracy": 0.150938680768013,
      "num_tokens": 39345622.0,
      "step": 19430
    },
    {
      "entropy": 6.2544622898101805,
      "epoch": 1.698418246963209,
      "grad_norm": 2.625,
      "learning_rate": 0.00047130170391260597,
      "loss": 5.9363,
      "mean_token_accuracy": 0.1494395315647125,
      "num_tokens": 39356868.0,
      "step": 19435
    },
    {
      "entropy": 6.225611543655395,
      "epoch": 1.6988551953159137,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004712864736082562,
      "loss": 5.8237,
      "mean_token_accuracy": 0.1584457516670227,
      "num_tokens": 39366866.0,
      "step": 19440
    },
    {
      "entropy": 6.198269891738891,
      "epoch": 1.6992921436686184,
      "grad_norm": 3.109375,
      "learning_rate": 0.00047127123953902907,
      "loss": 5.8984,
      "mean_token_accuracy": 0.15157056450843812,
      "num_tokens": 39377303.0,
      "step": 19445
    },
    {
      "entropy": 6.168129014968872,
      "epoch": 1.699729092021323,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004712560017052167,
      "loss": 5.8636,
      "mean_token_accuracy": 0.1560070976614952,
      "num_tokens": 39387191.0,
      "step": 19450
    },
    {
      "entropy": 6.14867787361145,
      "epoch": 1.7001660403740277,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004712407601071114,
      "loss": 5.8931,
      "mean_token_accuracy": 0.15525287240743638,
      "num_tokens": 39397416.0,
      "step": 19455
    },
    {
      "entropy": 6.203057336807251,
      "epoch": 1.7006029887267324,
      "grad_norm": 3.171875,
      "learning_rate": 0.00047122551474500555,
      "loss": 5.9025,
      "mean_token_accuracy": 0.1521309234201908,
      "num_tokens": 39407855.0,
      "step": 19460
    },
    {
      "entropy": 6.199628114700317,
      "epoch": 1.7010399370794373,
      "grad_norm": 2.6875,
      "learning_rate": 0.00047121026561919137,
      "loss": 5.953,
      "mean_token_accuracy": 0.14864318519830705,
      "num_tokens": 39417735.0,
      "step": 19465
    },
    {
      "entropy": 6.18361496925354,
      "epoch": 1.7014768854321418,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004711950127299616,
      "loss": 5.8879,
      "mean_token_accuracy": 0.15658994764089584,
      "num_tokens": 39427641.0,
      "step": 19470
    },
    {
      "entropy": 6.183968114852905,
      "epoch": 1.7019138337848467,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004711797560776087,
      "loss": 5.8241,
      "mean_token_accuracy": 0.16399957090616227,
      "num_tokens": 39437372.0,
      "step": 19475
    },
    {
      "entropy": 6.266106367111206,
      "epoch": 1.7023507821375512,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047116449566242516,
      "loss": 5.9023,
      "mean_token_accuracy": 0.1559259533882141,
      "num_tokens": 39447800.0,
      "step": 19480
    },
    {
      "entropy": 6.190653991699219,
      "epoch": 1.7027877304902561,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004711492314847038,
      "loss": 5.9338,
      "mean_token_accuracy": 0.15377873033285142,
      "num_tokens": 39457310.0,
      "step": 19485
    },
    {
      "entropy": 6.159643220901489,
      "epoch": 1.7032246788429608,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047113396354473753,
      "loss": 5.9242,
      "mean_token_accuracy": 0.14963654130697251,
      "num_tokens": 39468579.0,
      "step": 19490
    },
    {
      "entropy": 6.268629884719848,
      "epoch": 1.7036616271956655,
      "grad_norm": 2.25,
      "learning_rate": 0.000471118691842819,
      "loss": 5.8643,
      "mean_token_accuracy": 0.15711129903793336,
      "num_tokens": 39479027.0,
      "step": 19495
    },
    {
      "entropy": 6.187201976776123,
      "epoch": 1.7040985755483702,
      "grad_norm": 2.921875,
      "learning_rate": 0.00047110341637924126,
      "loss": 5.8623,
      "mean_token_accuracy": 0.1627778559923172,
      "num_tokens": 39488651.0,
      "step": 19500
    },
    {
      "entropy": 6.120381212234497,
      "epoch": 1.7045355239010749,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004710881371542972,
      "loss": 5.9226,
      "mean_token_accuracy": 0.15346940010786056,
      "num_tokens": 39498633.0,
      "step": 19505
    },
    {
      "entropy": 6.167008304595948,
      "epoch": 1.7049724722537796,
      "grad_norm": 2.34375,
      "learning_rate": 0.00047107285416827984,
      "loss": 5.8208,
      "mean_token_accuracy": 0.16597067415714264,
      "num_tokens": 39509351.0,
      "step": 19510
    },
    {
      "entropy": 6.2016847133636475,
      "epoch": 1.7054094206064843,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004710575674214824,
      "loss": 5.8748,
      "mean_token_accuracy": 0.15829478204250336,
      "num_tokens": 39519821.0,
      "step": 19515
    },
    {
      "entropy": 6.17425045967102,
      "epoch": 1.7058463689591892,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004710422769141982,
      "loss": 5.9139,
      "mean_token_accuracy": 0.15401314646005632,
      "num_tokens": 39530040.0,
      "step": 19520
    },
    {
      "entropy": 6.2030986785888675,
      "epoch": 1.7062833173118936,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004710269826467203,
      "loss": 5.9069,
      "mean_token_accuracy": 0.1531284749507904,
      "num_tokens": 39540669.0,
      "step": 19525
    },
    {
      "entropy": 6.228457832336426,
      "epoch": 1.7067202656645986,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047101168461934225,
      "loss": 5.8981,
      "mean_token_accuracy": 0.15268439650535584,
      "num_tokens": 39550651.0,
      "step": 19530
    },
    {
      "entropy": 6.200884199142456,
      "epoch": 1.707157214017303,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004709963828323573,
      "loss": 5.8919,
      "mean_token_accuracy": 0.13913417235016823,
      "num_tokens": 39561748.0,
      "step": 19535
    },
    {
      "entropy": 6.205925321578979,
      "epoch": 1.707594162370008,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004709810772860591,
      "loss": 5.8406,
      "mean_token_accuracy": 0.16295359134674073,
      "num_tokens": 39571328.0,
      "step": 19540
    },
    {
      "entropy": 6.142232608795166,
      "epoch": 1.7080311107227124,
      "grad_norm": 2.265625,
      "learning_rate": 0.000470965767980741,
      "loss": 5.8817,
      "mean_token_accuracy": 0.16303909718990325,
      "num_tokens": 39581931.0,
      "step": 19545
    },
    {
      "entropy": 6.239671993255615,
      "epoch": 1.7084680590754173,
      "grad_norm": 3.578125,
      "learning_rate": 0.00047095045491669696,
      "loss": 6.0175,
      "mean_token_accuracy": 0.14950380474328995,
      "num_tokens": 39592818.0,
      "step": 19550
    },
    {
      "entropy": 6.16671667098999,
      "epoch": 1.708905007428122,
      "grad_norm": 2.71875,
      "learning_rate": 0.00047093513809422043,
      "loss": 5.8817,
      "mean_token_accuracy": 0.1540909469127655,
      "num_tokens": 39602867.0,
      "step": 19555
    },
    {
      "entropy": 6.1982457637786865,
      "epoch": 1.7093419557808267,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004709198175136053,
      "loss": 5.9056,
      "mean_token_accuracy": 0.15097580403089522,
      "num_tokens": 39613185.0,
      "step": 19560
    },
    {
      "entropy": 6.200175762176514,
      "epoch": 1.7097789041335314,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004709044931751454,
      "loss": 5.8782,
      "mean_token_accuracy": 0.15892379879951476,
      "num_tokens": 39623262.0,
      "step": 19565
    },
    {
      "entropy": 6.2189844131469725,
      "epoch": 1.710215852486236,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004708891650791347,
      "loss": 5.8877,
      "mean_token_accuracy": 0.15514734238386155,
      "num_tokens": 39633184.0,
      "step": 19570
    },
    {
      "entropy": 6.143938493728638,
      "epoch": 1.710652800838941,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004708738332258671,
      "loss": 5.9399,
      "mean_token_accuracy": 0.15138588845729828,
      "num_tokens": 39643778.0,
      "step": 19575
    },
    {
      "entropy": 6.194218921661377,
      "epoch": 1.7110897491916455,
      "grad_norm": 2.375,
      "learning_rate": 0.0004708584976156368,
      "loss": 5.8497,
      "mean_token_accuracy": 0.15735856890678407,
      "num_tokens": 39653519.0,
      "step": 19580
    },
    {
      "entropy": 6.21487979888916,
      "epoch": 1.7115266975443504,
      "grad_norm": 2.875,
      "learning_rate": 0.00047084315824873793,
      "loss": 5.9248,
      "mean_token_accuracy": 0.15054146349430084,
      "num_tokens": 39663824.0,
      "step": 19585
    },
    {
      "entropy": 6.250456142425537,
      "epoch": 1.7119636458970549,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004708278151254646,
      "loss": 5.9164,
      "mean_token_accuracy": 0.14930497705936432,
      "num_tokens": 39673631.0,
      "step": 19590
    },
    {
      "entropy": 6.316969871520996,
      "epoch": 1.7124005942497598,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004708124682461112,
      "loss": 5.9679,
      "mean_token_accuracy": 0.14897105991840362,
      "num_tokens": 39683118.0,
      "step": 19595
    },
    {
      "entropy": 6.225621271133423,
      "epoch": 1.7128375426024642,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047079711761097197,
      "loss": 5.892,
      "mean_token_accuracy": 0.1513240449130535,
      "num_tokens": 39693947.0,
      "step": 19600
    },
    {
      "entropy": 6.183497142791748,
      "epoch": 1.7132744909551691,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004707817632203415,
      "loss": 5.9041,
      "mean_token_accuracy": 0.1537383034825325,
      "num_tokens": 39703874.0,
      "step": 19605
    },
    {
      "entropy": 6.170041942596436,
      "epoch": 1.7137114393078738,
      "grad_norm": 2.8125,
      "learning_rate": 0.00047076640507451425,
      "loss": 5.886,
      "mean_token_accuracy": 0.15816190093755722,
      "num_tokens": 39714997.0,
      "step": 19610
    },
    {
      "entropy": 6.243953895568848,
      "epoch": 1.7141483876605785,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004707510431737847,
      "loss": 5.8965,
      "mean_token_accuracy": 0.15344041138887404,
      "num_tokens": 39726042.0,
      "step": 19615
    },
    {
      "entropy": 6.173181819915771,
      "epoch": 1.7145853360132832,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004707356775184476,
      "loss": 5.9066,
      "mean_token_accuracy": 0.15139636844396592,
      "num_tokens": 39736733.0,
      "step": 19620
    },
    {
      "entropy": 6.232531404495239,
      "epoch": 1.715022284365988,
      "grad_norm": 3.0,
      "learning_rate": 0.0004707203081087977,
      "loss": 5.9957,
      "mean_token_accuracy": 0.14228763654828072,
      "num_tokens": 39746811.0,
      "step": 19625
    },
    {
      "entropy": 6.195516443252563,
      "epoch": 1.7154592327186926,
      "grad_norm": 2.125,
      "learning_rate": 0.0004707049349451297,
      "loss": 5.924,
      "mean_token_accuracy": 0.1504388228058815,
      "num_tokens": 39757371.0,
      "step": 19630
    },
    {
      "entropy": 6.200914669036865,
      "epoch": 1.7158961810713973,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004706895580277384,
      "loss": 5.8861,
      "mean_token_accuracy": 0.15823639929294586,
      "num_tokens": 39767430.0,
      "step": 19635
    },
    {
      "entropy": 6.2426833152771,
      "epoch": 1.7163331294241022,
      "grad_norm": 11.375,
      "learning_rate": 0.00047067417735691884,
      "loss": 5.9025,
      "mean_token_accuracy": 0.1557820424437523,
      "num_tokens": 39777270.0,
      "step": 19640
    },
    {
      "entropy": 6.184940814971924,
      "epoch": 1.7167700777768067,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004706587929329662,
      "loss": 5.8881,
      "mean_token_accuracy": 0.15841181129217147,
      "num_tokens": 39787647.0,
      "step": 19645
    },
    {
      "entropy": 6.20784068107605,
      "epoch": 1.7172070261295116,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047064340475617527,
      "loss": 5.943,
      "mean_token_accuracy": 0.15177740007638932,
      "num_tokens": 39797315.0,
      "step": 19650
    },
    {
      "entropy": 6.247117233276367,
      "epoch": 1.717643974482216,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004706280128268413,
      "loss": 5.8225,
      "mean_token_accuracy": 0.16040437817573547,
      "num_tokens": 39806625.0,
      "step": 19655
    },
    {
      "entropy": 6.172389888763428,
      "epoch": 1.718080922834921,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004706126171452596,
      "loss": 5.9545,
      "mean_token_accuracy": 0.1513059951364994,
      "num_tokens": 39818113.0,
      "step": 19660
    },
    {
      "entropy": 6.16670241355896,
      "epoch": 1.7185178711876257,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004705972177117254,
      "loss": 5.9342,
      "mean_token_accuracy": 0.15128545910120011,
      "num_tokens": 39827897.0,
      "step": 19665
    },
    {
      "entropy": 6.226880264282227,
      "epoch": 1.7189548195403304,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047058181452653403,
      "loss": 5.7712,
      "mean_token_accuracy": 0.16809355467557907,
      "num_tokens": 39837209.0,
      "step": 19670
    },
    {
      "entropy": 6.2112561702728275,
      "epoch": 1.719391767893035,
      "grad_norm": 2.8125,
      "learning_rate": 0.000470566407589981,
      "loss": 5.8854,
      "mean_token_accuracy": 0.158533738553524,
      "num_tokens": 39848023.0,
      "step": 19675
    },
    {
      "entropy": 6.17528190612793,
      "epoch": 1.7198287162457397,
      "grad_norm": 2.0,
      "learning_rate": 0.0004705509969023618,
      "loss": 5.961,
      "mean_token_accuracy": 0.1508388489484787,
      "num_tokens": 39857992.0,
      "step": 19680
    },
    {
      "entropy": 6.183531188964844,
      "epoch": 1.7202656645984444,
      "grad_norm": 2.390625,
      "learning_rate": 0.000470535582463972,
      "loss": 5.9365,
      "mean_token_accuracy": 0.15543587803840636,
      "num_tokens": 39867927.0,
      "step": 19685
    },
    {
      "entropy": 6.183828544616699,
      "epoch": 1.7207026129511491,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004705201642751073,
      "loss": 5.8588,
      "mean_token_accuracy": 0.15721919536590576,
      "num_tokens": 39877327.0,
      "step": 19690
    },
    {
      "entropy": 6.224378681182861,
      "epoch": 1.721139561303854,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047050474233606334,
      "loss": 5.8775,
      "mean_token_accuracy": 0.15297963768243789,
      "num_tokens": 39887369.0,
      "step": 19695
    },
    {
      "entropy": 6.238476514816284,
      "epoch": 1.7215765096565585,
      "grad_norm": 3.09375,
      "learning_rate": 0.000470489316647136,
      "loss": 5.932,
      "mean_token_accuracy": 0.1522274896502495,
      "num_tokens": 39897530.0,
      "step": 19700
    },
    {
      "entropy": 6.172255897521973,
      "epoch": 1.7220134580092634,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004704738872086212,
      "loss": 5.8893,
      "mean_token_accuracy": 0.1547755554318428,
      "num_tokens": 39907234.0,
      "step": 19705
    },
    {
      "entropy": 6.309378242492675,
      "epoch": 1.722450406361968,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004704584540208148,
      "loss": 5.9757,
      "mean_token_accuracy": 0.1458725966513157,
      "num_tokens": 39917940.0,
      "step": 19710
    },
    {
      "entropy": 6.191907835006714,
      "epoch": 1.7228873547146728,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004704430170840127,
      "loss": 5.8525,
      "mean_token_accuracy": 0.16579221934080124,
      "num_tokens": 39927569.0,
      "step": 19715
    },
    {
      "entropy": 6.10360951423645,
      "epoch": 1.7233243030673773,
      "grad_norm": 3.75,
      "learning_rate": 0.0004704275763985113,
      "loss": 5.9345,
      "mean_token_accuracy": 0.15377551466226577,
      "num_tokens": 39938302.0,
      "step": 19720
    },
    {
      "entropy": 6.246457767486572,
      "epoch": 1.7237612514200822,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004704121319646065,
      "loss": 5.8813,
      "mean_token_accuracy": 0.15690755993127822,
      "num_tokens": 39948133.0,
      "step": 19725
    },
    {
      "entropy": 6.210375261306763,
      "epoch": 1.7241981997727869,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047039668378259455,
      "loss": 5.8577,
      "mean_token_accuracy": 0.15972493439912797,
      "num_tokens": 39958924.0,
      "step": 19730
    },
    {
      "entropy": 6.195425701141358,
      "epoch": 1.7246351481254916,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047038123185277187,
      "loss": 5.834,
      "mean_token_accuracy": 0.15639737248420715,
      "num_tokens": 39968578.0,
      "step": 19735
    },
    {
      "entropy": 6.153471851348877,
      "epoch": 1.7250720964781963,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047036577617543484,
      "loss": 5.8417,
      "mean_token_accuracy": 0.1628642812371254,
      "num_tokens": 39978112.0,
      "step": 19740
    },
    {
      "entropy": 6.1818163871765135,
      "epoch": 1.725509044830901,
      "grad_norm": 3.375,
      "learning_rate": 0.00047035031675087983,
      "loss": 5.9179,
      "mean_token_accuracy": 0.14637974724173547,
      "num_tokens": 39988071.0,
      "step": 19745
    },
    {
      "entropy": 6.279864263534546,
      "epoch": 1.7259459931836056,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004703348535794034,
      "loss": 5.953,
      "mean_token_accuracy": 0.14586808457970618,
      "num_tokens": 39998112.0,
      "step": 19750
    },
    {
      "entropy": 6.125663137435913,
      "epoch": 1.7263829415363103,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047031938666130206,
      "loss": 5.9169,
      "mean_token_accuracy": 0.15324584022164345,
      "num_tokens": 40008869.0,
      "step": 19755
    },
    {
      "entropy": 6.214671277999878,
      "epoch": 1.7268198898890152,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004703039159968726,
      "loss": 5.9061,
      "mean_token_accuracy": 0.15851442217826844,
      "num_tokens": 40018711.0,
      "step": 19760
    },
    {
      "entropy": 6.249885225296021,
      "epoch": 1.7272568382417197,
      "grad_norm": 1.984375,
      "learning_rate": 0.00047028844158641166,
      "loss": 5.9015,
      "mean_token_accuracy": 0.1542143702507019,
      "num_tokens": 40029209.0,
      "step": 19765
    },
    {
      "entropy": 6.174735403060913,
      "epoch": 1.7276937865944246,
      "grad_norm": 2.1875,
      "learning_rate": 0.000470272963430216,
      "loss": 5.8839,
      "mean_token_accuracy": 0.15720602720975876,
      "num_tokens": 40039071.0,
      "step": 19770
    },
    {
      "entropy": 6.168638372421265,
      "epoch": 1.728130734947129,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004702574815285827,
      "loss": 5.8425,
      "mean_token_accuracy": 0.16035914421081543,
      "num_tokens": 40049765.0,
      "step": 19775
    },
    {
      "entropy": 6.159608221054077,
      "epoch": 1.728567683299834,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004702419958818085,
      "loss": 5.849,
      "mean_token_accuracy": 0.1567331075668335,
      "num_tokens": 40060020.0,
      "step": 19780
    },
    {
      "entropy": 6.245507001876831,
      "epoch": 1.7290046316525387,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047022650649019063,
      "loss": 5.8914,
      "mean_token_accuracy": 0.15106320679187774,
      "num_tokens": 40070081.0,
      "step": 19785
    },
    {
      "entropy": 6.24323616027832,
      "epoch": 1.7294415800052434,
      "grad_norm": 2.234375,
      "learning_rate": 0.000470211013354026,
      "loss": 5.9486,
      "mean_token_accuracy": 0.14734721034765244,
      "num_tokens": 40080532.0,
      "step": 19790
    },
    {
      "entropy": 6.174783039093017,
      "epoch": 1.729878528357948,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004701955164736118,
      "loss": 5.9147,
      "mean_token_accuracy": 0.1547049954533577,
      "num_tokens": 40091288.0,
      "step": 19795
    },
    {
      "entropy": 6.177815294265747,
      "epoch": 1.7303154767106528,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004701800158492454,
      "loss": 5.8778,
      "mean_token_accuracy": 0.1489587679505348,
      "num_tokens": 40100461.0,
      "step": 19800
    },
    {
      "entropy": 6.2234807968139645,
      "epoch": 1.7307524250633575,
      "grad_norm": 1.84375,
      "learning_rate": 0.000470164511481224,
      "loss": 5.8559,
      "mean_token_accuracy": 0.15290800780057906,
      "num_tokens": 40109627.0,
      "step": 19805
    },
    {
      "entropy": 6.193103551864624,
      "epoch": 1.7311893734160622,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000470149003369845,
      "loss": 5.8968,
      "mean_token_accuracy": 0.1531236633658409,
      "num_tokens": 40119183.0,
      "step": 19810
    },
    {
      "entropy": 6.172509002685547,
      "epoch": 1.731626321768767,
      "grad_norm": 1.875,
      "learning_rate": 0.0004701334915154058,
      "loss": 5.8983,
      "mean_token_accuracy": 0.15369946956634523,
      "num_tokens": 40129569.0,
      "step": 19815
    },
    {
      "entropy": 6.2103578567504885,
      "epoch": 1.7320632701214715,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047011797591820404,
      "loss": 5.8403,
      "mean_token_accuracy": 0.1522467017173767,
      "num_tokens": 40139618.0,
      "step": 19820
    },
    {
      "entropy": 6.113168096542358,
      "epoch": 1.7325002184741765,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047010245657853726,
      "loss": 5.7962,
      "mean_token_accuracy": 0.16175133287906646,
      "num_tokens": 40149349.0,
      "step": 19825
    },
    {
      "entropy": 6.199889326095581,
      "epoch": 1.732937166826881,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004700869334967031,
      "loss": 5.9075,
      "mean_token_accuracy": 0.1553854003548622,
      "num_tokens": 40159161.0,
      "step": 19830
    },
    {
      "entropy": 6.183363056182861,
      "epoch": 1.7333741151795858,
      "grad_norm": 2.34375,
      "learning_rate": 0.00047007140667299944,
      "loss": 5.8894,
      "mean_token_accuracy": 0.15276622027158737,
      "num_tokens": 40169006.0,
      "step": 19835
    },
    {
      "entropy": 6.226429414749146,
      "epoch": 1.7338110635322905,
      "grad_norm": 2.375,
      "learning_rate": 0.00047005587610772384,
      "loss": 5.8868,
      "mean_token_accuracy": 0.15136303454637529,
      "num_tokens": 40179964.0,
      "step": 19840
    },
    {
      "entropy": 6.217330360412598,
      "epoch": 1.7342480118849952,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047004034180117446,
      "loss": 5.9229,
      "mean_token_accuracy": 0.15100776106119157,
      "num_tokens": 40190598.0,
      "step": 19845
    },
    {
      "entropy": 6.23437385559082,
      "epoch": 1.7346849602377,
      "grad_norm": 2.6875,
      "learning_rate": 0.000470024803753649,
      "loss": 5.9051,
      "mean_token_accuracy": 0.15435452312231063,
      "num_tokens": 40199537.0,
      "step": 19850
    },
    {
      "entropy": 6.1629315376281735,
      "epoch": 1.7351219085904046,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004700092619654456,
      "loss": 5.875,
      "mean_token_accuracy": 0.1557350903749466,
      "num_tokens": 40209167.0,
      "step": 19855
    },
    {
      "entropy": 6.196388626098633,
      "epoch": 1.7355588569431093,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004699937164368625,
      "loss": 5.8802,
      "mean_token_accuracy": 0.1566798523068428,
      "num_tokens": 40218738.0,
      "step": 19860
    },
    {
      "entropy": 6.1942915439605715,
      "epoch": 1.735995805295814,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004699781671681977,
      "loss": 5.8526,
      "mean_token_accuracy": 0.15883951634168625,
      "num_tokens": 40227897.0,
      "step": 19865
    },
    {
      "entropy": 6.193858575820923,
      "epoch": 1.736432753648519,
      "grad_norm": 2.890625,
      "learning_rate": 0.00046996261415974946,
      "loss": 5.9319,
      "mean_token_accuracy": 0.15008512884378433,
      "num_tokens": 40238870.0,
      "step": 19870
    },
    {
      "entropy": 6.158765268325806,
      "epoch": 1.7368697020012234,
      "grad_norm": 2.84375,
      "learning_rate": 0.00046994705741181615,
      "loss": 5.8331,
      "mean_token_accuracy": 0.15819477885961533,
      "num_tokens": 40247832.0,
      "step": 19875
    },
    {
      "entropy": 6.226218843460083,
      "epoch": 1.7373066503539283,
      "grad_norm": 3.25,
      "learning_rate": 0.0004699314969246961,
      "loss": 5.8815,
      "mean_token_accuracy": 0.15299174934625626,
      "num_tokens": 40258696.0,
      "step": 19880
    },
    {
      "entropy": 6.152800464630127,
      "epoch": 1.7377435987066328,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046991593269868776,
      "loss": 5.8614,
      "mean_token_accuracy": 0.15949728488922119,
      "num_tokens": 40267767.0,
      "step": 19885
    },
    {
      "entropy": 6.2125342845916744,
      "epoch": 1.7381805470593377,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004699003647340898,
      "loss": 5.9273,
      "mean_token_accuracy": 0.1548434630036354,
      "num_tokens": 40277813.0,
      "step": 19890
    },
    {
      "entropy": 6.1712571144104,
      "epoch": 1.7386174954120421,
      "grad_norm": 2.25,
      "learning_rate": 0.00046988479303120066,
      "loss": 5.9101,
      "mean_token_accuracy": 0.15774208903312684,
      "num_tokens": 40289031.0,
      "step": 19895
    },
    {
      "entropy": 6.224442768096924,
      "epoch": 1.739054443764747,
      "grad_norm": 5.3125,
      "learning_rate": 0.00046986921759031906,
      "loss": 5.8724,
      "mean_token_accuracy": 0.1567473903298378,
      "num_tokens": 40299428.0,
      "step": 19900
    },
    {
      "entropy": 6.174745225906372,
      "epoch": 1.7394913921174517,
      "grad_norm": 2.75,
      "learning_rate": 0.0004698536384117438,
      "loss": 5.851,
      "mean_token_accuracy": 0.15421851724386215,
      "num_tokens": 40308996.0,
      "step": 19905
    },
    {
      "entropy": 6.201578283309937,
      "epoch": 1.7399283404701564,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046983805549577365,
      "loss": 5.9097,
      "mean_token_accuracy": 0.1522073432803154,
      "num_tokens": 40319753.0,
      "step": 19910
    },
    {
      "entropy": 6.312132883071899,
      "epoch": 1.7403652888228611,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004698224688427074,
      "loss": 6.0597,
      "mean_token_accuracy": 0.14743621423840522,
      "num_tokens": 40330729.0,
      "step": 19915
    },
    {
      "entropy": 6.272248029708862,
      "epoch": 1.7408022371755658,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004698068784528442,
      "loss": 5.9815,
      "mean_token_accuracy": 0.14679063037037848,
      "num_tokens": 40340246.0,
      "step": 19920
    },
    {
      "entropy": 6.203789329528808,
      "epoch": 1.7412391855282705,
      "grad_norm": 3.578125,
      "learning_rate": 0.000469791284326483,
      "loss": 5.8408,
      "mean_token_accuracy": 0.15252190977334976,
      "num_tokens": 40350099.0,
      "step": 19925
    },
    {
      "entropy": 6.196573686599732,
      "epoch": 1.7416761338809752,
      "grad_norm": 3.421875,
      "learning_rate": 0.00046977568646392284,
      "loss": 5.9087,
      "mean_token_accuracy": 0.15803659707307816,
      "num_tokens": 40360825.0,
      "step": 19930
    },
    {
      "entropy": 6.18964958190918,
      "epoch": 1.74211308223368,
      "grad_norm": 2.15625,
      "learning_rate": 0.000469760084865463,
      "loss": 5.9706,
      "mean_token_accuracy": 0.13983488604426383,
      "num_tokens": 40371025.0,
      "step": 19935
    },
    {
      "entropy": 6.1954038619995115,
      "epoch": 1.7425500305863846,
      "grad_norm": 2.0,
      "learning_rate": 0.0004697444795314027,
      "loss": 5.8302,
      "mean_token_accuracy": 0.16482615172863008,
      "num_tokens": 40380338.0,
      "step": 19940
    },
    {
      "entropy": 6.198279809951782,
      "epoch": 1.7429869789390895,
      "grad_norm": 3.71875,
      "learning_rate": 0.00046972887046204114,
      "loss": 5.9199,
      "mean_token_accuracy": 0.1512584276497364,
      "num_tokens": 40391814.0,
      "step": 19945
    },
    {
      "entropy": 6.151174163818359,
      "epoch": 1.743423927291794,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00046971325765767793,
      "loss": 5.8962,
      "mean_token_accuracy": 0.14727857261896132,
      "num_tokens": 40402013.0,
      "step": 19950
    },
    {
      "entropy": 6.2095475673675535,
      "epoch": 1.7438608756444989,
      "grad_norm": 3.890625,
      "learning_rate": 0.00046969764111861234,
      "loss": 5.8717,
      "mean_token_accuracy": 0.14947144836187362,
      "num_tokens": 40412205.0,
      "step": 19955
    },
    {
      "entropy": 6.120860958099366,
      "epoch": 1.7442978239972036,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046968202084514405,
      "loss": 5.8533,
      "mean_token_accuracy": 0.16615784615278245,
      "num_tokens": 40423212.0,
      "step": 19960
    },
    {
      "entropy": 6.181463146209717,
      "epoch": 1.7447347723499083,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004696663968375725,
      "loss": 5.8339,
      "mean_token_accuracy": 0.1566979244351387,
      "num_tokens": 40433272.0,
      "step": 19965
    },
    {
      "entropy": 6.191803789138794,
      "epoch": 1.745171720702613,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004696507690961975,
      "loss": 5.9491,
      "mean_token_accuracy": 0.15099540948867798,
      "num_tokens": 40443931.0,
      "step": 19970
    },
    {
      "entropy": 6.181142044067383,
      "epoch": 1.7456086690553176,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004696351376213188,
      "loss": 5.9008,
      "mean_token_accuracy": 0.1551302619278431,
      "num_tokens": 40453008.0,
      "step": 19975
    },
    {
      "entropy": 6.170594263076782,
      "epoch": 1.7460456174080223,
      "grad_norm": 2.28125,
      "learning_rate": 0.000469619502413236,
      "loss": 5.9496,
      "mean_token_accuracy": 0.15811190754175186,
      "num_tokens": 40463081.0,
      "step": 19980
    },
    {
      "entropy": 6.256664037704468,
      "epoch": 1.746482565760727,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004696038634722493,
      "loss": 5.7782,
      "mean_token_accuracy": 0.157869590818882,
      "num_tokens": 40472390.0,
      "step": 19985
    },
    {
      "entropy": 6.207471513748169,
      "epoch": 1.746919514113432,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046958822079865855,
      "loss": 5.9507,
      "mean_token_accuracy": 0.15404710173606873,
      "num_tokens": 40483068.0,
      "step": 19990
    },
    {
      "entropy": 6.258970022201538,
      "epoch": 1.7473564624661364,
      "grad_norm": 3.125,
      "learning_rate": 0.0004695725743927637,
      "loss": 5.9031,
      "mean_token_accuracy": 0.15802823305130004,
      "num_tokens": 40493125.0,
      "step": 19995
    },
    {
      "entropy": 6.211593914031982,
      "epoch": 1.7477934108188413,
      "grad_norm": 2.515625,
      "learning_rate": 0.00046955692425486496,
      "loss": 5.9027,
      "mean_token_accuracy": 0.15155962407588958,
      "num_tokens": 40502618.0,
      "step": 20000
    },
    {
      "entropy": 6.225067281723023,
      "epoch": 1.7482303591715458,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004695412703852624,
      "loss": 5.9186,
      "mean_token_accuracy": 0.15340645164251326,
      "num_tokens": 40512878.0,
      "step": 20005
    },
    {
      "entropy": 6.153727722167969,
      "epoch": 1.7486673075242507,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004695256127842563,
      "loss": 5.8832,
      "mean_token_accuracy": 0.1557719811797142,
      "num_tokens": 40521983.0,
      "step": 20010
    },
    {
      "entropy": 6.200105142593384,
      "epoch": 1.7491042558769554,
      "grad_norm": 3.0625,
      "learning_rate": 0.00046950995145214716,
      "loss": 5.9051,
      "mean_token_accuracy": 0.15498339235782624,
      "num_tokens": 40533126.0,
      "step": 20015
    },
    {
      "entropy": 6.1865214824676515,
      "epoch": 1.74954120422966,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004694942863892352,
      "loss": 5.8698,
      "mean_token_accuracy": 0.15204892605543135,
      "num_tokens": 40543321.0,
      "step": 20020
    },
    {
      "entropy": 6.172781562805175,
      "epoch": 1.7499781525823648,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004694786175958209,
      "loss": 5.9006,
      "mean_token_accuracy": 0.153663020581007,
      "num_tokens": 40554208.0,
      "step": 20025
    },
    {
      "entropy": 6.237401294708252,
      "epoch": 1.7504151009350695,
      "grad_norm": 2.578125,
      "learning_rate": 0.00046946294507220477,
      "loss": 5.8546,
      "mean_token_accuracy": 0.14973213374614716,
      "num_tokens": 40563365.0,
      "step": 20030
    },
    {
      "entropy": 6.101063776016235,
      "epoch": 1.7508520492877742,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004694472688186874,
      "loss": 5.7915,
      "mean_token_accuracy": 0.15756777226924895,
      "num_tokens": 40573765.0,
      "step": 20035
    },
    {
      "entropy": 6.159966516494751,
      "epoch": 1.7512889976404789,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004694315888355697,
      "loss": 5.8967,
      "mean_token_accuracy": 0.15862600207328797,
      "num_tokens": 40583786.0,
      "step": 20040
    },
    {
      "entropy": 6.188342475891114,
      "epoch": 1.7517259459931838,
      "grad_norm": 3.578125,
      "learning_rate": 0.00046941590512315215,
      "loss": 5.9359,
      "mean_token_accuracy": 0.15113548636436464,
      "num_tokens": 40593686.0,
      "step": 20045
    },
    {
      "entropy": 6.216604804992675,
      "epoch": 1.7521628943458882,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004694002176817356,
      "loss": 5.8864,
      "mean_token_accuracy": 0.15338338911533356,
      "num_tokens": 40604326.0,
      "step": 20050
    },
    {
      "entropy": 6.24794020652771,
      "epoch": 1.7525998426985931,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004693845265116211,
      "loss": 5.8643,
      "mean_token_accuracy": 0.1575102612376213,
      "num_tokens": 40613683.0,
      "step": 20055
    },
    {
      "entropy": 6.164059257507324,
      "epoch": 1.7530367910512976,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046936883161310944,
      "loss": 5.8804,
      "mean_token_accuracy": 0.1597310036420822,
      "num_tokens": 40623675.0,
      "step": 20060
    },
    {
      "entropy": 6.204534006118775,
      "epoch": 1.7534737394040025,
      "grad_norm": 3.078125,
      "learning_rate": 0.00046935313298650185,
      "loss": 5.9228,
      "mean_token_accuracy": 0.15348683595657348,
      "num_tokens": 40634006.0,
      "step": 20065
    },
    {
      "entropy": 6.173541784286499,
      "epoch": 1.753910687756707,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046933743063209927,
      "loss": 5.7912,
      "mean_token_accuracy": 0.1667817011475563,
      "num_tokens": 40643175.0,
      "step": 20070
    },
    {
      "entropy": 6.23266568183899,
      "epoch": 1.754347636109412,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046932172455020296,
      "loss": 5.9391,
      "mean_token_accuracy": 0.14552800357341766,
      "num_tokens": 40653447.0,
      "step": 20075
    },
    {
      "entropy": 6.188573980331421,
      "epoch": 1.7547845844621166,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004693060147411141,
      "loss": 5.858,
      "mean_token_accuracy": 0.1610131099820137,
      "num_tokens": 40663499.0,
      "step": 20080
    },
    {
      "entropy": 6.144504737854004,
      "epoch": 1.7552215328148213,
      "grad_norm": 2.0,
      "learning_rate": 0.0004692903012051341,
      "loss": 5.8477,
      "mean_token_accuracy": 0.15875036120414734,
      "num_tokens": 40672943.0,
      "step": 20085
    },
    {
      "entropy": 6.140041780471802,
      "epoch": 1.755658481167526,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004692745839425643,
      "loss": 5.912,
      "mean_token_accuracy": 0.15214159786701204,
      "num_tokens": 40683659.0,
      "step": 20090
    },
    {
      "entropy": 6.232609128952026,
      "epoch": 1.7560954295202307,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004692588629537062,
      "loss": 5.9288,
      "mean_token_accuracy": 0.15229168236255647,
      "num_tokens": 40693829.0,
      "step": 20095
    },
    {
      "entropy": 6.19420018196106,
      "epoch": 1.7565323778729354,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004692431382388613,
      "loss": 5.9262,
      "mean_token_accuracy": 0.15073085129261016,
      "num_tokens": 40704717.0,
      "step": 20100
    },
    {
      "entropy": 6.221127653121949,
      "epoch": 1.75696932622564,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046922740979833135,
      "loss": 5.8946,
      "mean_token_accuracy": 0.15092213451862335,
      "num_tokens": 40714970.0,
      "step": 20105
    },
    {
      "entropy": 6.195417737960815,
      "epoch": 1.757406274578345,
      "grad_norm": 2.25,
      "learning_rate": 0.00046921167763241777,
      "loss": 5.8653,
      "mean_token_accuracy": 0.15528983175754546,
      "num_tokens": 40723664.0,
      "step": 20110
    },
    {
      "entropy": 6.173594856262207,
      "epoch": 1.7578432229310494,
      "grad_norm": 32.5,
      "learning_rate": 0.00046919594174142247,
      "loss": 5.866,
      "mean_token_accuracy": 0.1532951697707176,
      "num_tokens": 40733735.0,
      "step": 20115
    },
    {
      "entropy": 6.171074533462525,
      "epoch": 1.7582801712837544,
      "grad_norm": 3.8125,
      "learning_rate": 0.00046918020212564725,
      "loss": 5.9329,
      "mean_token_accuracy": 0.15307766422629357,
      "num_tokens": 40743699.0,
      "step": 20120
    },
    {
      "entropy": 6.171239042282105,
      "epoch": 1.7587171196364588,
      "grad_norm": 3.453125,
      "learning_rate": 0.000469164458785394,
      "loss": 5.9616,
      "mean_token_accuracy": 0.1491508014500141,
      "num_tokens": 40754553.0,
      "step": 20125
    },
    {
      "entropy": 6.291165733337403,
      "epoch": 1.7591540679891637,
      "grad_norm": 2.515625,
      "learning_rate": 0.00046914871172096473,
      "loss": 5.9144,
      "mean_token_accuracy": 0.15184154063463212,
      "num_tokens": 40764660.0,
      "step": 20130
    },
    {
      "entropy": 6.192643547058106,
      "epoch": 1.7595910163418684,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004691329609326614,
      "loss": 5.7557,
      "mean_token_accuracy": 0.16703584492206575,
      "num_tokens": 40774237.0,
      "step": 20135
    },
    {
      "entropy": 6.131662797927857,
      "epoch": 1.7600279646945731,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004691172064207862,
      "loss": 5.8608,
      "mean_token_accuracy": 0.15884741991758347,
      "num_tokens": 40783608.0,
      "step": 20140
    },
    {
      "entropy": 6.207524967193604,
      "epoch": 1.7604649130472778,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004691014481856412,
      "loss": 5.9355,
      "mean_token_accuracy": 0.1470074698328972,
      "num_tokens": 40793987.0,
      "step": 20145
    },
    {
      "entropy": 6.273020601272583,
      "epoch": 1.7609018613999825,
      "grad_norm": 2.5,
      "learning_rate": 0.00046908568622752874,
      "loss": 5.9375,
      "mean_token_accuracy": 0.15230748802423477,
      "num_tokens": 40805600.0,
      "step": 20150
    },
    {
      "entropy": 6.205696725845337,
      "epoch": 1.7613388097526872,
      "grad_norm": 2.59375,
      "learning_rate": 0.00046906992054675115,
      "loss": 5.8911,
      "mean_token_accuracy": 0.15779782682657242,
      "num_tokens": 40815252.0,
      "step": 20155
    },
    {
      "entropy": 6.134473991394043,
      "epoch": 1.7617757581053919,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004690541511436107,
      "loss": 5.89,
      "mean_token_accuracy": 0.15745916217565536,
      "num_tokens": 40825945.0,
      "step": 20160
    },
    {
      "entropy": 6.199404239654541,
      "epoch": 1.7622127064580968,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004690383780184101,
      "loss": 5.9186,
      "mean_token_accuracy": 0.15090208053588866,
      "num_tokens": 40835811.0,
      "step": 20165
    },
    {
      "entropy": 6.220023775100708,
      "epoch": 1.7626496548108013,
      "grad_norm": 2.359375,
      "learning_rate": 0.00046902260117145163,
      "loss": 5.8922,
      "mean_token_accuracy": 0.15171571224927902,
      "num_tokens": 40845088.0,
      "step": 20170
    },
    {
      "entropy": 6.159562873840332,
      "epoch": 1.7630866031635062,
      "grad_norm": 3.828125,
      "learning_rate": 0.000469006820603038,
      "loss": 5.9492,
      "mean_token_accuracy": 0.15521355867385864,
      "num_tokens": 40854703.0,
      "step": 20175
    },
    {
      "entropy": 6.233977127075195,
      "epoch": 1.7635235515162107,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046899103631347196,
      "loss": 5.8944,
      "mean_token_accuracy": 0.15682866275310517,
      "num_tokens": 40865056.0,
      "step": 20180
    },
    {
      "entropy": 6.15766077041626,
      "epoch": 1.7639604998689156,
      "grad_norm": 2.125,
      "learning_rate": 0.0004689752483030561,
      "loss": 5.7863,
      "mean_token_accuracy": 0.15725178867578507,
      "num_tokens": 40874355.0,
      "step": 20185
    },
    {
      "entropy": 6.183209419250488,
      "epoch": 1.7643974482216203,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004689594565720934,
      "loss": 5.8747,
      "mean_token_accuracy": 0.15321042835712434,
      "num_tokens": 40884329.0,
      "step": 20190
    },
    {
      "entropy": 6.181976652145385,
      "epoch": 1.764834396574325,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004689436611208867,
      "loss": 5.8982,
      "mean_token_accuracy": 0.15286797136068345,
      "num_tokens": 40894702.0,
      "step": 20195
    },
    {
      "entropy": 6.216638565063477,
      "epoch": 1.7652713449270296,
      "grad_norm": 3.578125,
      "learning_rate": 0.00046892786194973897,
      "loss": 5.8699,
      "mean_token_accuracy": 0.14924490749835967,
      "num_tokens": 40905467.0,
      "step": 20200
    },
    {
      "entropy": 6.236700010299683,
      "epoch": 1.7657082932797343,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004689120590589532,
      "loss": 5.9316,
      "mean_token_accuracy": 0.15253936201334,
      "num_tokens": 40915915.0,
      "step": 20205
    },
    {
      "entropy": 6.170786142349243,
      "epoch": 1.766145241632439,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004688962524488325,
      "loss": 5.8703,
      "mean_token_accuracy": 0.1565911740064621,
      "num_tokens": 40925941.0,
      "step": 20210
    },
    {
      "entropy": 6.210948181152344,
      "epoch": 1.7665821899851437,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004688804421196801,
      "loss": 5.9233,
      "mean_token_accuracy": 0.15469201803207397,
      "num_tokens": 40935831.0,
      "step": 20215
    },
    {
      "entropy": 6.157115077972412,
      "epoch": 1.7670191383378486,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046886462807179926,
      "loss": 5.8587,
      "mean_token_accuracy": 0.1605278193950653,
      "num_tokens": 40945341.0,
      "step": 20220
    },
    {
      "entropy": 6.16046552658081,
      "epoch": 1.767456086690553,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004688488103054931,
      "loss": 5.9209,
      "mean_token_accuracy": 0.15260131806135177,
      "num_tokens": 40955759.0,
      "step": 20225
    },
    {
      "entropy": 6.215676593780517,
      "epoch": 1.767893035043258,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004688329888210654,
      "loss": 5.8338,
      "mean_token_accuracy": 0.15017721503973008,
      "num_tokens": 40966184.0,
      "step": 20230
    },
    {
      "entropy": 6.2154491424560545,
      "epoch": 1.7683299833959625,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004688171636188194,
      "loss": 5.8838,
      "mean_token_accuracy": 0.1527157962322235,
      "num_tokens": 40975797.0,
      "step": 20235
    },
    {
      "entropy": 6.233051824569702,
      "epoch": 1.7687669317486674,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004688013346990585,
      "loss": 5.8788,
      "mean_token_accuracy": 0.1540012314915657,
      "num_tokens": 40985830.0,
      "step": 20240
    },
    {
      "entropy": 6.136021661758423,
      "epoch": 1.7692038801013719,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046878550206208656,
      "loss": 5.7976,
      "mean_token_accuracy": 0.15833680480718612,
      "num_tokens": 40995692.0,
      "step": 20245
    },
    {
      "entropy": 6.172507667541504,
      "epoch": 1.7696408284540768,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004687696657082071,
      "loss": 5.9124,
      "mean_token_accuracy": 0.15045669823884963,
      "num_tokens": 41005506.0,
      "step": 20250
    },
    {
      "entropy": 6.176908349990844,
      "epoch": 1.7700777768067815,
      "grad_norm": 3.1875,
      "learning_rate": 0.00046875382563772396,
      "loss": 5.8168,
      "mean_token_accuracy": 0.15885351300239564,
      "num_tokens": 41014944.0,
      "step": 20255
    },
    {
      "entropy": 6.212207698822022,
      "epoch": 1.7705147251594862,
      "grad_norm": 2.59375,
      "learning_rate": 0.00046873798185094093,
      "loss": 5.9706,
      "mean_token_accuracy": 0.14580247551202774,
      "num_tokens": 41025424.0,
      "step": 20260
    },
    {
      "entropy": 6.138521432876587,
      "epoch": 1.7709516735121908,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004687221343481618,
      "loss": 5.8188,
      "mean_token_accuracy": 0.16110175400972365,
      "num_tokens": 41035817.0,
      "step": 20265
    },
    {
      "entropy": 6.166412258148194,
      "epoch": 1.7713886218648955,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046870628312969064,
      "loss": 5.8683,
      "mean_token_accuracy": 0.16103764176368712,
      "num_tokens": 41045730.0,
      "step": 20270
    },
    {
      "entropy": 6.284985733032227,
      "epoch": 1.7718255702176002,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004686904281958315,
      "loss": 5.9473,
      "mean_token_accuracy": 0.14517279118299484,
      "num_tokens": 41057921.0,
      "step": 20275
    },
    {
      "entropy": 6.2457959175109865,
      "epoch": 1.772262518570305,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046867456954688847,
      "loss": 5.9569,
      "mean_token_accuracy": 0.15388963520526885,
      "num_tokens": 41068879.0,
      "step": 20280
    },
    {
      "entropy": 6.216877222061157,
      "epoch": 1.7726994669230098,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004686587071831657,
      "loss": 5.9856,
      "mean_token_accuracy": 0.1487608090043068,
      "num_tokens": 41079443.0,
      "step": 20285
    },
    {
      "entropy": 6.209468650817871,
      "epoch": 1.7731364152757143,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004686428411049675,
      "loss": 5.9093,
      "mean_token_accuracy": 0.15278039276599883,
      "num_tokens": 41089932.0,
      "step": 20290
    },
    {
      "entropy": 6.2437201023101805,
      "epoch": 1.7735733636284192,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004686269713125981,
      "loss": 5.9382,
      "mean_token_accuracy": 0.15149659812450408,
      "num_tokens": 41100412.0,
      "step": 20295
    },
    {
      "entropy": 6.204418516159057,
      "epoch": 1.7740103119811237,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004686110978063619,
      "loss": 5.8946,
      "mean_token_accuracy": 0.15260340720415116,
      "num_tokens": 41110830.0,
      "step": 20300
    },
    {
      "entropy": 6.195499420166016,
      "epoch": 1.7744472603338286,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004685952205865634,
      "loss": 5.8643,
      "mean_token_accuracy": 0.15423081666231156,
      "num_tokens": 41120662.0,
      "step": 20305
    },
    {
      "entropy": 6.180319356918335,
      "epoch": 1.7748842086865333,
      "grad_norm": 2.984375,
      "learning_rate": 0.00046857933965350714,
      "loss": 5.8431,
      "mean_token_accuracy": 0.16294772773981095,
      "num_tokens": 41130523.0,
      "step": 20310
    },
    {
      "entropy": 6.255850791931152,
      "epoch": 1.775321157039238,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004685634550074977,
      "loss": 5.9713,
      "mean_token_accuracy": 0.15220875144004822,
      "num_tokens": 41141659.0,
      "step": 20315
    },
    {
      "entropy": 6.211370658874512,
      "epoch": 1.7757581053919427,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004685475666488398,
      "loss": 5.9948,
      "mean_token_accuracy": 0.1442561961710453,
      "num_tokens": 41151038.0,
      "step": 20320
    },
    {
      "entropy": 6.191706848144531,
      "epoch": 1.7761950537446474,
      "grad_norm": 3.40625,
      "learning_rate": 0.00046853167457783815,
      "loss": 5.9379,
      "mean_token_accuracy": 0.15674971789121628,
      "num_tokens": 41161888.0,
      "step": 20325
    },
    {
      "entropy": 6.32315411567688,
      "epoch": 1.776632002097352,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004685157787947975,
      "loss": 5.9562,
      "mean_token_accuracy": 0.1485387720167637,
      "num_tokens": 41172722.0,
      "step": 20330
    },
    {
      "entropy": 6.194577217102051,
      "epoch": 1.7770689504500568,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004684998793000229,
      "loss": 5.8274,
      "mean_token_accuracy": 0.15719235241413115,
      "num_tokens": 41181716.0,
      "step": 20335
    },
    {
      "entropy": 6.217396783828735,
      "epoch": 1.7775058988027617,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004684839760938193,
      "loss": 5.9437,
      "mean_token_accuracy": 0.14574080407619477,
      "num_tokens": 41192903.0,
      "step": 20340
    },
    {
      "entropy": 6.18211407661438,
      "epoch": 1.7779428471554661,
      "grad_norm": 2.6875,
      "learning_rate": 0.00046846806917649145,
      "loss": 5.907,
      "mean_token_accuracy": 0.1518358826637268,
      "num_tokens": 41203283.0,
      "step": 20345
    },
    {
      "entropy": 6.243850326538086,
      "epoch": 1.778379795508171,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004684521585483447,
      "loss": 5.9006,
      "mean_token_accuracy": 0.15490635484457016,
      "num_tokens": 41213615.0,
      "step": 20350
    },
    {
      "entropy": 6.174698162078857,
      "epoch": 1.7788167438608755,
      "grad_norm": 1.953125,
      "learning_rate": 0.00046843624420968435,
      "loss": 5.8104,
      "mean_token_accuracy": 0.16570706367492677,
      "num_tokens": 41222705.0,
      "step": 20355
    },
    {
      "entropy": 6.131819057464599,
      "epoch": 1.7792536922135804,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004684203261608153,
      "loss": 5.8596,
      "mean_token_accuracy": 0.1618562400341034,
      "num_tokens": 41233201.0,
      "step": 20360
    },
    {
      "entropy": 6.264989137649536,
      "epoch": 1.7796906405662851,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004684044044020431,
      "loss": 5.9036,
      "mean_token_accuracy": 0.15120690912008286,
      "num_tokens": 41243088.0,
      "step": 20365
    },
    {
      "entropy": 6.258667993545532,
      "epoch": 1.7801275889189898,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004683884789336731,
      "loss": 5.9109,
      "mean_token_accuracy": 0.1497899189591408,
      "num_tokens": 41253009.0,
      "step": 20370
    },
    {
      "entropy": 6.128800868988037,
      "epoch": 1.7805645372716945,
      "grad_norm": 2.359375,
      "learning_rate": 0.00046837254975601073,
      "loss": 5.923,
      "mean_token_accuracy": 0.16066857427358627,
      "num_tokens": 41262209.0,
      "step": 20375
    },
    {
      "entropy": 6.2701271057128904,
      "epoch": 1.7810014856243992,
      "grad_norm": 2.78125,
      "learning_rate": 0.00046835661686936154,
      "loss": 5.9214,
      "mean_token_accuracy": 0.15663807094097137,
      "num_tokens": 41272762.0,
      "step": 20380
    },
    {
      "entropy": 6.259342908859253,
      "epoch": 1.7814384339771039,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046834068027403115,
      "loss": 5.9783,
      "mean_token_accuracy": 0.14475687444210053,
      "num_tokens": 41284306.0,
      "step": 20385
    },
    {
      "entropy": 6.22597451210022,
      "epoch": 1.7818753823298086,
      "grad_norm": 2.8125,
      "learning_rate": 0.00046832473997032514,
      "loss": 5.928,
      "mean_token_accuracy": 0.1511193834245205,
      "num_tokens": 41295211.0,
      "step": 20390
    },
    {
      "entropy": 6.164445400238037,
      "epoch": 1.7823123306825135,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004683087959585493,
      "loss": 5.7743,
      "mean_token_accuracy": 0.16256022453308105,
      "num_tokens": 41303773.0,
      "step": 20395
    },
    {
      "entropy": 6.19306902885437,
      "epoch": 1.782749279035218,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004682928482390095,
      "loss": 5.8731,
      "mean_token_accuracy": 0.1608504205942154,
      "num_tokens": 41313536.0,
      "step": 20400
    },
    {
      "entropy": 6.220166397094727,
      "epoch": 1.7831862273879229,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046827689681201164,
      "loss": 5.9289,
      "mean_token_accuracy": 0.15338435918092727,
      "num_tokens": 41323831.0,
      "step": 20405
    },
    {
      "entropy": 6.191534090042114,
      "epoch": 1.7836231757406273,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004682609416778615,
      "loss": 5.8937,
      "mean_token_accuracy": 0.15655421763658522,
      "num_tokens": 41334295.0,
      "step": 20410
    },
    {
      "entropy": 6.229070997238159,
      "epoch": 1.7840601240933323,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004682449828368653,
      "loss": 5.9071,
      "mean_token_accuracy": 0.15033280700445176,
      "num_tokens": 41344672.0,
      "step": 20415
    },
    {
      "entropy": 6.261100959777832,
      "epoch": 1.7844970724460367,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004682290202893291,
      "loss": 5.949,
      "mean_token_accuracy": 0.14750506430864335,
      "num_tokens": 41355108.0,
      "step": 20420
    },
    {
      "entropy": 6.325549125671387,
      "epoch": 1.7849340207987416,
      "grad_norm": 2.71875,
      "learning_rate": 0.00046821305403555905,
      "loss": 6.0152,
      "mean_token_accuracy": 0.14685259908437728,
      "num_tokens": 41366277.0,
      "step": 20425
    },
    {
      "entropy": 6.195319366455078,
      "epoch": 1.7853709691514463,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046819708407586135,
      "loss": 5.8706,
      "mean_token_accuracy": 0.1600157931447029,
      "num_tokens": 41374757.0,
      "step": 20430
    },
    {
      "entropy": 6.22753415107727,
      "epoch": 1.785807917504151,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004681811104105423,
      "loss": 5.8703,
      "mean_token_accuracy": 0.16357674598693847,
      "num_tokens": 41384883.0,
      "step": 20435
    },
    {
      "entropy": 6.211219310760498,
      "epoch": 1.7862448658568557,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004681651330399084,
      "loss": 5.911,
      "mean_token_accuracy": 0.157825268805027,
      "num_tokens": 41396492.0,
      "step": 20440
    },
    {
      "entropy": 6.155817604064941,
      "epoch": 1.7866818142095604,
      "grad_norm": 2.0,
      "learning_rate": 0.000468149151964266,
      "loss": 5.865,
      "mean_token_accuracy": 0.1570971503853798,
      "num_tokens": 41406217.0,
      "step": 20445
    },
    {
      "entropy": 6.163604354858398,
      "epoch": 1.787118762562265,
      "grad_norm": 2.375,
      "learning_rate": 0.0004681331671839217,
      "loss": 5.7997,
      "mean_token_accuracy": 0.16388321816921234,
      "num_tokens": 41415895.0,
      "step": 20450
    },
    {
      "entropy": 6.223845720291138,
      "epoch": 1.7875557109149698,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046811717869918196,
      "loss": 5.9436,
      "mean_token_accuracy": 0.14631979614496232,
      "num_tokens": 41426021.0,
      "step": 20455
    },
    {
      "entropy": 6.241466426849366,
      "epoch": 1.7879926592676747,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004681011865103536,
      "loss": 5.8813,
      "mean_token_accuracy": 0.1546391114592552,
      "num_tokens": 41434656.0,
      "step": 20460
    },
    {
      "entropy": 6.173611879348755,
      "epoch": 1.7884296076203792,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046808519061774324,
      "loss": 5.9036,
      "mean_token_accuracy": 0.1543724998831749,
      "num_tokens": 41443940.0,
      "step": 20465
    },
    {
      "entropy": 6.213986539840699,
      "epoch": 1.788866555973084,
      "grad_norm": 2.65625,
      "learning_rate": 0.00046806919102165785,
      "loss": 5.9547,
      "mean_token_accuracy": 0.154327392578125,
      "num_tokens": 41454907.0,
      "step": 20470
    },
    {
      "entropy": 6.210531520843506,
      "epoch": 1.7893035043257886,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004680531877224042,
      "loss": 5.8232,
      "mean_token_accuracy": 0.15920254588127136,
      "num_tokens": 41464248.0,
      "step": 20475
    },
    {
      "entropy": 6.122627258300781,
      "epoch": 1.7897404526784935,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004680371807202892,
      "loss": 5.8428,
      "mean_token_accuracy": 0.15965242832899093,
      "num_tokens": 41475192.0,
      "step": 20480
    },
    {
      "entropy": 6.251329851150513,
      "epoch": 1.7901774010311982,
      "grad_norm": 3.171875,
      "learning_rate": 0.00046802117001561996,
      "loss": 5.9396,
      "mean_token_accuracy": 0.1448342964053154,
      "num_tokens": 41484753.0,
      "step": 20485
    },
    {
      "entropy": 6.259180021286011,
      "epoch": 1.7906143493839028,
      "grad_norm": 3.015625,
      "learning_rate": 0.00046800515560870346,
      "loss": 6.008,
      "mean_token_accuracy": 0.14531339704990387,
      "num_tokens": 41495885.0,
      "step": 20490
    },
    {
      "entropy": 6.363275718688965,
      "epoch": 1.7910512977366075,
      "grad_norm": 1.90625,
      "learning_rate": 0.00046798913749984694,
      "loss": 5.918,
      "mean_token_accuracy": 0.15477714389562608,
      "num_tokens": 41506330.0,
      "step": 20495
    },
    {
      "entropy": 6.1845911026000975,
      "epoch": 1.7914882460893122,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004679731156893577,
      "loss": 5.8978,
      "mean_token_accuracy": 0.15803170800209046,
      "num_tokens": 41515850.0,
      "step": 20500
    },
    {
      "entropy": 6.13728084564209,
      "epoch": 1.791925194442017,
      "grad_norm": 3.78125,
      "learning_rate": 0.000467957090177543,
      "loss": 5.8964,
      "mean_token_accuracy": 0.1511285647749901,
      "num_tokens": 41525910.0,
      "step": 20505
    },
    {
      "entropy": 6.247721862792969,
      "epoch": 1.7923621427947216,
      "grad_norm": 2.609375,
      "learning_rate": 0.00046794106096471015,
      "loss": 5.8155,
      "mean_token_accuracy": 0.15594040751457214,
      "num_tokens": 41536417.0,
      "step": 20510
    },
    {
      "entropy": 6.202562475204468,
      "epoch": 1.7927990911474265,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004679250280511667,
      "loss": 5.8924,
      "mean_token_accuracy": 0.15220615565776824,
      "num_tokens": 41547215.0,
      "step": 20515
    },
    {
      "entropy": 6.194731140136719,
      "epoch": 1.793236039500131,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004679089914372201,
      "loss": 5.91,
      "mean_token_accuracy": 0.15424472987651824,
      "num_tokens": 41556036.0,
      "step": 20520
    },
    {
      "entropy": 6.191279792785645,
      "epoch": 1.793672987852836,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046789295112317796,
      "loss": 5.9045,
      "mean_token_accuracy": 0.1552931159734726,
      "num_tokens": 41565534.0,
      "step": 20525
    },
    {
      "entropy": 6.164443397521973,
      "epoch": 1.7941099362055404,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004678769071093479,
      "loss": 5.8767,
      "mean_token_accuracy": 0.15233133733272552,
      "num_tokens": 41576150.0,
      "step": 20530
    },
    {
      "entropy": 6.162917995452881,
      "epoch": 1.7945468845582453,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00046786085939603785,
      "loss": 5.8414,
      "mean_token_accuracy": 0.1622470125555992,
      "num_tokens": 41585248.0,
      "step": 20535
    },
    {
      "entropy": 6.185472011566162,
      "epoch": 1.7949838329109498,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004678448079835552,
      "loss": 5.9003,
      "mean_token_accuracy": 0.15317177325487136,
      "num_tokens": 41595215.0,
      "step": 20540
    },
    {
      "entropy": 6.258771371841431,
      "epoch": 1.7954207812636547,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004678287528722083,
      "loss": 5.9452,
      "mean_token_accuracy": 0.1520175814628601,
      "num_tokens": 41605268.0,
      "step": 20545
    },
    {
      "entropy": 6.208068990707398,
      "epoch": 1.7958577296163594,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004678126940623048,
      "loss": 5.7967,
      "mean_token_accuracy": 0.1634898766875267,
      "num_tokens": 41614554.0,
      "step": 20550
    },
    {
      "entropy": 6.152574300765991,
      "epoch": 1.796294677969064,
      "grad_norm": 4.65625,
      "learning_rate": 0.00046779663155415287,
      "loss": 5.9435,
      "mean_token_accuracy": 0.1511387422680855,
      "num_tokens": 41624839.0,
      "step": 20555
    },
    {
      "entropy": 6.114175271987915,
      "epoch": 1.7967316263217687,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004677805653480604,
      "loss": 5.7773,
      "mean_token_accuracy": 0.16911624297499656,
      "num_tokens": 41634577.0,
      "step": 20560
    },
    {
      "entropy": 6.23898720741272,
      "epoch": 1.7971685746744734,
      "grad_norm": 2.78125,
      "learning_rate": 0.00046776449544433575,
      "loss": 5.927,
      "mean_token_accuracy": 0.14993715286254883,
      "num_tokens": 41645213.0,
      "step": 20565
    },
    {
      "entropy": 6.14775128364563,
      "epoch": 1.7976055230271784,
      "grad_norm": 2.25,
      "learning_rate": 0.000467748421843287,
      "loss": 5.8117,
      "mean_token_accuracy": 0.16186597645282746,
      "num_tokens": 41655160.0,
      "step": 20570
    },
    {
      "entropy": 6.2214559555053714,
      "epoch": 1.7980424713798828,
      "grad_norm": 1.96875,
      "learning_rate": 0.00046773234454522265,
      "loss": 5.8889,
      "mean_token_accuracy": 0.14825257956981658,
      "num_tokens": 41665861.0,
      "step": 20575
    },
    {
      "entropy": 6.241279983520508,
      "epoch": 1.7984794197325877,
      "grad_norm": 1.59375,
      "learning_rate": 0.00046771626355045085,
      "loss": 5.893,
      "mean_token_accuracy": 0.15878873318433762,
      "num_tokens": 41676699.0,
      "step": 20580
    },
    {
      "entropy": 6.232455396652222,
      "epoch": 1.7989163680852922,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004677001788592801,
      "loss": 5.9342,
      "mean_token_accuracy": 0.1530511274933815,
      "num_tokens": 41688162.0,
      "step": 20585
    },
    {
      "entropy": 6.211602830886841,
      "epoch": 1.7993533164379971,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000467684090472019,
      "loss": 5.9342,
      "mean_token_accuracy": 0.1522899881005287,
      "num_tokens": 41698088.0,
      "step": 20590
    },
    {
      "entropy": 6.221312570571899,
      "epoch": 1.7997902647907016,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004676679983889761,
      "loss": 5.8572,
      "mean_token_accuracy": 0.16074490845203399,
      "num_tokens": 41707446.0,
      "step": 20595
    },
    {
      "entropy": 6.265927076339722,
      "epoch": 1.8002272131434065,
      "grad_norm": 2.796875,
      "learning_rate": 0.00046765190261046,
      "loss": 5.9315,
      "mean_token_accuracy": 0.1508984923362732,
      "num_tokens": 41716388.0,
      "step": 20600
    },
    {
      "entropy": 6.188093423843384,
      "epoch": 1.8006641614961112,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004676358031367795,
      "loss": 5.9506,
      "mean_token_accuracy": 0.1561223953962326,
      "num_tokens": 41726328.0,
      "step": 20605
    },
    {
      "entropy": 6.231557178497314,
      "epoch": 1.8011011098488159,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046761969996824326,
      "loss": 5.8843,
      "mean_token_accuracy": 0.15412451922893525,
      "num_tokens": 41736128.0,
      "step": 20610
    },
    {
      "entropy": 6.227087688446045,
      "epoch": 1.8015380582015206,
      "grad_norm": 5.6875,
      "learning_rate": 0.0004676035931051604,
      "loss": 5.8378,
      "mean_token_accuracy": 0.1573920339345932,
      "num_tokens": 41746316.0,
      "step": 20615
    },
    {
      "entropy": 6.185740613937378,
      "epoch": 1.8019750065542253,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046758748254783957,
      "loss": 5.7912,
      "mean_token_accuracy": 0.16243599951267243,
      "num_tokens": 41756015.0,
      "step": 20620
    },
    {
      "entropy": 6.177681636810303,
      "epoch": 1.80241195490693,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046757136829658996,
      "loss": 5.8256,
      "mean_token_accuracy": 0.15692303329706192,
      "num_tokens": 41766360.0,
      "step": 20625
    },
    {
      "entropy": 6.2350109100341795,
      "epoch": 1.8028489032596346,
      "grad_norm": 2.828125,
      "learning_rate": 0.00046755525035172056,
      "loss": 5.9412,
      "mean_token_accuracy": 0.15095627382397653,
      "num_tokens": 41776640.0,
      "step": 20630
    },
    {
      "entropy": 6.163138198852539,
      "epoch": 1.8032858516123396,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004675391287135406,
      "loss": 5.9197,
      "mean_token_accuracy": 0.1554860144853592,
      "num_tokens": 41787329.0,
      "step": 20635
    },
    {
      "entropy": 6.223035478591919,
      "epoch": 1.803722799965044,
      "grad_norm": 4.09375,
      "learning_rate": 0.00046752300338235924,
      "loss": 5.9215,
      "mean_token_accuracy": 0.15009732097387313,
      "num_tokens": 41797890.0,
      "step": 20640
    },
    {
      "entropy": 6.204122924804688,
      "epoch": 1.804159748317749,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004675068743584858,
      "loss": 5.9187,
      "mean_token_accuracy": 0.15073023214936257,
      "num_tokens": 41808845.0,
      "step": 20645
    },
    {
      "entropy": 6.254212284088135,
      "epoch": 1.8045966966704534,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004674907416422297,
      "loss": 5.9115,
      "mean_token_accuracy": 0.1554583117365837,
      "num_tokens": 41819499.0,
      "step": 20650
    },
    {
      "entropy": 6.142949390411377,
      "epoch": 1.8050336450231583,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004674746052339002,
      "loss": 5.9069,
      "mean_token_accuracy": 0.15577002465724946,
      "num_tokens": 41827998.0,
      "step": 20655
    },
    {
      "entropy": 6.236325693130493,
      "epoch": 1.805470593375863,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046745846513380704,
      "loss": 5.9504,
      "mean_token_accuracy": 0.1496094986796379,
      "num_tokens": 41839610.0,
      "step": 20660
    },
    {
      "entropy": 6.319034576416016,
      "epoch": 1.8059075417285677,
      "grad_norm": 1.921875,
      "learning_rate": 0.00046744232134225954,
      "loss": 5.8716,
      "mean_token_accuracy": 0.1535409763455391,
      "num_tokens": 41850111.0,
      "step": 20665
    },
    {
      "entropy": 6.162566137313843,
      "epoch": 1.8063444900812724,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004674261738595676,
      "loss": 5.8537,
      "mean_token_accuracy": 0.15402898639440538,
      "num_tokens": 41859843.0,
      "step": 20670
    },
    {
      "entropy": 6.198745584487915,
      "epoch": 1.806781438433977,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046741002268604073,
      "loss": 5.8509,
      "mean_token_accuracy": 0.15794793963432313,
      "num_tokens": 41868864.0,
      "step": 20675
    },
    {
      "entropy": 6.191723823547363,
      "epoch": 1.8072183867866818,
      "grad_norm": 2.46875,
      "learning_rate": 0.00046739386782198883,
      "loss": 5.899,
      "mean_token_accuracy": 0.15280616730451585,
      "num_tokens": 41878274.0,
      "step": 20680
    },
    {
      "entropy": 6.17008376121521,
      "epoch": 1.8076553351393865,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046737770926772175,
      "loss": 5.8613,
      "mean_token_accuracy": 0.1534583032131195,
      "num_tokens": 41888005.0,
      "step": 20685
    },
    {
      "entropy": 6.249389743804931,
      "epoch": 1.8080922834920914,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004673615470235494,
      "loss": 5.9692,
      "mean_token_accuracy": 0.1532702386379242,
      "num_tokens": 41897415.0,
      "step": 20690
    },
    {
      "entropy": 6.189446926116943,
      "epoch": 1.8085292318447959,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004673453810897817,
      "loss": 5.8423,
      "mean_token_accuracy": 0.15866279155015944,
      "num_tokens": 41907021.0,
      "step": 20695
    },
    {
      "entropy": 6.156405115127564,
      "epoch": 1.8089661801975008,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004673292114667289,
      "loss": 5.9448,
      "mean_token_accuracy": 0.14989005029201508,
      "num_tokens": 41918358.0,
      "step": 20700
    },
    {
      "entropy": 6.272113704681397,
      "epoch": 1.8094031285502052,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046731303815470104,
      "loss": 5.9524,
      "mean_token_accuracy": 0.1397172160446644,
      "num_tokens": 41929311.0,
      "step": 20705
    },
    {
      "entropy": 6.239481449127197,
      "epoch": 1.8098400769029102,
      "grad_norm": 3.65625,
      "learning_rate": 0.00046729686115400826,
      "loss": 5.8964,
      "mean_token_accuracy": 0.1543360933661461,
      "num_tokens": 41939623.0,
      "step": 20710
    },
    {
      "entropy": 6.18283486366272,
      "epoch": 1.8102770252556146,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004672806804649611,
      "loss": 5.8725,
      "mean_token_accuracy": 0.15301173329353332,
      "num_tokens": 41949650.0,
      "step": 20715
    },
    {
      "entropy": 6.226338148117065,
      "epoch": 1.8107139736083195,
      "grad_norm": 2.359375,
      "learning_rate": 0.00046726449608786957,
      "loss": 5.8456,
      "mean_token_accuracy": 0.15791960656642914,
      "num_tokens": 41960262.0,
      "step": 20720
    },
    {
      "entropy": 6.172950744628906,
      "epoch": 1.8111509219610242,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004672483080230443,
      "loss": 5.8595,
      "mean_token_accuracy": 0.1489389196038246,
      "num_tokens": 41969874.0,
      "step": 20725
    },
    {
      "entropy": 6.181381273269653,
      "epoch": 1.811587870313729,
      "grad_norm": 1.90625,
      "learning_rate": 0.00046723211627079577,
      "loss": 5.88,
      "mean_token_accuracy": 0.15492943078279495,
      "num_tokens": 41978810.0,
      "step": 20730
    },
    {
      "entropy": 6.183534431457519,
      "epoch": 1.8120248186664336,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004672159208314344,
      "loss": 5.8331,
      "mean_token_accuracy": 0.15860722213983536,
      "num_tokens": 41988785.0,
      "step": 20735
    },
    {
      "entropy": 6.122798824310303,
      "epoch": 1.8124617670191383,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004671997217052711,
      "loss": 5.9007,
      "mean_token_accuracy": 0.15991397202014923,
      "num_tokens": 41998698.0,
      "step": 20740
    },
    {
      "entropy": 6.159543704986572,
      "epoch": 1.812898715371843,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004671835188926164,
      "loss": 5.8586,
      "mean_token_accuracy": 0.15961575359106064,
      "num_tokens": 42009082.0,
      "step": 20745
    },
    {
      "entropy": 6.247245264053345,
      "epoch": 1.8133356637245477,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004671673123937811,
      "loss": 5.877,
      "mean_token_accuracy": 0.15273915380239486,
      "num_tokens": 42019960.0,
      "step": 20750
    },
    {
      "entropy": 6.163627433776855,
      "epoch": 1.8137726120772526,
      "grad_norm": 2.015625,
      "learning_rate": 0.000467151102209076,
      "loss": 5.8066,
      "mean_token_accuracy": 0.16019020825624466,
      "num_tokens": 42028799.0,
      "step": 20755
    },
    {
      "entropy": 6.182877826690674,
      "epoch": 1.814209560429957,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046713488833881214,
      "loss": 5.8775,
      "mean_token_accuracy": 0.1495542325079441,
      "num_tokens": 42038967.0,
      "step": 20760
    },
    {
      "entropy": 6.200984859466553,
      "epoch": 1.814646508782662,
      "grad_norm": 3.546875,
      "learning_rate": 0.00046711867078330047,
      "loss": 5.8557,
      "mean_token_accuracy": 0.15792694464325904,
      "num_tokens": 42049982.0,
      "step": 20765
    },
    {
      "entropy": 6.194500732421875,
      "epoch": 1.8150834571353665,
      "grad_norm": 16.75,
      "learning_rate": 0.00046710244954285184,
      "loss": 5.9894,
      "mean_token_accuracy": 0.1522536739706993,
      "num_tokens": 42059835.0,
      "step": 20770
    },
    {
      "entropy": 6.149795770645142,
      "epoch": 1.8155204054880714,
      "grad_norm": 2.625,
      "learning_rate": 0.00046708622461777774,
      "loss": 5.776,
      "mean_token_accuracy": 0.1651137501001358,
      "num_tokens": 42069339.0,
      "step": 20775
    },
    {
      "entropy": 6.182058191299438,
      "epoch": 1.815957353840776,
      "grad_norm": 3.25,
      "learning_rate": 0.0004670699960083892,
      "loss": 5.8161,
      "mean_token_accuracy": 0.15455052107572556,
      "num_tokens": 42080618.0,
      "step": 20780
    },
    {
      "entropy": 6.2071380615234375,
      "epoch": 1.8163943021934807,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004670537637149974,
      "loss": 5.9671,
      "mean_token_accuracy": 0.14395468831062316,
      "num_tokens": 42091458.0,
      "step": 20785
    },
    {
      "entropy": 6.1999085426330565,
      "epoch": 1.8168312505461854,
      "grad_norm": 1.953125,
      "learning_rate": 0.00046703752773791387,
      "loss": 5.9706,
      "mean_token_accuracy": 0.14660705924034118,
      "num_tokens": 42102515.0,
      "step": 20790
    },
    {
      "entropy": 6.232903528213501,
      "epoch": 1.8172681988988901,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004670212880774499,
      "loss": 5.8968,
      "mean_token_accuracy": 0.14915742576122284,
      "num_tokens": 42113853.0,
      "step": 20795
    },
    {
      "entropy": 6.20604100227356,
      "epoch": 1.8177051472515948,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000467005044733917,
      "loss": 5.9122,
      "mean_token_accuracy": 0.14887262061238288,
      "num_tokens": 42123101.0,
      "step": 20800
    },
    {
      "entropy": 6.20085973739624,
      "epoch": 1.8181420956042995,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004669887977076267,
      "loss": 5.91,
      "mean_token_accuracy": 0.15145349726080895,
      "num_tokens": 42133051.0,
      "step": 20805
    },
    {
      "entropy": 6.199611520767212,
      "epoch": 1.8185790439570044,
      "grad_norm": 2.78125,
      "learning_rate": 0.00046697254699889064,
      "loss": 5.8992,
      "mean_token_accuracy": 0.15625961720943451,
      "num_tokens": 42142705.0,
      "step": 20810
    },
    {
      "entropy": 6.20850887298584,
      "epoch": 1.819015992309709,
      "grad_norm": 5.59375,
      "learning_rate": 0.00046695629260802064,
      "loss": 5.8781,
      "mean_token_accuracy": 0.15668742209672928,
      "num_tokens": 42152399.0,
      "step": 20815
    },
    {
      "entropy": 6.217681741714477,
      "epoch": 1.8194529406624138,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004669400345353283,
      "loss": 5.9117,
      "mean_token_accuracy": 0.1521599367260933,
      "num_tokens": 42162651.0,
      "step": 20820
    },
    {
      "entropy": 6.109353733062744,
      "epoch": 1.8198898890151183,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004669237727811255,
      "loss": 5.7877,
      "mean_token_accuracy": 0.16020817160606385,
      "num_tokens": 42172595.0,
      "step": 20825
    },
    {
      "entropy": 6.205196666717529,
      "epoch": 1.8203268373678232,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004669075073457242,
      "loss": 5.8816,
      "mean_token_accuracy": 0.1537886753678322,
      "num_tokens": 42182631.0,
      "step": 20830
    },
    {
      "entropy": 6.185060501098633,
      "epoch": 1.8207637857205279,
      "grad_norm": 2.359375,
      "learning_rate": 0.00046689123822943643,
      "loss": 5.8667,
      "mean_token_accuracy": 0.15246639549732208,
      "num_tokens": 42192743.0,
      "step": 20835
    },
    {
      "entropy": 6.198953914642334,
      "epoch": 1.8212007340732326,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004668749654325741,
      "loss": 5.8726,
      "mean_token_accuracy": 0.15402436405420303,
      "num_tokens": 42203090.0,
      "step": 20840
    },
    {
      "entropy": 6.195448923110962,
      "epoch": 1.8216376824259373,
      "grad_norm": 2.375,
      "learning_rate": 0.0004668586889554493,
      "loss": 5.9487,
      "mean_token_accuracy": 0.15168781578540802,
      "num_tokens": 42213341.0,
      "step": 20845
    },
    {
      "entropy": 6.204003047943115,
      "epoch": 1.822074630778642,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004668424087983744,
      "loss": 5.9475,
      "mean_token_accuracy": 0.15563365519046785,
      "num_tokens": 42222862.0,
      "step": 20850
    },
    {
      "entropy": 6.234261226654053,
      "epoch": 1.8225115791313466,
      "grad_norm": 3.0625,
      "learning_rate": 0.00046682612496166156,
      "loss": 5.9405,
      "mean_token_accuracy": 0.14986907169222832,
      "num_tokens": 42234045.0,
      "step": 20855
    },
    {
      "entropy": 6.3007261753082275,
      "epoch": 1.8229485274840513,
      "grad_norm": 2.546875,
      "learning_rate": 0.00046680983744562315,
      "loss": 5.8627,
      "mean_token_accuracy": 0.15463084280490874,
      "num_tokens": 42244093.0,
      "step": 20860
    },
    {
      "entropy": 6.153869152069092,
      "epoch": 1.8233854758367563,
      "grad_norm": 3.125,
      "learning_rate": 0.0004667935462505716,
      "loss": 5.8901,
      "mean_token_accuracy": 0.15651557147502898,
      "num_tokens": 42254865.0,
      "step": 20865
    },
    {
      "entropy": 6.1403931140899655,
      "epoch": 1.8238224241894607,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00046677725137681924,
      "loss": 5.8274,
      "mean_token_accuracy": 0.15917468518018724,
      "num_tokens": 42264565.0,
      "step": 20870
    },
    {
      "entropy": 6.267869281768799,
      "epoch": 1.8242593725421656,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00046676095282467873,
      "loss": 5.9256,
      "mean_token_accuracy": 0.1505710244178772,
      "num_tokens": 42274289.0,
      "step": 20875
    },
    {
      "entropy": 6.200950765609742,
      "epoch": 1.82469632089487,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004667446505944627,
      "loss": 5.897,
      "mean_token_accuracy": 0.15393013507127762,
      "num_tokens": 42283855.0,
      "step": 20880
    },
    {
      "entropy": 6.109321594238281,
      "epoch": 1.825133269247575,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046672834468648375,
      "loss": 5.9173,
      "mean_token_accuracy": 0.15863101780414582,
      "num_tokens": 42294871.0,
      "step": 20885
    },
    {
      "entropy": 6.244772624969483,
      "epoch": 1.8255702176002795,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004667120351010547,
      "loss": 5.8577,
      "mean_token_accuracy": 0.15649794936180114,
      "num_tokens": 42305315.0,
      "step": 20890
    },
    {
      "entropy": 6.21011176109314,
      "epoch": 1.8260071659529844,
      "grad_norm": 2.53125,
      "learning_rate": 0.00046669572183848833,
      "loss": 5.8854,
      "mean_token_accuracy": 0.1583572193980217,
      "num_tokens": 42316893.0,
      "step": 20895
    },
    {
      "entropy": 6.134417247772217,
      "epoch": 1.826444114305689,
      "grad_norm": 2.640625,
      "learning_rate": 0.00046667940489909766,
      "loss": 5.8045,
      "mean_token_accuracy": 0.1608392119407654,
      "num_tokens": 42326784.0,
      "step": 20900
    },
    {
      "entropy": 6.144859075546265,
      "epoch": 1.8268810626583938,
      "grad_norm": 2.375,
      "learning_rate": 0.00046666308428319547,
      "loss": 5.8685,
      "mean_token_accuracy": 0.16930171102285385,
      "num_tokens": 42336283.0,
      "step": 20905
    },
    {
      "entropy": 6.189002418518067,
      "epoch": 1.8273180110110985,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046664675999109494,
      "loss": 5.8895,
      "mean_token_accuracy": 0.15448684841394425,
      "num_tokens": 42346933.0,
      "step": 20910
    },
    {
      "entropy": 6.188580846786499,
      "epoch": 1.8277549593638032,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004666304320231091,
      "loss": 5.9342,
      "mean_token_accuracy": 0.15245856046676637,
      "num_tokens": 42357616.0,
      "step": 20915
    },
    {
      "entropy": 6.2367757797241214,
      "epoch": 1.8281919077165079,
      "grad_norm": 2.125,
      "learning_rate": 0.0004666141003795512,
      "loss": 5.9007,
      "mean_token_accuracy": 0.15505314022302627,
      "num_tokens": 42367566.0,
      "step": 20920
    },
    {
      "entropy": 6.277100229263306,
      "epoch": 1.8286288560692125,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046659776506073446,
      "loss": 5.9163,
      "mean_token_accuracy": 0.15684097707271577,
      "num_tokens": 42377412.0,
      "step": 20925
    },
    {
      "entropy": 6.221475076675415,
      "epoch": 1.8290658044219175,
      "grad_norm": 3.265625,
      "learning_rate": 0.00046658142606697216,
      "loss": 5.8634,
      "mean_token_accuracy": 0.15334687530994415,
      "num_tokens": 42387952.0,
      "step": 20930
    },
    {
      "entropy": 6.218066835403443,
      "epoch": 1.829502752774622,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004665650833985777,
      "loss": 5.9077,
      "mean_token_accuracy": 0.14830271154642105,
      "num_tokens": 42398488.0,
      "step": 20935
    },
    {
      "entropy": 6.204687595367432,
      "epoch": 1.8299397011273268,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046654873705586464,
      "loss": 5.8532,
      "mean_token_accuracy": 0.1593029424548149,
      "num_tokens": 42409137.0,
      "step": 20940
    },
    {
      "entropy": 6.224873447418213,
      "epoch": 1.8303766494800313,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004665323870391464,
      "loss": 5.9478,
      "mean_token_accuracy": 0.14620455652475356,
      "num_tokens": 42419867.0,
      "step": 20945
    },
    {
      "entropy": 6.210864162445068,
      "epoch": 1.8308135978327362,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004665160333487365,
      "loss": 5.9203,
      "mean_token_accuracy": 0.1486515946686268,
      "num_tokens": 42429590.0,
      "step": 20950
    },
    {
      "entropy": 6.2603325843811035,
      "epoch": 1.831250546185441,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004664996759849489,
      "loss": 5.8789,
      "mean_token_accuracy": 0.15790773630142213,
      "num_tokens": 42439777.0,
      "step": 20955
    },
    {
      "entropy": 6.138587570190429,
      "epoch": 1.8316874945381456,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004664833149480971,
      "loss": 5.8687,
      "mean_token_accuracy": 0.1580924093723297,
      "num_tokens": 42449843.0,
      "step": 20960
    },
    {
      "entropy": 6.1012523651123045,
      "epoch": 1.8321244428908503,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046646695023849506,
      "loss": 5.7386,
      "mean_token_accuracy": 0.16799938082695007,
      "num_tokens": 42459056.0,
      "step": 20965
    },
    {
      "entropy": 6.152824831008911,
      "epoch": 1.832561391243555,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046645058185645644,
      "loss": 5.8025,
      "mean_token_accuracy": 0.15238026678562164,
      "num_tokens": 42468946.0,
      "step": 20970
    },
    {
      "entropy": 6.163679790496826,
      "epoch": 1.8329983395962597,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004664342098022954,
      "loss": 5.8634,
      "mean_token_accuracy": 0.15472086519002914,
      "num_tokens": 42478894.0,
      "step": 20975
    },
    {
      "entropy": 6.1776971340179445,
      "epoch": 1.8334352879489644,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004664178340763259,
      "loss": 5.8623,
      "mean_token_accuracy": 0.15696797519922256,
      "num_tokens": 42490124.0,
      "step": 20980
    },
    {
      "entropy": 6.188824892044067,
      "epoch": 1.8338722363016693,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004664014546788621,
      "loss": 5.8734,
      "mean_token_accuracy": 0.15605505108833312,
      "num_tokens": 42499554.0,
      "step": 20985
    },
    {
      "entropy": 6.159797143936157,
      "epoch": 1.8343091846543738,
      "grad_norm": 2.90625,
      "learning_rate": 0.00046638507161021807,
      "loss": 5.7651,
      "mean_token_accuracy": 0.16428791135549545,
      "num_tokens": 42510364.0,
      "step": 20990
    },
    {
      "entropy": 6.160965442657471,
      "epoch": 1.8347461330070787,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046636868487070804,
      "loss": 5.9439,
      "mean_token_accuracy": 0.14997201561927795,
      "num_tokens": 42521671.0,
      "step": 20995
    },
    {
      "entropy": 6.191292572021484,
      "epoch": 1.8351830813597831,
      "grad_norm": 1.921875,
      "learning_rate": 0.00046635229446064647,
      "loss": 5.8646,
      "mean_token_accuracy": 0.15591911375522613,
      "num_tokens": 42532784.0,
      "step": 21000
    },
    {
      "epoch": 1.8351830813597831,
      "eval_entropy": 5.987726174344866,
      "eval_loss": 5.922273635864258,
      "eval_mean_token_accuracy": 0.16092439408401787,
      "eval_num_tokens": 42532784.0,
      "eval_runtime": 27.8048,
      "eval_samples_per_second": 1331.211,
      "eval_steps_per_second": 166.41,
      "step": 21000
    },
    {
      "entropy": 6.159976863861084,
      "epoch": 1.835620029712488,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004663359003803475,
      "loss": 5.7911,
      "mean_token_accuracy": 0.1662597320973873,
      "num_tokens": 42542502.0,
      "step": 21005
    },
    {
      "entropy": 6.113416767120361,
      "epoch": 1.8360569780651927,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004663195026301258,
      "loss": 5.8064,
      "mean_token_accuracy": 0.15519697219133377,
      "num_tokens": 42551837.0,
      "step": 21010
    },
    {
      "entropy": 6.21447982788086,
      "epoch": 1.8364939264178974,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046630310121029573,
      "loss": 5.8569,
      "mean_token_accuracy": 0.15790526792407036,
      "num_tokens": 42560537.0,
      "step": 21015
    },
    {
      "entropy": 6.151961612701416,
      "epoch": 1.8369308747706021,
      "grad_norm": 3.265625,
      "learning_rate": 0.00046628669612117194,
      "loss": 5.8687,
      "mean_token_accuracy": 0.14935640692710878,
      "num_tokens": 42570386.0,
      "step": 21020
    },
    {
      "entropy": 6.222688817977906,
      "epoch": 1.8373678231233068,
      "grad_norm": 4.3125,
      "learning_rate": 0.00046627028736306907,
      "loss": 5.9295,
      "mean_token_accuracy": 0.15200354382395745,
      "num_tokens": 42579860.0,
      "step": 21025
    },
    {
      "entropy": 6.260773611068726,
      "epoch": 1.8378047714760115,
      "grad_norm": 3.796875,
      "learning_rate": 0.00046625387493630187,
      "loss": 5.9264,
      "mean_token_accuracy": 0.15194553434848784,
      "num_tokens": 42590320.0,
      "step": 21030
    },
    {
      "entropy": 6.178484535217285,
      "epoch": 1.8382417198287162,
      "grad_norm": 10.625,
      "learning_rate": 0.0004662374588411852,
      "loss": 5.8706,
      "mean_token_accuracy": 0.1553271844983101,
      "num_tokens": 42600331.0,
      "step": 21035
    },
    {
      "entropy": 6.160251569747925,
      "epoch": 1.8386786681814211,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046622103907803384,
      "loss": 5.9069,
      "mean_token_accuracy": 0.154375097155571,
      "num_tokens": 42610860.0,
      "step": 21040
    },
    {
      "entropy": 6.191299772262573,
      "epoch": 1.8391156165341256,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046620461564716265,
      "loss": 5.9137,
      "mean_token_accuracy": 0.149410580098629,
      "num_tokens": 42620305.0,
      "step": 21045
    },
    {
      "entropy": 6.216998434066772,
      "epoch": 1.8395525648868305,
      "grad_norm": 2.5,
      "learning_rate": 0.0004661881885488869,
      "loss": 5.8161,
      "mean_token_accuracy": 0.1594637155532837,
      "num_tokens": 42629667.0,
      "step": 21050
    },
    {
      "entropy": 6.138246679306031,
      "epoch": 1.839989513239535,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004661717577835214,
      "loss": 5.9223,
      "mean_token_accuracy": 0.15716630443930626,
      "num_tokens": 42640116.0,
      "step": 21055
    },
    {
      "entropy": 6.214045524597168,
      "epoch": 1.8404264615922399,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004661553233513815,
      "loss": 5.8425,
      "mean_token_accuracy": 0.15920509099960328,
      "num_tokens": 42650014.0,
      "step": 21060
    },
    {
      "entropy": 6.265816879272461,
      "epoch": 1.8408634099449444,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004661388852527822,
      "loss": 5.9505,
      "mean_token_accuracy": 0.1481980174779892,
      "num_tokens": 42661280.0,
      "step": 21065
    },
    {
      "entropy": 6.1425638675689695,
      "epoch": 1.8413003582976493,
      "grad_norm": 2.84375,
      "learning_rate": 0.00046612244348803906,
      "loss": 5.9485,
      "mean_token_accuracy": 0.15513161122798919,
      "num_tokens": 42671530.0,
      "step": 21070
    },
    {
      "entropy": 6.274524402618408,
      "epoch": 1.841737306650354,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004661059980574672,
      "loss": 5.9521,
      "mean_token_accuracy": 0.1559941902756691,
      "num_tokens": 42682738.0,
      "step": 21075
    },
    {
      "entropy": 6.223133325576782,
      "epoch": 1.8421742550030586,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004660895489613822,
      "loss": 5.9152,
      "mean_token_accuracy": 0.14910192266106606,
      "num_tokens": 42694271.0,
      "step": 21080
    },
    {
      "entropy": 6.165912246704101,
      "epoch": 1.8426112033557633,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004660730962000996,
      "loss": 5.9411,
      "mean_token_accuracy": 0.1476719170808792,
      "num_tokens": 42704896.0,
      "step": 21085
    },
    {
      "entropy": 6.168588066101075,
      "epoch": 1.843048151708468,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004660566397739348,
      "loss": 5.873,
      "mean_token_accuracy": 0.15713151395320893,
      "num_tokens": 42715506.0,
      "step": 21090
    },
    {
      "entropy": 6.2327593803405765,
      "epoch": 1.8434851000611727,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046604017968320357,
      "loss": 5.853,
      "mean_token_accuracy": 0.1611639678478241,
      "num_tokens": 42725428.0,
      "step": 21095
    },
    {
      "entropy": 6.141726350784301,
      "epoch": 1.8439220484138774,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004660237159282216,
      "loss": 5.8587,
      "mean_token_accuracy": 0.15310736000537872,
      "num_tokens": 42735105.0,
      "step": 21100
    },
    {
      "entropy": 6.0939709663391115,
      "epoch": 1.8443589967665823,
      "grad_norm": 2.953125,
      "learning_rate": 0.00046600724850930466,
      "loss": 5.8001,
      "mean_token_accuracy": 0.15816112607717514,
      "num_tokens": 42744191.0,
      "step": 21105
    },
    {
      "entropy": 6.225602674484253,
      "epoch": 1.8447959451192868,
      "grad_norm": 3.0,
      "learning_rate": 0.0004659907774267686,
      "loss": 5.9875,
      "mean_token_accuracy": 0.1536152258515358,
      "num_tokens": 42755416.0,
      "step": 21110
    },
    {
      "entropy": 6.222778177261352,
      "epoch": 1.8452328934719917,
      "grad_norm": 2.46875,
      "learning_rate": 0.00046597430268092933,
      "loss": 5.8052,
      "mean_token_accuracy": 0.16265518218278885,
      "num_tokens": 42764974.0,
      "step": 21115
    },
    {
      "entropy": 6.1532816886901855,
      "epoch": 1.8456698418246962,
      "grad_norm": 2.359375,
      "learning_rate": 0.00046595782427210285,
      "loss": 5.7831,
      "mean_token_accuracy": 0.16372381895780563,
      "num_tokens": 42775080.0,
      "step": 21120
    },
    {
      "entropy": 6.075526762008667,
      "epoch": 1.846106790177401,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004659413422006053,
      "loss": 5.853,
      "mean_token_accuracy": 0.15297995433211325,
      "num_tokens": 42784333.0,
      "step": 21125
    },
    {
      "entropy": 6.224170446395874,
      "epoch": 1.8465437385301058,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004659248564667527,
      "loss": 5.8642,
      "mean_token_accuracy": 0.1572964072227478,
      "num_tokens": 42792912.0,
      "step": 21130
    },
    {
      "entropy": 6.189256191253662,
      "epoch": 1.8469806868828105,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046590836707086135,
      "loss": 5.8977,
      "mean_token_accuracy": 0.15011458620429038,
      "num_tokens": 42803812.0,
      "step": 21135
    },
    {
      "entropy": 6.242350625991821,
      "epoch": 1.8474176352355152,
      "grad_norm": 2.484375,
      "learning_rate": 0.00046589187401324753,
      "loss": 5.9371,
      "mean_token_accuracy": 0.15316127240657806,
      "num_tokens": 42815172.0,
      "step": 21140
    },
    {
      "entropy": 6.251559829711914,
      "epoch": 1.8478545835882199,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004658753772942274,
      "loss": 5.8609,
      "mean_token_accuracy": 0.15020973086357117,
      "num_tokens": 42824958.0,
      "step": 21145
    },
    {
      "entropy": 6.1344810962677006,
      "epoch": 1.8482915319409245,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046585887691411764,
      "loss": 5.8923,
      "mean_token_accuracy": 0.15356578677892685,
      "num_tokens": 42834940.0,
      "step": 21150
    },
    {
      "entropy": 6.170931482315064,
      "epoch": 1.8487284802936292,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004658423728732346,
      "loss": 5.8171,
      "mean_token_accuracy": 0.15696654766798018,
      "num_tokens": 42845102.0,
      "step": 21155
    },
    {
      "entropy": 6.155392217636108,
      "epoch": 1.8491654286463342,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004658258651718948,
      "loss": 5.8201,
      "mean_token_accuracy": 0.15791754722595214,
      "num_tokens": 42853876.0,
      "step": 21160
    },
    {
      "entropy": 6.249963808059692,
      "epoch": 1.8496023769990386,
      "grad_norm": 4.0625,
      "learning_rate": 0.00046580935381041496,
      "loss": 5.9961,
      "mean_token_accuracy": 0.14482576996088029,
      "num_tokens": 42863592.0,
      "step": 21165
    },
    {
      "entropy": 6.270920085906982,
      "epoch": 1.8500393253517435,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004657928387891117,
      "loss": 5.8318,
      "mean_token_accuracy": 0.1633957788348198,
      "num_tokens": 42873478.0,
      "step": 21170
    },
    {
      "entropy": 6.180761528015137,
      "epoch": 1.850476273704448,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046577632010830184,
      "loss": 5.7988,
      "mean_token_accuracy": 0.1599924609065056,
      "num_tokens": 42882591.0,
      "step": 21175
    },
    {
      "entropy": 6.086556625366211,
      "epoch": 1.850913222057153,
      "grad_norm": 2.125,
      "learning_rate": 0.0004657597977683022,
      "loss": 5.7669,
      "mean_token_accuracy": 0.16514393240213393,
      "num_tokens": 42892391.0,
      "step": 21180
    },
    {
      "entropy": 6.20089898109436,
      "epoch": 1.8513501704098576,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004657432717694297,
      "loss": 5.8624,
      "mean_token_accuracy": 0.16084311604499818,
      "num_tokens": 42902839.0,
      "step": 21185
    },
    {
      "entropy": 6.22453498840332,
      "epoch": 1.8517871187625623,
      "grad_norm": 2.625,
      "learning_rate": 0.0004657267421120012,
      "loss": 5.9237,
      "mean_token_accuracy": 0.15316448658704757,
      "num_tokens": 42912404.0,
      "step": 21190
    },
    {
      "entropy": 6.148573303222657,
      "epoch": 1.852224067115267,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004657102087963339,
      "loss": 5.8499,
      "mean_token_accuracy": 0.15888406634330748,
      "num_tokens": 42922939.0,
      "step": 21195
    },
    {
      "entropy": 6.114471912384033,
      "epoch": 1.8526610154679717,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00046569367182274494,
      "loss": 5.8697,
      "mean_token_accuracy": 0.16104934960603715,
      "num_tokens": 42932095.0,
      "step": 21200
    },
    {
      "entropy": 6.214882516860962,
      "epoch": 1.8530979638206764,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004656771311915513,
      "loss": 5.8693,
      "mean_token_accuracy": 0.15711315274238585,
      "num_tokens": 42941910.0,
      "step": 21205
    },
    {
      "entropy": 6.216891765594482,
      "epoch": 1.853534912173381,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004656605869030704,
      "loss": 5.8865,
      "mean_token_accuracy": 0.15631891340017318,
      "num_tokens": 42952869.0,
      "step": 21210
    },
    {
      "entropy": 6.239920520782471,
      "epoch": 1.853971860526086,
      "grad_norm": 2.75,
      "learning_rate": 0.00046564403895761966,
      "loss": 5.9811,
      "mean_token_accuracy": 0.15417468771338463,
      "num_tokens": 42963300.0,
      "step": 21215
    },
    {
      "entropy": 6.180042171478272,
      "epoch": 1.8544088088787904,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004656274873555162,
      "loss": 5.8191,
      "mean_token_accuracy": 0.15564295649528503,
      "num_tokens": 42973505.0,
      "step": 21220
    },
    {
      "entropy": 6.138100719451904,
      "epoch": 1.8548457572314954,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004656109320970778,
      "loss": 5.857,
      "mean_token_accuracy": 0.156312258541584,
      "num_tokens": 42982813.0,
      "step": 21225
    },
    {
      "entropy": 6.132159233093262,
      "epoch": 1.8552827055841998,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004655943731826217,
      "loss": 5.8768,
      "mean_token_accuracy": 0.15542505234479903,
      "num_tokens": 42991792.0,
      "step": 21230
    },
    {
      "entropy": 6.166652059555053,
      "epoch": 1.8557196539369047,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004655778106124658,
      "loss": 5.8147,
      "mean_token_accuracy": 0.16788267344236374,
      "num_tokens": 43002191.0,
      "step": 21235
    },
    {
      "entropy": 6.215726757049561,
      "epoch": 1.8561566022896092,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004655612443869275,
      "loss": 5.8396,
      "mean_token_accuracy": 0.16284652203321456,
      "num_tokens": 43013771.0,
      "step": 21240
    },
    {
      "entropy": 6.170625352859497,
      "epoch": 1.8565935506423141,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004655446745063247,
      "loss": 5.8985,
      "mean_token_accuracy": 0.14969685077667236,
      "num_tokens": 43023982.0,
      "step": 21245
    },
    {
      "entropy": 6.178741979598999,
      "epoch": 1.8570304989950188,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004655281009709753,
      "loss": 5.9605,
      "mean_token_accuracy": 0.15080476701259612,
      "num_tokens": 43034827.0,
      "step": 21250
    },
    {
      "entropy": 6.195120763778687,
      "epoch": 1.8574674473477235,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046551152378119696,
      "loss": 5.829,
      "mean_token_accuracy": 0.15844297856092454,
      "num_tokens": 43044811.0,
      "step": 21255
    },
    {
      "entropy": 6.193256044387818,
      "epoch": 1.8579043957004282,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004654949429373078,
      "loss": 5.8717,
      "mean_token_accuracy": 0.15671620070934295,
      "num_tokens": 43054513.0,
      "step": 21260
    },
    {
      "entropy": 6.147799730300903,
      "epoch": 1.858341344053133,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004654783584396258,
      "loss": 5.8779,
      "mean_token_accuracy": 0.15236560702323915,
      "num_tokens": 43064356.0,
      "step": 21265
    },
    {
      "entropy": 6.168254423141479,
      "epoch": 1.8587782924058376,
      "grad_norm": 2.53125,
      "learning_rate": 0.00046546177028846915,
      "loss": 5.7714,
      "mean_token_accuracy": 0.16335634738206864,
      "num_tokens": 43074663.0,
      "step": 21270
    },
    {
      "entropy": 6.213752555847168,
      "epoch": 1.8592152407585423,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004654451784841559,
      "loss": 5.8457,
      "mean_token_accuracy": 0.1590520530939102,
      "num_tokens": 43084542.0,
      "step": 21275
    },
    {
      "entropy": 6.080261135101319,
      "epoch": 1.8596521891112472,
      "grad_norm": 2.6875,
      "learning_rate": 0.00046542858302700443,
      "loss": 5.6768,
      "mean_token_accuracy": 0.1744646906852722,
      "num_tokens": 43093864.0,
      "step": 21280
    },
    {
      "entropy": 6.073693752288818,
      "epoch": 1.8600891374639517,
      "grad_norm": 1.90625,
      "learning_rate": 0.00046541198391733286,
      "loss": 5.8161,
      "mean_token_accuracy": 0.15882111340761185,
      "num_tokens": 43104541.0,
      "step": 21285
    },
    {
      "entropy": 6.172966814041137,
      "epoch": 1.8605260858166566,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004653953811554596,
      "loss": 5.8444,
      "mean_token_accuracy": 0.1601268544793129,
      "num_tokens": 43114378.0,
      "step": 21290
    },
    {
      "entropy": 6.253961753845215,
      "epoch": 1.860963034169361,
      "grad_norm": 2.875,
      "learning_rate": 0.0004653787747417033,
      "loss": 6.0381,
      "mean_token_accuracy": 0.14931707605719566,
      "num_tokens": 43125503.0,
      "step": 21295
    },
    {
      "entropy": 6.183517599105835,
      "epoch": 1.861399982522066,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046536216467638227,
      "loss": 5.8316,
      "mean_token_accuracy": 0.1614314630627632,
      "num_tokens": 43135034.0,
      "step": 21300
    },
    {
      "entropy": 6.118591403961181,
      "epoch": 1.8618369308747706,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004653455509598152,
      "loss": 5.8954,
      "mean_token_accuracy": 0.15902843549847603,
      "num_tokens": 43145359.0,
      "step": 21305
    },
    {
      "entropy": 6.1091536521911625,
      "epoch": 1.8622738792274753,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004653289335923206,
      "loss": 5.7809,
      "mean_token_accuracy": 0.1659266747534275,
      "num_tokens": 43154995.0,
      "step": 21310
    },
    {
      "entropy": 6.190190219879151,
      "epoch": 1.86271082758018,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004653123125742175,
      "loss": 5.8487,
      "mean_token_accuracy": 0.1614477202296257,
      "num_tokens": 43164703.0,
      "step": 21315
    },
    {
      "entropy": 6.19055438041687,
      "epoch": 1.8631477759328847,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004652956879058245,
      "loss": 5.8801,
      "mean_token_accuracy": 0.15788840278983116,
      "num_tokens": 43174193.0,
      "step": 21320
    },
    {
      "entropy": 6.144304943084717,
      "epoch": 1.8635847242855894,
      "grad_norm": 2.375,
      "learning_rate": 0.00046527905958746043,
      "loss": 5.8977,
      "mean_token_accuracy": 0.16102043390274048,
      "num_tokens": 43184236.0,
      "step": 21325
    },
    {
      "entropy": 6.1652045249938965,
      "epoch": 1.864021672638294,
      "grad_norm": 2.875,
      "learning_rate": 0.00046526242761944434,
      "loss": 5.8975,
      "mean_token_accuracy": 0.1602829709649086,
      "num_tokens": 43194421.0,
      "step": 21330
    },
    {
      "entropy": 6.1695435523986815,
      "epoch": 1.864458620990999,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046524579200209513,
      "loss": 5.9282,
      "mean_token_accuracy": 0.15546021163463591,
      "num_tokens": 43205200.0,
      "step": 21335
    },
    {
      "entropy": 6.164396476745606,
      "epoch": 1.8648955693437035,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046522915273573194,
      "loss": 5.8188,
      "mean_token_accuracy": 0.16378037482500077,
      "num_tokens": 43215415.0,
      "step": 21340
    },
    {
      "entropy": 6.208400726318359,
      "epoch": 1.8653325176964084,
      "grad_norm": 3.125,
      "learning_rate": 0.0004652125098206739,
      "loss": 5.879,
      "mean_token_accuracy": 0.15649597644805907,
      "num_tokens": 43225648.0,
      "step": 21345
    },
    {
      "entropy": 6.206456089019776,
      "epoch": 1.8657694660491129,
      "grad_norm": 2.125,
      "learning_rate": 0.00046519586325724033,
      "loss": 5.8642,
      "mean_token_accuracy": 0.1508763201534748,
      "num_tokens": 43236358.0,
      "step": 21350
    },
    {
      "entropy": 6.189830541610718,
      "epoch": 1.8662064144018178,
      "grad_norm": 2.546875,
      "learning_rate": 0.00046517921304575045,
      "loss": 5.8976,
      "mean_token_accuracy": 0.15497033596038817,
      "num_tokens": 43247022.0,
      "step": 21355
    },
    {
      "entropy": 6.187968778610229,
      "epoch": 1.8666433627545223,
      "grad_norm": 1.984375,
      "learning_rate": 0.00046516255918652356,
      "loss": 5.9527,
      "mean_token_accuracy": 0.15104374885559083,
      "num_tokens": 43258249.0,
      "step": 21360
    },
    {
      "entropy": 6.201366567611695,
      "epoch": 1.8670803111072272,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004651459016798791,
      "loss": 5.8883,
      "mean_token_accuracy": 0.15354067459702492,
      "num_tokens": 43268997.0,
      "step": 21365
    },
    {
      "entropy": 6.164216566085815,
      "epoch": 1.8675172594599319,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004651292405261368,
      "loss": 5.8216,
      "mean_token_accuracy": 0.15929169356822967,
      "num_tokens": 43279165.0,
      "step": 21370
    },
    {
      "entropy": 6.119032001495361,
      "epoch": 1.8679542078126365,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046511257572561584,
      "loss": 5.8859,
      "mean_token_accuracy": 0.1583026185631752,
      "num_tokens": 43289778.0,
      "step": 21375
    },
    {
      "entropy": 6.178448104858399,
      "epoch": 1.8683911561653412,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004650959072786362,
      "loss": 5.8129,
      "mean_token_accuracy": 0.155116967856884,
      "num_tokens": 43299865.0,
      "step": 21380
    },
    {
      "entropy": 6.133208894729615,
      "epoch": 1.868828104518046,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004650792351855173,
      "loss": 5.8231,
      "mean_token_accuracy": 0.1603497490286827,
      "num_tokens": 43309224.0,
      "step": 21385
    },
    {
      "entropy": 6.174136829376221,
      "epoch": 1.8692650528707508,
      "grad_norm": 2.375,
      "learning_rate": 0.0004650625594465792,
      "loss": 5.8139,
      "mean_token_accuracy": 0.1491766795516014,
      "num_tokens": 43320001.0,
      "step": 21390
    },
    {
      "entropy": 6.149684286117553,
      "epoch": 1.8697020012234553,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004650458800621415,
      "loss": 5.8216,
      "mean_token_accuracy": 0.15756140500307084,
      "num_tokens": 43329790.0,
      "step": 21395
    },
    {
      "entropy": 6.262713432312012,
      "epoch": 1.8701389495761602,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004650291970325244,
      "loss": 5.9393,
      "mean_token_accuracy": 0.1533758521080017,
      "num_tokens": 43340394.0,
      "step": 21400
    },
    {
      "entropy": 6.205081224441528,
      "epoch": 1.8705758979288647,
      "grad_norm": 2.125,
      "learning_rate": 0.0004650125103580476,
      "loss": 5.901,
      "mean_token_accuracy": 0.15313107818365096,
      "num_tokens": 43351404.0,
      "step": 21405
    },
    {
      "entropy": 6.060749340057373,
      "epoch": 1.8710128462815696,
      "grad_norm": 2.125,
      "learning_rate": 0.0004649958200390313,
      "loss": 5.6792,
      "mean_token_accuracy": 0.17394959926605225,
      "num_tokens": 43361017.0,
      "step": 21410
    },
    {
      "entropy": 6.17750563621521,
      "epoch": 1.871449794634274,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004649791260757956,
      "loss": 5.9225,
      "mean_token_accuracy": 0.15339571088552476,
      "num_tokens": 43371680.0,
      "step": 21415
    },
    {
      "entropy": 6.2090795040130615,
      "epoch": 1.871886742986979,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004649624284686608,
      "loss": 5.9102,
      "mean_token_accuracy": 0.16104961633682252,
      "num_tokens": 43381908.0,
      "step": 21420
    },
    {
      "entropy": 6.2231996059417725,
      "epoch": 1.8723236913396837,
      "grad_norm": 2.96875,
      "learning_rate": 0.000464945727217947,
      "loss": 5.9562,
      "mean_token_accuracy": 0.14688519090414048,
      "num_tokens": 43393234.0,
      "step": 21425
    },
    {
      "entropy": 6.166251373291016,
      "epoch": 1.8727606396923884,
      "grad_norm": 3.046875,
      "learning_rate": 0.00046492902232397464,
      "loss": 5.9584,
      "mean_token_accuracy": 0.1529166355729103,
      "num_tokens": 43404065.0,
      "step": 21430
    },
    {
      "entropy": 6.201070547103882,
      "epoch": 1.873197588045093,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046491231378706415,
      "loss": 5.853,
      "mean_token_accuracy": 0.1568079635500908,
      "num_tokens": 43414417.0,
      "step": 21435
    },
    {
      "entropy": 6.251335144042969,
      "epoch": 1.8736345363977978,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000464895601607536,
      "loss": 5.9083,
      "mean_token_accuracy": 0.14784293621778488,
      "num_tokens": 43424265.0,
      "step": 21440
    },
    {
      "entropy": 6.114007043838501,
      "epoch": 1.8740714847505024,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004648788857857106,
      "loss": 5.7943,
      "mean_token_accuracy": 0.16009590178728103,
      "num_tokens": 43434699.0,
      "step": 21445
    },
    {
      "entropy": 6.184651041030884,
      "epoch": 1.8745084331032071,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046486216632190874,
      "loss": 5.8995,
      "mean_token_accuracy": 0.15598782002925873,
      "num_tokens": 43444113.0,
      "step": 21450
    },
    {
      "entropy": 6.143470573425293,
      "epoch": 1.874945381455912,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046484544321645103,
      "loss": 5.8594,
      "mean_token_accuracy": 0.15668480694293976,
      "num_tokens": 43454256.0,
      "step": 21455
    },
    {
      "entropy": 6.211294174194336,
      "epoch": 1.8753823298086165,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004648287164696583,
      "loss": 5.8696,
      "mean_token_accuracy": 0.15715180188417435,
      "num_tokens": 43463182.0,
      "step": 21460
    },
    {
      "entropy": 6.17541356086731,
      "epoch": 1.8758192781613214,
      "grad_norm": 2.625,
      "learning_rate": 0.0004648119860818512,
      "loss": 5.831,
      "mean_token_accuracy": 0.15863404422998428,
      "num_tokens": 43473326.0,
      "step": 21465
    },
    {
      "entropy": 6.1408360481262205,
      "epoch": 1.876256226514026,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046479525205335084,
      "loss": 5.9009,
      "mean_token_accuracy": 0.15754251629114152,
      "num_tokens": 43483210.0,
      "step": 21470
    },
    {
      "entropy": 6.147503089904785,
      "epoch": 1.8766931748667308,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004647785143844781,
      "loss": 5.8689,
      "mean_token_accuracy": 0.15817331969738008,
      "num_tokens": 43493834.0,
      "step": 21475
    },
    {
      "entropy": 6.229709959030151,
      "epoch": 1.8771301232194355,
      "grad_norm": 2.53125,
      "learning_rate": 0.00046476177307555394,
      "loss": 5.8901,
      "mean_token_accuracy": 0.1521814063191414,
      "num_tokens": 43504522.0,
      "step": 21480
    },
    {
      "entropy": 6.247865056991577,
      "epoch": 1.8775670715721402,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046474502812689954,
      "loss": 5.9355,
      "mean_token_accuracy": 0.15563329309225082,
      "num_tokens": 43514950.0,
      "step": 21485
    },
    {
      "entropy": 6.200471353530884,
      "epoch": 1.878004019924845,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004647282795388361,
      "loss": 5.8795,
      "mean_token_accuracy": 0.15861220061779022,
      "num_tokens": 43526463.0,
      "step": 21490
    },
    {
      "entropy": 6.117213106155395,
      "epoch": 1.8784409682775496,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004647115273116849,
      "loss": 5.8739,
      "mean_token_accuracy": 0.15369050502777098,
      "num_tokens": 43536104.0,
      "step": 21495
    },
    {
      "entropy": 6.231101846694946,
      "epoch": 1.8788779166302543,
      "grad_norm": 2.78125,
      "learning_rate": 0.00046469477144576716,
      "loss": 6.026,
      "mean_token_accuracy": 0.14458683878183365,
      "num_tokens": 43547863.0,
      "step": 21500
    },
    {
      "entropy": 6.245478200912475,
      "epoch": 1.879314864982959,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004646780119414043,
      "loss": 5.8541,
      "mean_token_accuracy": 0.15576292127370833,
      "num_tokens": 43558024.0,
      "step": 21505
    },
    {
      "entropy": 6.219295930862427,
      "epoch": 1.8797518133356639,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004646612487989178,
      "loss": 5.8791,
      "mean_token_accuracy": 0.16175556629896165,
      "num_tokens": 43567464.0,
      "step": 21510
    },
    {
      "entropy": 6.115847206115722,
      "epoch": 1.8801887616883683,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004646444820186291,
      "loss": 5.799,
      "mean_token_accuracy": 0.16558459252119065,
      "num_tokens": 43576826.0,
      "step": 21515
    },
    {
      "entropy": 6.08799729347229,
      "epoch": 1.8806257100410733,
      "grad_norm": 3.0,
      "learning_rate": 0.00046462771160085995,
      "loss": 5.7922,
      "mean_token_accuracy": 0.16350083202123641,
      "num_tokens": 43586106.0,
      "step": 21520
    },
    {
      "entropy": 6.199208545684814,
      "epoch": 1.8810626583937777,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046461093754593185,
      "loss": 5.9362,
      "mean_token_accuracy": 0.14938604086637497,
      "num_tokens": 43597077.0,
      "step": 21525
    },
    {
      "entropy": 6.248563051223755,
      "epoch": 1.8814996067464826,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004645941598541667,
      "loss": 5.9527,
      "mean_token_accuracy": 0.1514609709382057,
      "num_tokens": 43608773.0,
      "step": 21530
    },
    {
      "entropy": 6.289806699752807,
      "epoch": 1.8819365550991871,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004645773785258862,
      "loss": 5.8645,
      "mean_token_accuracy": 0.15852869004011155,
      "num_tokens": 43618017.0,
      "step": 21535
    },
    {
      "entropy": 6.159722471237183,
      "epoch": 1.882373503451892,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004645605935614122,
      "loss": 5.8712,
      "mean_token_accuracy": 0.15411456152796746,
      "num_tokens": 43627865.0,
      "step": 21540
    },
    {
      "entropy": 6.188554096221924,
      "epoch": 1.8828104518045967,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046454380496106665,
      "loss": 5.8875,
      "mean_token_accuracy": 0.15000440031290055,
      "num_tokens": 43637772.0,
      "step": 21545
    },
    {
      "entropy": 6.13653597831726,
      "epoch": 1.8832474001573014,
      "grad_norm": 3.140625,
      "learning_rate": 0.00046452701272517167,
      "loss": 5.8258,
      "mean_token_accuracy": 0.159127376973629,
      "num_tokens": 43648242.0,
      "step": 21550
    },
    {
      "entropy": 6.168258619308472,
      "epoch": 1.883684348510006,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046451021685404927,
      "loss": 5.8938,
      "mean_token_accuracy": 0.15524711161851884,
      "num_tokens": 43658298.0,
      "step": 21555
    },
    {
      "entropy": 6.201591205596924,
      "epoch": 1.8841212968627108,
      "grad_norm": 2.375,
      "learning_rate": 0.00046449341734802155,
      "loss": 5.8856,
      "mean_token_accuracy": 0.16228315979242325,
      "num_tokens": 43667792.0,
      "step": 21560
    },
    {
      "entropy": 6.219060039520263,
      "epoch": 1.8845582452154155,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004644766142074108,
      "loss": 5.8113,
      "mean_token_accuracy": 0.16070643067359924,
      "num_tokens": 43678571.0,
      "step": 21565
    },
    {
      "entropy": 6.203215217590332,
      "epoch": 1.8849951935681202,
      "grad_norm": 3.078125,
      "learning_rate": 0.00046445980743253936,
      "loss": 5.9361,
      "mean_token_accuracy": 0.15266651660203934,
      "num_tokens": 43690851.0,
      "step": 21570
    },
    {
      "entropy": 6.138280010223388,
      "epoch": 1.885432141920825,
      "grad_norm": 4.96875,
      "learning_rate": 0.00046444299702372955,
      "loss": 5.8495,
      "mean_token_accuracy": 0.16308822631835937,
      "num_tokens": 43700960.0,
      "step": 21575
    },
    {
      "entropy": 6.1968433380126955,
      "epoch": 1.8858690902735296,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004644261829813038,
      "loss": 5.8342,
      "mean_token_accuracy": 0.15722841024398804,
      "num_tokens": 43710466.0,
      "step": 21580
    },
    {
      "entropy": 6.181585311889648,
      "epoch": 1.8863060386262345,
      "grad_norm": 2.359375,
      "learning_rate": 0.00046440936530558455,
      "loss": 5.9084,
      "mean_token_accuracy": 0.1571176290512085,
      "num_tokens": 43720261.0,
      "step": 21585
    },
    {
      "entropy": 6.174232530593872,
      "epoch": 1.886742986978939,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004643925439968944,
      "loss": 5.9213,
      "mean_token_accuracy": 0.15428656786680223,
      "num_tokens": 43731045.0,
      "step": 21590
    },
    {
      "entropy": 6.223461246490478,
      "epoch": 1.8871799353316439,
      "grad_norm": 2.53125,
      "learning_rate": 0.00046437571905555604,
      "loss": 5.8933,
      "mean_token_accuracy": 0.15739396288990976,
      "num_tokens": 43741873.0,
      "step": 21595
    },
    {
      "entropy": 6.188825464248657,
      "epoch": 1.8876168836843485,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046435889048189216,
      "loss": 5.9189,
      "mean_token_accuracy": 0.15276177376508712,
      "num_tokens": 43751473.0,
      "step": 21600
    },
    {
      "entropy": 6.277851200103759,
      "epoch": 1.8880538320370532,
      "grad_norm": 2.546875,
      "learning_rate": 0.00046434205827622555,
      "loss": 5.9218,
      "mean_token_accuracy": 0.1571374535560608,
      "num_tokens": 43762517.0,
      "step": 21605
    },
    {
      "entropy": 6.214324283599853,
      "epoch": 1.888490780389758,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004643252224388791,
      "loss": 5.9403,
      "mean_token_accuracy": 0.15002717971801757,
      "num_tokens": 43772491.0,
      "step": 21610
    },
    {
      "entropy": 6.196625089645385,
      "epoch": 1.8889277287424626,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046430838297017556,
      "loss": 5.881,
      "mean_token_accuracy": 0.15300368368625641,
      "num_tokens": 43781646.0,
      "step": 21615
    },
    {
      "entropy": 6.171273803710937,
      "epoch": 1.8893646770951673,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046429153987043815,
      "loss": 5.8079,
      "mean_token_accuracy": 0.16148501187562941,
      "num_tokens": 43792020.0,
      "step": 21620
    },
    {
      "entropy": 6.089026641845703,
      "epoch": 1.889801625447872,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004642746931399898,
      "loss": 5.8057,
      "mean_token_accuracy": 0.1655655950307846,
      "num_tokens": 43801638.0,
      "step": 21625
    },
    {
      "entropy": 6.2237961769104,
      "epoch": 1.890238573800577,
      "grad_norm": 2.71875,
      "learning_rate": 0.00046425784277915363,
      "loss": 5.9389,
      "mean_token_accuracy": 0.15109892636537553,
      "num_tokens": 43811475.0,
      "step": 21630
    },
    {
      "entropy": 6.208870458602905,
      "epoch": 1.8906755221532814,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004642409887882528,
      "loss": 5.8559,
      "mean_token_accuracy": 0.1583374857902527,
      "num_tokens": 43820811.0,
      "step": 21635
    },
    {
      "entropy": 6.227170515060425,
      "epoch": 1.8911124705059863,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004642241311676107,
      "loss": 5.9534,
      "mean_token_accuracy": 0.14206218644976615,
      "num_tokens": 43830804.0,
      "step": 21640
    },
    {
      "entropy": 6.107029819488526,
      "epoch": 1.8915494188586908,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004642072699175506,
      "loss": 5.7428,
      "mean_token_accuracy": 0.17078920751810073,
      "num_tokens": 43840156.0,
      "step": 21645
    },
    {
      "entropy": 6.136354303359985,
      "epoch": 1.8919863672113957,
      "grad_norm": 2.078125,
      "learning_rate": 0.000464190405038396,
      "loss": 5.8517,
      "mean_token_accuracy": 0.1560780167579651,
      "num_tokens": 43849622.0,
      "step": 21650
    },
    {
      "entropy": 6.228947544097901,
      "epoch": 1.8924233155641004,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004641735365304703,
      "loss": 5.9021,
      "mean_token_accuracy": 0.15283941328525544,
      "num_tokens": 43860184.0,
      "step": 21655
    },
    {
      "entropy": 6.045779418945313,
      "epoch": 1.892860263916805,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046415666439409694,
      "loss": 5.7458,
      "mean_token_accuracy": 0.17528384625911714,
      "num_tokens": 43870157.0,
      "step": 21660
    },
    {
      "entropy": 6.136982154846192,
      "epoch": 1.8932972122695098,
      "grad_norm": 2.84375,
      "learning_rate": 0.00046413978862959964,
      "loss": 5.8411,
      "mean_token_accuracy": 0.1575170561671257,
      "num_tokens": 43879867.0,
      "step": 21665
    },
    {
      "entropy": 6.117923974990845,
      "epoch": 1.8937341606222144,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004641229092373022,
      "loss": 5.8518,
      "mean_token_accuracy": 0.158186873793602,
      "num_tokens": 43890314.0,
      "step": 21670
    },
    {
      "entropy": 6.248541402816772,
      "epoch": 1.8941711089749191,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004641060262175282,
      "loss": 5.9123,
      "mean_token_accuracy": 0.15175431966781616,
      "num_tokens": 43901075.0,
      "step": 21675
    },
    {
      "entropy": 6.196884870529175,
      "epoch": 1.8946080573276238,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046408913957060144,
      "loss": 5.915,
      "mean_token_accuracy": 0.1498457133769989,
      "num_tokens": 43911777.0,
      "step": 21680
    },
    {
      "entropy": 6.205182600021362,
      "epoch": 1.8950450056803287,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00046407224929684597,
      "loss": 5.8193,
      "mean_token_accuracy": 0.15789983570575714,
      "num_tokens": 43921901.0,
      "step": 21685
    },
    {
      "entropy": 6.170269584655761,
      "epoch": 1.8954819540330332,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004640553553965857,
      "loss": 5.9242,
      "mean_token_accuracy": 0.1477438524365425,
      "num_tokens": 43931142.0,
      "step": 21690
    },
    {
      "entropy": 6.1818701267242435,
      "epoch": 1.8959189023857381,
      "grad_norm": 2.5,
      "learning_rate": 0.0004640384578701447,
      "loss": 5.8454,
      "mean_token_accuracy": 0.1577176794409752,
      "num_tokens": 43940642.0,
      "step": 21695
    },
    {
      "entropy": 6.238743877410888,
      "epoch": 1.8963558507384426,
      "grad_norm": 2.515625,
      "learning_rate": 0.00046402155671784696,
      "loss": 5.9114,
      "mean_token_accuracy": 0.1472873017191887,
      "num_tokens": 43950808.0,
      "step": 21700
    },
    {
      "entropy": 6.265205049514771,
      "epoch": 1.8967927990911475,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046400465194001674,
      "loss": 5.9998,
      "mean_token_accuracy": 0.14801618456840515,
      "num_tokens": 43961349.0,
      "step": 21705
    },
    {
      "entropy": 6.24491810798645,
      "epoch": 1.897229747443852,
      "grad_norm": 2.640625,
      "learning_rate": 0.00046398774353697826,
      "loss": 5.8381,
      "mean_token_accuracy": 0.15784917324781417,
      "num_tokens": 43970072.0,
      "step": 21710
    },
    {
      "entropy": 6.201720905303955,
      "epoch": 1.897666695796557,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004639708315090558,
      "loss": 5.9503,
      "mean_token_accuracy": 0.14970830604434013,
      "num_tokens": 43980076.0,
      "step": 21715
    },
    {
      "entropy": 6.140421342849732,
      "epoch": 1.8981036441492616,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004639539158565738,
      "loss": 5.7606,
      "mean_token_accuracy": 0.16991902738809586,
      "num_tokens": 43989876.0,
      "step": 21720
    },
    {
      "entropy": 6.258656454086304,
      "epoch": 1.8985405925019663,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004639369965798567,
      "loss": 5.9222,
      "mean_token_accuracy": 0.14939058125019072,
      "num_tokens": 44000920.0,
      "step": 21725
    },
    {
      "entropy": 6.125934839248657,
      "epoch": 1.898977540854671,
      "grad_norm": 2.359375,
      "learning_rate": 0.00046392007367922894,
      "loss": 5.7705,
      "mean_token_accuracy": 0.15964253544807433,
      "num_tokens": 44011539.0,
      "step": 21730
    },
    {
      "entropy": 6.189364576339722,
      "epoch": 1.8994144892073757,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004639031471550152,
      "loss": 5.8597,
      "mean_token_accuracy": 0.1558148592710495,
      "num_tokens": 44022624.0,
      "step": 21735
    },
    {
      "entropy": 6.20315113067627,
      "epoch": 1.8998514375600803,
      "grad_norm": 2.484375,
      "learning_rate": 0.00046388621700754014,
      "loss": 5.8506,
      "mean_token_accuracy": 0.15789655074477196,
      "num_tokens": 44033105.0,
      "step": 21740
    },
    {
      "entropy": 6.167607975006104,
      "epoch": 1.900288385912785,
      "grad_norm": 3.484375,
      "learning_rate": 0.00046386928323712847,
      "loss": 5.8504,
      "mean_token_accuracy": 0.15339830294251441,
      "num_tokens": 44042805.0,
      "step": 21745
    },
    {
      "entropy": 6.083519363403321,
      "epoch": 1.90072533426549,
      "grad_norm": 2.515625,
      "learning_rate": 0.000463852345844105,
      "loss": 5.793,
      "mean_token_accuracy": 0.17263589203357696,
      "num_tokens": 44052471.0,
      "step": 21750
    },
    {
      "entropy": 6.138082504272461,
      "epoch": 1.9011622826181944,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004638354048287945,
      "loss": 5.8962,
      "mean_token_accuracy": 0.15183840468525886,
      "num_tokens": 44063358.0,
      "step": 21755
    },
    {
      "entropy": 6.211869573593139,
      "epoch": 1.9015992309708993,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046381846019152194,
      "loss": 5.8943,
      "mean_token_accuracy": 0.15244804322719574,
      "num_tokens": 44073598.0,
      "step": 21760
    },
    {
      "entropy": 6.281509876251221,
      "epoch": 1.9020361793236038,
      "grad_norm": 2.640625,
      "learning_rate": 0.00046380151193261247,
      "loss": 5.9185,
      "mean_token_accuracy": 0.15340570360422134,
      "num_tokens": 44084752.0,
      "step": 21765
    },
    {
      "entropy": 6.173838663101196,
      "epoch": 1.9024731276763087,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004637845600523911,
      "loss": 5.8423,
      "mean_token_accuracy": 0.15962817817926406,
      "num_tokens": 44094857.0,
      "step": 21770
    },
    {
      "entropy": 6.245470666885376,
      "epoch": 1.9029100760290134,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004637676045511829,
      "loss": 5.898,
      "mean_token_accuracy": 0.15340993627905847,
      "num_tokens": 44105700.0,
      "step": 21775
    },
    {
      "entropy": 6.1734333515167235,
      "epoch": 1.903347024381718,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004637506454293131,
      "loss": 5.8523,
      "mean_token_accuracy": 0.15724046528339386,
      "num_tokens": 44116588.0,
      "step": 21780
    },
    {
      "entropy": 6.122375345230102,
      "epoch": 1.9037839727344228,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004637336826871071,
      "loss": 5.8906,
      "mean_token_accuracy": 0.15426090657711028,
      "num_tokens": 44126948.0,
      "step": 21785
    },
    {
      "entropy": 6.209274864196777,
      "epoch": 1.9042209210871275,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00046371671632489017,
      "loss": 5.9292,
      "mean_token_accuracy": 0.15152719616889954,
      "num_tokens": 44136239.0,
      "step": 21790
    },
    {
      "entropy": 6.18453049659729,
      "epoch": 1.9046578694398322,
      "grad_norm": 2.640625,
      "learning_rate": 0.00046369974634298766,
      "loss": 5.7691,
      "mean_token_accuracy": 0.16392500102519988,
      "num_tokens": 44146314.0,
      "step": 21795
    },
    {
      "entropy": 6.120741510391236,
      "epoch": 1.9050948177925369,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004636827727417252,
      "loss": 5.9223,
      "mean_token_accuracy": 0.15811474695801736,
      "num_tokens": 44155767.0,
      "step": 21800
    },
    {
      "entropy": 6.2563063621521,
      "epoch": 1.9055317661452418,
      "grad_norm": 2.265625,
      "learning_rate": 0.00046366579552142823,
      "loss": 5.9067,
      "mean_token_accuracy": 0.15626552253961562,
      "num_tokens": 44165830.0,
      "step": 21805
    },
    {
      "entropy": 6.077903795242309,
      "epoch": 1.9059687144979462,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004636488146824225,
      "loss": 5.7574,
      "mean_token_accuracy": 0.16165262460708618,
      "num_tokens": 44175526.0,
      "step": 21810
    },
    {
      "entropy": 6.185360765457153,
      "epoch": 1.9064056628506512,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004636318302250336,
      "loss": 5.9014,
      "mean_token_accuracy": 0.1556556113064289,
      "num_tokens": 44185950.0,
      "step": 21815
    },
    {
      "entropy": 6.230239534378052,
      "epoch": 1.9068426112033556,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004636148421495874,
      "loss": 5.8257,
      "mean_token_accuracy": 0.15975222438573838,
      "num_tokens": 44196518.0,
      "step": 21820
    },
    {
      "entropy": 6.203117752075196,
      "epoch": 1.9072795595560605,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046359785045640967,
      "loss": 5.9415,
      "mean_token_accuracy": 0.15730905085802077,
      "num_tokens": 44207709.0,
      "step": 21825
    },
    {
      "entropy": 6.230766868591308,
      "epoch": 1.9077165079087652,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004635808551458264,
      "loss": 5.9189,
      "mean_token_accuracy": 0.1477121353149414,
      "num_tokens": 44218311.0,
      "step": 21830
    },
    {
      "entropy": 6.198222970962524,
      "epoch": 1.90815345626147,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046356385621816343,
      "loss": 5.8464,
      "mean_token_accuracy": 0.15256570726633073,
      "num_tokens": 44229059.0,
      "step": 21835
    },
    {
      "entropy": 6.19588041305542,
      "epoch": 1.9085904046141746,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004635468536737469,
      "loss": 5.8674,
      "mean_token_accuracy": 0.15368787944316864,
      "num_tokens": 44239254.0,
      "step": 21840
    },
    {
      "entropy": 6.195323419570923,
      "epoch": 1.9090273529668793,
      "grad_norm": 1.90625,
      "learning_rate": 0.00046352984751290297,
      "loss": 5.7604,
      "mean_token_accuracy": 0.15872007310390474,
      "num_tokens": 44248465.0,
      "step": 21845
    },
    {
      "entropy": 6.118875932693482,
      "epoch": 1.909464301319584,
      "grad_norm": 3.46875,
      "learning_rate": 0.00046351283773595777,
      "loss": 5.8672,
      "mean_token_accuracy": 0.1514302283525467,
      "num_tokens": 44259121.0,
      "step": 21850
    },
    {
      "entropy": 6.224070262908936,
      "epoch": 1.9099012496722887,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046349582434323757,
      "loss": 5.8551,
      "mean_token_accuracy": 0.1657547041773796,
      "num_tokens": 44269291.0,
      "step": 21855
    },
    {
      "entropy": 6.113006114959717,
      "epoch": 1.9103381980249936,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046347880733506867,
      "loss": 5.8514,
      "mean_token_accuracy": 0.15636085867881774,
      "num_tokens": 44280985.0,
      "step": 21860
    },
    {
      "entropy": 6.143130445480347,
      "epoch": 1.910775146377698,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046346178671177746,
      "loss": 5.8386,
      "mean_token_accuracy": 0.15991266518831254,
      "num_tokens": 44290370.0,
      "step": 21865
    },
    {
      "entropy": 6.156775951385498,
      "epoch": 1.911212094730403,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046344476247369044,
      "loss": 5.8524,
      "mean_token_accuracy": 0.15415228754281998,
      "num_tokens": 44300670.0,
      "step": 21870
    },
    {
      "entropy": 6.145845699310303,
      "epoch": 1.9116490430831075,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00046342773462113414,
      "loss": 5.8678,
      "mean_token_accuracy": 0.15688926428556443,
      "num_tokens": 44311578.0,
      "step": 21875
    },
    {
      "entropy": 6.13451361656189,
      "epoch": 1.9120859914358124,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046341070315443516,
      "loss": 5.7947,
      "mean_token_accuracy": 0.15973854213953018,
      "num_tokens": 44321692.0,
      "step": 21880
    },
    {
      "entropy": 6.08477201461792,
      "epoch": 1.9125229397885168,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046339366807392017,
      "loss": 5.8254,
      "mean_token_accuracy": 0.1636738061904907,
      "num_tokens": 44331468.0,
      "step": 21885
    },
    {
      "entropy": 6.15170373916626,
      "epoch": 1.9129598881412218,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00046337662937991585,
      "loss": 5.859,
      "mean_token_accuracy": 0.15844744443893433,
      "num_tokens": 44341491.0,
      "step": 21890
    },
    {
      "entropy": 6.201627397537232,
      "epoch": 1.9133968364939264,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004633595870727492,
      "loss": 5.9348,
      "mean_token_accuracy": 0.1492035999894142,
      "num_tokens": 44351568.0,
      "step": 21895
    },
    {
      "entropy": 6.286606502532959,
      "epoch": 1.9138337848466311,
      "grad_norm": 2.78125,
      "learning_rate": 0.00046334254115274685,
      "loss": 5.9363,
      "mean_token_accuracy": 0.15731890723109246,
      "num_tokens": 44362009.0,
      "step": 21900
    },
    {
      "entropy": 6.154558753967285,
      "epoch": 1.9142707331993358,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004633254916202359,
      "loss": 5.817,
      "mean_token_accuracy": 0.16116126626729965,
      "num_tokens": 44372671.0,
      "step": 21905
    },
    {
      "entropy": 6.0915134906768795,
      "epoch": 1.9147076815520405,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004633084384755434,
      "loss": 5.8371,
      "mean_token_accuracy": 0.16064594089984893,
      "num_tokens": 44382658.0,
      "step": 21910
    },
    {
      "entropy": 6.2275307178497314,
      "epoch": 1.9151446299047452,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004632913817189963,
      "loss": 5.8234,
      "mean_token_accuracy": 0.16381541192531585,
      "num_tokens": 44392090.0,
      "step": 21915
    },
    {
      "entropy": 6.257759141921997,
      "epoch": 1.91558157825745,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004632743213509219,
      "loss": 5.9317,
      "mean_token_accuracy": 0.14731671735644342,
      "num_tokens": 44402896.0,
      "step": 21920
    },
    {
      "entropy": 6.204769039154053,
      "epoch": 1.9160185266101548,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046325725737164734,
      "loss": 5.8947,
      "mean_token_accuracy": 0.1539790078997612,
      "num_tokens": 44413700.0,
      "step": 21925
    },
    {
      "entropy": 6.182380962371826,
      "epoch": 1.9164554749628593,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004632401897814999,
      "loss": 5.7888,
      "mean_token_accuracy": 0.16058825254440307,
      "num_tokens": 44423442.0,
      "step": 21930
    },
    {
      "entropy": 6.130188417434693,
      "epoch": 1.9168924233155642,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004632231185808071,
      "loss": 5.8408,
      "mean_token_accuracy": 0.1557360142469406,
      "num_tokens": 44434197.0,
      "step": 21935
    },
    {
      "entropy": 6.109326696395874,
      "epoch": 1.9173293716682687,
      "grad_norm": 3.140625,
      "learning_rate": 0.00046320604376989604,
      "loss": 5.8492,
      "mean_token_accuracy": 0.15946876406669616,
      "num_tokens": 44444483.0,
      "step": 21940
    },
    {
      "entropy": 6.122554588317871,
      "epoch": 1.9177663200209736,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004631889653490946,
      "loss": 5.8631,
      "mean_token_accuracy": 0.15679526776075364,
      "num_tokens": 44456303.0,
      "step": 21945
    },
    {
      "entropy": 6.183895874023437,
      "epoch": 1.9182032683736783,
      "grad_norm": 5.375,
      "learning_rate": 0.00046317188331873024,
      "loss": 5.8953,
      "mean_token_accuracy": 0.15872961282730103,
      "num_tokens": 44467314.0,
      "step": 21950
    },
    {
      "entropy": 6.20928316116333,
      "epoch": 1.918640216726383,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004631547976791305,
      "loss": 5.8388,
      "mean_token_accuracy": 0.15519485771656036,
      "num_tokens": 44476466.0,
      "step": 21955
    },
    {
      "entropy": 6.140113687515258,
      "epoch": 1.9190771650790877,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004631377084306231,
      "loss": 5.8764,
      "mean_token_accuracy": 0.152776500582695,
      "num_tokens": 44486247.0,
      "step": 21960
    },
    {
      "entropy": 6.193143224716186,
      "epoch": 1.9195141134317923,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046312061557353585,
      "loss": 5.8524,
      "mean_token_accuracy": 0.15485877096652984,
      "num_tokens": 44497166.0,
      "step": 21965
    },
    {
      "entropy": 6.204368591308594,
      "epoch": 1.919951061784497,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004631035191081967,
      "loss": 5.8879,
      "mean_token_accuracy": 0.1520686462521553,
      "num_tokens": 44507677.0,
      "step": 21970
    },
    {
      "entropy": 6.182278060913086,
      "epoch": 1.9203880101372017,
      "grad_norm": 2.5,
      "learning_rate": 0.0004630864190349334,
      "loss": 5.8356,
      "mean_token_accuracy": 0.1580998495221138,
      "num_tokens": 44517392.0,
      "step": 21975
    },
    {
      "entropy": 6.249351739883423,
      "epoch": 1.9208249584899066,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046306931535407416,
      "loss": 5.8981,
      "mean_token_accuracy": 0.15908069759607316,
      "num_tokens": 44528914.0,
      "step": 21980
    },
    {
      "entropy": 6.248476123809814,
      "epoch": 1.9212619068426111,
      "grad_norm": 2.125,
      "learning_rate": 0.00046305220806594683,
      "loss": 5.9723,
      "mean_token_accuracy": 0.14499160796403884,
      "num_tokens": 44539419.0,
      "step": 21985
    },
    {
      "entropy": 6.177359294891358,
      "epoch": 1.921698855195316,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004630350971708797,
      "loss": 5.8479,
      "mean_token_accuracy": 0.16170360296964645,
      "num_tokens": 44548704.0,
      "step": 21990
    },
    {
      "entropy": 6.17561936378479,
      "epoch": 1.9221358035480205,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046301798266920076,
      "loss": 5.8477,
      "mean_token_accuracy": 0.1612362489104271,
      "num_tokens": 44559110.0,
      "step": 21995
    },
    {
      "entropy": 6.205255222320557,
      "epoch": 1.9225727519007254,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004630008645612384,
      "loss": 5.9345,
      "mean_token_accuracy": 0.14891254752874375,
      "num_tokens": 44570414.0,
      "step": 22000
    },
    {
      "entropy": 6.202343320846557,
      "epoch": 1.92300970025343,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046298374284732096,
      "loss": 5.9665,
      "mean_token_accuracy": 0.15269767865538597,
      "num_tokens": 44580194.0,
      "step": 22005
    },
    {
      "entropy": 6.231554889678955,
      "epoch": 1.9234466486061348,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004629666175277768,
      "loss": 5.839,
      "mean_token_accuracy": 0.16359364688396455,
      "num_tokens": 44590174.0,
      "step": 22010
    },
    {
      "entropy": 6.166368007659912,
      "epoch": 1.9238835969588395,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004629494886029345,
      "loss": 5.943,
      "mean_token_accuracy": 0.15100892558693885,
      "num_tokens": 44600940.0,
      "step": 22015
    },
    {
      "entropy": 6.210608386993409,
      "epoch": 1.9243205453115442,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004629323560731225,
      "loss": 5.9209,
      "mean_token_accuracy": 0.1548597954213619,
      "num_tokens": 44610852.0,
      "step": 22020
    },
    {
      "entropy": 6.261061477661133,
      "epoch": 1.9247574936642489,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00046291521993866943,
      "loss": 5.9621,
      "mean_token_accuracy": 0.15260894298553468,
      "num_tokens": 44621479.0,
      "step": 22025
    },
    {
      "entropy": 6.23016619682312,
      "epoch": 1.9251944420169536,
      "grad_norm": 2.0625,
      "learning_rate": 0.000462898080199904,
      "loss": 5.8668,
      "mean_token_accuracy": 0.1557646870613098,
      "num_tokens": 44631838.0,
      "step": 22030
    },
    {
      "entropy": 6.1093665599823,
      "epoch": 1.9256313903696585,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004628809368571549,
      "loss": 5.8587,
      "mean_token_accuracy": 0.1547582343220711,
      "num_tokens": 44643103.0,
      "step": 22035
    },
    {
      "entropy": 6.210915660858154,
      "epoch": 1.926068338722363,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00046286378991075095,
      "loss": 5.9919,
      "mean_token_accuracy": 0.1528694212436676,
      "num_tokens": 44653659.0,
      "step": 22040
    },
    {
      "entropy": 6.254080057144165,
      "epoch": 1.9265052870750679,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004628466393610211,
      "loss": 5.8553,
      "mean_token_accuracy": 0.15774400383234025,
      "num_tokens": 44665064.0,
      "step": 22045
    },
    {
      "entropy": 6.167032480239868,
      "epoch": 1.9269422354277723,
      "grad_norm": 2.8125,
      "learning_rate": 0.00046282948520829433,
      "loss": 5.8355,
      "mean_token_accuracy": 0.16270095854997635,
      "num_tokens": 44674256.0,
      "step": 22050
    },
    {
      "entropy": 6.103113126754761,
      "epoch": 1.9273791837804772,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004628123274528996,
      "loss": 5.8321,
      "mean_token_accuracy": 0.161521278321743,
      "num_tokens": 44683715.0,
      "step": 22055
    },
    {
      "entropy": 6.220614385604859,
      "epoch": 1.9278161321331817,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046279516609516594,
      "loss": 5.9813,
      "mean_token_accuracy": 0.14780441671609879,
      "num_tokens": 44694637.0,
      "step": 22060
    },
    {
      "entropy": 6.215032482147217,
      "epoch": 1.9282530804858866,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004627780011354226,
      "loss": 5.8507,
      "mean_token_accuracy": 0.15723320692777634,
      "num_tokens": 44705307.0,
      "step": 22065
    },
    {
      "entropy": 6.21521487236023,
      "epoch": 1.9286900288385913,
      "grad_norm": 2.375,
      "learning_rate": 0.0004627608325739989,
      "loss": 5.8903,
      "mean_token_accuracy": 0.1592691347002983,
      "num_tokens": 44715293.0,
      "step": 22070
    },
    {
      "entropy": 6.219334411621094,
      "epoch": 1.929126977191296,
      "grad_norm": 2.15625,
      "learning_rate": 0.000462743660411224,
      "loss": 5.9681,
      "mean_token_accuracy": 0.14105015769600868,
      "num_tokens": 44726406.0,
      "step": 22075
    },
    {
      "entropy": 6.2092327117919925,
      "epoch": 1.9295639255440007,
      "grad_norm": 2.375,
      "learning_rate": 0.00046272648464742733,
      "loss": 5.8756,
      "mean_token_accuracy": 0.15742866918444634,
      "num_tokens": 44737166.0,
      "step": 22080
    },
    {
      "entropy": 6.129435586929321,
      "epoch": 1.9300008738967054,
      "grad_norm": 1.65625,
      "learning_rate": 0.00046270930528293826,
      "loss": 5.7762,
      "mean_token_accuracy": 0.15855736136436463,
      "num_tokens": 44746693.0,
      "step": 22085
    },
    {
      "entropy": 6.176175785064697,
      "epoch": 1.93043782224941,
      "grad_norm": 3.0,
      "learning_rate": 0.0004626921223180864,
      "loss": 5.799,
      "mean_token_accuracy": 0.16505579501390458,
      "num_tokens": 44757499.0,
      "step": 22090
    },
    {
      "entropy": 6.162463521957397,
      "epoch": 1.9308747706021148,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046267493575320127,
      "loss": 5.8309,
      "mean_token_accuracy": 0.15850281119346618,
      "num_tokens": 44768112.0,
      "step": 22095
    },
    {
      "entropy": 6.090469884872436,
      "epoch": 1.9313117189548197,
      "grad_norm": 2.875,
      "learning_rate": 0.0004626577455886125,
      "loss": 5.7789,
      "mean_token_accuracy": 0.15493441224098206,
      "num_tokens": 44778317.0,
      "step": 22100
    },
    {
      "entropy": 6.158993721008301,
      "epoch": 1.9317486673075241,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004626405518246499,
      "loss": 5.9285,
      "mean_token_accuracy": 0.14833365753293037,
      "num_tokens": 44790483.0,
      "step": 22105
    },
    {
      "entropy": 6.150081300735474,
      "epoch": 1.932185615660229,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046262335446164315,
      "loss": 5.7786,
      "mean_token_accuracy": 0.16205852031707763,
      "num_tokens": 44799891.0,
      "step": 22110
    },
    {
      "entropy": 6.151667594909668,
      "epoch": 1.9326225640129335,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004626061534999222,
      "loss": 5.9016,
      "mean_token_accuracy": 0.15451744794845582,
      "num_tokens": 44810728.0,
      "step": 22115
    },
    {
      "entropy": 6.208214473724365,
      "epoch": 1.9330595123656384,
      "grad_norm": 2.71875,
      "learning_rate": 0.00046258894893981686,
      "loss": 5.8966,
      "mean_token_accuracy": 0.15502929240465163,
      "num_tokens": 44820597.0,
      "step": 22120
    },
    {
      "entropy": 6.199304676055908,
      "epoch": 1.9334964607183431,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004625717407816572,
      "loss": 5.8973,
      "mean_token_accuracy": 0.15357017517089844,
      "num_tokens": 44832570.0,
      "step": 22125
    },
    {
      "entropy": 6.223907041549682,
      "epoch": 1.9339334090710478,
      "grad_norm": 2.125,
      "learning_rate": 0.0004625545290257733,
      "loss": 5.912,
      "mean_token_accuracy": 0.15688930004835128,
      "num_tokens": 44841936.0,
      "step": 22130
    },
    {
      "entropy": 6.098055696487426,
      "epoch": 1.9343703574237525,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046253731367249527,
      "loss": 5.8185,
      "mean_token_accuracy": 0.16453840285539628,
      "num_tokens": 44852329.0,
      "step": 22135
    },
    {
      "entropy": 6.166409683227539,
      "epoch": 1.9348073057764572,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004625200947221532,
      "loss": 5.8647,
      "mean_token_accuracy": 0.15537972003221512,
      "num_tokens": 44862183.0,
      "step": 22140
    },
    {
      "entropy": 6.20058765411377,
      "epoch": 1.935244254129162,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004625028721750776,
      "loss": 5.8912,
      "mean_token_accuracy": 0.14867072850465773,
      "num_tokens": 44872712.0,
      "step": 22145
    },
    {
      "entropy": 6.119341516494751,
      "epoch": 1.9356812024818666,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046248564603159855,
      "loss": 5.9037,
      "mean_token_accuracy": 0.1552067294716835,
      "num_tokens": 44884423.0,
      "step": 22150
    },
    {
      "entropy": 6.199878549575805,
      "epoch": 1.9361181508345715,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004624684162920466,
      "loss": 5.8326,
      "mean_token_accuracy": 0.154467111825943,
      "num_tokens": 44894556.0,
      "step": 22155
    },
    {
      "entropy": 6.148905324935913,
      "epoch": 1.936555099187276,
      "grad_norm": 3.140625,
      "learning_rate": 0.00046245118295675213,
      "loss": 5.8051,
      "mean_token_accuracy": 0.15631384253501893,
      "num_tokens": 44905208.0,
      "step": 22160
    },
    {
      "entropy": 6.139513826370239,
      "epoch": 1.9369920475399809,
      "grad_norm": 3.90625,
      "learning_rate": 0.00046243394602604585,
      "loss": 5.8158,
      "mean_token_accuracy": 0.15859874933958054,
      "num_tokens": 44915674.0,
      "step": 22165
    },
    {
      "entropy": 6.146511173248291,
      "epoch": 1.9374289958926854,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004624167055002582,
      "loss": 5.8747,
      "mean_token_accuracy": 0.15596939474344254,
      "num_tokens": 44926382.0,
      "step": 22170
    },
    {
      "entropy": 6.149654674530029,
      "epoch": 1.9378659442453903,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046239946137971997,
      "loss": 5.7981,
      "mean_token_accuracy": 0.17003785073757172,
      "num_tokens": 44935963.0,
      "step": 22175
    },
    {
      "entropy": 6.2032581806182865,
      "epoch": 1.9383028925980947,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046238221366476183,
      "loss": 5.8888,
      "mean_token_accuracy": 0.15382368564605714,
      "num_tokens": 44946312.0,
      "step": 22180
    },
    {
      "entropy": 6.076856851577759,
      "epoch": 1.9387398409507997,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046236496235571475,
      "loss": 5.8212,
      "mean_token_accuracy": 0.1617642268538475,
      "num_tokens": 44956213.0,
      "step": 22185
    },
    {
      "entropy": 6.192263555526734,
      "epoch": 1.9391767893035043,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004623477074529094,
      "loss": 5.9887,
      "mean_token_accuracy": 0.14449533373117446,
      "num_tokens": 44966575.0,
      "step": 22190
    },
    {
      "entropy": 6.224356842041016,
      "epoch": 1.939613737656209,
      "grad_norm": 1.75,
      "learning_rate": 0.0004623304489566769,
      "loss": 5.8898,
      "mean_token_accuracy": 0.1565432593226433,
      "num_tokens": 44976353.0,
      "step": 22195
    },
    {
      "entropy": 6.160773754119873,
      "epoch": 1.9400506860089137,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00046231318686734826,
      "loss": 5.8446,
      "mean_token_accuracy": 0.15657754093408585,
      "num_tokens": 44987046.0,
      "step": 22200
    },
    {
      "entropy": 6.253946161270141,
      "epoch": 1.9404876343616184,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004622959211852545,
      "loss": 5.8345,
      "mean_token_accuracy": 0.15706214159727097,
      "num_tokens": 44996606.0,
      "step": 22205
    },
    {
      "entropy": 6.271708345413208,
      "epoch": 1.9409245827143233,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046227865191072684,
      "loss": 5.9093,
      "mean_token_accuracy": 0.15073804035782815,
      "num_tokens": 45006895.0,
      "step": 22210
    },
    {
      "entropy": 6.138206529617309,
      "epoch": 1.9413615310670278,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004622613790440965,
      "loss": 5.7804,
      "mean_token_accuracy": 0.16169863641262056,
      "num_tokens": 45017075.0,
      "step": 22215
    },
    {
      "entropy": 6.183891868591308,
      "epoch": 1.9417984794197327,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004622441025856948,
      "loss": 5.8425,
      "mean_token_accuracy": 0.1535644829273224,
      "num_tokens": 45027331.0,
      "step": 22220
    },
    {
      "entropy": 6.131240272521973,
      "epoch": 1.9422354277724372,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004622268225358531,
      "loss": 5.8346,
      "mean_token_accuracy": 0.14973184019327163,
      "num_tokens": 45037623.0,
      "step": 22225
    },
    {
      "entropy": 6.18422155380249,
      "epoch": 1.942672376125142,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00046220953889490287,
      "loss": 5.9003,
      "mean_token_accuracy": 0.1457845911383629,
      "num_tokens": 45048191.0,
      "step": 22230
    },
    {
      "entropy": 6.222772169113159,
      "epoch": 1.9431093244778466,
      "grad_norm": 2.125,
      "learning_rate": 0.00046219225166317554,
      "loss": 5.8884,
      "mean_token_accuracy": 0.15597901344299317,
      "num_tokens": 45057906.0,
      "step": 22235
    },
    {
      "entropy": 6.118486499786377,
      "epoch": 1.9435462728305515,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046217496084100267,
      "loss": 5.8626,
      "mean_token_accuracy": 0.16903566867113112,
      "num_tokens": 45067842.0,
      "step": 22240
    },
    {
      "entropy": 6.157311630249024,
      "epoch": 1.9439832211832562,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046215766642871607,
      "loss": 5.8375,
      "mean_token_accuracy": 0.15722315907478332,
      "num_tokens": 45077388.0,
      "step": 22245
    },
    {
      "entropy": 6.091760778427124,
      "epoch": 1.9444201695359609,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00046214036842664736,
      "loss": 5.7328,
      "mean_token_accuracy": 0.16780344843864442,
      "num_tokens": 45086001.0,
      "step": 22250
    },
    {
      "entropy": 6.1513701438903805,
      "epoch": 1.9448571178886656,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004621230668351283,
      "loss": 5.8061,
      "mean_token_accuracy": 0.16599617749452591,
      "num_tokens": 45096203.0,
      "step": 22255
    },
    {
      "entropy": 6.1212451457977295,
      "epoch": 1.9452940662413702,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00046210576165449076,
      "loss": 5.793,
      "mean_token_accuracy": 0.16427026242017745,
      "num_tokens": 45104997.0,
      "step": 22260
    },
    {
      "entropy": 6.164539432525634,
      "epoch": 1.945731014594075,
      "grad_norm": 2.5,
      "learning_rate": 0.00046208845288506667,
      "loss": 5.844,
      "mean_token_accuracy": 0.15977058336138725,
      "num_tokens": 45114150.0,
      "step": 22265
    },
    {
      "entropy": 6.05092978477478,
      "epoch": 1.9461679629467796,
      "grad_norm": 2.25,
      "learning_rate": 0.000462071140527188,
      "loss": 5.8446,
      "mean_token_accuracy": 0.16655290722846985,
      "num_tokens": 45124668.0,
      "step": 22270
    },
    {
      "entropy": 6.17544264793396,
      "epoch": 1.9466049112994845,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004620538245811869,
      "loss": 5.8799,
      "mean_token_accuracy": 0.15649853199720382,
      "num_tokens": 45135164.0,
      "step": 22275
    },
    {
      "entropy": 6.16636962890625,
      "epoch": 1.947041859652189,
      "grad_norm": 2.828125,
      "learning_rate": 0.00046203650504739543,
      "loss": 5.6972,
      "mean_token_accuracy": 0.16863479167222978,
      "num_tokens": 45144453.0,
      "step": 22280
    },
    {
      "entropy": 6.082061052322388,
      "epoch": 1.947478808004894,
      "grad_norm": 1.8125,
      "learning_rate": 0.00046201918192614573,
      "loss": 5.8851,
      "mean_token_accuracy": 0.15107959061861037,
      "num_tokens": 45154557.0,
      "step": 22285
    },
    {
      "entropy": 6.1157537460327145,
      "epoch": 1.9479157563575984,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004620018552177701,
      "loss": 5.8422,
      "mean_token_accuracy": 0.15758588165044785,
      "num_tokens": 45164938.0,
      "step": 22290
    },
    {
      "entropy": 6.261508226394653,
      "epoch": 1.9483527047103033,
      "grad_norm": 2.734375,
      "learning_rate": 0.00046198452492260096,
      "loss": 5.9929,
      "mean_token_accuracy": 0.1393781289458275,
      "num_tokens": 45175823.0,
      "step": 22295
    },
    {
      "entropy": 6.094382095336914,
      "epoch": 1.948789653063008,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004619671910409706,
      "loss": 5.7705,
      "mean_token_accuracy": 0.1771104231476784,
      "num_tokens": 45185589.0,
      "step": 22300
    },
    {
      "entropy": 6.15451078414917,
      "epoch": 1.9492266014157127,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004619498535732116,
      "loss": 5.9094,
      "mean_token_accuracy": 0.1566986471414566,
      "num_tokens": 45195654.0,
      "step": 22305
    },
    {
      "entropy": 6.254664182662964,
      "epoch": 1.9496635497684174,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046193251251965646,
      "loss": 5.937,
      "mean_token_accuracy": 0.15293463245034217,
      "num_tokens": 45205824.0,
      "step": 22310
    },
    {
      "entropy": 6.207081460952759,
      "epoch": 1.950100498121122,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004619151678806377,
      "loss": 5.8583,
      "mean_token_accuracy": 0.15920321494340897,
      "num_tokens": 45215228.0,
      "step": 22315
    },
    {
      "entropy": 6.15283784866333,
      "epoch": 1.9505374464738268,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004618978196564882,
      "loss": 5.7593,
      "mean_token_accuracy": 0.16300940811634063,
      "num_tokens": 45225043.0,
      "step": 22320
    },
    {
      "entropy": 6.183427858352661,
      "epoch": 1.9509743948265315,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004618804678475405,
      "loss": 5.8687,
      "mean_token_accuracy": 0.15568883419036866,
      "num_tokens": 45234837.0,
      "step": 22325
    },
    {
      "entropy": 6.114549684524536,
      "epoch": 1.9514113431792364,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046186311245412746,
      "loss": 5.7485,
      "mean_token_accuracy": 0.1618823826313019,
      "num_tokens": 45243540.0,
      "step": 22330
    },
    {
      "entropy": 6.141507148742676,
      "epoch": 1.9518482915319408,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004618457534765821,
      "loss": 5.8265,
      "mean_token_accuracy": 0.1566348433494568,
      "num_tokens": 45253053.0,
      "step": 22335
    },
    {
      "entropy": 6.07482590675354,
      "epoch": 1.9522852398846458,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004618283909152372,
      "loss": 5.7699,
      "mean_token_accuracy": 0.16847453862428666,
      "num_tokens": 45262973.0,
      "step": 22340
    },
    {
      "entropy": 6.07757043838501,
      "epoch": 1.9527221882373502,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004618110247704259,
      "loss": 5.7918,
      "mean_token_accuracy": 0.1661233201622963,
      "num_tokens": 45272951.0,
      "step": 22345
    },
    {
      "entropy": 6.227281951904297,
      "epoch": 1.9531591365900551,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046179365504248125,
      "loss": 5.9152,
      "mean_token_accuracy": 0.15174846798181535,
      "num_tokens": 45283596.0,
      "step": 22350
    },
    {
      "entropy": 6.221758413314819,
      "epoch": 1.9535960849427596,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046177628173173647,
      "loss": 5.7669,
      "mean_token_accuracy": 0.1677076444029808,
      "num_tokens": 45294712.0,
      "step": 22355
    },
    {
      "entropy": 6.138839864730835,
      "epoch": 1.9540330332954645,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004617589048385247,
      "loss": 5.8575,
      "mean_token_accuracy": 0.15590397417545318,
      "num_tokens": 45305681.0,
      "step": 22360
    },
    {
      "entropy": 6.135281372070312,
      "epoch": 1.9544699816481692,
      "grad_norm": 5.875,
      "learning_rate": 0.0004617415243631793,
      "loss": 5.7826,
      "mean_token_accuracy": 0.1644510805606842,
      "num_tokens": 45315332.0,
      "step": 22365
    },
    {
      "entropy": 6.076905107498169,
      "epoch": 1.954906930000874,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046172414030603356,
      "loss": 5.778,
      "mean_token_accuracy": 0.16083117797970772,
      "num_tokens": 45324526.0,
      "step": 22370
    },
    {
      "entropy": 6.177600717544555,
      "epoch": 1.9553438783535786,
      "grad_norm": 2.78125,
      "learning_rate": 0.000461706752667421,
      "loss": 5.8319,
      "mean_token_accuracy": 0.15915335565805436,
      "num_tokens": 45333989.0,
      "step": 22375
    },
    {
      "entropy": 6.2001838207244875,
      "epoch": 1.9557808267062833,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004616893614476751,
      "loss": 5.7802,
      "mean_token_accuracy": 0.15603892207145692,
      "num_tokens": 45344545.0,
      "step": 22380
    },
    {
      "entropy": 6.137746047973633,
      "epoch": 1.956217775058988,
      "grad_norm": 3.375,
      "learning_rate": 0.00046167196664712945,
      "loss": 5.8253,
      "mean_token_accuracy": 0.15759874731302262,
      "num_tokens": 45354125.0,
      "step": 22385
    },
    {
      "entropy": 6.211357116699219,
      "epoch": 1.9566547234116927,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00046165456826611763,
      "loss": 5.8774,
      "mean_token_accuracy": 0.1536942407488823,
      "num_tokens": 45364577.0,
      "step": 22390
    },
    {
      "entropy": 6.19125542640686,
      "epoch": 1.9570916717643976,
      "grad_norm": 1.921875,
      "learning_rate": 0.00046163716630497336,
      "loss": 5.8773,
      "mean_token_accuracy": 0.15919489338994025,
      "num_tokens": 45374655.0,
      "step": 22395
    },
    {
      "entropy": 6.119091892242432,
      "epoch": 1.957528620117102,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046161976076403047,
      "loss": 5.8292,
      "mean_token_accuracy": 0.15972487926483153,
      "num_tokens": 45384383.0,
      "step": 22400
    },
    {
      "entropy": 6.1395516872406,
      "epoch": 1.957965568469807,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004616023516436228,
      "loss": 5.842,
      "mean_token_accuracy": 0.15590830892324448,
      "num_tokens": 45394645.0,
      "step": 22405
    },
    {
      "entropy": 6.140950345993042,
      "epoch": 1.9584025168225114,
      "grad_norm": 3.15625,
      "learning_rate": 0.00046158493894408423,
      "loss": 5.7917,
      "mean_token_accuracy": 0.1599231630563736,
      "num_tokens": 45404661.0,
      "step": 22410
    },
    {
      "entropy": 6.164739274978638,
      "epoch": 1.9588394651752163,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004615675226657487,
      "loss": 5.919,
      "mean_token_accuracy": 0.1563268393278122,
      "num_tokens": 45413915.0,
      "step": 22415
    },
    {
      "entropy": 6.239946985244751,
      "epoch": 1.959276413527921,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004615501028089504,
      "loss": 5.8512,
      "mean_token_accuracy": 0.15476495027542114,
      "num_tokens": 45424019.0,
      "step": 22420
    },
    {
      "entropy": 6.174018383026123,
      "epoch": 1.9597133618806257,
      "grad_norm": 8.1875,
      "learning_rate": 0.0004615326793740233,
      "loss": 5.81,
      "mean_token_accuracy": 0.15897967517375947,
      "num_tokens": 45435087.0,
      "step": 22425
    },
    {
      "entropy": 6.192762613296509,
      "epoch": 1.9601503102333304,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00046151525236130174,
      "loss": 5.9521,
      "mean_token_accuracy": 0.15137489289045333,
      "num_tokens": 45446363.0,
      "step": 22430
    },
    {
      "entropy": 6.2385388851165775,
      "epoch": 1.960587258586035,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004614978217711199,
      "loss": 5.8211,
      "mean_token_accuracy": 0.15661779791116714,
      "num_tokens": 45456268.0,
      "step": 22435
    },
    {
      "entropy": 6.0659120082855225,
      "epoch": 1.9610242069387398,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046148038760381217,
      "loss": 5.7807,
      "mean_token_accuracy": 0.16689562499523164,
      "num_tokens": 45466335.0,
      "step": 22440
    },
    {
      "entropy": 6.123346424102783,
      "epoch": 1.9614611552914445,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004614629498597128,
      "loss": 5.8428,
      "mean_token_accuracy": 0.1534925267100334,
      "num_tokens": 45476128.0,
      "step": 22445
    },
    {
      "entropy": 6.1947009563446045,
      "epoch": 1.9618981036441494,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004614455085391564,
      "loss": 5.7858,
      "mean_token_accuracy": 0.15899542719125748,
      "num_tokens": 45485986.0,
      "step": 22450
    },
    {
      "entropy": 6.143411779403687,
      "epoch": 1.9623350519968539,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004614280636424774,
      "loss": 5.8539,
      "mean_token_accuracy": 0.15920880436897278,
      "num_tokens": 45497270.0,
      "step": 22455
    },
    {
      "entropy": 6.109233570098877,
      "epoch": 1.9627720003495588,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004614106151700105,
      "loss": 5.8429,
      "mean_token_accuracy": 0.16261110007762908,
      "num_tokens": 45507415.0,
      "step": 22460
    },
    {
      "entropy": 6.117492246627807,
      "epoch": 1.9632089487022633,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004613931631220903,
      "loss": 5.8724,
      "mean_token_accuracy": 0.15084838420152663,
      "num_tokens": 45518025.0,
      "step": 22465
    },
    {
      "entropy": 6.185597276687622,
      "epoch": 1.9636458970549682,
      "grad_norm": 1.984375,
      "learning_rate": 0.00046137570749905157,
      "loss": 5.9102,
      "mean_token_accuracy": 0.15359861701726912,
      "num_tokens": 45528834.0,
      "step": 22470
    },
    {
      "entropy": 6.197058296203613,
      "epoch": 1.9640828454076729,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004613582483012291,
      "loss": 5.7913,
      "mean_token_accuracy": 0.16080717444419862,
      "num_tokens": 45538543.0,
      "step": 22475
    },
    {
      "entropy": 6.158528614044189,
      "epoch": 1.9645197937603776,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004613407855289578,
      "loss": 5.9176,
      "mean_token_accuracy": 0.1550272136926651,
      "num_tokens": 45549298.0,
      "step": 22480
    },
    {
      "entropy": 6.210320377349854,
      "epoch": 1.9649567421130822,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004613233191825726,
      "loss": 5.9087,
      "mean_token_accuracy": 0.15596358776092528,
      "num_tokens": 45560064.0,
      "step": 22485
    },
    {
      "entropy": 6.206814813613891,
      "epoch": 1.965393690465787,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004613058492624084,
      "loss": 5.7668,
      "mean_token_accuracy": 0.16869568675756455,
      "num_tokens": 45569780.0,
      "step": 22490
    },
    {
      "entropy": 6.0283928394317625,
      "epoch": 1.9658306388184916,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004612883757688005,
      "loss": 5.7529,
      "mean_token_accuracy": 0.16148466467857361,
      "num_tokens": 45580256.0,
      "step": 22495
    },
    {
      "entropy": 6.151550579071045,
      "epoch": 1.9662675871711963,
      "grad_norm": 6.28125,
      "learning_rate": 0.00046127089870208395,
      "loss": 5.8535,
      "mean_token_accuracy": 0.1579982191324234,
      "num_tokens": 45590615.0,
      "step": 22500
    },
    {
      "entropy": 6.1368659973144535,
      "epoch": 1.9667045355239012,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004612534180625939,
      "loss": 5.7669,
      "mean_token_accuracy": 0.1597551003098488,
      "num_tokens": 45600215.0,
      "step": 22505
    },
    {
      "entropy": 6.114622974395752,
      "epoch": 1.9671414838766057,
      "grad_norm": 3.34375,
      "learning_rate": 0.00046123593385066574,
      "loss": 5.856,
      "mean_token_accuracy": 0.16159062683582306,
      "num_tokens": 45610484.0,
      "step": 22510
    },
    {
      "entropy": 6.1566473007202145,
      "epoch": 1.9675784322293106,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046121844606663475,
      "loss": 5.8667,
      "mean_token_accuracy": 0.15385104864835739,
      "num_tokens": 45620528.0,
      "step": 22515
    },
    {
      "entropy": 6.179942035675049,
      "epoch": 1.968015380582015,
      "grad_norm": 2.375,
      "learning_rate": 0.0004612009547108364,
      "loss": 5.8171,
      "mean_token_accuracy": 0.1596275433897972,
      "num_tokens": 45629985.0,
      "step": 22520
    },
    {
      "entropy": 6.174739360809326,
      "epoch": 1.96845232893472,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046118345978360615,
      "loss": 5.8756,
      "mean_token_accuracy": 0.1499561294913292,
      "num_tokens": 45639750.0,
      "step": 22525
    },
    {
      "entropy": 6.160564994812011,
      "epoch": 1.9688892772874245,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004611659612852796,
      "loss": 5.8478,
      "mean_token_accuracy": 0.15649292021989822,
      "num_tokens": 45650779.0,
      "step": 22530
    },
    {
      "entropy": 6.142979574203491,
      "epoch": 1.9693262256401294,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046114845921619234,
      "loss": 5.8037,
      "mean_token_accuracy": 0.15390663743019103,
      "num_tokens": 45660363.0,
      "step": 22535
    },
    {
      "entropy": 6.107254219055176,
      "epoch": 1.969763173992834,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004611309535766801,
      "loss": 5.9152,
      "mean_token_accuracy": 0.15373687148094178,
      "num_tokens": 45670474.0,
      "step": 22540
    },
    {
      "entropy": 6.150674724578858,
      "epoch": 1.9702001223455388,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004611134443670787,
      "loss": 5.7667,
      "mean_token_accuracy": 0.16771003603935242,
      "num_tokens": 45680157.0,
      "step": 22545
    },
    {
      "entropy": 6.206464767456055,
      "epoch": 1.9706370706982435,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004610959315877239,
      "loss": 5.8315,
      "mean_token_accuracy": 0.16163065508008004,
      "num_tokens": 45689340.0,
      "step": 22550
    },
    {
      "entropy": 6.2242378234863285,
      "epoch": 1.9710740190509481,
      "grad_norm": 3.125,
      "learning_rate": 0.00046107841523895153,
      "loss": 5.8585,
      "mean_token_accuracy": 0.15271333381533622,
      "num_tokens": 45699453.0,
      "step": 22555
    },
    {
      "entropy": 6.149085521697998,
      "epoch": 1.9715109674036528,
      "grad_norm": 3.28125,
      "learning_rate": 0.00046106089532109776,
      "loss": 5.8312,
      "mean_token_accuracy": 0.15881359875202178,
      "num_tokens": 45711513.0,
      "step": 22560
    },
    {
      "entropy": 6.196894216537475,
      "epoch": 1.9719479157563575,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004610433718344985,
      "loss": 5.9038,
      "mean_token_accuracy": 0.15674621611833572,
      "num_tokens": 45722939.0,
      "step": 22565
    },
    {
      "entropy": 6.1927818775177,
      "epoch": 1.9723848641090624,
      "grad_norm": 3.0,
      "learning_rate": 0.00046102584477948985,
      "loss": 5.8669,
      "mean_token_accuracy": 0.15491048693656922,
      "num_tokens": 45733099.0,
      "step": 22570
    },
    {
      "entropy": 6.137688493728637,
      "epoch": 1.972821812461767,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004610083141564079,
      "loss": 5.8958,
      "mean_token_accuracy": 0.1582522511482239,
      "num_tokens": 45743161.0,
      "step": 22575
    },
    {
      "entropy": 6.2205321311950685,
      "epoch": 1.9732587608144718,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046099077996558923,
      "loss": 5.9347,
      "mean_token_accuracy": 0.15454236418008804,
      "num_tokens": 45752903.0,
      "step": 22580
    },
    {
      "entropy": 6.1202386856079105,
      "epoch": 1.9736957091671763,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046097324220736993,
      "loss": 5.7488,
      "mean_token_accuracy": 0.16987803280353547,
      "num_tokens": 45763360.0,
      "step": 22585
    },
    {
      "entropy": 6.141084432601929,
      "epoch": 1.9741326575198812,
      "grad_norm": 2.125,
      "learning_rate": 0.0004609557008820863,
      "loss": 5.7395,
      "mean_token_accuracy": 0.16386134773492814,
      "num_tokens": 45773025.0,
      "step": 22590
    },
    {
      "entropy": 6.13622555732727,
      "epoch": 1.974569605872586,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004609381559900749,
      "loss": 5.7473,
      "mean_token_accuracy": 0.1641770288348198,
      "num_tokens": 45782844.0,
      "step": 22595
    },
    {
      "entropy": 6.118111515045166,
      "epoch": 1.9750065542252906,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046092060753167224,
      "loss": 5.8394,
      "mean_token_accuracy": 0.15988815277814866,
      "num_tokens": 45793058.0,
      "step": 22600
    },
    {
      "entropy": 6.20667405128479,
      "epoch": 1.9754435025779953,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004609030555072149,
      "loss": 5.8336,
      "mean_token_accuracy": 0.1579747051000595,
      "num_tokens": 45802865.0,
      "step": 22605
    },
    {
      "entropy": 6.147736692428589,
      "epoch": 1.9758804509307,
      "grad_norm": 2.671875,
      "learning_rate": 0.00046088549991703953,
      "loss": 5.7806,
      "mean_token_accuracy": 0.16558094471693038,
      "num_tokens": 45812981.0,
      "step": 22610
    },
    {
      "entropy": 6.085139179229737,
      "epoch": 1.9763173992834047,
      "grad_norm": 2.96875,
      "learning_rate": 0.000460867940761483,
      "loss": 5.7418,
      "mean_token_accuracy": 0.16667381674051285,
      "num_tokens": 45822477.0,
      "step": 22615
    },
    {
      "entropy": 6.0912535190582275,
      "epoch": 1.9767543476361094,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004608503780408818,
      "loss": 5.9267,
      "mean_token_accuracy": 0.15426602214574814,
      "num_tokens": 45833642.0,
      "step": 22620
    },
    {
      "entropy": 6.203613138198852,
      "epoch": 1.9771912959888143,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004608328117555731,
      "loss": 5.8144,
      "mean_token_accuracy": 0.15907812416553496,
      "num_tokens": 45843079.0,
      "step": 22625
    },
    {
      "entropy": 6.1847034931182865,
      "epoch": 1.9776282443415187,
      "grad_norm": 2.375,
      "learning_rate": 0.0004608152419058936,
      "loss": 5.8614,
      "mean_token_accuracy": 0.15452682226896286,
      "num_tokens": 45852987.0,
      "step": 22630
    },
    {
      "entropy": 6.184673929214478,
      "epoch": 1.9780651926942237,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046079766849218043,
      "loss": 5.8481,
      "mean_token_accuracy": 0.1553792543709278,
      "num_tokens": 45863148.0,
      "step": 22635
    },
    {
      "entropy": 6.113623189926147,
      "epoch": 1.9785021410469281,
      "grad_norm": 2.25,
      "learning_rate": 0.00046078009151477063,
      "loss": 5.8314,
      "mean_token_accuracy": 0.15932471081614494,
      "num_tokens": 45873430.0,
      "step": 22640
    },
    {
      "entropy": 6.231031513214111,
      "epoch": 1.978939089399633,
      "grad_norm": 3.25,
      "learning_rate": 0.00046076251097400126,
      "loss": 5.8226,
      "mean_token_accuracy": 0.15274938493967055,
      "num_tokens": 45883169.0,
      "step": 22645
    },
    {
      "entropy": 6.168148851394653,
      "epoch": 1.9793760377523377,
      "grad_norm": 4.9375,
      "learning_rate": 0.00046074492687020967,
      "loss": 5.8883,
      "mean_token_accuracy": 0.15789738446474075,
      "num_tokens": 45893948.0,
      "step": 22650
    },
    {
      "entropy": 6.0905163288116455,
      "epoch": 1.9798129861050424,
      "grad_norm": 2.71875,
      "learning_rate": 0.000460727339203733,
      "loss": 5.8607,
      "mean_token_accuracy": 0.16022102236747743,
      "num_tokens": 45904887.0,
      "step": 22655
    },
    {
      "entropy": 6.1656678199768065,
      "epoch": 1.980249934457747,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046070974797490875,
      "loss": 5.8507,
      "mean_token_accuracy": 0.15404208302497863,
      "num_tokens": 45914845.0,
      "step": 22660
    },
    {
      "entropy": 6.14314637184143,
      "epoch": 1.9806868828104518,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004606921531840741,
      "loss": 5.856,
      "mean_token_accuracy": 0.16169695258140565,
      "num_tokens": 45925104.0,
      "step": 22665
    },
    {
      "entropy": 6.072655391693115,
      "epoch": 1.9811238311631565,
      "grad_norm": 2.5,
      "learning_rate": 0.0004606745548315667,
      "loss": 5.7797,
      "mean_token_accuracy": 0.15863175988197326,
      "num_tokens": 45934934.0,
      "step": 22670
    },
    {
      "entropy": 6.120302963256836,
      "epoch": 1.9815607795158612,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004606569529177241,
      "loss": 5.8483,
      "mean_token_accuracy": 0.1597791939973831,
      "num_tokens": 45945449.0,
      "step": 22675
    },
    {
      "entropy": 6.2086395740509035,
      "epoch": 1.981997727868566,
      "grad_norm": 2.375,
      "learning_rate": 0.0004606393474428838,
      "loss": 5.8585,
      "mean_token_accuracy": 0.15306298583745956,
      "num_tokens": 45954897.0,
      "step": 22680
    },
    {
      "entropy": 6.209586811065674,
      "epoch": 1.9824346762212706,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046062173840738356,
      "loss": 5.8103,
      "mean_token_accuracy": 0.16279839128255844,
      "num_tokens": 45965085.0,
      "step": 22685
    },
    {
      "entropy": 6.088955974578857,
      "epoch": 1.9828716245739755,
      "grad_norm": 2.0,
      "learning_rate": 0.0004606041258115611,
      "loss": 5.8239,
      "mean_token_accuracy": 0.164753607660532,
      "num_tokens": 45974610.0,
      "step": 22690
    },
    {
      "entropy": 6.178389215469361,
      "epoch": 1.98330857292668,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004605865096557542,
      "loss": 5.8336,
      "mean_token_accuracy": 0.15523045510053635,
      "num_tokens": 45984922.0,
      "step": 22695
    },
    {
      "entropy": 6.147445201873779,
      "epoch": 1.9837455212793849,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004605688899403009,
      "loss": 5.8085,
      "mean_token_accuracy": 0.1617824599146843,
      "num_tokens": 45994591.0,
      "step": 22700
    },
    {
      "entropy": 6.178703022003174,
      "epoch": 1.9841824696320893,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046055126666553906,
      "loss": 5.943,
      "mean_token_accuracy": 0.1497579500079155,
      "num_tokens": 46005291.0,
      "step": 22705
    },
    {
      "entropy": 6.105194997787476,
      "epoch": 1.9846194179847942,
      "grad_norm": 3.375,
      "learning_rate": 0.00046053363983180657,
      "loss": 5.7995,
      "mean_token_accuracy": 0.1587808236479759,
      "num_tokens": 46014970.0,
      "step": 22710
    },
    {
      "entropy": 6.173995113372802,
      "epoch": 1.985056366337499,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046051600943944175,
      "loss": 5.8831,
      "mean_token_accuracy": 0.1572568953037262,
      "num_tokens": 46025042.0,
      "step": 22715
    },
    {
      "entropy": 6.2055806636810305,
      "epoch": 1.9854933146902036,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046049837548878264,
      "loss": 5.8914,
      "mean_token_accuracy": 0.15105953961610794,
      "num_tokens": 46035440.0,
      "step": 22720
    },
    {
      "entropy": 6.0994456768035885,
      "epoch": 1.9859302630429083,
      "grad_norm": 3.96875,
      "learning_rate": 0.00046048073798016746,
      "loss": 5.7903,
      "mean_token_accuracy": 0.1667264625430107,
      "num_tokens": 46045344.0,
      "step": 22725
    },
    {
      "entropy": 6.229736423492431,
      "epoch": 1.986367211395613,
      "grad_norm": 2.578125,
      "learning_rate": 0.00046046309691393455,
      "loss": 5.8844,
      "mean_token_accuracy": 0.15495181679725648,
      "num_tokens": 46055054.0,
      "step": 22730
    },
    {
      "entropy": 6.161386013031006,
      "epoch": 1.9868041597483177,
      "grad_norm": 3.125,
      "learning_rate": 0.00046044545229042226,
      "loss": 5.8573,
      "mean_token_accuracy": 0.15734347328543663,
      "num_tokens": 46063835.0,
      "step": 22735
    },
    {
      "entropy": 6.135680150985718,
      "epoch": 1.9872411081010224,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046042780410996906,
      "loss": 5.8787,
      "mean_token_accuracy": 0.15352381616830826,
      "num_tokens": 46074256.0,
      "step": 22740
    },
    {
      "entropy": 6.282417821884155,
      "epoch": 1.9876780564537273,
      "grad_norm": 2.515625,
      "learning_rate": 0.00046041015237291346,
      "loss": 5.9663,
      "mean_token_accuracy": 0.15703093707561494,
      "num_tokens": 46085563.0,
      "step": 22745
    },
    {
      "entropy": 6.192596054077148,
      "epoch": 1.9881150048064318,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046039249707959386,
      "loss": 5.8425,
      "mean_token_accuracy": 0.16281472966074945,
      "num_tokens": 46094956.0,
      "step": 22750
    },
    {
      "entropy": 6.154790592193604,
      "epoch": 1.9885519531591367,
      "grad_norm": 2.78125,
      "learning_rate": 0.00046037483823034906,
      "loss": 5.8456,
      "mean_token_accuracy": 0.16504292339086532,
      "num_tokens": 46105292.0,
      "step": 22755
    },
    {
      "entropy": 6.237738752365113,
      "epoch": 1.9889889015118412,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004603571758255177,
      "loss": 5.991,
      "mean_token_accuracy": 0.14883336126804353,
      "num_tokens": 46117254.0,
      "step": 22760
    },
    {
      "entropy": 6.198213720321656,
      "epoch": 1.989425849864546,
      "grad_norm": 2.375,
      "learning_rate": 0.00046033950986543865,
      "loss": 5.9216,
      "mean_token_accuracy": 0.15218820422887802,
      "num_tokens": 46127846.0,
      "step": 22765
    },
    {
      "entropy": 6.163524866104126,
      "epoch": 1.9898627982172508,
      "grad_norm": 4.09375,
      "learning_rate": 0.00046032184035045066,
      "loss": 5.8919,
      "mean_token_accuracy": 0.15595585852861404,
      "num_tokens": 46138476.0,
      "step": 22770
    },
    {
      "entropy": 6.1187413215637205,
      "epoch": 1.9902997465699555,
      "grad_norm": 3.21875,
      "learning_rate": 0.00046030416728089267,
      "loss": 5.772,
      "mean_token_accuracy": 0.16642215251922607,
      "num_tokens": 46147789.0,
      "step": 22775
    },
    {
      "entropy": 6.138396167755127,
      "epoch": 1.9907366949226601,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046028649065710367,
      "loss": 5.9187,
      "mean_token_accuracy": 0.15555381029844284,
      "num_tokens": 46158484.0,
      "step": 22780
    },
    {
      "entropy": 6.206918382644654,
      "epoch": 1.9911736432753648,
      "grad_norm": 1.875,
      "learning_rate": 0.00046026881047942273,
      "loss": 5.8743,
      "mean_token_accuracy": 0.15732206404209137,
      "num_tokens": 46168062.0,
      "step": 22785
    },
    {
      "entropy": 6.234512901306152,
      "epoch": 1.9916105916280695,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046025112674818883,
      "loss": 5.8305,
      "mean_token_accuracy": 0.1602164939045906,
      "num_tokens": 46178710.0,
      "step": 22790
    },
    {
      "entropy": 6.129577875137329,
      "epoch": 1.9920475399807742,
      "grad_norm": 2.734375,
      "learning_rate": 0.00046023343946374127,
      "loss": 5.8789,
      "mean_token_accuracy": 0.15484800338745117,
      "num_tokens": 46188728.0,
      "step": 22795
    },
    {
      "entropy": 6.116234874725341,
      "epoch": 1.9924844883334791,
      "grad_norm": 5.65625,
      "learning_rate": 0.00046021574862641925,
      "loss": 5.8602,
      "mean_token_accuracy": 0.1578035071492195,
      "num_tokens": 46199716.0,
      "step": 22800
    },
    {
      "entropy": 6.238975572586059,
      "epoch": 1.9929214366861836,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046019805423656217,
      "loss": 5.9279,
      "mean_token_accuracy": 0.1500931575894356,
      "num_tokens": 46210885.0,
      "step": 22805
    },
    {
      "entropy": 6.171203660964966,
      "epoch": 1.9933583850388885,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004601803562945094,
      "loss": 5.8517,
      "mean_token_accuracy": 0.1642708510160446,
      "num_tokens": 46220879.0,
      "step": 22810
    },
    {
      "entropy": 6.143125295639038,
      "epoch": 1.993795333391593,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004601626548006003,
      "loss": 5.86,
      "mean_token_accuracy": 0.15307185873389245,
      "num_tokens": 46230420.0,
      "step": 22815
    },
    {
      "entropy": 6.093084907531738,
      "epoch": 1.994232281744298,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004601449497551745,
      "loss": 5.8283,
      "mean_token_accuracy": 0.16307453960180282,
      "num_tokens": 46240548.0,
      "step": 22820
    },
    {
      "entropy": 6.185442781448364,
      "epoch": 1.9946692300970026,
      "grad_norm": 3.453125,
      "learning_rate": 0.00046012724115857153,
      "loss": 5.8294,
      "mean_token_accuracy": 0.16359004825353624,
      "num_tokens": 46250484.0,
      "step": 22825
    },
    {
      "entropy": 6.22773585319519,
      "epoch": 1.9951061784497073,
      "grad_norm": 2.75,
      "learning_rate": 0.00046010952901113107,
      "loss": 5.8789,
      "mean_token_accuracy": 0.1587586909532547,
      "num_tokens": 46261184.0,
      "step": 22830
    },
    {
      "entropy": 6.0908068180084225,
      "epoch": 1.995543126802412,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004600918133131928,
      "loss": 5.8296,
      "mean_token_accuracy": 0.16444566547870637,
      "num_tokens": 46271301.0,
      "step": 22835
    },
    {
      "entropy": 6.211796808242798,
      "epoch": 1.9959800751551167,
      "grad_norm": 2.890625,
      "learning_rate": 0.00046007409406509663,
      "loss": 5.8841,
      "mean_token_accuracy": 0.1540208339691162,
      "num_tokens": 46281201.0,
      "step": 22840
    },
    {
      "entropy": 6.185043716430664,
      "epoch": 1.9964170235078214,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004600563712671823,
      "loss": 5.8954,
      "mean_token_accuracy": 0.16124220490455626,
      "num_tokens": 46291969.0,
      "step": 22845
    },
    {
      "entropy": 6.034311294555664,
      "epoch": 1.996853971860526,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046003864491978984,
      "loss": 5.6901,
      "mean_token_accuracy": 0.17924728542566298,
      "num_tokens": 46301308.0,
      "step": 22850
    },
    {
      "entropy": 6.166991758346557,
      "epoch": 1.997290920213231,
      "grad_norm": 2.546875,
      "learning_rate": 0.00046002091502325926,
      "loss": 5.8552,
      "mean_token_accuracy": 0.15315467193722726,
      "num_tokens": 46311564.0,
      "step": 22855
    },
    {
      "entropy": 6.134119701385498,
      "epoch": 1.9977278685659354,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004600031815779306,
      "loss": 5.8161,
      "mean_token_accuracy": 0.16249994039535523,
      "num_tokens": 46321843.0,
      "step": 22860
    },
    {
      "entropy": 6.119950962066651,
      "epoch": 1.9981648169186403,
      "grad_norm": 3.328125,
      "learning_rate": 0.00045998544458414385,
      "loss": 5.772,
      "mean_token_accuracy": 0.16070916652679443,
      "num_tokens": 46331274.0,
      "step": 22865
    },
    {
      "entropy": 6.190118741989136,
      "epoch": 1.9986017652713448,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045996770404223937,
      "loss": 5.9007,
      "mean_token_accuracy": 0.14945872500538826,
      "num_tokens": 46341385.0,
      "step": 22870
    },
    {
      "entropy": 6.161245727539063,
      "epoch": 1.9990387136240497,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045994995995255746,
      "loss": 5.8257,
      "mean_token_accuracy": 0.1606876239180565,
      "num_tokens": 46350385.0,
      "step": 22875
    },
    {
      "entropy": 6.080937099456787,
      "epoch": 1.9994756619767542,
      "grad_norm": 2.21875,
      "learning_rate": 0.00045993221231543837,
      "loss": 5.7377,
      "mean_token_accuracy": 0.16089848726987838,
      "num_tokens": 46361111.0,
      "step": 22880
    },
    {
      "entropy": 6.144779348373413,
      "epoch": 1.999912610329459,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004599144611312226,
      "loss": 5.8822,
      "mean_token_accuracy": 0.15225949436426162,
      "num_tokens": 46371316.0,
      "step": 22885
    },
    {
      "entropy": 6.202335453033447,
      "epoch": 2.0003495586821636,
      "grad_norm": 1.90625,
      "learning_rate": 0.00045989670640025047,
      "loss": 5.8486,
      "mean_token_accuracy": 0.15764326304197313,
      "num_tokens": 46381151.0,
      "step": 22890
    },
    {
      "entropy": 6.194754362106323,
      "epoch": 2.0007865070348685,
      "grad_norm": 2.5625,
      "learning_rate": 0.00045987894812286276,
      "loss": 5.7941,
      "mean_token_accuracy": 0.16122581213712692,
      "num_tokens": 46391420.0,
      "step": 22895
    },
    {
      "entropy": 6.131979465484619,
      "epoch": 2.0012234553875734,
      "grad_norm": 2.140625,
      "learning_rate": 0.00045986118629939984,
      "loss": 5.7787,
      "mean_token_accuracy": 0.16208028495311738,
      "num_tokens": 46401575.0,
      "step": 22900
    },
    {
      "entropy": 6.180108690261841,
      "epoch": 2.001660403740278,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004598434209302025,
      "loss": 5.7745,
      "mean_token_accuracy": 0.16275254338979722,
      "num_tokens": 46411275.0,
      "step": 22905
    },
    {
      "entropy": 6.16912579536438,
      "epoch": 2.002097352092983,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004598256520156116,
      "loss": 5.8132,
      "mean_token_accuracy": 0.16121154129505158,
      "num_tokens": 46421179.0,
      "step": 22910
    },
    {
      "entropy": 6.125225400924682,
      "epoch": 2.0025343004456873,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004598078795559678,
      "loss": 5.85,
      "mean_token_accuracy": 0.1532897651195526,
      "num_tokens": 46431865.0,
      "step": 22915
    },
    {
      "entropy": 6.163967609405518,
      "epoch": 2.002971248798392,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045979010355161196,
      "loss": 5.9245,
      "mean_token_accuracy": 0.15001446157693862,
      "num_tokens": 46443398.0,
      "step": 22920
    },
    {
      "entropy": 6.282079458236694,
      "epoch": 2.0034081971510966,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004597723240028852,
      "loss": 5.9186,
      "mean_token_accuracy": 0.15054334700107574,
      "num_tokens": 46454818.0,
      "step": 22925
    },
    {
      "entropy": 6.257069444656372,
      "epoch": 2.0038451455038016,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004597545409101283,
      "loss": 5.8746,
      "mean_token_accuracy": 0.15505069494247437,
      "num_tokens": 46465231.0,
      "step": 22930
    },
    {
      "entropy": 6.216408538818359,
      "epoch": 2.004282093856506,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004597367542736826,
      "loss": 5.8186,
      "mean_token_accuracy": 0.16462078392505647,
      "num_tokens": 46476323.0,
      "step": 22935
    },
    {
      "entropy": 6.108573532104492,
      "epoch": 2.004719042209211,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004597189640938892,
      "loss": 5.783,
      "mean_token_accuracy": 0.15926224887371063,
      "num_tokens": 46486573.0,
      "step": 22940
    },
    {
      "entropy": 6.056864833831787,
      "epoch": 2.0051559905619154,
      "grad_norm": 3.375,
      "learning_rate": 0.0004597011703710892,
      "loss": 5.7835,
      "mean_token_accuracy": 0.1611382856965065,
      "num_tokens": 46497779.0,
      "step": 22945
    },
    {
      "entropy": 6.211396980285644,
      "epoch": 2.0055929389146203,
      "grad_norm": 2.625,
      "learning_rate": 0.000459683373105624,
      "loss": 5.8521,
      "mean_token_accuracy": 0.15294089764356614,
      "num_tokens": 46509074.0,
      "step": 22950
    },
    {
      "entropy": 6.104520511627197,
      "epoch": 2.006029887267325,
      "grad_norm": 1.921875,
      "learning_rate": 0.00045966557229783493,
      "loss": 5.6778,
      "mean_token_accuracy": 0.16837961450219155,
      "num_tokens": 46519292.0,
      "step": 22955
    },
    {
      "entropy": 6.08701639175415,
      "epoch": 2.0064668356200297,
      "grad_norm": 2.625,
      "learning_rate": 0.0004596477679480634,
      "loss": 5.7448,
      "mean_token_accuracy": 0.16718475967645646,
      "num_tokens": 46529689.0,
      "step": 22960
    },
    {
      "entropy": 6.179895687103271,
      "epoch": 2.0069037839727346,
      "grad_norm": 2.421875,
      "learning_rate": 0.00045962996005665096,
      "loss": 5.8187,
      "mean_token_accuracy": 0.15670831203460694,
      "num_tokens": 46538616.0,
      "step": 22965
    },
    {
      "entropy": 6.092795372009277,
      "epoch": 2.007340732325439,
      "grad_norm": 2.765625,
      "learning_rate": 0.00045961214862393914,
      "loss": 5.7268,
      "mean_token_accuracy": 0.16788861751556397,
      "num_tokens": 46548372.0,
      "step": 22970
    },
    {
      "entropy": 6.100507640838623,
      "epoch": 2.007777680678144,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004595943336502695,
      "loss": 5.7962,
      "mean_token_accuracy": 0.15275149345397948,
      "num_tokens": 46558991.0,
      "step": 22975
    },
    {
      "entropy": 6.178472805023193,
      "epoch": 2.0082146290308485,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004595765151359839,
      "loss": 5.8184,
      "mean_token_accuracy": 0.15995914936065675,
      "num_tokens": 46569531.0,
      "step": 22980
    },
    {
      "entropy": 6.22849850654602,
      "epoch": 2.0086515773835534,
      "grad_norm": 2.234375,
      "learning_rate": 0.000459558693081424,
      "loss": 5.839,
      "mean_token_accuracy": 0.16309329569339753,
      "num_tokens": 46579774.0,
      "step": 22985
    },
    {
      "entropy": 6.163317012786865,
      "epoch": 2.009088525736258,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004595408674869317,
      "loss": 5.7992,
      "mean_token_accuracy": 0.15803786367177963,
      "num_tokens": 46589810.0,
      "step": 22990
    },
    {
      "entropy": 6.142231369018555,
      "epoch": 2.0095254740889628,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004595230383528488,
      "loss": 5.7803,
      "mean_token_accuracy": 0.16549627780914306,
      "num_tokens": 46599347.0,
      "step": 22995
    },
    {
      "entropy": 6.3051018714904785,
      "epoch": 2.0099624224416672,
      "grad_norm": 2.375,
      "learning_rate": 0.00045950520567951734,
      "loss": 5.885,
      "mean_token_accuracy": 0.15258360132575036,
      "num_tokens": 46609926.0,
      "step": 23000
    },
    {
      "entropy": 6.1784782886505125,
      "epoch": 2.010399370794372,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004594873694672794,
      "loss": 5.8049,
      "mean_token_accuracy": 0.16225646585226058,
      "num_tokens": 46620736.0,
      "step": 23005
    },
    {
      "entropy": 6.120838737487793,
      "epoch": 2.0108363191470766,
      "grad_norm": 2.59375,
      "learning_rate": 0.00045946952971647696,
      "loss": 5.7575,
      "mean_token_accuracy": 0.15760584622621537,
      "num_tokens": 46630713.0,
      "step": 23010
    },
    {
      "entropy": 6.163970613479615,
      "epoch": 2.0112732674997815,
      "grad_norm": 3.125,
      "learning_rate": 0.0004594516864274524,
      "loss": 5.8011,
      "mean_token_accuracy": 0.1591937519609928,
      "num_tokens": 46641197.0,
      "step": 23015
    },
    {
      "entropy": 6.047294044494629,
      "epoch": 2.0117102158524864,
      "grad_norm": 2.953125,
      "learning_rate": 0.00045943383960054775,
      "loss": 5.6395,
      "mean_token_accuracy": 0.18268658816814423,
      "num_tokens": 46651325.0,
      "step": 23020
    },
    {
      "entropy": 6.1234128952026365,
      "epoch": 2.012147164205191,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045941598923610543,
      "loss": 5.8072,
      "mean_token_accuracy": 0.16258950233459474,
      "num_tokens": 46662100.0,
      "step": 23025
    },
    {
      "entropy": 6.126599502563477,
      "epoch": 2.012584112557896,
      "grad_norm": 2.234375,
      "learning_rate": 0.00045939813533446783,
      "loss": 5.7916,
      "mean_token_accuracy": 0.1651959612965584,
      "num_tokens": 46671460.0,
      "step": 23030
    },
    {
      "entropy": 6.21024522781372,
      "epoch": 2.0130210609106003,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004593802778959773,
      "loss": 5.8821,
      "mean_token_accuracy": 0.155816188454628,
      "num_tokens": 46681835.0,
      "step": 23035
    },
    {
      "entropy": 6.106692457199097,
      "epoch": 2.013458009263305,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004593624169209765,
      "loss": 5.7869,
      "mean_token_accuracy": 0.1587640807032585,
      "num_tokens": 46691976.0,
      "step": 23040
    },
    {
      "entropy": 6.135254192352295,
      "epoch": 2.0138949576160097,
      "grad_norm": 2.78125,
      "learning_rate": 0.00045934455240980793,
      "loss": 5.7951,
      "mean_token_accuracy": 0.16161486208438874,
      "num_tokens": 46702589.0,
      "step": 23045
    },
    {
      "entropy": 6.210773944854736,
      "epoch": 2.0143319059687146,
      "grad_norm": 2.625,
      "learning_rate": 0.0004593266843628143,
      "loss": 5.8883,
      "mean_token_accuracy": 0.1551184684038162,
      "num_tokens": 46712167.0,
      "step": 23050
    },
    {
      "entropy": 6.2066868305206295,
      "epoch": 2.014768854321419,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004593088127803382,
      "loss": 5.8232,
      "mean_token_accuracy": 0.15819681584835052,
      "num_tokens": 46722684.0,
      "step": 23055
    },
    {
      "entropy": 6.148368883132934,
      "epoch": 2.015205802674124,
      "grad_norm": 2.140625,
      "learning_rate": 0.00045929093766272253,
      "loss": 5.8165,
      "mean_token_accuracy": 0.15742237269878387,
      "num_tokens": 46733856.0,
      "step": 23060
    },
    {
      "entropy": 6.221851205825805,
      "epoch": 2.0156427510268284,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004592730590103101,
      "loss": 5.8675,
      "mean_token_accuracy": 0.15509819760918617,
      "num_tokens": 46744210.0,
      "step": 23065
    },
    {
      "entropy": 6.189997720718384,
      "epoch": 2.0160796993795334,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004592551768234438,
      "loss": 5.8129,
      "mean_token_accuracy": 0.16024121642112732,
      "num_tokens": 46754611.0,
      "step": 23070
    },
    {
      "entropy": 6.180946779251099,
      "epoch": 2.0165166477322383,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004592372911024667,
      "loss": 5.8832,
      "mean_token_accuracy": 0.1505200073122978,
      "num_tokens": 46765647.0,
      "step": 23075
    },
    {
      "entropy": 6.1566534519195555,
      "epoch": 2.0169535960849427,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00045921940184772184,
      "loss": 5.8691,
      "mean_token_accuracy": 0.15600946992635728,
      "num_tokens": 46776052.0,
      "step": 23080
    },
    {
      "entropy": 6.188437747955322,
      "epoch": 2.0173905444376476,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004592015090595523,
      "loss": 5.7923,
      "mean_token_accuracy": 0.1628482773900032,
      "num_tokens": 46786255.0,
      "step": 23085
    },
    {
      "entropy": 6.149437475204468,
      "epoch": 2.017827492790352,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004591836127383013,
      "loss": 5.788,
      "mean_token_accuracy": 0.16428371667861938,
      "num_tokens": 46795958.0,
      "step": 23090
    },
    {
      "entropy": 6.1469696998596195,
      "epoch": 2.018264441143057,
      "grad_norm": 3.171875,
      "learning_rate": 0.00045916571288431206,
      "loss": 5.8119,
      "mean_token_accuracy": 0.1559152953326702,
      "num_tokens": 46806508.0,
      "step": 23095
    },
    {
      "entropy": 6.1937150955200195,
      "epoch": 2.0187013894957615,
      "grad_norm": 2.390625,
      "learning_rate": 0.00045914780949792797,
      "loss": 5.8842,
      "mean_token_accuracy": 0.14996794015169143,
      "num_tokens": 46816868.0,
      "step": 23100
    },
    {
      "entropy": 6.207612562179565,
      "epoch": 2.0191383378484664,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004591299025794924,
      "loss": 5.8243,
      "mean_token_accuracy": 0.15816390961408616,
      "num_tokens": 46827591.0,
      "step": 23105
    },
    {
      "entropy": 6.143208694458008,
      "epoch": 2.019575286201171,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004591119921293488,
      "loss": 5.7621,
      "mean_token_accuracy": 0.16008330732584,
      "num_tokens": 46837394.0,
      "step": 23110
    },
    {
      "entropy": 6.13141713142395,
      "epoch": 2.020012234553876,
      "grad_norm": 3.765625,
      "learning_rate": 0.00045909407814784074,
      "loss": 5.7596,
      "mean_token_accuracy": 0.16560789048671723,
      "num_tokens": 46847881.0,
      "step": 23115
    },
    {
      "entropy": 6.1572651863098145,
      "epoch": 2.0204491829065803,
      "grad_norm": 1.6875,
      "learning_rate": 0.00045907616063531185,
      "loss": 5.8083,
      "mean_token_accuracy": 0.1582823485136032,
      "num_tokens": 46857728.0,
      "step": 23120
    },
    {
      "entropy": 6.226747131347656,
      "epoch": 2.020886131259285,
      "grad_norm": 5.46875,
      "learning_rate": 0.0004590582395921057,
      "loss": 5.8519,
      "mean_token_accuracy": 0.15773721039295197,
      "num_tokens": 46866991.0,
      "step": 23125
    },
    {
      "entropy": 6.1372754096984865,
      "epoch": 2.0213230796119896,
      "grad_norm": 2.328125,
      "learning_rate": 0.000459040315018566,
      "loss": 5.8561,
      "mean_token_accuracy": 0.1560954213142395,
      "num_tokens": 46876253.0,
      "step": 23130
    },
    {
      "entropy": 6.155477285385132,
      "epoch": 2.0217600279646946,
      "grad_norm": 2.75,
      "learning_rate": 0.00045902238691503675,
      "loss": 5.8177,
      "mean_token_accuracy": 0.15432549864053727,
      "num_tokens": 46887309.0,
      "step": 23135
    },
    {
      "entropy": 6.18152871131897,
      "epoch": 2.0221969763173995,
      "grad_norm": 2.140625,
      "learning_rate": 0.00045900445528186163,
      "loss": 5.8458,
      "mean_token_accuracy": 0.15418187230825425,
      "num_tokens": 46896722.0,
      "step": 23140
    },
    {
      "entropy": 6.150671577453613,
      "epoch": 2.022633924670104,
      "grad_norm": 2.609375,
      "learning_rate": 0.00045898652011938456,
      "loss": 5.8633,
      "mean_token_accuracy": 0.16035813614726066,
      "num_tokens": 46907519.0,
      "step": 23145
    },
    {
      "entropy": 6.120505142211914,
      "epoch": 2.023070873022809,
      "grad_norm": 1.796875,
      "learning_rate": 0.00045896858142794976,
      "loss": 5.772,
      "mean_token_accuracy": 0.1609134018421173,
      "num_tokens": 46917387.0,
      "step": 23150
    },
    {
      "entropy": 6.168593549728394,
      "epoch": 2.0235078213755133,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004589506392079011,
      "loss": 5.8001,
      "mean_token_accuracy": 0.16737936586141586,
      "num_tokens": 46926477.0,
      "step": 23155
    },
    {
      "entropy": 6.1706404209136965,
      "epoch": 2.0239447697282182,
      "grad_norm": 2.53125,
      "learning_rate": 0.00045893269345958284,
      "loss": 5.8163,
      "mean_token_accuracy": 0.16519944369792938,
      "num_tokens": 46935807.0,
      "step": 23160
    },
    {
      "entropy": 6.105763721466064,
      "epoch": 2.0243817180809227,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004589147441833391,
      "loss": 5.8096,
      "mean_token_accuracy": 0.15833812355995178,
      "num_tokens": 46946843.0,
      "step": 23165
    },
    {
      "entropy": 6.126455211639405,
      "epoch": 2.0248186664336276,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004588967913795142,
      "loss": 5.7948,
      "mean_token_accuracy": 0.16279045939445497,
      "num_tokens": 46956289.0,
      "step": 23170
    },
    {
      "entropy": 6.218300008773804,
      "epoch": 2.025255614786332,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045887883504845244,
      "loss": 5.7727,
      "mean_token_accuracy": 0.16329470574855803,
      "num_tokens": 46965606.0,
      "step": 23175
    },
    {
      "entropy": 6.099500322341919,
      "epoch": 2.025692563139037,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045886087519049837,
      "loss": 5.7139,
      "mean_token_accuracy": 0.17282634526491164,
      "num_tokens": 46974769.0,
      "step": 23180
    },
    {
      "entropy": 6.0933372497558596,
      "epoch": 2.0261295114917415,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045884291180599626,
      "loss": 5.8241,
      "mean_token_accuracy": 0.15294387936592102,
      "num_tokens": 46984753.0,
      "step": 23185
    },
    {
      "entropy": 6.117982673645019,
      "epoch": 2.0265664598444464,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00045882494489529085,
      "loss": 5.7721,
      "mean_token_accuracy": 0.16050271838903427,
      "num_tokens": 46995403.0,
      "step": 23190
    },
    {
      "entropy": 6.1582481384277346,
      "epoch": 2.0270034081971513,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004588069744587267,
      "loss": 5.7968,
      "mean_token_accuracy": 0.1626962959766388,
      "num_tokens": 47005054.0,
      "step": 23195
    },
    {
      "entropy": 6.149623203277588,
      "epoch": 2.0274403565498558,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004587890004966484,
      "loss": 5.8328,
      "mean_token_accuracy": 0.15874386578798294,
      "num_tokens": 47016315.0,
      "step": 23200
    },
    {
      "entropy": 6.187961626052856,
      "epoch": 2.0278773049025607,
      "grad_norm": 2.140625,
      "learning_rate": 0.00045877102300940077,
      "loss": 5.745,
      "mean_token_accuracy": 0.15337959975004195,
      "num_tokens": 47025981.0,
      "step": 23205
    },
    {
      "entropy": 6.108480834960938,
      "epoch": 2.028314253255265,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045875304199732863,
      "loss": 5.7584,
      "mean_token_accuracy": 0.16251400113105774,
      "num_tokens": 47035906.0,
      "step": 23210
    },
    {
      "entropy": 6.091802644729614,
      "epoch": 2.02875120160797,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045873505746077685,
      "loss": 5.8128,
      "mean_token_accuracy": 0.1556391105055809,
      "num_tokens": 47045884.0,
      "step": 23215
    },
    {
      "entropy": 6.194464635848999,
      "epoch": 2.0291881499606745,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004587170694000904,
      "loss": 5.807,
      "mean_token_accuracy": 0.16337863653898238,
      "num_tokens": 47055575.0,
      "step": 23220
    },
    {
      "entropy": 6.080639314651489,
      "epoch": 2.0296250983133794,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004586990778156144,
      "loss": 5.8391,
      "mean_token_accuracy": 0.15741609707474707,
      "num_tokens": 47065484.0,
      "step": 23225
    },
    {
      "entropy": 6.2175740718841555,
      "epoch": 2.030062046666084,
      "grad_norm": 1.75,
      "learning_rate": 0.00045868108270769373,
      "loss": 5.7784,
      "mean_token_accuracy": 0.16319801211357116,
      "num_tokens": 47076604.0,
      "step": 23230
    },
    {
      "entropy": 6.183162260055542,
      "epoch": 2.030498995018789,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004586630840766736,
      "loss": 5.7801,
      "mean_token_accuracy": 0.16354157030582428,
      "num_tokens": 47086452.0,
      "step": 23235
    },
    {
      "entropy": 6.073337554931641,
      "epoch": 2.0309359433714933,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045864508192289933,
      "loss": 5.8638,
      "mean_token_accuracy": 0.15942336320877076,
      "num_tokens": 47096425.0,
      "step": 23240
    },
    {
      "entropy": 6.16868109703064,
      "epoch": 2.031372891724198,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045862707624671616,
      "loss": 5.8254,
      "mean_token_accuracy": 0.15929411053657533,
      "num_tokens": 47107214.0,
      "step": 23245
    },
    {
      "entropy": 6.191125345230103,
      "epoch": 2.031809840076903,
      "grad_norm": 1.890625,
      "learning_rate": 0.00045860906704846944,
      "loss": 5.7942,
      "mean_token_accuracy": 0.1570572480559349,
      "num_tokens": 47117498.0,
      "step": 23250
    },
    {
      "entropy": 6.115081167221069,
      "epoch": 2.0322467884296076,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045859105432850464,
      "loss": 5.7604,
      "mean_token_accuracy": 0.16413535326719284,
      "num_tokens": 47128031.0,
      "step": 23255
    },
    {
      "entropy": 6.192137384414673,
      "epoch": 2.0326837367823125,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004585730380871672,
      "loss": 5.8182,
      "mean_token_accuracy": 0.16113633215427398,
      "num_tokens": 47138347.0,
      "step": 23260
    },
    {
      "entropy": 6.079470443725586,
      "epoch": 2.033120685135017,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004585550183248027,
      "loss": 5.7195,
      "mean_token_accuracy": 0.16117189228534698,
      "num_tokens": 47147642.0,
      "step": 23265
    },
    {
      "entropy": 6.113587665557861,
      "epoch": 2.033557633487722,
      "grad_norm": 2.734375,
      "learning_rate": 0.00045853699504175673,
      "loss": 5.8204,
      "mean_token_accuracy": 0.16280135214328767,
      "num_tokens": 47156855.0,
      "step": 23270
    },
    {
      "entropy": 6.176898860931397,
      "epoch": 2.0339945818404264,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045851896823837505,
      "loss": 5.7965,
      "mean_token_accuracy": 0.1565664991736412,
      "num_tokens": 47167030.0,
      "step": 23275
    },
    {
      "entropy": 6.179897832870483,
      "epoch": 2.0344315301931313,
      "grad_norm": 3.109375,
      "learning_rate": 0.00045850093791500334,
      "loss": 5.8285,
      "mean_token_accuracy": 0.1587355002760887,
      "num_tokens": 47177471.0,
      "step": 23280
    },
    {
      "entropy": 6.186842441558838,
      "epoch": 2.0348684785458357,
      "grad_norm": 5.0625,
      "learning_rate": 0.00045848290407198754,
      "loss": 5.8779,
      "mean_token_accuracy": 0.1528676263988018,
      "num_tokens": 47187911.0,
      "step": 23285
    },
    {
      "entropy": 6.133584833145141,
      "epoch": 2.0353054268985407,
      "grad_norm": 2.0,
      "learning_rate": 0.0004584648667096736,
      "loss": 5.774,
      "mean_token_accuracy": 0.16082483232021333,
      "num_tokens": 47198622.0,
      "step": 23290
    },
    {
      "entropy": 6.163207769393921,
      "epoch": 2.035742375251245,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004584468258284072,
      "loss": 5.8486,
      "mean_token_accuracy": 0.16173560917377472,
      "num_tokens": 47207593.0,
      "step": 23295
    },
    {
      "entropy": 6.194510459899902,
      "epoch": 2.03617932360395,
      "grad_norm": 2.125,
      "learning_rate": 0.0004584287814285347,
      "loss": 5.8219,
      "mean_token_accuracy": 0.15437747687101364,
      "num_tokens": 47218734.0,
      "step": 23300
    },
    {
      "entropy": 6.178821229934693,
      "epoch": 2.0366162719566545,
      "grad_norm": 2.265625,
      "learning_rate": 0.00045841073351040196,
      "loss": 5.8668,
      "mean_token_accuracy": 0.15442686527967453,
      "num_tokens": 47228662.0,
      "step": 23305
    },
    {
      "entropy": 6.2735456943511965,
      "epoch": 2.0370532203093594,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004583926820743553,
      "loss": 5.9325,
      "mean_token_accuracy": 0.153537617623806,
      "num_tokens": 47238460.0,
      "step": 23310
    },
    {
      "entropy": 6.1613153457641605,
      "epoch": 2.0374901686620643,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004583746271207409,
      "loss": 5.8328,
      "mean_token_accuracy": 0.1484882190823555,
      "num_tokens": 47247789.0,
      "step": 23315
    },
    {
      "entropy": 6.156906414031982,
      "epoch": 2.037927117014769,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045835656864990504,
      "loss": 5.8224,
      "mean_token_accuracy": 0.1644187316298485,
      "num_tokens": 47257901.0,
      "step": 23320
    },
    {
      "entropy": 6.273704481124878,
      "epoch": 2.0383640653674737,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004583385066621941,
      "loss": 5.8203,
      "mean_token_accuracy": 0.1575300171971321,
      "num_tokens": 47268362.0,
      "step": 23325
    },
    {
      "entropy": 6.165722465515136,
      "epoch": 2.038801013720178,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045832044115795464,
      "loss": 5.815,
      "mean_token_accuracy": 0.163201205432415,
      "num_tokens": 47278297.0,
      "step": 23330
    },
    {
      "entropy": 6.117983198165893,
      "epoch": 2.039237962072883,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045830237213753287,
      "loss": 5.8437,
      "mean_token_accuracy": 0.15382062941789626,
      "num_tokens": 47288795.0,
      "step": 23335
    },
    {
      "entropy": 6.154676294326782,
      "epoch": 2.0396749104255876,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004582842996012757,
      "loss": 5.774,
      "mean_token_accuracy": 0.1639147147536278,
      "num_tokens": 47298373.0,
      "step": 23340
    },
    {
      "entropy": 6.14943904876709,
      "epoch": 2.0401118587782925,
      "grad_norm": 2.328125,
      "learning_rate": 0.00045826622354952956,
      "loss": 5.8671,
      "mean_token_accuracy": 0.15499431490898133,
      "num_tokens": 47308160.0,
      "step": 23345
    },
    {
      "entropy": 6.136783695220947,
      "epoch": 2.040548807130997,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045824814398264126,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16289520114660264,
      "num_tokens": 47317626.0,
      "step": 23350
    },
    {
      "entropy": 6.195802068710327,
      "epoch": 2.040985755483702,
      "grad_norm": 2.375,
      "learning_rate": 0.00045823006090095757,
      "loss": 5.8781,
      "mean_token_accuracy": 0.1599137470126152,
      "num_tokens": 47327221.0,
      "step": 23355
    },
    {
      "entropy": 6.162030029296875,
      "epoch": 2.0414227038364063,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004582119743048253,
      "loss": 5.8373,
      "mean_token_accuracy": 0.1602051943540573,
      "num_tokens": 47337385.0,
      "step": 23360
    },
    {
      "entropy": 6.105412769317627,
      "epoch": 2.0418596521891113,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004581938841945912,
      "loss": 5.801,
      "mean_token_accuracy": 0.15931429266929625,
      "num_tokens": 47346894.0,
      "step": 23365
    },
    {
      "entropy": 6.201750040054321,
      "epoch": 2.042296600541816,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045817579057060254,
      "loss": 5.8263,
      "mean_token_accuracy": 0.15982370674610138,
      "num_tokens": 47356192.0,
      "step": 23370
    },
    {
      "entropy": 6.167966079711914,
      "epoch": 2.0427335488945206,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004581576934332063,
      "loss": 5.7562,
      "mean_token_accuracy": 0.15810691714286804,
      "num_tokens": 47366704.0,
      "step": 23375
    },
    {
      "entropy": 6.120054292678833,
      "epoch": 2.0431704972472255,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004581395927827494,
      "loss": 5.7763,
      "mean_token_accuracy": 0.16458745151758195,
      "num_tokens": 47377068.0,
      "step": 23380
    },
    {
      "entropy": 6.10514087677002,
      "epoch": 2.04360744559993,
      "grad_norm": 2.125,
      "learning_rate": 0.00045812148861957923,
      "loss": 5.7891,
      "mean_token_accuracy": 0.1684432104229927,
      "num_tokens": 47387268.0,
      "step": 23385
    },
    {
      "entropy": 6.182196998596192,
      "epoch": 2.044044393952635,
      "grad_norm": 2.625,
      "learning_rate": 0.0004581033809440429,
      "loss": 5.7687,
      "mean_token_accuracy": 0.16317813992500305,
      "num_tokens": 47397508.0,
      "step": 23390
    },
    {
      "entropy": 6.102484846115113,
      "epoch": 2.0444813423053394,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004580852697564878,
      "loss": 5.8173,
      "mean_token_accuracy": 0.15522830933332443,
      "num_tokens": 47407871.0,
      "step": 23395
    },
    {
      "entropy": 6.077707672119141,
      "epoch": 2.0449182906580443,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00045806715505726135,
      "loss": 5.8746,
      "mean_token_accuracy": 0.156168369948864,
      "num_tokens": 47417089.0,
      "step": 23400
    },
    {
      "entropy": 6.173311758041382,
      "epoch": 2.045355239010749,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045804903684671083,
      "loss": 5.7756,
      "mean_token_accuracy": 0.16449107974767685,
      "num_tokens": 47426746.0,
      "step": 23405
    },
    {
      "entropy": 6.218811321258545,
      "epoch": 2.0457921873634537,
      "grad_norm": 2.59375,
      "learning_rate": 0.00045803091512518397,
      "loss": 5.8455,
      "mean_token_accuracy": 0.16153171360492707,
      "num_tokens": 47437873.0,
      "step": 23410
    },
    {
      "entropy": 6.101855754852295,
      "epoch": 2.046229135716158,
      "grad_norm": 2.4375,
      "learning_rate": 0.00045801278989302815,
      "loss": 5.8031,
      "mean_token_accuracy": 0.16233762949705124,
      "num_tokens": 47448520.0,
      "step": 23415
    },
    {
      "entropy": 6.158153867721557,
      "epoch": 2.046666084068863,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004579946611505912,
      "loss": 5.8264,
      "mean_token_accuracy": 0.15771035403013228,
      "num_tokens": 47458323.0,
      "step": 23420
    },
    {
      "entropy": 6.193030738830567,
      "epoch": 2.0471030324215675,
      "grad_norm": 2.625,
      "learning_rate": 0.0004579765288982207,
      "loss": 5.9041,
      "mean_token_accuracy": 0.15633908957242965,
      "num_tokens": 47468061.0,
      "step": 23425
    },
    {
      "entropy": 6.163549184799194,
      "epoch": 2.0475399807742725,
      "grad_norm": 3.15625,
      "learning_rate": 0.00045795839313626455,
      "loss": 5.8413,
      "mean_token_accuracy": 0.15857867747545243,
      "num_tokens": 47477541.0,
      "step": 23430
    },
    {
      "entropy": 6.0528160572052006,
      "epoch": 2.0479769291269774,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004579402538650705,
      "loss": 5.7882,
      "mean_token_accuracy": 0.1588955208659172,
      "num_tokens": 47488570.0,
      "step": 23435
    },
    {
      "entropy": 6.126381254196167,
      "epoch": 2.048413877479682,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004579221110849866,
      "loss": 5.804,
      "mean_token_accuracy": 0.15793004631996155,
      "num_tokens": 47498386.0,
      "step": 23440
    },
    {
      "entropy": 6.18155426979065,
      "epoch": 2.0488508258323868,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004579039647963607,
      "loss": 5.7068,
      "mean_token_accuracy": 0.168092480301857,
      "num_tokens": 47507292.0,
      "step": 23445
    },
    {
      "entropy": 6.143343210220337,
      "epoch": 2.0492877741850912,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045788581499954094,
      "loss": 5.7887,
      "mean_token_accuracy": 0.16382915377616883,
      "num_tokens": 47516971.0,
      "step": 23450
    },
    {
      "entropy": 6.048208427429199,
      "epoch": 2.049724722537796,
      "grad_norm": 2.140625,
      "learning_rate": 0.00045786766169487536,
      "loss": 5.6688,
      "mean_token_accuracy": 0.17484075874090194,
      "num_tokens": 47527233.0,
      "step": 23455
    },
    {
      "entropy": 6.115447473526001,
      "epoch": 2.0501616708905006,
      "grad_norm": 2.4375,
      "learning_rate": 0.00045784950488271227,
      "loss": 5.767,
      "mean_token_accuracy": 0.15939561873674393,
      "num_tokens": 47537542.0,
      "step": 23460
    },
    {
      "entropy": 6.1705821514129635,
      "epoch": 2.0505986192432055,
      "grad_norm": 2.328125,
      "learning_rate": 0.00045783134456339985,
      "loss": 5.7566,
      "mean_token_accuracy": 0.16165938079357148,
      "num_tokens": 47546613.0,
      "step": 23465
    },
    {
      "entropy": 6.1173656463623045,
      "epoch": 2.05103556759591,
      "grad_norm": 2.0625,
      "learning_rate": 0.00045781318073728647,
      "loss": 5.7648,
      "mean_token_accuracy": 0.16748571693897246,
      "num_tokens": 47555462.0,
      "step": 23470
    },
    {
      "entropy": 6.1729748249053955,
      "epoch": 2.051472515948615,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004577950134047204,
      "loss": 5.7607,
      "mean_token_accuracy": 0.1577707201242447,
      "num_tokens": 47565429.0,
      "step": 23475
    },
    {
      "entropy": 6.183677721023559,
      "epoch": 2.0519094643013194,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004577768425660503,
      "loss": 5.9343,
      "mean_token_accuracy": 0.14996827095746995,
      "num_tokens": 47576021.0,
      "step": 23480
    },
    {
      "entropy": 6.166540288925171,
      "epoch": 2.0523464126540243,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004577586682216245,
      "loss": 5.8261,
      "mean_token_accuracy": 0.1556169293820858,
      "num_tokens": 47586016.0,
      "step": 23485
    },
    {
      "entropy": 6.109238719940185,
      "epoch": 2.052783361006729,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004577404903717917,
      "loss": 5.7896,
      "mean_token_accuracy": 0.15864626318216324,
      "num_tokens": 47595673.0,
      "step": 23490
    },
    {
      "entropy": 6.0972710132598875,
      "epoch": 2.0532203093594337,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004577223090169006,
      "loss": 5.7378,
      "mean_token_accuracy": 0.1644854411482811,
      "num_tokens": 47605252.0,
      "step": 23495
    },
    {
      "entropy": 6.103595018386841,
      "epoch": 2.0536572577121386,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004577041241572998,
      "loss": 5.7806,
      "mean_token_accuracy": 0.16001876890659333,
      "num_tokens": 47614726.0,
      "step": 23500
    },
    {
      "entropy": 6.1249924659729,
      "epoch": 2.054094206064843,
      "grad_norm": 3.125,
      "learning_rate": 0.00045768593579333816,
      "loss": 5.8191,
      "mean_token_accuracy": 0.16426118612289428,
      "num_tokens": 47625618.0,
      "step": 23505
    },
    {
      "entropy": 6.200565052032471,
      "epoch": 2.054531154417548,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004576677439253646,
      "loss": 5.8507,
      "mean_token_accuracy": 0.157753287255764,
      "num_tokens": 47635702.0,
      "step": 23510
    },
    {
      "entropy": 6.125229215621948,
      "epoch": 2.0549681027702524,
      "grad_norm": 2.90625,
      "learning_rate": 0.00045764954855372796,
      "loss": 5.8491,
      "mean_token_accuracy": 0.16131159886717797,
      "num_tokens": 47647301.0,
      "step": 23515
    },
    {
      "entropy": 6.138525915145874,
      "epoch": 2.0554050511229573,
      "grad_norm": 2.125,
      "learning_rate": 0.0004576313496787772,
      "loss": 5.9016,
      "mean_token_accuracy": 0.15745027214288712,
      "num_tokens": 47656844.0,
      "step": 23520
    },
    {
      "entropy": 6.231021690368652,
      "epoch": 2.055841999475662,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004576131473008616,
      "loss": 5.7669,
      "mean_token_accuracy": 0.16236383020877837,
      "num_tokens": 47666196.0,
      "step": 23525
    },
    {
      "entropy": 6.125605201721191,
      "epoch": 2.0562789478283667,
      "grad_norm": 2.375,
      "learning_rate": 0.0004575949414203301,
      "loss": 5.7665,
      "mean_token_accuracy": 0.163436858355999,
      "num_tokens": 47675329.0,
      "step": 23530
    },
    {
      "entropy": 6.040912055969239,
      "epoch": 2.056715896181071,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045757673203753195,
      "loss": 5.7824,
      "mean_token_accuracy": 0.15985868722200394,
      "num_tokens": 47686806.0,
      "step": 23535
    },
    {
      "entropy": 6.154395818710327,
      "epoch": 2.057152844533776,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004575585191528164,
      "loss": 5.7791,
      "mean_token_accuracy": 0.16334619969129563,
      "num_tokens": 47696664.0,
      "step": 23540
    },
    {
      "entropy": 6.125362157821655,
      "epoch": 2.057589792886481,
      "grad_norm": 1.859375,
      "learning_rate": 0.00045754030276653275,
      "loss": 5.8583,
      "mean_token_accuracy": 0.15476511418819427,
      "num_tokens": 47707257.0,
      "step": 23545
    },
    {
      "entropy": 6.171913766860962,
      "epoch": 2.0580267412391855,
      "grad_norm": 2.125,
      "learning_rate": 0.0004575220828790305,
      "loss": 5.7534,
      "mean_token_accuracy": 0.16428215205669403,
      "num_tokens": 47716345.0,
      "step": 23550
    },
    {
      "entropy": 6.120296478271484,
      "epoch": 2.0584636895918904,
      "grad_norm": 4.34375,
      "learning_rate": 0.00045750385949065904,
      "loss": 5.7879,
      "mean_token_accuracy": 0.1588292360305786,
      "num_tokens": 47727779.0,
      "step": 23555
    },
    {
      "entropy": 6.101216554641724,
      "epoch": 2.058900637944595,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004574856326017679,
      "loss": 5.8427,
      "mean_token_accuracy": 0.1605902463197708,
      "num_tokens": 47737591.0,
      "step": 23560
    },
    {
      "entropy": 6.201908779144287,
      "epoch": 2.0593375862973,
      "grad_norm": 2.578125,
      "learning_rate": 0.00045746740221270664,
      "loss": 5.7803,
      "mean_token_accuracy": 0.16222186684608458,
      "num_tokens": 47746871.0,
      "step": 23565
    },
    {
      "entropy": 6.19092755317688,
      "epoch": 2.0597745346500043,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004574491683238251,
      "loss": 5.8349,
      "mean_token_accuracy": 0.15914518162608146,
      "num_tokens": 47757089.0,
      "step": 23570
    },
    {
      "entropy": 6.167212772369385,
      "epoch": 2.060211483002709,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004574309309354729,
      "loss": 5.8305,
      "mean_token_accuracy": 0.16021551489830016,
      "num_tokens": 47766853.0,
      "step": 23575
    },
    {
      "entropy": 6.153849697113037,
      "epoch": 2.0606484313554136,
      "grad_norm": 2.515625,
      "learning_rate": 0.00045741269004799984,
      "loss": 5.7951,
      "mean_token_accuracy": 0.1617000013589859,
      "num_tokens": 47776549.0,
      "step": 23580
    },
    {
      "entropy": 6.1530732154846195,
      "epoch": 2.0610853797081186,
      "grad_norm": 1.9375,
      "learning_rate": 0.00045739444566175584,
      "loss": 5.8221,
      "mean_token_accuracy": 0.15731338784098625,
      "num_tokens": 47786641.0,
      "step": 23585
    },
    {
      "entropy": 6.166473388671875,
      "epoch": 2.061522328060823,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045737619777709077,
      "loss": 5.8507,
      "mean_token_accuracy": 0.15677983909845353,
      "num_tokens": 47796678.0,
      "step": 23590
    },
    {
      "entropy": 6.196117353439331,
      "epoch": 2.061959276413528,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004573579463943546,
      "loss": 5.8028,
      "mean_token_accuracy": 0.1584724709391594,
      "num_tokens": 47806678.0,
      "step": 23595
    },
    {
      "entropy": 6.16224217414856,
      "epoch": 2.062396224766233,
      "grad_norm": 2.875,
      "learning_rate": 0.00045733969151389754,
      "loss": 5.8083,
      "mean_token_accuracy": 0.16224002838134766,
      "num_tokens": 47816619.0,
      "step": 23600
    },
    {
      "entropy": 6.068272447586059,
      "epoch": 2.0628331731189373,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004573214331360696,
      "loss": 5.8053,
      "mean_token_accuracy": 0.15947069674730302,
      "num_tokens": 47825437.0,
      "step": 23605
    },
    {
      "entropy": 6.134217357635498,
      "epoch": 2.0632701214716422,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004573031712612211,
      "loss": 5.7602,
      "mean_token_accuracy": 0.16240641623735427,
      "num_tokens": 47835604.0,
      "step": 23610
    },
    {
      "entropy": 6.138091230392456,
      "epoch": 2.0637070698243467,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004572849058897022,
      "loss": 5.7386,
      "mean_token_accuracy": 0.17278169095516205,
      "num_tokens": 47845299.0,
      "step": 23615
    },
    {
      "entropy": 6.042770051956177,
      "epoch": 2.0641440181770516,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004572666370218633,
      "loss": 5.7892,
      "mean_token_accuracy": 0.1675105407834053,
      "num_tokens": 47855293.0,
      "step": 23620
    },
    {
      "entropy": 6.110277271270752,
      "epoch": 2.064580966529756,
      "grad_norm": 3.3125,
      "learning_rate": 0.00045724836465805483,
      "loss": 5.8452,
      "mean_token_accuracy": 0.15855987817049028,
      "num_tokens": 47864809.0,
      "step": 23625
    },
    {
      "entropy": 6.137280416488648,
      "epoch": 2.065017914882461,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004572300887986272,
      "loss": 5.8307,
      "mean_token_accuracy": 0.15469652712345122,
      "num_tokens": 47874032.0,
      "step": 23630
    },
    {
      "entropy": 6.205067300796509,
      "epoch": 2.0654548632351655,
      "grad_norm": 1.84375,
      "learning_rate": 0.00045721180944393095,
      "loss": 5.7771,
      "mean_token_accuracy": 0.16715073734521865,
      "num_tokens": 47884465.0,
      "step": 23635
    },
    {
      "entropy": 6.056624937057495,
      "epoch": 2.0658918115878704,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004571935265943168,
      "loss": 5.7478,
      "mean_token_accuracy": 0.1754244178533554,
      "num_tokens": 47894030.0,
      "step": 23640
    },
    {
      "entropy": 6.131896066665649,
      "epoch": 2.066328759940575,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045717524025013526,
      "loss": 5.8062,
      "mean_token_accuracy": 0.15609943196177484,
      "num_tokens": 47904338.0,
      "step": 23645
    },
    {
      "entropy": 6.2024681091308596,
      "epoch": 2.0667657082932798,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004571569504117372,
      "loss": 5.8627,
      "mean_token_accuracy": 0.15901922434568405,
      "num_tokens": 47914384.0,
      "step": 23650
    },
    {
      "entropy": 6.149119853973389,
      "epoch": 2.0672026566459842,
      "grad_norm": 2.0,
      "learning_rate": 0.0004571386570794734,
      "loss": 5.7361,
      "mean_token_accuracy": 0.15901260823011398,
      "num_tokens": 47924386.0,
      "step": 23655
    },
    {
      "entropy": 6.11936321258545,
      "epoch": 2.067639604998689,
      "grad_norm": 3.4375,
      "learning_rate": 0.00045712036025369465,
      "loss": 5.8112,
      "mean_token_accuracy": 0.16088489443063736,
      "num_tokens": 47935379.0,
      "step": 23660
    },
    {
      "entropy": 6.101821041107177,
      "epoch": 2.068076553351394,
      "grad_norm": 2.6875,
      "learning_rate": 0.000457102059934752,
      "loss": 5.7822,
      "mean_token_accuracy": 0.16332819759845735,
      "num_tokens": 47945657.0,
      "step": 23665
    },
    {
      "entropy": 6.108335876464844,
      "epoch": 2.0685135017040985,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045708375612299646,
      "loss": 5.7847,
      "mean_token_accuracy": 0.15956311970949172,
      "num_tokens": 47956294.0,
      "step": 23670
    },
    {
      "entropy": 6.16575403213501,
      "epoch": 2.0689504500568034,
      "grad_norm": 3.015625,
      "learning_rate": 0.00045706544881877895,
      "loss": 5.8835,
      "mean_token_accuracy": 0.1523364193737507,
      "num_tokens": 47967362.0,
      "step": 23675
    },
    {
      "entropy": 6.20620436668396,
      "epoch": 2.069387398409508,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004570471380224508,
      "loss": 5.8012,
      "mean_token_accuracy": 0.16327320039272308,
      "num_tokens": 47977521.0,
      "step": 23680
    },
    {
      "entropy": 6.114765644073486,
      "epoch": 2.069824346762213,
      "grad_norm": 2.546875,
      "learning_rate": 0.00045702882373436317,
      "loss": 5.8229,
      "mean_token_accuracy": 0.16038476973772048,
      "num_tokens": 47988485.0,
      "step": 23685
    },
    {
      "entropy": 6.105095624923706,
      "epoch": 2.0702612951149173,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004570105059548673,
      "loss": 5.8405,
      "mean_token_accuracy": 0.15482570230960846,
      "num_tokens": 47998696.0,
      "step": 23690
    },
    {
      "entropy": 6.176537656784058,
      "epoch": 2.070698243467622,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004569921846843146,
      "loss": 5.6482,
      "mean_token_accuracy": 0.17268780767917633,
      "num_tokens": 48007843.0,
      "step": 23695
    },
    {
      "entropy": 6.115229749679566,
      "epoch": 2.0711351918203267,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004569738599230563,
      "loss": 5.7641,
      "mean_token_accuracy": 0.16025924384593965,
      "num_tokens": 48017579.0,
      "step": 23700
    },
    {
      "entropy": 6.0986837387084964,
      "epoch": 2.0715721401730316,
      "grad_norm": 2.453125,
      "learning_rate": 0.00045695553167144417,
      "loss": 5.7755,
      "mean_token_accuracy": 0.16430808901786803,
      "num_tokens": 48027267.0,
      "step": 23705
    },
    {
      "entropy": 6.123270273208618,
      "epoch": 2.072009088525736,
      "grad_norm": 2.140625,
      "learning_rate": 0.00045693719992982956,
      "loss": 5.8058,
      "mean_token_accuracy": 0.1611434042453766,
      "num_tokens": 48038064.0,
      "step": 23710
    },
    {
      "entropy": 6.21777753829956,
      "epoch": 2.072446036878441,
      "grad_norm": 2.75,
      "learning_rate": 0.00045691886469856406,
      "loss": 5.8883,
      "mean_token_accuracy": 0.15099416226148604,
      "num_tokens": 48048316.0,
      "step": 23715
    },
    {
      "entropy": 6.139502286911011,
      "epoch": 2.072882985231146,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004569005259779995,
      "loss": 5.8306,
      "mean_token_accuracy": 0.16075243055820465,
      "num_tokens": 48059083.0,
      "step": 23720
    },
    {
      "entropy": 6.102025556564331,
      "epoch": 2.0733199335838504,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00045688218376848744,
      "loss": 5.7648,
      "mean_token_accuracy": 0.1604068696498871,
      "num_tokens": 48069184.0,
      "step": 23725
    },
    {
      "entropy": 6.116828918457031,
      "epoch": 2.0737568819365553,
      "grad_norm": 1.625,
      "learning_rate": 0.0004568638380703799,
      "loss": 5.8057,
      "mean_token_accuracy": 0.15720739662647248,
      "num_tokens": 48078737.0,
      "step": 23730
    },
    {
      "entropy": 6.16328911781311,
      "epoch": 2.0741938302892597,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004568454888840285,
      "loss": 5.8076,
      "mean_token_accuracy": 0.15637186765670777,
      "num_tokens": 48088581.0,
      "step": 23735
    },
    {
      "entropy": 6.0626777648925785,
      "epoch": 2.0746307786419647,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004568271362097855,
      "loss": 5.673,
      "mean_token_accuracy": 0.18589871674776076,
      "num_tokens": 48098941.0,
      "step": 23740
    },
    {
      "entropy": 6.145608234405517,
      "epoch": 2.075067726994669,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00045680878004800265,
      "loss": 5.8224,
      "mean_token_accuracy": 0.15179096013307572,
      "num_tokens": 48108773.0,
      "step": 23745
    },
    {
      "entropy": 6.176486492156982,
      "epoch": 2.075504675347374,
      "grad_norm": 2.0,
      "learning_rate": 0.0004567904203990322,
      "loss": 5.819,
      "mean_token_accuracy": 0.168433877825737,
      "num_tokens": 48117699.0,
      "step": 23750
    },
    {
      "entropy": 6.034690093994141,
      "epoch": 2.0759416237000785,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004567720572632261,
      "loss": 5.7287,
      "mean_token_accuracy": 0.16658338457345961,
      "num_tokens": 48127693.0,
      "step": 23755
    },
    {
      "entropy": 6.105871915817261,
      "epoch": 2.0763785720527834,
      "grad_norm": 1.84375,
      "learning_rate": 0.00045675369064093686,
      "loss": 5.8622,
      "mean_token_accuracy": 0.15401278287172318,
      "num_tokens": 48138073.0,
      "step": 23760
    },
    {
      "entropy": 6.109974718093872,
      "epoch": 2.076815520405488,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004567353205325165,
      "loss": 5.718,
      "mean_token_accuracy": 0.17248254418373107,
      "num_tokens": 48147504.0,
      "step": 23765
    },
    {
      "entropy": 6.173751306533814,
      "epoch": 2.077252468758193,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004567169469383175,
      "loss": 5.7805,
      "mean_token_accuracy": 0.15525670498609542,
      "num_tokens": 48156739.0,
      "step": 23770
    },
    {
      "entropy": 6.111525917053223,
      "epoch": 2.0776894171108973,
      "grad_norm": 2.125,
      "learning_rate": 0.00045669856985869225,
      "loss": 5.7762,
      "mean_token_accuracy": 0.16051966845989227,
      "num_tokens": 48166883.0,
      "step": 23775
    },
    {
      "entropy": 6.119926881790161,
      "epoch": 2.078126365463602,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004566801892939932,
      "loss": 5.7419,
      "mean_token_accuracy": 0.16560746729373932,
      "num_tokens": 48177738.0,
      "step": 23780
    },
    {
      "entropy": 6.1283904075622555,
      "epoch": 2.078563313816307,
      "grad_norm": 2.625,
      "learning_rate": 0.0004566618052445729,
      "loss": 5.7374,
      "mean_token_accuracy": 0.16230679154396058,
      "num_tokens": 48187979.0,
      "step": 23785
    },
    {
      "entropy": 6.091846609115601,
      "epoch": 2.0790002621690116,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000456643417710784,
      "loss": 5.8161,
      "mean_token_accuracy": 0.15725044012069703,
      "num_tokens": 48198579.0,
      "step": 23790
    },
    {
      "entropy": 6.171621561050415,
      "epoch": 2.0794372105217165,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00045662502669297915,
      "loss": 5.8875,
      "mean_token_accuracy": 0.1525994837284088,
      "num_tokens": 48208860.0,
      "step": 23795
    },
    {
      "entropy": 6.21701226234436,
      "epoch": 2.079874158874421,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004566066321915112,
      "loss": 5.8088,
      "mean_token_accuracy": 0.1629228338599205,
      "num_tokens": 48218842.0,
      "step": 23800
    },
    {
      "entropy": 6.131508016586304,
      "epoch": 2.080311107227126,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004565882342067328,
      "loss": 5.761,
      "mean_token_accuracy": 0.15796410739421846,
      "num_tokens": 48229082.0,
      "step": 23805
    },
    {
      "entropy": 6.1365217685699465,
      "epoch": 2.0807480555798303,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00045656983273899694,
      "loss": 5.8472,
      "mean_token_accuracy": 0.1580294832587242,
      "num_tokens": 48239158.0,
      "step": 23810
    },
    {
      "entropy": 6.1310357570648195,
      "epoch": 2.0811850039325352,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004565514277886565,
      "loss": 5.8228,
      "mean_token_accuracy": 0.15331375747919082,
      "num_tokens": 48249663.0,
      "step": 23815
    },
    {
      "entropy": 6.175390100479126,
      "epoch": 2.0816219522852397,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045653301935606456,
      "loss": 5.8444,
      "mean_token_accuracy": 0.15147224813699722,
      "num_tokens": 48260319.0,
      "step": 23820
    },
    {
      "entropy": 6.179495620727539,
      "epoch": 2.0820589006379446,
      "grad_norm": 2.71875,
      "learning_rate": 0.00045651460744157424,
      "loss": 5.8421,
      "mean_token_accuracy": 0.15112654566764833,
      "num_tokens": 48270107.0,
      "step": 23825
    },
    {
      "entropy": 6.069668388366699,
      "epoch": 2.082495848990649,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004564961920455387,
      "loss": 5.8217,
      "mean_token_accuracy": 0.16202236860990524,
      "num_tokens": 48280529.0,
      "step": 23830
    },
    {
      "entropy": 6.160116147994995,
      "epoch": 2.082932797343354,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004564777731683109,
      "loss": 5.8623,
      "mean_token_accuracy": 0.15439659357070923,
      "num_tokens": 48290672.0,
      "step": 23835
    },
    {
      "entropy": 6.239656639099121,
      "epoch": 2.083369745696059,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004564593508102445,
      "loss": 5.8653,
      "mean_token_accuracy": 0.15614313781261444,
      "num_tokens": 48301327.0,
      "step": 23840
    },
    {
      "entropy": 6.130482196807861,
      "epoch": 2.0838066940487634,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004564409249716925,
      "loss": 5.8208,
      "mean_token_accuracy": 0.15363506078720093,
      "num_tokens": 48310861.0,
      "step": 23845
    },
    {
      "entropy": 6.082393836975098,
      "epoch": 2.0842436424014683,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00045642249565300866,
      "loss": 5.7944,
      "mean_token_accuracy": 0.15544576495885848,
      "num_tokens": 48320381.0,
      "step": 23850
    },
    {
      "entropy": 6.150322151184082,
      "epoch": 2.084680590754173,
      "grad_norm": 2.25,
      "learning_rate": 0.00045640406285454616,
      "loss": 5.7468,
      "mean_token_accuracy": 0.164085553586483,
      "num_tokens": 48330566.0,
      "step": 23855
    },
    {
      "entropy": 6.129547548294068,
      "epoch": 2.0851175391068777,
      "grad_norm": 2.0625,
      "learning_rate": 0.00045638562657665885,
      "loss": 5.8142,
      "mean_token_accuracy": 0.15932340323925018,
      "num_tokens": 48340773.0,
      "step": 23860
    },
    {
      "entropy": 6.135800743103028,
      "epoch": 2.085554487459582,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045636718681969996,
      "loss": 5.7869,
      "mean_token_accuracy": 0.157858707010746,
      "num_tokens": 48350972.0,
      "step": 23865
    },
    {
      "entropy": 6.162114524841309,
      "epoch": 2.085991435812287,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004563487435840235,
      "loss": 5.8043,
      "mean_token_accuracy": 0.1558579757809639,
      "num_tokens": 48360606.0,
      "step": 23870
    },
    {
      "entropy": 6.098880338668823,
      "epoch": 2.0864283841649915,
      "grad_norm": 2.0,
      "learning_rate": 0.00045633029686998304,
      "loss": 5.8636,
      "mean_token_accuracy": 0.16265986263751983,
      "num_tokens": 48371970.0,
      "step": 23875
    },
    {
      "entropy": 6.128022718429565,
      "epoch": 2.0868653325176965,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004563118466779326,
      "loss": 5.7539,
      "mean_token_accuracy": 0.16514307484030724,
      "num_tokens": 48382917.0,
      "step": 23880
    },
    {
      "entropy": 6.157595872879028,
      "epoch": 2.087302280870401,
      "grad_norm": 1.84375,
      "learning_rate": 0.00045629339300822593,
      "loss": 5.8443,
      "mean_token_accuracy": 0.16168017238378524,
      "num_tokens": 48393428.0,
      "step": 23885
    },
    {
      "entropy": 6.256121253967285,
      "epoch": 2.087739229223106,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004562749358612168,
      "loss": 5.9025,
      "mean_token_accuracy": 0.15070503950119019,
      "num_tokens": 48404616.0,
      "step": 23890
    },
    {
      "entropy": 6.128035259246826,
      "epoch": 2.0881761775758108,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004562564752372596,
      "loss": 5.7172,
      "mean_token_accuracy": 0.1703226923942566,
      "num_tokens": 48414753.0,
      "step": 23895
    },
    {
      "entropy": 6.146278238296508,
      "epoch": 2.0886131259285152,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004562380111367082,
      "loss": 5.8577,
      "mean_token_accuracy": 0.15729131251573564,
      "num_tokens": 48424759.0,
      "step": 23900
    },
    {
      "entropy": 6.161119270324707,
      "epoch": 2.08905007428122,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045621954355991675,
      "loss": 5.8185,
      "mean_token_accuracy": 0.16103170961141586,
      "num_tokens": 48435158.0,
      "step": 23905
    },
    {
      "entropy": 6.109135723114013,
      "epoch": 2.0894870226339246,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00045620107250723946,
      "loss": 5.7887,
      "mean_token_accuracy": 0.16193486750125885,
      "num_tokens": 48445234.0,
      "step": 23910
    },
    {
      "entropy": 6.1733620166778564,
      "epoch": 2.0899239709866295,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045618259797903073,
      "loss": 5.7806,
      "mean_token_accuracy": 0.16289247423410416,
      "num_tokens": 48456053.0,
      "step": 23915
    },
    {
      "entropy": 6.159033536911011,
      "epoch": 2.090360919339334,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004561641199756447,
      "loss": 5.7863,
      "mean_token_accuracy": 0.15874427258968354,
      "num_tokens": 48466293.0,
      "step": 23920
    },
    {
      "entropy": 6.065924263000488,
      "epoch": 2.090797867692039,
      "grad_norm": 3.84375,
      "learning_rate": 0.00045614563849743605,
      "loss": 5.765,
      "mean_token_accuracy": 0.1650603547692299,
      "num_tokens": 48476170.0,
      "step": 23925
    },
    {
      "entropy": 6.132977151870728,
      "epoch": 2.0912348160447434,
      "grad_norm": 2.921875,
      "learning_rate": 0.00045612715354475907,
      "loss": 5.7734,
      "mean_token_accuracy": 0.1547189772129059,
      "num_tokens": 48485728.0,
      "step": 23930
    },
    {
      "entropy": 6.114084720611572,
      "epoch": 2.0916717643974483,
      "grad_norm": 2.296875,
      "learning_rate": 0.00045610866511796845,
      "loss": 5.8045,
      "mean_token_accuracy": 0.16163035333156586,
      "num_tokens": 48496531.0,
      "step": 23935
    },
    {
      "entropy": 6.096020936965942,
      "epoch": 2.0921087127501528,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045609017321741863,
      "loss": 5.7521,
      "mean_token_accuracy": 0.16086275577545167,
      "num_tokens": 48506006.0,
      "step": 23940
    },
    {
      "entropy": 6.154129362106323,
      "epoch": 2.0925456611028577,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004560716778434644,
      "loss": 5.8194,
      "mean_token_accuracy": 0.1599178433418274,
      "num_tokens": 48515902.0,
      "step": 23945
    },
    {
      "entropy": 6.193798160552978,
      "epoch": 2.0929826094555626,
      "grad_norm": 2.9375,
      "learning_rate": 0.00045605317899646066,
      "loss": 5.8438,
      "mean_token_accuracy": 0.15808311551809312,
      "num_tokens": 48526263.0,
      "step": 23950
    },
    {
      "entropy": 6.1690774917602536,
      "epoch": 2.093419557808267,
      "grad_norm": 1.640625,
      "learning_rate": 0.00045603467667676195,
      "loss": 5.8018,
      "mean_token_accuracy": 0.16499936133623122,
      "num_tokens": 48537415.0,
      "step": 23955
    },
    {
      "entropy": 6.160503101348877,
      "epoch": 2.093856506160972,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004560161708847233,
      "loss": 5.8729,
      "mean_token_accuracy": 0.15711526870727538,
      "num_tokens": 48548911.0,
      "step": 23960
    },
    {
      "entropy": 6.150792741775513,
      "epoch": 2.0942934545136764,
      "grad_norm": 3.0,
      "learning_rate": 0.0004559976616206996,
      "loss": 5.82,
      "mean_token_accuracy": 0.1580877870321274,
      "num_tokens": 48560266.0,
      "step": 23965
    },
    {
      "entropy": 6.2053759574890135,
      "epoch": 2.0947304028663813,
      "grad_norm": 2.546875,
      "learning_rate": 0.00045597914888504597,
      "loss": 5.8902,
      "mean_token_accuracy": 0.15257944613695146,
      "num_tokens": 48570696.0,
      "step": 23970
    },
    {
      "entropy": 6.1706099033355715,
      "epoch": 2.095167351219086,
      "grad_norm": 2.875,
      "learning_rate": 0.00045596063267811746,
      "loss": 5.7786,
      "mean_token_accuracy": 0.16238964200019837,
      "num_tokens": 48580725.0,
      "step": 23975
    },
    {
      "entropy": 6.108142232894897,
      "epoch": 2.0956042995717907,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004559421130002691,
      "loss": 5.8443,
      "mean_token_accuracy": 0.163856203854084,
      "num_tokens": 48591968.0,
      "step": 23980
    },
    {
      "entropy": 6.165525674819946,
      "epoch": 2.096041247924495,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004559235898518563,
      "loss": 5.7725,
      "mean_token_accuracy": 0.17220778614282609,
      "num_tokens": 48602074.0,
      "step": 23985
    },
    {
      "entropy": 6.163488245010376,
      "epoch": 2.0964781962772,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004559050632332342,
      "loss": 5.8057,
      "mean_token_accuracy": 0.16481990069150926,
      "num_tokens": 48613323.0,
      "step": 23990
    },
    {
      "entropy": 6.156353569030761,
      "epoch": 2.0969151446299046,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004558865331447581,
      "loss": 5.827,
      "mean_token_accuracy": 0.15559408366680144,
      "num_tokens": 48623386.0,
      "step": 23995
    },
    {
      "entropy": 6.049243927001953,
      "epoch": 2.0973520929826095,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004558679995867837,
      "loss": 5.7497,
      "mean_token_accuracy": 0.16954871118068696,
      "num_tokens": 48633379.0,
      "step": 24000
    },
    {
      "epoch": 2.0973520929826095,
      "eval_entropy": 5.909023268732418,
      "eval_loss": 5.862783432006836,
      "eval_mean_token_accuracy": 0.16594386259780244,
      "eval_num_tokens": 48633379.0,
      "eval_runtime": 27.7684,
      "eval_samples_per_second": 1332.952,
      "eval_steps_per_second": 166.628,
      "step": 24000
    },
    {
      "entropy": 6.133021259307862,
      "epoch": 2.097789041335314,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004558494625596662,
      "loss": 5.7712,
      "mean_token_accuracy": 0.1646033078432083,
      "num_tokens": 48643918.0,
      "step": 24005
    },
    {
      "entropy": 6.16729211807251,
      "epoch": 2.098225989688019,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004558309220637612,
      "loss": 5.8875,
      "mean_token_accuracy": 0.15530220419168472,
      "num_tokens": 48654534.0,
      "step": 24010
    },
    {
      "entropy": 6.151780271530152,
      "epoch": 2.098662938040724,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004558123780994244,
      "loss": 5.7815,
      "mean_token_accuracy": 0.16142997443675994,
      "num_tokens": 48664928.0,
      "step": 24015
    },
    {
      "entropy": 6.100732517242432,
      "epoch": 2.0990998863934283,
      "grad_norm": 1.875,
      "learning_rate": 0.0004557938306670115,
      "loss": 5.7675,
      "mean_token_accuracy": 0.16654305383563042,
      "num_tokens": 48675441.0,
      "step": 24020
    },
    {
      "entropy": 6.094825792312622,
      "epoch": 2.099536834746133,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045577527976687813,
      "loss": 5.7094,
      "mean_token_accuracy": 0.16648484915494918,
      "num_tokens": 48685173.0,
      "step": 24025
    },
    {
      "entropy": 6.042048025131225,
      "epoch": 2.0999737830988376,
      "grad_norm": 2.625,
      "learning_rate": 0.00045575672539938014,
      "loss": 5.7723,
      "mean_token_accuracy": 0.16567939668893814,
      "num_tokens": 48695539.0,
      "step": 24030
    },
    {
      "entropy": 6.093257761001587,
      "epoch": 2.1004107314515426,
      "grad_norm": 2.125,
      "learning_rate": 0.00045573816756487345,
      "loss": 5.8176,
      "mean_token_accuracy": 0.16516663432121276,
      "num_tokens": 48705977.0,
      "step": 24035
    },
    {
      "entropy": 6.131457328796387,
      "epoch": 2.100847679804247,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004557196062637141,
      "loss": 5.7324,
      "mean_token_accuracy": 0.16550699025392532,
      "num_tokens": 48715924.0,
      "step": 24040
    },
    {
      "entropy": 6.17737603187561,
      "epoch": 2.101284628156952,
      "grad_norm": 2.875,
      "learning_rate": 0.0004557010414962579,
      "loss": 5.8418,
      "mean_token_accuracy": 0.15230527222156526,
      "num_tokens": 48727071.0,
      "step": 24045
    },
    {
      "entropy": 6.158092212677002,
      "epoch": 2.1017215765096564,
      "grad_norm": 4.4375,
      "learning_rate": 0.00045568247326286103,
      "loss": 5.8141,
      "mean_token_accuracy": 0.15453498363494872,
      "num_tokens": 48736959.0,
      "step": 24050
    },
    {
      "entropy": 6.112994289398193,
      "epoch": 2.1021585248623613,
      "grad_norm": 2.5625,
      "learning_rate": 0.00045566390156387974,
      "loss": 5.8169,
      "mean_token_accuracy": 0.15812865793704986,
      "num_tokens": 48748419.0,
      "step": 24055
    },
    {
      "entropy": 6.136483860015869,
      "epoch": 2.102595473215066,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004556453263996701,
      "loss": 5.773,
      "mean_token_accuracy": 0.16617968678474426,
      "num_tokens": 48758480.0,
      "step": 24060
    },
    {
      "entropy": 6.198967218399048,
      "epoch": 2.1030324215677707,
      "grad_norm": 2.390625,
      "learning_rate": 0.00045562674777058844,
      "loss": 5.8875,
      "mean_token_accuracy": 0.1560373678803444,
      "num_tokens": 48768182.0,
      "step": 24065
    },
    {
      "entropy": 6.177241420745849,
      "epoch": 2.103469369920475,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004556081656769911,
      "loss": 5.7828,
      "mean_token_accuracy": 0.15777845829725265,
      "num_tokens": 48777729.0,
      "step": 24070
    },
    {
      "entropy": 6.096408176422119,
      "epoch": 2.10390631827318,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045558958011923456,
      "loss": 5.7579,
      "mean_token_accuracy": 0.16987152993679047,
      "num_tokens": 48787109.0,
      "step": 24075
    },
    {
      "entropy": 6.144212293624878,
      "epoch": 2.104343266625885,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004555709910976752,
      "loss": 5.8079,
      "mean_token_accuracy": 0.15732285678386687,
      "num_tokens": 48797152.0,
      "step": 24080
    },
    {
      "entropy": 6.1199622631073,
      "epoch": 2.1047802149785895,
      "grad_norm": 3.09375,
      "learning_rate": 0.00045555239861266964,
      "loss": 5.7771,
      "mean_token_accuracy": 0.16345112919807434,
      "num_tokens": 48806288.0,
      "step": 24085
    },
    {
      "entropy": 6.016192865371704,
      "epoch": 2.1052171633312944,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004555338026645744,
      "loss": 5.7213,
      "mean_token_accuracy": 0.17334125339984893,
      "num_tokens": 48816025.0,
      "step": 24090
    },
    {
      "entropy": 6.151209783554077,
      "epoch": 2.105654111683999,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004555152032537463,
      "loss": 5.8319,
      "mean_token_accuracy": 0.15525479465723038,
      "num_tokens": 48827150.0,
      "step": 24095
    },
    {
      "entropy": 6.190832233428955,
      "epoch": 2.1060910600367038,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045549660038054205,
      "loss": 5.7559,
      "mean_token_accuracy": 0.16156779527664183,
      "num_tokens": 48837190.0,
      "step": 24100
    },
    {
      "entropy": 6.2160295963287355,
      "epoch": 2.1065280083894082,
      "grad_norm": 2.546875,
      "learning_rate": 0.00045547799404531826,
      "loss": 5.8239,
      "mean_token_accuracy": 0.15915861278772353,
      "num_tokens": 48847573.0,
      "step": 24105
    },
    {
      "entropy": 6.060466003417969,
      "epoch": 2.106964956742113,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045545938424843214,
      "loss": 5.6875,
      "mean_token_accuracy": 0.1633274659514427,
      "num_tokens": 48858966.0,
      "step": 24110
    },
    {
      "entropy": 6.110295677185059,
      "epoch": 2.1074019050948176,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004554407709902404,
      "loss": 5.8217,
      "mean_token_accuracy": 0.15711138844490052,
      "num_tokens": 48869027.0,
      "step": 24115
    },
    {
      "entropy": 6.10790638923645,
      "epoch": 2.1078388534475225,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004554221542711001,
      "loss": 5.8217,
      "mean_token_accuracy": 0.1662372827529907,
      "num_tokens": 48879325.0,
      "step": 24120
    },
    {
      "entropy": 6.058655834197998,
      "epoch": 2.108275801800227,
      "grad_norm": 2.0625,
      "learning_rate": 0.00045540353409136836,
      "loss": 5.7106,
      "mean_token_accuracy": 0.17198172062635422,
      "num_tokens": 48888939.0,
      "step": 24125
    },
    {
      "entropy": 6.085688066482544,
      "epoch": 2.108712750152932,
      "grad_norm": 2.265625,
      "learning_rate": 0.00045538491045140234,
      "loss": 5.7839,
      "mean_token_accuracy": 0.15988143533468246,
      "num_tokens": 48898997.0,
      "step": 24130
    },
    {
      "entropy": 6.1399623394012455,
      "epoch": 2.109149698505637,
      "grad_norm": 2.328125,
      "learning_rate": 0.00045536628335155914,
      "loss": 5.7711,
      "mean_token_accuracy": 0.16360228955745698,
      "num_tokens": 48909063.0,
      "step": 24135
    },
    {
      "entropy": 6.062327718734741,
      "epoch": 2.1095866468583413,
      "grad_norm": 2.1875,
      "learning_rate": 0.00045534765279219626,
      "loss": 5.6654,
      "mean_token_accuracy": 0.17838005423545839,
      "num_tokens": 48918694.0,
      "step": 24140
    },
    {
      "entropy": 6.138954830169678,
      "epoch": 2.110023595211046,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004553290187736708,
      "loss": 5.8415,
      "mean_token_accuracy": 0.1598917007446289,
      "num_tokens": 48928223.0,
      "step": 24145
    },
    {
      "entropy": 6.1307906150817875,
      "epoch": 2.1104605435637507,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004553103812963402,
      "loss": 5.7954,
      "mean_token_accuracy": 0.16307459622621537,
      "num_tokens": 48938193.0,
      "step": 24150
    },
    {
      "entropy": 6.177183818817139,
      "epoch": 2.1108974919164556,
      "grad_norm": 2.359375,
      "learning_rate": 0.00045529174036056217,
      "loss": 5.8844,
      "mean_token_accuracy": 0.1480637200176716,
      "num_tokens": 48949693.0,
      "step": 24155
    },
    {
      "entropy": 6.106791687011719,
      "epoch": 2.11133444026916,
      "grad_norm": 3.265625,
      "learning_rate": 0.00045527309596669397,
      "loss": 5.7683,
      "mean_token_accuracy": 0.16647205501794815,
      "num_tokens": 48960199.0,
      "step": 24160
    },
    {
      "entropy": 6.13969349861145,
      "epoch": 2.111771388621865,
      "grad_norm": 2.453125,
      "learning_rate": 0.00045525444811509336,
      "loss": 5.8018,
      "mean_token_accuracy": 0.16208005845546722,
      "num_tokens": 48969380.0,
      "step": 24165
    },
    {
      "entropy": 6.098386621475219,
      "epoch": 2.1122083369745694,
      "grad_norm": 2.921875,
      "learning_rate": 0.000455235796806118,
      "loss": 5.718,
      "mean_token_accuracy": 0.16776890754699708,
      "num_tokens": 48978963.0,
      "step": 24170
    },
    {
      "entropy": 6.11563286781311,
      "epoch": 2.1126452853272744,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004552171420401256,
      "loss": 5.8523,
      "mean_token_accuracy": 0.16490355879068375,
      "num_tokens": 48989568.0,
      "step": 24175
    },
    {
      "entropy": 6.150411081314087,
      "epoch": 2.113082233679979,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045519848381747396,
      "loss": 5.8294,
      "mean_token_accuracy": 0.15826844274997712,
      "num_tokens": 49000684.0,
      "step": 24180
    },
    {
      "entropy": 6.120581197738647,
      "epoch": 2.1135191820326837,
      "grad_norm": 4.46875,
      "learning_rate": 0.000455179822138521,
      "loss": 5.8694,
      "mean_token_accuracy": 0.14880899339914322,
      "num_tokens": 49010194.0,
      "step": 24185
    },
    {
      "entropy": 6.068571662902832,
      "epoch": 2.1139561303853887,
      "grad_norm": 1.90625,
      "learning_rate": 0.00045516115700362466,
      "loss": 5.7577,
      "mean_token_accuracy": 0.16768087595701217,
      "num_tokens": 49020180.0,
      "step": 24190
    },
    {
      "entropy": 6.132275581359863,
      "epoch": 2.114393078738093,
      "grad_norm": 2.0625,
      "learning_rate": 0.00045514248841314295,
      "loss": 5.8172,
      "mean_token_accuracy": 0.16499296426773072,
      "num_tokens": 49030665.0,
      "step": 24195
    },
    {
      "entropy": 6.216008806228638,
      "epoch": 2.114830027090798,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004551238163674338,
      "loss": 5.8087,
      "mean_token_accuracy": 0.16382874250411988,
      "num_tokens": 49040197.0,
      "step": 24200
    },
    {
      "entropy": 6.127198219299316,
      "epoch": 2.1152669754435025,
      "grad_norm": 2.53125,
      "learning_rate": 0.00045510514086685553,
      "loss": 5.7714,
      "mean_token_accuracy": 0.16007289439439773,
      "num_tokens": 49049996.0,
      "step": 24205
    },
    {
      "entropy": 6.17523684501648,
      "epoch": 2.1157039237962074,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004550864619117664,
      "loss": 5.88,
      "mean_token_accuracy": 0.1569709934294224,
      "num_tokens": 49061691.0,
      "step": 24210
    },
    {
      "entropy": 6.23469386100769,
      "epoch": 2.116140872148912,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004550677795025244,
      "loss": 5.8976,
      "mean_token_accuracy": 0.15461632013320922,
      "num_tokens": 49071605.0,
      "step": 24215
    },
    {
      "entropy": 6.125311946868896,
      "epoch": 2.116577820501617,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004550490936394881,
      "loss": 5.7361,
      "mean_token_accuracy": 0.17150014638900757,
      "num_tokens": 49081317.0,
      "step": 24220
    },
    {
      "entropy": 6.064651966094971,
      "epoch": 2.1170147688543213,
      "grad_norm": 2.46875,
      "learning_rate": 0.00045503040432301586,
      "loss": 5.7436,
      "mean_token_accuracy": 0.1609647586941719,
      "num_tokens": 49090662.0,
      "step": 24225
    },
    {
      "entropy": 6.2205853939056395,
      "epoch": 2.117451717207026,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004550117115534661,
      "loss": 5.9285,
      "mean_token_accuracy": 0.14656810611486434,
      "num_tokens": 49102046.0,
      "step": 24230
    },
    {
      "entropy": 6.221758842468262,
      "epoch": 2.1178886655597307,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045499301533119736,
      "loss": 5.7442,
      "mean_token_accuracy": 0.16828668415546416,
      "num_tokens": 49111047.0,
      "step": 24235
    },
    {
      "entropy": 6.08589448928833,
      "epoch": 2.1183256139124356,
      "grad_norm": 2.375,
      "learning_rate": 0.0004549743156565684,
      "loss": 5.7628,
      "mean_token_accuracy": 0.16037226021289824,
      "num_tokens": 49120606.0,
      "step": 24240
    },
    {
      "entropy": 6.130853891372681,
      "epoch": 2.1187625622651405,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004549556125299377,
      "loss": 5.8276,
      "mean_token_accuracy": 0.16027093231678008,
      "num_tokens": 49130981.0,
      "step": 24245
    },
    {
      "entropy": 6.138224172592163,
      "epoch": 2.119199510617845,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004549369059516639,
      "loss": 5.8501,
      "mean_token_accuracy": 0.15824138075113298,
      "num_tokens": 49140843.0,
      "step": 24250
    },
    {
      "entropy": 6.201585912704468,
      "epoch": 2.11963645897055,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004549181959221061,
      "loss": 5.8373,
      "mean_token_accuracy": 0.1519739508628845,
      "num_tokens": 49150492.0,
      "step": 24255
    },
    {
      "entropy": 6.105088901519776,
      "epoch": 2.1200734073232543,
      "grad_norm": 1.796875,
      "learning_rate": 0.00045489948244162304,
      "loss": 5.7922,
      "mean_token_accuracy": 0.1626556321978569,
      "num_tokens": 49159973.0,
      "step": 24260
    },
    {
      "entropy": 6.197771835327148,
      "epoch": 2.1205103556759592,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004548807655105736,
      "loss": 5.8791,
      "mean_token_accuracy": 0.15381792336702346,
      "num_tokens": 49170711.0,
      "step": 24265
    },
    {
      "entropy": 6.2146648406982425,
      "epoch": 2.1209473040286637,
      "grad_norm": 4.0,
      "learning_rate": 0.00045486204512931684,
      "loss": 5.8388,
      "mean_token_accuracy": 0.15497995167970657,
      "num_tokens": 49181121.0,
      "step": 24270
    },
    {
      "entropy": 6.008566951751709,
      "epoch": 2.1213842523813686,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004548433212982118,
      "loss": 5.6742,
      "mean_token_accuracy": 0.17899947315454484,
      "num_tokens": 49191528.0,
      "step": 24275
    },
    {
      "entropy": 6.124616861343384,
      "epoch": 2.121821200734073,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004548245940176176,
      "loss": 5.8307,
      "mean_token_accuracy": 0.15468672439455985,
      "num_tokens": 49202244.0,
      "step": 24280
    },
    {
      "entropy": 6.119020318984985,
      "epoch": 2.122258149086778,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004548058632878935,
      "loss": 5.8114,
      "mean_token_accuracy": 0.16280713230371474,
      "num_tokens": 49212964.0,
      "step": 24285
    },
    {
      "entropy": 6.191755294799805,
      "epoch": 2.1226950974394825,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045478712910939874,
      "loss": 5.8273,
      "mean_token_accuracy": 0.157781021296978,
      "num_tokens": 49223478.0,
      "step": 24290
    },
    {
      "entropy": 6.171230697631836,
      "epoch": 2.1231320457921874,
      "grad_norm": 2.125,
      "learning_rate": 0.00045476839148249256,
      "loss": 5.8853,
      "mean_token_accuracy": 0.15325332432985306,
      "num_tokens": 49233933.0,
      "step": 24295
    },
    {
      "entropy": 6.117204809188843,
      "epoch": 2.123568994144892,
      "grad_norm": 2.1875,
      "learning_rate": 0.00045474965040753453,
      "loss": 5.7397,
      "mean_token_accuracy": 0.16596123874187468,
      "num_tokens": 49243147.0,
      "step": 24300
    },
    {
      "entropy": 6.1066724300384525,
      "epoch": 2.1240059424975968,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004547309058848839,
      "loss": 5.7266,
      "mean_token_accuracy": 0.16735248565673827,
      "num_tokens": 49252524.0,
      "step": 24305
    },
    {
      "entropy": 6.1274168491363525,
      "epoch": 2.1244428908503017,
      "grad_norm": 2.328125,
      "learning_rate": 0.00045471215791490047,
      "loss": 5.799,
      "mean_token_accuracy": 0.1618364378809929,
      "num_tokens": 49261845.0,
      "step": 24310
    },
    {
      "entropy": 6.095428848266602,
      "epoch": 2.124879839203006,
      "grad_norm": 2.25,
      "learning_rate": 0.00045469340649794353,
      "loss": 5.7637,
      "mean_token_accuracy": 0.16472189426422118,
      "num_tokens": 49272437.0,
      "step": 24315
    },
    {
      "entropy": 6.110375499725341,
      "epoch": 2.125316787555711,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000454674651634373,
      "loss": 5.7368,
      "mean_token_accuracy": 0.17218557596206666,
      "num_tokens": 49281687.0,
      "step": 24320
    },
    {
      "entropy": 6.108816862106323,
      "epoch": 2.1257537359084155,
      "grad_norm": 2.4375,
      "learning_rate": 0.00045465589332454844,
      "loss": 5.7014,
      "mean_token_accuracy": 0.16242825984954834,
      "num_tokens": 49291524.0,
      "step": 24325
    },
    {
      "entropy": 6.1500317573547365,
      "epoch": 2.1261906842611205,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00045463713156882976,
      "loss": 5.8331,
      "mean_token_accuracy": 0.15694678127765654,
      "num_tokens": 49303124.0,
      "step": 24330
    },
    {
      "entropy": 6.176592874526977,
      "epoch": 2.126627632613825,
      "grad_norm": 3.984375,
      "learning_rate": 0.00045461836636757674,
      "loss": 5.7305,
      "mean_token_accuracy": 0.16584524810314177,
      "num_tokens": 49313050.0,
      "step": 24335
    },
    {
      "entropy": 6.089331912994385,
      "epoch": 2.12706458096653,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045459959772114935,
      "loss": 5.7207,
      "mean_token_accuracy": 0.1703559935092926,
      "num_tokens": 49321652.0,
      "step": 24340
    },
    {
      "entropy": 6.061080980300903,
      "epoch": 2.1275015293192343,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045458082562990754,
      "loss": 5.7416,
      "mean_token_accuracy": 0.17160896956920624,
      "num_tokens": 49331713.0,
      "step": 24345
    },
    {
      "entropy": 6.088311433792114,
      "epoch": 2.127938477671939,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004545620500942114,
      "loss": 5.7721,
      "mean_token_accuracy": 0.1659550741314888,
      "num_tokens": 49340370.0,
      "step": 24350
    },
    {
      "entropy": 6.1837334632873535,
      "epoch": 2.1283754260246437,
      "grad_norm": 2.0,
      "learning_rate": 0.0004545432711144211,
      "loss": 5.8019,
      "mean_token_accuracy": 0.16196324154734612,
      "num_tokens": 49349894.0,
      "step": 24355
    },
    {
      "entropy": 6.117650079727173,
      "epoch": 2.1288123743773486,
      "grad_norm": 2.875,
      "learning_rate": 0.0004545244886908967,
      "loss": 5.6917,
      "mean_token_accuracy": 0.1663397341966629,
      "num_tokens": 49359738.0,
      "step": 24360
    },
    {
      "entropy": 6.09303846359253,
      "epoch": 2.129249322730053,
      "grad_norm": 1.703125,
      "learning_rate": 0.00045450570282399856,
      "loss": 5.7911,
      "mean_token_accuracy": 0.1603309005498886,
      "num_tokens": 49370495.0,
      "step": 24365
    },
    {
      "entropy": 6.167524719238282,
      "epoch": 2.129686271082758,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000454486913514087,
      "loss": 5.8221,
      "mean_token_accuracy": 0.16587257534265518,
      "num_tokens": 49382191.0,
      "step": 24370
    },
    {
      "entropy": 6.151463365554809,
      "epoch": 2.130123219435463,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004544681207615224,
      "loss": 5.8333,
      "mean_token_accuracy": 0.1587792828679085,
      "num_tokens": 49391724.0,
      "step": 24375
    },
    {
      "entropy": 6.153978300094605,
      "epoch": 2.1305601677881674,
      "grad_norm": 1.75,
      "learning_rate": 0.0004544493245666652,
      "loss": 5.7765,
      "mean_token_accuracy": 0.1679410457611084,
      "num_tokens": 49401117.0,
      "step": 24380
    },
    {
      "entropy": 6.088162469863891,
      "epoch": 2.1309971161408723,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004544305249298759,
      "loss": 5.7421,
      "mean_token_accuracy": 0.16621051877737045,
      "num_tokens": 49410731.0,
      "step": 24385
    },
    {
      "entropy": 6.135039901733398,
      "epoch": 2.1314340644935768,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004544117218515152,
      "loss": 5.7999,
      "mean_token_accuracy": 0.16650499105453492,
      "num_tokens": 49420373.0,
      "step": 24390
    },
    {
      "entropy": 6.10662693977356,
      "epoch": 2.1318710128462817,
      "grad_norm": 2.578125,
      "learning_rate": 0.00045439291533194356,
      "loss": 5.746,
      "mean_token_accuracy": 0.16702510714530944,
      "num_tokens": 49430021.0,
      "step": 24395
    },
    {
      "entropy": 6.081175327301025,
      "epoch": 2.132307961198986,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004543741053715218,
      "loss": 5.7229,
      "mean_token_accuracy": 0.16428142935037612,
      "num_tokens": 49439687.0,
      "step": 24400
    },
    {
      "entropy": 6.1511331558227536,
      "epoch": 2.132744909551691,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004543552919706107,
      "loss": 5.9051,
      "mean_token_accuracy": 0.15116332620382308,
      "num_tokens": 49450320.0,
      "step": 24405
    },
    {
      "entropy": 6.153339290618897,
      "epoch": 2.1331818579043955,
      "grad_norm": 2.25,
      "learning_rate": 0.0004543364751295712,
      "loss": 5.7251,
      "mean_token_accuracy": 0.1616206794977188,
      "num_tokens": 49461272.0,
      "step": 24410
    },
    {
      "entropy": 6.186064434051514,
      "epoch": 2.1336188062571004,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045431765484876406,
      "loss": 5.8455,
      "mean_token_accuracy": 0.15796141922473908,
      "num_tokens": 49473087.0,
      "step": 24415
    },
    {
      "entropy": 6.076219606399536,
      "epoch": 2.134055754609805,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004542988311285505,
      "loss": 5.7633,
      "mean_token_accuracy": 0.16527504324913025,
      "num_tokens": 49483174.0,
      "step": 24420
    },
    {
      "entropy": 6.147305631637574,
      "epoch": 2.13449270296251,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045428000396929125,
      "loss": 5.7588,
      "mean_token_accuracy": 0.16245170533657075,
      "num_tokens": 49492267.0,
      "step": 24425
    },
    {
      "entropy": 6.076569271087647,
      "epoch": 2.1349296513152147,
      "grad_norm": 2.9375,
      "learning_rate": 0.00045426117337134757,
      "loss": 5.7337,
      "mean_token_accuracy": 0.16121915727853775,
      "num_tokens": 49502338.0,
      "step": 24430
    },
    {
      "entropy": 6.12009015083313,
      "epoch": 2.135366599667919,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045424233933508075,
      "loss": 5.8245,
      "mean_token_accuracy": 0.16175890266895293,
      "num_tokens": 49512991.0,
      "step": 24435
    },
    {
      "entropy": 6.0873785495758055,
      "epoch": 2.135803548020624,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004542235018608519,
      "loss": 5.6729,
      "mean_token_accuracy": 0.1725426807999611,
      "num_tokens": 49522224.0,
      "step": 24440
    },
    {
      "entropy": 6.097157144546509,
      "epoch": 2.1362404963733286,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045420466094902234,
      "loss": 5.8006,
      "mean_token_accuracy": 0.16104185432195664,
      "num_tokens": 49531982.0,
      "step": 24445
    },
    {
      "entropy": 6.168701934814453,
      "epoch": 2.1366774447260335,
      "grad_norm": 3.0,
      "learning_rate": 0.0004541858165999535,
      "loss": 5.7291,
      "mean_token_accuracy": 0.1671820431947708,
      "num_tokens": 49541487.0,
      "step": 24450
    },
    {
      "entropy": 6.163399076461792,
      "epoch": 2.137114393078738,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004541669688140068,
      "loss": 5.8584,
      "mean_token_accuracy": 0.1509682722389698,
      "num_tokens": 49552052.0,
      "step": 24455
    },
    {
      "entropy": 6.109228801727295,
      "epoch": 2.137551341431443,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004541481175915438,
      "loss": 5.8013,
      "mean_token_accuracy": 0.15786446407437324,
      "num_tokens": 49561807.0,
      "step": 24460
    },
    {
      "entropy": 6.201771974563599,
      "epoch": 2.1379882897841473,
      "grad_norm": 2.09375,
      "learning_rate": 0.000454129262932926,
      "loss": 5.8143,
      "mean_token_accuracy": 0.1624198481440544,
      "num_tokens": 49570687.0,
      "step": 24465
    },
    {
      "entropy": 6.182664632797241,
      "epoch": 2.1384252381368523,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045411040483851506,
      "loss": 5.8062,
      "mean_token_accuracy": 0.1638040140271187,
      "num_tokens": 49581000.0,
      "step": 24470
    },
    {
      "entropy": 6.07001838684082,
      "epoch": 2.1388621864895567,
      "grad_norm": 2.90625,
      "learning_rate": 0.00045409154330867274,
      "loss": 5.8393,
      "mean_token_accuracy": 0.16619377136230468,
      "num_tokens": 49590442.0,
      "step": 24475
    },
    {
      "entropy": 6.157864999771118,
      "epoch": 2.1392991348422616,
      "grad_norm": 3.03125,
      "learning_rate": 0.00045407267834376075,
      "loss": 5.7868,
      "mean_token_accuracy": 0.16424607038497924,
      "num_tokens": 49600068.0,
      "step": 24480
    },
    {
      "entropy": 6.201351261138916,
      "epoch": 2.1397360831949666,
      "grad_norm": 2.6875,
      "learning_rate": 0.000454053809944141,
      "loss": 5.8381,
      "mean_token_accuracy": 0.15322006940841676,
      "num_tokens": 49610696.0,
      "step": 24485
    },
    {
      "entropy": 6.126097774505615,
      "epoch": 2.140173031547671,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004540349381101753,
      "loss": 5.8044,
      "mean_token_accuracy": 0.15984151065349578,
      "num_tokens": 49619489.0,
      "step": 24490
    },
    {
      "entropy": 6.17208948135376,
      "epoch": 2.140609979900376,
      "grad_norm": 1.9375,
      "learning_rate": 0.00045401606284222563,
      "loss": 5.804,
      "mean_token_accuracy": 0.15770916491746903,
      "num_tokens": 49629572.0,
      "step": 24495
    },
    {
      "entropy": 6.171817588806152,
      "epoch": 2.1410469282530804,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00045399718414065415,
      "loss": 5.7493,
      "mean_token_accuracy": 0.1620365709066391,
      "num_tokens": 49640035.0,
      "step": 24500
    },
    {
      "entropy": 6.092524480819702,
      "epoch": 2.1414838766057853,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004539783020058229,
      "loss": 5.729,
      "mean_token_accuracy": 0.17079070806503296,
      "num_tokens": 49649671.0,
      "step": 24505
    },
    {
      "entropy": 6.107431793212891,
      "epoch": 2.14192082495849,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000453959416438094,
      "loss": 5.8059,
      "mean_token_accuracy": 0.1643483191728592,
      "num_tokens": 49659122.0,
      "step": 24510
    },
    {
      "entropy": 6.193722820281982,
      "epoch": 2.1423577733111947,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004539405274378297,
      "loss": 5.7437,
      "mean_token_accuracy": 0.16266967207193375,
      "num_tokens": 49669938.0,
      "step": 24515
    },
    {
      "entropy": 6.172129821777344,
      "epoch": 2.142794721663899,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004539216350053924,
      "loss": 5.7646,
      "mean_token_accuracy": 0.15950179994106292,
      "num_tokens": 49680783.0,
      "step": 24520
    },
    {
      "entropy": 6.10248703956604,
      "epoch": 2.143231670016604,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004539027391411443,
      "loss": 5.8236,
      "mean_token_accuracy": 0.15986089557409286,
      "num_tokens": 49691271.0,
      "step": 24525
    },
    {
      "entropy": 6.127971029281616,
      "epoch": 2.1436686183693086,
      "grad_norm": 2.484375,
      "learning_rate": 0.00045388383984544794,
      "loss": 5.8023,
      "mean_token_accuracy": 0.15900971591472626,
      "num_tokens": 49702496.0,
      "step": 24530
    },
    {
      "entropy": 6.16137318611145,
      "epoch": 2.1441055667220135,
      "grad_norm": 2.546875,
      "learning_rate": 0.00045386493711866577,
      "loss": 5.8373,
      "mean_token_accuracy": 0.16026654690504075,
      "num_tokens": 49712145.0,
      "step": 24535
    },
    {
      "entropy": 6.151933097839356,
      "epoch": 2.1445425150747184,
      "grad_norm": 2.21875,
      "learning_rate": 0.00045384603096116044,
      "loss": 5.7631,
      "mean_token_accuracy": 0.16526898741722107,
      "num_tokens": 49722243.0,
      "step": 24540
    },
    {
      "entropy": 6.164535427093506,
      "epoch": 2.144979463427423,
      "grad_norm": 2.0,
      "learning_rate": 0.0004538271213732945,
      "loss": 5.8234,
      "mean_token_accuracy": 0.16345806270837784,
      "num_tokens": 49731804.0,
      "step": 24545
    },
    {
      "entropy": 6.13788914680481,
      "epoch": 2.1454164117801278,
      "grad_norm": 3.03125,
      "learning_rate": 0.00045380820835543066,
      "loss": 5.7348,
      "mean_token_accuracy": 0.16279738694429396,
      "num_tokens": 49741275.0,
      "step": 24550
    },
    {
      "entropy": 6.049767589569091,
      "epoch": 2.1458533601328322,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004537892919079317,
      "loss": 5.6938,
      "mean_token_accuracy": 0.1646263763308525,
      "num_tokens": 49751140.0,
      "step": 24555
    },
    {
      "entropy": 6.115736865997315,
      "epoch": 2.146290308485537,
      "grad_norm": 2.296875,
      "learning_rate": 0.00045377037203116053,
      "loss": 5.816,
      "mean_token_accuracy": 0.1603135958313942,
      "num_tokens": 49760747.0,
      "step": 24560
    },
    {
      "entropy": 6.176042652130127,
      "epoch": 2.1467272568382416,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045375144872547975,
      "loss": 5.8498,
      "mean_token_accuracy": 0.1530046656727791,
      "num_tokens": 49771491.0,
      "step": 24565
    },
    {
      "entropy": 6.171210050582886,
      "epoch": 2.1471642051909465,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004537325219912527,
      "loss": 5.7385,
      "mean_token_accuracy": 0.16872286498546601,
      "num_tokens": 49781523.0,
      "step": 24570
    },
    {
      "entropy": 6.205987501144409,
      "epoch": 2.147601153543651,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004537135918288421,
      "loss": 5.8665,
      "mean_token_accuracy": 0.15675292015075684,
      "num_tokens": 49791560.0,
      "step": 24575
    },
    {
      "entropy": 6.073527145385742,
      "epoch": 2.148038101896356,
      "grad_norm": 2.125,
      "learning_rate": 0.0004536946582386112,
      "loss": 5.714,
      "mean_token_accuracy": 0.15974147617816925,
      "num_tokens": 49800588.0,
      "step": 24580
    },
    {
      "entropy": 6.081054258346557,
      "epoch": 2.1484750502490604,
      "grad_norm": 3.28125,
      "learning_rate": 0.00045367572122092316,
      "loss": 5.7887,
      "mean_token_accuracy": 0.16638316810131074,
      "num_tokens": 49810381.0,
      "step": 24585
    },
    {
      "entropy": 6.144349050521851,
      "epoch": 2.1489119986017653,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004536567807761411,
      "loss": 5.8216,
      "mean_token_accuracy": 0.1622787356376648,
      "num_tokens": 49820931.0,
      "step": 24590
    },
    {
      "entropy": 6.101977300643921,
      "epoch": 2.14934894695447,
      "grad_norm": 3.5,
      "learning_rate": 0.00045363783690462845,
      "loss": 5.8118,
      "mean_token_accuracy": 0.16049007326364517,
      "num_tokens": 49831264.0,
      "step": 24595
    },
    {
      "entropy": 6.190305519104004,
      "epoch": 2.1497858953071747,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004536188896067483,
      "loss": 5.8136,
      "mean_token_accuracy": 0.16231686472892762,
      "num_tokens": 49841656.0,
      "step": 24600
    },
    {
      "entropy": 6.19525032043457,
      "epoch": 2.1502228436598796,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004535999388828645,
      "loss": 5.6853,
      "mean_token_accuracy": 0.1714086875319481,
      "num_tokens": 49851127.0,
      "step": 24605
    },
    {
      "entropy": 6.0963554859161375,
      "epoch": 2.150659792012584,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004535809847333402,
      "loss": 5.782,
      "mean_token_accuracy": 0.1601368561387062,
      "num_tokens": 49860563.0,
      "step": 24610
    },
    {
      "entropy": 6.109686756134034,
      "epoch": 2.151096740365289,
      "grad_norm": 2.3125,
      "learning_rate": 0.000453562027158539,
      "loss": 5.7052,
      "mean_token_accuracy": 0.16218193024396896,
      "num_tokens": 49869741.0,
      "step": 24615
    },
    {
      "entropy": 6.129808044433593,
      "epoch": 2.1515336887179934,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004535430661588245,
      "loss": 5.816,
      "mean_token_accuracy": 0.15960391461849213,
      "num_tokens": 49879664.0,
      "step": 24620
    },
    {
      "entropy": 6.126880407333374,
      "epoch": 2.1519706370706984,
      "grad_norm": 2.328125,
      "learning_rate": 0.00045352410173456057,
      "loss": 5.7782,
      "mean_token_accuracy": 0.1595143586397171,
      "num_tokens": 49889322.0,
      "step": 24625
    },
    {
      "entropy": 6.188329458236694,
      "epoch": 2.152407585423403,
      "grad_norm": 2.640625,
      "learning_rate": 0.00045350513388611074,
      "loss": 5.8375,
      "mean_token_accuracy": 0.15660687685012817,
      "num_tokens": 49899172.0,
      "step": 24630
    },
    {
      "entropy": 6.1552001476287845,
      "epoch": 2.1528445337761077,
      "grad_norm": 3.046875,
      "learning_rate": 0.00045348616261383897,
      "loss": 5.7787,
      "mean_token_accuracy": 0.16078456491231918,
      "num_tokens": 49909211.0,
      "step": 24635
    },
    {
      "entropy": 6.157931756973267,
      "epoch": 2.153281482128812,
      "grad_norm": 2.1875,
      "learning_rate": 0.00045346718791810896,
      "loss": 5.8039,
      "mean_token_accuracy": 0.16693703681230546,
      "num_tokens": 49918884.0,
      "step": 24640
    },
    {
      "entropy": 6.106607818603516,
      "epoch": 2.153718430481517,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045344820979928484,
      "loss": 5.7777,
      "mean_token_accuracy": 0.1650645300745964,
      "num_tokens": 49929558.0,
      "step": 24645
    },
    {
      "entropy": 6.146615266799927,
      "epoch": 2.1541553788342216,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045342922825773057,
      "loss": 5.8511,
      "mean_token_accuracy": 0.1614081606268883,
      "num_tokens": 49939831.0,
      "step": 24650
    },
    {
      "entropy": 6.170341300964355,
      "epoch": 2.1545923271869265,
      "grad_norm": 1.890625,
      "learning_rate": 0.00045341024329381024,
      "loss": 5.7605,
      "mean_token_accuracy": 0.16318252682685852,
      "num_tokens": 49949690.0,
      "step": 24655
    },
    {
      "entropy": 6.114606666564941,
      "epoch": 2.1550292755396314,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004533912549078879,
      "loss": 5.8091,
      "mean_token_accuracy": 0.15740249007940293,
      "num_tokens": 49959154.0,
      "step": 24660
    },
    {
      "entropy": 6.207797050476074,
      "epoch": 2.155466223892336,
      "grad_norm": 3.0625,
      "learning_rate": 0.00045337226310032783,
      "loss": 5.8137,
      "mean_token_accuracy": 0.16161729991436005,
      "num_tokens": 49970185.0,
      "step": 24665
    },
    {
      "entropy": 6.176773262023926,
      "epoch": 2.155903172245041,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004533532678714943,
      "loss": 5.8589,
      "mean_token_accuracy": 0.15883582159876825,
      "num_tokens": 49980489.0,
      "step": 24670
    },
    {
      "entropy": 6.115427827835083,
      "epoch": 2.1563401205977453,
      "grad_norm": 1.953125,
      "learning_rate": 0.00045333426922175156,
      "loss": 5.7963,
      "mean_token_accuracy": 0.1619945228099823,
      "num_tokens": 49990038.0,
      "step": 24675
    },
    {
      "entropy": 6.142610120773315,
      "epoch": 2.15677706895045,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004533152671514641,
      "loss": 5.7619,
      "mean_token_accuracy": 0.17011360675096512,
      "num_tokens": 49998888.0,
      "step": 24680
    },
    {
      "entropy": 6.060665082931519,
      "epoch": 2.1572140173031547,
      "grad_norm": 2.390625,
      "learning_rate": 0.00045329626166099653,
      "loss": 5.7578,
      "mean_token_accuracy": 0.16813307255506516,
      "num_tokens": 50009349.0,
      "step": 24685
    },
    {
      "entropy": 6.147572803497314,
      "epoch": 2.1576509656558596,
      "grad_norm": 2.484375,
      "learning_rate": 0.00045327725275071313,
      "loss": 5.8682,
      "mean_token_accuracy": 0.15451956242322923,
      "num_tokens": 50019626.0,
      "step": 24690
    },
    {
      "entropy": 6.19269061088562,
      "epoch": 2.158087914008564,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004532582404209786,
      "loss": 5.7507,
      "mean_token_accuracy": 0.1654902622103691,
      "num_tokens": 50029281.0,
      "step": 24695
    },
    {
      "entropy": 6.165007019042969,
      "epoch": 2.158524862361269,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004532392246721576,
      "loss": 5.7854,
      "mean_token_accuracy": 0.16426008492708205,
      "num_tokens": 50038700.0,
      "step": 24700
    },
    {
      "entropy": 6.154060745239258,
      "epoch": 2.1589618107139734,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004532202055046149,
      "loss": 5.7505,
      "mean_token_accuracy": 0.16433951556682586,
      "num_tokens": 50048904.0,
      "step": 24705
    },
    {
      "entropy": 6.107552623748779,
      "epoch": 2.1593987590666783,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004532011829187153,
      "loss": 5.7663,
      "mean_token_accuracy": 0.16192354410886764,
      "num_tokens": 50059188.0,
      "step": 24710
    },
    {
      "entropy": 6.049364137649536,
      "epoch": 2.159835707419383,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004531821569148236,
      "loss": 5.7359,
      "mean_token_accuracy": 0.1694171518087387,
      "num_tokens": 50069069.0,
      "step": 24715
    },
    {
      "entropy": 6.148147487640381,
      "epoch": 2.1602726557720877,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045316312749330473,
      "loss": 5.7634,
      "mean_token_accuracy": 0.16528197079896928,
      "num_tokens": 50078904.0,
      "step": 24720
    },
    {
      "entropy": 6.097108793258667,
      "epoch": 2.1607096041247926,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00045314409465452385,
      "loss": 5.7549,
      "mean_token_accuracy": 0.16322440207004546,
      "num_tokens": 50089897.0,
      "step": 24725
    },
    {
      "entropy": 6.140368938446045,
      "epoch": 2.161146552477497,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045312505839884573,
      "loss": 5.834,
      "mean_token_accuracy": 0.1578768253326416,
      "num_tokens": 50100030.0,
      "step": 24730
    },
    {
      "entropy": 6.171216678619385,
      "epoch": 2.161583500830202,
      "grad_norm": 5.15625,
      "learning_rate": 0.00045310601872663574,
      "loss": 5.7756,
      "mean_token_accuracy": 0.16482087224721909,
      "num_tokens": 50110043.0,
      "step": 24735
    },
    {
      "entropy": 6.114184713363647,
      "epoch": 2.1620204491829065,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00045308697563825894,
      "loss": 5.7757,
      "mean_token_accuracy": 0.1654902547597885,
      "num_tokens": 50120081.0,
      "step": 24740
    },
    {
      "entropy": 6.1478691577911375,
      "epoch": 2.1624573975356114,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004530679291340807,
      "loss": 5.7892,
      "mean_token_accuracy": 0.16738145798444748,
      "num_tokens": 50129834.0,
      "step": 24745
    },
    {
      "entropy": 6.150910806655884,
      "epoch": 2.162894345888316,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004530488792144662,
      "loss": 5.843,
      "mean_token_accuracy": 0.15216249376535415,
      "num_tokens": 50140853.0,
      "step": 24750
    },
    {
      "entropy": 6.161650800704956,
      "epoch": 2.1633312942410208,
      "grad_norm": 2.265625,
      "learning_rate": 0.000453029825879781,
      "loss": 5.8688,
      "mean_token_accuracy": 0.16252606958150864,
      "num_tokens": 50151341.0,
      "step": 24755
    },
    {
      "entropy": 6.144836854934693,
      "epoch": 2.1637682425937252,
      "grad_norm": 1.875,
      "learning_rate": 0.00045301076913039035,
      "loss": 5.7289,
      "mean_token_accuracy": 0.17165855318307877,
      "num_tokens": 50160826.0,
      "step": 24760
    },
    {
      "entropy": 6.1121909618377686,
      "epoch": 2.16420519094643,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004529917089666599,
      "loss": 5.7348,
      "mean_token_accuracy": 0.16186702102422715,
      "num_tokens": 50170720.0,
      "step": 24765
    },
    {
      "entropy": 6.136308050155639,
      "epoch": 2.1646421392991346,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004529726453889552,
      "loss": 5.783,
      "mean_token_accuracy": 0.16133645474910735,
      "num_tokens": 50180707.0,
      "step": 24770
    },
    {
      "entropy": 6.039327907562256,
      "epoch": 2.1650790876518395,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045295357839764193,
      "loss": 5.7483,
      "mean_token_accuracy": 0.17086519300937653,
      "num_tokens": 50190714.0,
      "step": 24775
    },
    {
      "entropy": 6.177323341369629,
      "epoch": 2.1655160360045445,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004529345079930858,
      "loss": 5.8574,
      "mean_token_accuracy": 0.1574186220765114,
      "num_tokens": 50201023.0,
      "step": 24780
    },
    {
      "entropy": 6.193730258941651,
      "epoch": 2.165952984357249,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004529154341756525,
      "loss": 5.7908,
      "mean_token_accuracy": 0.15708331912755966,
      "num_tokens": 50211064.0,
      "step": 24785
    },
    {
      "entropy": 6.185175180435181,
      "epoch": 2.166389932709954,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00045289635694570796,
      "loss": 5.8349,
      "mean_token_accuracy": 0.15737575441598892,
      "num_tokens": 50220696.0,
      "step": 24790
    },
    {
      "entropy": 6.0644454002380375,
      "epoch": 2.1668268810626583,
      "grad_norm": 1.859375,
      "learning_rate": 0.00045287727630361814,
      "loss": 5.7657,
      "mean_token_accuracy": 0.16498317271471025,
      "num_tokens": 50230629.0,
      "step": 24795
    },
    {
      "entropy": 6.08183479309082,
      "epoch": 2.167263829415363,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045285819224974894,
      "loss": 5.7554,
      "mean_token_accuracy": 0.16630676835775376,
      "num_tokens": 50241042.0,
      "step": 24800
    },
    {
      "entropy": 6.130773973464966,
      "epoch": 2.1677007777680677,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004528391047844664,
      "loss": 5.7143,
      "mean_token_accuracy": 0.16635574847459794,
      "num_tokens": 50251731.0,
      "step": 24805
    },
    {
      "entropy": 6.172975635528564,
      "epoch": 2.1681377261207726,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004528200139081366,
      "loss": 5.8279,
      "mean_token_accuracy": 0.15807994753122329,
      "num_tokens": 50262893.0,
      "step": 24810
    },
    {
      "entropy": 6.144801235198974,
      "epoch": 2.168574674473477,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004528009196211258,
      "loss": 5.9148,
      "mean_token_accuracy": 0.15343134999275207,
      "num_tokens": 50273666.0,
      "step": 24815
    },
    {
      "entropy": 6.1593918800354,
      "epoch": 2.169011622826182,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004527818219238002,
      "loss": 5.7401,
      "mean_token_accuracy": 0.1623699799180031,
      "num_tokens": 50283165.0,
      "step": 24820
    },
    {
      "entropy": 6.141596078872681,
      "epoch": 2.1694485711788865,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004527627208165261,
      "loss": 5.6968,
      "mean_token_accuracy": 0.1663254588842392,
      "num_tokens": 50291958.0,
      "step": 24825
    },
    {
      "entropy": 6.119465160369873,
      "epoch": 2.1698855195315914,
      "grad_norm": 1.84375,
      "learning_rate": 0.00045274361629966985,
      "loss": 5.8525,
      "mean_token_accuracy": 0.1578680768609047,
      "num_tokens": 50302944.0,
      "step": 24830
    },
    {
      "entropy": 6.186633205413818,
      "epoch": 2.1703224678842963,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045272450837359783,
      "loss": 5.784,
      "mean_token_accuracy": 0.15957930386066438,
      "num_tokens": 50313088.0,
      "step": 24835
    },
    {
      "entropy": 6.135117292404175,
      "epoch": 2.1707594162370007,
      "grad_norm": 2.9375,
      "learning_rate": 0.00045270539703867675,
      "loss": 5.8141,
      "mean_token_accuracy": 0.1537604346871376,
      "num_tokens": 50323121.0,
      "step": 24840
    },
    {
      "entropy": 6.132446050643921,
      "epoch": 2.1711963645897057,
      "grad_norm": 2.109375,
      "learning_rate": 0.000452686282295273,
      "loss": 5.809,
      "mean_token_accuracy": 0.1629273772239685,
      "num_tokens": 50333091.0,
      "step": 24845
    },
    {
      "entropy": 6.1767065048217775,
      "epoch": 2.17163331294241,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00045266716414375324,
      "loss": 5.7135,
      "mean_token_accuracy": 0.16795559227466583,
      "num_tokens": 50342752.0,
      "step": 24850
    },
    {
      "entropy": 6.102734518051148,
      "epoch": 2.172070261295115,
      "grad_norm": 3.34375,
      "learning_rate": 0.00045264804258448416,
      "loss": 5.7441,
      "mean_token_accuracy": 0.16696255803108215,
      "num_tokens": 50352213.0,
      "step": 24855
    },
    {
      "entropy": 6.080652236938477,
      "epoch": 2.1725072096478195,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004526289176178325,
      "loss": 5.8196,
      "mean_token_accuracy": 0.15774286538362503,
      "num_tokens": 50362586.0,
      "step": 24860
    },
    {
      "entropy": 6.116003322601318,
      "epoch": 2.1729441580005244,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004526097892441652,
      "loss": 5.6902,
      "mean_token_accuracy": 0.16732027977705002,
      "num_tokens": 50372742.0,
      "step": 24865
    },
    {
      "entropy": 6.035664844512939,
      "epoch": 2.173381106353229,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045259065746384904,
      "loss": 5.685,
      "mean_token_accuracy": 0.17128758281469345,
      "num_tokens": 50382857.0,
      "step": 24870
    },
    {
      "entropy": 6.149796724319458,
      "epoch": 2.173818054705934,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000452571522277251,
      "loss": 5.7893,
      "mean_token_accuracy": 0.16124936789274216,
      "num_tokens": 50393638.0,
      "step": 24875
    },
    {
      "entropy": 6.152971315383911,
      "epoch": 2.1742550030586383,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004525523836847382,
      "loss": 5.8289,
      "mean_token_accuracy": 0.16068097800016404,
      "num_tokens": 50404928.0,
      "step": 24880
    },
    {
      "entropy": 6.100503635406494,
      "epoch": 2.174691951411343,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004525332416866776,
      "loss": 5.6872,
      "mean_token_accuracy": 0.1697850286960602,
      "num_tokens": 50414382.0,
      "step": 24885
    },
    {
      "entropy": 6.123734855651856,
      "epoch": 2.175128899764048,
      "grad_norm": 2.25,
      "learning_rate": 0.0004525140962834363,
      "loss": 5.777,
      "mean_token_accuracy": 0.16675321459770204,
      "num_tokens": 50425203.0,
      "step": 24890
    },
    {
      "entropy": 6.186834716796875,
      "epoch": 2.1755658481167526,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004524949474753817,
      "loss": 5.9305,
      "mean_token_accuracy": 0.14710071235895156,
      "num_tokens": 50436395.0,
      "step": 24895
    },
    {
      "entropy": 6.114315795898437,
      "epoch": 2.1760027964694575,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004524757952628811,
      "loss": 5.7634,
      "mean_token_accuracy": 0.1714218944311142,
      "num_tokens": 50446990.0,
      "step": 24900
    },
    {
      "entropy": 6.140411615371704,
      "epoch": 2.176439744822162,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00045245663964630164,
      "loss": 5.8417,
      "mean_token_accuracy": 0.15442856252193451,
      "num_tokens": 50457964.0,
      "step": 24905
    },
    {
      "entropy": 6.187087821960449,
      "epoch": 2.176876693174867,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004524374806260108,
      "loss": 5.7538,
      "mean_token_accuracy": 0.1611899331212044,
      "num_tokens": 50467329.0,
      "step": 24910
    },
    {
      "entropy": 6.123509550094605,
      "epoch": 2.1773136415275713,
      "grad_norm": 2.125,
      "learning_rate": 0.0004524183182023762,
      "loss": 5.7723,
      "mean_token_accuracy": 0.16551902294158935,
      "num_tokens": 50476852.0,
      "step": 24915
    },
    {
      "entropy": 6.099613428115845,
      "epoch": 2.1777505898802763,
      "grad_norm": 3.953125,
      "learning_rate": 0.00045239915237576525,
      "loss": 5.7512,
      "mean_token_accuracy": 0.16198043525218964,
      "num_tokens": 50486207.0,
      "step": 24920
    },
    {
      "entropy": 6.140814113616943,
      "epoch": 2.1781875382329807,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00045237998314654555,
      "loss": 5.8512,
      "mean_token_accuracy": 0.16337123811244963,
      "num_tokens": 50496457.0,
      "step": 24925
    },
    {
      "entropy": 6.1625080585479735,
      "epoch": 2.1786244865856856,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004523608105150849,
      "loss": 5.838,
      "mean_token_accuracy": 0.15290412902832032,
      "num_tokens": 50506430.0,
      "step": 24930
    },
    {
      "entropy": 6.151779270172119,
      "epoch": 2.17906143493839,
      "grad_norm": 2.390625,
      "learning_rate": 0.000452341634481751,
      "loss": 5.7239,
      "mean_token_accuracy": 0.15993465185165406,
      "num_tokens": 50516204.0,
      "step": 24935
    },
    {
      "entropy": 6.122723150253296,
      "epoch": 2.179498383291095,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004523224550469115,
      "loss": 5.7795,
      "mean_token_accuracy": 0.15817153155803682,
      "num_tokens": 50525978.0,
      "step": 24940
    },
    {
      "entropy": 6.097207403182983,
      "epoch": 2.1799353316438,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004523032722109345,
      "loss": 5.7562,
      "mean_token_accuracy": 0.15931738764047623,
      "num_tokens": 50535579.0,
      "step": 24945
    },
    {
      "entropy": 6.130011415481567,
      "epoch": 2.1803722799965044,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004522840859741878,
      "loss": 5.7529,
      "mean_token_accuracy": 0.16249380856752396,
      "num_tokens": 50545765.0,
      "step": 24950
    },
    {
      "entropy": 6.191589641571045,
      "epoch": 2.1808092283492093,
      "grad_norm": 2.546875,
      "learning_rate": 0.00045226489633703944,
      "loss": 5.9208,
      "mean_token_accuracy": 0.15862150490283966,
      "num_tokens": 50556237.0,
      "step": 24955
    },
    {
      "entropy": 6.254256820678711,
      "epoch": 2.181246176701914,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00045224570329985747,
      "loss": 5.8676,
      "mean_token_accuracy": 0.15987957566976546,
      "num_tokens": 50567720.0,
      "step": 24960
    },
    {
      "entropy": 6.14825701713562,
      "epoch": 2.1816831250546187,
      "grad_norm": 1.453125,
      "learning_rate": 0.00045222650686301,
      "loss": 5.7559,
      "mean_token_accuracy": 0.16713852435350418,
      "num_tokens": 50577234.0,
      "step": 24965
    },
    {
      "entropy": 6.132482671737671,
      "epoch": 2.182120073407323,
      "grad_norm": 4.0,
      "learning_rate": 0.0004522073070268654,
      "loss": 5.8293,
      "mean_token_accuracy": 0.1624958857893944,
      "num_tokens": 50587416.0,
      "step": 24970
    },
    {
      "entropy": 6.0945484161376955,
      "epoch": 2.182557021760028,
      "grad_norm": 2.75,
      "learning_rate": 0.0004521881037917917,
      "loss": 5.7218,
      "mean_token_accuracy": 0.17031861692667008,
      "num_tokens": 50598317.0,
      "step": 24975
    },
    {
      "entropy": 6.151479196548462,
      "epoch": 2.1829939701127326,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004521688971581573,
      "loss": 5.8281,
      "mean_token_accuracy": 0.1588055856525898,
      "num_tokens": 50608507.0,
      "step": 24980
    },
    {
      "entropy": 6.123809909820556,
      "epoch": 2.1834309184654375,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045214968712633066,
      "loss": 5.7624,
      "mean_token_accuracy": 0.1640541136264801,
      "num_tokens": 50618625.0,
      "step": 24985
    },
    {
      "entropy": 6.172607040405273,
      "epoch": 2.183867866818142,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004521304736966802,
      "loss": 5.7888,
      "mean_token_accuracy": 0.16223382204771042,
      "num_tokens": 50628696.0,
      "step": 24990
    },
    {
      "entropy": 6.2112956047058105,
      "epoch": 2.184304815170847,
      "grad_norm": 2.328125,
      "learning_rate": 0.00045211125686957443,
      "loss": 5.8462,
      "mean_token_accuracy": 0.15745509266853333,
      "num_tokens": 50639313.0,
      "step": 24995
    },
    {
      "entropy": 6.1532135009765625,
      "epoch": 2.1847417635235513,
      "grad_norm": 2.1875,
      "learning_rate": 0.00045209203664538195,
      "loss": 5.7562,
      "mean_token_accuracy": 0.16347122341394424,
      "num_tokens": 50650715.0,
      "step": 25000
    },
    {
      "entropy": 6.13451075553894,
      "epoch": 2.1851787118762562,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004520728130244714,
      "loss": 5.771,
      "mean_token_accuracy": 0.16391534656286239,
      "num_tokens": 50660224.0,
      "step": 25005
    },
    {
      "entropy": 6.122021484375,
      "epoch": 2.185615660228961,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004520535860072115,
      "loss": 5.7656,
      "mean_token_accuracy": 0.16573483496904373,
      "num_tokens": 50670346.0,
      "step": 25010
    },
    {
      "entropy": 6.089285087585449,
      "epoch": 2.1860526085816656,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045203435559397113,
      "loss": 5.7728,
      "mean_token_accuracy": 0.16445882469415665,
      "num_tokens": 50680274.0,
      "step": 25015
    },
    {
      "entropy": 6.096529531478882,
      "epoch": 2.1864895569343705,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004520151217851189,
      "loss": 5.7592,
      "mean_token_accuracy": 0.16453305035829544,
      "num_tokens": 50690360.0,
      "step": 25020
    },
    {
      "entropy": 6.148435878753662,
      "epoch": 2.186926505287075,
      "grad_norm": 1.90625,
      "learning_rate": 0.000451995884581024,
      "loss": 5.8071,
      "mean_token_accuracy": 0.162690170109272,
      "num_tokens": 50700798.0,
      "step": 25025
    },
    {
      "entropy": 6.115629816055298,
      "epoch": 2.18736345363978,
      "grad_norm": 3.625,
      "learning_rate": 0.0004519766439820552,
      "loss": 5.8084,
      "mean_token_accuracy": 0.1594453528523445,
      "num_tokens": 50710120.0,
      "step": 25030
    },
    {
      "entropy": 6.203378057479858,
      "epoch": 2.1878004019924844,
      "grad_norm": 4.5,
      "learning_rate": 0.0004519573999885817,
      "loss": 5.8893,
      "mean_token_accuracy": 0.16404750049114228,
      "num_tokens": 50719903.0,
      "step": 25035
    },
    {
      "entropy": 6.1327643394470215,
      "epoch": 2.1882373503451893,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045193815260097253,
      "loss": 5.8139,
      "mean_token_accuracy": 0.16141761243343353,
      "num_tokens": 50730632.0,
      "step": 25040
    },
    {
      "entropy": 6.205726671218872,
      "epoch": 2.1886742986978938,
      "grad_norm": 2.703125,
      "learning_rate": 0.00045191890181959696,
      "loss": 5.8226,
      "mean_token_accuracy": 0.15366312563419343,
      "num_tokens": 50741263.0,
      "step": 25045
    },
    {
      "entropy": 6.126460313796997,
      "epoch": 2.1891112470505987,
      "grad_norm": 3.640625,
      "learning_rate": 0.00045189964764482407,
      "loss": 5.7676,
      "mean_token_accuracy": 0.16021917313337325,
      "num_tokens": 50751394.0,
      "step": 25050
    },
    {
      "entropy": 6.123516511917114,
      "epoch": 2.189548195403303,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004518803900770232,
      "loss": 5.8058,
      "mean_token_accuracy": 0.15901520401239394,
      "num_tokens": 50762038.0,
      "step": 25055
    },
    {
      "entropy": 6.198383951187134,
      "epoch": 2.189985143756008,
      "grad_norm": 2.328125,
      "learning_rate": 0.00045186112911656387,
      "loss": 5.8557,
      "mean_token_accuracy": 0.15164461433887483,
      "num_tokens": 50774063.0,
      "step": 25060
    },
    {
      "entropy": 6.218735885620117,
      "epoch": 2.1904220921087125,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004518418647638153,
      "loss": 5.8304,
      "mean_token_accuracy": 0.1549309842288494,
      "num_tokens": 50784606.0,
      "step": 25065
    },
    {
      "entropy": 6.167850017547607,
      "epoch": 2.1908590404614174,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004518225970191471,
      "loss": 5.8052,
      "mean_token_accuracy": 0.15589399337768556,
      "num_tokens": 50794892.0,
      "step": 25070
    },
    {
      "entropy": 6.072506999969482,
      "epoch": 2.1912959888141224,
      "grad_norm": 2.25,
      "learning_rate": 0.00045180332588292896,
      "loss": 5.8198,
      "mean_token_accuracy": 0.1607952207326889,
      "num_tokens": 50805903.0,
      "step": 25075
    },
    {
      "entropy": 6.092518949508667,
      "epoch": 2.191732937166827,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004517840513555303,
      "loss": 5.6681,
      "mean_token_accuracy": 0.16844437271356583,
      "num_tokens": 50815339.0,
      "step": 25080
    },
    {
      "entropy": 6.178649187088013,
      "epoch": 2.1921698855195317,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004517647734373209,
      "loss": 5.8159,
      "mean_token_accuracy": 0.1629050001502037,
      "num_tokens": 50825575.0,
      "step": 25085
    },
    {
      "entropy": 6.166235303878784,
      "epoch": 2.192606833872236,
      "grad_norm": 2.625,
      "learning_rate": 0.00045174549212867053,
      "loss": 5.846,
      "mean_token_accuracy": 0.15281661674380304,
      "num_tokens": 50836106.0,
      "step": 25090
    },
    {
      "entropy": 6.117022943496704,
      "epoch": 2.193043782224941,
      "grad_norm": 2.25,
      "learning_rate": 0.00045172620742994904,
      "loss": 5.8156,
      "mean_token_accuracy": 0.15703723356127738,
      "num_tokens": 50847350.0,
      "step": 25095
    },
    {
      "entropy": 6.171441125869751,
      "epoch": 2.1934807305776456,
      "grad_norm": 2.0,
      "learning_rate": 0.0004517069193415263,
      "loss": 5.6975,
      "mean_token_accuracy": 0.1673838108778,
      "num_tokens": 50857543.0,
      "step": 25100
    },
    {
      "entropy": 6.181086874008178,
      "epoch": 2.1939176789303505,
      "grad_norm": 3.03125,
      "learning_rate": 0.00045168762786377214,
      "loss": 5.8058,
      "mean_token_accuracy": 0.1565091535449028,
      "num_tokens": 50868791.0,
      "step": 25105
    },
    {
      "entropy": 6.177199602127075,
      "epoch": 2.194354627283055,
      "grad_norm": 2.890625,
      "learning_rate": 0.00045166833299705686,
      "loss": 5.7797,
      "mean_token_accuracy": 0.15851620435714722,
      "num_tokens": 50879244.0,
      "step": 25110
    },
    {
      "entropy": 6.2024986743927,
      "epoch": 2.19479157563576,
      "grad_norm": 4.5,
      "learning_rate": 0.00045164903474175026,
      "loss": 5.8015,
      "mean_token_accuracy": 0.16427187919616698,
      "num_tokens": 50888935.0,
      "step": 25115
    },
    {
      "entropy": 6.1697096824646,
      "epoch": 2.1952285239884644,
      "grad_norm": 2.125,
      "learning_rate": 0.00045162973309822264,
      "loss": 5.773,
      "mean_token_accuracy": 0.16581937819719314,
      "num_tokens": 50898152.0,
      "step": 25120
    },
    {
      "entropy": 6.102431344985962,
      "epoch": 2.1956654723411693,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004516104280668442,
      "loss": 5.7951,
      "mean_token_accuracy": 0.15648173093795775,
      "num_tokens": 50908555.0,
      "step": 25125
    },
    {
      "entropy": 6.178210830688476,
      "epoch": 2.196102420693874,
      "grad_norm": 2.703125,
      "learning_rate": 0.00045159111964798514,
      "loss": 5.9041,
      "mean_token_accuracy": 0.15144950598478318,
      "num_tokens": 50919078.0,
      "step": 25130
    },
    {
      "entropy": 6.165487813949585,
      "epoch": 2.1965393690465786,
      "grad_norm": 4.125,
      "learning_rate": 0.00045157180784201594,
      "loss": 5.7776,
      "mean_token_accuracy": 0.16360928416252135,
      "num_tokens": 50928879.0,
      "step": 25135
    },
    {
      "entropy": 6.167407369613647,
      "epoch": 2.1969763173992836,
      "grad_norm": 1.8125,
      "learning_rate": 0.000451552492649307,
      "loss": 5.7907,
      "mean_token_accuracy": 0.16346344351768494,
      "num_tokens": 50939456.0,
      "step": 25140
    },
    {
      "entropy": 6.132479906082153,
      "epoch": 2.197413265751988,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045153317407022866,
      "loss": 5.7012,
      "mean_token_accuracy": 0.1711938798427582,
      "num_tokens": 50948789.0,
      "step": 25145
    },
    {
      "entropy": 6.077538728713989,
      "epoch": 2.197850214104693,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045151385210515153,
      "loss": 5.7277,
      "mean_token_accuracy": 0.15712295174598695,
      "num_tokens": 50958492.0,
      "step": 25150
    },
    {
      "entropy": 6.1080340385437015,
      "epoch": 2.1982871624573974,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045149452675444625,
      "loss": 5.7732,
      "mean_token_accuracy": 0.16581079959869385,
      "num_tokens": 50968814.0,
      "step": 25155
    },
    {
      "entropy": 6.173214483261108,
      "epoch": 2.1987241108101023,
      "grad_norm": 2.125,
      "learning_rate": 0.00045147519801848346,
      "loss": 5.777,
      "mean_token_accuracy": 0.15995731204748154,
      "num_tokens": 50979680.0,
      "step": 25160
    },
    {
      "entropy": 6.115719413757324,
      "epoch": 2.199161059162807,
      "grad_norm": 4.90625,
      "learning_rate": 0.00045145586589763394,
      "loss": 5.8014,
      "mean_token_accuracy": 0.16113118082284927,
      "num_tokens": 50990827.0,
      "step": 25165
    },
    {
      "entropy": 6.113433790206909,
      "epoch": 2.1995980075155117,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004514365303922684,
      "loss": 5.7858,
      "mean_token_accuracy": 0.1618678316473961,
      "num_tokens": 51001065.0,
      "step": 25170
    },
    {
      "entropy": 6.0920172214508055,
      "epoch": 2.200034955868216,
      "grad_norm": 1.953125,
      "learning_rate": 0.00045141719150275773,
      "loss": 5.7437,
      "mean_token_accuracy": 0.17024223357439042,
      "num_tokens": 51011348.0,
      "step": 25175
    },
    {
      "entropy": 6.138101959228516,
      "epoch": 2.200471904220921,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045139784922947296,
      "loss": 5.7444,
      "mean_token_accuracy": 0.16221659630537033,
      "num_tokens": 51022033.0,
      "step": 25180
    },
    {
      "entropy": 6.103479719161987,
      "epoch": 2.200908852573626,
      "grad_norm": 2.53125,
      "learning_rate": 0.000451378503572785,
      "loss": 5.8043,
      "mean_token_accuracy": 0.15817313343286515,
      "num_tokens": 51032001.0,
      "step": 25185
    },
    {
      "entropy": 6.1210743427276615,
      "epoch": 2.2013458009263305,
      "grad_norm": 2.921875,
      "learning_rate": 0.00045135915453306487,
      "loss": 5.8425,
      "mean_token_accuracy": 0.16085945516824723,
      "num_tokens": 51041227.0,
      "step": 25190
    },
    {
      "entropy": 6.024200296401977,
      "epoch": 2.2017827492790354,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00045133980211068374,
      "loss": 5.6212,
      "mean_token_accuracy": 0.17677876055240632,
      "num_tokens": 51050642.0,
      "step": 25195
    },
    {
      "entropy": 6.189979982376099,
      "epoch": 2.20221969763174,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004513204463060129,
      "loss": 5.7119,
      "mean_token_accuracy": 0.1634989470243454,
      "num_tokens": 51060646.0,
      "step": 25200
    },
    {
      "entropy": 6.116321992874146,
      "epoch": 2.2026566459844448,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004513010871194234,
      "loss": 5.8446,
      "mean_token_accuracy": 0.16332965940237046,
      "num_tokens": 51071762.0,
      "step": 25205
    },
    {
      "entropy": 6.095517587661743,
      "epoch": 2.2030935943371492,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00045128172455128674,
      "loss": 5.7964,
      "mean_token_accuracy": 0.16646843254566193,
      "num_tokens": 51080802.0,
      "step": 25210
    },
    {
      "entropy": 6.163711071014404,
      "epoch": 2.203530542689854,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004512623586019742,
      "loss": 5.8427,
      "mean_token_accuracy": 0.15103527009487153,
      "num_tokens": 51091802.0,
      "step": 25215
    },
    {
      "entropy": 6.113758277893067,
      "epoch": 2.2039674910425586,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004512429892718573,
      "loss": 5.7328,
      "mean_token_accuracy": 0.17165565639734268,
      "num_tokens": 51100939.0,
      "step": 25220
    },
    {
      "entropy": 6.063447618484497,
      "epoch": 2.2044044393952635,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004512236165613074,
      "loss": 5.7577,
      "mean_token_accuracy": 0.16111077666282653,
      "num_tokens": 51110928.0,
      "step": 25225
    },
    {
      "entropy": 6.129931545257568,
      "epoch": 2.204841387747968,
      "grad_norm": 2.609375,
      "learning_rate": 0.00045120424047069633,
      "loss": 5.8014,
      "mean_token_accuracy": 0.15778490602970124,
      "num_tokens": 51121126.0,
      "step": 25230
    },
    {
      "entropy": 6.171729707717896,
      "epoch": 2.205278336100673,
      "grad_norm": 2.890625,
      "learning_rate": 0.00045118486100039553,
      "loss": 5.7015,
      "mean_token_accuracy": 0.16990143060684204,
      "num_tokens": 51130425.0,
      "step": 25235
    },
    {
      "entropy": 6.115838670730591,
      "epoch": 2.205715284453378,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004511654781507768,
      "loss": 5.7539,
      "mean_token_accuracy": 0.16205742806196213,
      "num_tokens": 51139994.0,
      "step": 25240
    },
    {
      "entropy": 6.101830244064331,
      "epoch": 2.2061522328060823,
      "grad_norm": 2.375,
      "learning_rate": 0.00045114609192221196,
      "loss": 5.7976,
      "mean_token_accuracy": 0.15661717355251312,
      "num_tokens": 51149762.0,
      "step": 25245
    },
    {
      "entropy": 6.036141633987427,
      "epoch": 2.206589181158787,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004511267023150727,
      "loss": 5.7548,
      "mean_token_accuracy": 0.17089859694242476,
      "num_tokens": 51161091.0,
      "step": 25250
    },
    {
      "entropy": 6.161953115463257,
      "epoch": 2.2070261295114917,
      "grad_norm": 2.03125,
      "learning_rate": 0.000451107309329731,
      "loss": 5.762,
      "mean_token_accuracy": 0.16462596803903579,
      "num_tokens": 51170749.0,
      "step": 25255
    },
    {
      "entropy": 6.153527021408081,
      "epoch": 2.2074630778641966,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045108791296655877,
      "loss": 5.7364,
      "mean_token_accuracy": 0.16257859766483307,
      "num_tokens": 51179515.0,
      "step": 25260
    },
    {
      "entropy": 6.073777675628662,
      "epoch": 2.207900026216901,
      "grad_norm": 2.6875,
      "learning_rate": 0.00045106851322592817,
      "loss": 5.8147,
      "mean_token_accuracy": 0.167317758500576,
      "num_tokens": 51189036.0,
      "step": 25265
    },
    {
      "entropy": 6.138585901260376,
      "epoch": 2.208336974569606,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004510491101082112,
      "loss": 5.8303,
      "mean_token_accuracy": 0.16058240383863448,
      "num_tokens": 51199967.0,
      "step": 25270
    },
    {
      "entropy": 6.113816928863526,
      "epoch": 2.2087739229223105,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004510297036137801,
      "loss": 5.7919,
      "mean_token_accuracy": 0.1650155231356621,
      "num_tokens": 51210281.0,
      "step": 25275
    },
    {
      "entropy": 6.159410524368286,
      "epoch": 2.2092108712750154,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045101029374300703,
      "loss": 5.8407,
      "mean_token_accuracy": 0.15952381640672683,
      "num_tokens": 51220209.0,
      "step": 25280
    },
    {
      "entropy": 6.215707874298095,
      "epoch": 2.20964781962772,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004509908804962643,
      "loss": 5.7896,
      "mean_token_accuracy": 0.16523460000753404,
      "num_tokens": 51231108.0,
      "step": 25285
    },
    {
      "entropy": 6.068835258483887,
      "epoch": 2.2100847679804247,
      "grad_norm": 2.859375,
      "learning_rate": 0.00045097146387392423,
      "loss": 5.7434,
      "mean_token_accuracy": 0.16433213502168656,
      "num_tokens": 51241316.0,
      "step": 25290
    },
    {
      "entropy": 6.102057981491089,
      "epoch": 2.210521716333129,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004509520438763593,
      "loss": 5.7824,
      "mean_token_accuracy": 0.16462171226739883,
      "num_tokens": 51250854.0,
      "step": 25295
    },
    {
      "entropy": 6.12411732673645,
      "epoch": 2.210958664685834,
      "grad_norm": 3.3125,
      "learning_rate": 0.00045093262050394193,
      "loss": 5.7742,
      "mean_token_accuracy": 0.158060921728611,
      "num_tokens": 51261001.0,
      "step": 25300
    },
    {
      "entropy": 6.120763921737671,
      "epoch": 2.211395613038539,
      "grad_norm": 2.703125,
      "learning_rate": 0.00045091319375704475,
      "loss": 5.806,
      "mean_token_accuracy": 0.1660653382539749,
      "num_tokens": 51270815.0,
      "step": 25305
    },
    {
      "entropy": 6.180273008346558,
      "epoch": 2.2118325613912435,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045089376363604037,
      "loss": 5.7818,
      "mean_token_accuracy": 0.1610511839389801,
      "num_tokens": 51281370.0,
      "step": 25310
    },
    {
      "entropy": 6.164554309844971,
      "epoch": 2.2122695097439484,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045087433014130143,
      "loss": 5.797,
      "mean_token_accuracy": 0.1734825924038887,
      "num_tokens": 51290147.0,
      "step": 25315
    },
    {
      "entropy": 6.1689615726470945,
      "epoch": 2.212706458096653,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004508548932732006,
      "loss": 5.7745,
      "mean_token_accuracy": 0.165625761449337,
      "num_tokens": 51299487.0,
      "step": 25320
    },
    {
      "entropy": 6.17425799369812,
      "epoch": 2.213143406449358,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004508354530321108,
      "loss": 5.7397,
      "mean_token_accuracy": 0.1677071139216423,
      "num_tokens": 51309425.0,
      "step": 25325
    },
    {
      "entropy": 6.137548875808716,
      "epoch": 2.2135803548020623,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004508160094184049,
      "loss": 5.7694,
      "mean_token_accuracy": 0.15933065637946128,
      "num_tokens": 51319962.0,
      "step": 25330
    },
    {
      "entropy": 6.1414713859558105,
      "epoch": 2.214017303154767,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004507965624324558,
      "loss": 5.7805,
      "mean_token_accuracy": 0.16383811533451081,
      "num_tokens": 51330892.0,
      "step": 25335
    },
    {
      "entropy": 6.1723411083221436,
      "epoch": 2.2144542515074717,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045077711207463654,
      "loss": 5.81,
      "mean_token_accuracy": 0.1556896910071373,
      "num_tokens": 51342783.0,
      "step": 25340
    },
    {
      "entropy": 6.2059091091156,
      "epoch": 2.2148911998601766,
      "grad_norm": 3.515625,
      "learning_rate": 0.00045075765834532024,
      "loss": 5.8315,
      "mean_token_accuracy": 0.15537831038236619,
      "num_tokens": 51352890.0,
      "step": 25345
    },
    {
      "entropy": 6.147296380996704,
      "epoch": 2.215328148212881,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004507382012448798,
      "loss": 5.8645,
      "mean_token_accuracy": 0.15379691123962402,
      "num_tokens": 51363479.0,
      "step": 25350
    },
    {
      "entropy": 6.167978572845459,
      "epoch": 2.215765096565586,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004507187407736886,
      "loss": 5.8064,
      "mean_token_accuracy": 0.1618042677640915,
      "num_tokens": 51374304.0,
      "step": 25355
    },
    {
      "entropy": 6.124129581451416,
      "epoch": 2.2162020449182904,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004506992769321199,
      "loss": 5.8318,
      "mean_token_accuracy": 0.15626233220100402,
      "num_tokens": 51384649.0,
      "step": 25360
    },
    {
      "entropy": 6.137325096130371,
      "epoch": 2.2166389932709953,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045067980972054707,
      "loss": 5.7695,
      "mean_token_accuracy": 0.16066195368766784,
      "num_tokens": 51394582.0,
      "step": 25365
    },
    {
      "entropy": 6.177107954025269,
      "epoch": 2.2170759416237003,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004506603391393434,
      "loss": 5.775,
      "mean_token_accuracy": 0.1618586853146553,
      "num_tokens": 51404618.0,
      "step": 25370
    },
    {
      "entropy": 6.111138248443604,
      "epoch": 2.2175128899764047,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004506408651888824,
      "loss": 5.7921,
      "mean_token_accuracy": 0.16030359268188477,
      "num_tokens": 51414810.0,
      "step": 25375
    },
    {
      "entropy": 6.121479129791259,
      "epoch": 2.2179498383291096,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004506213878695375,
      "loss": 5.8677,
      "mean_token_accuracy": 0.1609814703464508,
      "num_tokens": 51426118.0,
      "step": 25380
    },
    {
      "entropy": 6.180749320983887,
      "epoch": 2.218386786681814,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045060190718168244,
      "loss": 5.7843,
      "mean_token_accuracy": 0.16862816959619523,
      "num_tokens": 51435260.0,
      "step": 25385
    },
    {
      "entropy": 6.147540187835693,
      "epoch": 2.218823735034519,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004505824231256907,
      "loss": 5.7403,
      "mean_token_accuracy": 0.16758888810873032,
      "num_tokens": 51444778.0,
      "step": 25390
    },
    {
      "entropy": 6.109844493865967,
      "epoch": 2.2192606833872235,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004505629357019361,
      "loss": 5.8101,
      "mean_token_accuracy": 0.16392170935869216,
      "num_tokens": 51456030.0,
      "step": 25395
    },
    {
      "entropy": 6.175970029830933,
      "epoch": 2.2196976317399284,
      "grad_norm": 1.75,
      "learning_rate": 0.0004505434449107925,
      "loss": 5.7801,
      "mean_token_accuracy": 0.16062992662191392,
      "num_tokens": 51465836.0,
      "step": 25400
    },
    {
      "entropy": 6.113465261459351,
      "epoch": 2.220134580092633,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004505239507526335,
      "loss": 5.7223,
      "mean_token_accuracy": 0.16692662835121155,
      "num_tokens": 51475330.0,
      "step": 25405
    },
    {
      "entropy": 6.060976171493531,
      "epoch": 2.220571528445338,
      "grad_norm": 1.96875,
      "learning_rate": 0.00045050445322783326,
      "loss": 5.718,
      "mean_token_accuracy": 0.16712878197431563,
      "num_tokens": 51486646.0,
      "step": 25410
    },
    {
      "entropy": 6.078708600997925,
      "epoch": 2.2210084767980423,
      "grad_norm": 3.25,
      "learning_rate": 0.0004504849523367656,
      "loss": 5.7566,
      "mean_token_accuracy": 0.16831666976213455,
      "num_tokens": 51496878.0,
      "step": 25415
    },
    {
      "entropy": 6.205249881744384,
      "epoch": 2.221445425150747,
      "grad_norm": 3.25,
      "learning_rate": 0.0004504654480798046,
      "loss": 5.8716,
      "mean_token_accuracy": 0.15232149958610536,
      "num_tokens": 51507092.0,
      "step": 25420
    },
    {
      "entropy": 6.187704610824585,
      "epoch": 2.221882373503452,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004504459404573243,
      "loss": 5.7958,
      "mean_token_accuracy": 0.16288021802902222,
      "num_tokens": 51517521.0,
      "step": 25425
    },
    {
      "entropy": 6.140657806396485,
      "epoch": 2.2223193218561565,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004504264294696991,
      "loss": 5.8137,
      "mean_token_accuracy": 0.15665873140096664,
      "num_tokens": 51527902.0,
      "step": 25430
    },
    {
      "entropy": 6.149126482009888,
      "epoch": 2.2227562702088615,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004504069151173029,
      "loss": 5.7974,
      "mean_token_accuracy": 0.15883081257343293,
      "num_tokens": 51539062.0,
      "step": 25435
    },
    {
      "entropy": 6.168322944641114,
      "epoch": 2.223193218561566,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004503873974005102,
      "loss": 5.7162,
      "mean_token_accuracy": 0.171208193898201,
      "num_tokens": 51548938.0,
      "step": 25440
    },
    {
      "entropy": 6.104405307769776,
      "epoch": 2.223630166914271,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004503678763196953,
      "loss": 5.7775,
      "mean_token_accuracy": 0.16519924104213715,
      "num_tokens": 51559330.0,
      "step": 25445
    },
    {
      "entropy": 6.051292085647583,
      "epoch": 2.2240671152669753,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004503483518752327,
      "loss": 5.7994,
      "mean_token_accuracy": 0.16119832694530487,
      "num_tokens": 51569617.0,
      "step": 25450
    },
    {
      "entropy": 6.084255266189575,
      "epoch": 2.2245040636196802,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004503288240674968,
      "loss": 5.7581,
      "mean_token_accuracy": 0.16223194301128388,
      "num_tokens": 51579545.0,
      "step": 25455
    },
    {
      "entropy": 6.156206274032593,
      "epoch": 2.2249410119723847,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004503092928968622,
      "loss": 5.7602,
      "mean_token_accuracy": 0.16295454800128936,
      "num_tokens": 51589375.0,
      "step": 25460
    },
    {
      "entropy": 6.148745918273926,
      "epoch": 2.2253779603250896,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004502897583637035,
      "loss": 5.7757,
      "mean_token_accuracy": 0.1637924611568451,
      "num_tokens": 51600586.0,
      "step": 25465
    },
    {
      "entropy": 6.09916934967041,
      "epoch": 2.225814908677794,
      "grad_norm": 2.75,
      "learning_rate": 0.00045027022046839534,
      "loss": 5.7069,
      "mean_token_accuracy": 0.16737261414527893,
      "num_tokens": 51609568.0,
      "step": 25470
    },
    {
      "entropy": 6.109530878067017,
      "epoch": 2.226251857030499,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004502506792113126,
      "loss": 5.7374,
      "mean_token_accuracy": 0.16780151128768922,
      "num_tokens": 51619450.0,
      "step": 25475
    },
    {
      "entropy": 6.1216083526611325,
      "epoch": 2.226688805383204,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004502311345928299,
      "loss": 5.7443,
      "mean_token_accuracy": 0.16450161337852479,
      "num_tokens": 51630528.0,
      "step": 25480
    },
    {
      "entropy": 6.126740455627441,
      "epoch": 2.2271257537359084,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00045021158661332227,
      "loss": 5.7626,
      "mean_token_accuracy": 0.16685846894979478,
      "num_tokens": 51640204.0,
      "step": 25485
    },
    {
      "entropy": 6.161543893814087,
      "epoch": 2.2275627020886133,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004501920352731646,
      "loss": 5.8973,
      "mean_token_accuracy": 0.16177313923835754,
      "num_tokens": 51649531.0,
      "step": 25490
    },
    {
      "entropy": 6.183104228973389,
      "epoch": 2.2279996504413178,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004501724805727319,
      "loss": 5.8768,
      "mean_token_accuracy": 0.15489027500152588,
      "num_tokens": 51660943.0,
      "step": 25495
    },
    {
      "entropy": 6.1730705261230465,
      "epoch": 2.2284365987940227,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004501529225123992,
      "loss": 5.6929,
      "mean_token_accuracy": 0.16997389793395995,
      "num_tokens": 51670606.0,
      "step": 25500
    },
    {
      "entropy": 6.103829050064087,
      "epoch": 2.228873547146727,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045013336109254174,
      "loss": 5.7691,
      "mean_token_accuracy": 0.15993960201740265,
      "num_tokens": 51681634.0,
      "step": 25505
    },
    {
      "entropy": 6.157010173797607,
      "epoch": 2.229310495499432,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045011379631353457,
      "loss": 5.7811,
      "mean_token_accuracy": 0.16067007631063462,
      "num_tokens": 51691579.0,
      "step": 25510
    },
    {
      "entropy": 6.11133131980896,
      "epoch": 2.2297474438521365,
      "grad_norm": 2.96875,
      "learning_rate": 0.00045009422817575304,
      "loss": 5.681,
      "mean_token_accuracy": 0.17738389372825622,
      "num_tokens": 51700739.0,
      "step": 25515
    },
    {
      "entropy": 6.144833421707153,
      "epoch": 2.2301843922048414,
      "grad_norm": 2.125,
      "learning_rate": 0.0004500746566795725,
      "loss": 5.8071,
      "mean_token_accuracy": 0.16344086229801177,
      "num_tokens": 51711905.0,
      "step": 25520
    },
    {
      "entropy": 6.149466133117675,
      "epoch": 2.230621340557546,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00045005508182536824,
      "loss": 5.7133,
      "mean_token_accuracy": 0.16331608444452286,
      "num_tokens": 51722020.0,
      "step": 25525
    },
    {
      "entropy": 6.11451210975647,
      "epoch": 2.231058288910251,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004500355036135158,
      "loss": 5.821,
      "mean_token_accuracy": 0.16408296972513198,
      "num_tokens": 51731763.0,
      "step": 25530
    },
    {
      "entropy": 6.159789896011352,
      "epoch": 2.2314952372629557,
      "grad_norm": 2.390625,
      "learning_rate": 0.00045001592204439077,
      "loss": 5.8378,
      "mean_token_accuracy": 0.1528629332780838,
      "num_tokens": 51741425.0,
      "step": 25535
    },
    {
      "entropy": 6.197350120544433,
      "epoch": 2.23193218561566,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004499963371183686,
      "loss": 5.8296,
      "mean_token_accuracy": 0.15793461054563523,
      "num_tokens": 51751788.0,
      "step": 25540
    },
    {
      "entropy": 6.130553197860718,
      "epoch": 2.232369133968365,
      "grad_norm": 2.15625,
      "learning_rate": 0.00044997674883582494,
      "loss": 5.8291,
      "mean_token_accuracy": 0.15757920295000077,
      "num_tokens": 51763205.0,
      "step": 25545
    },
    {
      "entropy": 6.214443635940552,
      "epoch": 2.2328060823210696,
      "grad_norm": 2.0,
      "learning_rate": 0.0004499571571971356,
      "loss": 5.8045,
      "mean_token_accuracy": 0.16023496240377427,
      "num_tokens": 51772817.0,
      "step": 25550
    },
    {
      "entropy": 6.183890295028687,
      "epoch": 2.2332430306737745,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004499375622026763,
      "loss": 5.7845,
      "mean_token_accuracy": 0.16768895238637924,
      "num_tokens": 51781596.0,
      "step": 25555
    },
    {
      "entropy": 6.133799505233765,
      "epoch": 2.233679979026479,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004499179638528229,
      "loss": 5.7556,
      "mean_token_accuracy": 0.16338135302066803,
      "num_tokens": 51792452.0,
      "step": 25560
    },
    {
      "entropy": 6.155353832244873,
      "epoch": 2.234116927379184,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004498983621479513,
      "loss": 5.8036,
      "mean_token_accuracy": 0.1602183222770691,
      "num_tokens": 51802554.0,
      "step": 25565
    },
    {
      "entropy": 6.164522790908814,
      "epoch": 2.2345538757318884,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004498787570884374,
      "loss": 5.8098,
      "mean_token_accuracy": 0.1555229678750038,
      "num_tokens": 51812855.0,
      "step": 25570
    },
    {
      "entropy": 6.180279159545899,
      "epoch": 2.2349908240845933,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004498591486746573,
      "loss": 5.7938,
      "mean_token_accuracy": 0.16226059198379517,
      "num_tokens": 51822306.0,
      "step": 25575
    },
    {
      "entropy": 6.198747968673706,
      "epoch": 2.2354277724372977,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044983953690698724,
      "loss": 5.8062,
      "mean_token_accuracy": 0.16102661043405533,
      "num_tokens": 51832209.0,
      "step": 25580
    },
    {
      "entropy": 6.171001148223877,
      "epoch": 2.2358647207900026,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004498199217858032,
      "loss": 5.8217,
      "mean_token_accuracy": 0.15965196043252944,
      "num_tokens": 51841794.0,
      "step": 25585
    },
    {
      "entropy": 6.1531086444854735,
      "epoch": 2.2363016691427076,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00044980030331148144,
      "loss": 5.8073,
      "mean_token_accuracy": 0.15651230663061141,
      "num_tokens": 51851973.0,
      "step": 25590
    },
    {
      "entropy": 6.224698543548584,
      "epoch": 2.236738617495412,
      "grad_norm": 4.28125,
      "learning_rate": 0.00044978068148439826,
      "loss": 5.8064,
      "mean_token_accuracy": 0.1570983648300171,
      "num_tokens": 51862173.0,
      "step": 25595
    },
    {
      "entropy": 6.225810623168945,
      "epoch": 2.237175565848117,
      "grad_norm": 2.671875,
      "learning_rate": 0.00044976105630493,
      "loss": 5.8059,
      "mean_token_accuracy": 0.15908589512109755,
      "num_tokens": 51872960.0,
      "step": 25600
    },
    {
      "entropy": 6.116137456893921,
      "epoch": 2.2376125142008214,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004497414277734532,
      "loss": 5.7454,
      "mean_token_accuracy": 0.16497665792703628,
      "num_tokens": 51883911.0,
      "step": 25605
    },
    {
      "entropy": 6.134789896011353,
      "epoch": 2.2380494625535263,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00044972179589034407,
      "loss": 5.7712,
      "mean_token_accuracy": 0.16276127994060516,
      "num_tokens": 51893982.0,
      "step": 25610
    },
    {
      "entropy": 6.10467562675476,
      "epoch": 2.238486410906231,
      "grad_norm": 3.234375,
      "learning_rate": 0.00044970216065597947,
      "loss": 5.6968,
      "mean_token_accuracy": 0.1681345596909523,
      "num_tokens": 51903808.0,
      "step": 25615
    },
    {
      "entropy": 6.011141157150268,
      "epoch": 2.2389233592589357,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00044968252207073584,
      "loss": 5.6878,
      "mean_token_accuracy": 0.17324769794940947,
      "num_tokens": 51912918.0,
      "step": 25620
    },
    {
      "entropy": 6.169246053695678,
      "epoch": 2.23936030761164,
      "grad_norm": 2.5,
      "learning_rate": 0.00044966288013499,
      "loss": 5.8321,
      "mean_token_accuracy": 0.16441337317228316,
      "num_tokens": 51923894.0,
      "step": 25625
    },
    {
      "entropy": 6.2201837539672855,
      "epoch": 2.239797255964345,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004496432348491185,
      "loss": 5.8289,
      "mean_token_accuracy": 0.1614087387919426,
      "num_tokens": 51934569.0,
      "step": 25630
    },
    {
      "entropy": 6.150409078598022,
      "epoch": 2.2402342043170496,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004496235862134982,
      "loss": 5.7891,
      "mean_token_accuracy": 0.15825730860233306,
      "num_tokens": 51944678.0,
      "step": 25635
    },
    {
      "entropy": 6.198678255081177,
      "epoch": 2.2406711526697545,
      "grad_norm": 3.3125,
      "learning_rate": 0.00044960393422850616,
      "loss": 5.8699,
      "mean_token_accuracy": 0.1553519368171692,
      "num_tokens": 51954785.0,
      "step": 25640
    },
    {
      "entropy": 6.15028715133667,
      "epoch": 2.241108101022459,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004495842788945191,
      "loss": 5.7242,
      "mean_token_accuracy": 0.1635920748114586,
      "num_tokens": 51964427.0,
      "step": 25645
    },
    {
      "entropy": 6.112845325469971,
      "epoch": 2.241545049375164,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00044956462021191406,
      "loss": 5.8817,
      "mean_token_accuracy": 0.15734345018863677,
      "num_tokens": 51974367.0,
      "step": 25650
    },
    {
      "entropy": 6.139910793304443,
      "epoch": 2.2419819977278688,
      "grad_norm": 2.796875,
      "learning_rate": 0.00044954495818106814,
      "loss": 5.7869,
      "mean_token_accuracy": 0.16542850583791732,
      "num_tokens": 51984935.0,
      "step": 25655
    },
    {
      "entropy": 6.171747398376465,
      "epoch": 2.2424189460805732,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004495252928023585,
      "loss": 5.7849,
      "mean_token_accuracy": 0.1635998696088791,
      "num_tokens": 51995132.0,
      "step": 25660
    },
    {
      "entropy": 6.1566187858581545,
      "epoch": 2.242855894433278,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004495056240761623,
      "loss": 5.8312,
      "mean_token_accuracy": 0.16193651407957077,
      "num_tokens": 52005828.0,
      "step": 25665
    },
    {
      "entropy": 6.108979225158691,
      "epoch": 2.2432928427859826,
      "grad_norm": 4.3125,
      "learning_rate": 0.00044948595200285677,
      "loss": 5.6248,
      "mean_token_accuracy": 0.1718690350651741,
      "num_tokens": 52015381.0,
      "step": 25670
    },
    {
      "entropy": 6.137670850753784,
      "epoch": 2.2437297911386875,
      "grad_norm": 2.125,
      "learning_rate": 0.00044946627658281924,
      "loss": 5.8225,
      "mean_token_accuracy": 0.15946702808141708,
      "num_tokens": 52025699.0,
      "step": 25675
    },
    {
      "entropy": 6.160710668563842,
      "epoch": 2.244166739491392,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004494465978164271,
      "loss": 5.73,
      "mean_token_accuracy": 0.1660478115081787,
      "num_tokens": 52035903.0,
      "step": 25680
    },
    {
      "entropy": 6.142135143280029,
      "epoch": 2.244603687844097,
      "grad_norm": 4.75,
      "learning_rate": 0.0004494269157040579,
      "loss": 5.7272,
      "mean_token_accuracy": 0.170460444688797,
      "num_tokens": 52046039.0,
      "step": 25685
    },
    {
      "entropy": 6.16995906829834,
      "epoch": 2.2450406361968014,
      "grad_norm": 2.640625,
      "learning_rate": 0.00044940723024608893,
      "loss": 5.8157,
      "mean_token_accuracy": 0.1541767381131649,
      "num_tokens": 52057146.0,
      "step": 25690
    },
    {
      "entropy": 6.135122680664063,
      "epoch": 2.2454775845495063,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00044938754144289793,
      "loss": 5.7703,
      "mean_token_accuracy": 0.16444438546895981,
      "num_tokens": 52067355.0,
      "step": 25695
    },
    {
      "entropy": 6.135501718521118,
      "epoch": 2.2459145329022108,
      "grad_norm": 4.03125,
      "learning_rate": 0.00044936784929486257,
      "loss": 5.7494,
      "mean_token_accuracy": 0.1655629739165306,
      "num_tokens": 52077062.0,
      "step": 25700
    },
    {
      "entropy": 6.129872512817383,
      "epoch": 2.2463514812549157,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004493481538023604,
      "loss": 5.7487,
      "mean_token_accuracy": 0.16596405059099198,
      "num_tokens": 52087918.0,
      "step": 25705
    },
    {
      "entropy": 6.1420325756073,
      "epoch": 2.24678842960762,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00044932845496576936,
      "loss": 5.7728,
      "mean_token_accuracy": 0.1615591824054718,
      "num_tokens": 52097283.0,
      "step": 25710
    },
    {
      "entropy": 6.127867794036865,
      "epoch": 2.247225377960325,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004493087527854672,
      "loss": 5.8131,
      "mean_token_accuracy": 0.15918889045715331,
      "num_tokens": 52109037.0,
      "step": 25715
    },
    {
      "entropy": 6.163216495513916,
      "epoch": 2.24766232631303,
      "grad_norm": 3.375,
      "learning_rate": 0.0004492890472618318,
      "loss": 5.7972,
      "mean_token_accuracy": 0.16115583181381227,
      "num_tokens": 52119809.0,
      "step": 25720
    },
    {
      "entropy": 6.1293816566467285,
      "epoch": 2.2480992746657344,
      "grad_norm": 2.546875,
      "learning_rate": 0.00044926933839524116,
      "loss": 5.6994,
      "mean_token_accuracy": 0.1647676259279251,
      "num_tokens": 52129641.0,
      "step": 25725
    },
    {
      "entropy": 6.042558670043945,
      "epoch": 2.2485362230184394,
      "grad_norm": 2.484375,
      "learning_rate": 0.00044924962618607324,
      "loss": 5.6805,
      "mean_token_accuracy": 0.1736478865146637,
      "num_tokens": 52139701.0,
      "step": 25730
    },
    {
      "entropy": 6.071463108062744,
      "epoch": 2.248973171371144,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004492299106347062,
      "loss": 5.6984,
      "mean_token_accuracy": 0.17167677432298661,
      "num_tokens": 52148402.0,
      "step": 25735
    },
    {
      "entropy": 6.072716808319091,
      "epoch": 2.2494101197238487,
      "grad_norm": 6.3125,
      "learning_rate": 0.0004492101917415183,
      "loss": 5.7741,
      "mean_token_accuracy": 0.15646433234214782,
      "num_tokens": 52159238.0,
      "step": 25740
    },
    {
      "entropy": 6.1715480327606205,
      "epoch": 2.249847068076553,
      "grad_norm": 2.625,
      "learning_rate": 0.00044919046950688744,
      "loss": 5.721,
      "mean_token_accuracy": 0.17045436352491378,
      "num_tokens": 52169172.0,
      "step": 25745
    },
    {
      "entropy": 6.118143463134766,
      "epoch": 2.250284016429258,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044917074393119224,
      "loss": 5.7355,
      "mean_token_accuracy": 0.16414201483130456,
      "num_tokens": 52179432.0,
      "step": 25750
    },
    {
      "entropy": 6.108035612106323,
      "epoch": 2.2507209647819626,
      "grad_norm": 2.890625,
      "learning_rate": 0.00044915101501481086,
      "loss": 5.7883,
      "mean_token_accuracy": 0.16209824234247208,
      "num_tokens": 52190939.0,
      "step": 25755
    },
    {
      "entropy": 6.052841520309448,
      "epoch": 2.2511579131346675,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00044913128275812184,
      "loss": 5.7464,
      "mean_token_accuracy": 0.16309450268745423,
      "num_tokens": 52201644.0,
      "step": 25760
    },
    {
      "entropy": 6.151989412307739,
      "epoch": 2.251594861487372,
      "grad_norm": 1.875,
      "learning_rate": 0.0004491115471615035,
      "loss": 5.7436,
      "mean_token_accuracy": 0.16480934619903564,
      "num_tokens": 52211353.0,
      "step": 25765
    },
    {
      "entropy": 6.142731094360352,
      "epoch": 2.252031809840077,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004490918082253344,
      "loss": 5.8134,
      "mean_token_accuracy": 0.15809272676706315,
      "num_tokens": 52221717.0,
      "step": 25770
    },
    {
      "entropy": 6.103111362457275,
      "epoch": 2.252468758192782,
      "grad_norm": 2.703125,
      "learning_rate": 0.00044907206594999325,
      "loss": 5.8221,
      "mean_token_accuracy": 0.1621400907635689,
      "num_tokens": 52232419.0,
      "step": 25775
    },
    {
      "entropy": 6.140601539611817,
      "epoch": 2.2529057065454863,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044905232033585867,
      "loss": 5.8107,
      "mean_token_accuracy": 0.16146874874830247,
      "num_tokens": 52242386.0,
      "step": 25780
    },
    {
      "entropy": 6.174315452575684,
      "epoch": 2.253342654898191,
      "grad_norm": 2.5,
      "learning_rate": 0.0004490325713833094,
      "loss": 5.7789,
      "mean_token_accuracy": 0.16056858897209167,
      "num_tokens": 52252569.0,
      "step": 25785
    },
    {
      "entropy": 6.044464063644409,
      "epoch": 2.2537796032508957,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004490128190927242,
      "loss": 5.653,
      "mean_token_accuracy": 0.17724292129278182,
      "num_tokens": 52263101.0,
      "step": 25790
    },
    {
      "entropy": 6.111260080337525,
      "epoch": 2.2542165516036006,
      "grad_norm": 1.9375,
      "learning_rate": 0.00044899306346448195,
      "loss": 5.748,
      "mean_token_accuracy": 0.16421585083007811,
      "num_tokens": 52273250.0,
      "step": 25795
    },
    {
      "entropy": 6.098425817489624,
      "epoch": 2.254653499956305,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004489733044989616,
      "loss": 5.8488,
      "mean_token_accuracy": 0.1571675047278404,
      "num_tokens": 52284351.0,
      "step": 25800
    },
    {
      "entropy": 6.080093240737915,
      "epoch": 2.25509044830901,
      "grad_norm": 2.203125,
      "learning_rate": 0.00044895354219654204,
      "loss": 5.7479,
      "mean_token_accuracy": 0.16819257736206056,
      "num_tokens": 52294642.0,
      "step": 25805
    },
    {
      "entropy": 6.077001667022705,
      "epoch": 2.2555273966617144,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00044893377655760245,
      "loss": 5.6715,
      "mean_token_accuracy": 0.17005294263362886,
      "num_tokens": 52303793.0,
      "step": 25810
    },
    {
      "entropy": 6.082685089111328,
      "epoch": 2.2559643450144193,
      "grad_norm": 2.34375,
      "learning_rate": 0.00044891400758252187,
      "loss": 5.7753,
      "mean_token_accuracy": 0.15689354091882707,
      "num_tokens": 52313541.0,
      "step": 25815
    },
    {
      "entropy": 6.136448574066162,
      "epoch": 2.256401293367124,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004488942352716795,
      "loss": 5.7828,
      "mean_token_accuracy": 0.1582068234682083,
      "num_tokens": 52325428.0,
      "step": 25820
    },
    {
      "entropy": 6.103855323791504,
      "epoch": 2.2568382417198287,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004488744596254545,
      "loss": 5.7814,
      "mean_token_accuracy": 0.16288510859012603,
      "num_tokens": 52335837.0,
      "step": 25825
    },
    {
      "entropy": 6.089674091339111,
      "epoch": 2.2572751900725336,
      "grad_norm": 2.5625,
      "learning_rate": 0.00044885468064422636,
      "loss": 5.7226,
      "mean_token_accuracy": 0.1641608238220215,
      "num_tokens": 52346925.0,
      "step": 25830
    },
    {
      "entropy": 6.110829877853393,
      "epoch": 2.257712138425238,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004488348983283743,
      "loss": 5.8665,
      "mean_token_accuracy": 0.15874343663454055,
      "num_tokens": 52357244.0,
      "step": 25835
    },
    {
      "entropy": 6.093083143234253,
      "epoch": 2.258149086777943,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004488151126782779,
      "loss": 5.73,
      "mean_token_accuracy": 0.16587564051151277,
      "num_tokens": 52367657.0,
      "step": 25840
    },
    {
      "entropy": 6.189214849472046,
      "epoch": 2.2585860351306475,
      "grad_norm": 3.171875,
      "learning_rate": 0.00044879532369431635,
      "loss": 5.7936,
      "mean_token_accuracy": 0.15854572951793672,
      "num_tokens": 52377298.0,
      "step": 25845
    },
    {
      "entropy": 6.178944492340088,
      "epoch": 2.2590229834833524,
      "grad_norm": 2.984375,
      "learning_rate": 0.00044877553137686953,
      "loss": 5.8347,
      "mean_token_accuracy": 0.1640269011259079,
      "num_tokens": 52387829.0,
      "step": 25850
    },
    {
      "entropy": 6.084576988220215,
      "epoch": 2.259459931836057,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044875573572631706,
      "loss": 5.766,
      "mean_token_accuracy": 0.16736888289451599,
      "num_tokens": 52398214.0,
      "step": 25855
    },
    {
      "entropy": 6.07977933883667,
      "epoch": 2.259896880188762,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004487359367430384,
      "loss": 5.7232,
      "mean_token_accuracy": 0.16696680337190628,
      "num_tokens": 52407619.0,
      "step": 25860
    },
    {
      "entropy": 6.161075830459595,
      "epoch": 2.2603338285414663,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004487161344274135,
      "loss": 5.7853,
      "mean_token_accuracy": 0.16023899093270302,
      "num_tokens": 52417465.0,
      "step": 25865
    },
    {
      "entropy": 6.081514453887939,
      "epoch": 2.260770776894171,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004486963287798221,
      "loss": 5.7262,
      "mean_token_accuracy": 0.17354840636253357,
      "num_tokens": 52427000.0,
      "step": 25870
    },
    {
      "entropy": 6.141672611236572,
      "epoch": 2.2612077252468756,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004486765198006441,
      "loss": 5.8074,
      "mean_token_accuracy": 0.1601485639810562,
      "num_tokens": 52437449.0,
      "step": 25875
    },
    {
      "entropy": 6.238683366775513,
      "epoch": 2.2616446735995805,
      "grad_norm": 4.59375,
      "learning_rate": 0.00044865670749025943,
      "loss": 5.7925,
      "mean_token_accuracy": 0.15623376294970512,
      "num_tokens": 52446946.0,
      "step": 25880
    },
    {
      "entropy": 6.105078792572021,
      "epoch": 2.2620816219522855,
      "grad_norm": 1.984375,
      "learning_rate": 0.00044863689184904815,
      "loss": 5.7485,
      "mean_token_accuracy": 0.163386207818985,
      "num_tokens": 52456993.0,
      "step": 25885
    },
    {
      "entropy": 6.1600080013275145,
      "epoch": 2.26251857030499,
      "grad_norm": 2.875,
      "learning_rate": 0.0004486170728773903,
      "loss": 5.8559,
      "mean_token_accuracy": 0.15127549916505814,
      "num_tokens": 52467999.0,
      "step": 25890
    },
    {
      "entropy": 6.2005260467529295,
      "epoch": 2.262955518657695,
      "grad_norm": 1.796875,
      "learning_rate": 0.000448597250575666,
      "loss": 5.836,
      "mean_token_accuracy": 0.15507406517863273,
      "num_tokens": 52479556.0,
      "step": 25895
    },
    {
      "entropy": 6.15119366645813,
      "epoch": 2.2633924670103993,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004485774249442555,
      "loss": 5.7883,
      "mean_token_accuracy": 0.15990700125694274,
      "num_tokens": 52489160.0,
      "step": 25900
    },
    {
      "entropy": 6.076725864410401,
      "epoch": 2.2638294153631042,
      "grad_norm": 2.578125,
      "learning_rate": 0.000448557595983539,
      "loss": 5.7098,
      "mean_token_accuracy": 0.16386072039604188,
      "num_tokens": 52500102.0,
      "step": 25905
    },
    {
      "entropy": 6.087321805953979,
      "epoch": 2.2642663637158087,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00044853776369389685,
      "loss": 5.7475,
      "mean_token_accuracy": 0.16724371165037155,
      "num_tokens": 52509615.0,
      "step": 25910
    },
    {
      "entropy": 6.13729419708252,
      "epoch": 2.2647033120685136,
      "grad_norm": 3.875,
      "learning_rate": 0.0004485179280757095,
      "loss": 5.7539,
      "mean_token_accuracy": 0.1677501305937767,
      "num_tokens": 52520796.0,
      "step": 25915
    },
    {
      "entropy": 6.174475479125976,
      "epoch": 2.265140260421218,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00044849808912935735,
      "loss": 5.7326,
      "mean_token_accuracy": 0.16420014351606368,
      "num_tokens": 52530641.0,
      "step": 25920
    },
    {
      "entropy": 6.14914026260376,
      "epoch": 2.265577208773923,
      "grad_norm": 2.40625,
      "learning_rate": 0.000448478246855221,
      "loss": 5.7902,
      "mean_token_accuracy": 0.16253169625997543,
      "num_tokens": 52541626.0,
      "step": 25925
    },
    {
      "entropy": 6.150815677642822,
      "epoch": 2.2660141571266275,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004484584012536809,
      "loss": 5.8178,
      "mean_token_accuracy": 0.1623900607228279,
      "num_tokens": 52552301.0,
      "step": 25930
    },
    {
      "entropy": 6.0189801216125485,
      "epoch": 2.2664511054793324,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004484385523251177,
      "loss": 5.6788,
      "mean_token_accuracy": 0.18161962032318116,
      "num_tokens": 52562463.0,
      "step": 25935
    },
    {
      "entropy": 6.111992454528808,
      "epoch": 2.2668880538320373,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004484187000699123,
      "loss": 5.786,
      "mean_token_accuracy": 0.16175806820392608,
      "num_tokens": 52572782.0,
      "step": 25940
    },
    {
      "entropy": 6.14350209236145,
      "epoch": 2.2673250021847418,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044839884448844534,
      "loss": 5.7218,
      "mean_token_accuracy": 0.16989077925682067,
      "num_tokens": 52582293.0,
      "step": 25945
    },
    {
      "entropy": 6.107601881027222,
      "epoch": 2.2677619505374462,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004483789855810977,
      "loss": 5.7716,
      "mean_token_accuracy": 0.15509573817253114,
      "num_tokens": 52591950.0,
      "step": 25950
    },
    {
      "entropy": 6.116190814971924,
      "epoch": 2.268198898890151,
      "grad_norm": 1.75,
      "learning_rate": 0.00044835912334825024,
      "loss": 5.8063,
      "mean_token_accuracy": 0.1613449364900589,
      "num_tokens": 52603998.0,
      "step": 25955
    },
    {
      "entropy": 6.200264883041382,
      "epoch": 2.268635847242856,
      "grad_norm": 2.078125,
      "learning_rate": 0.00044833925779028393,
      "loss": 5.7978,
      "mean_token_accuracy": 0.15692956298589705,
      "num_tokens": 52614955.0,
      "step": 25960
    },
    {
      "entropy": 6.237612295150757,
      "epoch": 2.2690727955955605,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004483193889075799,
      "loss": 5.8092,
      "mean_token_accuracy": 0.16132107526063919,
      "num_tokens": 52625305.0,
      "step": 25965
    },
    {
      "entropy": 6.063508129119873,
      "epoch": 2.2695097439482654,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004482995167005191,
      "loss": 5.7185,
      "mean_token_accuracy": 0.16880927383899688,
      "num_tokens": 52635310.0,
      "step": 25970
    },
    {
      "entropy": 6.080681610107422,
      "epoch": 2.26994669230097,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00044827964116948284,
      "loss": 5.7658,
      "mean_token_accuracy": 0.16130001097917557,
      "num_tokens": 52644946.0,
      "step": 25975
    },
    {
      "entropy": 6.154069280624389,
      "epoch": 2.270383640653675,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004482597623148522,
      "loss": 5.7631,
      "mean_token_accuracy": 0.16104332506656646,
      "num_tokens": 52655056.0,
      "step": 25980
    },
    {
      "entropy": 6.126058530807495,
      "epoch": 2.2708205890063793,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004482398801370085,
      "loss": 5.7796,
      "mean_token_accuracy": 0.16075320541858673,
      "num_tokens": 52665423.0,
      "step": 25985
    },
    {
      "entropy": 6.147169399261474,
      "epoch": 2.271257537359084,
      "grad_norm": 1.6875,
      "learning_rate": 0.00044821999463633323,
      "loss": 5.8103,
      "mean_token_accuracy": 0.16205511689186097,
      "num_tokens": 52675300.0,
      "step": 25990
    },
    {
      "entropy": 6.1269676208496096,
      "epoch": 2.271694485711789,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00044820010581320764,
      "loss": 5.7611,
      "mean_token_accuracy": 0.1622126206755638,
      "num_tokens": 52685431.0,
      "step": 25995
    },
    {
      "entropy": 6.1901881217956545,
      "epoch": 2.2721314340644936,
      "grad_norm": 1.84375,
      "learning_rate": 0.00044818021366801325,
      "loss": 5.8524,
      "mean_token_accuracy": 0.15857587307691573,
      "num_tokens": 52695756.0,
      "step": 26000
    },
    {
      "entropy": 6.11140456199646,
      "epoch": 2.272568382417198,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004481603182011317,
      "loss": 5.7352,
      "mean_token_accuracy": 0.16163264513015746,
      "num_tokens": 52704889.0,
      "step": 26005
    },
    {
      "entropy": 6.075212526321411,
      "epoch": 2.273005330769903,
      "grad_norm": 1.890625,
      "learning_rate": 0.00044814041941294435,
      "loss": 5.6804,
      "mean_token_accuracy": 0.1729758709669113,
      "num_tokens": 52714227.0,
      "step": 26010
    },
    {
      "entropy": 6.10250153541565,
      "epoch": 2.273442279122608,
      "grad_norm": 2.28125,
      "learning_rate": 0.00044812051730383316,
      "loss": 5.7159,
      "mean_token_accuracy": 0.17349123805761338,
      "num_tokens": 52725150.0,
      "step": 26015
    },
    {
      "entropy": 6.154853677749633,
      "epoch": 2.2738792274753123,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004481006118741798,
      "loss": 5.7347,
      "mean_token_accuracy": 0.16198684126138688,
      "num_tokens": 52734825.0,
      "step": 26020
    },
    {
      "entropy": 6.110104084014893,
      "epoch": 2.2743161758280173,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004480807031243658,
      "loss": 5.7218,
      "mean_token_accuracy": 0.17108402401208878,
      "num_tokens": 52745185.0,
      "step": 26025
    },
    {
      "entropy": 6.1184814929962155,
      "epoch": 2.2747531241807217,
      "grad_norm": 2.125,
      "learning_rate": 0.00044806079105477335,
      "loss": 5.8031,
      "mean_token_accuracy": 0.1615259200334549,
      "num_tokens": 52755122.0,
      "step": 26030
    },
    {
      "entropy": 6.096545314788818,
      "epoch": 2.2751900725334266,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004480408756657842,
      "loss": 5.7651,
      "mean_token_accuracy": 0.16145117431879044,
      "num_tokens": 52765338.0,
      "step": 26035
    },
    {
      "entropy": 6.215092992782592,
      "epoch": 2.275627020886131,
      "grad_norm": 1.796875,
      "learning_rate": 0.00044802095695778047,
      "loss": 5.829,
      "mean_token_accuracy": 0.15670665651559829,
      "num_tokens": 52775045.0,
      "step": 26040
    },
    {
      "entropy": 6.123351335525513,
      "epoch": 2.276063969238836,
      "grad_norm": 2.0,
      "learning_rate": 0.00044800103493114403,
      "loss": 5.7695,
      "mean_token_accuracy": 0.16487954705953597,
      "num_tokens": 52785942.0,
      "step": 26045
    },
    {
      "entropy": 6.03672456741333,
      "epoch": 2.2765009175915405,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004479811095862571,
      "loss": 5.7295,
      "mean_token_accuracy": 0.17210182994604112,
      "num_tokens": 52794905.0,
      "step": 26050
    },
    {
      "entropy": 6.1460793018341064,
      "epoch": 2.2769378659442454,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004479611809235019,
      "loss": 5.6876,
      "mean_token_accuracy": 0.1721363067626953,
      "num_tokens": 52804091.0,
      "step": 26055
    },
    {
      "entropy": 6.181627941131592,
      "epoch": 2.27737481429695,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004479412489432605,
      "loss": 5.7998,
      "mean_token_accuracy": 0.1532753586769104,
      "num_tokens": 52814400.0,
      "step": 26060
    },
    {
      "entropy": 6.10326042175293,
      "epoch": 2.277811762649655,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00044792131364591544,
      "loss": 5.7911,
      "mean_token_accuracy": 0.1653194472193718,
      "num_tokens": 52824423.0,
      "step": 26065
    },
    {
      "entropy": 6.172331142425537,
      "epoch": 2.2782487110023597,
      "grad_norm": 1.484375,
      "learning_rate": 0.00044790137503184895,
      "loss": 5.7844,
      "mean_token_accuracy": 0.16741949915885926,
      "num_tokens": 52833852.0,
      "step": 26070
    },
    {
      "entropy": 6.162242317199707,
      "epoch": 2.278685659355064,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004478814331014435,
      "loss": 5.7887,
      "mean_token_accuracy": 0.16142166554927825,
      "num_tokens": 52844759.0,
      "step": 26075
    },
    {
      "entropy": 6.0833258628845215,
      "epoch": 2.279122607707769,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004478614878550815,
      "loss": 5.7484,
      "mean_token_accuracy": 0.16299889832735062,
      "num_tokens": 52854302.0,
      "step": 26080
    },
    {
      "entropy": 6.100016832351685,
      "epoch": 2.2795595560604736,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004478415392931457,
      "loss": 5.7637,
      "mean_token_accuracy": 0.1612307161092758,
      "num_tokens": 52864086.0,
      "step": 26085
    },
    {
      "entropy": 6.136908721923828,
      "epoch": 2.2799965044131785,
      "grad_norm": 2.234375,
      "learning_rate": 0.00044782158741601854,
      "loss": 5.6884,
      "mean_token_accuracy": 0.16437434554100036,
      "num_tokens": 52874894.0,
      "step": 26090
    },
    {
      "entropy": 6.049416637420654,
      "epoch": 2.280433452765883,
      "grad_norm": 2.078125,
      "learning_rate": 0.00044780163222408277,
      "loss": 5.7639,
      "mean_token_accuracy": 0.16193112581968308,
      "num_tokens": 52885418.0,
      "step": 26095
    },
    {
      "entropy": 6.138898468017578,
      "epoch": 2.280870401118588,
      "grad_norm": 1.890625,
      "learning_rate": 0.00044778167371772117,
      "loss": 5.8089,
      "mean_token_accuracy": 0.16407772302627563,
      "num_tokens": 52895515.0,
      "step": 26100
    },
    {
      "entropy": 6.118705129623413,
      "epoch": 2.2813073494712923,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004477617118973165,
      "loss": 5.7672,
      "mean_token_accuracy": 0.17028350681066512,
      "num_tokens": 52905599.0,
      "step": 26105
    },
    {
      "entropy": 6.144879674911499,
      "epoch": 2.2817442978239972,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004477417467632517,
      "loss": 5.8486,
      "mean_token_accuracy": 0.15702340602874756,
      "num_tokens": 52916599.0,
      "step": 26110
    },
    {
      "entropy": 6.184028625488281,
      "epoch": 2.2821812461767017,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00044772177831590965,
      "loss": 5.777,
      "mean_token_accuracy": 0.16953660994768144,
      "num_tokens": 52928131.0,
      "step": 26115
    },
    {
      "entropy": 6.173533773422241,
      "epoch": 2.2826181945294066,
      "grad_norm": 1.5,
      "learning_rate": 0.00044770180655567343,
      "loss": 5.8161,
      "mean_token_accuracy": 0.1549117907881737,
      "num_tokens": 52939305.0,
      "step": 26120
    },
    {
      "entropy": 6.082504034042358,
      "epoch": 2.2830551428821115,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004476818314829261,
      "loss": 5.6667,
      "mean_token_accuracy": 0.16927100718021393,
      "num_tokens": 52948814.0,
      "step": 26125
    },
    {
      "entropy": 6.0803240776062015,
      "epoch": 2.283492091234816,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004476618530980506,
      "loss": 5.8254,
      "mean_token_accuracy": 0.15931920260190963,
      "num_tokens": 52958964.0,
      "step": 26130
    },
    {
      "entropy": 6.12164192199707,
      "epoch": 2.283929039587521,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004476418714014305,
      "loss": 5.8578,
      "mean_token_accuracy": 0.156445974111557,
      "num_tokens": 52969837.0,
      "step": 26135
    },
    {
      "entropy": 6.129214715957642,
      "epoch": 2.2843659879402254,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044762188639344875,
      "loss": 5.7228,
      "mean_token_accuracy": 0.1671435609459877,
      "num_tokens": 52979184.0,
      "step": 26140
    },
    {
      "entropy": 6.079606246948242,
      "epoch": 2.2848029362929303,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044760189807448874,
      "loss": 5.8633,
      "mean_token_accuracy": 0.15994543433189393,
      "num_tokens": 52989867.0,
      "step": 26145
    },
    {
      "entropy": 6.142032527923584,
      "epoch": 2.2852398846456348,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004475819064449338,
      "loss": 5.7833,
      "mean_token_accuracy": 0.1625167429447174,
      "num_tokens": 53000177.0,
      "step": 26150
    },
    {
      "entropy": 6.06935224533081,
      "epoch": 2.2856768329983397,
      "grad_norm": 2.46875,
      "learning_rate": 0.00044756191150516764,
      "loss": 5.696,
      "mean_token_accuracy": 0.1645074814558029,
      "num_tokens": 53009372.0,
      "step": 26155
    },
    {
      "entropy": 6.141605186462402,
      "epoch": 2.286113781351044,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004475419132555735,
      "loss": 5.7736,
      "mean_token_accuracy": 0.16082814931869507,
      "num_tokens": 53020405.0,
      "step": 26160
    },
    {
      "entropy": 6.1140772819519045,
      "epoch": 2.286550729703749,
      "grad_norm": 2.140625,
      "learning_rate": 0.00044752191169653504,
      "loss": 5.7936,
      "mean_token_accuracy": 0.1630957007408142,
      "num_tokens": 53030713.0,
      "step": 26165
    },
    {
      "entropy": 6.123281240463257,
      "epoch": 2.2869876780564535,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00044750190682843596,
      "loss": 5.773,
      "mean_token_accuracy": 0.16676330864429473,
      "num_tokens": 53040868.0,
      "step": 26170
    },
    {
      "entropy": 6.072976684570312,
      "epoch": 2.2874246264091584,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004474818986516599,
      "loss": 5.624,
      "mean_token_accuracy": 0.16805089861154557,
      "num_tokens": 53050349.0,
      "step": 26175
    },
    {
      "entropy": 6.0813840389251705,
      "epoch": 2.2878615747618634,
      "grad_norm": 2.4375,
      "learning_rate": 0.00044746188716659064,
      "loss": 5.7254,
      "mean_token_accuracy": 0.16700114905834199,
      "num_tokens": 53060446.0,
      "step": 26180
    },
    {
      "entropy": 6.138211965560913,
      "epoch": 2.288298523114568,
      "grad_norm": 2.9375,
      "learning_rate": 0.00044744187237361207,
      "loss": 5.799,
      "mean_token_accuracy": 0.16022028475999833,
      "num_tokens": 53071099.0,
      "step": 26185
    },
    {
      "entropy": 6.100603914260864,
      "epoch": 2.2887354714672727,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044742185427310797,
      "loss": 5.7187,
      "mean_token_accuracy": 0.16798291504383087,
      "num_tokens": 53080712.0,
      "step": 26190
    },
    {
      "entropy": 6.1378261089324955,
      "epoch": 2.289172419819977,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004474018328654623,
      "loss": 5.8302,
      "mean_token_accuracy": 0.15751507878303528,
      "num_tokens": 53091160.0,
      "step": 26195
    },
    {
      "entropy": 6.020037651062012,
      "epoch": 2.289609368172682,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004473818081510592,
      "loss": 5.7707,
      "mean_token_accuracy": 0.16245096921920776,
      "num_tokens": 53100767.0,
      "step": 26200
    },
    {
      "entropy": 6.15551643371582,
      "epoch": 2.2900463165253866,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00044736178013028276,
      "loss": 5.7836,
      "mean_token_accuracy": 0.1684655874967575,
      "num_tokens": 53111641.0,
      "step": 26205
    },
    {
      "entropy": 6.136761236190796,
      "epoch": 2.2904832648780915,
      "grad_norm": 2.03125,
      "learning_rate": 0.000447341748803517,
      "loss": 5.7501,
      "mean_token_accuracy": 0.16575844138860701,
      "num_tokens": 53121982.0,
      "step": 26210
    },
    {
      "entropy": 6.128815460205078,
      "epoch": 2.290920213230796,
      "grad_norm": 2.28125,
      "learning_rate": 0.00044732171417114614,
      "loss": 5.819,
      "mean_token_accuracy": 0.16164244413375856,
      "num_tokens": 53132649.0,
      "step": 26215
    },
    {
      "entropy": 6.133703327178955,
      "epoch": 2.291357161583501,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004473016762335546,
      "loss": 5.812,
      "mean_token_accuracy": 0.16514074951410293,
      "num_tokens": 53142249.0,
      "step": 26220
    },
    {
      "entropy": 6.181513929367066,
      "epoch": 2.2917941099362054,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004472816349911266,
      "loss": 5.7399,
      "mean_token_accuracy": 0.17201398462057113,
      "num_tokens": 53152375.0,
      "step": 26225
    },
    {
      "entropy": 6.2043054580688475,
      "epoch": 2.2922310582889103,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004472615904442465,
      "loss": 5.8305,
      "mean_token_accuracy": 0.15774222165346147,
      "num_tokens": 53162699.0,
      "step": 26230
    },
    {
      "entropy": 6.104915189743042,
      "epoch": 2.292668006641615,
      "grad_norm": 2.125,
      "learning_rate": 0.00044724154259329893,
      "loss": 5.7977,
      "mean_token_accuracy": 0.15652577430009842,
      "num_tokens": 53172890.0,
      "step": 26235
    },
    {
      "entropy": 6.183963441848755,
      "epoch": 2.2931049549943197,
      "grad_norm": 34.0,
      "learning_rate": 0.0004472214914386682,
      "loss": 5.8214,
      "mean_token_accuracy": 0.16386860609054565,
      "num_tokens": 53183735.0,
      "step": 26240
    },
    {
      "entropy": 6.144765377044678,
      "epoch": 2.2935419033470246,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004472014369807391,
      "loss": 5.7731,
      "mean_token_accuracy": 0.15972821563482284,
      "num_tokens": 53193046.0,
      "step": 26245
    },
    {
      "entropy": 6.0334409236907955,
      "epoch": 2.293978851699729,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004471813792198962,
      "loss": 5.7374,
      "mean_token_accuracy": 0.1734500050544739,
      "num_tokens": 53203367.0,
      "step": 26250
    },
    {
      "entropy": 6.1560142993927,
      "epoch": 2.294415800052434,
      "grad_norm": 2.71875,
      "learning_rate": 0.00044716131815652425,
      "loss": 5.8949,
      "mean_token_accuracy": 0.15677228271961213,
      "num_tokens": 53214207.0,
      "step": 26255
    },
    {
      "entropy": 6.16264910697937,
      "epoch": 2.2948527484051384,
      "grad_norm": 2.875,
      "learning_rate": 0.000447141253791008,
      "loss": 5.7615,
      "mean_token_accuracy": 0.16300004720687866,
      "num_tokens": 53225061.0,
      "step": 26260
    },
    {
      "entropy": 6.18325662612915,
      "epoch": 2.2952896967578433,
      "grad_norm": 1.515625,
      "learning_rate": 0.00044712118612373223,
      "loss": 5.819,
      "mean_token_accuracy": 0.16108712255954744,
      "num_tokens": 53235463.0,
      "step": 26265
    },
    {
      "entropy": 6.1525359630584715,
      "epoch": 2.295726645110548,
      "grad_norm": 1.46875,
      "learning_rate": 0.000447101115155082,
      "loss": 5.8193,
      "mean_token_accuracy": 0.15847906321287156,
      "num_tokens": 53245923.0,
      "step": 26270
    },
    {
      "entropy": 6.173504972457886,
      "epoch": 2.2961635934632527,
      "grad_norm": 2.046875,
      "learning_rate": 0.00044708104088544216,
      "loss": 5.7535,
      "mean_token_accuracy": 0.15893026441335678,
      "num_tokens": 53255756.0,
      "step": 26275
    },
    {
      "entropy": 6.096993494033813,
      "epoch": 2.296600541815957,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004470609633151978,
      "loss": 5.7874,
      "mean_token_accuracy": 0.16051169782876967,
      "num_tokens": 53266176.0,
      "step": 26280
    },
    {
      "entropy": 6.085315179824829,
      "epoch": 2.297037490168662,
      "grad_norm": 1.875,
      "learning_rate": 0.00044704088244473404,
      "loss": 5.7721,
      "mean_token_accuracy": 0.164838907122612,
      "num_tokens": 53276703.0,
      "step": 26285
    },
    {
      "entropy": 6.154263544082641,
      "epoch": 2.297474438521367,
      "grad_norm": 1.8828125,
      "learning_rate": 0.000447020798274436,
      "loss": 5.8646,
      "mean_token_accuracy": 0.15338767170906067,
      "num_tokens": 53286511.0,
      "step": 26290
    },
    {
      "entropy": 6.169029426574707,
      "epoch": 2.2979113868740715,
      "grad_norm": 2.84375,
      "learning_rate": 0.00044700071080468884,
      "loss": 5.7908,
      "mean_token_accuracy": 0.16125189065933226,
      "num_tokens": 53297261.0,
      "step": 26295
    },
    {
      "entropy": 6.152741384506226,
      "epoch": 2.298348335226776,
      "grad_norm": 1.765625,
      "learning_rate": 0.000446980620035878,
      "loss": 5.7725,
      "mean_token_accuracy": 0.16200512051582336,
      "num_tokens": 53307054.0,
      "step": 26300
    },
    {
      "entropy": 6.193601751327515,
      "epoch": 2.298785283579481,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044696052596838873,
      "loss": 5.7845,
      "mean_token_accuracy": 0.1618684485554695,
      "num_tokens": 53318022.0,
      "step": 26305
    },
    {
      "entropy": 6.136290836334228,
      "epoch": 2.299222231932186,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004469404286026064,
      "loss": 5.7726,
      "mean_token_accuracy": 0.1682145744562149,
      "num_tokens": 53328462.0,
      "step": 26310
    },
    {
      "entropy": 6.072665596008301,
      "epoch": 2.2996591802848902,
      "grad_norm": 1.359375,
      "learning_rate": 0.00044692032793891665,
      "loss": 5.7405,
      "mean_token_accuracy": 0.1660195469856262,
      "num_tokens": 53337837.0,
      "step": 26315
    },
    {
      "entropy": 6.064973306655884,
      "epoch": 2.300096128637595,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044690022397770484,
      "loss": 5.7209,
      "mean_token_accuracy": 0.17015330493450165,
      "num_tokens": 53348026.0,
      "step": 26320
    },
    {
      "entropy": 6.155670690536499,
      "epoch": 2.3005330769902996,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004468801167193567,
      "loss": 5.7892,
      "mean_token_accuracy": 0.16339804828166962,
      "num_tokens": 53358881.0,
      "step": 26325
    },
    {
      "entropy": 6.080107831954956,
      "epoch": 2.3009700253430045,
      "grad_norm": 2.515625,
      "learning_rate": 0.00044686000616425795,
      "loss": 5.7634,
      "mean_token_accuracy": 0.16441864222288133,
      "num_tokens": 53369732.0,
      "step": 26330
    },
    {
      "entropy": 6.170788049697876,
      "epoch": 2.301406973695709,
      "grad_norm": 3.6875,
      "learning_rate": 0.00044683989231279406,
      "loss": 5.7935,
      "mean_token_accuracy": 0.16043317317962646,
      "num_tokens": 53379607.0,
      "step": 26335
    },
    {
      "entropy": 6.253824186325073,
      "epoch": 2.301843922048414,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044681977516535106,
      "loss": 5.8334,
      "mean_token_accuracy": 0.15278440117835998,
      "num_tokens": 53390630.0,
      "step": 26340
    },
    {
      "entropy": 6.026923608779907,
      "epoch": 2.3022808704011184,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004467996547223148,
      "loss": 5.6803,
      "mean_token_accuracy": 0.17710861265659333,
      "num_tokens": 53400830.0,
      "step": 26345
    },
    {
      "entropy": 6.049797058105469,
      "epoch": 2.3027178187538233,
      "grad_norm": 2.34375,
      "learning_rate": 0.00044677953098407107,
      "loss": 5.7603,
      "mean_token_accuracy": 0.16045739501714706,
      "num_tokens": 53411150.0,
      "step": 26350
    },
    {
      "entropy": 6.078828811645508,
      "epoch": 2.303154767106528,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004467594039510059,
      "loss": 5.7547,
      "mean_token_accuracy": 0.16814842373132705,
      "num_tokens": 53421072.0,
      "step": 26355
    },
    {
      "entropy": 6.14873514175415,
      "epoch": 2.3035917154592327,
      "grad_norm": 2.6875,
      "learning_rate": 0.00044673927362350545,
      "loss": 5.7149,
      "mean_token_accuracy": 0.16814006567001344,
      "num_tokens": 53430659.0,
      "step": 26360
    },
    {
      "entropy": 6.10139741897583,
      "epoch": 2.3040286638119376,
      "grad_norm": 2.125,
      "learning_rate": 0.00044671914000195566,
      "loss": 5.6625,
      "mean_token_accuracy": 0.16845917105674743,
      "num_tokens": 53440883.0,
      "step": 26365
    },
    {
      "entropy": 6.060765218734741,
      "epoch": 2.304465612164642,
      "grad_norm": 2.078125,
      "learning_rate": 0.00044669900308674283,
      "loss": 5.7558,
      "mean_token_accuracy": 0.16412233263254167,
      "num_tokens": 53450491.0,
      "step": 26370
    },
    {
      "entropy": 6.106438636779785,
      "epoch": 2.304902560517347,
      "grad_norm": 3.546875,
      "learning_rate": 0.00044667886287825316,
      "loss": 5.7774,
      "mean_token_accuracy": 0.16450200825929642,
      "num_tokens": 53460864.0,
      "step": 26375
    },
    {
      "entropy": 6.0928874015808105,
      "epoch": 2.3053395088700515,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044665871937687296,
      "loss": 5.7537,
      "mean_token_accuracy": 0.16130056530237197,
      "num_tokens": 53470965.0,
      "step": 26380
    },
    {
      "entropy": 6.078852796554566,
      "epoch": 2.3057764572227564,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004466385725829885,
      "loss": 5.6784,
      "mean_token_accuracy": 0.1750039741396904,
      "num_tokens": 53480272.0,
      "step": 26385
    },
    {
      "entropy": 6.037267684936523,
      "epoch": 2.306213405575461,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004466184224969863,
      "loss": 5.7879,
      "mean_token_accuracy": 0.1667797550559044,
      "num_tokens": 53490285.0,
      "step": 26390
    },
    {
      "entropy": 6.098877811431885,
      "epoch": 2.3066503539281658,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004465982691192529,
      "loss": 5.6713,
      "mean_token_accuracy": 0.17178015112876893,
      "num_tokens": 53499712.0,
      "step": 26395
    },
    {
      "entropy": 6.110355520248413,
      "epoch": 2.3070873022808702,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004465781124501747,
      "loss": 5.7546,
      "mean_token_accuracy": 0.16631364971399307,
      "num_tokens": 53510144.0,
      "step": 26400
    },
    {
      "entropy": 6.122933721542358,
      "epoch": 2.307524250633575,
      "grad_norm": 1.578125,
      "learning_rate": 0.00044655795249013845,
      "loss": 5.7703,
      "mean_token_accuracy": 0.16576137840747834,
      "num_tokens": 53519936.0,
      "step": 26405
    },
    {
      "entropy": 6.119088888168335,
      "epoch": 2.3079611989862796,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004465377892395308,
      "loss": 5.7823,
      "mean_token_accuracy": 0.16403750628232955,
      "num_tokens": 53530447.0,
      "step": 26410
    },
    {
      "entropy": 6.168364858627319,
      "epoch": 2.3083981473389845,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004465176226987384,
      "loss": 5.8185,
      "mean_token_accuracy": 0.16038477569818496,
      "num_tokens": 53540929.0,
      "step": 26415
    },
    {
      "entropy": 6.083198738098145,
      "epoch": 2.3088350956916894,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004464974528681481,
      "loss": 5.7183,
      "mean_token_accuracy": 0.17195321172475814,
      "num_tokens": 53551575.0,
      "step": 26420
    },
    {
      "entropy": 6.102803325653076,
      "epoch": 2.309272044044394,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004464772797481469,
      "loss": 5.8543,
      "mean_token_accuracy": 0.15863524824380876,
      "num_tokens": 53563081.0,
      "step": 26425
    },
    {
      "entropy": 6.162367343902588,
      "epoch": 2.309708992397099,
      "grad_norm": 2.59375,
      "learning_rate": 0.00044645710333912153,
      "loss": 5.8018,
      "mean_token_accuracy": 0.163744455575943,
      "num_tokens": 53572303.0,
      "step": 26430
    },
    {
      "entropy": 6.140973949432373,
      "epoch": 2.3101459407498033,
      "grad_norm": 2.28125,
      "learning_rate": 0.00044643692364145914,
      "loss": 5.7292,
      "mean_token_accuracy": 0.16648439466953277,
      "num_tokens": 53582033.0,
      "step": 26435
    },
    {
      "entropy": 5.934871435165405,
      "epoch": 2.310582889102508,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044641674065554665,
      "loss": 5.5429,
      "mean_token_accuracy": 0.19486366212368011,
      "num_tokens": 53592140.0,
      "step": 26440
    },
    {
      "entropy": 6.080536365509033,
      "epoch": 2.3110198374552127,
      "grad_norm": 2.515625,
      "learning_rate": 0.00044639655438177134,
      "loss": 5.7476,
      "mean_token_accuracy": 0.16243237107992173,
      "num_tokens": 53602601.0,
      "step": 26445
    },
    {
      "entropy": 6.127170133590698,
      "epoch": 2.3114567858079176,
      "grad_norm": 1.890625,
      "learning_rate": 0.00044637636482052027,
      "loss": 5.7536,
      "mean_token_accuracy": 0.1613859787583351,
      "num_tokens": 53613292.0,
      "step": 26450
    },
    {
      "entropy": 6.153541898727417,
      "epoch": 2.311893734160622,
      "grad_norm": 1.375,
      "learning_rate": 0.0004463561719721807,
      "loss": 5.732,
      "mean_token_accuracy": 0.16798051148653032,
      "num_tokens": 53624493.0,
      "step": 26455
    },
    {
      "entropy": 6.109717178344726,
      "epoch": 2.312330682513327,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00044633597583713995,
      "loss": 5.7701,
      "mean_token_accuracy": 0.1667689248919487,
      "num_tokens": 53633739.0,
      "step": 26460
    },
    {
      "entropy": 6.086873245239258,
      "epoch": 2.3127676308660314,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004463157764157855,
      "loss": 5.7423,
      "mean_token_accuracy": 0.16692711263895035,
      "num_tokens": 53643025.0,
      "step": 26465
    },
    {
      "entropy": 6.102715730667114,
      "epoch": 2.3132045792187363,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004462955737085046,
      "loss": 5.7164,
      "mean_token_accuracy": 0.16712764799594879,
      "num_tokens": 53652905.0,
      "step": 26470
    },
    {
      "entropy": 6.061807298660279,
      "epoch": 2.3136415275714413,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004462753677156849,
      "loss": 5.7171,
      "mean_token_accuracy": 0.17114118188619615,
      "num_tokens": 53663189.0,
      "step": 26475
    },
    {
      "entropy": 6.149008512496948,
      "epoch": 2.3140784759241457,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004462551584377138,
      "loss": 5.8028,
      "mean_token_accuracy": 0.16421856731176376,
      "num_tokens": 53673068.0,
      "step": 26480
    },
    {
      "entropy": 6.125166940689087,
      "epoch": 2.3145154242768506,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004462349458749791,
      "loss": 5.8425,
      "mean_token_accuracy": 0.15504914820194243,
      "num_tokens": 53683929.0,
      "step": 26485
    },
    {
      "entropy": 6.1650763034820555,
      "epoch": 2.314952372629555,
      "grad_norm": 2.40625,
      "learning_rate": 0.00044621473002786846,
      "loss": 5.8143,
      "mean_token_accuracy": 0.16009239107370377,
      "num_tokens": 53694170.0,
      "step": 26490
    },
    {
      "entropy": 6.152600240707398,
      "epoch": 2.31538932098226,
      "grad_norm": 1.765625,
      "learning_rate": 0.00044619451089676946,
      "loss": 5.7393,
      "mean_token_accuracy": 0.17006041705608368,
      "num_tokens": 53704491.0,
      "step": 26495
    },
    {
      "entropy": 6.136135864257812,
      "epoch": 2.3158262693349645,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004461742884820701,
      "loss": 5.7968,
      "mean_token_accuracy": 0.16544368118047714,
      "num_tokens": 53714971.0,
      "step": 26500
    },
    {
      "entropy": 5.818529891967773,
      "epoch": 2.3162632176876694,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004461540627841582,
      "loss": 5.5241,
      "mean_token_accuracy": 0.20188093781471253,
      "num_tokens": 53723965.0,
      "step": 26505
    },
    {
      "entropy": 6.183325481414795,
      "epoch": 2.316700166040374,
      "grad_norm": 2.109375,
      "learning_rate": 0.00044613383380342163,
      "loss": 5.7601,
      "mean_token_accuracy": 0.1613331124186516,
      "num_tokens": 53734561.0,
      "step": 26510
    },
    {
      "entropy": 6.166078042984009,
      "epoch": 2.317137114393079,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004461136015402485,
      "loss": 5.7424,
      "mean_token_accuracy": 0.1634400174021721,
      "num_tokens": 53745115.0,
      "step": 26515
    },
    {
      "entropy": 6.0411115169525145,
      "epoch": 2.3175740627457833,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004460933659950269,
      "loss": 5.6867,
      "mean_token_accuracy": 0.16949982792139054,
      "num_tokens": 53754884.0,
      "step": 26520
    },
    {
      "entropy": 6.082873296737671,
      "epoch": 2.318011011098488,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004460731271681448,
      "loss": 5.7511,
      "mean_token_accuracy": 0.16326964646577835,
      "num_tokens": 53764772.0,
      "step": 26525
    },
    {
      "entropy": 6.162495422363281,
      "epoch": 2.318447959451193,
      "grad_norm": 1.59375,
      "learning_rate": 0.00044605288505999047,
      "loss": 5.8383,
      "mean_token_accuracy": 0.15441009253263474,
      "num_tokens": 53775391.0,
      "step": 26530
    },
    {
      "entropy": 6.179100131988525,
      "epoch": 2.3188849078038976,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044603263967095224,
      "loss": 5.7741,
      "mean_token_accuracy": 0.16370278745889663,
      "num_tokens": 53785698.0,
      "step": 26535
    },
    {
      "entropy": 6.159629154205322,
      "epoch": 2.3193218561566025,
      "grad_norm": 1.921875,
      "learning_rate": 0.00044601239100141833,
      "loss": 5.7456,
      "mean_token_accuracy": 0.1640851005911827,
      "num_tokens": 53795432.0,
      "step": 26540
    },
    {
      "entropy": 6.084775066375732,
      "epoch": 2.319758804509307,
      "grad_norm": 1.828125,
      "learning_rate": 0.00044599213905177724,
      "loss": 5.803,
      "mean_token_accuracy": 0.16202486008405687,
      "num_tokens": 53805295.0,
      "step": 26545
    },
    {
      "entropy": 6.121294689178467,
      "epoch": 2.320195752862012,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004459718838224171,
      "loss": 5.7596,
      "mean_token_accuracy": 0.16937350928783418,
      "num_tokens": 53814765.0,
      "step": 26550
    },
    {
      "entropy": 6.157626295089722,
      "epoch": 2.3206327012147163,
      "grad_norm": 1.84375,
      "learning_rate": 0.00044595162531372684,
      "loss": 5.7305,
      "mean_token_accuracy": 0.16751553863286972,
      "num_tokens": 53824154.0,
      "step": 26555
    },
    {
      "entropy": 6.117071342468262,
      "epoch": 2.3210696495674212,
      "grad_norm": 3.671875,
      "learning_rate": 0.00044593136352609475,
      "loss": 5.7972,
      "mean_token_accuracy": 0.16033383458852768,
      "num_tokens": 53834867.0,
      "step": 26560
    },
    {
      "entropy": 6.176008844375611,
      "epoch": 2.3215065979201257,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004459110984599096,
      "loss": 5.8391,
      "mean_token_accuracy": 0.15412599146366118,
      "num_tokens": 53844901.0,
      "step": 26565
    },
    {
      "entropy": 6.145382452011108,
      "epoch": 2.3219435462728306,
      "grad_norm": 2.515625,
      "learning_rate": 0.00044589083011555995,
      "loss": 5.8432,
      "mean_token_accuracy": 0.16144897639751435,
      "num_tokens": 53854897.0,
      "step": 26570
    },
    {
      "entropy": 6.156482839584351,
      "epoch": 2.322380494625535,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004458705584934347,
      "loss": 5.7898,
      "mean_token_accuracy": 0.1589842915534973,
      "num_tokens": 53865152.0,
      "step": 26575
    },
    {
      "entropy": 6.17560133934021,
      "epoch": 2.32281744297824,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00044585028359392256,
      "loss": 5.8107,
      "mean_token_accuracy": 0.15631030797958373,
      "num_tokens": 53875261.0,
      "step": 26580
    },
    {
      "entropy": 6.1509857177734375,
      "epoch": 2.323254391330945,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004458300054174125,
      "loss": 5.7467,
      "mean_token_accuracy": 0.16342332363128662,
      "num_tokens": 53885281.0,
      "step": 26585
    },
    {
      "entropy": 6.088864803314209,
      "epoch": 2.3236913396836494,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004458097239642933,
      "loss": 5.7375,
      "mean_token_accuracy": 0.16745404601097108,
      "num_tokens": 53895282.0,
      "step": 26590
    },
    {
      "entropy": 6.1490638732910154,
      "epoch": 2.3241282880363543,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004457894392349542,
      "loss": 5.794,
      "mean_token_accuracy": 0.16404519975185394,
      "num_tokens": 53905843.0,
      "step": 26595
    },
    {
      "entropy": 6.111294603347778,
      "epoch": 2.3245652363890588,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004457691512297841,
      "loss": 5.8327,
      "mean_token_accuracy": 0.15963328033685684,
      "num_tokens": 53915270.0,
      "step": 26600
    },
    {
      "entropy": 6.1665808200836185,
      "epoch": 2.3250021847417637,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004457488599491723,
      "loss": 5.8341,
      "mean_token_accuracy": 0.15892481803894043,
      "num_tokens": 53924548.0,
      "step": 26605
    },
    {
      "entropy": 6.1557094097137455,
      "epoch": 2.325439133094468,
      "grad_norm": 4.9375,
      "learning_rate": 0.00044572856539350784,
      "loss": 5.8417,
      "mean_token_accuracy": 0.16056790798902512,
      "num_tokens": 53935696.0,
      "step": 26610
    },
    {
      "entropy": 6.1574877262115475,
      "epoch": 2.325876081447173,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00044570826756318006,
      "loss": 5.7664,
      "mean_token_accuracy": 0.16626745015382766,
      "num_tokens": 53945499.0,
      "step": 26615
    },
    {
      "entropy": 6.113559198379517,
      "epoch": 2.3263130297998775,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004456879664585782,
      "loss": 5.7046,
      "mean_token_accuracy": 0.16801749616861344,
      "num_tokens": 53955478.0,
      "step": 26620
    },
    {
      "entropy": 6.128007936477661,
      "epoch": 2.3267499781525824,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004456676620800917,
      "loss": 5.7038,
      "mean_token_accuracy": 0.16503314822912216,
      "num_tokens": 53964908.0,
      "step": 26625
    },
    {
      "entropy": 6.081296825408936,
      "epoch": 2.327186926505287,
      "grad_norm": 2.25,
      "learning_rate": 0.0004456473544281101,
      "loss": 5.7076,
      "mean_token_accuracy": 0.1676509350538254,
      "num_tokens": 53974850.0,
      "step": 26630
    },
    {
      "entropy": 6.101322174072266,
      "epoch": 2.327623874857992,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004456270435030228,
      "loss": 5.7213,
      "mean_token_accuracy": 0.16594717651605606,
      "num_tokens": 53984458.0,
      "step": 26635
    },
    {
      "entropy": 6.0960320949554445,
      "epoch": 2.3280608232106967,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004456067293052193,
      "loss": 5.7495,
      "mean_token_accuracy": 0.16151269376277924,
      "num_tokens": 53994560.0,
      "step": 26640
    },
    {
      "entropy": 6.1216198921203615,
      "epoch": 2.328497771563401,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044558641183508945,
      "loss": 5.7325,
      "mean_token_accuracy": 0.16581966131925582,
      "num_tokens": 54003698.0,
      "step": 26645
    },
    {
      "entropy": 6.105107641220092,
      "epoch": 2.3289347199161057,
      "grad_norm": 1.6875,
      "learning_rate": 0.00044556609109302277,
      "loss": 5.6778,
      "mean_token_accuracy": 0.17886544167995452,
      "num_tokens": 54013405.0,
      "step": 26650
    },
    {
      "entropy": 6.161891126632691,
      "epoch": 2.3293716682688106,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004455457670794092,
      "loss": 5.7649,
      "mean_token_accuracy": 0.16663281470537186,
      "num_tokens": 54023140.0,
      "step": 26655
    },
    {
      "entropy": 6.098981046676636,
      "epoch": 2.3298086166215155,
      "grad_norm": 1.421875,
      "learning_rate": 0.00044552543979463826,
      "loss": 5.7305,
      "mean_token_accuracy": 0.16340069472789764,
      "num_tokens": 54032506.0,
      "step": 26660
    },
    {
      "entropy": 6.064685440063476,
      "epoch": 2.33024556497422,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004455051092391002,
      "loss": 5.6774,
      "mean_token_accuracy": 0.1698750913143158,
      "num_tokens": 54042113.0,
      "step": 26665
    },
    {
      "entropy": 6.099151372909546,
      "epoch": 2.330682513326925,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00044548477541318473,
      "loss": 5.7723,
      "mean_token_accuracy": 0.1646581619977951,
      "num_tokens": 54051139.0,
      "step": 26670
    },
    {
      "entropy": 6.116441440582276,
      "epoch": 2.3311194616796294,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004454644383172819,
      "loss": 5.745,
      "mean_token_accuracy": 0.1639465570449829,
      "num_tokens": 54060929.0,
      "step": 26675
    },
    {
      "entropy": 6.198477840423584,
      "epoch": 2.3315564100323343,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004454440979517819,
      "loss": 5.7925,
      "mean_token_accuracy": 0.1667448952794075,
      "num_tokens": 54071654.0,
      "step": 26680
    },
    {
      "entropy": 6.06194486618042,
      "epoch": 2.3319933583850387,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004454237543170747,
      "loss": 5.7174,
      "mean_token_accuracy": 0.16745162457227708,
      "num_tokens": 54081497.0,
      "step": 26685
    },
    {
      "entropy": 6.124838399887085,
      "epoch": 2.3324303067377437,
      "grad_norm": 3.796875,
      "learning_rate": 0.00044540340741355063,
      "loss": 5.7894,
      "mean_token_accuracy": 0.1639810934662819,
      "num_tokens": 54091899.0,
      "step": 26690
    },
    {
      "entropy": 6.099967622756958,
      "epoch": 2.332867255090448,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004453830572416,
      "loss": 5.7256,
      "mean_token_accuracy": 0.16493989378213883,
      "num_tokens": 54101734.0,
      "step": 26695
    },
    {
      "entropy": 6.143844842910767,
      "epoch": 2.333304203443153,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000445362703801613,
      "loss": 5.7774,
      "mean_token_accuracy": 0.1629488557577133,
      "num_tokens": 54112633.0,
      "step": 26700
    },
    {
      "entropy": 6.086287879943848,
      "epoch": 2.3337411517958575,
      "grad_norm": 2.671875,
      "learning_rate": 0.00044534234709398,
      "loss": 5.7347,
      "mean_token_accuracy": 0.16674728244543074,
      "num_tokens": 54122182.0,
      "step": 26705
    },
    {
      "entropy": 6.0737498760223385,
      "epoch": 2.3341781001485624,
      "grad_norm": 2.03125,
      "learning_rate": 0.00044532198711909153,
      "loss": 5.7673,
      "mean_token_accuracy": 0.1584361672401428,
      "num_tokens": 54131410.0,
      "step": 26710
    },
    {
      "entropy": 6.100105667114258,
      "epoch": 2.3346150485012673,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004453016238773382,
      "loss": 5.7208,
      "mean_token_accuracy": 0.1687342882156372,
      "num_tokens": 54141890.0,
      "step": 26715
    },
    {
      "entropy": 6.091963434219361,
      "epoch": 2.335051996853972,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004452812573691104,
      "loss": 5.7503,
      "mean_token_accuracy": 0.167882476747036,
      "num_tokens": 54151579.0,
      "step": 26720
    },
    {
      "entropy": 6.070718908309937,
      "epoch": 2.3354889452066767,
      "grad_norm": 2.21875,
      "learning_rate": 0.00044526088759479893,
      "loss": 5.7883,
      "mean_token_accuracy": 0.16802774071693422,
      "num_tokens": 54161824.0,
      "step": 26725
    },
    {
      "entropy": 6.086431694030762,
      "epoch": 2.335925893559381,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00044524051455479435,
      "loss": 5.6931,
      "mean_token_accuracy": 0.17380696386098862,
      "num_tokens": 54172110.0,
      "step": 26730
    },
    {
      "entropy": 6.133514881134033,
      "epoch": 2.336362841912086,
      "grad_norm": 1.875,
      "learning_rate": 0.00044522013824948756,
      "loss": 5.6961,
      "mean_token_accuracy": 0.16993848383426666,
      "num_tokens": 54181852.0,
      "step": 26735
    },
    {
      "entropy": 6.094956588745117,
      "epoch": 2.3367997902647906,
      "grad_norm": 2.078125,
      "learning_rate": 0.00044519975867926933,
      "loss": 5.7912,
      "mean_token_accuracy": 0.16119859516620635,
      "num_tokens": 54192703.0,
      "step": 26740
    },
    {
      "entropy": 6.079411411285401,
      "epoch": 2.3372367386174955,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004451793758445305,
      "loss": 5.6559,
      "mean_token_accuracy": 0.17096028178930284,
      "num_tokens": 54201613.0,
      "step": 26745
    },
    {
      "entropy": 6.104124212265015,
      "epoch": 2.3376736869702,
      "grad_norm": 1.921875,
      "learning_rate": 0.000445158989745662,
      "loss": 5.8279,
      "mean_token_accuracy": 0.1620839059352875,
      "num_tokens": 54211717.0,
      "step": 26750
    },
    {
      "entropy": 5.997409820556641,
      "epoch": 2.338110635322905,
      "grad_norm": 2.109375,
      "learning_rate": 0.00044513860038305505,
      "loss": 5.6634,
      "mean_token_accuracy": 0.1735346108675003,
      "num_tokens": 54221481.0,
      "step": 26755
    },
    {
      "entropy": 6.122946166992188,
      "epoch": 2.3385475836756093,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004451182077571005,
      "loss": 5.7442,
      "mean_token_accuracy": 0.16729093641042708,
      "num_tokens": 54231558.0,
      "step": 26760
    },
    {
      "entropy": 6.068603897094727,
      "epoch": 2.3389845320283142,
      "grad_norm": 2.3125,
      "learning_rate": 0.00044509781186818963,
      "loss": 5.7235,
      "mean_token_accuracy": 0.16603551357984542,
      "num_tokens": 54241150.0,
      "step": 26765
    },
    {
      "entropy": 6.104074001312256,
      "epoch": 2.339421480381019,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004450774127167136,
      "loss": 5.7441,
      "mean_token_accuracy": 0.16880864202976226,
      "num_tokens": 54251323.0,
      "step": 26770
    },
    {
      "entropy": 6.134990310668945,
      "epoch": 2.3398584287337236,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004450570103030637,
      "loss": 5.7671,
      "mean_token_accuracy": 0.16114602833986283,
      "num_tokens": 54260956.0,
      "step": 26775
    },
    {
      "entropy": 6.133560562133789,
      "epoch": 2.3402953770864285,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044503660462763116,
      "loss": 5.727,
      "mean_token_accuracy": 0.16838705390691758,
      "num_tokens": 54270382.0,
      "step": 26780
    },
    {
      "entropy": 6.109395170211792,
      "epoch": 2.340732325439133,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004450161956908075,
      "loss": 5.756,
      "mean_token_accuracy": 0.16471577435731888,
      "num_tokens": 54280380.0,
      "step": 26785
    },
    {
      "entropy": 6.104592800140381,
      "epoch": 2.341169273791838,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004449957834929841,
      "loss": 5.7405,
      "mean_token_accuracy": 0.16652151346206664,
      "num_tokens": 54290536.0,
      "step": 26790
    },
    {
      "entropy": 6.070264720916748,
      "epoch": 2.3416062221445424,
      "grad_norm": 1.75,
      "learning_rate": 0.00044497536803455245,
      "loss": 5.6657,
      "mean_token_accuracy": 0.16970127075910568,
      "num_tokens": 54299835.0,
      "step": 26795
    },
    {
      "entropy": 6.132720804214477,
      "epoch": 2.3420431704972473,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004449549493159043,
      "loss": 5.7415,
      "mean_token_accuracy": 0.17460979521274567,
      "num_tokens": 54309496.0,
      "step": 26800
    },
    {
      "entropy": 6.074022531509399,
      "epoch": 2.3424801188499518,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004449345273374311,
      "loss": 5.6888,
      "mean_token_accuracy": 0.16812217980623245,
      "num_tokens": 54318591.0,
      "step": 26805
    },
    {
      "entropy": 6.04746994972229,
      "epoch": 2.3429170672026567,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004449141020995246,
      "loss": 5.7313,
      "mean_token_accuracy": 0.16848801225423812,
      "num_tokens": 54327995.0,
      "step": 26810
    },
    {
      "entropy": 6.164483833312988,
      "epoch": 2.343354015555361,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004448936736025765,
      "loss": 5.7549,
      "mean_token_accuracy": 0.1607406996190548,
      "num_tokens": 54339122.0,
      "step": 26815
    },
    {
      "entropy": 6.148275089263916,
      "epoch": 2.343790963908066,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004448732418469789,
      "loss": 5.7748,
      "mean_token_accuracy": 0.16707601994276047,
      "num_tokens": 54349004.0,
      "step": 26820
    },
    {
      "entropy": 6.122946691513062,
      "epoch": 2.344227912260771,
      "grad_norm": 4.375,
      "learning_rate": 0.00044485280683312345,
      "loss": 5.7604,
      "mean_token_accuracy": 0.1700206145644188,
      "num_tokens": 54359859.0,
      "step": 26825
    },
    {
      "entropy": 6.160425901412964,
      "epoch": 2.3446648606134755,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004448323685614021,
      "loss": 5.7465,
      "mean_token_accuracy": 0.1615842968225479,
      "num_tokens": 54369386.0,
      "step": 26830
    },
    {
      "entropy": 6.092472934722901,
      "epoch": 2.3451018089661804,
      "grad_norm": 1.5,
      "learning_rate": 0.00044481192703220695,
      "loss": 5.6947,
      "mean_token_accuracy": 0.1711404636502266,
      "num_tokens": 54379250.0,
      "step": 26835
    },
    {
      "entropy": 6.102591323852539,
      "epoch": 2.345538757318885,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044479148224593014,
      "loss": 5.7563,
      "mean_token_accuracy": 0.16593141555786134,
      "num_tokens": 54388980.0,
      "step": 26840
    },
    {
      "entropy": 6.023606061935425,
      "epoch": 2.3459757056715898,
      "grad_norm": 3.46875,
      "learning_rate": 0.00044477103420296365,
      "loss": 5.6289,
      "mean_token_accuracy": 0.1816928580403328,
      "num_tokens": 54398849.0,
      "step": 26845
    },
    {
      "entropy": 6.1675934314727785,
      "epoch": 2.346412654024294,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004447505829036998,
      "loss": 5.809,
      "mean_token_accuracy": 0.1599281057715416,
      "num_tokens": 54409426.0,
      "step": 26850
    },
    {
      "entropy": 6.17660813331604,
      "epoch": 2.346849602376999,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004447301283485308,
      "loss": 5.8147,
      "mean_token_accuracy": 0.16743225157260894,
      "num_tokens": 54420212.0,
      "step": 26855
    },
    {
      "entropy": 6.133490991592407,
      "epoch": 2.3472865507297036,
      "grad_norm": 2.25,
      "learning_rate": 0.00044470967053784904,
      "loss": 5.864,
      "mean_token_accuracy": 0.16108891367912292,
      "num_tokens": 54431179.0,
      "step": 26860
    },
    {
      "entropy": 6.132482862472534,
      "epoch": 2.3477234990824085,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004446892094720468,
      "loss": 5.7591,
      "mean_token_accuracy": 0.17231343388557435,
      "num_tokens": 54442247.0,
      "step": 26865
    },
    {
      "entropy": 6.083278465270996,
      "epoch": 2.348160447435113,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00044466874515151667,
      "loss": 5.6864,
      "mean_token_accuracy": 0.17886364907026292,
      "num_tokens": 54451475.0,
      "step": 26870
    },
    {
      "entropy": 6.055406999588013,
      "epoch": 2.348597395787818,
      "grad_norm": 2.046875,
      "learning_rate": 0.000444648277576651,
      "loss": 5.8703,
      "mean_token_accuracy": 0.15414660498499871,
      "num_tokens": 54463259.0,
      "step": 26875
    },
    {
      "entropy": 6.131029462814331,
      "epoch": 2.349034344140523,
      "grad_norm": 2.0,
      "learning_rate": 0.00044462780674784264,
      "loss": 5.7219,
      "mean_token_accuracy": 0.16645563542842864,
      "num_tokens": 54473573.0,
      "step": 26880
    },
    {
      "entropy": 6.160770750045776,
      "epoch": 2.3494712924932273,
      "grad_norm": 1.390625,
      "learning_rate": 0.000444607332665484,
      "loss": 5.8493,
      "mean_token_accuracy": 0.1608395144343376,
      "num_tokens": 54482907.0,
      "step": 26885
    },
    {
      "entropy": 6.144451999664307,
      "epoch": 2.349908240845932,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004445868553299678,
      "loss": 5.7787,
      "mean_token_accuracy": 0.16771869063377381,
      "num_tokens": 54492993.0,
      "step": 26890
    },
    {
      "entropy": 6.109993267059326,
      "epoch": 2.3503451891986367,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004445663747416868,
      "loss": 5.8189,
      "mean_token_accuracy": 0.15772685706615447,
      "num_tokens": 54503826.0,
      "step": 26895
    },
    {
      "entropy": 6.1188023567199705,
      "epoch": 2.3507821375513416,
      "grad_norm": 3.25,
      "learning_rate": 0.00044454589090103406,
      "loss": 5.7918,
      "mean_token_accuracy": 0.15717767179012299,
      "num_tokens": 54513839.0,
      "step": 26900
    },
    {
      "entropy": 6.1463830947875975,
      "epoch": 2.351219085904046,
      "grad_norm": 1.890625,
      "learning_rate": 0.00044452540380840217,
      "loss": 5.6767,
      "mean_token_accuracy": 0.17050101310014726,
      "num_tokens": 54523742.0,
      "step": 26905
    },
    {
      "entropy": 6.223521947860718,
      "epoch": 2.351656034256751,
      "grad_norm": 1.6875,
      "learning_rate": 0.00044450491346418426,
      "loss": 5.8257,
      "mean_token_accuracy": 0.15917447209358215,
      "num_tokens": 54534017.0,
      "step": 26910
    },
    {
      "entropy": 6.057462596893311,
      "epoch": 2.3520929826094554,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004444844198687732,
      "loss": 5.7078,
      "mean_token_accuracy": 0.1670807495713234,
      "num_tokens": 54544445.0,
      "step": 26915
    },
    {
      "entropy": 6.071163320541382,
      "epoch": 2.3525299309621603,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004444639230225622,
      "loss": 5.7166,
      "mean_token_accuracy": 0.16619838625192643,
      "num_tokens": 54554966.0,
      "step": 26920
    },
    {
      "entropy": 6.124922800064087,
      "epoch": 2.352966879314865,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004444434229259444,
      "loss": 5.7258,
      "mean_token_accuracy": 0.1688815951347351,
      "num_tokens": 54565005.0,
      "step": 26925
    },
    {
      "entropy": 6.104650449752808,
      "epoch": 2.3534038276675697,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004444229195793129,
      "loss": 5.7337,
      "mean_token_accuracy": 0.16412517577409744,
      "num_tokens": 54575242.0,
      "step": 26930
    },
    {
      "entropy": 6.088753747940063,
      "epoch": 2.3538407760202746,
      "grad_norm": 1.5,
      "learning_rate": 0.0004444024129830611,
      "loss": 5.7375,
      "mean_token_accuracy": 0.16295743435621263,
      "num_tokens": 54585448.0,
      "step": 26935
    },
    {
      "entropy": 6.066171646118164,
      "epoch": 2.354277724372979,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004443819031375822,
      "loss": 5.6902,
      "mean_token_accuracy": 0.16824013441801072,
      "num_tokens": 54595891.0,
      "step": 26940
    },
    {
      "entropy": 6.152268648147583,
      "epoch": 2.3547146727256836,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004443613900432697,
      "loss": 5.808,
      "mean_token_accuracy": 0.15487637221813202,
      "num_tokens": 54607407.0,
      "step": 26945
    },
    {
      "entropy": 6.110444831848144,
      "epoch": 2.3551516210783885,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044434087370051693,
      "loss": 5.7633,
      "mean_token_accuracy": 0.1608761504292488,
      "num_tokens": 54617337.0,
      "step": 26950
    },
    {
      "entropy": 6.106645774841309,
      "epoch": 2.3555885694310934,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004443203541097175,
      "loss": 5.7353,
      "mean_token_accuracy": 0.16762966066598892,
      "num_tokens": 54627914.0,
      "step": 26955
    },
    {
      "entropy": 6.080085515975952,
      "epoch": 2.356025517783798,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004442998312712649,
      "loss": 5.641,
      "mean_token_accuracy": 0.17345616221427917,
      "num_tokens": 54636709.0,
      "step": 26960
    },
    {
      "entropy": 6.022604942321777,
      "epoch": 2.356462466136503,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044427930518555294,
      "loss": 5.694,
      "mean_token_accuracy": 0.16897065564990044,
      "num_tokens": 54646167.0,
      "step": 26965
    },
    {
      "entropy": 6.103919982910156,
      "epoch": 2.3568994144892073,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004442587758529751,
      "loss": 5.7683,
      "mean_token_accuracy": 0.16288076490163803,
      "num_tokens": 54656663.0,
      "step": 26970
    },
    {
      "entropy": 6.162918281555176,
      "epoch": 2.357336362841912,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004442382432739253,
      "loss": 5.7862,
      "mean_token_accuracy": 0.16106356978416442,
      "num_tokens": 54667123.0,
      "step": 26975
    },
    {
      "entropy": 6.1397425651550295,
      "epoch": 2.3577733111946166,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00044421770744879725,
      "loss": 5.7907,
      "mean_token_accuracy": 0.16578477770090103,
      "num_tokens": 54677381.0,
      "step": 26980
    },
    {
      "entropy": 6.117559289932251,
      "epoch": 2.3582102595473216,
      "grad_norm": 2.1875,
      "learning_rate": 0.000444197168377985,
      "loss": 5.727,
      "mean_token_accuracy": 0.1662319168448448,
      "num_tokens": 54687190.0,
      "step": 26985
    },
    {
      "entropy": 6.158610486984253,
      "epoch": 2.3586472079000265,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00044417662606188236,
      "loss": 5.7882,
      "mean_token_accuracy": 0.16552705615758895,
      "num_tokens": 54697102.0,
      "step": 26990
    },
    {
      "entropy": 6.1186912059783936,
      "epoch": 2.359084156252731,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004441560805008834,
      "loss": 5.7694,
      "mean_token_accuracy": 0.16712108254432678,
      "num_tokens": 54707315.0,
      "step": 26995
    },
    {
      "entropy": 6.126333570480346,
      "epoch": 2.3595211046054354,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004441355316953822,
      "loss": 5.8343,
      "mean_token_accuracy": 0.15604881197214127,
      "num_tokens": 54717873.0,
      "step": 27000
    },
    {
      "epoch": 2.3595211046054354,
      "eval_entropy": 5.843953955835623,
      "eval_loss": 5.817349433898926,
      "eval_mean_token_accuracy": 0.17057172124325803,
      "eval_num_tokens": 54717873.0,
      "eval_runtime": 27.7783,
      "eval_samples_per_second": 1332.478,
      "eval_steps_per_second": 166.569,
      "step": 27000
    },
    {
      "entropy": 6.132651042938233,
      "epoch": 2.3599580529581403,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004441149796457728,
      "loss": 5.8006,
      "mean_token_accuracy": 0.16085389107465745,
      "num_tokens": 54727723.0,
      "step": 27005
    },
    {
      "entropy": 6.117178630828858,
      "epoch": 2.3603950013108452,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00044409442435244947,
      "loss": 5.7193,
      "mean_token_accuracy": 0.1709327593445778,
      "num_tokens": 54738108.0,
      "step": 27010
    },
    {
      "entropy": 6.116751623153687,
      "epoch": 2.3608319496635497,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004440738658158065,
      "loss": 5.8243,
      "mean_token_accuracy": 0.1626998320221901,
      "num_tokens": 54749011.0,
      "step": 27015
    },
    {
      "entropy": 6.1564007759094235,
      "epoch": 2.3612688980162546,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004440533040362382,
      "loss": 5.7507,
      "mean_token_accuracy": 0.16390021443367003,
      "num_tokens": 54759922.0,
      "step": 27020
    },
    {
      "entropy": 6.092589712142944,
      "epoch": 2.361705846368959,
      "grad_norm": 2.84375,
      "learning_rate": 0.00044403273901413897,
      "loss": 5.7608,
      "mean_token_accuracy": 0.1576153367757797,
      "num_tokens": 54770134.0,
      "step": 27025
    },
    {
      "entropy": 6.121652746200562,
      "epoch": 2.362142794721664,
      "grad_norm": 1.546875,
      "learning_rate": 0.00044401217074990323,
      "loss": 5.6985,
      "mean_token_accuracy": 0.16920360028743744,
      "num_tokens": 54780131.0,
      "step": 27030
    },
    {
      "entropy": 6.096730899810791,
      "epoch": 2.3625797430743685,
      "grad_norm": 2.578125,
      "learning_rate": 0.00044399159924392545,
      "loss": 5.7904,
      "mean_token_accuracy": 0.16466324627399445,
      "num_tokens": 54790601.0,
      "step": 27035
    },
    {
      "entropy": 6.131312513351441,
      "epoch": 2.3630166914270734,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004439710244966003,
      "loss": 5.7245,
      "mean_token_accuracy": 0.16603925824165344,
      "num_tokens": 54799995.0,
      "step": 27040
    },
    {
      "entropy": 6.143316411972046,
      "epoch": 2.363453639779778,
      "grad_norm": 1.625,
      "learning_rate": 0.0004439504465083223,
      "loss": 5.7417,
      "mean_token_accuracy": 0.16401946395635605,
      "num_tokens": 54810299.0,
      "step": 27045
    },
    {
      "entropy": 6.127547979354858,
      "epoch": 2.3638905881324828,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004439298652794863,
      "loss": 5.7149,
      "mean_token_accuracy": 0.16457904875278473,
      "num_tokens": 54820802.0,
      "step": 27050
    },
    {
      "entropy": 6.086389350891113,
      "epoch": 2.3643275364851872,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00044390928081048693,
      "loss": 5.7551,
      "mean_token_accuracy": 0.16736960411071777,
      "num_tokens": 54831077.0,
      "step": 27055
    },
    {
      "entropy": 6.081876039505005,
      "epoch": 2.364764484837892,
      "grad_norm": 2.1875,
      "learning_rate": 0.000443888693101719,
      "loss": 5.7514,
      "mean_token_accuracy": 0.16782679855823518,
      "num_tokens": 54842574.0,
      "step": 27060
    },
    {
      "entropy": 6.119289827346802,
      "epoch": 2.365201433190597,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044386810215357747,
      "loss": 5.6828,
      "mean_token_accuracy": 0.16804996579885484,
      "num_tokens": 54852365.0,
      "step": 27065
    },
    {
      "entropy": 6.063072395324707,
      "epoch": 2.3656383815433015,
      "grad_norm": 1.796875,
      "learning_rate": 0.00044384750796645724,
      "loss": 5.7554,
      "mean_token_accuracy": 0.1613278403878212,
      "num_tokens": 54863008.0,
      "step": 27070
    },
    {
      "entropy": 6.09646258354187,
      "epoch": 2.3660753298960064,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004438269105407534,
      "loss": 5.7477,
      "mean_token_accuracy": 0.16488235741853713,
      "num_tokens": 54873134.0,
      "step": 27075
    },
    {
      "entropy": 6.165373086929321,
      "epoch": 2.366512278248711,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00044380630987686086,
      "loss": 5.8035,
      "mean_token_accuracy": 0.1644013248383999,
      "num_tokens": 54883075.0,
      "step": 27080
    },
    {
      "entropy": 6.108354282379151,
      "epoch": 2.366949226601416,
      "grad_norm": 2.171875,
      "learning_rate": 0.00044378570597517495,
      "loss": 5.7088,
      "mean_token_accuracy": 0.16800790876150132,
      "num_tokens": 54892788.0,
      "step": 27085
    },
    {
      "entropy": 6.014555597305298,
      "epoch": 2.3673861749541203,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004437650988360906,
      "loss": 5.7008,
      "mean_token_accuracy": 0.16720739752054214,
      "num_tokens": 54902708.0,
      "step": 27090
    },
    {
      "entropy": 6.065566635131836,
      "epoch": 2.367823123306825,
      "grad_norm": 2.625,
      "learning_rate": 0.0004437444884600034,
      "loss": 5.6937,
      "mean_token_accuracy": 0.1703638330101967,
      "num_tokens": 54912906.0,
      "step": 27095
    },
    {
      "entropy": 6.081116151809693,
      "epoch": 2.3682600716595297,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00044372387484730835,
      "loss": 5.7589,
      "mean_token_accuracy": 0.16371235698461534,
      "num_tokens": 54922802.0,
      "step": 27100
    },
    {
      "entropy": 6.140959453582764,
      "epoch": 2.3686970200122346,
      "grad_norm": 1.828125,
      "learning_rate": 0.00044370325799840106,
      "loss": 5.8009,
      "mean_token_accuracy": 0.16207725703716278,
      "num_tokens": 54933203.0,
      "step": 27105
    },
    {
      "entropy": 6.1456695079803465,
      "epoch": 2.369133968364939,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00044368263791367684,
      "loss": 5.7241,
      "mean_token_accuracy": 0.16792073845863342,
      "num_tokens": 54944081.0,
      "step": 27110
    },
    {
      "entropy": 6.076117134094238,
      "epoch": 2.369570916717644,
      "grad_norm": 2.484375,
      "learning_rate": 0.00044366201459353124,
      "loss": 5.7496,
      "mean_token_accuracy": 0.16795005798339843,
      "num_tokens": 54953421.0,
      "step": 27115
    },
    {
      "entropy": 6.062148237228394,
      "epoch": 2.370007865070349,
      "grad_norm": 1.828125,
      "learning_rate": 0.00044364138803835984,
      "loss": 5.761,
      "mean_token_accuracy": 0.16398418098688125,
      "num_tokens": 54964127.0,
      "step": 27120
    },
    {
      "entropy": 6.093556213378906,
      "epoch": 2.3704448134230534,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00044362075824855816,
      "loss": 5.7022,
      "mean_token_accuracy": 0.1713394895195961,
      "num_tokens": 54974636.0,
      "step": 27125
    },
    {
      "entropy": 6.134442090988159,
      "epoch": 2.3708817617757583,
      "grad_norm": 2.21875,
      "learning_rate": 0.00044360012522452215,
      "loss": 5.7974,
      "mean_token_accuracy": 0.16102064549922943,
      "num_tokens": 54985542.0,
      "step": 27130
    },
    {
      "entropy": 6.109624576568604,
      "epoch": 2.3713187101284627,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004435794889666472,
      "loss": 5.7246,
      "mean_token_accuracy": 0.16860554963350297,
      "num_tokens": 54994973.0,
      "step": 27135
    },
    {
      "entropy": 6.123045539855957,
      "epoch": 2.3717556584811677,
      "grad_norm": 1.8125,
      "learning_rate": 0.00044355884947532947,
      "loss": 5.815,
      "mean_token_accuracy": 0.16150278300046922,
      "num_tokens": 55006510.0,
      "step": 27140
    },
    {
      "entropy": 6.146032381057739,
      "epoch": 2.372192606833872,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004435382067509647,
      "loss": 5.8043,
      "mean_token_accuracy": 0.16557089239358902,
      "num_tokens": 55016375.0,
      "step": 27145
    },
    {
      "entropy": 6.05098705291748,
      "epoch": 2.372629555186577,
      "grad_norm": 2.234375,
      "learning_rate": 0.00044351756079394865,
      "loss": 5.7334,
      "mean_token_accuracy": 0.17163571566343308,
      "num_tokens": 55026213.0,
      "step": 27150
    },
    {
      "entropy": 6.204241609573364,
      "epoch": 2.3730665035392815,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004434969116046776,
      "loss": 5.8359,
      "mean_token_accuracy": 0.15655337572097777,
      "num_tokens": 55036836.0,
      "step": 27155
    },
    {
      "entropy": 6.129290676116943,
      "epoch": 2.3735034518919864,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004434762591835475,
      "loss": 5.675,
      "mean_token_accuracy": 0.1665548115968704,
      "num_tokens": 55046243.0,
      "step": 27160
    },
    {
      "entropy": 6.052304744720459,
      "epoch": 2.373940400244691,
      "grad_norm": 2.109375,
      "learning_rate": 0.00044345560353095437,
      "loss": 5.7287,
      "mean_token_accuracy": 0.16715483814477922,
      "num_tokens": 55055882.0,
      "step": 27165
    },
    {
      "entropy": 6.038439083099365,
      "epoch": 2.374377348597396,
      "grad_norm": 2.125,
      "learning_rate": 0.0004434349446472946,
      "loss": 5.6753,
      "mean_token_accuracy": 0.16554410010576248,
      "num_tokens": 55066398.0,
      "step": 27170
    },
    {
      "entropy": 6.135483837127685,
      "epoch": 2.3748142969501007,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044341428253296436,
      "loss": 5.8104,
      "mean_token_accuracy": 0.15811321437358855,
      "num_tokens": 55077053.0,
      "step": 27175
    },
    {
      "entropy": 6.117898464202881,
      "epoch": 2.375251245302805,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004433936171883598,
      "loss": 5.7406,
      "mean_token_accuracy": 0.16786708384752275,
      "num_tokens": 55086973.0,
      "step": 27180
    },
    {
      "entropy": 6.105523300170899,
      "epoch": 2.37568819365551,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004433729486138775,
      "loss": 5.7149,
      "mean_token_accuracy": 0.16717524081468582,
      "num_tokens": 55096162.0,
      "step": 27185
    },
    {
      "entropy": 6.083913850784302,
      "epoch": 2.3761251420082146,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00044335227680991385,
      "loss": 5.7144,
      "mean_token_accuracy": 0.16725060790777208,
      "num_tokens": 55105852.0,
      "step": 27190
    },
    {
      "entropy": 6.102247953414917,
      "epoch": 2.3765620903609195,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004433316017768653,
      "loss": 5.6641,
      "mean_token_accuracy": 0.17710405439138413,
      "num_tokens": 55116092.0,
      "step": 27195
    },
    {
      "entropy": 6.1314475536346436,
      "epoch": 2.376999038713624,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004433109235151284,
      "loss": 5.7324,
      "mean_token_accuracy": 0.1679028958082199,
      "num_tokens": 55125439.0,
      "step": 27200
    },
    {
      "entropy": 6.103550910949707,
      "epoch": 2.377435987066329,
      "grad_norm": 3.25,
      "learning_rate": 0.00044329024202509977,
      "loss": 5.7721,
      "mean_token_accuracy": 0.16253959238529206,
      "num_tokens": 55134684.0,
      "step": 27205
    },
    {
      "entropy": 6.091237640380859,
      "epoch": 2.3778729354190333,
      "grad_norm": 3.1875,
      "learning_rate": 0.00044326955730717614,
      "loss": 5.6974,
      "mean_token_accuracy": 0.16963720470666885,
      "num_tokens": 55143853.0,
      "step": 27210
    },
    {
      "entropy": 6.198922109603882,
      "epoch": 2.3783098837717382,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004432488693617542,
      "loss": 5.9192,
      "mean_token_accuracy": 0.14730891436338425,
      "num_tokens": 55154190.0,
      "step": 27215
    },
    {
      "entropy": 6.161638164520264,
      "epoch": 2.3787468321244427,
      "grad_norm": 1.609375,
      "learning_rate": 0.00044322817818923083,
      "loss": 5.7688,
      "mean_token_accuracy": 0.16888174414634705,
      "num_tokens": 55164158.0,
      "step": 27220
    },
    {
      "entropy": 6.11997013092041,
      "epoch": 2.3791837804771476,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00044320748379000284,
      "loss": 5.8198,
      "mean_token_accuracy": 0.1543012171983719,
      "num_tokens": 55173635.0,
      "step": 27225
    },
    {
      "entropy": 6.13445520401001,
      "epoch": 2.3796207288298525,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044318678616446717,
      "loss": 5.7263,
      "mean_token_accuracy": 0.17354104965925216,
      "num_tokens": 55182972.0,
      "step": 27230
    },
    {
      "entropy": 6.164822769165039,
      "epoch": 2.380057677182557,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00044316608531302085,
      "loss": 5.778,
      "mean_token_accuracy": 0.1629917338490486,
      "num_tokens": 55192769.0,
      "step": 27235
    },
    {
      "entropy": 6.088997459411621,
      "epoch": 2.380494625535262,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044314538123606083,
      "loss": 5.764,
      "mean_token_accuracy": 0.1656671240925789,
      "num_tokens": 55203241.0,
      "step": 27240
    },
    {
      "entropy": 6.049153995513916,
      "epoch": 2.3809315738879664,
      "grad_norm": 3.34375,
      "learning_rate": 0.00044312467393398434,
      "loss": 5.7147,
      "mean_token_accuracy": 0.1704629674553871,
      "num_tokens": 55214272.0,
      "step": 27245
    },
    {
      "entropy": 6.091970157623291,
      "epoch": 2.3813685222406713,
      "grad_norm": 1.78125,
      "learning_rate": 0.00044310396340718853,
      "loss": 5.651,
      "mean_token_accuracy": 0.17257196009159087,
      "num_tokens": 55223947.0,
      "step": 27250
    },
    {
      "entropy": 6.0924906730651855,
      "epoch": 2.3818054705933758,
      "grad_norm": 2.109375,
      "learning_rate": 0.00044308324965607054,
      "loss": 5.8094,
      "mean_token_accuracy": 0.15821968391537666,
      "num_tokens": 55234906.0,
      "step": 27255
    },
    {
      "entropy": 6.068419122695923,
      "epoch": 2.3822424189460807,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004430625326810278,
      "loss": 5.6368,
      "mean_token_accuracy": 0.1692095845937729,
      "num_tokens": 55244785.0,
      "step": 27260
    },
    {
      "entropy": 6.12485556602478,
      "epoch": 2.382679367298785,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044304181248245754,
      "loss": 5.7986,
      "mean_token_accuracy": 0.16326246410608292,
      "num_tokens": 55255053.0,
      "step": 27265
    },
    {
      "entropy": 6.17266616821289,
      "epoch": 2.38311631565149,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004430210890607574,
      "loss": 5.8138,
      "mean_token_accuracy": 0.16410293132066728,
      "num_tokens": 55264525.0,
      "step": 27270
    },
    {
      "entropy": 6.189845991134644,
      "epoch": 2.3835532640041945,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044300036241632463,
      "loss": 5.7416,
      "mean_token_accuracy": 0.16580468267202378,
      "num_tokens": 55273486.0,
      "step": 27275
    },
    {
      "entropy": 6.113778114318848,
      "epoch": 2.3839902123568995,
      "grad_norm": 7.96875,
      "learning_rate": 0.00044297963254955687,
      "loss": 5.7217,
      "mean_token_accuracy": 0.16905015259981154,
      "num_tokens": 55283164.0,
      "step": 27280
    },
    {
      "entropy": 6.088917255401611,
      "epoch": 2.3844271607096044,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00044295889946085185,
      "loss": 5.7904,
      "mean_token_accuracy": 0.1609894037246704,
      "num_tokens": 55294657.0,
      "step": 27285
    },
    {
      "entropy": 6.21134386062622,
      "epoch": 2.384864109062309,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00044293816315060704,
      "loss": 5.7815,
      "mean_token_accuracy": 0.1629866451025009,
      "num_tokens": 55304743.0,
      "step": 27290
    },
    {
      "entropy": 6.08080906867981,
      "epoch": 2.3853010574150133,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004429174236192203,
      "loss": 5.6929,
      "mean_token_accuracy": 0.18190576136112213,
      "num_tokens": 55314106.0,
      "step": 27295
    },
    {
      "entropy": 6.145316123962402,
      "epoch": 2.385738005767718,
      "grad_norm": 2.3125,
      "learning_rate": 0.00044289668086708935,
      "loss": 5.7288,
      "mean_token_accuracy": 0.16519947797060014,
      "num_tokens": 55324407.0,
      "step": 27300
    },
    {
      "entropy": 6.082515382766724,
      "epoch": 2.386174954120423,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00044287593489461205,
      "loss": 5.7518,
      "mean_token_accuracy": 0.16830275058746338,
      "num_tokens": 55335853.0,
      "step": 27305
    },
    {
      "entropy": 6.0855419635772705,
      "epoch": 2.3866119024731276,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044285518570218635,
      "loss": 5.7386,
      "mean_token_accuracy": 0.16572066247463227,
      "num_tokens": 55345947.0,
      "step": 27310
    },
    {
      "entropy": 6.144145965576172,
      "epoch": 2.3870488508258325,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004428344332902103,
      "loss": 5.7398,
      "mean_token_accuracy": 0.16362067312002182,
      "num_tokens": 55355709.0,
      "step": 27315
    },
    {
      "entropy": 6.154619836807251,
      "epoch": 2.387485799178537,
      "grad_norm": 2.296875,
      "learning_rate": 0.00044281367765908186,
      "loss": 5.6824,
      "mean_token_accuracy": 0.16676439642906188,
      "num_tokens": 55365597.0,
      "step": 27320
    },
    {
      "entropy": 6.106455898284912,
      "epoch": 2.387922747531242,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044279291880919915,
      "loss": 5.7394,
      "mean_token_accuracy": 0.1620962381362915,
      "num_tokens": 55375508.0,
      "step": 27325
    },
    {
      "entropy": 6.040557813644409,
      "epoch": 2.3883596958839464,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044277215674096035,
      "loss": 5.7191,
      "mean_token_accuracy": 0.1662188097834587,
      "num_tokens": 55384928.0,
      "step": 27330
    },
    {
      "entropy": 6.1230895042419435,
      "epoch": 2.3887966442366513,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004427513914547637,
      "loss": 5.7424,
      "mean_token_accuracy": 0.16833312809467316,
      "num_tokens": 55394805.0,
      "step": 27335
    },
    {
      "entropy": 6.119432163238526,
      "epoch": 2.3892335925893557,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004427306229510074,
      "loss": 5.779,
      "mean_token_accuracy": 0.16935325413942337,
      "num_tokens": 55405360.0,
      "step": 27340
    },
    {
      "entropy": 6.117374134063721,
      "epoch": 2.3896705409420607,
      "grad_norm": 1.640625,
      "learning_rate": 0.00044270985123008984,
      "loss": 5.7259,
      "mean_token_accuracy": 0.1638546586036682,
      "num_tokens": 55415124.0,
      "step": 27345
    },
    {
      "entropy": 6.152176427841186,
      "epoch": 2.390107489294765,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004426890762924096,
      "loss": 5.7557,
      "mean_token_accuracy": 0.16583392918109893,
      "num_tokens": 55425621.0,
      "step": 27350
    },
    {
      "entropy": 6.167156982421875,
      "epoch": 2.39054443764747,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004426682981383648,
      "loss": 5.7981,
      "mean_token_accuracy": 0.15737639516592025,
      "num_tokens": 55435812.0,
      "step": 27355
    },
    {
      "entropy": 6.137868595123291,
      "epoch": 2.390981386000175,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004426475167683543,
      "loss": 5.7716,
      "mean_token_accuracy": 0.1609954923391342,
      "num_tokens": 55445462.0,
      "step": 27360
    },
    {
      "entropy": 6.122380685806275,
      "epoch": 2.3914183343528794,
      "grad_norm": 2.328125,
      "learning_rate": 0.00044262673218277654,
      "loss": 5.7528,
      "mean_token_accuracy": 0.1610233500599861,
      "num_tokens": 55455992.0,
      "step": 27365
    },
    {
      "entropy": 6.166995239257813,
      "epoch": 2.3918552827055843,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00044260594438203025,
      "loss": 5.7925,
      "mean_token_accuracy": 0.1611805722117424,
      "num_tokens": 55466715.0,
      "step": 27370
    },
    {
      "entropy": 6.158024263381958,
      "epoch": 2.392292231058289,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044258515336651416,
      "loss": 5.7463,
      "mean_token_accuracy": 0.16754869967699051,
      "num_tokens": 55476280.0,
      "step": 27375
    },
    {
      "entropy": 6.08666353225708,
      "epoch": 2.3927291794109937,
      "grad_norm": 1.625,
      "learning_rate": 0.00044256435913662696,
      "loss": 5.7206,
      "mean_token_accuracy": 0.1685633435845375,
      "num_tokens": 55486105.0,
      "step": 27380
    },
    {
      "entropy": 6.152891492843628,
      "epoch": 2.393166127763698,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004425435616927676,
      "loss": 5.7425,
      "mean_token_accuracy": 0.16303757727146148,
      "num_tokens": 55496152.0,
      "step": 27385
    },
    {
      "entropy": 6.117990303039551,
      "epoch": 2.393603076116403,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004425227610353349,
      "loss": 5.7711,
      "mean_token_accuracy": 0.16863640695810317,
      "num_tokens": 55507104.0,
      "step": 27390
    },
    {
      "entropy": 6.100887250900269,
      "epoch": 2.3940400244691076,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004425019571647279,
      "loss": 5.7504,
      "mean_token_accuracy": 0.17108944952487945,
      "num_tokens": 55517114.0,
      "step": 27395
    },
    {
      "entropy": 6.0673308849334715,
      "epoch": 2.3944769728218125,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00044248115008134557,
      "loss": 5.6535,
      "mean_token_accuracy": 0.1743907406926155,
      "num_tokens": 55526719.0,
      "step": 27400
    },
    {
      "entropy": 6.13311538696289,
      "epoch": 2.394913921174517,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044246033978558697,
      "loss": 5.7066,
      "mean_token_accuracy": 0.16790243685245515,
      "num_tokens": 55536760.0,
      "step": 27405
    },
    {
      "entropy": 6.082996129989624,
      "epoch": 2.395350869527222,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044243952627785136,
      "loss": 5.7577,
      "mean_token_accuracy": 0.16544368416070937,
      "num_tokens": 55547193.0,
      "step": 27410
    },
    {
      "entropy": 6.118897008895874,
      "epoch": 2.395787817879927,
      "grad_norm": 2.0,
      "learning_rate": 0.00044241870955853793,
      "loss": 5.694,
      "mean_token_accuracy": 0.16929915845394133,
      "num_tokens": 55557257.0,
      "step": 27415
    },
    {
      "entropy": 6.094059753417969,
      "epoch": 2.3962247662326313,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00044239788962804584,
      "loss": 5.7388,
      "mean_token_accuracy": 0.1634899765253067,
      "num_tokens": 55568191.0,
      "step": 27420
    },
    {
      "entropy": 6.111023950576782,
      "epoch": 2.396661714585336,
      "grad_norm": 1.578125,
      "learning_rate": 0.00044237706648677456,
      "loss": 5.757,
      "mean_token_accuracy": 0.15868191421031952,
      "num_tokens": 55578983.0,
      "step": 27425
    },
    {
      "entropy": 6.088877630233765,
      "epoch": 2.3970986629380406,
      "grad_norm": 1.625,
      "learning_rate": 0.0004423562401351234,
      "loss": 5.7117,
      "mean_token_accuracy": 0.16774897873401642,
      "num_tokens": 55588772.0,
      "step": 27430
    },
    {
      "entropy": 6.055570936203003,
      "epoch": 2.3975356112907455,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004423354105734919,
      "loss": 5.745,
      "mean_token_accuracy": 0.16710472702980042,
      "num_tokens": 55599514.0,
      "step": 27435
    },
    {
      "entropy": 6.113922882080078,
      "epoch": 2.39797255964345,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004423145778022795,
      "loss": 5.7648,
      "mean_token_accuracy": 0.15752217173576355,
      "num_tokens": 55609183.0,
      "step": 27440
    },
    {
      "entropy": 6.1196294784545895,
      "epoch": 2.398409507996155,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004422937418218859,
      "loss": 5.7618,
      "mean_token_accuracy": 0.16847009658813478,
      "num_tokens": 55618836.0,
      "step": 27445
    },
    {
      "entropy": 6.136512517929077,
      "epoch": 2.3988464563488594,
      "grad_norm": 1.71875,
      "learning_rate": 0.00044227290263271066,
      "loss": 5.7984,
      "mean_token_accuracy": 0.16257413625717163,
      "num_tokens": 55629314.0,
      "step": 27450
    },
    {
      "entropy": 6.130738925933838,
      "epoch": 2.3992834047015643,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004422520602351535,
      "loss": 5.6761,
      "mean_token_accuracy": 0.18548360764980315,
      "num_tokens": 55640542.0,
      "step": 27455
    },
    {
      "entropy": 6.101352024078369,
      "epoch": 2.399720353054269,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004422312146296141,
      "loss": 5.8034,
      "mean_token_accuracy": 0.1602294459939003,
      "num_tokens": 55650165.0,
      "step": 27460
    },
    {
      "entropy": 6.104769897460938,
      "epoch": 2.4001573014069737,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044221036581649245,
      "loss": 5.7437,
      "mean_token_accuracy": 0.1630367711186409,
      "num_tokens": 55658962.0,
      "step": 27465
    },
    {
      "entropy": 6.081813907623291,
      "epoch": 2.4005942497596786,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004421895137961884,
      "loss": 5.7297,
      "mean_token_accuracy": 0.15928539037704467,
      "num_tokens": 55668357.0,
      "step": 27470
    },
    {
      "entropy": 6.166380977630615,
      "epoch": 2.401031198112383,
      "grad_norm": 1.625,
      "learning_rate": 0.00044216865856910176,
      "loss": 5.8333,
      "mean_token_accuracy": 0.1643952399492264,
      "num_tokens": 55678313.0,
      "step": 27475
    },
    {
      "entropy": 6.103434896469116,
      "epoch": 2.401468146465088,
      "grad_norm": 1.5,
      "learning_rate": 0.00044214780013563277,
      "loss": 5.7478,
      "mean_token_accuracy": 0.16137870997190476,
      "num_tokens": 55689332.0,
      "step": 27480
    },
    {
      "entropy": 6.090963888168335,
      "epoch": 2.4019050948177925,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00044212693849618136,
      "loss": 5.733,
      "mean_token_accuracy": 0.1704863578081131,
      "num_tokens": 55699397.0,
      "step": 27485
    },
    {
      "entropy": 6.13179292678833,
      "epoch": 2.4023420431704974,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004421060736511476,
      "loss": 5.7984,
      "mean_token_accuracy": 0.16132167726755142,
      "num_tokens": 55708548.0,
      "step": 27490
    },
    {
      "entropy": 6.18438310623169,
      "epoch": 2.402778991523202,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044208520560093183,
      "loss": 5.7877,
      "mean_token_accuracy": 0.16231499910354613,
      "num_tokens": 55718043.0,
      "step": 27495
    },
    {
      "entropy": 6.142577362060547,
      "epoch": 2.4032159398759068,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004420643343459343,
      "loss": 5.7444,
      "mean_token_accuracy": 0.16370736807584763,
      "num_tokens": 55727889.0,
      "step": 27500
    },
    {
      "entropy": 6.098811626434326,
      "epoch": 2.4036528882286112,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004420434598865553,
      "loss": 5.7765,
      "mean_token_accuracy": 0.17216399013996125,
      "num_tokens": 55738993.0,
      "step": 27505
    },
    {
      "entropy": 6.107850360870361,
      "epoch": 2.404089836581316,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004420225822231952,
      "loss": 5.7707,
      "mean_token_accuracy": 0.16577051430940629,
      "num_tokens": 55748858.0,
      "step": 27510
    },
    {
      "entropy": 6.1087583065032955,
      "epoch": 2.4045267849340206,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00044200170135625446,
      "loss": 5.751,
      "mean_token_accuracy": 0.17046456933021545,
      "num_tokens": 55758639.0,
      "step": 27515
    },
    {
      "entropy": 6.202047348022461,
      "epoch": 2.4049637332867255,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004419808172861336,
      "loss": 5.802,
      "mean_token_accuracy": 0.15898158997297288,
      "num_tokens": 55769632.0,
      "step": 27520
    },
    {
      "entropy": 6.152137327194214,
      "epoch": 2.4054006816394304,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004419599300132331,
      "loss": 5.7965,
      "mean_token_accuracy": 0.16236904859542847,
      "num_tokens": 55780147.0,
      "step": 27525
    },
    {
      "entropy": 6.076031351089478,
      "epoch": 2.405837629992135,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044193903953795367,
      "loss": 5.6893,
      "mean_token_accuracy": 0.1636853963136673,
      "num_tokens": 55790933.0,
      "step": 27530
    },
    {
      "entropy": 6.116088962554931,
      "epoch": 2.40627457834484,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044191814586069605,
      "loss": 5.7053,
      "mean_token_accuracy": 0.17439066022634506,
      "num_tokens": 55800479.0,
      "step": 27535
    },
    {
      "entropy": 6.070321846008301,
      "epoch": 2.4067115266975443,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004418972489818609,
      "loss": 5.7243,
      "mean_token_accuracy": 0.1740534409880638,
      "num_tokens": 55810613.0,
      "step": 27540
    },
    {
      "entropy": 6.134655141830445,
      "epoch": 2.407148475050249,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044187634890184904,
      "loss": 5.7942,
      "mean_token_accuracy": 0.16379427015781403,
      "num_tokens": 55821507.0,
      "step": 27545
    },
    {
      "entropy": 6.105913352966309,
      "epoch": 2.4075854234029537,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004418554456210614,
      "loss": 5.8174,
      "mean_token_accuracy": 0.1586725428700447,
      "num_tokens": 55831878.0,
      "step": 27550
    },
    {
      "entropy": 6.126907062530518,
      "epoch": 2.4080223717556586,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00044183453913989885,
      "loss": 5.6802,
      "mean_token_accuracy": 0.1693064033985138,
      "num_tokens": 55841314.0,
      "step": 27555
    },
    {
      "entropy": 6.1247725009918215,
      "epoch": 2.408459320108363,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004418136294587624,
      "loss": 5.7357,
      "mean_token_accuracy": 0.16736264526844025,
      "num_tokens": 55851056.0,
      "step": 27560
    },
    {
      "entropy": 6.097383451461792,
      "epoch": 2.408896268461068,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004417927165780531,
      "loss": 5.7359,
      "mean_token_accuracy": 0.16657669246196746,
      "num_tokens": 55860705.0,
      "step": 27565
    },
    {
      "entropy": 6.119924831390381,
      "epoch": 2.4093332168137724,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004417718004981723,
      "loss": 5.7591,
      "mean_token_accuracy": 0.16645372807979583,
      "num_tokens": 55870749.0,
      "step": 27570
    },
    {
      "entropy": 6.178444242477417,
      "epoch": 2.4097701651664774,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004417508812195207,
      "loss": 5.8451,
      "mean_token_accuracy": 0.15860468596220018,
      "num_tokens": 55880135.0,
      "step": 27575
    },
    {
      "entropy": 6.137647914886474,
      "epoch": 2.4102071135191823,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004417299587425001,
      "loss": 5.7001,
      "mean_token_accuracy": 0.17040138691663742,
      "num_tokens": 55890357.0,
      "step": 27580
    },
    {
      "entropy": 6.161546659469605,
      "epoch": 2.4106440618718867,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004417090330675114,
      "loss": 5.7738,
      "mean_token_accuracy": 0.16281794458627702,
      "num_tokens": 55900080.0,
      "step": 27585
    },
    {
      "entropy": 6.0730620384216305,
      "epoch": 2.4110810102245916,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00044168810419495597,
      "loss": 5.771,
      "mean_token_accuracy": 0.16732670962810517,
      "num_tokens": 55911326.0,
      "step": 27590
    },
    {
      "entropy": 6.117193984985351,
      "epoch": 2.411517958577296,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004416671721252355,
      "loss": 5.6958,
      "mean_token_accuracy": 0.17092116326093673,
      "num_tokens": 55921872.0,
      "step": 27595
    },
    {
      "entropy": 6.0810218334198,
      "epoch": 2.411954906930001,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00044164623685875133,
      "loss": 5.7335,
      "mean_token_accuracy": 0.17043702602386473,
      "num_tokens": 55932539.0,
      "step": 27600
    },
    {
      "entropy": 6.101160049438477,
      "epoch": 2.4123918552827055,
      "grad_norm": 1.453125,
      "learning_rate": 0.000441625298395905,
      "loss": 5.7604,
      "mean_token_accuracy": 0.17012778371572496,
      "num_tokens": 55942087.0,
      "step": 27605
    },
    {
      "entropy": 6.109452104568481,
      "epoch": 2.4128288036354104,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004416043567370981,
      "loss": 5.7364,
      "mean_token_accuracy": 0.16601086258888245,
      "num_tokens": 55951604.0,
      "step": 27610
    },
    {
      "entropy": 6.144398307800293,
      "epoch": 2.413265751988115,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004415834118827323,
      "loss": 5.7056,
      "mean_token_accuracy": 0.1705930933356285,
      "num_tokens": 55961687.0,
      "step": 27615
    },
    {
      "entropy": 6.137257862091064,
      "epoch": 2.41370270034082,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004415624638332095,
      "loss": 5.7836,
      "mean_token_accuracy": 0.15888985097408295,
      "num_tokens": 55972088.0,
      "step": 27620
    },
    {
      "entropy": 6.056908369064331,
      "epoch": 2.4141396486935243,
      "grad_norm": 1.890625,
      "learning_rate": 0.00044154151258893125,
      "loss": 5.7807,
      "mean_token_accuracy": 0.15634377896785737,
      "num_tokens": 55981857.0,
      "step": 27625
    },
    {
      "entropy": 6.088106679916382,
      "epoch": 2.414576597046229,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004415205581502996,
      "loss": 5.6802,
      "mean_token_accuracy": 0.1750894919037819,
      "num_tokens": 55992196.0,
      "step": 27630
    },
    {
      "entropy": 6.130224895477295,
      "epoch": 2.415013545398934,
      "grad_norm": 1.703125,
      "learning_rate": 0.00044149960051771643,
      "loss": 5.692,
      "mean_token_accuracy": 0.16772665530443193,
      "num_tokens": 56002295.0,
      "step": 27635
    },
    {
      "entropy": 6.0987818241119385,
      "epoch": 2.4154504937516386,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044147863969158353,
      "loss": 5.7157,
      "mean_token_accuracy": 0.16625651717185974,
      "num_tokens": 56011847.0,
      "step": 27640
    },
    {
      "entropy": 6.086707544326782,
      "epoch": 2.415887442104343,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004414576756723032,
      "loss": 5.6584,
      "mean_token_accuracy": 0.17317849546670913,
      "num_tokens": 56021717.0,
      "step": 27645
    },
    {
      "entropy": 6.124586009979248,
      "epoch": 2.416324390457048,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004414367084602774,
      "loss": 5.7781,
      "mean_token_accuracy": 0.16664691269397736,
      "num_tokens": 56031845.0,
      "step": 27650
    },
    {
      "entropy": 6.080695390701294,
      "epoch": 2.416761338809753,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004414157380559083,
      "loss": 5.7341,
      "mean_token_accuracy": 0.16675712764263154,
      "num_tokens": 56042082.0,
      "step": 27655
    },
    {
      "entropy": 6.142419481277466,
      "epoch": 2.4171982871624573,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00044139476445959817,
      "loss": 5.7413,
      "mean_token_accuracy": 0.16034905463457108,
      "num_tokens": 56052592.0,
      "step": 27660
    },
    {
      "entropy": 6.120674991607666,
      "epoch": 2.4176352355151622,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004413737876717492,
      "loss": 5.8323,
      "mean_token_accuracy": 0.15478274673223497,
      "num_tokens": 56062500.0,
      "step": 27665
    },
    {
      "entropy": 6.095399999618531,
      "epoch": 2.4180721838678667,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004413528076927639,
      "loss": 5.6784,
      "mean_token_accuracy": 0.16773297637701035,
      "num_tokens": 56072912.0,
      "step": 27670
    },
    {
      "entropy": 6.121125173568726,
      "epoch": 2.4185091322205716,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004413318245230446,
      "loss": 5.772,
      "mean_token_accuracy": 0.161003740131855,
      "num_tokens": 56082591.0,
      "step": 27675
    },
    {
      "entropy": 6.077878761291504,
      "epoch": 2.418946080573276,
      "grad_norm": 2.0,
      "learning_rate": 0.0004413108381629936,
      "loss": 5.6612,
      "mean_token_accuracy": 0.1714988499879837,
      "num_tokens": 56091738.0,
      "step": 27680
    },
    {
      "entropy": 6.10207200050354,
      "epoch": 2.419383028925981,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004412898486130137,
      "loss": 5.7642,
      "mean_token_accuracy": 0.1613841399550438,
      "num_tokens": 56102011.0,
      "step": 27685
    },
    {
      "entropy": 6.125308704376221,
      "epoch": 2.4198199772786855,
      "grad_norm": 2.703125,
      "learning_rate": 0.00044126885587350735,
      "loss": 5.7485,
      "mean_token_accuracy": 0.17180779129266738,
      "num_tokens": 56112901.0,
      "step": 27690
    },
    {
      "entropy": 6.098558378219605,
      "epoch": 2.4202569256313904,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004412478599448772,
      "loss": 5.7475,
      "mean_token_accuracy": 0.16935478001832963,
      "num_tokens": 56123598.0,
      "step": 27695
    },
    {
      "entropy": 6.064134120941162,
      "epoch": 2.420693873984095,
      "grad_norm": 2.046875,
      "learning_rate": 0.000441226860827526,
      "loss": 5.764,
      "mean_token_accuracy": 0.16110283583402635,
      "num_tokens": 56133686.0,
      "step": 27700
    },
    {
      "entropy": 6.150255823135376,
      "epoch": 2.4211308223367998,
      "grad_norm": 2.109375,
      "learning_rate": 0.00044120585852185654,
      "loss": 5.7642,
      "mean_token_accuracy": 0.16273897141218185,
      "num_tokens": 56143722.0,
      "step": 27705
    },
    {
      "entropy": 6.104858493804931,
      "epoch": 2.4215677706895047,
      "grad_norm": 1.640625,
      "learning_rate": 0.00044118485302827156,
      "loss": 5.6637,
      "mean_token_accuracy": 0.17395438253879547,
      "num_tokens": 56153781.0,
      "step": 27710
    },
    {
      "entropy": 6.035255861282349,
      "epoch": 2.422004719042209,
      "grad_norm": 2.0,
      "learning_rate": 0.00044116384434717405,
      "loss": 5.697,
      "mean_token_accuracy": 0.1715370923280716,
      "num_tokens": 56163398.0,
      "step": 27715
    },
    {
      "entropy": 6.12542314529419,
      "epoch": 2.422441667394914,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004411428324789669,
      "loss": 5.8033,
      "mean_token_accuracy": 0.16449213474988938,
      "num_tokens": 56173418.0,
      "step": 27720
    },
    {
      "entropy": 6.1616229057312015,
      "epoch": 2.4228786157476185,
      "grad_norm": 1.5,
      "learning_rate": 0.00044112181742405324,
      "loss": 5.8284,
      "mean_token_accuracy": 0.1605752170085907,
      "num_tokens": 56184547.0,
      "step": 27725
    },
    {
      "entropy": 6.161860656738281,
      "epoch": 2.4233155641003234,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000441100799182836,
      "loss": 5.7395,
      "mean_token_accuracy": 0.1591387152671814,
      "num_tokens": 56195404.0,
      "step": 27730
    },
    {
      "entropy": 6.0495306015014645,
      "epoch": 2.423752512453028,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004410797777557184,
      "loss": 5.6822,
      "mean_token_accuracy": 0.17306947112083435,
      "num_tokens": 56205552.0,
      "step": 27735
    },
    {
      "entropy": 6.093865966796875,
      "epoch": 2.424189460805733,
      "grad_norm": 2.203125,
      "learning_rate": 0.00044105875314310374,
      "loss": 5.7219,
      "mean_token_accuracy": 0.16898130625486374,
      "num_tokens": 56215581.0,
      "step": 27740
    },
    {
      "entropy": 6.074378395080567,
      "epoch": 2.4246264091584373,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044103772534539505,
      "loss": 5.6712,
      "mean_token_accuracy": 0.17609635144472122,
      "num_tokens": 56225428.0,
      "step": 27745
    },
    {
      "entropy": 6.083625602722168,
      "epoch": 2.425063357511142,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004410166943629959,
      "loss": 5.7554,
      "mean_token_accuracy": 0.1624819055199623,
      "num_tokens": 56235284.0,
      "step": 27750
    },
    {
      "entropy": 6.078255653381348,
      "epoch": 2.4255003058638467,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004409956601963095,
      "loss": 5.6923,
      "mean_token_accuracy": 0.17330564558506012,
      "num_tokens": 56244444.0,
      "step": 27755
    },
    {
      "entropy": 6.097892951965332,
      "epoch": 2.4259372542165516,
      "grad_norm": 3.75,
      "learning_rate": 0.0004409746228457393,
      "loss": 5.745,
      "mean_token_accuracy": 0.17343166023492812,
      "num_tokens": 56254983.0,
      "step": 27760
    },
    {
      "entropy": 6.115110254287719,
      "epoch": 2.4263742025692565,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004409535823116889,
      "loss": 5.7289,
      "mean_token_accuracy": 0.17281022816896438,
      "num_tokens": 56264307.0,
      "step": 27765
    },
    {
      "entropy": 6.120898580551147,
      "epoch": 2.426811150921961,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004409325385945619,
      "loss": 5.7591,
      "mean_token_accuracy": 0.16034931242465972,
      "num_tokens": 56274418.0,
      "step": 27770
    },
    {
      "entropy": 6.220321369171143,
      "epoch": 2.427248099274666,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004409114916947618,
      "loss": 5.818,
      "mean_token_accuracy": 0.15898640751838683,
      "num_tokens": 56285665.0,
      "step": 27775
    },
    {
      "entropy": 6.126898384094238,
      "epoch": 2.4276850476273704,
      "grad_norm": 1.609375,
      "learning_rate": 0.00044089044161269237,
      "loss": 5.7325,
      "mean_token_accuracy": 0.16972580999135972,
      "num_tokens": 56296320.0,
      "step": 27780
    },
    {
      "entropy": 6.105062627792359,
      "epoch": 2.4281219959800753,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004408693883487574,
      "loss": 5.8248,
      "mean_token_accuracy": 0.1629883423447609,
      "num_tokens": 56308009.0,
      "step": 27785
    },
    {
      "entropy": 6.078344106674194,
      "epoch": 2.4285589443327797,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004408483319033606,
      "loss": 5.7431,
      "mean_token_accuracy": 0.16607182770967482,
      "num_tokens": 56318341.0,
      "step": 27790
    },
    {
      "entropy": 6.10351300239563,
      "epoch": 2.4289958926854847,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004408272722769059,
      "loss": 5.6854,
      "mean_token_accuracy": 0.17496393024921417,
      "num_tokens": 56328808.0,
      "step": 27795
    },
    {
      "entropy": 6.168734312057495,
      "epoch": 2.429432841038189,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004408062094697972,
      "loss": 5.7634,
      "mean_token_accuracy": 0.1673481807112694,
      "num_tokens": 56339043.0,
      "step": 27800
    },
    {
      "entropy": 6.050416088104248,
      "epoch": 2.429869789390894,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00044078514348243846,
      "loss": 5.7291,
      "mean_token_accuracy": 0.16561201214790344,
      "num_tokens": 56349518.0,
      "step": 27805
    },
    {
      "entropy": 6.078094720840454,
      "epoch": 2.4303067377435985,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044076407431523394,
      "loss": 5.6882,
      "mean_token_accuracy": 0.16243896931409835,
      "num_tokens": 56358739.0,
      "step": 27810
    },
    {
      "entropy": 6.136951160430908,
      "epoch": 2.4307436860963034,
      "grad_norm": 1.625,
      "learning_rate": 0.0004407430019685875,
      "loss": 5.7321,
      "mean_token_accuracy": 0.1687646195292473,
      "num_tokens": 56368049.0,
      "step": 27815
    },
    {
      "entropy": 6.141361808776855,
      "epoch": 2.4311806344490083,
      "grad_norm": 1.96875,
      "learning_rate": 0.00044072192644290355,
      "loss": 5.7314,
      "mean_token_accuracy": 0.1621536687016487,
      "num_tokens": 56377607.0,
      "step": 27820
    },
    {
      "entropy": 6.0767927169799805,
      "epoch": 2.431617582801713,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004407008477385862,
      "loss": 5.7274,
      "mean_token_accuracy": 0.170142824947834,
      "num_tokens": 56387188.0,
      "step": 27825
    },
    {
      "entropy": 6.112976884841919,
      "epoch": 2.4320545311544177,
      "grad_norm": 2.90625,
      "learning_rate": 0.00044067976585603964,
      "loss": 5.8329,
      "mean_token_accuracy": 0.15736746340990065,
      "num_tokens": 56396389.0,
      "step": 27830
    },
    {
      "entropy": 6.138528823852539,
      "epoch": 2.432491479507122,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004406586807956684,
      "loss": 5.7877,
      "mean_token_accuracy": 0.1647431567311287,
      "num_tokens": 56406909.0,
      "step": 27835
    },
    {
      "entropy": 6.1199195861816404,
      "epoch": 2.432928427859827,
      "grad_norm": 1.796875,
      "learning_rate": 0.00044063759255787686,
      "loss": 5.7498,
      "mean_token_accuracy": 0.16339132636785508,
      "num_tokens": 56416373.0,
      "step": 27840
    },
    {
      "entropy": 6.181586456298828,
      "epoch": 2.4333653762125316,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044061650114306954,
      "loss": 5.8065,
      "mean_token_accuracy": 0.1613627091050148,
      "num_tokens": 56426759.0,
      "step": 27845
    },
    {
      "entropy": 6.136765193939209,
      "epoch": 2.4338023245652365,
      "grad_norm": 2.34375,
      "learning_rate": 0.00044059540655165097,
      "loss": 5.7776,
      "mean_token_accuracy": 0.16385326236486436,
      "num_tokens": 56436783.0,
      "step": 27850
    },
    {
      "entropy": 6.107071828842163,
      "epoch": 2.434239272917941,
      "grad_norm": 1.609375,
      "learning_rate": 0.00044057430878402564,
      "loss": 5.7337,
      "mean_token_accuracy": 0.160229854285717,
      "num_tokens": 56447191.0,
      "step": 27855
    },
    {
      "entropy": 6.05881667137146,
      "epoch": 2.434676221270646,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004405532078405984,
      "loss": 5.8009,
      "mean_token_accuracy": 0.15853200256824493,
      "num_tokens": 56457119.0,
      "step": 27860
    },
    {
      "entropy": 6.186550188064575,
      "epoch": 2.4351131696233503,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004405321037217739,
      "loss": 5.6877,
      "mean_token_accuracy": 0.17259850054979325,
      "num_tokens": 56467139.0,
      "step": 27865
    },
    {
      "entropy": 6.093527364730835,
      "epoch": 2.4355501179760553,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004405109964279568,
      "loss": 5.7339,
      "mean_token_accuracy": 0.16499125063419343,
      "num_tokens": 56477188.0,
      "step": 27870
    },
    {
      "entropy": 6.138102340698242,
      "epoch": 2.43598706632876,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044048988595955216,
      "loss": 5.8675,
      "mean_token_accuracy": 0.1571480005979538,
      "num_tokens": 56487594.0,
      "step": 27875
    },
    {
      "entropy": 6.162228155136108,
      "epoch": 2.4364240146814646,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004404687723169647,
      "loss": 5.7643,
      "mean_token_accuracy": 0.16692930161952974,
      "num_tokens": 56498008.0,
      "step": 27880
    },
    {
      "entropy": 6.056178283691406,
      "epoch": 2.4368609630341695,
      "grad_norm": 1.984375,
      "learning_rate": 0.00044044765550059955,
      "loss": 5.6687,
      "mean_token_accuracy": 0.17507675886154175,
      "num_tokens": 56508409.0,
      "step": 27885
    },
    {
      "entropy": 6.097272825241089,
      "epoch": 2.437297911386874,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004404265355108616,
      "loss": 5.7606,
      "mean_token_accuracy": 0.16361431479454042,
      "num_tokens": 56519513.0,
      "step": 27890
    },
    {
      "entropy": 6.132414436340332,
      "epoch": 2.437734859739579,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004404054123481561,
      "loss": 5.6878,
      "mean_token_accuracy": 0.16918692290782927,
      "num_tokens": 56528549.0,
      "step": 27895
    },
    {
      "entropy": 6.073365497589111,
      "epoch": 2.4381718080922834,
      "grad_norm": 2.34375,
      "learning_rate": 0.00044038428601288817,
      "loss": 5.7231,
      "mean_token_accuracy": 0.16958190947771073,
      "num_tokens": 56537859.0,
      "step": 27900
    },
    {
      "entropy": 6.112896156311035,
      "epoch": 2.4386087564449883,
      "grad_norm": 1.6875,
      "learning_rate": 0.00044036315650546287,
      "loss": 5.7668,
      "mean_token_accuracy": 0.15916139781475067,
      "num_tokens": 56547326.0,
      "step": 27905
    },
    {
      "entropy": 6.154233360290528,
      "epoch": 2.439045704797693,
      "grad_norm": 1.609375,
      "learning_rate": 0.00044034202382628565,
      "loss": 5.7966,
      "mean_token_accuracy": 0.16218268424272536,
      "num_tokens": 56557460.0,
      "step": 27910
    },
    {
      "entropy": 6.078154706954956,
      "epoch": 2.4394826531503977,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004403208879757617,
      "loss": 5.706,
      "mean_token_accuracy": 0.17163472771644592,
      "num_tokens": 56566928.0,
      "step": 27915
    },
    {
      "entropy": 6.005050563812256,
      "epoch": 2.439919601503102,
      "grad_norm": 1.875,
      "learning_rate": 0.0004402997489542966,
      "loss": 5.6784,
      "mean_token_accuracy": 0.18006877452135087,
      "num_tokens": 56577576.0,
      "step": 27920
    },
    {
      "entropy": 6.122775745391846,
      "epoch": 2.440356549855807,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004402786067622955,
      "loss": 5.8156,
      "mean_token_accuracy": 0.1612702265381813,
      "num_tokens": 56587744.0,
      "step": 27925
    },
    {
      "entropy": 6.121011972427368,
      "epoch": 2.440793498208512,
      "grad_norm": 1.625,
      "learning_rate": 0.0004402574614001643,
      "loss": 5.7629,
      "mean_token_accuracy": 0.163192917406559,
      "num_tokens": 56598272.0,
      "step": 27930
    },
    {
      "entropy": 6.088539886474609,
      "epoch": 2.4412304465612165,
      "grad_norm": 1.703125,
      "learning_rate": 0.00044023631286830835,
      "loss": 5.6993,
      "mean_token_accuracy": 0.16928116381168365,
      "num_tokens": 56609346.0,
      "step": 27935
    },
    {
      "entropy": 6.163487482070923,
      "epoch": 2.441667394913921,
      "grad_norm": 1.375,
      "learning_rate": 0.00044021516116713334,
      "loss": 5.7048,
      "mean_token_accuracy": 0.16798498183488847,
      "num_tokens": 56619210.0,
      "step": 27940
    },
    {
      "entropy": 6.0698143482208256,
      "epoch": 2.442104343266626,
      "grad_norm": 1.4375,
      "learning_rate": 0.000440194006297045,
      "loss": 5.7603,
      "mean_token_accuracy": 0.16535133123397827,
      "num_tokens": 56629135.0,
      "step": 27945
    },
    {
      "entropy": 6.120367527008057,
      "epoch": 2.4425412916193308,
      "grad_norm": 2.3125,
      "learning_rate": 0.000440172848258449,
      "loss": 5.7322,
      "mean_token_accuracy": 0.16741303205490113,
      "num_tokens": 56638900.0,
      "step": 27950
    },
    {
      "entropy": 6.141907835006714,
      "epoch": 2.4429782399720352,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044015168705175117,
      "loss": 5.7564,
      "mean_token_accuracy": 0.1647174373269081,
      "num_tokens": 56648605.0,
      "step": 27955
    },
    {
      "entropy": 6.0952308177948,
      "epoch": 2.44341518832474,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004401305226773576,
      "loss": 5.8026,
      "mean_token_accuracy": 0.16661955118179322,
      "num_tokens": 56659556.0,
      "step": 27960
    },
    {
      "entropy": 6.21298360824585,
      "epoch": 2.4438521366774446,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004401093551356739,
      "loss": 5.8605,
      "mean_token_accuracy": 0.15764699131250381,
      "num_tokens": 56669727.0,
      "step": 27965
    },
    {
      "entropy": 6.190178346633911,
      "epoch": 2.4442890850301495,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004400881844271064,
      "loss": 5.8116,
      "mean_token_accuracy": 0.15887829810380935,
      "num_tokens": 56681308.0,
      "step": 27970
    },
    {
      "entropy": 6.134395790100098,
      "epoch": 2.444726033382854,
      "grad_norm": 2.328125,
      "learning_rate": 0.00044006701055206095,
      "loss": 5.7683,
      "mean_token_accuracy": 0.1634887635707855,
      "num_tokens": 56691089.0,
      "step": 27975
    },
    {
      "entropy": 6.075980520248413,
      "epoch": 2.445162981735559,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004400458335109437,
      "loss": 5.7111,
      "mean_token_accuracy": 0.16951862573623658,
      "num_tokens": 56701599.0,
      "step": 27980
    },
    {
      "entropy": 6.043613433837891,
      "epoch": 2.4455999300882634,
      "grad_norm": 1.453125,
      "learning_rate": 0.00044002465330416095,
      "loss": 5.6176,
      "mean_token_accuracy": 0.17849229723215104,
      "num_tokens": 56711890.0,
      "step": 27985
    },
    {
      "entropy": 6.081659126281738,
      "epoch": 2.4460368784409683,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004400034699321189,
      "loss": 5.6942,
      "mean_token_accuracy": 0.1723373517394066,
      "num_tokens": 56722003.0,
      "step": 27990
    },
    {
      "entropy": 6.123232412338257,
      "epoch": 2.4464738267936728,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004399822833952238,
      "loss": 5.6469,
      "mean_token_accuracy": 0.16956235617399215,
      "num_tokens": 56731636.0,
      "step": 27995
    },
    {
      "entropy": 6.00853853225708,
      "epoch": 2.4469107751463777,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000439961093693882,
      "loss": 5.6564,
      "mean_token_accuracy": 0.171569761633873,
      "num_tokens": 56741973.0,
      "step": 28000
    },
    {
      "entropy": 6.1667595386505125,
      "epoch": 2.4473477234990826,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004399399008285001,
      "loss": 5.8027,
      "mean_token_accuracy": 0.16097283959388733,
      "num_tokens": 56751441.0,
      "step": 28005
    },
    {
      "entropy": 6.1545384407043455,
      "epoch": 2.447784671851787,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004399187047994844,
      "loss": 5.7477,
      "mean_token_accuracy": 0.16659095883369446,
      "num_tokens": 56761156.0,
      "step": 28010
    },
    {
      "entropy": 6.121378707885742,
      "epoch": 2.448221620204492,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043989750560724153,
      "loss": 5.7654,
      "mean_token_accuracy": 0.16000129133462906,
      "num_tokens": 56771414.0,
      "step": 28015
    },
    {
      "entropy": 5.98272008895874,
      "epoch": 2.4486585685571964,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004398763032521781,
      "loss": 5.6184,
      "mean_token_accuracy": 0.18304491192102432,
      "num_tokens": 56781572.0,
      "step": 28020
    },
    {
      "entropy": 6.117282104492188,
      "epoch": 2.4490955169099013,
      "grad_norm": 1.765625,
      "learning_rate": 0.00043985509773470077,
      "loss": 5.7866,
      "mean_token_accuracy": 0.15214873775839804,
      "num_tokens": 56791299.0,
      "step": 28025
    },
    {
      "entropy": 6.12339563369751,
      "epoch": 2.449532465262606,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004398338890552163,
      "loss": 5.7157,
      "mean_token_accuracy": 0.16372289508581161,
      "num_tokens": 56802485.0,
      "step": 28030
    },
    {
      "entropy": 6.12536211013794,
      "epoch": 2.4499694136153107,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004398126772141315,
      "loss": 5.7683,
      "mean_token_accuracy": 0.16079675406217575,
      "num_tokens": 56811694.0,
      "step": 28035
    },
    {
      "entropy": 6.047839546203614,
      "epoch": 2.450406361968015,
      "grad_norm": 2.359375,
      "learning_rate": 0.00043979146221185313,
      "loss": 5.672,
      "mean_token_accuracy": 0.16791390031576156,
      "num_tokens": 56821814.0,
      "step": 28040
    },
    {
      "entropy": 6.127837753295898,
      "epoch": 2.45084331032072,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004397702440487882,
      "loss": 5.7467,
      "mean_token_accuracy": 0.16463910788297653,
      "num_tokens": 56832257.0,
      "step": 28045
    },
    {
      "entropy": 6.117665863037109,
      "epoch": 2.4512802586734246,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004397490227253436,
      "loss": 5.7635,
      "mean_token_accuracy": 0.16537929624319075,
      "num_tokens": 56841789.0,
      "step": 28050
    },
    {
      "entropy": 6.070604562759399,
      "epoch": 2.4517172070261295,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004397277982419265,
      "loss": 5.741,
      "mean_token_accuracy": 0.1682926282286644,
      "num_tokens": 56851612.0,
      "step": 28055
    },
    {
      "entropy": 6.116510534286499,
      "epoch": 2.4521541553788344,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043970657059894385,
      "loss": 5.6829,
      "mean_token_accuracy": 0.17126362770795822,
      "num_tokens": 56861812.0,
      "step": 28060
    },
    {
      "entropy": 6.176416730880737,
      "epoch": 2.452591103731539,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000439685339796803,
      "loss": 5.6983,
      "mean_token_accuracy": 0.17379953265190123,
      "num_tokens": 56871873.0,
      "step": 28065
    },
    {
      "entropy": 6.1187786102294925,
      "epoch": 2.453028052084244,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004396641058359109,
      "loss": 5.7742,
      "mean_token_accuracy": 0.1603071480989456,
      "num_tokens": 56883579.0,
      "step": 28070
    },
    {
      "entropy": 6.038225746154785,
      "epoch": 2.4534650004369483,
      "grad_norm": 2.171875,
      "learning_rate": 0.00043964286871667495,
      "loss": 5.7238,
      "mean_token_accuracy": 0.16422489434480667,
      "num_tokens": 56894305.0,
      "step": 28075
    },
    {
      "entropy": 6.121295738220215,
      "epoch": 2.453901948789653,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004396216284395026,
      "loss": 5.7194,
      "mean_token_accuracy": 0.1653195470571518,
      "num_tokens": 56904393.0,
      "step": 28080
    },
    {
      "entropy": 6.158559370040893,
      "epoch": 2.4543388971423576,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004396003850048011,
      "loss": 5.7317,
      "mean_token_accuracy": 0.16900288462638854,
      "num_tokens": 56914600.0,
      "step": 28085
    },
    {
      "entropy": 6.110575485229492,
      "epoch": 2.4547758454950626,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004395791384129779,
      "loss": 5.7986,
      "mean_token_accuracy": 0.16839957982301712,
      "num_tokens": 56925948.0,
      "step": 28090
    },
    {
      "entropy": 6.081804275512695,
      "epoch": 2.455212793847767,
      "grad_norm": 1.53125,
      "learning_rate": 0.00043955788866444073,
      "loss": 5.6766,
      "mean_token_accuracy": 0.16526653617620468,
      "num_tokens": 56935395.0,
      "step": 28095
    },
    {
      "entropy": 6.151046085357666,
      "epoch": 2.455649742200472,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00043953663575959696,
      "loss": 5.67,
      "mean_token_accuracy": 0.17165251225233077,
      "num_tokens": 56944769.0,
      "step": 28100
    },
    {
      "entropy": 6.017258405685425,
      "epoch": 2.4560866905531764,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004395153796988543,
      "loss": 5.6692,
      "mean_token_accuracy": 0.17158620655536652,
      "num_tokens": 56954352.0,
      "step": 28105
    },
    {
      "entropy": 6.031219816207885,
      "epoch": 2.4565236389058813,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004394941204826205,
      "loss": 5.6798,
      "mean_token_accuracy": 0.16519540846347808,
      "num_tokens": 56963759.0,
      "step": 28110
    },
    {
      "entropy": 6.124006509780884,
      "epoch": 2.4569605872585862,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00043947285811130315,
      "loss": 5.6837,
      "mean_token_accuracy": 0.1698813855648041,
      "num_tokens": 56974962.0,
      "step": 28115
    },
    {
      "entropy": 6.136958980560303,
      "epoch": 2.4573975356112907,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00043945159258531035,
      "loss": 5.764,
      "mean_token_accuracy": 0.16496959328651428,
      "num_tokens": 56985637.0,
      "step": 28120
    },
    {
      "entropy": 6.105902576446534,
      "epoch": 2.4578344839639956,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004394303239050498,
      "loss": 5.7341,
      "mean_token_accuracy": 0.16837209165096284,
      "num_tokens": 56995779.0,
      "step": 28125
    },
    {
      "entropy": 6.114929676055908,
      "epoch": 2.4582714323167,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004394090520709294,
      "loss": 5.6886,
      "mean_token_accuracy": 0.17130339443683623,
      "num_tokens": 57005506.0,
      "step": 28130
    },
    {
      "entropy": 6.080328559875488,
      "epoch": 2.458708380669405,
      "grad_norm": 1.859375,
      "learning_rate": 0.00043938777708335726,
      "loss": 5.7871,
      "mean_token_accuracy": 0.16083515733480452,
      "num_tokens": 57016001.0,
      "step": 28135
    },
    {
      "entropy": 6.057065439224243,
      "epoch": 2.4591453290221095,
      "grad_norm": 2.109375,
      "learning_rate": 0.00043936649894274146,
      "loss": 5.7466,
      "mean_token_accuracy": 0.16990187019109726,
      "num_tokens": 57026418.0,
      "step": 28140
    },
    {
      "entropy": 6.200252199172974,
      "epoch": 2.4595822773748144,
      "grad_norm": 2.46875,
      "learning_rate": 0.00043934521764949,
      "loss": 5.7514,
      "mean_token_accuracy": 0.17046775072813034,
      "num_tokens": 57037170.0,
      "step": 28145
    },
    {
      "entropy": 6.127634429931641,
      "epoch": 2.460019225727519,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004393239332040113,
      "loss": 5.6923,
      "mean_token_accuracy": 0.17315335124731063,
      "num_tokens": 57046775.0,
      "step": 28150
    },
    {
      "entropy": 6.055511856079102,
      "epoch": 2.4604561740802238,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043930264560671336,
      "loss": 5.7105,
      "mean_token_accuracy": 0.16804729551076888,
      "num_tokens": 57056360.0,
      "step": 28155
    },
    {
      "entropy": 6.082494640350342,
      "epoch": 2.4608931224329282,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004392813548580045,
      "loss": 5.6922,
      "mean_token_accuracy": 0.16672088205814362,
      "num_tokens": 57066355.0,
      "step": 28160
    },
    {
      "entropy": 6.066829061508178,
      "epoch": 2.461330070785633,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004392600609582933,
      "loss": 5.7336,
      "mean_token_accuracy": 0.17046307772397995,
      "num_tokens": 57077158.0,
      "step": 28165
    },
    {
      "entropy": 6.163929033279419,
      "epoch": 2.461767019138338,
      "grad_norm": 1.8125,
      "learning_rate": 0.000439238763907988,
      "loss": 5.7908,
      "mean_token_accuracy": 0.16573076844215393,
      "num_tokens": 57088005.0,
      "step": 28170
    },
    {
      "entropy": 6.098238563537597,
      "epoch": 2.4622039674910425,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004392174637074972,
      "loss": 5.7087,
      "mean_token_accuracy": 0.16778565794229508,
      "num_tokens": 57098399.0,
      "step": 28175
    },
    {
      "entropy": 6.086942195892334,
      "epoch": 2.4626409158437474,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004391961603572294,
      "loss": 5.7127,
      "mean_token_accuracy": 0.17017977982759475,
      "num_tokens": 57108949.0,
      "step": 28180
    },
    {
      "entropy": 6.097065448760986,
      "epoch": 2.463077864196452,
      "grad_norm": 2.625,
      "learning_rate": 0.00043917485385759316,
      "loss": 5.7467,
      "mean_token_accuracy": 0.16437918841838836,
      "num_tokens": 57119441.0,
      "step": 28185
    },
    {
      "entropy": 6.1488518714904785,
      "epoch": 2.463514812549157,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004391535442089971,
      "loss": 5.7211,
      "mean_token_accuracy": 0.17584417015314102,
      "num_tokens": 57129507.0,
      "step": 28190
    },
    {
      "entropy": 6.109136152267456,
      "epoch": 2.4639517609018613,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004391322314118502,
      "loss": 5.7206,
      "mean_token_accuracy": 0.1677968308329582,
      "num_tokens": 57140490.0,
      "step": 28195
    },
    {
      "entropy": 6.070234632492065,
      "epoch": 2.464388709254566,
      "grad_norm": 2.15625,
      "learning_rate": 0.00043911091546656105,
      "loss": 5.6489,
      "mean_token_accuracy": 0.17555813789367675,
      "num_tokens": 57150706.0,
      "step": 28200
    },
    {
      "entropy": 6.044542360305786,
      "epoch": 2.4648256576072707,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00043908959637353864,
      "loss": 5.682,
      "mean_token_accuracy": 0.17200292646884918,
      "num_tokens": 57160259.0,
      "step": 28205
    },
    {
      "entropy": 6.05194845199585,
      "epoch": 2.4652626059599756,
      "grad_norm": 1.515625,
      "learning_rate": 0.00043906827413319166,
      "loss": 5.675,
      "mean_token_accuracy": 0.17596423178911208,
      "num_tokens": 57171831.0,
      "step": 28210
    },
    {
      "entropy": 6.135050964355469,
      "epoch": 2.46569955431268,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004390469487459292,
      "loss": 5.7992,
      "mean_token_accuracy": 0.16078699976205826,
      "num_tokens": 57182172.0,
      "step": 28215
    },
    {
      "entropy": 6.055383729934692,
      "epoch": 2.466136502665385,
      "grad_norm": 1.5,
      "learning_rate": 0.00043902562021216033,
      "loss": 5.6966,
      "mean_token_accuracy": 0.17489081770181655,
      "num_tokens": 57191633.0,
      "step": 28220
    },
    {
      "entropy": 6.135259914398193,
      "epoch": 2.46657345101809,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004390042885322942,
      "loss": 5.7124,
      "mean_token_accuracy": 0.17090965211391448,
      "num_tokens": 57200953.0,
      "step": 28225
    },
    {
      "entropy": 6.161620426177978,
      "epoch": 2.4670103993707944,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00043898295370673986,
      "loss": 5.792,
      "mean_token_accuracy": 0.1630767658352852,
      "num_tokens": 57210893.0,
      "step": 28230
    },
    {
      "entropy": 6.046744775772095,
      "epoch": 2.4674473477234993,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00043896161573590646,
      "loss": 5.7156,
      "mean_token_accuracy": 0.17467929869890214,
      "num_tokens": 57221458.0,
      "step": 28235
    },
    {
      "entropy": 6.141603183746338,
      "epoch": 2.4678842960762037,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004389402746202034,
      "loss": 5.7802,
      "mean_token_accuracy": 0.16218274533748628,
      "num_tokens": 57231450.0,
      "step": 28240
    },
    {
      "entropy": 6.181050729751587,
      "epoch": 2.4683212444289087,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00043891893036004,
      "loss": 5.7447,
      "mean_token_accuracy": 0.16724757850170135,
      "num_tokens": 57242820.0,
      "step": 28245
    },
    {
      "entropy": 6.178207159042358,
      "epoch": 2.468758192781613,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004388975829558256,
      "loss": 5.7656,
      "mean_token_accuracy": 0.16079156398773192,
      "num_tokens": 57253465.0,
      "step": 28250
    },
    {
      "entropy": 6.067081069946289,
      "epoch": 2.469195141134318,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00043887623240796965,
      "loss": 5.7432,
      "mean_token_accuracy": 0.17009436339139938,
      "num_tokens": 57263521.0,
      "step": 28255
    },
    {
      "entropy": 6.130150318145752,
      "epoch": 2.4696320894870225,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004388548787168817,
      "loss": 5.7447,
      "mean_token_accuracy": 0.15751346498727797,
      "num_tokens": 57273568.0,
      "step": 28260
    },
    {
      "entropy": 6.156837511062622,
      "epoch": 2.4700690378397274,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004388335218829713,
      "loss": 5.7849,
      "mean_token_accuracy": 0.1688705116510391,
      "num_tokens": 57283960.0,
      "step": 28265
    },
    {
      "entropy": 6.084983539581299,
      "epoch": 2.470505986192432,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004388121619066482,
      "loss": 5.7302,
      "mean_token_accuracy": 0.1700523555278778,
      "num_tokens": 57294672.0,
      "step": 28270
    },
    {
      "entropy": 6.118184280395508,
      "epoch": 2.470942934545137,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004387907987883219,
      "loss": 5.7373,
      "mean_token_accuracy": 0.17126729786396028,
      "num_tokens": 57305594.0,
      "step": 28275
    },
    {
      "entropy": 6.123820972442627,
      "epoch": 2.4713798828978417,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004387694325284023,
      "loss": 5.6812,
      "mean_token_accuracy": 0.1782374322414398,
      "num_tokens": 57315250.0,
      "step": 28280
    },
    {
      "entropy": 6.168384552001953,
      "epoch": 2.471816831250546,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004387480631272991,
      "loss": 5.7487,
      "mean_token_accuracy": 0.16716889590024947,
      "num_tokens": 57326750.0,
      "step": 28285
    },
    {
      "entropy": 6.11349139213562,
      "epoch": 2.4722537796032507,
      "grad_norm": 1.5,
      "learning_rate": 0.0004387266905854223,
      "loss": 5.7102,
      "mean_token_accuracy": 0.16442713886499405,
      "num_tokens": 57336151.0,
      "step": 28290
    },
    {
      "entropy": 6.090876150131225,
      "epoch": 2.4726907279559556,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004387053149031819,
      "loss": 5.6988,
      "mean_token_accuracy": 0.16922986805438994,
      "num_tokens": 57346761.0,
      "step": 28295
    },
    {
      "entropy": 6.100080156326294,
      "epoch": 2.4731276763086605,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004386839360809876,
      "loss": 5.7451,
      "mean_token_accuracy": 0.16896583288908004,
      "num_tokens": 57356654.0,
      "step": 28300
    },
    {
      "entropy": 6.141951227188111,
      "epoch": 2.473564624661365,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004386625541192497,
      "loss": 5.7738,
      "mean_token_accuracy": 0.164067080616951,
      "num_tokens": 57367888.0,
      "step": 28305
    },
    {
      "entropy": 6.083452796936035,
      "epoch": 2.47400157301407,
      "grad_norm": 1.625,
      "learning_rate": 0.00043864116901837826,
      "loss": 5.6944,
      "mean_token_accuracy": 0.17601428627967836,
      "num_tokens": 57377167.0,
      "step": 28310
    },
    {
      "entropy": 6.029130458831787,
      "epoch": 2.4744385213667743,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00043861978077878343,
      "loss": 5.6904,
      "mean_token_accuracy": 0.16934041827917098,
      "num_tokens": 57389076.0,
      "step": 28315
    },
    {
      "entropy": 6.154650640487671,
      "epoch": 2.4748754697194792,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004385983894008755,
      "loss": 5.7342,
      "mean_token_accuracy": 0.16596973687410355,
      "num_tokens": 57401136.0,
      "step": 28320
    },
    {
      "entropy": 6.094171619415283,
      "epoch": 2.4753124180721837,
      "grad_norm": 2.28125,
      "learning_rate": 0.00043857699488506467,
      "loss": 5.7065,
      "mean_token_accuracy": 0.16650061756372453,
      "num_tokens": 57410548.0,
      "step": 28325
    },
    {
      "entropy": 6.0894958019256595,
      "epoch": 2.4757493664248886,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004385555972317615,
      "loss": 5.7567,
      "mean_token_accuracy": 0.16345023810863496,
      "num_tokens": 57420371.0,
      "step": 28330
    },
    {
      "entropy": 6.182074403762817,
      "epoch": 2.476186314777593,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043853419644137616,
      "loss": 5.7655,
      "mean_token_accuracy": 0.1723088577389717,
      "num_tokens": 57430079.0,
      "step": 28335
    },
    {
      "entropy": 6.111065101623535,
      "epoch": 2.476623263130298,
      "grad_norm": 4.40625,
      "learning_rate": 0.00043851279251431937,
      "loss": 5.7238,
      "mean_token_accuracy": 0.17048347294330596,
      "num_tokens": 57440763.0,
      "step": 28340
    },
    {
      "entropy": 6.197044134140015,
      "epoch": 2.4770602114830025,
      "grad_norm": 2.515625,
      "learning_rate": 0.00043849138545100136,
      "loss": 5.8116,
      "mean_token_accuracy": 0.16417123526334762,
      "num_tokens": 57451008.0,
      "step": 28345
    },
    {
      "entropy": 6.157305765151977,
      "epoch": 2.4774971598357074,
      "grad_norm": 2.5,
      "learning_rate": 0.00043846997525183305,
      "loss": 5.775,
      "mean_token_accuracy": 0.16130539178848266,
      "num_tokens": 57460852.0,
      "step": 28350
    },
    {
      "entropy": 6.140087270736695,
      "epoch": 2.4779341081884123,
      "grad_norm": 1.59375,
      "learning_rate": 0.00043844856191722486,
      "loss": 5.755,
      "mean_token_accuracy": 0.16005523800849913,
      "num_tokens": 57470027.0,
      "step": 28355
    },
    {
      "entropy": 6.155585479736328,
      "epoch": 2.478371056541117,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043842714544758764,
      "loss": 5.6504,
      "mean_token_accuracy": 0.16650094613432884,
      "num_tokens": 57479846.0,
      "step": 28360
    },
    {
      "entropy": 6.1079222679138185,
      "epoch": 2.4788080048938217,
      "grad_norm": 2.078125,
      "learning_rate": 0.00043840572584333216,
      "loss": 5.7862,
      "mean_token_accuracy": 0.16217874884605407,
      "num_tokens": 57490569.0,
      "step": 28365
    },
    {
      "entropy": 6.086313915252686,
      "epoch": 2.479244953246526,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004383843031048692,
      "loss": 5.743,
      "mean_token_accuracy": 0.16363697797060012,
      "num_tokens": 57501048.0,
      "step": 28370
    },
    {
      "entropy": 6.134778356552124,
      "epoch": 2.479681901599231,
      "grad_norm": 1.953125,
      "learning_rate": 0.00043836287723260977,
      "loss": 5.8023,
      "mean_token_accuracy": 0.1582293003797531,
      "num_tokens": 57511510.0,
      "step": 28375
    },
    {
      "entropy": 6.103812122344971,
      "epoch": 2.4801188499519355,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00043834144822696463,
      "loss": 5.7528,
      "mean_token_accuracy": 0.16453388035297395,
      "num_tokens": 57522229.0,
      "step": 28380
    },
    {
      "entropy": 6.093504524230957,
      "epoch": 2.4805557983046405,
      "grad_norm": 2.5625,
      "learning_rate": 0.000438320016088345,
      "loss": 5.6914,
      "mean_token_accuracy": 0.1629055991768837,
      "num_tokens": 57532618.0,
      "step": 28385
    },
    {
      "entropy": 6.123232221603393,
      "epoch": 2.480992746657345,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004382985808171619,
      "loss": 5.7592,
      "mean_token_accuracy": 0.16773559004068375,
      "num_tokens": 57542698.0,
      "step": 28390
    },
    {
      "entropy": 6.068953037261963,
      "epoch": 2.48142969501005,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043827714241382644,
      "loss": 5.6292,
      "mean_token_accuracy": 0.16906940340995788,
      "num_tokens": 57552556.0,
      "step": 28395
    },
    {
      "entropy": 6.180788946151734,
      "epoch": 2.4818666433627543,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004382557008787498,
      "loss": 5.7785,
      "mean_token_accuracy": 0.1634097456932068,
      "num_tokens": 57562830.0,
      "step": 28400
    },
    {
      "entropy": 6.178267145156861,
      "epoch": 2.4823035917154592,
      "grad_norm": 1.875,
      "learning_rate": 0.0004382342562123433,
      "loss": 5.7146,
      "mean_token_accuracy": 0.1683570072054863,
      "num_tokens": 57572584.0,
      "step": 28405
    },
    {
      "entropy": 6.046517658233642,
      "epoch": 2.482740540068164,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004382128084150182,
      "loss": 5.7006,
      "mean_token_accuracy": 0.1702185571193695,
      "num_tokens": 57582571.0,
      "step": 28410
    },
    {
      "entropy": 6.089315557479859,
      "epoch": 2.4831774884208686,
      "grad_norm": 2.359375,
      "learning_rate": 0.00043819135748718595,
      "loss": 5.7664,
      "mean_token_accuracy": 0.16789860129356385,
      "num_tokens": 57593267.0,
      "step": 28415
    },
    {
      "entropy": 6.11922607421875,
      "epoch": 2.4836144367735735,
      "grad_norm": 1.59375,
      "learning_rate": 0.00043816990342925797,
      "loss": 5.7611,
      "mean_token_accuracy": 0.1635598883032799,
      "num_tokens": 57603708.0,
      "step": 28420
    },
    {
      "entropy": 6.143654012680054,
      "epoch": 2.484051385126278,
      "grad_norm": 1.625,
      "learning_rate": 0.00043814844624164576,
      "loss": 5.801,
      "mean_token_accuracy": 0.15858690589666366,
      "num_tokens": 57613523.0,
      "step": 28425
    },
    {
      "entropy": 6.098622798919678,
      "epoch": 2.484488333478983,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004381269859247607,
      "loss": 5.7224,
      "mean_token_accuracy": 0.17315996140241624,
      "num_tokens": 57624181.0,
      "step": 28430
    },
    {
      "entropy": 6.1033101081848145,
      "epoch": 2.4849252818316874,
      "grad_norm": 2.96875,
      "learning_rate": 0.00043810552247901475,
      "loss": 5.6803,
      "mean_token_accuracy": 0.1660487100481987,
      "num_tokens": 57633674.0,
      "step": 28435
    },
    {
      "entropy": 6.1282511234283445,
      "epoch": 2.4853622301843923,
      "grad_norm": 1.59375,
      "learning_rate": 0.00043808405590481934,
      "loss": 5.7479,
      "mean_token_accuracy": 0.16916636377573013,
      "num_tokens": 57644723.0,
      "step": 28440
    },
    {
      "entropy": 6.121088886260987,
      "epoch": 2.4857991785370968,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004380625862025863,
      "loss": 5.6967,
      "mean_token_accuracy": 0.16926486939191818,
      "num_tokens": 57654274.0,
      "step": 28445
    },
    {
      "entropy": 6.00963339805603,
      "epoch": 2.4862361268898017,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004380411133727274,
      "loss": 5.681,
      "mean_token_accuracy": 0.16458612978458403,
      "num_tokens": 57664572.0,
      "step": 28450
    },
    {
      "entropy": 6.088610982894897,
      "epoch": 2.486673075242506,
      "grad_norm": 1.734375,
      "learning_rate": 0.00043801963741565454,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1562821686267853,
      "num_tokens": 57674048.0,
      "step": 28455
    },
    {
      "entropy": 6.2326360702514645,
      "epoch": 2.487110023595211,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004379981583317796,
      "loss": 5.7692,
      "mean_token_accuracy": 0.1606403797864914,
      "num_tokens": 57685488.0,
      "step": 28460
    },
    {
      "entropy": 6.141277933120728,
      "epoch": 2.487546971947916,
      "grad_norm": 2.140625,
      "learning_rate": 0.00043797667612151454,
      "loss": 5.7467,
      "mean_token_accuracy": 0.16577474772930145,
      "num_tokens": 57696044.0,
      "step": 28465
    },
    {
      "entropy": 6.062156438827515,
      "epoch": 2.4879839203006204,
      "grad_norm": 2.09375,
      "learning_rate": 0.00043795519078527144,
      "loss": 5.7098,
      "mean_token_accuracy": 0.17400626838207245,
      "num_tokens": 57705702.0,
      "step": 28470
    },
    {
      "entropy": 6.143585968017578,
      "epoch": 2.4884208686533253,
      "grad_norm": 4.15625,
      "learning_rate": 0.00043793370232346246,
      "loss": 5.8008,
      "mean_token_accuracy": 0.16136892586946489,
      "num_tokens": 57715628.0,
      "step": 28475
    },
    {
      "entropy": 6.16754846572876,
      "epoch": 2.48885781700603,
      "grad_norm": 3.28125,
      "learning_rate": 0.00043791221073649964,
      "loss": 5.7535,
      "mean_token_accuracy": 0.16332018822431565,
      "num_tokens": 57725319.0,
      "step": 28480
    },
    {
      "entropy": 6.063836097717285,
      "epoch": 2.4892947653587347,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00043789071602479525,
      "loss": 5.7672,
      "mean_token_accuracy": 0.16198882311582566,
      "num_tokens": 57736483.0,
      "step": 28485
    },
    {
      "entropy": 6.133048963546753,
      "epoch": 2.489731713711439,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004378692181887616,
      "loss": 5.7445,
      "mean_token_accuracy": 0.1614668533205986,
      "num_tokens": 57746667.0,
      "step": 28490
    },
    {
      "entropy": 6.076351547241211,
      "epoch": 2.490168662064144,
      "grad_norm": 2.015625,
      "learning_rate": 0.00043784771722881095,
      "loss": 5.6364,
      "mean_token_accuracy": 0.17929122149944304,
      "num_tokens": 57755971.0,
      "step": 28495
    },
    {
      "entropy": 6.049137687683105,
      "epoch": 2.4906056104168486,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004378262131453558,
      "loss": 5.6848,
      "mean_token_accuracy": 0.1699887827038765,
      "num_tokens": 57765858.0,
      "step": 28500
    },
    {
      "entropy": 6.097113943099975,
      "epoch": 2.4910425587695535,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004378047059388086,
      "loss": 5.6989,
      "mean_token_accuracy": 0.16596155762672424,
      "num_tokens": 57775154.0,
      "step": 28505
    },
    {
      "entropy": 6.0911133766174315,
      "epoch": 2.491479507122258,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004377831956095817,
      "loss": 5.6985,
      "mean_token_accuracy": 0.17340469360351562,
      "num_tokens": 57785416.0,
      "step": 28510
    },
    {
      "entropy": 6.098613834381103,
      "epoch": 2.491916455474963,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000437761682158088,
      "loss": 5.728,
      "mean_token_accuracy": 0.16896307617425918,
      "num_tokens": 57795483.0,
      "step": 28515
    },
    {
      "entropy": 6.086117219924927,
      "epoch": 2.492353403827668,
      "grad_norm": 2.765625,
      "learning_rate": 0.00043774016558473976,
      "loss": 5.7319,
      "mean_token_accuracy": 0.16873013824224473,
      "num_tokens": 57805772.0,
      "step": 28520
    },
    {
      "entropy": 6.09727201461792,
      "epoch": 2.4927903521803723,
      "grad_norm": 3.59375,
      "learning_rate": 0.00043771864588994995,
      "loss": 5.7546,
      "mean_token_accuracy": 0.1666936069726944,
      "num_tokens": 57815514.0,
      "step": 28525
    },
    {
      "entropy": 6.064947128295898,
      "epoch": 2.493227300533077,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004376971230741313,
      "loss": 5.6984,
      "mean_token_accuracy": 0.17340996265411376,
      "num_tokens": 57826312.0,
      "step": 28530
    },
    {
      "entropy": 6.110623931884765,
      "epoch": 2.4936642488857816,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004376755971376965,
      "loss": 5.7778,
      "mean_token_accuracy": 0.1642627239227295,
      "num_tokens": 57836718.0,
      "step": 28535
    },
    {
      "entropy": 6.157033538818359,
      "epoch": 2.4941011972384866,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004376540680810585,
      "loss": 5.7108,
      "mean_token_accuracy": 0.17148173749446868,
      "num_tokens": 57846833.0,
      "step": 28540
    },
    {
      "entropy": 6.170121479034424,
      "epoch": 2.494538145591191,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004376325359046304,
      "loss": 5.8126,
      "mean_token_accuracy": 0.15933283418416977,
      "num_tokens": 57856716.0,
      "step": 28545
    },
    {
      "entropy": 6.153274869918823,
      "epoch": 2.494975093943896,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043761100060882486,
      "loss": 5.7726,
      "mean_token_accuracy": 0.16213457882404328,
      "num_tokens": 57867079.0,
      "step": 28550
    },
    {
      "entropy": 6.099008464813233,
      "epoch": 2.4954120422966004,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004375894621940552,
      "loss": 5.7067,
      "mean_token_accuracy": 0.16774236857891084,
      "num_tokens": 57876593.0,
      "step": 28555
    },
    {
      "entropy": 6.080410051345825,
      "epoch": 2.4958489906493053,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004375679206607344,
      "loss": 5.7187,
      "mean_token_accuracy": 0.175180184841156,
      "num_tokens": 57886834.0,
      "step": 28560
    },
    {
      "entropy": 6.051687574386596,
      "epoch": 2.49628593900201,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004375463760092758,
      "loss": 5.6794,
      "mean_token_accuracy": 0.17484518885612488,
      "num_tokens": 57896766.0,
      "step": 28565
    },
    {
      "entropy": 6.1326884746551515,
      "epoch": 2.4967228873547147,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004375248282400925,
      "loss": 5.7122,
      "mean_token_accuracy": 0.16821678578853608,
      "num_tokens": 57907259.0,
      "step": 28570
    },
    {
      "entropy": 6.149142265319824,
      "epoch": 2.4971598357074196,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004375032773535977,
      "loss": 5.8378,
      "mean_token_accuracy": 0.15374866276979446,
      "num_tokens": 57917940.0,
      "step": 28575
    },
    {
      "entropy": 6.1699295997619625,
      "epoch": 2.497596784060124,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000437481723350205,
      "loss": 5.7104,
      "mean_token_accuracy": 0.1673874333500862,
      "num_tokens": 57927256.0,
      "step": 28580
    },
    {
      "entropy": 6.055751085281372,
      "epoch": 2.4980337324128286,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043746016623032776,
      "loss": 5.6966,
      "mean_token_accuracy": 0.1654731661081314,
      "num_tokens": 57936718.0,
      "step": 28585
    },
    {
      "entropy": 6.144385576248169,
      "epoch": 2.4984706807655335,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004374386059943794,
      "loss": 5.7523,
      "mean_token_accuracy": 0.16133656054735185,
      "num_tokens": 57947548.0,
      "step": 28590
    },
    {
      "entropy": 6.164455842971802,
      "epoch": 2.4989076291182384,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004374170426427734,
      "loss": 5.697,
      "mean_token_accuracy": 0.1707790121436119,
      "num_tokens": 57957598.0,
      "step": 28595
    },
    {
      "entropy": 6.063155317306519,
      "epoch": 2.499344577470943,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043739547617592344,
      "loss": 5.8032,
      "mean_token_accuracy": 0.16245734095573425,
      "num_tokens": 57967790.0,
      "step": 28600
    },
    {
      "entropy": 6.102785921096801,
      "epoch": 2.4997815258236478,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00043737390659424316,
      "loss": 5.7451,
      "mean_token_accuracy": 0.16992301046848296,
      "num_tokens": 57977904.0,
      "step": 28605
    },
    {
      "entropy": 6.098716402053833,
      "epoch": 2.5002184741763522,
      "grad_norm": 1.828125,
      "learning_rate": 0.00043735233389814623,
      "loss": 5.599,
      "mean_token_accuracy": 0.17997690439224243,
      "num_tokens": 57987882.0,
      "step": 28610
    },
    {
      "entropy": 5.982296562194824,
      "epoch": 2.500655422529057,
      "grad_norm": 1.609375,
      "learning_rate": 0.00043733075808804643,
      "loss": 5.723,
      "mean_token_accuracy": 0.1734093591570854,
      "num_tokens": 57997376.0,
      "step": 28615
    },
    {
      "entropy": 6.023700094223022,
      "epoch": 2.5010923708817616,
      "grad_norm": 2.59375,
      "learning_rate": 0.00043730917916435776,
      "loss": 5.6578,
      "mean_token_accuracy": 0.17524690926074982,
      "num_tokens": 58008115.0,
      "step": 28620
    },
    {
      "entropy": 6.17190990447998,
      "epoch": 2.5015293192344665,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004372875971274939,
      "loss": 5.8149,
      "mean_token_accuracy": 0.15404974669218063,
      "num_tokens": 58017922.0,
      "step": 28625
    },
    {
      "entropy": 6.1061193466186525,
      "epoch": 2.5019662675871714,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004372660119778689,
      "loss": 5.7455,
      "mean_token_accuracy": 0.16680692434310912,
      "num_tokens": 58028208.0,
      "step": 28630
    },
    {
      "entropy": 6.140908050537109,
      "epoch": 2.502403215939876,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004372444237158968,
      "loss": 5.7667,
      "mean_token_accuracy": 0.16259835064411163,
      "num_tokens": 58037237.0,
      "step": 28635
    },
    {
      "entropy": 6.112023735046387,
      "epoch": 2.5028401642925804,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004372228323419916,
      "loss": 5.7014,
      "mean_token_accuracy": 0.17374918460845948,
      "num_tokens": 58048083.0,
      "step": 28640
    },
    {
      "entropy": 6.102118110656738,
      "epoch": 2.5032771126452853,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004372012378565674,
      "loss": 5.7837,
      "mean_token_accuracy": 0.16065812855958939,
      "num_tokens": 58058593.0,
      "step": 28645
    },
    {
      "entropy": 6.089183044433594,
      "epoch": 2.50371406099799,
      "grad_norm": 1.875,
      "learning_rate": 0.00043717964026003856,
      "loss": 5.7493,
      "mean_token_accuracy": 0.17259377837181092,
      "num_tokens": 58068996.0,
      "step": 28650
    },
    {
      "entropy": 6.126375961303711,
      "epoch": 2.5041510093506947,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004371580395528192,
      "loss": 5.7098,
      "mean_token_accuracy": 0.1701649010181427,
      "num_tokens": 58078546.0,
      "step": 28655
    },
    {
      "entropy": 6.155629301071167,
      "epoch": 2.5045879577033996,
      "grad_norm": 1.703125,
      "learning_rate": 0.00043713643573532377,
      "loss": 5.7951,
      "mean_token_accuracy": 0.1568273723125458,
      "num_tokens": 58087824.0,
      "step": 28660
    },
    {
      "entropy": 6.11942138671875,
      "epoch": 2.505024906056104,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043711482880796643,
      "loss": 5.7859,
      "mean_token_accuracy": 0.167328342795372,
      "num_tokens": 58097940.0,
      "step": 28665
    },
    {
      "entropy": 6.146848392486572,
      "epoch": 2.505461854408809,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004370932187711618,
      "loss": 5.7759,
      "mean_token_accuracy": 0.1572338193655014,
      "num_tokens": 58109274.0,
      "step": 28670
    },
    {
      "entropy": 6.14003472328186,
      "epoch": 2.5058988027615134,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004370716056253242,
      "loss": 5.7197,
      "mean_token_accuracy": 0.1624738708138466,
      "num_tokens": 58119061.0,
      "step": 28675
    },
    {
      "entropy": 6.107754564285278,
      "epoch": 2.5063357511142184,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00043704998937086844,
      "loss": 5.7614,
      "mean_token_accuracy": 0.1602608934044838,
      "num_tokens": 58128949.0,
      "step": 28680
    },
    {
      "entropy": 6.098017978668213,
      "epoch": 2.5067726994669233,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004370283700082088,
      "loss": 5.6553,
      "mean_token_accuracy": 0.16889359056949615,
      "num_tokens": 58139807.0,
      "step": 28685
    },
    {
      "entropy": 6.084874296188355,
      "epoch": 2.5072096478196277,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004370067475377602,
      "loss": 5.7587,
      "mean_token_accuracy": 0.16640831977128984,
      "num_tokens": 58150878.0,
      "step": 28690
    },
    {
      "entropy": 6.040452289581299,
      "epoch": 2.507646596172332,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004369851219599373,
      "loss": 5.7335,
      "mean_token_accuracy": 0.17150481343269347,
      "num_tokens": 58160805.0,
      "step": 28695
    },
    {
      "entropy": 6.137100315093994,
      "epoch": 2.508083544525037,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00043696349327515484,
      "loss": 5.6435,
      "mean_token_accuracy": 0.17161210924386977,
      "num_tokens": 58170298.0,
      "step": 28700
    },
    {
      "entropy": 6.036894607543945,
      "epoch": 2.508520492877742,
      "grad_norm": 2.390625,
      "learning_rate": 0.00043694186148382773,
      "loss": 5.7075,
      "mean_token_accuracy": 0.17630729377269744,
      "num_tokens": 58180497.0,
      "step": 28705
    },
    {
      "entropy": 6.128305435180664,
      "epoch": 2.5089574412304465,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00043692022658637085,
      "loss": 5.7641,
      "mean_token_accuracy": 0.16782361567020415,
      "num_tokens": 58190871.0,
      "step": 28710
    },
    {
      "entropy": 6.100801229476929,
      "epoch": 2.5093943895831514,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00043689858858319923,
      "loss": 5.7369,
      "mean_token_accuracy": 0.168955235183239,
      "num_tokens": 58201654.0,
      "step": 28715
    },
    {
      "entropy": 6.1489029884338375,
      "epoch": 2.509831337935856,
      "grad_norm": 1.8125,
      "learning_rate": 0.00043687694747472785,
      "loss": 5.7617,
      "mean_token_accuracy": 0.16795263886451722,
      "num_tokens": 58211647.0,
      "step": 28720
    },
    {
      "entropy": 6.015167474746704,
      "epoch": 2.510268286288561,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043685530326137173,
      "loss": 5.7299,
      "mean_token_accuracy": 0.17271275371313094,
      "num_tokens": 58221945.0,
      "step": 28725
    },
    {
      "entropy": 6.1276257038116455,
      "epoch": 2.5107052346412653,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00043683365594354614,
      "loss": 5.7593,
      "mean_token_accuracy": 0.1707206681370735,
      "num_tokens": 58231702.0,
      "step": 28730
    },
    {
      "entropy": 6.158997631072998,
      "epoch": 2.51114218299397,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004368120055216662,
      "loss": 5.789,
      "mean_token_accuracy": 0.16149349212646485,
      "num_tokens": 58242822.0,
      "step": 28735
    },
    {
      "entropy": 6.101110029220581,
      "epoch": 2.511579131346675,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004367903519961472,
      "loss": 5.6931,
      "mean_token_accuracy": 0.17061678916215897,
      "num_tokens": 58253221.0,
      "step": 28740
    },
    {
      "entropy": 6.077698850631714,
      "epoch": 2.5120160796993796,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004367686953674045,
      "loss": 5.6949,
      "mean_token_accuracy": 0.17220184803009034,
      "num_tokens": 58263739.0,
      "step": 28745
    },
    {
      "entropy": 6.021663475036621,
      "epoch": 2.512453028052084,
      "grad_norm": 1.953125,
      "learning_rate": 0.00043674703563585345,
      "loss": 5.68,
      "mean_token_accuracy": 0.17394344061613082,
      "num_tokens": 58273526.0,
      "step": 28750
    },
    {
      "entropy": 6.049652576446533,
      "epoch": 2.512889976404789,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004367253728019095,
      "loss": 5.7199,
      "mean_token_accuracy": 0.1768442362546921,
      "num_tokens": 58283373.0,
      "step": 28755
    },
    {
      "entropy": 6.1330726146698,
      "epoch": 2.513326924757494,
      "grad_norm": 2.296875,
      "learning_rate": 0.00043670370686598814,
      "loss": 5.7713,
      "mean_token_accuracy": 0.1647837996482849,
      "num_tokens": 58293365.0,
      "step": 28760
    },
    {
      "entropy": 6.071543645858765,
      "epoch": 2.5137638731101983,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00043668203782850495,
      "loss": 5.7275,
      "mean_token_accuracy": 0.17642978578805923,
      "num_tokens": 58303750.0,
      "step": 28765
    },
    {
      "entropy": 6.13848295211792,
      "epoch": 2.5142008214629032,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004366603656898755,
      "loss": 5.7494,
      "mean_token_accuracy": 0.17032788395881654,
      "num_tokens": 58313209.0,
      "step": 28770
    },
    {
      "entropy": 6.12419319152832,
      "epoch": 2.5146377698156077,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00043663869045051565,
      "loss": 5.7526,
      "mean_token_accuracy": 0.17141420394182205,
      "num_tokens": 58322641.0,
      "step": 28775
    },
    {
      "entropy": 6.076434564590454,
      "epoch": 2.5150747181683126,
      "grad_norm": 1.84375,
      "learning_rate": 0.00043661701211084096,
      "loss": 5.626,
      "mean_token_accuracy": 0.17809172868728637,
      "num_tokens": 58331865.0,
      "step": 28780
    },
    {
      "entropy": 6.1137199878692625,
      "epoch": 2.515511666521017,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043659533067126735,
      "loss": 5.7754,
      "mean_token_accuracy": 0.1661386027932167,
      "num_tokens": 58341188.0,
      "step": 28785
    },
    {
      "entropy": 6.102758979797363,
      "epoch": 2.515948614873722,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004365736461322105,
      "loss": 5.7923,
      "mean_token_accuracy": 0.16507168412208556,
      "num_tokens": 58351856.0,
      "step": 28790
    },
    {
      "entropy": 6.151340532302856,
      "epoch": 2.5163855632264265,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004365519584940866,
      "loss": 5.7716,
      "mean_token_accuracy": 0.16765367835760117,
      "num_tokens": 58363066.0,
      "step": 28795
    },
    {
      "entropy": 6.163084411621094,
      "epoch": 2.5168225115791314,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004365302677573114,
      "loss": 5.7646,
      "mean_token_accuracy": 0.15863908380270003,
      "num_tokens": 58374646.0,
      "step": 28800
    },
    {
      "entropy": 6.0817187309265135,
      "epoch": 2.517259459931836,
      "grad_norm": 1.453125,
      "learning_rate": 0.000436508573922301,
      "loss": 5.7189,
      "mean_token_accuracy": 0.1657414734363556,
      "num_tokens": 58385192.0,
      "step": 28805
    },
    {
      "entropy": 6.143931245803833,
      "epoch": 2.5176964082845408,
      "grad_norm": 1.875,
      "learning_rate": 0.00043648687698947153,
      "loss": 5.7426,
      "mean_token_accuracy": 0.1733117625117302,
      "num_tokens": 58394237.0,
      "step": 28810
    },
    {
      "entropy": 6.046315765380859,
      "epoch": 2.5181333566372457,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043646517695923923,
      "loss": 5.6947,
      "mean_token_accuracy": 0.17188163101673126,
      "num_tokens": 58404229.0,
      "step": 28815
    },
    {
      "entropy": 6.082351160049439,
      "epoch": 2.51857030498995,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004364434738320201,
      "loss": 5.7525,
      "mean_token_accuracy": 0.16743903160095214,
      "num_tokens": 58414706.0,
      "step": 28820
    },
    {
      "entropy": 6.158907127380371,
      "epoch": 2.5190072533426546,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00043642176760823066,
      "loss": 5.7731,
      "mean_token_accuracy": 0.16042548716068267,
      "num_tokens": 58425699.0,
      "step": 28825
    },
    {
      "entropy": 6.160825157165528,
      "epoch": 2.5194442016953595,
      "grad_norm": 2.1875,
      "learning_rate": 0.000436400058288287,
      "loss": 5.7564,
      "mean_token_accuracy": 0.16030147075653076,
      "num_tokens": 58435923.0,
      "step": 28830
    },
    {
      "entropy": 6.038008069992065,
      "epoch": 2.5198811500480645,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004363783458726058,
      "loss": 5.7158,
      "mean_token_accuracy": 0.1678299531340599,
      "num_tokens": 58445569.0,
      "step": 28835
    },
    {
      "entropy": 6.163236379623413,
      "epoch": 2.520318098400769,
      "grad_norm": 1.703125,
      "learning_rate": 0.00043635663036160324,
      "loss": 5.7146,
      "mean_token_accuracy": 0.17524854689836503,
      "num_tokens": 58456686.0,
      "step": 28840
    },
    {
      "entropy": 6.0593505859375,
      "epoch": 2.520755046753474,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000436334911755696,
      "loss": 5.6505,
      "mean_token_accuracy": 0.17898094654083252,
      "num_tokens": 58466148.0,
      "step": 28845
    },
    {
      "entropy": 6.097918558120727,
      "epoch": 2.5211919951061783,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004363131900553006,
      "loss": 5.7157,
      "mean_token_accuracy": 0.16519160121679305,
      "num_tokens": 58475161.0,
      "step": 28850
    },
    {
      "entropy": 6.099815034866333,
      "epoch": 2.521628943458883,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004362914652608338,
      "loss": 5.6555,
      "mean_token_accuracy": 0.17631271183490754,
      "num_tokens": 58483631.0,
      "step": 28855
    },
    {
      "entropy": 5.975007724761963,
      "epoch": 2.5220658918115877,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004362697373727121,
      "loss": 5.6816,
      "mean_token_accuracy": 0.16618525832891465,
      "num_tokens": 58493063.0,
      "step": 28860
    },
    {
      "entropy": 6.106382608413696,
      "epoch": 2.5225028401642926,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004362480063913524,
      "loss": 5.8081,
      "mean_token_accuracy": 0.15738761574029922,
      "num_tokens": 58503658.0,
      "step": 28865
    },
    {
      "entropy": 6.160551309585571,
      "epoch": 2.5229397885169975,
      "grad_norm": 1.765625,
      "learning_rate": 0.00043622627231717126,
      "loss": 5.6864,
      "mean_token_accuracy": 0.17053182423114777,
      "num_tokens": 58515083.0,
      "step": 28870
    },
    {
      "entropy": 6.185694980621338,
      "epoch": 2.523376736869702,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004362045351505859,
      "loss": 5.7463,
      "mean_token_accuracy": 0.1678892970085144,
      "num_tokens": 58524984.0,
      "step": 28875
    },
    {
      "entropy": 6.19119701385498,
      "epoch": 2.5238136852224065,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004361827948920131,
      "loss": 5.8449,
      "mean_token_accuracy": 0.15692123025655746,
      "num_tokens": 58535651.0,
      "step": 28880
    },
    {
      "entropy": 6.100747299194336,
      "epoch": 2.5242506335751114,
      "grad_norm": 2.15625,
      "learning_rate": 0.00043616105154186974,
      "loss": 5.7819,
      "mean_token_accuracy": 0.1654674604535103,
      "num_tokens": 58546380.0,
      "step": 28885
    },
    {
      "entropy": 6.066555976867676,
      "epoch": 2.5246875819278163,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00043613930510057305,
      "loss": 5.6932,
      "mean_token_accuracy": 0.17490607649087905,
      "num_tokens": 58555639.0,
      "step": 28890
    },
    {
      "entropy": 6.125019598007202,
      "epoch": 2.5251245302805208,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043611755556853993,
      "loss": 5.7486,
      "mean_token_accuracy": 0.16746176332235335,
      "num_tokens": 58564948.0,
      "step": 28895
    },
    {
      "entropy": 6.138490056991577,
      "epoch": 2.5255614786332257,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004360958029461878,
      "loss": 5.797,
      "mean_token_accuracy": 0.16198670864105225,
      "num_tokens": 58574146.0,
      "step": 28900
    },
    {
      "entropy": 6.1633847713470455,
      "epoch": 2.52599842698593,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004360740472339336,
      "loss": 5.7023,
      "mean_token_accuracy": 0.17735922187566758,
      "num_tokens": 58583438.0,
      "step": 28905
    },
    {
      "entropy": 6.10577621459961,
      "epoch": 2.526435375338635,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00043605228843219494,
      "loss": 5.7039,
      "mean_token_accuracy": 0.16858933866024017,
      "num_tokens": 58594552.0,
      "step": 28910
    },
    {
      "entropy": 6.122722959518432,
      "epoch": 2.5268723236913395,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004360305265413889,
      "loss": 5.7127,
      "mean_token_accuracy": 0.16804221272468567,
      "num_tokens": 58604162.0,
      "step": 28915
    },
    {
      "entropy": 6.054220247268677,
      "epoch": 2.5273092720440444,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004360087615619331,
      "loss": 5.6572,
      "mean_token_accuracy": 0.17255546003580094,
      "num_tokens": 58613487.0,
      "step": 28920
    },
    {
      "entropy": 5.986214351654053,
      "epoch": 2.5277462203967493,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043598699349424474,
      "loss": 5.6009,
      "mean_token_accuracy": 0.186163929104805,
      "num_tokens": 58623252.0,
      "step": 28925
    },
    {
      "entropy": 6.114772129058838,
      "epoch": 2.528183168749454,
      "grad_norm": 1.25,
      "learning_rate": 0.0004359652223387415,
      "loss": 5.7928,
      "mean_token_accuracy": 0.15375191420316697,
      "num_tokens": 58633585.0,
      "step": 28930
    },
    {
      "entropy": 6.078793239593506,
      "epoch": 2.5286201171021583,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004359434480958411,
      "loss": 5.7455,
      "mean_token_accuracy": 0.16730255484580994,
      "num_tokens": 58643202.0,
      "step": 28935
    },
    {
      "entropy": 6.041677236557007,
      "epoch": 2.529057065454863,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004359216707659608,
      "loss": 5.6661,
      "mean_token_accuracy": 0.17884819507598876,
      "num_tokens": 58653153.0,
      "step": 28940
    },
    {
      "entropy": 6.1243359565734865,
      "epoch": 2.529494013807568,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004358998903495188,
      "loss": 5.6509,
      "mean_token_accuracy": 0.17599302381277085,
      "num_tokens": 58663835.0,
      "step": 28945
    },
    {
      "entropy": 6.15927357673645,
      "epoch": 2.5299309621602726,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004358781068469324,
      "loss": 5.7319,
      "mean_token_accuracy": 0.16999367922544478,
      "num_tokens": 58673114.0,
      "step": 28950
    },
    {
      "entropy": 6.095506811141968,
      "epoch": 2.5303679105129775,
      "grad_norm": 2.15625,
      "learning_rate": 0.00043585632025861963,
      "loss": 5.7121,
      "mean_token_accuracy": 0.17098892629146575,
      "num_tokens": 58683121.0,
      "step": 28955
    },
    {
      "entropy": 6.106607103347779,
      "epoch": 2.530804858865682,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004358345305849984,
      "loss": 5.7522,
      "mean_token_accuracy": 0.16612544655799866,
      "num_tokens": 58693656.0,
      "step": 28960
    },
    {
      "entropy": 6.093336868286133,
      "epoch": 2.531241807218387,
      "grad_norm": 2.46875,
      "learning_rate": 0.00043581273782648664,
      "loss": 5.6977,
      "mean_token_accuracy": 0.17669406086206435,
      "num_tokens": 58703061.0,
      "step": 28965
    },
    {
      "entropy": 6.084471893310547,
      "epoch": 2.5316787555710913,
      "grad_norm": 2.5625,
      "learning_rate": 0.00043579094198350224,
      "loss": 5.7364,
      "mean_token_accuracy": 0.17037638574838637,
      "num_tokens": 58712463.0,
      "step": 28970
    },
    {
      "entropy": 6.081583166122437,
      "epoch": 2.5321157039237963,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043576914305646334,
      "loss": 5.6726,
      "mean_token_accuracy": 0.1712366297841072,
      "num_tokens": 58722214.0,
      "step": 28975
    },
    {
      "entropy": 6.055950784683228,
      "epoch": 2.532552652276501,
      "grad_norm": 1.546875,
      "learning_rate": 0.000435747341045788,
      "loss": 5.7373,
      "mean_token_accuracy": 0.17179748862981797,
      "num_tokens": 58732042.0,
      "step": 28980
    },
    {
      "entropy": 6.111783885955811,
      "epoch": 2.5329896006292056,
      "grad_norm": 2.015625,
      "learning_rate": 0.00043572553595189444,
      "loss": 5.8195,
      "mean_token_accuracy": 0.15835743993520737,
      "num_tokens": 58743139.0,
      "step": 28985
    },
    {
      "entropy": 6.187650346755982,
      "epoch": 2.53342654898191,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00043570372777520096,
      "loss": 5.7675,
      "mean_token_accuracy": 0.16577517092227936,
      "num_tokens": 58752994.0,
      "step": 28990
    },
    {
      "entropy": 6.100604724884033,
      "epoch": 2.533863497334615,
      "grad_norm": 2.046875,
      "learning_rate": 0.00043568191651612573,
      "loss": 5.7193,
      "mean_token_accuracy": 0.16468655616044997,
      "num_tokens": 58762664.0,
      "step": 28995
    },
    {
      "entropy": 6.093721532821656,
      "epoch": 2.53430044568732,
      "grad_norm": 2.109375,
      "learning_rate": 0.00043566010217508706,
      "loss": 5.7598,
      "mean_token_accuracy": 0.16438718885183334,
      "num_tokens": 58772992.0,
      "step": 29000
    },
    {
      "entropy": 6.083069515228272,
      "epoch": 2.5347373940400244,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004356382847525035,
      "loss": 5.7408,
      "mean_token_accuracy": 0.1681981384754181,
      "num_tokens": 58783157.0,
      "step": 29005
    },
    {
      "entropy": 6.10103669166565,
      "epoch": 2.5351743423927293,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004356164642487935,
      "loss": 5.7641,
      "mean_token_accuracy": 0.1658971980214119,
      "num_tokens": 58794020.0,
      "step": 29010
    },
    {
      "entropy": 6.057534456253052,
      "epoch": 2.535611290745434,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004355946406643755,
      "loss": 5.6779,
      "mean_token_accuracy": 0.17810027450323104,
      "num_tokens": 58804524.0,
      "step": 29015
    },
    {
      "entropy": 6.087020301818848,
      "epoch": 2.5360482390981387,
      "grad_norm": 1.796875,
      "learning_rate": 0.00043557281399966814,
      "loss": 5.7444,
      "mean_token_accuracy": 0.16654877066612245,
      "num_tokens": 58814278.0,
      "step": 29020
    },
    {
      "entropy": 6.068491840362549,
      "epoch": 2.536485187450843,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004355509842550901,
      "loss": 5.679,
      "mean_token_accuracy": 0.165374992787838,
      "num_tokens": 58824258.0,
      "step": 29025
    },
    {
      "entropy": 6.093202400207519,
      "epoch": 2.536922135803548,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004355291514310601,
      "loss": 5.7584,
      "mean_token_accuracy": 0.1671905428171158,
      "num_tokens": 58834316.0,
      "step": 29030
    },
    {
      "entropy": 6.104451847076416,
      "epoch": 2.537359084156253,
      "grad_norm": 1.859375,
      "learning_rate": 0.00043550731552799676,
      "loss": 5.7514,
      "mean_token_accuracy": 0.16467293053865434,
      "num_tokens": 58844201.0,
      "step": 29035
    },
    {
      "entropy": 6.112790584564209,
      "epoch": 2.5377960325089575,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00043548547654631917,
      "loss": 5.6696,
      "mean_token_accuracy": 0.1741873636841774,
      "num_tokens": 58853675.0,
      "step": 29040
    },
    {
      "entropy": 6.112831735610962,
      "epoch": 2.538232980861662,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004354636344864459,
      "loss": 5.7079,
      "mean_token_accuracy": 0.1683295652270317,
      "num_tokens": 58863316.0,
      "step": 29045
    },
    {
      "entropy": 6.057488489151001,
      "epoch": 2.538669929214367,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004354417893487961,
      "loss": 5.6766,
      "mean_token_accuracy": 0.16620831042528153,
      "num_tokens": 58873479.0,
      "step": 29050
    },
    {
      "entropy": 6.010253477096557,
      "epoch": 2.5391068775670718,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004354199411337886,
      "loss": 5.693,
      "mean_token_accuracy": 0.16646684408187867,
      "num_tokens": 58882540.0,
      "step": 29055
    },
    {
      "entropy": 6.103668117523194,
      "epoch": 2.5395438259197762,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004353980898418427,
      "loss": 5.6994,
      "mean_token_accuracy": 0.16649235039949417,
      "num_tokens": 58891565.0,
      "step": 29060
    },
    {
      "entropy": 6.118711042404175,
      "epoch": 2.539980774272481,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004353762354733773,
      "loss": 5.6879,
      "mean_token_accuracy": 0.1672006905078888,
      "num_tokens": 58900689.0,
      "step": 29065
    },
    {
      "entropy": 6.0792444229125975,
      "epoch": 2.5404177226251856,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043535437802881175,
      "loss": 5.6794,
      "mean_token_accuracy": 0.17405762523412704,
      "num_tokens": 58910588.0,
      "step": 29070
    },
    {
      "entropy": 6.063365840911866,
      "epoch": 2.5408546709778905,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004353325175085651,
      "loss": 5.7953,
      "mean_token_accuracy": 0.16209391355514527,
      "num_tokens": 58921189.0,
      "step": 29075
    },
    {
      "entropy": 6.0658759593963625,
      "epoch": 2.541291619330595,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043531065391305685,
      "loss": 5.6959,
      "mean_token_accuracy": 0.16645718812942506,
      "num_tokens": 58930941.0,
      "step": 29080
    },
    {
      "entropy": 6.138298988342285,
      "epoch": 2.5417285676833,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004352887872427063,
      "loss": 5.8554,
      "mean_token_accuracy": 0.16820737719535828,
      "num_tokens": 58940371.0,
      "step": 29085
    },
    {
      "entropy": 6.203558540344238,
      "epoch": 2.5421655160360044,
      "grad_norm": 2.234375,
      "learning_rate": 0.00043526691749793273,
      "loss": 5.7257,
      "mean_token_accuracy": 0.16567156016826629,
      "num_tokens": 58950446.0,
      "step": 29090
    },
    {
      "entropy": 6.059780645370483,
      "epoch": 2.5426024643887093,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004352450446791557,
      "loss": 5.6891,
      "mean_token_accuracy": 0.16857760399580002,
      "num_tokens": 58960757.0,
      "step": 29095
    },
    {
      "entropy": 6.079778099060059,
      "epoch": 2.5430394127414138,
      "grad_norm": 2.0625,
      "learning_rate": 0.00043522316878679475,
      "loss": 5.7644,
      "mean_token_accuracy": 0.16771948784589769,
      "num_tokens": 58971621.0,
      "step": 29100
    },
    {
      "entropy": 6.1762432098388675,
      "epoch": 2.5434763610941187,
      "grad_norm": 2.21875,
      "learning_rate": 0.00043520128982126944,
      "loss": 5.7462,
      "mean_token_accuracy": 0.16571457982063292,
      "num_tokens": 58982505.0,
      "step": 29105
    },
    {
      "entropy": 6.062740612030029,
      "epoch": 2.5439133094468236,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004351794077829995,
      "loss": 5.641,
      "mean_token_accuracy": 0.1738319382071495,
      "num_tokens": 58992270.0,
      "step": 29110
    },
    {
      "entropy": 6.011745357513428,
      "epoch": 2.544350257799528,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004351575226724046,
      "loss": 5.6231,
      "mean_token_accuracy": 0.17486979216337203,
      "num_tokens": 59002510.0,
      "step": 29115
    },
    {
      "entropy": 6.091541862487793,
      "epoch": 2.544787206152233,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004351356344899045,
      "loss": 5.7367,
      "mean_token_accuracy": 0.1668363779783249,
      "num_tokens": 59012942.0,
      "step": 29120
    },
    {
      "entropy": 6.0989336490631105,
      "epoch": 2.5452241545049374,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00043511374323591904,
      "loss": 5.817,
      "mean_token_accuracy": 0.1634032368659973,
      "num_tokens": 59023509.0,
      "step": 29125
    },
    {
      "entropy": 6.147858190536499,
      "epoch": 2.5456611028576424,
      "grad_norm": 1.53125,
      "learning_rate": 0.00043509184891086795,
      "loss": 5.7296,
      "mean_token_accuracy": 0.16557153910398484,
      "num_tokens": 59033782.0,
      "step": 29130
    },
    {
      "entropy": 6.133325529098511,
      "epoch": 2.546098051210347,
      "grad_norm": 2.609375,
      "learning_rate": 0.00043506995151517135,
      "loss": 5.7563,
      "mean_token_accuracy": 0.16247304826974868,
      "num_tokens": 59045791.0,
      "step": 29135
    },
    {
      "entropy": 6.038221120834351,
      "epoch": 2.5465349995630517,
      "grad_norm": 1.75,
      "learning_rate": 0.00043504805104924925,
      "loss": 5.6726,
      "mean_token_accuracy": 0.17335552275180816,
      "num_tokens": 59056578.0,
      "step": 29140
    },
    {
      "entropy": 6.069480895996094,
      "epoch": 2.546971947915756,
      "grad_norm": 1.25,
      "learning_rate": 0.0004350261475135217,
      "loss": 5.7541,
      "mean_token_accuracy": 0.16096172779798507,
      "num_tokens": 59066712.0,
      "step": 29145
    },
    {
      "entropy": 6.11282639503479,
      "epoch": 2.547408896268461,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004350042409084088,
      "loss": 5.6943,
      "mean_token_accuracy": 0.17067378163337707,
      "num_tokens": 59077557.0,
      "step": 29150
    },
    {
      "entropy": 6.083163499832153,
      "epoch": 2.5478458446211656,
      "grad_norm": 1.703125,
      "learning_rate": 0.00043498233123433064,
      "loss": 5.7007,
      "mean_token_accuracy": 0.16960372924804687,
      "num_tokens": 59086649.0,
      "step": 29155
    },
    {
      "entropy": 6.108971834182739,
      "epoch": 2.5482827929738705,
      "grad_norm": 1.59375,
      "learning_rate": 0.00043496041849170757,
      "loss": 5.672,
      "mean_token_accuracy": 0.1707042545080185,
      "num_tokens": 59096838.0,
      "step": 29160
    },
    {
      "entropy": 6.142396306991577,
      "epoch": 2.5487197413265754,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004349385026809599,
      "loss": 5.734,
      "mean_token_accuracy": 0.17197115570306779,
      "num_tokens": 59107884.0,
      "step": 29165
    },
    {
      "entropy": 6.084319162368774,
      "epoch": 2.54915668967928,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043491658380250784,
      "loss": 5.7534,
      "mean_token_accuracy": 0.16237785518169404,
      "num_tokens": 59118315.0,
      "step": 29170
    },
    {
      "entropy": 6.076199054718018,
      "epoch": 2.5495936380319844,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043489466185677193,
      "loss": 5.7053,
      "mean_token_accuracy": 0.1717744767665863,
      "num_tokens": 59128331.0,
      "step": 29175
    },
    {
      "entropy": 6.078016090393066,
      "epoch": 2.5500305863846893,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004348727368441727,
      "loss": 5.7373,
      "mean_token_accuracy": 0.16760917901992797,
      "num_tokens": 59138643.0,
      "step": 29180
    },
    {
      "entropy": 6.11273341178894,
      "epoch": 2.550467534737394,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004348508087651306,
      "loss": 5.6691,
      "mean_token_accuracy": 0.1702326327562332,
      "num_tokens": 59148364.0,
      "step": 29185
    },
    {
      "entropy": 6.068109321594238,
      "epoch": 2.5509044830900987,
      "grad_norm": 1.890625,
      "learning_rate": 0.00043482887762006614,
      "loss": 5.7215,
      "mean_token_accuracy": 0.169105164706707,
      "num_tokens": 59158317.0,
      "step": 29190
    },
    {
      "entropy": 6.147952890396118,
      "epoch": 2.5513414314428036,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043480694340940017,
      "loss": 5.7974,
      "mean_token_accuracy": 0.16575304865837098,
      "num_tokens": 59168879.0,
      "step": 29195
    },
    {
      "entropy": 6.148684406280518,
      "epoch": 2.551778379795508,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004347850061335532,
      "loss": 5.7294,
      "mean_token_accuracy": 0.1682370811700821,
      "num_tokens": 59179248.0,
      "step": 29200
    },
    {
      "entropy": 5.985152006149292,
      "epoch": 2.552215328148213,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004347630657929461,
      "loss": 5.6714,
      "mean_token_accuracy": 0.1753835454583168,
      "num_tokens": 59189217.0,
      "step": 29205
    },
    {
      "entropy": 6.10281457901001,
      "epoch": 2.5526522765009174,
      "grad_norm": 1.578125,
      "learning_rate": 0.00043474112238799974,
      "loss": 5.7254,
      "mean_token_accuracy": 0.17049014568328857,
      "num_tokens": 59199857.0,
      "step": 29210
    },
    {
      "entropy": 6.176831388473511,
      "epoch": 2.5530892248536223,
      "grad_norm": 2.6875,
      "learning_rate": 0.00043471917591913486,
      "loss": 5.7283,
      "mean_token_accuracy": 0.16778160333633424,
      "num_tokens": 59210119.0,
      "step": 29215
    },
    {
      "entropy": 6.076271057128906,
      "epoch": 2.5535261732063272,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043469722638677256,
      "loss": 5.6497,
      "mean_token_accuracy": 0.17279819250106812,
      "num_tokens": 59219258.0,
      "step": 29220
    },
    {
      "entropy": 6.0759392261505125,
      "epoch": 2.5539631215590317,
      "grad_norm": 1.75,
      "learning_rate": 0.00043467527379133377,
      "loss": 5.7594,
      "mean_token_accuracy": 0.1674110159277916,
      "num_tokens": 59229664.0,
      "step": 29225
    },
    {
      "entropy": 6.190113639831543,
      "epoch": 2.554400069911736,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004346533181332395,
      "loss": 5.7673,
      "mean_token_accuracy": 0.16447471231222152,
      "num_tokens": 59239699.0,
      "step": 29230
    },
    {
      "entropy": 6.05651741027832,
      "epoch": 2.554837018264441,
      "grad_norm": 1.625,
      "learning_rate": 0.00043463135941291105,
      "loss": 5.7023,
      "mean_token_accuracy": 0.17206832766532898,
      "num_tokens": 59249891.0,
      "step": 29235
    },
    {
      "entropy": 6.075493812561035,
      "epoch": 2.555273966617146,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004346093976307694,
      "loss": 5.7506,
      "mean_token_accuracy": 0.16577305644750595,
      "num_tokens": 59260092.0,
      "step": 29240
    },
    {
      "entropy": 6.048191261291504,
      "epoch": 2.5557109149698505,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004345874327872359,
      "loss": 5.7604,
      "mean_token_accuracy": 0.16670387834310532,
      "num_tokens": 59268844.0,
      "step": 29245
    },
    {
      "entropy": 6.155736970901489,
      "epoch": 2.5561478633225554,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004345654648827318,
      "loss": 5.8074,
      "mean_token_accuracy": 0.16046487540006638,
      "num_tokens": 59279275.0,
      "step": 29250
    },
    {
      "entropy": 6.151616954803467,
      "epoch": 2.55658481167526,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004345434939176784,
      "loss": 5.7591,
      "mean_token_accuracy": 0.16487863957881926,
      "num_tokens": 59290596.0,
      "step": 29255
    },
    {
      "entropy": 6.069549465179444,
      "epoch": 2.5570217600279648,
      "grad_norm": 1.53125,
      "learning_rate": 0.00043452151989249727,
      "loss": 5.6817,
      "mean_token_accuracy": 0.16582526713609697,
      "num_tokens": 59300642.0,
      "step": 29260
    },
    {
      "entropy": 6.042895030975342,
      "epoch": 2.5574587083806692,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004344995428076098,
      "loss": 5.7514,
      "mean_token_accuracy": 0.16912768483161927,
      "num_tokens": 59310064.0,
      "step": 29265
    },
    {
      "entropy": 6.138311290740967,
      "epoch": 2.557895656733374,
      "grad_norm": 1.71875,
      "learning_rate": 0.00043447756266343755,
      "loss": 5.7296,
      "mean_token_accuracy": 0.16710765659809113,
      "num_tokens": 59319114.0,
      "step": 29270
    },
    {
      "entropy": 6.105566310882568,
      "epoch": 2.558332605086079,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00043445557946040207,
      "loss": 5.7948,
      "mean_token_accuracy": 0.16486831456422807,
      "num_tokens": 59329414.0,
      "step": 29275
    },
    {
      "entropy": 6.142416143417359,
      "epoch": 2.5587695534387835,
      "grad_norm": 1.375,
      "learning_rate": 0.00043443359319892495,
      "loss": 5.8011,
      "mean_token_accuracy": 0.1591491535305977,
      "num_tokens": 59339980.0,
      "step": 29280
    },
    {
      "entropy": 6.171898603439331,
      "epoch": 2.559206501791488,
      "grad_norm": 1.890625,
      "learning_rate": 0.000434411603879428,
      "loss": 5.78,
      "mean_token_accuracy": 0.16187877357006072,
      "num_tokens": 59349754.0,
      "step": 29285
    },
    {
      "entropy": 6.061413383483886,
      "epoch": 2.559643450144193,
      "grad_norm": 2.015625,
      "learning_rate": 0.000434389611502333,
      "loss": 5.5776,
      "mean_token_accuracy": 0.18531085848808287,
      "num_tokens": 59358542.0,
      "step": 29290
    },
    {
      "entropy": 6.101986885070801,
      "epoch": 2.560080398496898,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004343676160680617,
      "loss": 5.7159,
      "mean_token_accuracy": 0.17077200412750243,
      "num_tokens": 59367743.0,
      "step": 29295
    },
    {
      "entropy": 6.095064401626587,
      "epoch": 2.5605173468496023,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000434345617577036,
      "loss": 5.7351,
      "mean_token_accuracy": 0.16542245149612428,
      "num_tokens": 59377622.0,
      "step": 29300
    },
    {
      "entropy": 6.042428302764892,
      "epoch": 2.560954295202307,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00043432361602967787,
      "loss": 5.685,
      "mean_token_accuracy": 0.1646775409579277,
      "num_tokens": 59388661.0,
      "step": 29305
    },
    {
      "entropy": 6.130915355682373,
      "epoch": 2.5613912435550117,
      "grad_norm": 2.171875,
      "learning_rate": 0.00043430161142640934,
      "loss": 5.7643,
      "mean_token_accuracy": 0.166842083632946,
      "num_tokens": 59399935.0,
      "step": 29310
    },
    {
      "entropy": 6.123586654663086,
      "epoch": 2.5618281919077166,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004342796037676524,
      "loss": 5.6821,
      "mean_token_accuracy": 0.17601296603679656,
      "num_tokens": 59409513.0,
      "step": 29315
    },
    {
      "entropy": 6.089656257629395,
      "epoch": 2.562265140260421,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004342575930538292,
      "loss": 5.7117,
      "mean_token_accuracy": 0.16760994493961334,
      "num_tokens": 59420158.0,
      "step": 29320
    },
    {
      "entropy": 6.089406299591064,
      "epoch": 2.562702088613126,
      "grad_norm": 2.015625,
      "learning_rate": 0.00043423557928536197,
      "loss": 5.7673,
      "mean_token_accuracy": 0.17189732193946838,
      "num_tokens": 59429440.0,
      "step": 29325
    },
    {
      "entropy": 6.0219520092010494,
      "epoch": 2.563139036965831,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004342135624626728,
      "loss": 5.6441,
      "mean_token_accuracy": 0.17820007354021072,
      "num_tokens": 59439014.0,
      "step": 29330
    },
    {
      "entropy": 6.056343030929566,
      "epoch": 2.5635759853185354,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004341915425861841,
      "loss": 5.688,
      "mean_token_accuracy": 0.17280870378017427,
      "num_tokens": 59448714.0,
      "step": 29335
    },
    {
      "entropy": 6.038217544555664,
      "epoch": 2.56401293367124,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004341695196563183,
      "loss": 5.7321,
      "mean_token_accuracy": 0.17020404040813447,
      "num_tokens": 59459577.0,
      "step": 29340
    },
    {
      "entropy": 6.161348295211792,
      "epoch": 2.5644498820239447,
      "grad_norm": 2.1875,
      "learning_rate": 0.00043414749367349755,
      "loss": 5.7112,
      "mean_token_accuracy": 0.17270047664642335,
      "num_tokens": 59470313.0,
      "step": 29345
    },
    {
      "entropy": 6.121923112869263,
      "epoch": 2.5648868303766497,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004341254646381446,
      "loss": 5.7334,
      "mean_token_accuracy": 0.1720060348510742,
      "num_tokens": 59479484.0,
      "step": 29350
    },
    {
      "entropy": 6.063084936141967,
      "epoch": 2.565323778729354,
      "grad_norm": 4.46875,
      "learning_rate": 0.00043410343255068177,
      "loss": 5.6898,
      "mean_token_accuracy": 0.16496320813894272,
      "num_tokens": 59490290.0,
      "step": 29355
    },
    {
      "entropy": 6.0945220470428465,
      "epoch": 2.565760727082059,
      "grad_norm": 2.390625,
      "learning_rate": 0.00043408139741153175,
      "loss": 5.7586,
      "mean_token_accuracy": 0.1669430449604988,
      "num_tokens": 59501080.0,
      "step": 29360
    },
    {
      "entropy": 6.053455972671509,
      "epoch": 2.5661976754347635,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00043405935922111725,
      "loss": 5.6311,
      "mean_token_accuracy": 0.18372388780117035,
      "num_tokens": 59511453.0,
      "step": 29365
    },
    {
      "entropy": 6.046661376953125,
      "epoch": 2.5666346237874684,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004340373179798608,
      "loss": 5.7407,
      "mean_token_accuracy": 0.17002422660589217,
      "num_tokens": 59521438.0,
      "step": 29370
    },
    {
      "entropy": 6.1291642665863035,
      "epoch": 2.567071572140173,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004340152736881853,
      "loss": 5.8099,
      "mean_token_accuracy": 0.1627928137779236,
      "num_tokens": 59531763.0,
      "step": 29375
    },
    {
      "entropy": 6.232885694503784,
      "epoch": 2.567508520492878,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043399322634651345,
      "loss": 5.8754,
      "mean_token_accuracy": 0.1557070717215538,
      "num_tokens": 59542857.0,
      "step": 29380
    },
    {
      "entropy": 6.131737995147705,
      "epoch": 2.5679454688455827,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004339711759552684,
      "loss": 5.6822,
      "mean_token_accuracy": 0.17000928968191148,
      "num_tokens": 59552353.0,
      "step": 29385
    },
    {
      "entropy": 6.116338109970092,
      "epoch": 2.568382417198287,
      "grad_norm": 1.625,
      "learning_rate": 0.0004339491225148727,
      "loss": 5.7769,
      "mean_token_accuracy": 0.16909224689006805,
      "num_tokens": 59562377.0,
      "step": 29390
    },
    {
      "entropy": 6.129556703567505,
      "epoch": 2.5688193655509917,
      "grad_norm": 1.484375,
      "learning_rate": 0.00043392706602574967,
      "loss": 5.7094,
      "mean_token_accuracy": 0.16964516192674636,
      "num_tokens": 59572141.0,
      "step": 29395
    },
    {
      "entropy": 6.102687072753906,
      "epoch": 2.5692563139036966,
      "grad_norm": 1.375,
      "learning_rate": 0.00043390500648832214,
      "loss": 5.6642,
      "mean_token_accuracy": 0.16835127174854278,
      "num_tokens": 59581883.0,
      "step": 29400
    },
    {
      "entropy": 6.109248924255371,
      "epoch": 2.5696932622564015,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004338829439030134,
      "loss": 5.7449,
      "mean_token_accuracy": 0.16414839178323745,
      "num_tokens": 59591798.0,
      "step": 29405
    },
    {
      "entropy": 6.163287925720215,
      "epoch": 2.570130210609106,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004338608782702465,
      "loss": 5.6707,
      "mean_token_accuracy": 0.169481298327446,
      "num_tokens": 59601525.0,
      "step": 29410
    },
    {
      "entropy": 6.103771495819092,
      "epoch": 2.570567158961811,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004338388095904447,
      "loss": 5.7955,
      "mean_token_accuracy": 0.16636169701814651,
      "num_tokens": 59611335.0,
      "step": 29415
    },
    {
      "entropy": 6.062351131439209,
      "epoch": 2.5710041073145153,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004338167378640314,
      "loss": 5.7753,
      "mean_token_accuracy": 0.1674634575843811,
      "num_tokens": 59620962.0,
      "step": 29420
    },
    {
      "entropy": 6.116451168060303,
      "epoch": 2.5714410556672203,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004337946630914297,
      "loss": 5.7199,
      "mean_token_accuracy": 0.16407141089439392,
      "num_tokens": 59631568.0,
      "step": 29425
    },
    {
      "entropy": 6.116264152526855,
      "epoch": 2.5718780040199247,
      "grad_norm": 1.546875,
      "learning_rate": 0.00043377258527306316,
      "loss": 5.7585,
      "mean_token_accuracy": 0.16744166761636733,
      "num_tokens": 59642249.0,
      "step": 29430
    },
    {
      "entropy": 6.170928955078125,
      "epoch": 2.5723149523726296,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004337505044093554,
      "loss": 5.7228,
      "mean_token_accuracy": 0.1684716209769249,
      "num_tokens": 59652516.0,
      "step": 29435
    },
    {
      "entropy": 6.075858545303345,
      "epoch": 2.572751900725334,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043372842050072965,
      "loss": 5.5853,
      "mean_token_accuracy": 0.18151693195104598,
      "num_tokens": 59662487.0,
      "step": 29440
    },
    {
      "entropy": 6.015738868713379,
      "epoch": 2.573188849078039,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004337063335476097,
      "loss": 5.7492,
      "mean_token_accuracy": 0.16693507581949235,
      "num_tokens": 59674061.0,
      "step": 29445
    },
    {
      "entropy": 6.078819417953492,
      "epoch": 2.5736257974307435,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000433684243550419,
      "loss": 5.7557,
      "mean_token_accuracy": 0.1667622834444046,
      "num_tokens": 59684159.0,
      "step": 29450
    },
    {
      "entropy": 6.165730953216553,
      "epoch": 2.5740627457834484,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004336621505095815,
      "loss": 5.7473,
      "mean_token_accuracy": 0.16172108203172683,
      "num_tokens": 59693742.0,
      "step": 29455
    },
    {
      "entropy": 6.184835147857666,
      "epoch": 2.5744996941361533,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043364005442552073,
      "loss": 5.8107,
      "mean_token_accuracy": 0.16089756190776824,
      "num_tokens": 59703613.0,
      "step": 29460
    },
    {
      "entropy": 6.094794750213623,
      "epoch": 2.574936642488858,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004336179552986606,
      "loss": 5.6561,
      "mean_token_accuracy": 0.17281133979558944,
      "num_tokens": 59713409.0,
      "step": 29465
    },
    {
      "entropy": 6.0442530632019045,
      "epoch": 2.5753735908415623,
      "grad_norm": 1.65625,
      "learning_rate": 0.00043359585312942496,
      "loss": 5.6665,
      "mean_token_accuracy": 0.1812247946858406,
      "num_tokens": 59723502.0,
      "step": 29470
    },
    {
      "entropy": 6.109486675262451,
      "epoch": 2.575810539194267,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00043357374791823784,
      "loss": 5.7668,
      "mean_token_accuracy": 0.1671688064932823,
      "num_tokens": 59734069.0,
      "step": 29475
    },
    {
      "entropy": 6.135386323928833,
      "epoch": 2.576247487546972,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043355163966552303,
      "loss": 5.7169,
      "mean_token_accuracy": 0.16307394057512284,
      "num_tokens": 59743793.0,
      "step": 29480
    },
    {
      "entropy": 6.0817443370819095,
      "epoch": 2.5766844358996766,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00043352952837170467,
      "loss": 5.758,
      "mean_token_accuracy": 0.16357874125242233,
      "num_tokens": 59754524.0,
      "step": 29485
    },
    {
      "entropy": 6.143489027023316,
      "epoch": 2.5771213842523815,
      "grad_norm": 1.9453125,
      "learning_rate": 0.000433507414037207,
      "loss": 5.7254,
      "mean_token_accuracy": 0.1678919166326523,
      "num_tokens": 59765437.0,
      "step": 29490
    },
    {
      "entropy": 6.152441835403442,
      "epoch": 2.577558332605086,
      "grad_norm": 3.046875,
      "learning_rate": 0.000433485296662454,
      "loss": 5.7432,
      "mean_token_accuracy": 0.1668240785598755,
      "num_tokens": 59776136.0,
      "step": 29495
    },
    {
      "entropy": 6.072850179672241,
      "epoch": 2.577995280957791,
      "grad_norm": 1.859375,
      "learning_rate": 0.00043346317624787007,
      "loss": 5.6923,
      "mean_token_accuracy": 0.17092108577489853,
      "num_tokens": 59786375.0,
      "step": 29500
    },
    {
      "entropy": 6.0206804275512695,
      "epoch": 2.5784322293104953,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004334410527938793,
      "loss": 5.6406,
      "mean_token_accuracy": 0.1691017135977745,
      "num_tokens": 59796937.0,
      "step": 29505
    },
    {
      "entropy": 6.08823938369751,
      "epoch": 2.5788691776632002,
      "grad_norm": 1.78125,
      "learning_rate": 0.00043341892630090607,
      "loss": 5.7158,
      "mean_token_accuracy": 0.17037342488765717,
      "num_tokens": 59806625.0,
      "step": 29510
    },
    {
      "entropy": 6.137568807601928,
      "epoch": 2.579306126015905,
      "grad_norm": 2.5,
      "learning_rate": 0.0004333967967693749,
      "loss": 5.7776,
      "mean_token_accuracy": 0.16523916870355607,
      "num_tokens": 59816618.0,
      "step": 29515
    },
    {
      "entropy": 6.161610555648804,
      "epoch": 2.5797430743686096,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004333746641997102,
      "loss": 5.7376,
      "mean_token_accuracy": 0.1682370826601982,
      "num_tokens": 59827666.0,
      "step": 29520
    },
    {
      "entropy": 6.138539838790893,
      "epoch": 2.580180022721314,
      "grad_norm": 2.125,
      "learning_rate": 0.0004333525285923363,
      "loss": 5.7366,
      "mean_token_accuracy": 0.17016009539365767,
      "num_tokens": 59837704.0,
      "step": 29525
    },
    {
      "entropy": 6.067703485488892,
      "epoch": 2.580616971074019,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004333303899476781,
      "loss": 5.6517,
      "mean_token_accuracy": 0.1711758241057396,
      "num_tokens": 59848186.0,
      "step": 29530
    },
    {
      "entropy": 6.075818729400635,
      "epoch": 2.581053919426724,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043330824826616,
      "loss": 5.6827,
      "mean_token_accuracy": 0.16918137669563293,
      "num_tokens": 59858453.0,
      "step": 29535
    },
    {
      "entropy": 6.0458899974823,
      "epoch": 2.5814908677794284,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004332861035482067,
      "loss": 5.7103,
      "mean_token_accuracy": 0.16728637218475342,
      "num_tokens": 59867863.0,
      "step": 29540
    },
    {
      "entropy": 6.072792100906372,
      "epoch": 2.5819278161321333,
      "grad_norm": 2.71875,
      "learning_rate": 0.000433263955794243,
      "loss": 5.7066,
      "mean_token_accuracy": 0.17295838743448258,
      "num_tokens": 59879932.0,
      "step": 29545
    },
    {
      "entropy": 6.127572393417358,
      "epoch": 2.5823647644848378,
      "grad_norm": 1.75,
      "learning_rate": 0.0004332418050046937,
      "loss": 5.7152,
      "mean_token_accuracy": 0.16975524723529817,
      "num_tokens": 59889586.0,
      "step": 29550
    },
    {
      "entropy": 6.049217367172242,
      "epoch": 2.5828017128375427,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004332196511799837,
      "loss": 5.628,
      "mean_token_accuracy": 0.17637585252523422,
      "num_tokens": 59898789.0,
      "step": 29555
    },
    {
      "entropy": 6.034009838104248,
      "epoch": 2.583238661190247,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004331974943205379,
      "loss": 5.7189,
      "mean_token_accuracy": 0.17196426838636397,
      "num_tokens": 59909346.0,
      "step": 29560
    },
    {
      "entropy": 6.10962028503418,
      "epoch": 2.583675609542952,
      "grad_norm": 1.859375,
      "learning_rate": 0.00043317533442678124,
      "loss": 5.6436,
      "mean_token_accuracy": 0.17730285823345185,
      "num_tokens": 59919979.0,
      "step": 29565
    },
    {
      "entropy": 6.075351285934448,
      "epoch": 2.584112557895657,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00043315317149913876,
      "loss": 5.6859,
      "mean_token_accuracy": 0.16963583827018738,
      "num_tokens": 59929855.0,
      "step": 29570
    },
    {
      "entropy": 6.045836496353149,
      "epoch": 2.5845495062483614,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00043313100553803564,
      "loss": 5.708,
      "mean_token_accuracy": 0.1697233349084854,
      "num_tokens": 59939148.0,
      "step": 29575
    },
    {
      "entropy": 5.991641473770142,
      "epoch": 2.584986454601066,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004331088365438969,
      "loss": 5.6491,
      "mean_token_accuracy": 0.17239840626716613,
      "num_tokens": 59948645.0,
      "step": 29580
    },
    {
      "entropy": 6.112098932266235,
      "epoch": 2.585423402953771,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004330866645171479,
      "loss": 5.6894,
      "mean_token_accuracy": 0.1708008036017418,
      "num_tokens": 59958777.0,
      "step": 29585
    },
    {
      "entropy": 6.1003203868865965,
      "epoch": 2.5858603513064757,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004330644894582138,
      "loss": 5.6562,
      "mean_token_accuracy": 0.17906805276870727,
      "num_tokens": 59969176.0,
      "step": 29590
    },
    {
      "entropy": 5.999762916564942,
      "epoch": 2.58629729965918,
      "grad_norm": 2.515625,
      "learning_rate": 0.00043304231136751995,
      "loss": 5.7332,
      "mean_token_accuracy": 0.1649943232536316,
      "num_tokens": 59979263.0,
      "step": 29595
    },
    {
      "entropy": 6.08463134765625,
      "epoch": 2.586734248011885,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043302013024549174,
      "loss": 5.6689,
      "mean_token_accuracy": 0.17394621372222902,
      "num_tokens": 59989313.0,
      "step": 29600
    },
    {
      "entropy": 6.114327049255371,
      "epoch": 2.5871711963645896,
      "grad_norm": 1.375,
      "learning_rate": 0.00043299794609255474,
      "loss": 5.7495,
      "mean_token_accuracy": 0.16580477505922317,
      "num_tokens": 59999410.0,
      "step": 29605
    },
    {
      "entropy": 6.048985004425049,
      "epoch": 2.5876081447172945,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004329757589091342,
      "loss": 5.6512,
      "mean_token_accuracy": 0.1762889802455902,
      "num_tokens": 60010206.0,
      "step": 29610
    },
    {
      "entropy": 5.9372594356536865,
      "epoch": 2.588045093069999,
      "grad_norm": 1.625,
      "learning_rate": 0.00043295356869565586,
      "loss": 5.5611,
      "mean_token_accuracy": 0.18505351692438127,
      "num_tokens": 60018692.0,
      "step": 29615
    },
    {
      "entropy": 6.059975862503052,
      "epoch": 2.588482041422704,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043293137545254536,
      "loss": 5.7836,
      "mean_token_accuracy": 0.1658503979444504,
      "num_tokens": 60028486.0,
      "step": 29620
    },
    {
      "entropy": 6.105821895599365,
      "epoch": 2.588918989775409,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004329091791802283,
      "loss": 5.7328,
      "mean_token_accuracy": 0.16746872663497925,
      "num_tokens": 60039347.0,
      "step": 29625
    },
    {
      "entropy": 6.096479654312134,
      "epoch": 2.5893559381281133,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004328869798791305,
      "loss": 5.6343,
      "mean_token_accuracy": 0.18006803542375566,
      "num_tokens": 60049891.0,
      "step": 29630
    },
    {
      "entropy": 6.089907884597778,
      "epoch": 2.5897928864808177,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043286477754967757,
      "loss": 5.744,
      "mean_token_accuracy": 0.167370043694973,
      "num_tokens": 60061264.0,
      "step": 29635
    },
    {
      "entropy": 6.091320514678955,
      "epoch": 2.5902298348335226,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00043284257219229557,
      "loss": 5.7373,
      "mean_token_accuracy": 0.16792935580015184,
      "num_tokens": 60072453.0,
      "step": 29640
    },
    {
      "entropy": 5.972164106369019,
      "epoch": 2.5906667831862276,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00043282036380741035,
      "loss": 5.7129,
      "mean_token_accuracy": 0.16636434495449065,
      "num_tokens": 60083619.0,
      "step": 29645
    },
    {
      "entropy": 6.141052722930908,
      "epoch": 2.591103731538932,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004327981523954478,
      "loss": 5.6463,
      "mean_token_accuracy": 0.1741870641708374,
      "num_tokens": 60093231.0,
      "step": 29650
    },
    {
      "entropy": 6.029804515838623,
      "epoch": 2.591540679891637,
      "grad_norm": 1.6875,
      "learning_rate": 0.000432775937956834,
      "loss": 5.6359,
      "mean_token_accuracy": 0.17032478153705596,
      "num_tokens": 60103017.0,
      "step": 29655
    },
    {
      "entropy": 6.084715080261231,
      "epoch": 2.5919776282443414,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004327537204919951,
      "loss": 5.6973,
      "mean_token_accuracy": 0.1623273268342018,
      "num_tokens": 60113095.0,
      "step": 29660
    },
    {
      "entropy": 6.129330825805664,
      "epoch": 2.5924145765970463,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004327315000013571,
      "loss": 5.7588,
      "mean_token_accuracy": 0.16718512326478957,
      "num_tokens": 60123506.0,
      "step": 29665
    },
    {
      "entropy": 6.07743501663208,
      "epoch": 2.592851524949751,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004327092764853463,
      "loss": 5.7952,
      "mean_token_accuracy": 0.16324707269668579,
      "num_tokens": 60134349.0,
      "step": 29670
    },
    {
      "entropy": 6.090957736968994,
      "epoch": 2.5932884733024557,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00043268704994438897,
      "loss": 5.6952,
      "mean_token_accuracy": 0.16576803475618362,
      "num_tokens": 60143970.0,
      "step": 29675
    },
    {
      "entropy": 6.024298095703125,
      "epoch": 2.5937254216551606,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004326648203789113,
      "loss": 5.6933,
      "mean_token_accuracy": 0.16810683757066727,
      "num_tokens": 60154904.0,
      "step": 29680
    },
    {
      "entropy": 6.084721279144287,
      "epoch": 2.594162370007865,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004326425877893399,
      "loss": 5.7024,
      "mean_token_accuracy": 0.17034658640623093,
      "num_tokens": 60164747.0,
      "step": 29685
    },
    {
      "entropy": 6.176058387756347,
      "epoch": 2.5945993183605696,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00043262035217610104,
      "loss": 5.7093,
      "mean_token_accuracy": 0.1717853367328644,
      "num_tokens": 60175487.0,
      "step": 29690
    },
    {
      "entropy": 6.085010004043579,
      "epoch": 2.5950362667132745,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043259811353962114,
      "loss": 5.7042,
      "mean_token_accuracy": 0.1652551919221878,
      "num_tokens": 60185216.0,
      "step": 29695
    },
    {
      "entropy": 6.139104175567627,
      "epoch": 2.5954732150659794,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004325758718803269,
      "loss": 5.6988,
      "mean_token_accuracy": 0.16556959152221679,
      "num_tokens": 60195495.0,
      "step": 29700
    },
    {
      "entropy": 6.108751440048218,
      "epoch": 2.595910163418684,
      "grad_norm": 1.5,
      "learning_rate": 0.00043255362719864487,
      "loss": 5.7603,
      "mean_token_accuracy": 0.16537190079689026,
      "num_tokens": 60205556.0,
      "step": 29705
    },
    {
      "entropy": 6.06081600189209,
      "epoch": 2.5963471117713888,
      "grad_norm": 2.359375,
      "learning_rate": 0.00043253137949500166,
      "loss": 5.6946,
      "mean_token_accuracy": 0.1721880167722702,
      "num_tokens": 60215904.0,
      "step": 29710
    },
    {
      "entropy": 6.01737904548645,
      "epoch": 2.5967840601240932,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043250912876982415,
      "loss": 5.6254,
      "mean_token_accuracy": 0.17714870274066924,
      "num_tokens": 60224762.0,
      "step": 29715
    },
    {
      "entropy": 6.022523212432861,
      "epoch": 2.597221008476798,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000432486875023539,
      "loss": 5.6848,
      "mean_token_accuracy": 0.1767497256398201,
      "num_tokens": 60235284.0,
      "step": 29720
    },
    {
      "entropy": 6.130961990356445,
      "epoch": 2.5976579568295026,
      "grad_norm": 1.578125,
      "learning_rate": 0.000432464618256573,
      "loss": 5.7467,
      "mean_token_accuracy": 0.16810048520565032,
      "num_tokens": 60245939.0,
      "step": 29725
    },
    {
      "entropy": 6.16960334777832,
      "epoch": 2.5980949051822075,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004324423584693531,
      "loss": 5.7349,
      "mean_token_accuracy": 0.16690510958433152,
      "num_tokens": 60256903.0,
      "step": 29730
    },
    {
      "entropy": 6.119984531402588,
      "epoch": 2.5985318535349125,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004324200956623063,
      "loss": 5.6783,
      "mean_token_accuracy": 0.1656678080558777,
      "num_tokens": 60267046.0,
      "step": 29735
    },
    {
      "entropy": 6.146673107147217,
      "epoch": 2.598968801887617,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043239782983585964,
      "loss": 5.7241,
      "mean_token_accuracy": 0.16994871497154235,
      "num_tokens": 60276858.0,
      "step": 29740
    },
    {
      "entropy": 6.056886148452759,
      "epoch": 2.5994057502403214,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043237556099044,
      "loss": 5.7027,
      "mean_token_accuracy": 0.17112268060445784,
      "num_tokens": 60285874.0,
      "step": 29745
    },
    {
      "entropy": 6.087872648239136,
      "epoch": 2.5998426985930263,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004323532891264748,
      "loss": 5.7524,
      "mean_token_accuracy": 0.16764552742242814,
      "num_tokens": 60295580.0,
      "step": 29750
    },
    {
      "entropy": 6.127909231185913,
      "epoch": 2.600279646945731,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004323310142443909,
      "loss": 5.723,
      "mean_token_accuracy": 0.1723291590809822,
      "num_tokens": 60305425.0,
      "step": 29755
    },
    {
      "entropy": 6.106584548950195,
      "epoch": 2.6007165952984357,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043230873634461577,
      "loss": 5.6919,
      "mean_token_accuracy": 0.1674206495285034,
      "num_tokens": 60315578.0,
      "step": 29760
    },
    {
      "entropy": 6.114890623092651,
      "epoch": 2.6011535436511406,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004322864554275767,
      "loss": 5.6992,
      "mean_token_accuracy": 0.1691162943840027,
      "num_tokens": 60325345.0,
      "step": 29765
    },
    {
      "entropy": 6.081954669952393,
      "epoch": 2.601590492003845,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00043226417149370093,
      "loss": 5.6739,
      "mean_token_accuracy": 0.16777902096509933,
      "num_tokens": 60334608.0,
      "step": 29770
    },
    {
      "entropy": 6.055254602432251,
      "epoch": 2.60202744035655,
      "grad_norm": 1.5,
      "learning_rate": 0.00043224188454341593,
      "loss": 5.6732,
      "mean_token_accuracy": 0.1691708967089653,
      "num_tokens": 60346909.0,
      "step": 29775
    },
    {
      "entropy": 6.015150499343872,
      "epoch": 2.6024643887092545,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004322195945771493,
      "loss": 5.6426,
      "mean_token_accuracy": 0.17941097915172577,
      "num_tokens": 60355829.0,
      "step": 29780
    },
    {
      "entropy": 6.005622386932373,
      "epoch": 2.6029013370619594,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004321973015953284,
      "loss": 5.649,
      "mean_token_accuracy": 0.167841836810112,
      "num_tokens": 60364954.0,
      "step": 29785
    },
    {
      "entropy": 6.177559661865234,
      "epoch": 2.603338285414664,
      "grad_norm": 1.796875,
      "learning_rate": 0.00043217500559838084,
      "loss": 5.8353,
      "mean_token_accuracy": 0.15874366760253905,
      "num_tokens": 60375575.0,
      "step": 29790
    },
    {
      "entropy": 6.168684625625611,
      "epoch": 2.6037752337673687,
      "grad_norm": 2.25,
      "learning_rate": 0.00043215270658673437,
      "loss": 5.7411,
      "mean_token_accuracy": 0.1651345208287239,
      "num_tokens": 60385534.0,
      "step": 29795
    },
    {
      "entropy": 6.087994337081909,
      "epoch": 2.604212182120073,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004321304045608166,
      "loss": 5.6999,
      "mean_token_accuracy": 0.16629879474639891,
      "num_tokens": 60395517.0,
      "step": 29800
    },
    {
      "entropy": 6.160995054244995,
      "epoch": 2.604649130472778,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004321080995210554,
      "loss": 5.7748,
      "mean_token_accuracy": 0.166327366232872,
      "num_tokens": 60404781.0,
      "step": 29805
    },
    {
      "entropy": 6.069845390319824,
      "epoch": 2.605086078825483,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004320857914678785,
      "loss": 5.7207,
      "mean_token_accuracy": 0.17176818251609802,
      "num_tokens": 60415323.0,
      "step": 29810
    },
    {
      "entropy": 6.139312505722046,
      "epoch": 2.6055230271781875,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004320634804017138,
      "loss": 5.8219,
      "mean_token_accuracy": 0.15852820575237275,
      "num_tokens": 60425328.0,
      "step": 29815
    },
    {
      "entropy": 6.10943489074707,
      "epoch": 2.605959975530892,
      "grad_norm": 2.125,
      "learning_rate": 0.0004320411663229892,
      "loss": 5.7091,
      "mean_token_accuracy": 0.16666437834501266,
      "num_tokens": 60435391.0,
      "step": 29820
    },
    {
      "entropy": 6.112839221954346,
      "epoch": 2.606396923883597,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004320188492321328,
      "loss": 5.7369,
      "mean_token_accuracy": 0.16044698357582093,
      "num_tokens": 60445155.0,
      "step": 29825
    },
    {
      "entropy": 6.119149971008301,
      "epoch": 2.606833872236302,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004319965291295726,
      "loss": 5.6771,
      "mean_token_accuracy": 0.1676318272948265,
      "num_tokens": 60455655.0,
      "step": 29830
    },
    {
      "entropy": 6.062680101394653,
      "epoch": 2.6072708205890063,
      "grad_norm": 1.484375,
      "learning_rate": 0.00043197420601573665,
      "loss": 5.5785,
      "mean_token_accuracy": 0.170647594332695,
      "num_tokens": 60465080.0,
      "step": 29835
    },
    {
      "entropy": 6.041018724441528,
      "epoch": 2.607707768941711,
      "grad_norm": 1.375,
      "learning_rate": 0.00043195187989105327,
      "loss": 5.7133,
      "mean_token_accuracy": 0.16831654012203218,
      "num_tokens": 60474970.0,
      "step": 29840
    },
    {
      "entropy": 6.1333842277526855,
      "epoch": 2.6081447172944157,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043192955075595045,
      "loss": 5.6611,
      "mean_token_accuracy": 0.17251189351081847,
      "num_tokens": 60486599.0,
      "step": 29845
    },
    {
      "entropy": 6.0969321727752686,
      "epoch": 2.6085816656471206,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004319072186108567,
      "loss": 5.7591,
      "mean_token_accuracy": 0.16089060604572297,
      "num_tokens": 60496849.0,
      "step": 29850
    },
    {
      "entropy": 6.069681978225708,
      "epoch": 2.609018613999825,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004318848834562003,
      "loss": 5.7442,
      "mean_token_accuracy": 0.15960148200392724,
      "num_tokens": 60506708.0,
      "step": 29855
    },
    {
      "entropy": 5.979191398620605,
      "epoch": 2.60945556235253,
      "grad_norm": 2.09375,
      "learning_rate": 0.00043186254529240957,
      "loss": 5.6031,
      "mean_token_accuracy": 0.18529140949249268,
      "num_tokens": 60516538.0,
      "step": 29860
    },
    {
      "entropy": 6.0934412479400635,
      "epoch": 2.609892510705235,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00043184020411991315,
      "loss": 5.6405,
      "mean_token_accuracy": 0.17861660867929458,
      "num_tokens": 60526206.0,
      "step": 29865
    },
    {
      "entropy": 6.104588890075684,
      "epoch": 2.6103294590579393,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004318178599391393,
      "loss": 5.8197,
      "mean_token_accuracy": 0.1579548165202141,
      "num_tokens": 60536878.0,
      "step": 29870
    },
    {
      "entropy": 6.142395496368408,
      "epoch": 2.610766407410644,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00043179551275051675,
      "loss": 5.8786,
      "mean_token_accuracy": 0.15814365968108177,
      "num_tokens": 60547709.0,
      "step": 29875
    },
    {
      "entropy": 6.181494808197021,
      "epoch": 2.6112033557633487,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004317731625544742,
      "loss": 5.6436,
      "mean_token_accuracy": 0.17299097180366516,
      "num_tokens": 60557257.0,
      "step": 29880
    },
    {
      "entropy": 6.100407838821411,
      "epoch": 2.6116403041160536,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043175080935144015,
      "loss": 5.7622,
      "mean_token_accuracy": 0.16915906965732574,
      "num_tokens": 60568663.0,
      "step": 29885
    },
    {
      "entropy": 6.0503092288970945,
      "epoch": 2.612077252468758,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004317284531418434,
      "loss": 5.6877,
      "mean_token_accuracy": 0.1726618453860283,
      "num_tokens": 60579168.0,
      "step": 29890
    },
    {
      "entropy": 6.085426473617554,
      "epoch": 2.612514200821463,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043170609392611294,
      "loss": 5.7615,
      "mean_token_accuracy": 0.1611308053135872,
      "num_tokens": 60589893.0,
      "step": 29895
    },
    {
      "entropy": 6.081459856033325,
      "epoch": 2.6129511491741675,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043168373170467735,
      "loss": 5.6474,
      "mean_token_accuracy": 0.17665571123361587,
      "num_tokens": 60599406.0,
      "step": 29900
    },
    {
      "entropy": 6.020694017410278,
      "epoch": 2.6133880975268724,
      "grad_norm": 2.15625,
      "learning_rate": 0.00043166136647796577,
      "loss": 5.7445,
      "mean_token_accuracy": 0.1811906576156616,
      "num_tokens": 60610435.0,
      "step": 29905
    },
    {
      "entropy": 6.107263517379761,
      "epoch": 2.613825045879577,
      "grad_norm": 1.53125,
      "learning_rate": 0.000431638998246407,
      "loss": 5.7072,
      "mean_token_accuracy": 0.1698666349053383,
      "num_tokens": 60621405.0,
      "step": 29910
    },
    {
      "entropy": 6.103879165649414,
      "epoch": 2.614261994232282,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004316166270104303,
      "loss": 5.7036,
      "mean_token_accuracy": 0.16848973780870438,
      "num_tokens": 60632119.0,
      "step": 29915
    },
    {
      "entropy": 6.091409492492676,
      "epoch": 2.6146989425849867,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043159425277046454,
      "loss": 5.686,
      "mean_token_accuracy": 0.17136004269123079,
      "num_tokens": 60642163.0,
      "step": 29920
    },
    {
      "entropy": 6.066697788238526,
      "epoch": 2.615135890937691,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000431571875526939,
      "loss": 5.6735,
      "mean_token_accuracy": 0.1723114624619484,
      "num_tokens": 60652050.0,
      "step": 29925
    },
    {
      "entropy": 6.032119989395142,
      "epoch": 2.6155728392903956,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004315494952802828,
      "loss": 5.6676,
      "mean_token_accuracy": 0.17054408937692642,
      "num_tokens": 60661843.0,
      "step": 29930
    },
    {
      "entropy": 6.0475667953491214,
      "epoch": 2.6160097876431005,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043152711203092524,
      "loss": 5.6148,
      "mean_token_accuracy": 0.17548797726631166,
      "num_tokens": 60671618.0,
      "step": 29935
    },
    {
      "entropy": 6.156650447845459,
      "epoch": 2.6164467359958055,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004315047257792957,
      "loss": 5.8307,
      "mean_token_accuracy": 0.1621253915131092,
      "num_tokens": 60682930.0,
      "step": 29940
    },
    {
      "entropy": 6.201664018630981,
      "epoch": 2.61688368434851,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004314823365258235,
      "loss": 5.7481,
      "mean_token_accuracy": 0.1666848435997963,
      "num_tokens": 60692864.0,
      "step": 29945
    },
    {
      "entropy": 6.145039844512939,
      "epoch": 2.617320632701215,
      "grad_norm": 1.578125,
      "learning_rate": 0.00043145994427093806,
      "loss": 5.7295,
      "mean_token_accuracy": 0.1698861002922058,
      "num_tokens": 60702934.0,
      "step": 29950
    },
    {
      "entropy": 6.09082932472229,
      "epoch": 2.6177575810539193,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00043143754901506893,
      "loss": 5.7021,
      "mean_token_accuracy": 0.16682856678962707,
      "num_tokens": 60712289.0,
      "step": 29955
    },
    {
      "entropy": 6.0851599216461185,
      "epoch": 2.6181945294066242,
      "grad_norm": 1.59375,
      "learning_rate": 0.00043141515075864563,
      "loss": 5.655,
      "mean_token_accuracy": 0.1705339252948761,
      "num_tokens": 60723046.0,
      "step": 29960
    },
    {
      "entropy": 6.10382981300354,
      "epoch": 2.6186314777593287,
      "grad_norm": 2.015625,
      "learning_rate": 0.00043139274950209776,
      "loss": 5.7252,
      "mean_token_accuracy": 0.16149101108312608,
      "num_tokens": 60731969.0,
      "step": 29965
    },
    {
      "entropy": 6.117691516876221,
      "epoch": 2.6190684261120336,
      "grad_norm": 2.578125,
      "learning_rate": 0.00043137034524585506,
      "loss": 5.8287,
      "mean_token_accuracy": 0.16146129071712495,
      "num_tokens": 60743021.0,
      "step": 29970
    },
    {
      "entropy": 6.1234519481658936,
      "epoch": 2.6195053744647385,
      "grad_norm": 1.625,
      "learning_rate": 0.0004313479379903472,
      "loss": 5.712,
      "mean_token_accuracy": 0.17430299818515776,
      "num_tokens": 60751854.0,
      "step": 29975
    },
    {
      "entropy": 6.157818794250488,
      "epoch": 2.619942322817443,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004313255277360039,
      "loss": 5.7014,
      "mean_token_accuracy": 0.16314127743244172,
      "num_tokens": 60761572.0,
      "step": 29980
    },
    {
      "entropy": 6.018813991546631,
      "epoch": 2.6203792711701475,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004313031144832551,
      "loss": 5.6616,
      "mean_token_accuracy": 0.1699492394924164,
      "num_tokens": 60771618.0,
      "step": 29985
    },
    {
      "entropy": 6.105633401870728,
      "epoch": 2.6208162195228524,
      "grad_norm": 2.09375,
      "learning_rate": 0.00043128069823253064,
      "loss": 5.6639,
      "mean_token_accuracy": 0.17439507246017455,
      "num_tokens": 60781038.0,
      "step": 29990
    },
    {
      "entropy": 6.105250835418701,
      "epoch": 2.6212531678755573,
      "grad_norm": 2.796875,
      "learning_rate": 0.00043125827898426056,
      "loss": 5.7375,
      "mean_token_accuracy": 0.16617934703826903,
      "num_tokens": 60792433.0,
      "step": 29995
    },
    {
      "entropy": 6.055356216430664,
      "epoch": 2.6216901162282618,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043123585673887475,
      "loss": 5.6724,
      "mean_token_accuracy": 0.17386500984430314,
      "num_tokens": 60801995.0,
      "step": 30000
    },
    {
      "epoch": 2.6216901162282618,
      "eval_entropy": 5.839139696689669,
      "eval_loss": 5.758738040924072,
      "eval_mean_token_accuracy": 0.17481668004276996,
      "eval_num_tokens": 60801995.0,
      "eval_runtime": 27.8928,
      "eval_samples_per_second": 1327.01,
      "eval_steps_per_second": 165.885,
      "step": 30000
    },
    {
      "entropy": 6.057730627059937,
      "epoch": 2.6221270645809667,
      "grad_norm": 1.125,
      "learning_rate": 0.0004312134314968034,
      "loss": 5.7024,
      "mean_token_accuracy": 0.17013068795204161,
      "num_tokens": 60811784.0,
      "step": 30005
    },
    {
      "entropy": 6.072100448608398,
      "epoch": 2.622564012933671,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043119100325847654,
      "loss": 5.6639,
      "mean_token_accuracy": 0.17628273367881775,
      "num_tokens": 60822018.0,
      "step": 30010
    },
    {
      "entropy": 6.068133115768433,
      "epoch": 2.623000961286376,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004311685720243243,
      "loss": 5.7075,
      "mean_token_accuracy": 0.16851950585842132,
      "num_tokens": 60832204.0,
      "step": 30015
    },
    {
      "entropy": 6.076402568817139,
      "epoch": 2.6234379096390805,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004311461377947772,
      "loss": 5.6838,
      "mean_token_accuracy": 0.1692726045846939,
      "num_tokens": 60843293.0,
      "step": 30020
    },
    {
      "entropy": 6.098486232757568,
      "epoch": 2.6238748579917854,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004311237005702652,
      "loss": 5.6459,
      "mean_token_accuracy": 0.1684374988079071,
      "num_tokens": 60852323.0,
      "step": 30025
    },
    {
      "entropy": 6.012970447540283,
      "epoch": 2.6243118063444903,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004311012603512189,
      "loss": 5.7028,
      "mean_token_accuracy": 0.17108615040779113,
      "num_tokens": 60862788.0,
      "step": 30030
    },
    {
      "entropy": 6.094996213912964,
      "epoch": 2.624748754697195,
      "grad_norm": 2.421875,
      "learning_rate": 0.00043107881713806863,
      "loss": 5.7113,
      "mean_token_accuracy": 0.16905009597539902,
      "num_tokens": 60872793.0,
      "step": 30035
    },
    {
      "entropy": 6.0796403884887695,
      "epoch": 2.6251857030498993,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043105637093124484,
      "loss": 5.6702,
      "mean_token_accuracy": 0.17324941754341125,
      "num_tokens": 60882712.0,
      "step": 30040
    },
    {
      "entropy": 6.103579664230347,
      "epoch": 2.625622651402604,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00043103392173117814,
      "loss": 5.7059,
      "mean_token_accuracy": 0.17402534931898117,
      "num_tokens": 60892400.0,
      "step": 30045
    },
    {
      "entropy": 6.097170734405518,
      "epoch": 2.626059599755309,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043101146953829895,
      "loss": 5.6577,
      "mean_token_accuracy": 0.17465276569128035,
      "num_tokens": 60902480.0,
      "step": 30050
    },
    {
      "entropy": 6.0618688583374025,
      "epoch": 2.6264965481080136,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004309890143530382,
      "loss": 5.7115,
      "mean_token_accuracy": 0.17143333256244658,
      "num_tokens": 60912183.0,
      "step": 30055
    },
    {
      "entropy": 6.140414190292359,
      "epoch": 2.6269334964607185,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004309665561758262,
      "loss": 5.6936,
      "mean_token_accuracy": 0.17062972486019135,
      "num_tokens": 60922166.0,
      "step": 30060
    },
    {
      "entropy": 6.085996007919311,
      "epoch": 2.627370444813423,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00043094409500709404,
      "loss": 5.647,
      "mean_token_accuracy": 0.17352185249328614,
      "num_tokens": 60931980.0,
      "step": 30065
    },
    {
      "entropy": 6.036578130722046,
      "epoch": 2.627807393166128,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004309216308472725,
      "loss": 5.8576,
      "mean_token_accuracy": 0.16856524646282195,
      "num_tokens": 60943382.0,
      "step": 30070
    },
    {
      "entropy": 6.125305604934693,
      "epoch": 2.6282443415188323,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004308991636967923,
      "loss": 5.7327,
      "mean_token_accuracy": 0.16484911292791365,
      "num_tokens": 60953505.0,
      "step": 30075
    },
    {
      "entropy": 6.164077281951904,
      "epoch": 2.6286812898715373,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004308766935560844,
      "loss": 5.7246,
      "mean_token_accuracy": 0.1729841262102127,
      "num_tokens": 60963851.0,
      "step": 30080
    },
    {
      "entropy": 6.117432260513306,
      "epoch": 2.6291182382242417,
      "grad_norm": 1.40625,
      "learning_rate": 0.00043085422042557985,
      "loss": 5.7642,
      "mean_token_accuracy": 0.16490877270698548,
      "num_tokens": 60974919.0,
      "step": 30085
    },
    {
      "entropy": 6.125683116912842,
      "epoch": 2.6295551865769466,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00043083174430570974,
      "loss": 5.7393,
      "mean_token_accuracy": 0.16832350045442582,
      "num_tokens": 60983866.0,
      "step": 30090
    },
    {
      "entropy": 6.048496055603027,
      "epoch": 2.629992134929651,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00043080926519690517,
      "loss": 5.6639,
      "mean_token_accuracy": 0.17213201969861985,
      "num_tokens": 60994393.0,
      "step": 30095
    },
    {
      "entropy": 6.1524176597595215,
      "epoch": 2.630429083282356,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043078678309959716,
      "loss": 5.7502,
      "mean_token_accuracy": 0.1763453811407089,
      "num_tokens": 61004634.0,
      "step": 30100
    },
    {
      "entropy": 6.029833221435547,
      "epoch": 2.630866031635061,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004307642980142171,
      "loss": 5.6958,
      "mean_token_accuracy": 0.17050536423921586,
      "num_tokens": 61014771.0,
      "step": 30105
    },
    {
      "entropy": 6.098295211791992,
      "epoch": 2.6313029799877654,
      "grad_norm": 1.734375,
      "learning_rate": 0.00043074180994119604,
      "loss": 5.7538,
      "mean_token_accuracy": 0.1669932037591934,
      "num_tokens": 61025213.0,
      "step": 30110
    },
    {
      "entropy": 6.151889038085938,
      "epoch": 2.63173992834047,
      "grad_norm": 1.25,
      "learning_rate": 0.0004307193188809656,
      "loss": 5.6489,
      "mean_token_accuracy": 0.17466482967138292,
      "num_tokens": 61035601.0,
      "step": 30115
    },
    {
      "entropy": 5.916748428344727,
      "epoch": 2.632176876693175,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043069682483395694,
      "loss": 5.6881,
      "mean_token_accuracy": 0.17276469618082047,
      "num_tokens": 61045862.0,
      "step": 30120
    },
    {
      "entropy": 6.1435178279876705,
      "epoch": 2.6326138250458797,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00043067432780060163,
      "loss": 5.6844,
      "mean_token_accuracy": 0.17637910395860673,
      "num_tokens": 61056702.0,
      "step": 30125
    },
    {
      "entropy": 6.112952661514282,
      "epoch": 2.633050773398584,
      "grad_norm": 1.5,
      "learning_rate": 0.00043065182778133116,
      "loss": 5.7376,
      "mean_token_accuracy": 0.16361093372106553,
      "num_tokens": 61066353.0,
      "step": 30130
    },
    {
      "entropy": 6.1124502658844,
      "epoch": 2.633487721751289,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000430629324776577,
      "loss": 5.6719,
      "mean_token_accuracy": 0.1687804013490677,
      "num_tokens": 61076534.0,
      "step": 30135
    },
    {
      "entropy": 6.080886316299439,
      "epoch": 2.6339246701039936,
      "grad_norm": 1.671875,
      "learning_rate": 0.00043060681878677096,
      "loss": 5.6007,
      "mean_token_accuracy": 0.17626486122608184,
      "num_tokens": 61086635.0,
      "step": 30140
    },
    {
      "entropy": 6.062822723388672,
      "epoch": 2.6343616184566985,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00043058430981234453,
      "loss": 5.7133,
      "mean_token_accuracy": 0.16352597624063492,
      "num_tokens": 61096605.0,
      "step": 30145
    },
    {
      "entropy": 6.066860151290894,
      "epoch": 2.634798566809403,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004305617978537295,
      "loss": 5.6935,
      "mean_token_accuracy": 0.1751556098461151,
      "num_tokens": 61106465.0,
      "step": 30150
    },
    {
      "entropy": 6.118218660354614,
      "epoch": 2.635235515162108,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004305392829113577,
      "loss": 5.7634,
      "mean_token_accuracy": 0.17068621516227722,
      "num_tokens": 61116063.0,
      "step": 30155
    },
    {
      "entropy": 6.112340497970581,
      "epoch": 2.6356724635148128,
      "grad_norm": 1.5625,
      "learning_rate": 0.00043051676498566095,
      "loss": 5.7125,
      "mean_token_accuracy": 0.16257734149694442,
      "num_tokens": 61125794.0,
      "step": 30160
    },
    {
      "entropy": 6.096731328964234,
      "epoch": 2.6361094118675172,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004304942440770712,
      "loss": 5.709,
      "mean_token_accuracy": 0.1723858654499054,
      "num_tokens": 61135734.0,
      "step": 30165
    },
    {
      "entropy": 6.048110008239746,
      "epoch": 2.6365463602202217,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004304717201860203,
      "loss": 5.6763,
      "mean_token_accuracy": 0.17458571046590804,
      "num_tokens": 61146071.0,
      "step": 30170
    },
    {
      "entropy": 6.057622337341309,
      "epoch": 2.6369833085729266,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004304491933129404,
      "loss": 5.6602,
      "mean_token_accuracy": 0.17482249438762665,
      "num_tokens": 61154931.0,
      "step": 30175
    },
    {
      "entropy": 6.096342515945435,
      "epoch": 2.6374202569256315,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004304266634582634,
      "loss": 5.7316,
      "mean_token_accuracy": 0.17057165801525115,
      "num_tokens": 61164464.0,
      "step": 30180
    },
    {
      "entropy": 6.145320320129395,
      "epoch": 2.637857205278336,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004304041306224217,
      "loss": 5.7169,
      "mean_token_accuracy": 0.16747799813747405,
      "num_tokens": 61175379.0,
      "step": 30185
    },
    {
      "entropy": 6.056752490997314,
      "epoch": 2.638294153631041,
      "grad_norm": 2.28125,
      "learning_rate": 0.00043038159480584723,
      "loss": 5.6452,
      "mean_token_accuracy": 0.1743887260556221,
      "num_tokens": 61186340.0,
      "step": 30190
    },
    {
      "entropy": 6.073797225952148,
      "epoch": 2.6387311019837454,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00043035905600897237,
      "loss": 5.6318,
      "mean_token_accuracy": 0.17872558087110518,
      "num_tokens": 61196495.0,
      "step": 30195
    },
    {
      "entropy": 6.116700839996338,
      "epoch": 2.6391680503364503,
      "grad_norm": 1.796875,
      "learning_rate": 0.00043033651423222947,
      "loss": 5.7423,
      "mean_token_accuracy": 0.16871413290500642,
      "num_tokens": 61206555.0,
      "step": 30200
    },
    {
      "entropy": 6.144318532943726,
      "epoch": 2.6396049986891548,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004303139694760508,
      "loss": 5.7604,
      "mean_token_accuracy": 0.16331293061375618,
      "num_tokens": 61215818.0,
      "step": 30205
    },
    {
      "entropy": 6.123166131973266,
      "epoch": 2.6400419470418597,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00043029142174086875,
      "loss": 5.7126,
      "mean_token_accuracy": 0.1690562605857849,
      "num_tokens": 61226325.0,
      "step": 30210
    },
    {
      "entropy": 6.102904891967773,
      "epoch": 2.6404788953945646,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004302688710271159,
      "loss": 5.7197,
      "mean_token_accuracy": 0.1634292796254158,
      "num_tokens": 61236179.0,
      "step": 30215
    },
    {
      "entropy": 6.018401050567627,
      "epoch": 2.640915843747269,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00043024631733522484,
      "loss": 5.6564,
      "mean_token_accuracy": 0.17611150443553925,
      "num_tokens": 61245680.0,
      "step": 30220
    },
    {
      "entropy": 6.10023341178894,
      "epoch": 2.6413527920999735,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000430223760665628,
      "loss": 5.754,
      "mean_token_accuracy": 0.16287420690059662,
      "num_tokens": 61256369.0,
      "step": 30225
    },
    {
      "entropy": 6.0136888980865475,
      "epoch": 2.6417897404526784,
      "grad_norm": 1.546875,
      "learning_rate": 0.00043020120101875807,
      "loss": 5.6146,
      "mean_token_accuracy": 0.17174480259418487,
      "num_tokens": 61266395.0,
      "step": 30230
    },
    {
      "entropy": 6.1216278076171875,
      "epoch": 2.6422266888053834,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043017863839504784,
      "loss": 5.7449,
      "mean_token_accuracy": 0.17304136902093886,
      "num_tokens": 61276442.0,
      "step": 30235
    },
    {
      "entropy": 6.124312448501587,
      "epoch": 2.642663637158088,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00043015607279493,
      "loss": 5.7652,
      "mean_token_accuracy": 0.1650170385837555,
      "num_tokens": 61286230.0,
      "step": 30240
    },
    {
      "entropy": 6.08577880859375,
      "epoch": 2.6431005855107927,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004301335042188373,
      "loss": 5.6665,
      "mean_token_accuracy": 0.16676109731197358,
      "num_tokens": 61297853.0,
      "step": 30245
    },
    {
      "entropy": 6.060123062133789,
      "epoch": 2.643537533863497,
      "grad_norm": 2.34375,
      "learning_rate": 0.00043011093266720273,
      "loss": 5.6304,
      "mean_token_accuracy": 0.17449421137571336,
      "num_tokens": 61307976.0,
      "step": 30250
    },
    {
      "entropy": 6.1142106533050535,
      "epoch": 2.643974482216202,
      "grad_norm": 1.546875,
      "learning_rate": 0.00043008835814045926,
      "loss": 5.7223,
      "mean_token_accuracy": 0.16669000089168548,
      "num_tokens": 61319101.0,
      "step": 30255
    },
    {
      "entropy": 6.130772590637207,
      "epoch": 2.6444114305689066,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004300657806390398,
      "loss": 5.6981,
      "mean_token_accuracy": 0.16953193098306657,
      "num_tokens": 61329492.0,
      "step": 30260
    },
    {
      "entropy": 6.0963622570037845,
      "epoch": 2.6448483789216115,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043004320016337737,
      "loss": 5.7678,
      "mean_token_accuracy": 0.1632098913192749,
      "num_tokens": 61340480.0,
      "step": 30265
    },
    {
      "entropy": 6.111527585983277,
      "epoch": 2.6452853272743164,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043002061671390513,
      "loss": 5.6617,
      "mean_token_accuracy": 0.1711018979549408,
      "num_tokens": 61350572.0,
      "step": 30270
    },
    {
      "entropy": 6.157809400558472,
      "epoch": 2.645722275627021,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004299980302910563,
      "loss": 5.7286,
      "mean_token_accuracy": 0.1678188219666481,
      "num_tokens": 61361329.0,
      "step": 30275
    },
    {
      "entropy": 6.14020094871521,
      "epoch": 2.6461592239797254,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004299754408952639,
      "loss": 5.7861,
      "mean_token_accuracy": 0.1633751228451729,
      "num_tokens": 61372618.0,
      "step": 30280
    },
    {
      "entropy": 6.141544961929322,
      "epoch": 2.6465961723324303,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004299528485269614,
      "loss": 5.8087,
      "mean_token_accuracy": 0.15703094601631165,
      "num_tokens": 61383831.0,
      "step": 30285
    },
    {
      "entropy": 6.05940523147583,
      "epoch": 2.647033120685135,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004299302531865821,
      "loss": 5.6682,
      "mean_token_accuracy": 0.16942809224128724,
      "num_tokens": 61394527.0,
      "step": 30290
    },
    {
      "entropy": 6.069051265716553,
      "epoch": 2.6474700690378397,
      "grad_norm": 2.203125,
      "learning_rate": 0.00042990765487455935,
      "loss": 5.7802,
      "mean_token_accuracy": 0.1641063928604126,
      "num_tokens": 61404787.0,
      "step": 30295
    },
    {
      "entropy": 6.058274936676026,
      "epoch": 2.6479070173905446,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004298850535913266,
      "loss": 5.5785,
      "mean_token_accuracy": 0.1864791542291641,
      "num_tokens": 61413922.0,
      "step": 30300
    },
    {
      "entropy": 6.120040941238403,
      "epoch": 2.648343965743249,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004298624493373174,
      "loss": 5.6955,
      "mean_token_accuracy": 0.16904531866312028,
      "num_tokens": 61424750.0,
      "step": 30305
    },
    {
      "entropy": 6.0524452209472654,
      "epoch": 2.648780914095954,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004298398421129653,
      "loss": 5.6382,
      "mean_token_accuracy": 0.17793624252080917,
      "num_tokens": 61435222.0,
      "step": 30310
    },
    {
      "entropy": 6.049088001251221,
      "epoch": 2.6492178624486584,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004298172319187038,
      "loss": 5.6911,
      "mean_token_accuracy": 0.17791264206171037,
      "num_tokens": 61444230.0,
      "step": 30315
    },
    {
      "entropy": 6.056098699569702,
      "epoch": 2.6496548108013633,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004297946187549667,
      "loss": 5.7169,
      "mean_token_accuracy": 0.16537636667490005,
      "num_tokens": 61454975.0,
      "step": 30320
    },
    {
      "entropy": 6.151819181442261,
      "epoch": 2.6500917591540682,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042977200262218776,
      "loss": 5.79,
      "mean_token_accuracy": 0.1662588283419609,
      "num_tokens": 61465563.0,
      "step": 30325
    },
    {
      "entropy": 6.0966259956359865,
      "epoch": 2.6505287075067727,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004297493835208006,
      "loss": 5.7305,
      "mean_token_accuracy": 0.16922746300697328,
      "num_tokens": 61475330.0,
      "step": 30330
    },
    {
      "entropy": 6.119055986404419,
      "epoch": 2.650965655859477,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00042972676145123926,
      "loss": 5.7711,
      "mean_token_accuracy": 0.16394346356391906,
      "num_tokens": 61485577.0,
      "step": 30335
    },
    {
      "entropy": 6.1502806663513185,
      "epoch": 2.651402604212182,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042970413641393754,
      "loss": 5.6867,
      "mean_token_accuracy": 0.1687193289399147,
      "num_tokens": 61495502.0,
      "step": 30340
    },
    {
      "entropy": 6.082031965255737,
      "epoch": 2.651839552564887,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004296815084093294,
      "loss": 5.735,
      "mean_token_accuracy": 0.16486389338970184,
      "num_tokens": 61505383.0,
      "step": 30345
    },
    {
      "entropy": 6.034027194976806,
      "epoch": 2.6522765009175915,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042965887743784886,
      "loss": 5.7037,
      "mean_token_accuracy": 0.1754722610116005,
      "num_tokens": 61515052.0,
      "step": 30350
    },
    {
      "entropy": 6.092453670501709,
      "epoch": 2.6527134492702964,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004296362434999301,
      "loss": 5.6655,
      "mean_token_accuracy": 0.17466034442186357,
      "num_tokens": 61525139.0,
      "step": 30355
    },
    {
      "entropy": 6.057743692398072,
      "epoch": 2.653150397623001,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042961360659600706,
      "loss": 5.6112,
      "mean_token_accuracy": 0.1753731846809387,
      "num_tokens": 61534901.0,
      "step": 30360
    },
    {
      "entropy": 6.055743741989136,
      "epoch": 2.653587345975706,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004295909667265141,
      "loss": 5.6962,
      "mean_token_accuracy": 0.16789068430662155,
      "num_tokens": 61543258.0,
      "step": 30365
    },
    {
      "entropy": 6.056825780868531,
      "epoch": 2.6540242943284102,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00042956832389188534,
      "loss": 5.6728,
      "mean_token_accuracy": 0.1704175978899002,
      "num_tokens": 61553294.0,
      "step": 30370
    },
    {
      "entropy": 6.1096436977386475,
      "epoch": 2.654461242681115,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004295456780925551,
      "loss": 5.6743,
      "mean_token_accuracy": 0.1786956638097763,
      "num_tokens": 61562986.0,
      "step": 30375
    },
    {
      "entropy": 6.032594823837281,
      "epoch": 2.65489819103382,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042952302932895784,
      "loss": 5.7368,
      "mean_token_accuracy": 0.17598538845777512,
      "num_tokens": 61573994.0,
      "step": 30380
    },
    {
      "entropy": 6.107252073287964,
      "epoch": 2.6553351393865245,
      "grad_norm": 2.28125,
      "learning_rate": 0.00042950037760152793,
      "loss": 5.6671,
      "mean_token_accuracy": 0.17260402143001558,
      "num_tokens": 61584279.0,
      "step": 30385
    },
    {
      "entropy": 6.153750801086426,
      "epoch": 2.655772087739229,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004294777229106998,
      "loss": 5.7715,
      "mean_token_accuracy": 0.16389036774635315,
      "num_tokens": 61595188.0,
      "step": 30390
    },
    {
      "entropy": 6.052868461608886,
      "epoch": 2.656209036091934,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000429455065256908,
      "loss": 5.7727,
      "mean_token_accuracy": 0.1678144872188568,
      "num_tokens": 61605321.0,
      "step": 30395
    },
    {
      "entropy": 6.143194770812988,
      "epoch": 2.656645984444639,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004294324046405871,
      "loss": 5.7392,
      "mean_token_accuracy": 0.1684527650475502,
      "num_tokens": 61615137.0,
      "step": 30400
    },
    {
      "entropy": 6.066177034378052,
      "epoch": 2.6570829327973433,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042940974106217176,
      "loss": 5.6005,
      "mean_token_accuracy": 0.17232774049043656,
      "num_tokens": 61625912.0,
      "step": 30405
    },
    {
      "entropy": 6.065654277801514,
      "epoch": 2.6575198811500482,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004293870745220967,
      "loss": 5.6516,
      "mean_token_accuracy": 0.17138356864452362,
      "num_tokens": 61635836.0,
      "step": 30410
    },
    {
      "entropy": 6.075814628601075,
      "epoch": 2.6579568295027527,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004293644050207966,
      "loss": 5.6975,
      "mean_token_accuracy": 0.1696704924106598,
      "num_tokens": 61645151.0,
      "step": 30415
    },
    {
      "entropy": 6.130489635467529,
      "epoch": 2.6583937778554576,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004293417325587064,
      "loss": 5.7211,
      "mean_token_accuracy": 0.16937721967697145,
      "num_tokens": 61654266.0,
      "step": 30420
    },
    {
      "entropy": 6.117292261123657,
      "epoch": 2.658830726208162,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004293190571362608,
      "loss": 5.8598,
      "mean_token_accuracy": 0.157978917658329,
      "num_tokens": 61664355.0,
      "step": 30425
    },
    {
      "entropy": 6.169596338272095,
      "epoch": 2.659267674560867,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042929637875389486,
      "loss": 5.7777,
      "mean_token_accuracy": 0.1684722788631916,
      "num_tokens": 61674130.0,
      "step": 30430
    },
    {
      "entropy": 6.170914125442505,
      "epoch": 2.6597046229135715,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042927369741204357,
      "loss": 5.7448,
      "mean_token_accuracy": 0.16827378720045089,
      "num_tokens": 61683612.0,
      "step": 30435
    },
    {
      "entropy": 6.079743385314941,
      "epoch": 2.6601415712662764,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004292510131111419,
      "loss": 5.7488,
      "mean_token_accuracy": 0.17512442469596862,
      "num_tokens": 61693156.0,
      "step": 30440
    },
    {
      "entropy": 6.046451568603516,
      "epoch": 2.660578519618981,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042922832585162496,
      "loss": 5.5907,
      "mean_token_accuracy": 0.17186235040426254,
      "num_tokens": 61701925.0,
      "step": 30445
    },
    {
      "entropy": 6.048460817337036,
      "epoch": 2.6610154679716858,
      "grad_norm": 1.625,
      "learning_rate": 0.00042920563563392803,
      "loss": 5.7003,
      "mean_token_accuracy": 0.17680263072252272,
      "num_tokens": 61712388.0,
      "step": 30450
    },
    {
      "entropy": 6.08034381866455,
      "epoch": 2.6614524163243907,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00042918294245848605,
      "loss": 5.6882,
      "mean_token_accuracy": 0.1658626079559326,
      "num_tokens": 61722402.0,
      "step": 30455
    },
    {
      "entropy": 6.155047130584717,
      "epoch": 2.661889364677095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042916024632573454,
      "loss": 5.7625,
      "mean_token_accuracy": 0.16373366117477417,
      "num_tokens": 61733160.0,
      "step": 30460
    },
    {
      "entropy": 6.111729097366333,
      "epoch": 2.6623263130297996,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042913754723610865,
      "loss": 5.6819,
      "mean_token_accuracy": 0.17478953748941423,
      "num_tokens": 61742536.0,
      "step": 30465
    },
    {
      "entropy": 6.099606275558472,
      "epoch": 2.6627632613825045,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004291148451900439,
      "loss": 5.7107,
      "mean_token_accuracy": 0.16653266847133635,
      "num_tokens": 61753439.0,
      "step": 30470
    },
    {
      "entropy": 6.075577640533448,
      "epoch": 2.6632002097352094,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00042909214018797565,
      "loss": 5.721,
      "mean_token_accuracy": 0.1745551571249962,
      "num_tokens": 61763994.0,
      "step": 30475
    },
    {
      "entropy": 6.09922308921814,
      "epoch": 2.663637158087914,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004290694322303394,
      "loss": 5.7459,
      "mean_token_accuracy": 0.16643654853105544,
      "num_tokens": 61774082.0,
      "step": 30480
    },
    {
      "entropy": 6.095183181762695,
      "epoch": 2.664074106440619,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042904672131757065,
      "loss": 5.6779,
      "mean_token_accuracy": 0.17525109201669692,
      "num_tokens": 61784528.0,
      "step": 30485
    },
    {
      "entropy": 6.093565559387207,
      "epoch": 2.6645110547933233,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004290240074501051,
      "loss": 5.6723,
      "mean_token_accuracy": 0.1723364546895027,
      "num_tokens": 61794267.0,
      "step": 30490
    },
    {
      "entropy": 6.06688084602356,
      "epoch": 2.664948003146028,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004290012906283784,
      "loss": 5.6811,
      "mean_token_accuracy": 0.17172384709119798,
      "num_tokens": 61804842.0,
      "step": 30495
    },
    {
      "entropy": 6.06651554107666,
      "epoch": 2.6653849514987327,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042897857085282627,
      "loss": 5.6769,
      "mean_token_accuracy": 0.1689901754260063,
      "num_tokens": 61814316.0,
      "step": 30500
    },
    {
      "entropy": 6.152177476882935,
      "epoch": 2.6658218998514376,
      "grad_norm": 2.578125,
      "learning_rate": 0.00042895584812388445,
      "loss": 5.6988,
      "mean_token_accuracy": 0.1719392016530037,
      "num_tokens": 61823828.0,
      "step": 30505
    },
    {
      "entropy": 6.106168651580811,
      "epoch": 2.6662588482041425,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004289331224419888,
      "loss": 5.7066,
      "mean_token_accuracy": 0.17039956897497177,
      "num_tokens": 61833878.0,
      "step": 30510
    },
    {
      "entropy": 6.0800230503082275,
      "epoch": 2.666695796556847,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00042891039380757516,
      "loss": 5.6631,
      "mean_token_accuracy": 0.17121603339910507,
      "num_tokens": 61844591.0,
      "step": 30515
    },
    {
      "entropy": 6.068757390975952,
      "epoch": 2.6671327449095514,
      "grad_norm": 1.65625,
      "learning_rate": 0.00042888766222107954,
      "loss": 5.7148,
      "mean_token_accuracy": 0.17324168235063553,
      "num_tokens": 61854791.0,
      "step": 30520
    },
    {
      "entropy": 6.085429906845093,
      "epoch": 2.6675696932622563,
      "grad_norm": 1.671875,
      "learning_rate": 0.000428864927682938,
      "loss": 5.7334,
      "mean_token_accuracy": 0.17206884622573854,
      "num_tokens": 61865194.0,
      "step": 30525
    },
    {
      "entropy": 6.054072618484497,
      "epoch": 2.6680066416149613,
      "grad_norm": 1.421875,
      "learning_rate": 0.00042884219019358644,
      "loss": 5.6476,
      "mean_token_accuracy": 0.16845704466104508,
      "num_tokens": 61874827.0,
      "step": 30530
    },
    {
      "entropy": 6.124270105361939,
      "epoch": 2.6684435899676657,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004288194497534611,
      "loss": 5.7325,
      "mean_token_accuracy": 0.16577837914228438,
      "num_tokens": 61885596.0,
      "step": 30535
    },
    {
      "entropy": 6.082011270523071,
      "epoch": 2.6688805383203706,
      "grad_norm": 1.84375,
      "learning_rate": 0.00042879670636299805,
      "loss": 5.5735,
      "mean_token_accuracy": 0.17508003264665603,
      "num_tokens": 61895732.0,
      "step": 30540
    },
    {
      "entropy": 6.051146697998047,
      "epoch": 2.669317486673075,
      "grad_norm": 1.640625,
      "learning_rate": 0.00042877396002263357,
      "loss": 5.644,
      "mean_token_accuracy": 0.1664162501692772,
      "num_tokens": 61905581.0,
      "step": 30545
    },
    {
      "entropy": 6.030200052261352,
      "epoch": 2.66975443502578,
      "grad_norm": 2.5,
      "learning_rate": 0.00042875121073280404,
      "loss": 5.6196,
      "mean_token_accuracy": 0.17660651206970215,
      "num_tokens": 61914926.0,
      "step": 30550
    },
    {
      "entropy": 6.086882543563843,
      "epoch": 2.6701913833784845,
      "grad_norm": 7.90625,
      "learning_rate": 0.00042872845849394563,
      "loss": 5.6871,
      "mean_token_accuracy": 0.17159554809331895,
      "num_tokens": 61926501.0,
      "step": 30555
    },
    {
      "entropy": 6.13486065864563,
      "epoch": 2.6706283317311894,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042870570330649495,
      "loss": 5.7501,
      "mean_token_accuracy": 0.16903429478406906,
      "num_tokens": 61937161.0,
      "step": 30560
    },
    {
      "entropy": 6.080095958709717,
      "epoch": 2.6710652800838943,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004286829451708883,
      "loss": 5.6648,
      "mean_token_accuracy": 0.17020538896322251,
      "num_tokens": 61949125.0,
      "step": 30565
    },
    {
      "entropy": 6.012630224227905,
      "epoch": 2.671502228436599,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004286601840875622,
      "loss": 5.5521,
      "mean_token_accuracy": 0.181418976187706,
      "num_tokens": 61958110.0,
      "step": 30570
    },
    {
      "entropy": 6.023348188400268,
      "epoch": 2.6719391767893033,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004286374200569533,
      "loss": 5.6707,
      "mean_token_accuracy": 0.17073200047016143,
      "num_tokens": 61967805.0,
      "step": 30575
    },
    {
      "entropy": 6.063754415512085,
      "epoch": 2.672376125142008,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042861465307949815,
      "loss": 5.6893,
      "mean_token_accuracy": 0.17760273814201355,
      "num_tokens": 61977189.0,
      "step": 30580
    },
    {
      "entropy": 6.113606929779053,
      "epoch": 2.672813073494713,
      "grad_norm": 1.8125,
      "learning_rate": 0.00042859188315563345,
      "loss": 5.7785,
      "mean_token_accuracy": 0.17113002985715867,
      "num_tokens": 61988568.0,
      "step": 30585
    },
    {
      "entropy": 6.153583097457886,
      "epoch": 2.6732500218474176,
      "grad_norm": 2.734375,
      "learning_rate": 0.00042856911028579597,
      "loss": 5.6438,
      "mean_token_accuracy": 0.1712859183549881,
      "num_tokens": 61999316.0,
      "step": 30590
    },
    {
      "entropy": 6.091546821594238,
      "epoch": 2.6736869702001225,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00042854633447042247,
      "loss": 5.7059,
      "mean_token_accuracy": 0.17300911098718644,
      "num_tokens": 62009345.0,
      "step": 30595
    },
    {
      "entropy": 6.04045262336731,
      "epoch": 2.674123918552827,
      "grad_norm": 4.3125,
      "learning_rate": 0.00042852355570994983,
      "loss": 5.6821,
      "mean_token_accuracy": 0.16911975890398026,
      "num_tokens": 62020847.0,
      "step": 30600
    },
    {
      "entropy": 6.036492729187012,
      "epoch": 2.674560866905532,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00042850077400481495,
      "loss": 5.6447,
      "mean_token_accuracy": 0.17648502588272094,
      "num_tokens": 62030506.0,
      "step": 30605
    },
    {
      "entropy": 6.142688131332397,
      "epoch": 2.6749978152582363,
      "grad_norm": 2.03125,
      "learning_rate": 0.00042847798935545484,
      "loss": 5.7744,
      "mean_token_accuracy": 0.1661285638809204,
      "num_tokens": 62040238.0,
      "step": 30610
    },
    {
      "entropy": 6.079429864883423,
      "epoch": 2.6754347636109412,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004284552017623064,
      "loss": 5.6723,
      "mean_token_accuracy": 0.1745912715792656,
      "num_tokens": 62050192.0,
      "step": 30615
    },
    {
      "entropy": 6.096246910095215,
      "epoch": 2.675871711963646,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004284324112258068,
      "loss": 5.6888,
      "mean_token_accuracy": 0.17719174772500992,
      "num_tokens": 62059475.0,
      "step": 30620
    },
    {
      "entropy": 6.076967191696167,
      "epoch": 2.6763086603163506,
      "grad_norm": 2.375,
      "learning_rate": 0.00042840961774639314,
      "loss": 5.6694,
      "mean_token_accuracy": 0.17043835669755936,
      "num_tokens": 62068799.0,
      "step": 30625
    },
    {
      "entropy": 6.109461450576783,
      "epoch": 2.676745608669055,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004283868213245026,
      "loss": 5.7522,
      "mean_token_accuracy": 0.1641609013080597,
      "num_tokens": 62078205.0,
      "step": 30630
    },
    {
      "entropy": 6.039348840713501,
      "epoch": 2.67718255702176,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042836402196057255,
      "loss": 5.6063,
      "mean_token_accuracy": 0.1778814122080803,
      "num_tokens": 62088084.0,
      "step": 30635
    },
    {
      "entropy": 6.065054035186767,
      "epoch": 2.677619505374465,
      "grad_norm": 2.171875,
      "learning_rate": 0.00042834121965504014,
      "loss": 5.6248,
      "mean_token_accuracy": 0.17826693058013915,
      "num_tokens": 62098709.0,
      "step": 30640
    },
    {
      "entropy": 5.96101450920105,
      "epoch": 2.6780564537271694,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004283184144083428,
      "loss": 5.6089,
      "mean_token_accuracy": 0.17637152671813966,
      "num_tokens": 62108949.0,
      "step": 30645
    },
    {
      "entropy": 6.041886329650879,
      "epoch": 2.6784934020798743,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00042829560622091795,
      "loss": 5.689,
      "mean_token_accuracy": 0.16724461913108826,
      "num_tokens": 62118922.0,
      "step": 30650
    },
    {
      "entropy": 6.131612777709961,
      "epoch": 2.6789303504325788,
      "grad_norm": 1.4375,
      "learning_rate": 0.00042827279509320297,
      "loss": 5.6809,
      "mean_token_accuracy": 0.17342716455459595,
      "num_tokens": 62128396.0,
      "step": 30655
    },
    {
      "entropy": 6.1301689624786375,
      "epoch": 2.6793672987852837,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004282499810256356,
      "loss": 5.7372,
      "mean_token_accuracy": 0.16341442316770555,
      "num_tokens": 62139223.0,
      "step": 30660
    },
    {
      "entropy": 5.987939023971558,
      "epoch": 2.679804247137988,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00042822716401865314,
      "loss": 5.6273,
      "mean_token_accuracy": 0.1750589057803154,
      "num_tokens": 62149432.0,
      "step": 30665
    },
    {
      "entropy": 6.051985454559326,
      "epoch": 2.680241195490693,
      "grad_norm": 1.421875,
      "learning_rate": 0.00042820434407269347,
      "loss": 5.6969,
      "mean_token_accuracy": 0.16734425723552704,
      "num_tokens": 62160945.0,
      "step": 30670
    },
    {
      "entropy": 6.140903186798096,
      "epoch": 2.680678143843398,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004281815211881942,
      "loss": 5.6919,
      "mean_token_accuracy": 0.16290522068738938,
      "num_tokens": 62171514.0,
      "step": 30675
    },
    {
      "entropy": 6.0806769847869875,
      "epoch": 2.6811150921961024,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004281586953655931,
      "loss": 5.7135,
      "mean_token_accuracy": 0.16590942293405533,
      "num_tokens": 62181011.0,
      "step": 30680
    },
    {
      "entropy": 6.072387790679931,
      "epoch": 2.681552040548807,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004281358666053279,
      "loss": 5.6701,
      "mean_token_accuracy": 0.17861635386943817,
      "num_tokens": 62190915.0,
      "step": 30685
    },
    {
      "entropy": 6.037027978897095,
      "epoch": 2.681988988901512,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004281130349078366,
      "loss": 5.6855,
      "mean_token_accuracy": 0.16784376800060272,
      "num_tokens": 62200660.0,
      "step": 30690
    },
    {
      "entropy": 5.992252397537231,
      "epoch": 2.6824259372542167,
      "grad_norm": 1.1875,
      "learning_rate": 0.000428090200273557,
      "loss": 5.6866,
      "mean_token_accuracy": 0.17490296214818954,
      "num_tokens": 62210425.0,
      "step": 30695
    },
    {
      "entropy": 6.13086051940918,
      "epoch": 2.682862885606921,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004280673627029272,
      "loss": 5.6521,
      "mean_token_accuracy": 0.17362223118543624,
      "num_tokens": 62220921.0,
      "step": 30700
    },
    {
      "entropy": 6.077115440368653,
      "epoch": 2.683299833959626,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004280445221963852,
      "loss": 5.6679,
      "mean_token_accuracy": 0.17036591470241547,
      "num_tokens": 62230561.0,
      "step": 30705
    },
    {
      "entropy": 6.113948249816895,
      "epoch": 2.6837367823123306,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000428021678754369,
      "loss": 5.7275,
      "mean_token_accuracy": 0.17105989456176757,
      "num_tokens": 62240744.0,
      "step": 30710
    },
    {
      "entropy": 6.031236982345581,
      "epoch": 2.6841737306650355,
      "grad_norm": 1.703125,
      "learning_rate": 0.00042799883237731683,
      "loss": 5.622,
      "mean_token_accuracy": 0.174707293510437,
      "num_tokens": 62250902.0,
      "step": 30715
    },
    {
      "entropy": 6.127515172958374,
      "epoch": 2.68461067901774,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042797598306566694,
      "loss": 5.718,
      "mean_token_accuracy": 0.17089358419179917,
      "num_tokens": 62260459.0,
      "step": 30720
    },
    {
      "entropy": 6.123210477828979,
      "epoch": 2.685047627370445,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004279531308198575,
      "loss": 5.5974,
      "mean_token_accuracy": 0.17402150779962539,
      "num_tokens": 62269001.0,
      "step": 30725
    },
    {
      "entropy": 5.960367059707641,
      "epoch": 2.6854845757231494,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00042793027564032684,
      "loss": 5.6149,
      "mean_token_accuracy": 0.17320917546749115,
      "num_tokens": 62279197.0,
      "step": 30730
    },
    {
      "entropy": 6.032531356811523,
      "epoch": 2.6859215240758543,
      "grad_norm": 2.09375,
      "learning_rate": 0.00042790741752751337,
      "loss": 5.6288,
      "mean_token_accuracy": 0.1724080741405487,
      "num_tokens": 62289229.0,
      "step": 30735
    },
    {
      "entropy": 6.034072923660278,
      "epoch": 2.6863584724285587,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00042788455648185556,
      "loss": 5.6591,
      "mean_token_accuracy": 0.1797256961464882,
      "num_tokens": 62298819.0,
      "step": 30740
    },
    {
      "entropy": 6.076828336715698,
      "epoch": 2.6867954207812637,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042786169250379175,
      "loss": 5.6676,
      "mean_token_accuracy": 0.17109323889017106,
      "num_tokens": 62308567.0,
      "step": 30745
    },
    {
      "entropy": 6.110143661499023,
      "epoch": 2.6872323691339686,
      "grad_norm": 1.4375,
      "learning_rate": 0.00042783882559376067,
      "loss": 5.6959,
      "mean_token_accuracy": 0.17366999834775926,
      "num_tokens": 62318552.0,
      "step": 30750
    },
    {
      "entropy": 6.079264545440674,
      "epoch": 2.687669317486673,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042781595575220086,
      "loss": 5.6298,
      "mean_token_accuracy": 0.1786590039730072,
      "num_tokens": 62328430.0,
      "step": 30755
    },
    {
      "entropy": 6.058809757232666,
      "epoch": 2.688106265839378,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00042779308297955084,
      "loss": 5.75,
      "mean_token_accuracy": 0.16577178388834,
      "num_tokens": 62338540.0,
      "step": 30760
    },
    {
      "entropy": 6.002349758148194,
      "epoch": 2.6885432141920824,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00042777020727624945,
      "loss": 5.6892,
      "mean_token_accuracy": 0.16809403598308564,
      "num_tokens": 62348161.0,
      "step": 30765
    },
    {
      "entropy": 6.187067556381225,
      "epoch": 2.6889801625447873,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004277473286427354,
      "loss": 5.7614,
      "mean_token_accuracy": 0.16205809861421586,
      "num_tokens": 62358217.0,
      "step": 30770
    },
    {
      "entropy": 6.0865013122558596,
      "epoch": 2.689417110897492,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00042772444707944767,
      "loss": 5.6764,
      "mean_token_accuracy": 0.17103562206029893,
      "num_tokens": 62368801.0,
      "step": 30775
    },
    {
      "entropy": 6.055420446395874,
      "epoch": 2.6898540592501967,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004277015625868249,
      "loss": 5.6575,
      "mean_token_accuracy": 0.16860650926828386,
      "num_tokens": 62377632.0,
      "step": 30780
    },
    {
      "entropy": 6.0963921546936035,
      "epoch": 2.690291007602901,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004276786751653062,
      "loss": 5.7173,
      "mean_token_accuracy": 0.16862081438302995,
      "num_tokens": 62388056.0,
      "step": 30785
    },
    {
      "entropy": 6.0630988597869875,
      "epoch": 2.690727955955606,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004276557848153305,
      "loss": 5.6615,
      "mean_token_accuracy": 0.1676306962966919,
      "num_tokens": 62398015.0,
      "step": 30790
    },
    {
      "entropy": 6.058300638198853,
      "epoch": 2.6911649043083106,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00042763289153733685,
      "loss": 5.7342,
      "mean_token_accuracy": 0.1682770296931267,
      "num_tokens": 62408174.0,
      "step": 30795
    },
    {
      "entropy": 6.091994524002075,
      "epoch": 2.6916018526610155,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042760999533176435,
      "loss": 5.6728,
      "mean_token_accuracy": 0.1691396251320839,
      "num_tokens": 62417571.0,
      "step": 30800
    },
    {
      "entropy": 6.029355192184449,
      "epoch": 2.6920388010137204,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004275870961990523,
      "loss": 5.7107,
      "mean_token_accuracy": 0.17591763138771058,
      "num_tokens": 62427734.0,
      "step": 30805
    },
    {
      "entropy": 6.08669958114624,
      "epoch": 2.692475749366425,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00042756419413963965,
      "loss": 5.6786,
      "mean_token_accuracy": 0.1722591772675514,
      "num_tokens": 62437109.0,
      "step": 30810
    },
    {
      "entropy": 6.0084222793579105,
      "epoch": 2.6929126977191293,
      "grad_norm": 2.453125,
      "learning_rate": 0.00042754128915396583,
      "loss": 5.655,
      "mean_token_accuracy": 0.17137160003185273,
      "num_tokens": 62448163.0,
      "step": 30815
    },
    {
      "entropy": 6.0276368141174315,
      "epoch": 2.6933496460718342,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00042751838124247015,
      "loss": 5.7029,
      "mean_token_accuracy": 0.1674744814634323,
      "num_tokens": 62457993.0,
      "step": 30820
    },
    {
      "entropy": 6.117703866958618,
      "epoch": 2.693786594424539,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00042749547040559204,
      "loss": 5.7819,
      "mean_token_accuracy": 0.16431726962327958,
      "num_tokens": 62468554.0,
      "step": 30825
    },
    {
      "entropy": 6.142559146881103,
      "epoch": 2.6942235427772436,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004274725566437709,
      "loss": 5.6447,
      "mean_token_accuracy": 0.17458235919475557,
      "num_tokens": 62479204.0,
      "step": 30830
    },
    {
      "entropy": 6.106191635131836,
      "epoch": 2.6946604911299485,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004274496399574463,
      "loss": 5.6803,
      "mean_token_accuracy": 0.1682222381234169,
      "num_tokens": 62491391.0,
      "step": 30835
    },
    {
      "entropy": 5.968324518203735,
      "epoch": 2.695097439482653,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004274267203470576,
      "loss": 5.6151,
      "mean_token_accuracy": 0.17945559620857238,
      "num_tokens": 62500796.0,
      "step": 30840
    },
    {
      "entropy": 6.010866355895996,
      "epoch": 2.695534387835358,
      "grad_norm": 1.640625,
      "learning_rate": 0.00042740379781304455,
      "loss": 5.6539,
      "mean_token_accuracy": 0.17423538416624068,
      "num_tokens": 62509996.0,
      "step": 30845
    },
    {
      "entropy": 6.061119270324707,
      "epoch": 2.6959713361880624,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00042738087235584684,
      "loss": 5.6497,
      "mean_token_accuracy": 0.17329831570386886,
      "num_tokens": 62519676.0,
      "step": 30850
    },
    {
      "entropy": 6.083054637908935,
      "epoch": 2.6964082845407673,
      "grad_norm": 1.515625,
      "learning_rate": 0.00042735794397590407,
      "loss": 5.6924,
      "mean_token_accuracy": 0.16884765177965164,
      "num_tokens": 62530775.0,
      "step": 30855
    },
    {
      "entropy": 6.019345140457153,
      "epoch": 2.696845232893472,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004273350126736562,
      "loss": 5.6584,
      "mean_token_accuracy": 0.17210778146982192,
      "num_tokens": 62540643.0,
      "step": 30860
    },
    {
      "entropy": 6.078518056869507,
      "epoch": 2.6972821812461767,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004273120784495429,
      "loss": 5.6291,
      "mean_token_accuracy": 0.17797606587409973,
      "num_tokens": 62549766.0,
      "step": 30865
    },
    {
      "entropy": 6.101632022857666,
      "epoch": 2.697719129598881,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004272891413040042,
      "loss": 5.6386,
      "mean_token_accuracy": 0.16828909665346145,
      "num_tokens": 62560130.0,
      "step": 30870
    },
    {
      "entropy": 6.0968749046325685,
      "epoch": 2.698156077951586,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004272662012374799,
      "loss": 5.7273,
      "mean_token_accuracy": 0.17246001809835435,
      "num_tokens": 62571256.0,
      "step": 30875
    },
    {
      "entropy": 6.027985763549805,
      "epoch": 2.698593026304291,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004272432582504101,
      "loss": 5.607,
      "mean_token_accuracy": 0.18006455302238464,
      "num_tokens": 62582269.0,
      "step": 30880
    },
    {
      "entropy": 6.042442369461059,
      "epoch": 2.6990299746569955,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042722031234323485,
      "loss": 5.6682,
      "mean_token_accuracy": 0.17751594185829161,
      "num_tokens": 62591875.0,
      "step": 30885
    },
    {
      "entropy": 6.139891290664673,
      "epoch": 2.6994669230097004,
      "grad_norm": 1.953125,
      "learning_rate": 0.00042719736351639416,
      "loss": 5.7458,
      "mean_token_accuracy": 0.1637553572654724,
      "num_tokens": 62602467.0,
      "step": 30890
    },
    {
      "entropy": 6.035722541809082,
      "epoch": 2.699903871362405,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004271744117703284,
      "loss": 5.6192,
      "mean_token_accuracy": 0.17292855232954024,
      "num_tokens": 62611191.0,
      "step": 30895
    },
    {
      "entropy": 6.029976892471313,
      "epoch": 2.7003408197151098,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00042715145710547767,
      "loss": 5.6906,
      "mean_token_accuracy": 0.17756191194057463,
      "num_tokens": 62620961.0,
      "step": 30900
    },
    {
      "entropy": 6.052337121963501,
      "epoch": 2.700777768067814,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004271284995222822,
      "loss": 5.6832,
      "mean_token_accuracy": 0.17148202955722808,
      "num_tokens": 62631972.0,
      "step": 30905
    },
    {
      "entropy": 6.076364040374756,
      "epoch": 2.701214716420519,
      "grad_norm": 1.5,
      "learning_rate": 0.0004271055390211824,
      "loss": 5.6327,
      "mean_token_accuracy": 0.17794415801763536,
      "num_tokens": 62641073.0,
      "step": 30910
    },
    {
      "entropy": 6.062402868270874,
      "epoch": 2.701651664773224,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004270825756026187,
      "loss": 5.633,
      "mean_token_accuracy": 0.17402727752923966,
      "num_tokens": 62650364.0,
      "step": 30915
    },
    {
      "entropy": 6.068820381164551,
      "epoch": 2.7020886131259285,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004270596092670315,
      "loss": 5.7193,
      "mean_token_accuracy": 0.16872722208499907,
      "num_tokens": 62659846.0,
      "step": 30920
    },
    {
      "entropy": 6.070168972015381,
      "epoch": 2.702525561478633,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004270366400148613,
      "loss": 5.7023,
      "mean_token_accuracy": 0.17659879028797149,
      "num_tokens": 62670466.0,
      "step": 30925
    },
    {
      "entropy": 6.047671270370484,
      "epoch": 2.702962509831338,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004270136678465487,
      "loss": 5.6222,
      "mean_token_accuracy": 0.17338587939739228,
      "num_tokens": 62680330.0,
      "step": 30930
    },
    {
      "entropy": 6.14805383682251,
      "epoch": 2.703399458184043,
      "grad_norm": 1.453125,
      "learning_rate": 0.00042699069276253427,
      "loss": 5.7143,
      "mean_token_accuracy": 0.16473622173070906,
      "num_tokens": 62690177.0,
      "step": 30935
    },
    {
      "entropy": 6.078595209121704,
      "epoch": 2.7038364065367473,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004269677147632587,
      "loss": 5.6623,
      "mean_token_accuracy": 0.16886278092861176,
      "num_tokens": 62699365.0,
      "step": 30940
    },
    {
      "entropy": 5.977872562408447,
      "epoch": 2.704273354889452,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004269447338491628,
      "loss": 5.6452,
      "mean_token_accuracy": 0.18000553995370866,
      "num_tokens": 62709825.0,
      "step": 30945
    },
    {
      "entropy": 6.14600977897644,
      "epoch": 2.7047103032421567,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004269217500206873,
      "loss": 5.7159,
      "mean_token_accuracy": 0.1755475953221321,
      "num_tokens": 62719848.0,
      "step": 30950
    },
    {
      "entropy": 6.116764736175537,
      "epoch": 2.7051472515948616,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042689876327827293,
      "loss": 5.7733,
      "mean_token_accuracy": 0.1671248823404312,
      "num_tokens": 62730405.0,
      "step": 30955
    },
    {
      "entropy": 6.0538982391357425,
      "epoch": 2.705584199947566,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00042687577362236073,
      "loss": 5.6948,
      "mean_token_accuracy": 0.16863808035850525,
      "num_tokens": 62740275.0,
      "step": 30960
    },
    {
      "entropy": 6.078907775878906,
      "epoch": 2.706021148300271,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004268527810533917,
      "loss": 5.6853,
      "mean_token_accuracy": 0.17570008784532548,
      "num_tokens": 62750847.0,
      "step": 30965
    },
    {
      "entropy": 6.048583221435547,
      "epoch": 2.706458096652976,
      "grad_norm": 1.765625,
      "learning_rate": 0.00042682978557180665,
      "loss": 5.7255,
      "mean_token_accuracy": 0.1647748365998268,
      "num_tokens": 62761688.0,
      "step": 30970
    },
    {
      "entropy": 6.105798959732056,
      "epoch": 2.7068950450056803,
      "grad_norm": 1.46875,
      "learning_rate": 0.00042680678717804676,
      "loss": 5.7414,
      "mean_token_accuracy": 0.17029416263103486,
      "num_tokens": 62771168.0,
      "step": 30975
    },
    {
      "entropy": 6.0315755844116214,
      "epoch": 2.707331993358385,
      "grad_norm": 1.59375,
      "learning_rate": 0.00042678378587255315,
      "loss": 5.6146,
      "mean_token_accuracy": 0.18101221323013306,
      "num_tokens": 62780412.0,
      "step": 30980
    },
    {
      "entropy": 6.067301511764526,
      "epoch": 2.7077689417110897,
      "grad_norm": 1.671875,
      "learning_rate": 0.00042676078165576706,
      "loss": 5.7259,
      "mean_token_accuracy": 0.17335516810417176,
      "num_tokens": 62791673.0,
      "step": 30985
    },
    {
      "entropy": 6.128275060653687,
      "epoch": 2.7082058900637946,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004267377745281296,
      "loss": 5.6656,
      "mean_token_accuracy": 0.17026082873344422,
      "num_tokens": 62800980.0,
      "step": 30990
    },
    {
      "entropy": 6.027099227905273,
      "epoch": 2.708642838416499,
      "grad_norm": 2.078125,
      "learning_rate": 0.00042671476449008205,
      "loss": 5.687,
      "mean_token_accuracy": 0.1751496598124504,
      "num_tokens": 62813183.0,
      "step": 30995
    },
    {
      "entropy": 6.058013010025024,
      "epoch": 2.709079786769204,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000426691751542066,
      "loss": 5.6806,
      "mean_token_accuracy": 0.17014610320329665,
      "num_tokens": 62823170.0,
      "step": 31000
    },
    {
      "entropy": 6.125219058990479,
      "epoch": 2.7095167351219085,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004266687356845225,
      "loss": 5.683,
      "mean_token_accuracy": 0.17016275003552436,
      "num_tokens": 62833024.0,
      "step": 31005
    },
    {
      "entropy": 6.036967468261719,
      "epoch": 2.7099536834746134,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00042664571691789323,
      "loss": 5.6777,
      "mean_token_accuracy": 0.1725839525461197,
      "num_tokens": 62842533.0,
      "step": 31010
    },
    {
      "entropy": 6.1209142208099365,
      "epoch": 2.710390631827318,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004266226952426197,
      "loss": 5.7206,
      "mean_token_accuracy": 0.16993316560983657,
      "num_tokens": 62852984.0,
      "step": 31015
    },
    {
      "entropy": 6.10080852508545,
      "epoch": 2.710827580180023,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004265996706591434,
      "loss": 5.6935,
      "mean_token_accuracy": 0.1678627461194992,
      "num_tokens": 62861926.0,
      "step": 31020
    },
    {
      "entropy": 6.0998903751373295,
      "epoch": 2.7112645285327277,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042657664316790594,
      "loss": 5.6004,
      "mean_token_accuracy": 0.18235344290733338,
      "num_tokens": 62872194.0,
      "step": 31025
    },
    {
      "entropy": 6.016545152664184,
      "epoch": 2.711701476885432,
      "grad_norm": 3.765625,
      "learning_rate": 0.000426553612769349,
      "loss": 5.6371,
      "mean_token_accuracy": 0.17053132504224777,
      "num_tokens": 62882887.0,
      "step": 31030
    },
    {
      "entropy": 6.050682020187378,
      "epoch": 2.7121384252381366,
      "grad_norm": 1.5625,
      "learning_rate": 0.00042653057946391446,
      "loss": 5.6731,
      "mean_token_accuracy": 0.16985175907611846,
      "num_tokens": 62893563.0,
      "step": 31035
    },
    {
      "entropy": 6.0815245628356935,
      "epoch": 2.7125753735908416,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042650754325204395,
      "loss": 5.6798,
      "mean_token_accuracy": 0.17800917327404023,
      "num_tokens": 62902691.0,
      "step": 31040
    },
    {
      "entropy": 6.06647047996521,
      "epoch": 2.7130123219435465,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042648450413417937,
      "loss": 5.6361,
      "mean_token_accuracy": 0.1765693724155426,
      "num_tokens": 62912261.0,
      "step": 31045
    },
    {
      "entropy": 6.152713108062744,
      "epoch": 2.713449270296251,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00042646146211076263,
      "loss": 5.7558,
      "mean_token_accuracy": 0.16732678413391114,
      "num_tokens": 62922438.0,
      "step": 31050
    },
    {
      "entropy": 6.056459236145019,
      "epoch": 2.713886218648956,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004264384171822357,
      "loss": 5.6339,
      "mean_token_accuracy": 0.17674618363380432,
      "num_tokens": 62932378.0,
      "step": 31055
    },
    {
      "entropy": 6.131450557708741,
      "epoch": 2.7143231670016603,
      "grad_norm": 2.34375,
      "learning_rate": 0.00042641536934904064,
      "loss": 5.7966,
      "mean_token_accuracy": 0.16010637432336808,
      "num_tokens": 62943590.0,
      "step": 31060
    },
    {
      "entropy": 6.0785240650177,
      "epoch": 2.7147601153543652,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004263923186116193,
      "loss": 5.657,
      "mean_token_accuracy": 0.18036405891180038,
      "num_tokens": 62953393.0,
      "step": 31065
    },
    {
      "entropy": 6.077919912338257,
      "epoch": 2.7151970637070697,
      "grad_norm": 2.40625,
      "learning_rate": 0.00042636926497041406,
      "loss": 5.6891,
      "mean_token_accuracy": 0.17074144929647445,
      "num_tokens": 62964407.0,
      "step": 31070
    },
    {
      "entropy": 6.068600654602051,
      "epoch": 2.7156340120597746,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000426346208425867,
      "loss": 5.7216,
      "mean_token_accuracy": 0.168058380484581,
      "num_tokens": 62974968.0,
      "step": 31075
    },
    {
      "entropy": 6.083576822280884,
      "epoch": 2.716070960412479,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004263231489784203,
      "loss": 5.6215,
      "mean_token_accuracy": 0.1727873668074608,
      "num_tokens": 62984978.0,
      "step": 31080
    },
    {
      "entropy": 6.0675191402435305,
      "epoch": 2.716507908765184,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004263000866285164,
      "loss": 5.7347,
      "mean_token_accuracy": 0.16999080330133437,
      "num_tokens": 62995139.0,
      "step": 31085
    },
    {
      "entropy": 6.057342100143432,
      "epoch": 2.7169448571178885,
      "grad_norm": 1.890625,
      "learning_rate": 0.00042627702137659746,
      "loss": 5.7344,
      "mean_token_accuracy": 0.16430379301309586,
      "num_tokens": 63006127.0,
      "step": 31090
    },
    {
      "entropy": 6.121850919723511,
      "epoch": 2.7173818054705934,
      "grad_norm": 1.484375,
      "learning_rate": 0.00042625395322310605,
      "loss": 5.7358,
      "mean_token_accuracy": 0.17106882631778716,
      "num_tokens": 63015844.0,
      "step": 31095
    },
    {
      "entropy": 6.127876472473145,
      "epoch": 2.7178187538232983,
      "grad_norm": 1.875,
      "learning_rate": 0.00042623088216848446,
      "loss": 5.6922,
      "mean_token_accuracy": 0.1739245042204857,
      "num_tokens": 63025815.0,
      "step": 31100
    },
    {
      "entropy": 6.088843917846679,
      "epoch": 2.7182557021760028,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042620780821317543,
      "loss": 5.756,
      "mean_token_accuracy": 0.16788410842418672,
      "num_tokens": 63036108.0,
      "step": 31105
    },
    {
      "entropy": 6.113855218887329,
      "epoch": 2.7186926505287072,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004261847313576214,
      "loss": 5.6347,
      "mean_token_accuracy": 0.17565803825855256,
      "num_tokens": 63046305.0,
      "step": 31110
    },
    {
      "entropy": 6.152360630035401,
      "epoch": 2.719129598881412,
      "grad_norm": 2.46875,
      "learning_rate": 0.000426161651602265,
      "loss": 5.8185,
      "mean_token_accuracy": 0.16148675829172135,
      "num_tokens": 63057115.0,
      "step": 31115
    },
    {
      "entropy": 6.103050804138183,
      "epoch": 2.719566547234117,
      "grad_norm": 3.09375,
      "learning_rate": 0.00042613856894754886,
      "loss": 5.6665,
      "mean_token_accuracy": 0.17469296157360076,
      "num_tokens": 63067730.0,
      "step": 31120
    },
    {
      "entropy": 6.081635761260986,
      "epoch": 2.7200034955868215,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042611548339391574,
      "loss": 5.6612,
      "mean_token_accuracy": 0.17184237539768218,
      "num_tokens": 63077510.0,
      "step": 31125
    },
    {
      "entropy": 5.968316173553466,
      "epoch": 2.7204404439395264,
      "grad_norm": 1.59375,
      "learning_rate": 0.00042609239494180857,
      "loss": 5.5971,
      "mean_token_accuracy": 0.17657592594623567,
      "num_tokens": 63087160.0,
      "step": 31130
    },
    {
      "entropy": 6.03191909790039,
      "epoch": 2.720877392292231,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004260693035916701,
      "loss": 5.7254,
      "mean_token_accuracy": 0.17050208747386933,
      "num_tokens": 63097284.0,
      "step": 31135
    },
    {
      "entropy": 6.115389966964722,
      "epoch": 2.721314340644936,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042604620934394313,
      "loss": 5.6849,
      "mean_token_accuracy": 0.16964139118790628,
      "num_tokens": 63106910.0,
      "step": 31140
    },
    {
      "entropy": 6.084158706665039,
      "epoch": 2.7217512889976403,
      "grad_norm": 1.625,
      "learning_rate": 0.00042602311219907076,
      "loss": 5.5787,
      "mean_token_accuracy": 0.17542151659727095,
      "num_tokens": 63116129.0,
      "step": 31145
    },
    {
      "entropy": 5.985779333114624,
      "epoch": 2.722188237350345,
      "grad_norm": 2.921875,
      "learning_rate": 0.00042600001215749593,
      "loss": 5.6828,
      "mean_token_accuracy": 0.16735044568777085,
      "num_tokens": 63125602.0,
      "step": 31150
    },
    {
      "entropy": 6.060534286499023,
      "epoch": 2.72262518570305,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004259769092196618,
      "loss": 5.6807,
      "mean_token_accuracy": 0.1714681014418602,
      "num_tokens": 63136592.0,
      "step": 31155
    },
    {
      "entropy": 6.161821985244751,
      "epoch": 2.7230621340557546,
      "grad_norm": 2.96875,
      "learning_rate": 0.00042595380338601144,
      "loss": 5.6778,
      "mean_token_accuracy": 0.17541448175907134,
      "num_tokens": 63147040.0,
      "step": 31160
    },
    {
      "entropy": 5.999886131286621,
      "epoch": 2.723499082408459,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000425930694656988,
      "loss": 5.6713,
      "mean_token_accuracy": 0.16704075783491135,
      "num_tokens": 63156766.0,
      "step": 31165
    },
    {
      "entropy": 6.045689487457276,
      "epoch": 2.723936030761164,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004259075830330348,
      "loss": 5.6871,
      "mean_token_accuracy": 0.17193338721990586,
      "num_tokens": 63166253.0,
      "step": 31170
    },
    {
      "entropy": 6.123299837112427,
      "epoch": 2.724372979113869,
      "grad_norm": 1.609375,
      "learning_rate": 0.000425884468514595,
      "loss": 5.759,
      "mean_token_accuracy": 0.16451838314533235,
      "num_tokens": 63177352.0,
      "step": 31175
    },
    {
      "entropy": 6.127122688293457,
      "epoch": 2.7248099274665734,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004258613511021121,
      "loss": 5.7613,
      "mean_token_accuracy": 0.16282810270786285,
      "num_tokens": 63188092.0,
      "step": 31180
    },
    {
      "entropy": 6.075649785995483,
      "epoch": 2.7252468758192783,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004258382307960294,
      "loss": 5.6017,
      "mean_token_accuracy": 0.1764643132686615,
      "num_tokens": 63197677.0,
      "step": 31185
    },
    {
      "entropy": 6.124463653564453,
      "epoch": 2.7256838241719827,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004258151075967904,
      "loss": 5.7975,
      "mean_token_accuracy": 0.1624595433473587,
      "num_tokens": 63208536.0,
      "step": 31190
    },
    {
      "entropy": 6.050848579406738,
      "epoch": 2.7261207725246877,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004257919815048387,
      "loss": 5.6292,
      "mean_token_accuracy": 0.17259140014648439,
      "num_tokens": 63217966.0,
      "step": 31195
    },
    {
      "entropy": 6.05983419418335,
      "epoch": 2.726557720877392,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004257688525206176,
      "loss": 5.6489,
      "mean_token_accuracy": 0.169308540225029,
      "num_tokens": 63227420.0,
      "step": 31200
    },
    {
      "entropy": 6.112405157089233,
      "epoch": 2.726994669230097,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004257457206445711,
      "loss": 5.7432,
      "mean_token_accuracy": 0.16791827976703644,
      "num_tokens": 63237887.0,
      "step": 31205
    },
    {
      "entropy": 6.064127779006958,
      "epoch": 2.727431617582802,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00042572258587714255,
      "loss": 5.6413,
      "mean_token_accuracy": 0.1753789395093918,
      "num_tokens": 63247539.0,
      "step": 31210
    },
    {
      "entropy": 6.054610395431519,
      "epoch": 2.7278685659355064,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004256994482187759,
      "loss": 5.6143,
      "mean_token_accuracy": 0.17625898271799087,
      "num_tokens": 63257839.0,
      "step": 31215
    },
    {
      "entropy": 6.027532863616943,
      "epoch": 2.728305514288211,
      "grad_norm": 2.78125,
      "learning_rate": 0.00042567630766991477,
      "loss": 5.7448,
      "mean_token_accuracy": 0.16735746562480927,
      "num_tokens": 63268667.0,
      "step": 31220
    },
    {
      "entropy": 6.0635966777801515,
      "epoch": 2.728742462640916,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004256531642310032,
      "loss": 5.7255,
      "mean_token_accuracy": 0.1728343054652214,
      "num_tokens": 63277413.0,
      "step": 31225
    },
    {
      "entropy": 6.1195262432098385,
      "epoch": 2.7291794109936207,
      "grad_norm": 2.453125,
      "learning_rate": 0.000425630017902485,
      "loss": 5.7152,
      "mean_token_accuracy": 0.17487910091876985,
      "num_tokens": 63287225.0,
      "step": 31230
    },
    {
      "entropy": 6.081623315811157,
      "epoch": 2.729616359346325,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004256068686848041,
      "loss": 5.6833,
      "mean_token_accuracy": 0.1741498038172722,
      "num_tokens": 63297211.0,
      "step": 31235
    },
    {
      "entropy": 6.0312361240386965,
      "epoch": 2.73005330769903,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004255837165784045,
      "loss": 5.6554,
      "mean_token_accuracy": 0.17936736792325975,
      "num_tokens": 63307443.0,
      "step": 31240
    },
    {
      "entropy": 6.0378259181976315,
      "epoch": 2.7304902560517346,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004255605615837304,
      "loss": 5.7445,
      "mean_token_accuracy": 0.1688588798046112,
      "num_tokens": 63317951.0,
      "step": 31245
    },
    {
      "entropy": 6.086174440383911,
      "epoch": 2.7309272044044395,
      "grad_norm": 1.46875,
      "learning_rate": 0.00042553740370122574,
      "loss": 5.6491,
      "mean_token_accuracy": 0.17736051082611085,
      "num_tokens": 63328138.0,
      "step": 31250
    },
    {
      "entropy": 6.063856172561645,
      "epoch": 2.731364152757144,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004255142429313348,
      "loss": 5.5652,
      "mean_token_accuracy": 0.18576750606298448,
      "num_tokens": 63338229.0,
      "step": 31255
    },
    {
      "entropy": 6.076606035232544,
      "epoch": 2.731801101109849,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004254910792745019,
      "loss": 5.7227,
      "mean_token_accuracy": 0.17182790189981462,
      "num_tokens": 63348454.0,
      "step": 31260
    },
    {
      "entropy": 6.011942052841187,
      "epoch": 2.7322380494625538,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042546791273117125,
      "loss": 5.5489,
      "mean_token_accuracy": 0.1808629095554352,
      "num_tokens": 63357346.0,
      "step": 31265
    },
    {
      "entropy": 6.05736403465271,
      "epoch": 2.7326749978152582,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004254447433017871,
      "loss": 5.729,
      "mean_token_accuracy": 0.1708102345466614,
      "num_tokens": 63368329.0,
      "step": 31270
    },
    {
      "entropy": 6.1528548240661625,
      "epoch": 2.7331119461679627,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042542157098679397,
      "loss": 5.7525,
      "mean_token_accuracy": 0.16405082046985625,
      "num_tokens": 63378815.0,
      "step": 31275
    },
    {
      "entropy": 6.09550666809082,
      "epoch": 2.7335488945206676,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004253983957866364,
      "loss": 5.6563,
      "mean_token_accuracy": 0.17466020882129668,
      "num_tokens": 63390099.0,
      "step": 31280
    },
    {
      "entropy": 6.0530109882354735,
      "epoch": 2.7339858428733725,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042537521770175856,
      "loss": 5.7375,
      "mean_token_accuracy": 0.1633340299129486,
      "num_tokens": 63401244.0,
      "step": 31285
    },
    {
      "entropy": 6.061932706832886,
      "epoch": 2.734422791226077,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042535203673260544,
      "loss": 5.6604,
      "mean_token_accuracy": 0.1747384175658226,
      "num_tokens": 63410750.0,
      "step": 31290
    },
    {
      "entropy": 6.0293971538543705,
      "epoch": 2.734859739578782,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042532885287962146,
      "loss": 5.6405,
      "mean_token_accuracy": 0.17398273646831514,
      "num_tokens": 63420936.0,
      "step": 31295
    },
    {
      "entropy": 6.066473960876465,
      "epoch": 2.7352966879314864,
      "grad_norm": 1.75,
      "learning_rate": 0.00042530566614325125,
      "loss": 5.7047,
      "mean_token_accuracy": 0.1681433230638504,
      "num_tokens": 63431614.0,
      "step": 31300
    },
    {
      "entropy": 6.051823949813842,
      "epoch": 2.7357336362841913,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042528247652393964,
      "loss": 5.6284,
      "mean_token_accuracy": 0.17666591107845306,
      "num_tokens": 63441298.0,
      "step": 31305
    },
    {
      "entropy": 6.080903625488281,
      "epoch": 2.7361705846368958,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042525928402213127,
      "loss": 5.6663,
      "mean_token_accuracy": 0.1712244689464569,
      "num_tokens": 63451542.0,
      "step": 31310
    },
    {
      "entropy": 6.074837875366211,
      "epoch": 2.7366075329896007,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004252360886382712,
      "loss": 5.7254,
      "mean_token_accuracy": 0.16654887944459915,
      "num_tokens": 63460863.0,
      "step": 31315
    },
    {
      "entropy": 6.104344367980957,
      "epoch": 2.7370444813423056,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00042521289037280425,
      "loss": 5.705,
      "mean_token_accuracy": 0.16731063425540924,
      "num_tokens": 63470499.0,
      "step": 31320
    },
    {
      "entropy": 6.17957034111023,
      "epoch": 2.73748142969501,
      "grad_norm": 1.765625,
      "learning_rate": 0.00042518968922617534,
      "loss": 5.7338,
      "mean_token_accuracy": 0.16897505819797515,
      "num_tokens": 63480803.0,
      "step": 31325
    },
    {
      "entropy": 6.028335905075073,
      "epoch": 2.7379183780477145,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004251664851988295,
      "loss": 5.6734,
      "mean_token_accuracy": 0.17281928658485413,
      "num_tokens": 63491544.0,
      "step": 31330
    },
    {
      "entropy": 6.028273248672486,
      "epoch": 2.7383553264004195,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042514327829121175,
      "loss": 5.568,
      "mean_token_accuracy": 0.18003298193216324,
      "num_tokens": 63501086.0,
      "step": 31335
    },
    {
      "entropy": 6.111183834075928,
      "epoch": 2.7387922747531244,
      "grad_norm": 3.203125,
      "learning_rate": 0.00042512006850376724,
      "loss": 5.7082,
      "mean_token_accuracy": 0.17039864212274553,
      "num_tokens": 63512408.0,
      "step": 31340
    },
    {
      "entropy": 6.0761223316192625,
      "epoch": 2.739229223105829,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042509685583694116,
      "loss": 5.661,
      "mean_token_accuracy": 0.17121562361717224,
      "num_tokens": 63521735.0,
      "step": 31345
    },
    {
      "entropy": 6.095906162261963,
      "epoch": 2.7396661714585337,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004250736402911788,
      "loss": 5.6454,
      "mean_token_accuracy": 0.1782265394926071,
      "num_tokens": 63532045.0,
      "step": 31350
    },
    {
      "entropy": 6.053582286834716,
      "epoch": 2.740103119811238,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004250504218669253,
      "loss": 5.6292,
      "mean_token_accuracy": 0.1762608230113983,
      "num_tokens": 63541286.0,
      "step": 31355
    },
    {
      "entropy": 5.997816896438598,
      "epoch": 2.740540068163943,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004250272005646261,
      "loss": 5.6333,
      "mean_token_accuracy": 0.17646502405405046,
      "num_tokens": 63552195.0,
      "step": 31360
    },
    {
      "entropy": 6.052409696578979,
      "epoch": 2.7409770165166476,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004250039763847266,
      "loss": 5.621,
      "mean_token_accuracy": 0.17577073574066163,
      "num_tokens": 63561318.0,
      "step": 31365
    },
    {
      "entropy": 6.025645112991333,
      "epoch": 2.7414139648693525,
      "grad_norm": 1.734375,
      "learning_rate": 0.00042498074932767227,
      "loss": 5.6858,
      "mean_token_accuracy": 0.17272817194461823,
      "num_tokens": 63571637.0,
      "step": 31370
    },
    {
      "entropy": 6.136585998535156,
      "epoch": 2.7418509132220574,
      "grad_norm": 1.25,
      "learning_rate": 0.00042495751939390853,
      "loss": 5.7292,
      "mean_token_accuracy": 0.16477167159318923,
      "num_tokens": 63581618.0,
      "step": 31375
    },
    {
      "entropy": 5.990885925292969,
      "epoch": 2.742287861574762,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042493428658388094,
      "loss": 5.5602,
      "mean_token_accuracy": 0.18755961060523987,
      "num_tokens": 63592129.0,
      "step": 31380
    },
    {
      "entropy": 6.0818675518035885,
      "epoch": 2.7427248099274664,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004249110508980353,
      "loss": 5.7568,
      "mean_token_accuracy": 0.1647134006023407,
      "num_tokens": 63602861.0,
      "step": 31385
    },
    {
      "entropy": 6.099954414367676,
      "epoch": 2.7431617582801713,
      "grad_norm": 1.65625,
      "learning_rate": 0.00042488781233681705,
      "loss": 5.6425,
      "mean_token_accuracy": 0.1727519154548645,
      "num_tokens": 63612423.0,
      "step": 31390
    },
    {
      "entropy": 6.046270275115967,
      "epoch": 2.743598706632876,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00042486457090067206,
      "loss": 5.6874,
      "mean_token_accuracy": 0.17297804057598115,
      "num_tokens": 63622351.0,
      "step": 31395
    },
    {
      "entropy": 6.067964887619018,
      "epoch": 2.7440356549855807,
      "grad_norm": 2.0,
      "learning_rate": 0.00042484132659004604,
      "loss": 5.6246,
      "mean_token_accuracy": 0.18267279863357544,
      "num_tokens": 63632024.0,
      "step": 31400
    },
    {
      "entropy": 6.024618625640869,
      "epoch": 2.7444726033382856,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004248180794053849,
      "loss": 5.6307,
      "mean_token_accuracy": 0.17269423454999924,
      "num_tokens": 63641962.0,
      "step": 31405
    },
    {
      "entropy": 6.007831573486328,
      "epoch": 2.74490955169099,
      "grad_norm": 2.765625,
      "learning_rate": 0.00042479482934713446,
      "loss": 5.6813,
      "mean_token_accuracy": 0.1755543276667595,
      "num_tokens": 63652095.0,
      "step": 31410
    },
    {
      "entropy": 6.162137651443482,
      "epoch": 2.745346500043695,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004247715764157407,
      "loss": 5.7426,
      "mean_token_accuracy": 0.16974060386419296,
      "num_tokens": 63661392.0,
      "step": 31415
    },
    {
      "entropy": 6.077269697189331,
      "epoch": 2.7457834483963994,
      "grad_norm": 1.625,
      "learning_rate": 0.00042474832061164966,
      "loss": 5.6663,
      "mean_token_accuracy": 0.17006362676620485,
      "num_tokens": 63672383.0,
      "step": 31420
    },
    {
      "entropy": 6.066772985458374,
      "epoch": 2.7462203967491043,
      "grad_norm": 1.25,
      "learning_rate": 0.00042472506193530734,
      "loss": 5.6959,
      "mean_token_accuracy": 0.17081729471683502,
      "num_tokens": 63682522.0,
      "step": 31425
    },
    {
      "entropy": 6.009948015213013,
      "epoch": 2.746657345101809,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004247018003871599,
      "loss": 5.62,
      "mean_token_accuracy": 0.17807588279247283,
      "num_tokens": 63692443.0,
      "step": 31430
    },
    {
      "entropy": 6.008032035827637,
      "epoch": 2.7470942934545137,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042467853596765345,
      "loss": 5.6099,
      "mean_token_accuracy": 0.17877248227596282,
      "num_tokens": 63702268.0,
      "step": 31435
    },
    {
      "entropy": 5.953257894515991,
      "epoch": 2.747531241807218,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00042465526867723423,
      "loss": 5.577,
      "mean_token_accuracy": 0.18242289870977402,
      "num_tokens": 63712621.0,
      "step": 31440
    },
    {
      "entropy": 6.120203065872192,
      "epoch": 2.747968190159923,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004246319985163487,
      "loss": 5.7154,
      "mean_token_accuracy": 0.17177113592624665,
      "num_tokens": 63722319.0,
      "step": 31445
    },
    {
      "entropy": 6.1637256145477295,
      "epoch": 2.748405138512628,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004246087254854428,
      "loss": 5.7258,
      "mean_token_accuracy": 0.17496036142110824,
      "num_tokens": 63732726.0,
      "step": 31450
    },
    {
      "entropy": 6.043777275085449,
      "epoch": 2.7488420868653325,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042458544958496333,
      "loss": 5.6203,
      "mean_token_accuracy": 0.18018554896116257,
      "num_tokens": 63741197.0,
      "step": 31455
    },
    {
      "entropy": 6.042870569229126,
      "epoch": 2.749279035218037,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042456217081535645,
      "loss": 5.6312,
      "mean_token_accuracy": 0.17707999795675278,
      "num_tokens": 63751316.0,
      "step": 31460
    },
    {
      "entropy": 6.086418342590332,
      "epoch": 2.749715983570742,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042453888917706885,
      "loss": 5.7077,
      "mean_token_accuracy": 0.16800007224082947,
      "num_tokens": 63761246.0,
      "step": 31465
    },
    {
      "entropy": 6.096024417877198,
      "epoch": 2.750152931923447,
      "grad_norm": 1.75,
      "learning_rate": 0.00042451560467054696,
      "loss": 5.7195,
      "mean_token_accuracy": 0.16740930080413818,
      "num_tokens": 63771442.0,
      "step": 31470
    },
    {
      "entropy": 6.137561178207397,
      "epoch": 2.7505898802761513,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042449231729623735,
      "loss": 5.759,
      "mean_token_accuracy": 0.17123211175203323,
      "num_tokens": 63782840.0,
      "step": 31475
    },
    {
      "entropy": 6.032206058502197,
      "epoch": 2.751026828628856,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042446902705458683,
      "loss": 5.6434,
      "mean_token_accuracy": 0.16998863220214844,
      "num_tokens": 63793089.0,
      "step": 31480
    },
    {
      "entropy": 6.09352674484253,
      "epoch": 2.7514637769815606,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004244457339460421,
      "loss": 5.7141,
      "mean_token_accuracy": 0.1678785726428032,
      "num_tokens": 63804125.0,
      "step": 31485
    },
    {
      "entropy": 6.082953405380249,
      "epoch": 2.7519007253342656,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00042442243797104976,
      "loss": 5.7024,
      "mean_token_accuracy": 0.17372019737958908,
      "num_tokens": 63814713.0,
      "step": 31490
    },
    {
      "entropy": 6.103717422485351,
      "epoch": 2.75233767368697,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004243991391300569,
      "loss": 5.7373,
      "mean_token_accuracy": 0.1621850162744522,
      "num_tokens": 63824403.0,
      "step": 31495
    },
    {
      "entropy": 6.0827796459198,
      "epoch": 2.752774622039675,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004243758374235102,
      "loss": 5.7107,
      "mean_token_accuracy": 0.17158357799053192,
      "num_tokens": 63834182.0,
      "step": 31500
    },
    {
      "entropy": 6.076861095428467,
      "epoch": 2.75321157039238,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004243525328518567,
      "loss": 5.6862,
      "mean_token_accuracy": 0.17371679991483688,
      "num_tokens": 63844282.0,
      "step": 31505
    },
    {
      "entropy": 6.111580944061279,
      "epoch": 2.7536485187450843,
      "grad_norm": 1.5,
      "learning_rate": 0.0004243292254155433,
      "loss": 5.6372,
      "mean_token_accuracy": 0.17605184018611908,
      "num_tokens": 63854898.0,
      "step": 31510
    },
    {
      "entropy": 5.991305017471314,
      "epoch": 2.754085467097789,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00042430591511501717,
      "loss": 5.5968,
      "mean_token_accuracy": 0.1830749914050102,
      "num_tokens": 63864501.0,
      "step": 31515
    },
    {
      "entropy": 6.047938013076783,
      "epoch": 2.7545224154504937,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00042428260195072527,
      "loss": 5.6795,
      "mean_token_accuracy": 0.17609192728996276,
      "num_tokens": 63874812.0,
      "step": 31520
    },
    {
      "entropy": 6.084621858596802,
      "epoch": 2.7549593638031986,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004242592859231149,
      "loss": 5.6398,
      "mean_token_accuracy": 0.17340619266033172,
      "num_tokens": 63884028.0,
      "step": 31525
    },
    {
      "entropy": 6.011166048049927,
      "epoch": 2.755396312155903,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00042423596703263324,
      "loss": 5.6355,
      "mean_token_accuracy": 0.1814631402492523,
      "num_tokens": 63893991.0,
      "step": 31530
    },
    {
      "entropy": 6.074416828155518,
      "epoch": 2.755833260508608,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004242126452797275,
      "loss": 5.7434,
      "mean_token_accuracy": 0.16680092811584474,
      "num_tokens": 63904168.0,
      "step": 31535
    },
    {
      "entropy": 6.091437768936157,
      "epoch": 2.7562702088613125,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042418932066484505,
      "loss": 5.6941,
      "mean_token_accuracy": 0.16945927441120148,
      "num_tokens": 63913881.0,
      "step": 31540
    },
    {
      "entropy": 6.074461841583252,
      "epoch": 2.7567071572140174,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042416599318843324,
      "loss": 5.762,
      "mean_token_accuracy": 0.1631566822528839,
      "num_tokens": 63924521.0,
      "step": 31545
    },
    {
      "entropy": 6.134071493148804,
      "epoch": 2.757144105566722,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004241426628509395,
      "loss": 5.7135,
      "mean_token_accuracy": 0.1619485765695572,
      "num_tokens": 63935046.0,
      "step": 31550
    },
    {
      "entropy": 6.05696668624878,
      "epoch": 2.7575810539194268,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042411932965281143,
      "loss": 5.6734,
      "mean_token_accuracy": 0.17068079560995103,
      "num_tokens": 63944048.0,
      "step": 31555
    },
    {
      "entropy": 6.097636604309082,
      "epoch": 2.7580180022721317,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004240959935944964,
      "loss": 5.6455,
      "mean_token_accuracy": 0.1780053347349167,
      "num_tokens": 63954943.0,
      "step": 31560
    },
    {
      "entropy": 6.007201623916626,
      "epoch": 2.758454950624836,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004240726546764421,
      "loss": 5.6381,
      "mean_token_accuracy": 0.18044515550136567,
      "num_tokens": 63965929.0,
      "step": 31565
    },
    {
      "entropy": 6.057356309890747,
      "epoch": 2.7588918989775406,
      "grad_norm": 2.015625,
      "learning_rate": 0.00042404931289909614,
      "loss": 5.6729,
      "mean_token_accuracy": 0.16635266989469527,
      "num_tokens": 63976353.0,
      "step": 31570
    },
    {
      "entropy": 6.174076318740845,
      "epoch": 2.7593288473302455,
      "grad_norm": 2.296875,
      "learning_rate": 0.00042402596826290634,
      "loss": 5.7002,
      "mean_token_accuracy": 0.17261008471250533,
      "num_tokens": 63987073.0,
      "step": 31575
    },
    {
      "entropy": 6.043337488174439,
      "epoch": 2.7597657956829504,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004240026207683203,
      "loss": 5.6712,
      "mean_token_accuracy": 0.1764322727918625,
      "num_tokens": 63996631.0,
      "step": 31580
    },
    {
      "entropy": 6.053286027908325,
      "epoch": 2.760202744035655,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042397927041578596,
      "loss": 5.6921,
      "mean_token_accuracy": 0.1747522994875908,
      "num_tokens": 64007662.0,
      "step": 31585
    },
    {
      "entropy": 6.02344651222229,
      "epoch": 2.76063969238836,
      "grad_norm": 2.015625,
      "learning_rate": 0.00042395591720575113,
      "loss": 5.6185,
      "mean_token_accuracy": 0.17688922286033631,
      "num_tokens": 64017868.0,
      "step": 31590
    },
    {
      "entropy": 6.104690647125244,
      "epoch": 2.7610766407410643,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042393256113866377,
      "loss": 5.7062,
      "mean_token_accuracy": 0.16598058938980104,
      "num_tokens": 64027464.0,
      "step": 31595
    },
    {
      "entropy": 6.062092590332031,
      "epoch": 2.761513589093769,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004239092022149718,
      "loss": 5.7034,
      "mean_token_accuracy": 0.1692276582121849,
      "num_tokens": 64037795.0,
      "step": 31600
    },
    {
      "entropy": 6.098039197921753,
      "epoch": 2.7619505374464737,
      "grad_norm": 2.609375,
      "learning_rate": 0.00042388584043512333,
      "loss": 5.691,
      "mean_token_accuracy": 0.16782463788986207,
      "num_tokens": 64048055.0,
      "step": 31605
    },
    {
      "entropy": 6.035210609436035,
      "epoch": 2.7623874857991786,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004238624757995664,
      "loss": 5.597,
      "mean_token_accuracy": 0.18038821816444398,
      "num_tokens": 64057531.0,
      "step": 31610
    },
    {
      "entropy": 6.06682276725769,
      "epoch": 2.7628244341518835,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004238391083087493,
      "loss": 5.6323,
      "mean_token_accuracy": 0.17083710730075835,
      "num_tokens": 64066476.0,
      "step": 31615
    },
    {
      "entropy": 6.057695388793945,
      "epoch": 2.763261382504588,
      "grad_norm": 1.46875,
      "learning_rate": 0.00042381573796312,
      "loss": 5.6408,
      "mean_token_accuracy": 0.18172757625579833,
      "num_tokens": 64077069.0,
      "step": 31620
    },
    {
      "entropy": 6.036224889755249,
      "epoch": 2.7636983308572924,
      "grad_norm": 1.375,
      "learning_rate": 0.0004237923647631269,
      "loss": 5.6376,
      "mean_token_accuracy": 0.17785332202911378,
      "num_tokens": 64087325.0,
      "step": 31625
    },
    {
      "entropy": 6.0801270484924315,
      "epoch": 2.7641352792099974,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004237689887092183,
      "loss": 5.6953,
      "mean_token_accuracy": 0.1707962393760681,
      "num_tokens": 64097040.0,
      "step": 31630
    },
    {
      "entropy": 6.046498775482178,
      "epoch": 2.7645722275627023,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00042374560980184254,
      "loss": 5.6874,
      "mean_token_accuracy": 0.17390618324279786,
      "num_tokens": 64107798.0,
      "step": 31635
    },
    {
      "entropy": 6.158783674240112,
      "epoch": 2.7650091759154067,
      "grad_norm": 1.265625,
      "learning_rate": 0.000423722228041448,
      "loss": 5.7089,
      "mean_token_accuracy": 0.16724420636892318,
      "num_tokens": 64118395.0,
      "step": 31640
    },
    {
      "entropy": 6.10649356842041,
      "epoch": 2.7654461242681116,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042369884342848327,
      "loss": 5.7075,
      "mean_token_accuracy": 0.16791164577007295,
      "num_tokens": 64128040.0,
      "step": 31645
    },
    {
      "entropy": 6.021181535720825,
      "epoch": 2.765883072620816,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042367545596339674,
      "loss": 5.6614,
      "mean_token_accuracy": 0.17341413348913193,
      "num_tokens": 64137660.0,
      "step": 31650
    },
    {
      "entropy": 5.987086153030395,
      "epoch": 2.766320020973521,
      "grad_norm": 2.25,
      "learning_rate": 0.00042365206564663705,
      "loss": 5.6784,
      "mean_token_accuracy": 0.1685689389705658,
      "num_tokens": 64147632.0,
      "step": 31655
    },
    {
      "entropy": 6.031203985214233,
      "epoch": 2.7667569693262255,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004236286724786529,
      "loss": 5.6643,
      "mean_token_accuracy": 0.17475526183843612,
      "num_tokens": 64157604.0,
      "step": 31660
    },
    {
      "entropy": 6.1325194358825685,
      "epoch": 2.7671939176789304,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00042360527645989287,
      "loss": 5.7533,
      "mean_token_accuracy": 0.16392782777547837,
      "num_tokens": 64167626.0,
      "step": 31665
    },
    {
      "entropy": 6.139720487594604,
      "epoch": 2.7676308660316353,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004235818775908059,
      "loss": 5.7453,
      "mean_token_accuracy": 0.16847760528326033,
      "num_tokens": 64178436.0,
      "step": 31670
    },
    {
      "entropy": 5.963484239578247,
      "epoch": 2.76806781438434,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042355847587184056,
      "loss": 5.5952,
      "mean_token_accuracy": 0.1841944456100464,
      "num_tokens": 64188911.0,
      "step": 31675
    },
    {
      "entropy": 6.0489753723144535,
      "epoch": 2.7685047627370443,
      "grad_norm": 1.53125,
      "learning_rate": 0.00042353507130344584,
      "loss": 5.5899,
      "mean_token_accuracy": 0.1836236000061035,
      "num_tokens": 64198539.0,
      "step": 31680
    },
    {
      "entropy": 6.107727098464966,
      "epoch": 2.768941711089749,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004235116638860705,
      "loss": 5.7126,
      "mean_token_accuracy": 0.17174604088068007,
      "num_tokens": 64209121.0,
      "step": 31685
    },
    {
      "entropy": 6.060094547271729,
      "epoch": 2.769378659442454,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004234882536201638,
      "loss": 5.6565,
      "mean_token_accuracy": 0.17051102072000504,
      "num_tokens": 64218610.0,
      "step": 31690
    },
    {
      "entropy": 6.023504972457886,
      "epoch": 2.7698156077951586,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004234648405061745,
      "loss": 5.6793,
      "mean_token_accuracy": 0.17259233593940734,
      "num_tokens": 64228626.0,
      "step": 31695
    },
    {
      "entropy": 6.093149089813233,
      "epoch": 2.7702525561478635,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042344142454455174,
      "loss": 5.6289,
      "mean_token_accuracy": 0.17446598112583162,
      "num_tokens": 64238575.0,
      "step": 31700
    },
    {
      "entropy": 6.042654514312744,
      "epoch": 2.770689504500568,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004234180057357448,
      "loss": 5.6829,
      "mean_token_accuracy": 0.17268838733434677,
      "num_tokens": 64248551.0,
      "step": 31705
    },
    {
      "entropy": 6.099766540527344,
      "epoch": 2.771126452853273,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042339458408020267,
      "loss": 5.691,
      "mean_token_accuracy": 0.1705164149403572,
      "num_tokens": 64259315.0,
      "step": 31710
    },
    {
      "entropy": 6.08087944984436,
      "epoch": 2.7715634012059773,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004233711595783746,
      "loss": 5.691,
      "mean_token_accuracy": 0.1700661987066269,
      "num_tokens": 64269742.0,
      "step": 31715
    },
    {
      "entropy": 6.091532373428345,
      "epoch": 2.7720003495586822,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042334773223071,
      "loss": 5.6752,
      "mean_token_accuracy": 0.17546179592609407,
      "num_tokens": 64279266.0,
      "step": 31720
    },
    {
      "entropy": 6.040677738189697,
      "epoch": 2.7724372979113867,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042332430203765816,
      "loss": 5.629,
      "mean_token_accuracy": 0.17072993665933608,
      "num_tokens": 64289567.0,
      "step": 31725
    },
    {
      "entropy": 6.01774582862854,
      "epoch": 2.7728742462640916,
      "grad_norm": 1.25,
      "learning_rate": 0.0004233008689996685,
      "loss": 5.7268,
      "mean_token_accuracy": 0.17314556241035461,
      "num_tokens": 64299126.0,
      "step": 31730
    },
    {
      "entropy": 6.107919979095459,
      "epoch": 2.773311194616796,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004232774331171904,
      "loss": 5.6179,
      "mean_token_accuracy": 0.18445609211921693,
      "num_tokens": 64308964.0,
      "step": 31735
    },
    {
      "entropy": 6.1125486373901365,
      "epoch": 2.773748142969501,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042325399439067354,
      "loss": 5.6975,
      "mean_token_accuracy": 0.17171114236116408,
      "num_tokens": 64318463.0,
      "step": 31740
    },
    {
      "entropy": 6.151124572753906,
      "epoch": 2.774185091322206,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004232305528205673,
      "loss": 5.8445,
      "mean_token_accuracy": 0.16414942294359208,
      "num_tokens": 64328841.0,
      "step": 31745
    },
    {
      "entropy": 6.101870679855347,
      "epoch": 2.7746220396749104,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004232071084073214,
      "loss": 5.6835,
      "mean_token_accuracy": 0.17489487677812576,
      "num_tokens": 64339037.0,
      "step": 31750
    },
    {
      "entropy": 6.103510475158691,
      "epoch": 2.775058988027615,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00042318366115138553,
      "loss": 5.8319,
      "mean_token_accuracy": 0.16840914487838746,
      "num_tokens": 64350048.0,
      "step": 31755
    },
    {
      "entropy": 6.14383897781372,
      "epoch": 2.7754959363803198,
      "grad_norm": 2.140625,
      "learning_rate": 0.00042316021105320937,
      "loss": 5.7704,
      "mean_token_accuracy": 0.1635637938976288,
      "num_tokens": 64361066.0,
      "step": 31760
    },
    {
      "entropy": 6.121114253997803,
      "epoch": 2.7759328847330247,
      "grad_norm": 1.453125,
      "learning_rate": 0.00042313675811324276,
      "loss": 5.7259,
      "mean_token_accuracy": 0.17072082310914993,
      "num_tokens": 64371064.0,
      "step": 31765
    },
    {
      "entropy": 6.058264636993409,
      "epoch": 2.776369833085729,
      "grad_norm": 3.421875,
      "learning_rate": 0.00042311330233193545,
      "loss": 5.6217,
      "mean_token_accuracy": 0.18488989472389222,
      "num_tokens": 64380820.0,
      "step": 31770
    },
    {
      "entropy": 5.966573667526245,
      "epoch": 2.776806781438434,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004230898437097374,
      "loss": 5.6272,
      "mean_token_accuracy": 0.17457189708948134,
      "num_tokens": 64389777.0,
      "step": 31775
    },
    {
      "entropy": 6.109703826904297,
      "epoch": 2.7772437297911385,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004230663822470986,
      "loss": 5.777,
      "mean_token_accuracy": 0.16768710911273957,
      "num_tokens": 64399542.0,
      "step": 31780
    },
    {
      "entropy": 6.120810604095459,
      "epoch": 2.7776806781438435,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004230429179444689,
      "loss": 5.6829,
      "mean_token_accuracy": 0.16895513534545897,
      "num_tokens": 64410254.0,
      "step": 31785
    },
    {
      "entropy": 6.0578429222106935,
      "epoch": 2.778117626496548,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004230194508022986,
      "loss": 5.6583,
      "mean_token_accuracy": 0.1760959282517433,
      "num_tokens": 64420487.0,
      "step": 31790
    },
    {
      "entropy": 6.056322383880615,
      "epoch": 2.778554574849253,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042299598082103757,
      "loss": 5.6612,
      "mean_token_accuracy": 0.17435630410909653,
      "num_tokens": 64430171.0,
      "step": 31795
    },
    {
      "entropy": 6.088501834869385,
      "epoch": 2.7789915232019577,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042297250800113614,
      "loss": 5.7549,
      "mean_token_accuracy": 0.16745554357767106,
      "num_tokens": 64439960.0,
      "step": 31800
    },
    {
      "entropy": 6.1625494956970215,
      "epoch": 2.779428471554662,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004229490323430444,
      "loss": 5.7083,
      "mean_token_accuracy": 0.1704344555735588,
      "num_tokens": 64450238.0,
      "step": 31805
    },
    {
      "entropy": 6.088925075531006,
      "epoch": 2.7798654199073667,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00042292555384721277,
      "loss": 5.7047,
      "mean_token_accuracy": 0.16868208199739457,
      "num_tokens": 64460137.0,
      "step": 31810
    },
    {
      "entropy": 6.078416872024536,
      "epoch": 2.7803023682600716,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004229020725140915,
      "loss": 5.6524,
      "mean_token_accuracy": 0.17733215689659118,
      "num_tokens": 64470138.0,
      "step": 31815
    },
    {
      "entropy": 6.054874277114868,
      "epoch": 2.7807393166127765,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004228785883441309,
      "loss": 5.7364,
      "mean_token_accuracy": 0.1670811727643013,
      "num_tokens": 64480536.0,
      "step": 31820
    },
    {
      "entropy": 6.092967462539673,
      "epoch": 2.781176264965481,
      "grad_norm": 1.5,
      "learning_rate": 0.0004228551013377816,
      "loss": 5.6465,
      "mean_token_accuracy": 0.17734636962413788,
      "num_tokens": 64489899.0,
      "step": 31825
    },
    {
      "entropy": 6.0865977764129635,
      "epoch": 2.781613213318186,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042283161149549387,
      "loss": 5.6474,
      "mean_token_accuracy": 0.1796392872929573,
      "num_tokens": 64500272.0,
      "step": 31830
    },
    {
      "entropy": 6.1040290832519535,
      "epoch": 2.7820501616708904,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042280811881771846,
      "loss": 5.704,
      "mean_token_accuracy": 0.17201757431030273,
      "num_tokens": 64510507.0,
      "step": 31835
    },
    {
      "entropy": 5.968771028518677,
      "epoch": 2.7824871100235953,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042278462330490584,
      "loss": 5.616,
      "mean_token_accuracy": 0.1864620789885521,
      "num_tokens": 64521105.0,
      "step": 31840
    },
    {
      "entropy": 6.0239307403564455,
      "epoch": 2.7829240583762997,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042276112495750676,
      "loss": 5.7129,
      "mean_token_accuracy": 0.17058697640895842,
      "num_tokens": 64531828.0,
      "step": 31845
    },
    {
      "entropy": 6.103089141845703,
      "epoch": 2.7833610067290047,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004227376237759718,
      "loss": 5.5574,
      "mean_token_accuracy": 0.17617130577564238,
      "num_tokens": 64540783.0,
      "step": 31850
    },
    {
      "entropy": 6.024695301055909,
      "epoch": 2.7837979550817096,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004227141197607518,
      "loss": 5.6459,
      "mean_token_accuracy": 0.17666448205709456,
      "num_tokens": 64550958.0,
      "step": 31855
    },
    {
      "entropy": 6.010292005538941,
      "epoch": 2.784234903434414,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042269061291229766,
      "loss": 5.626,
      "mean_token_accuracy": 0.18296886831521988,
      "num_tokens": 64561666.0,
      "step": 31860
    },
    {
      "entropy": 6.06015248298645,
      "epoch": 2.7846718517871185,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042266710323106004,
      "loss": 5.689,
      "mean_token_accuracy": 0.16512232422828674,
      "num_tokens": 64570908.0,
      "step": 31865
    },
    {
      "entropy": 6.044925737380981,
      "epoch": 2.7851088001398234,
      "grad_norm": 1.796875,
      "learning_rate": 0.00042264359071749017,
      "loss": 5.5825,
      "mean_token_accuracy": 0.18252583593130112,
      "num_tokens": 64580981.0,
      "step": 31870
    },
    {
      "entropy": 6.092601108551025,
      "epoch": 2.7855457484925283,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042262007537203876,
      "loss": 5.6893,
      "mean_token_accuracy": 0.1676912158727646,
      "num_tokens": 64591776.0,
      "step": 31875
    },
    {
      "entropy": 6.044848918914795,
      "epoch": 2.785982696845233,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00042259655719515706,
      "loss": 5.6031,
      "mean_token_accuracy": 0.1737485036253929,
      "num_tokens": 64601201.0,
      "step": 31880
    },
    {
      "entropy": 6.0509411811828615,
      "epoch": 2.7864196451979377,
      "grad_norm": 1.375,
      "learning_rate": 0.0004225730361872959,
      "loss": 5.6953,
      "mean_token_accuracy": 0.17582377791404724,
      "num_tokens": 64610998.0,
      "step": 31885
    },
    {
      "entropy": 6.1085347652435305,
      "epoch": 2.786856593550642,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004225495123489067,
      "loss": 5.7609,
      "mean_token_accuracy": 0.1643397852778435,
      "num_tokens": 64620503.0,
      "step": 31890
    },
    {
      "entropy": 6.081197595596313,
      "epoch": 2.787293541903347,
      "grad_norm": 2.125,
      "learning_rate": 0.0004225259856804405,
      "loss": 5.6104,
      "mean_token_accuracy": 0.18137480914592743,
      "num_tokens": 64630172.0,
      "step": 31895
    },
    {
      "entropy": 6.0111918449401855,
      "epoch": 2.7877304902560516,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004225024561823486,
      "loss": 5.6555,
      "mean_token_accuracy": 0.17481929063796997,
      "num_tokens": 64639737.0,
      "step": 31900
    },
    {
      "entropy": 6.03213062286377,
      "epoch": 2.7881674386087565,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004224789238550823,
      "loss": 5.5971,
      "mean_token_accuracy": 0.17592462450265883,
      "num_tokens": 64650324.0,
      "step": 31905
    },
    {
      "entropy": 6.141432285308838,
      "epoch": 2.7886043869614614,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004224553886990929,
      "loss": 5.7714,
      "mean_token_accuracy": 0.16431823074817659,
      "num_tokens": 64661938.0,
      "step": 31910
    },
    {
      "entropy": 6.039334154129028,
      "epoch": 2.789041335314166,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00042243185071483193,
      "loss": 5.5919,
      "mean_token_accuracy": 0.1766808345913887,
      "num_tokens": 64672042.0,
      "step": 31915
    },
    {
      "entropy": 6.008403682708741,
      "epoch": 2.7894782836668703,
      "grad_norm": 1.9375,
      "learning_rate": 0.00042240830990275085,
      "loss": 5.6717,
      "mean_token_accuracy": 0.1730005547404289,
      "num_tokens": 64681798.0,
      "step": 31920
    },
    {
      "entropy": 6.087389230728149,
      "epoch": 2.7899152320195753,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000422384766263301,
      "loss": 5.6915,
      "mean_token_accuracy": 0.17039335668087005,
      "num_tokens": 64692329.0,
      "step": 31925
    },
    {
      "entropy": 6.052012157440186,
      "epoch": 2.79035218037228,
      "grad_norm": 1.703125,
      "learning_rate": 0.00042236121979693424,
      "loss": 5.6377,
      "mean_token_accuracy": 0.176263028383255,
      "num_tokens": 64701605.0,
      "step": 31930
    },
    {
      "entropy": 6.017575693130493,
      "epoch": 2.7907891287249846,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00042233767050410206,
      "loss": 5.7339,
      "mean_token_accuracy": 0.1699380725622177,
      "num_tokens": 64711649.0,
      "step": 31935
    },
    {
      "entropy": 6.11150541305542,
      "epoch": 2.7912260770776895,
      "grad_norm": 1.125,
      "learning_rate": 0.0004223141183852562,
      "loss": 5.6589,
      "mean_token_accuracy": 0.16957022845745087,
      "num_tokens": 64721451.0,
      "step": 31940
    },
    {
      "entropy": 6.03673529624939,
      "epoch": 2.791663025430394,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004222905634408482,
      "loss": 5.6471,
      "mean_token_accuracy": 0.17250234335660936,
      "num_tokens": 64731725.0,
      "step": 31945
    },
    {
      "entropy": 6.125047302246093,
      "epoch": 2.792099973783099,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00042226700567133015,
      "loss": 5.6607,
      "mean_token_accuracy": 0.16961756944656373,
      "num_tokens": 64741731.0,
      "step": 31950
    },
    {
      "entropy": 5.978402757644654,
      "epoch": 2.7925369221358034,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004222434450771537,
      "loss": 5.7099,
      "mean_token_accuracy": 0.17142774611711503,
      "num_tokens": 64753324.0,
      "step": 31955
    },
    {
      "entropy": 6.034575271606445,
      "epoch": 2.7929738704885083,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042221988165877083,
      "loss": 5.6174,
      "mean_token_accuracy": 0.18045715987682343,
      "num_tokens": 64763209.0,
      "step": 31960
    },
    {
      "entropy": 6.015704584121704,
      "epoch": 2.7934108188412132,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004221963154166335,
      "loss": 5.6628,
      "mean_token_accuracy": 0.17342773526906968,
      "num_tokens": 64773867.0,
      "step": 31965
    },
    {
      "entropy": 6.11203818321228,
      "epoch": 2.7938477671939177,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004221727463511937,
      "loss": 5.6505,
      "mean_token_accuracy": 0.16671427935361863,
      "num_tokens": 64785242.0,
      "step": 31970
    },
    {
      "entropy": 6.014839935302734,
      "epoch": 2.794284715546622,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004221491744629035,
      "loss": 5.639,
      "mean_token_accuracy": 0.1757766380906105,
      "num_tokens": 64795852.0,
      "step": 31975
    },
    {
      "entropy": 6.068985939025879,
      "epoch": 2.794721663899327,
      "grad_norm": 1.640625,
      "learning_rate": 0.00042212559975221504,
      "loss": 5.5775,
      "mean_token_accuracy": 0.18174809366464614,
      "num_tokens": 64805982.0,
      "step": 31980
    },
    {
      "entropy": 6.012110996246338,
      "epoch": 2.795158612252032,
      "grad_norm": 1.4375,
      "learning_rate": 0.00042210202221958055,
      "loss": 5.6948,
      "mean_token_accuracy": 0.17196368128061296,
      "num_tokens": 64816997.0,
      "step": 31985
    },
    {
      "entropy": 6.093861532211304,
      "epoch": 2.7955955606047365,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004220784418654521,
      "loss": 5.651,
      "mean_token_accuracy": 0.17162202596664428,
      "num_tokens": 64826597.0,
      "step": 31990
    },
    {
      "entropy": 6.011832141876221,
      "epoch": 2.7960325089574414,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042205485869028207,
      "loss": 5.5913,
      "mean_token_accuracy": 0.1757073074579239,
      "num_tokens": 64835917.0,
      "step": 31995
    },
    {
      "entropy": 6.098517513275146,
      "epoch": 2.796469457310146,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042203127269452284,
      "loss": 5.6353,
      "mean_token_accuracy": 0.17376898527145385,
      "num_tokens": 64846889.0,
      "step": 32000
    },
    {
      "entropy": 6.075806474685669,
      "epoch": 2.7969064056628508,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042200768387862676,
      "loss": 5.7159,
      "mean_token_accuracy": 0.1740233153104782,
      "num_tokens": 64857321.0,
      "step": 32005
    },
    {
      "entropy": 5.9598592758178714,
      "epoch": 2.7973433540155552,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042198409224304624,
      "loss": 5.679,
      "mean_token_accuracy": 0.17005496174097062,
      "num_tokens": 64867255.0,
      "step": 32010
    },
    {
      "entropy": 6.104970216751099,
      "epoch": 2.79778030236826,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004219604977882338,
      "loss": 5.6271,
      "mean_token_accuracy": 0.17481823563575744,
      "num_tokens": 64876179.0,
      "step": 32015
    },
    {
      "entropy": 6.1104878902435305,
      "epoch": 2.798217250720965,
      "grad_norm": 2.09375,
      "learning_rate": 0.00042193690051464206,
      "loss": 5.7489,
      "mean_token_accuracy": 0.16880867630243301,
      "num_tokens": 64886387.0,
      "step": 32020
    },
    {
      "entropy": 6.078713607788086,
      "epoch": 2.7986541990736695,
      "grad_norm": 1.671875,
      "learning_rate": 0.00042191330042272353,
      "loss": 5.7013,
      "mean_token_accuracy": 0.1729468286037445,
      "num_tokens": 64896816.0,
      "step": 32025
    },
    {
      "entropy": 6.011476707458496,
      "epoch": 2.799091147426374,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042188969751293093,
      "loss": 5.6355,
      "mean_token_accuracy": 0.1807342365384102,
      "num_tokens": 64907747.0,
      "step": 32030
    },
    {
      "entropy": 6.068430471420288,
      "epoch": 2.799528095779079,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042186609178571696,
      "loss": 5.6756,
      "mean_token_accuracy": 0.16877198368310928,
      "num_tokens": 64918066.0,
      "step": 32035
    },
    {
      "entropy": 6.121248865127564,
      "epoch": 2.799965044131784,
      "grad_norm": 1.5,
      "learning_rate": 0.0004218424832415344,
      "loss": 5.7312,
      "mean_token_accuracy": 0.16758372336626054,
      "num_tokens": 64927021.0,
      "step": 32040
    },
    {
      "entropy": 6.045506143569947,
      "epoch": 2.8004019924844883,
      "grad_norm": 2.4375,
      "learning_rate": 0.00042181887188083605,
      "loss": 5.5395,
      "mean_token_accuracy": 0.18428452163934708,
      "num_tokens": 64937477.0,
      "step": 32045
    },
    {
      "entropy": 5.995224380493164,
      "epoch": 2.800838940837193,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042179525770407486,
      "loss": 5.6363,
      "mean_token_accuracy": 0.17690736800432205,
      "num_tokens": 64947178.0,
      "step": 32050
    },
    {
      "entropy": 6.032656240463257,
      "epoch": 2.8012758891898977,
      "grad_norm": 1.65625,
      "learning_rate": 0.00042177164071170373,
      "loss": 5.7092,
      "mean_token_accuracy": 0.1653865933418274,
      "num_tokens": 64956921.0,
      "step": 32055
    },
    {
      "entropy": 6.103523397445679,
      "epoch": 2.8017128375426026,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004217480209041755,
      "loss": 5.6075,
      "mean_token_accuracy": 0.17417120635509492,
      "num_tokens": 64966688.0,
      "step": 32060
    },
    {
      "entropy": 6.0474120616912845,
      "epoch": 2.802149785895307,
      "grad_norm": 1.375,
      "learning_rate": 0.0004217243982819434,
      "loss": 5.61,
      "mean_token_accuracy": 0.17195419371128082,
      "num_tokens": 64975577.0,
      "step": 32065
    },
    {
      "entropy": 6.095047378540039,
      "epoch": 2.802586734248012,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004217007728454605,
      "loss": 5.6787,
      "mean_token_accuracy": 0.17171462625265121,
      "num_tokens": 64986121.0,
      "step": 32070
    },
    {
      "entropy": 6.093849754333496,
      "epoch": 2.8030236826007164,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004216771445951798,
      "loss": 5.6967,
      "mean_token_accuracy": 0.1685756504535675,
      "num_tokens": 64996623.0,
      "step": 32075
    },
    {
      "entropy": 6.026836109161377,
      "epoch": 2.8034606309534214,
      "grad_norm": 2.171875,
      "learning_rate": 0.00042165351353155476,
      "loss": 5.6169,
      "mean_token_accuracy": 0.170261150598526,
      "num_tokens": 65006756.0,
      "step": 32080
    },
    {
      "entropy": 6.027230072021484,
      "epoch": 2.803897579306126,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042162987965503834,
      "loss": 5.7014,
      "mean_token_accuracy": 0.17287039160728454,
      "num_tokens": 65017453.0,
      "step": 32085
    },
    {
      "entropy": 6.085272598266601,
      "epoch": 2.8043345276588307,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00042160624296608403,
      "loss": 5.7124,
      "mean_token_accuracy": 0.16392510533332824,
      "num_tokens": 65028090.0,
      "step": 32090
    },
    {
      "entropy": 6.074222898483276,
      "epoch": 2.8047714760115356,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004215826034651452,
      "loss": 5.7518,
      "mean_token_accuracy": 0.1684478387236595,
      "num_tokens": 65038422.0,
      "step": 32095
    },
    {
      "entropy": 6.072141838073731,
      "epoch": 2.80520842436424,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004215589611526752,
      "loss": 5.5747,
      "mean_token_accuracy": 0.18288754373788835,
      "num_tokens": 65048244.0,
      "step": 32100
    },
    {
      "entropy": 6.077329158782959,
      "epoch": 2.8056453727169446,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004215353160291275,
      "loss": 5.7011,
      "mean_token_accuracy": 0.16944419145584105,
      "num_tokens": 65058411.0,
      "step": 32105
    },
    {
      "entropy": 5.991480970382691,
      "epoch": 2.8060823210696495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004215116680949558,
      "loss": 5.6319,
      "mean_token_accuracy": 0.17420675307512284,
      "num_tokens": 65068148.0,
      "step": 32110
    },
    {
      "entropy": 6.060543489456177,
      "epoch": 2.8065192694223544,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004214880173506133,
      "loss": 5.6656,
      "mean_token_accuracy": 0.1728103369474411,
      "num_tokens": 65078212.0,
      "step": 32115
    },
    {
      "entropy": 6.076410484313965,
      "epoch": 2.806956217775059,
      "grad_norm": 1.578125,
      "learning_rate": 0.00042146436379655405,
      "loss": 5.6746,
      "mean_token_accuracy": 0.1713567391037941,
      "num_tokens": 65088133.0,
      "step": 32120
    },
    {
      "entropy": 6.019612884521484,
      "epoch": 2.807393166127764,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042144070743323155,
      "loss": 5.6064,
      "mean_token_accuracy": 0.16855982691049576,
      "num_tokens": 65097671.0,
      "step": 32125
    },
    {
      "entropy": 6.131534481048584,
      "epoch": 2.8078301144804683,
      "grad_norm": 1.25,
      "learning_rate": 0.0004214170482610995,
      "loss": 5.7592,
      "mean_token_accuracy": 0.16307077705860137,
      "num_tokens": 65107833.0,
      "step": 32130
    },
    {
      "entropy": 6.0671303272247314,
      "epoch": 2.808267062833173,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042139338628061177,
      "loss": 5.6389,
      "mean_token_accuracy": 0.17687022536993027,
      "num_tokens": 65118414.0,
      "step": 32135
    },
    {
      "entropy": 6.11484055519104,
      "epoch": 2.8087040111858776,
      "grad_norm": 2.078125,
      "learning_rate": 0.00042136972149222203,
      "loss": 5.7681,
      "mean_token_accuracy": 0.16336193382740022,
      "num_tokens": 65128920.0,
      "step": 32140
    },
    {
      "entropy": 6.019538736343383,
      "epoch": 2.8091409595385826,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004213460538963845,
      "loss": 5.6082,
      "mean_token_accuracy": 0.17908649295568466,
      "num_tokens": 65138869.0,
      "step": 32145
    },
    {
      "entropy": 6.056176471710205,
      "epoch": 2.8095779078912875,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004213223834935529,
      "loss": 5.6663,
      "mean_token_accuracy": 0.17373833656311036,
      "num_tokens": 65148737.0,
      "step": 32150
    },
    {
      "entropy": 6.080652856826783,
      "epoch": 2.810014856243992,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042129871028418137,
      "loss": 5.7068,
      "mean_token_accuracy": 0.16692305654287337,
      "num_tokens": 65158075.0,
      "step": 32155
    },
    {
      "entropy": 6.1393575191497805,
      "epoch": 2.8104518045966964,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004212750342687239,
      "loss": 5.7263,
      "mean_token_accuracy": 0.16629866510629654,
      "num_tokens": 65167788.0,
      "step": 32160
    },
    {
      "entropy": 6.037963342666626,
      "epoch": 2.8108887529494013,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004212513554476346,
      "loss": 5.6713,
      "mean_token_accuracy": 0.17235735654830933,
      "num_tokens": 65177581.0,
      "step": 32165
    },
    {
      "entropy": 6.099259710311889,
      "epoch": 2.8113257013021062,
      "grad_norm": 1.65625,
      "learning_rate": 0.00042122767382136766,
      "loss": 5.6558,
      "mean_token_accuracy": 0.17977633625268935,
      "num_tokens": 65186546.0,
      "step": 32170
    },
    {
      "entropy": 5.944136285781861,
      "epoch": 2.8117626496548107,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004212039893903774,
      "loss": 5.5303,
      "mean_token_accuracy": 0.18905836790800096,
      "num_tokens": 65195717.0,
      "step": 32175
    },
    {
      "entropy": 6.066419219970703,
      "epoch": 2.8121995980075156,
      "grad_norm": 1.609375,
      "learning_rate": 0.00042118030215511793,
      "loss": 5.724,
      "mean_token_accuracy": 0.1660353347659111,
      "num_tokens": 65206090.0,
      "step": 32180
    },
    {
      "entropy": 6.0723710536956785,
      "epoch": 2.81263654636022,
      "grad_norm": 1.375,
      "learning_rate": 0.00042115661211604373,
      "loss": 5.7087,
      "mean_token_accuracy": 0.17600794285535812,
      "num_tokens": 65217194.0,
      "step": 32185
    },
    {
      "entropy": 6.178366327285767,
      "epoch": 2.813073494712925,
      "grad_norm": 1.984375,
      "learning_rate": 0.000421132919273609,
      "loss": 5.7773,
      "mean_token_accuracy": 0.16716358065605164,
      "num_tokens": 65228906.0,
      "step": 32190
    },
    {
      "entropy": 6.084787273406983,
      "epoch": 2.8135104430656295,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004211092236282685,
      "loss": 5.721,
      "mean_token_accuracy": 0.1715310648083687,
      "num_tokens": 65238606.0,
      "step": 32195
    },
    {
      "entropy": 6.026787042617798,
      "epoch": 2.8139473914183344,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00042108552518047644,
      "loss": 5.6334,
      "mean_token_accuracy": 0.17519766837358475,
      "num_tokens": 65248259.0,
      "step": 32200
    },
    {
      "entropy": 6.118014764785767,
      "epoch": 2.8143843397710393,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004210618239306874,
      "loss": 5.7236,
      "mean_token_accuracy": 0.17079000771045685,
      "num_tokens": 65259294.0,
      "step": 32205
    },
    {
      "entropy": 6.125420093536377,
      "epoch": 2.8148212881237438,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004210381198793561,
      "loss": 5.7228,
      "mean_token_accuracy": 0.16535661071538926,
      "num_tokens": 65269498.0,
      "step": 32210
    },
    {
      "entropy": 6.1530091762542725,
      "epoch": 2.8152582364764482,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004210144130269371,
      "loss": 5.771,
      "mean_token_accuracy": 0.16032553762197493,
      "num_tokens": 65280027.0,
      "step": 32215
    },
    {
      "entropy": 6.073758840560913,
      "epoch": 2.815695184829153,
      "grad_norm": 2.4375,
      "learning_rate": 0.00042099070337388517,
      "loss": 5.6547,
      "mean_token_accuracy": 0.170989890396595,
      "num_tokens": 65290451.0,
      "step": 32220
    },
    {
      "entropy": 6.028297090530396,
      "epoch": 2.816132133181858,
      "grad_norm": 1.890625,
      "learning_rate": 0.00042096699092065504,
      "loss": 5.6207,
      "mean_token_accuracy": 0.17391049563884736,
      "num_tokens": 65299906.0,
      "step": 32225
    },
    {
      "entropy": 6.014467000961304,
      "epoch": 2.8165690815345625,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004209432756677015,
      "loss": 5.6383,
      "mean_token_accuracy": 0.17114360481500626,
      "num_tokens": 65310197.0,
      "step": 32230
    },
    {
      "entropy": 6.039159822463989,
      "epoch": 2.8170060298872674,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042091955761547963,
      "loss": 5.7229,
      "mean_token_accuracy": 0.17089447975158692,
      "num_tokens": 65320779.0,
      "step": 32235
    },
    {
      "entropy": 6.004044437408448,
      "epoch": 2.817442978239972,
      "grad_norm": 1.328125,
      "learning_rate": 0.000420895836764444,
      "loss": 5.556,
      "mean_token_accuracy": 0.18233199864625932,
      "num_tokens": 65330578.0,
      "step": 32240
    },
    {
      "entropy": 6.007945775985718,
      "epoch": 2.817879926592677,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042087211311504986,
      "loss": 5.678,
      "mean_token_accuracy": 0.1695188671350479,
      "num_tokens": 65341202.0,
      "step": 32245
    },
    {
      "entropy": 6.083621644973755,
      "epoch": 2.8183168749453813,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004208483866677521,
      "loss": 5.6553,
      "mean_token_accuracy": 0.17632312923669816,
      "num_tokens": 65352584.0,
      "step": 32250
    },
    {
      "entropy": 6.027787208557129,
      "epoch": 2.818753823298086,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004208246574230059,
      "loss": 5.6215,
      "mean_token_accuracy": 0.1736265242099762,
      "num_tokens": 65363191.0,
      "step": 32255
    },
    {
      "entropy": 5.996337604522705,
      "epoch": 2.819190771650791,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042080092538126633,
      "loss": 5.6309,
      "mean_token_accuracy": 0.17550832629203797,
      "num_tokens": 65373880.0,
      "step": 32260
    },
    {
      "entropy": 6.061549186706543,
      "epoch": 2.8196277200034956,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042077719054298867,
      "loss": 5.7104,
      "mean_token_accuracy": 0.17090259492397308,
      "num_tokens": 65383149.0,
      "step": 32265
    },
    {
      "entropy": 6.023580932617188,
      "epoch": 2.8200646683562,
      "grad_norm": 2.125,
      "learning_rate": 0.0004207534529086281,
      "loss": 5.6175,
      "mean_token_accuracy": 0.18424074947834015,
      "num_tokens": 65393005.0,
      "step": 32270
    },
    {
      "entropy": 6.020419502258301,
      "epoch": 2.820501616708905,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042072971247863984,
      "loss": 5.6833,
      "mean_token_accuracy": 0.1679721876978874,
      "num_tokens": 65402640.0,
      "step": 32275
    },
    {
      "entropy": 6.086766624450684,
      "epoch": 2.82093856506161,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004207059692534794,
      "loss": 5.7102,
      "mean_token_accuracy": 0.16519012451171874,
      "num_tokens": 65412863.0,
      "step": 32280
    },
    {
      "entropy": 6.060936117172242,
      "epoch": 2.8213755134143144,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042068222323360214,
      "loss": 5.7203,
      "mean_token_accuracy": 0.17432352602481843,
      "num_tokens": 65422581.0,
      "step": 32285
    },
    {
      "entropy": 6.086578416824341,
      "epoch": 2.8218124617670193,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042065847441946354,
      "loss": 5.7144,
      "mean_token_accuracy": 0.1721046730875969,
      "num_tokens": 65433604.0,
      "step": 32290
    },
    {
      "entropy": 6.01609468460083,
      "epoch": 2.8222494101197237,
      "grad_norm": 1.625,
      "learning_rate": 0.000420634722811519,
      "loss": 5.5903,
      "mean_token_accuracy": 0.17577413320541382,
      "num_tokens": 65442956.0,
      "step": 32295
    },
    {
      "entropy": 6.031525278091431,
      "epoch": 2.8226863584724287,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042061096841022416,
      "loss": 5.6774,
      "mean_token_accuracy": 0.17433386743068696,
      "num_tokens": 65452441.0,
      "step": 32300
    },
    {
      "entropy": 6.059484052658081,
      "epoch": 2.823123306825133,
      "grad_norm": 1.375,
      "learning_rate": 0.00042058721121603467,
      "loss": 5.7444,
      "mean_token_accuracy": 0.17174785882234572,
      "num_tokens": 65463628.0,
      "step": 32305
    },
    {
      "entropy": 6.030871248245239,
      "epoch": 2.823560255177838,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004205634512294062,
      "loss": 5.6303,
      "mean_token_accuracy": 0.17811108827590943,
      "num_tokens": 65472771.0,
      "step": 32310
    },
    {
      "entropy": 6.065471935272217,
      "epoch": 2.823997203530543,
      "grad_norm": 1.5,
      "learning_rate": 0.00042053968845079436,
      "loss": 5.6983,
      "mean_token_accuracy": 0.17017098367214203,
      "num_tokens": 65482678.0,
      "step": 32315
    },
    {
      "entropy": 6.1999359130859375,
      "epoch": 2.8244341518832474,
      "grad_norm": 1.640625,
      "learning_rate": 0.00042051592288065514,
      "loss": 5.7161,
      "mean_token_accuracy": 0.16791172474622726,
      "num_tokens": 65493127.0,
      "step": 32320
    },
    {
      "entropy": 6.083051300048828,
      "epoch": 2.824871100235952,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004204921545194443,
      "loss": 5.7124,
      "mean_token_accuracy": 0.16892116963863374,
      "num_tokens": 65502508.0,
      "step": 32325
    },
    {
      "entropy": 6.037751340866089,
      "epoch": 2.825308048588657,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004204683833676176,
      "loss": 5.7119,
      "mean_token_accuracy": 0.17071794867515563,
      "num_tokens": 65512336.0,
      "step": 32330
    },
    {
      "entropy": 6.078132963180542,
      "epoch": 2.8257449969413617,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004204446094256312,
      "loss": 5.6413,
      "mean_token_accuracy": 0.1679587408900261,
      "num_tokens": 65523139.0,
      "step": 32335
    },
    {
      "entropy": 6.0570920467376705,
      "epoch": 2.826181945294066,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042042083269394083,
      "loss": 5.6312,
      "mean_token_accuracy": 0.17600588947534562,
      "num_tokens": 65532846.0,
      "step": 32340
    },
    {
      "entropy": 6.002562761306763,
      "epoch": 2.826618893646771,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004203970531730028,
      "loss": 5.6429,
      "mean_token_accuracy": 0.16798023581504823,
      "num_tokens": 65544493.0,
      "step": 32345
    },
    {
      "entropy": 6.084763050079346,
      "epoch": 2.8270558419994756,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042037327086327304,
      "loss": 5.7241,
      "mean_token_accuracy": 0.17316241711378097,
      "num_tokens": 65554183.0,
      "step": 32350
    },
    {
      "entropy": 6.097694301605225,
      "epoch": 2.8274927903521805,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004203494857652078,
      "loss": 5.6469,
      "mean_token_accuracy": 0.17781269252300264,
      "num_tokens": 65564068.0,
      "step": 32355
    },
    {
      "entropy": 5.990041065216064,
      "epoch": 2.827929738704885,
      "grad_norm": 1.5625,
      "learning_rate": 0.00042032569787926326,
      "loss": 5.616,
      "mean_token_accuracy": 0.1829100802540779,
      "num_tokens": 65574272.0,
      "step": 32360
    },
    {
      "entropy": 6.065588903427124,
      "epoch": 2.82836668705759,
      "grad_norm": 1.640625,
      "learning_rate": 0.00042030190720589573,
      "loss": 5.6907,
      "mean_token_accuracy": 0.17033076286315918,
      "num_tokens": 65584890.0,
      "step": 32365
    },
    {
      "entropy": 6.121286678314209,
      "epoch": 2.8288036354102943,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004202781137455614,
      "loss": 5.6831,
      "mean_token_accuracy": 0.17354161590337752,
      "num_tokens": 65595776.0,
      "step": 32370
    },
    {
      "entropy": 6.001954793930054,
      "epoch": 2.8292405837629993,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004202543174987168,
      "loss": 5.6254,
      "mean_token_accuracy": 0.18214049339294433,
      "num_tokens": 65606234.0,
      "step": 32375
    },
    {
      "entropy": 6.053787899017334,
      "epoch": 2.8296775321157037,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042023051846581824,
      "loss": 5.7737,
      "mean_token_accuracy": 0.16956043243408203,
      "num_tokens": 65616576.0,
      "step": 32380
    },
    {
      "entropy": 6.045841789245605,
      "epoch": 2.8301144804684086,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004202067166473222,
      "loss": 5.6055,
      "mean_token_accuracy": 0.1841089427471161,
      "num_tokens": 65626792.0,
      "step": 32385
    },
    {
      "entropy": 6.075949430465698,
      "epoch": 2.8305514288211135,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004201829120436854,
      "loss": 5.6482,
      "mean_token_accuracy": 0.17668502479791642,
      "num_tokens": 65637041.0,
      "step": 32390
    },
    {
      "entropy": 6.013948440551758,
      "epoch": 2.830988377173818,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004201591046553641,
      "loss": 5.7021,
      "mean_token_accuracy": 0.16932215839624404,
      "num_tokens": 65647132.0,
      "step": 32395
    },
    {
      "entropy": 6.111507415771484,
      "epoch": 2.831425325526523,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042013529448281517,
      "loss": 5.7023,
      "mean_token_accuracy": 0.17012154906988144,
      "num_tokens": 65657372.0,
      "step": 32400
    },
    {
      "entropy": 6.059190845489502,
      "epoch": 2.8318622738792274,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004201114815264952,
      "loss": 5.6652,
      "mean_token_accuracy": 0.17678629010915756,
      "num_tokens": 65667118.0,
      "step": 32405
    },
    {
      "entropy": 6.080765867233277,
      "epoch": 2.8322992222319323,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000420087665786861,
      "loss": 5.7218,
      "mean_token_accuracy": 0.16729039400815965,
      "num_tokens": 65677830.0,
      "step": 32410
    },
    {
      "entropy": 6.0059648036956785,
      "epoch": 2.832736170584637,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042006384726436943,
      "loss": 5.6579,
      "mean_token_accuracy": 0.1745026245713234,
      "num_tokens": 65687238.0,
      "step": 32415
    },
    {
      "entropy": 6.0230687141418455,
      "epoch": 2.8331731189373417,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042004002595947713,
      "loss": 5.6193,
      "mean_token_accuracy": 0.1773090422153473,
      "num_tokens": 65697547.0,
      "step": 32420
    },
    {
      "entropy": 6.047022676467895,
      "epoch": 2.833610067290046,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004200162018726412,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1697848528623581,
      "num_tokens": 65709272.0,
      "step": 32425
    },
    {
      "entropy": 6.070788860321045,
      "epoch": 2.834047015642751,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004199923750043185,
      "loss": 5.698,
      "mean_token_accuracy": 0.1749678820371628,
      "num_tokens": 65720986.0,
      "step": 32430
    },
    {
      "entropy": 6.134988069534302,
      "epoch": 2.8344839639954555,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041996854535496616,
      "loss": 5.7072,
      "mean_token_accuracy": 0.17154998928308487,
      "num_tokens": 65731339.0,
      "step": 32435
    },
    {
      "entropy": 6.095059728622436,
      "epoch": 2.8349209123481605,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004199447129250411,
      "loss": 5.6856,
      "mean_token_accuracy": 0.17461057454347612,
      "num_tokens": 65741826.0,
      "step": 32440
    },
    {
      "entropy": 6.042696714401245,
      "epoch": 2.8353578607008654,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004199208777150004,
      "loss": 5.712,
      "mean_token_accuracy": 0.170701564848423,
      "num_tokens": 65752925.0,
      "step": 32445
    },
    {
      "entropy": 6.073705005645752,
      "epoch": 2.83579480905357,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004198970397253015,
      "loss": 5.6972,
      "mean_token_accuracy": 0.17088002860546112,
      "num_tokens": 65763948.0,
      "step": 32450
    },
    {
      "entropy": 6.077982664108276,
      "epoch": 2.8362317574062743,
      "grad_norm": 1.25,
      "learning_rate": 0.0004198731989564014,
      "loss": 5.6614,
      "mean_token_accuracy": 0.1737953782081604,
      "num_tokens": 65774086.0,
      "step": 32455
    },
    {
      "entropy": 5.999837064743042,
      "epoch": 2.8366687057589792,
      "grad_norm": 1.25,
      "learning_rate": 0.0004198493554087574,
      "loss": 5.5672,
      "mean_token_accuracy": 0.1858089029788971,
      "num_tokens": 65783630.0,
      "step": 32460
    },
    {
      "entropy": 6.022516250610352,
      "epoch": 2.837105654111684,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041982550908282686,
      "loss": 5.683,
      "mean_token_accuracy": 0.17201855182647705,
      "num_tokens": 65793903.0,
      "step": 32465
    },
    {
      "entropy": 6.09469952583313,
      "epoch": 2.8375426024643886,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00041980165997906714,
      "loss": 5.592,
      "mean_token_accuracy": 0.18385812193155288,
      "num_tokens": 65803483.0,
      "step": 32470
    },
    {
      "entropy": 6.032762241363526,
      "epoch": 2.8379795508170935,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004197778080979358,
      "loss": 5.6866,
      "mean_token_accuracy": 0.17417719066143036,
      "num_tokens": 65814336.0,
      "step": 32475
    },
    {
      "entropy": 6.045508480072021,
      "epoch": 2.838416499169798,
      "grad_norm": 1.484375,
      "learning_rate": 0.00041975395343989017,
      "loss": 5.7272,
      "mean_token_accuracy": 0.1672474816441536,
      "num_tokens": 65825525.0,
      "step": 32480
    },
    {
      "entropy": 6.07170000076294,
      "epoch": 2.838853447522503,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004197300960053879,
      "loss": 5.662,
      "mean_token_accuracy": 0.17949808090925218,
      "num_tokens": 65836276.0,
      "step": 32485
    },
    {
      "entropy": 6.076539659500122,
      "epoch": 2.8392903958752074,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004197062357948865,
      "loss": 5.6824,
      "mean_token_accuracy": 0.17157225161790848,
      "num_tokens": 65846164.0,
      "step": 32490
    },
    {
      "entropy": 6.057997226715088,
      "epoch": 2.8397273442279123,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041968237280884364,
      "loss": 5.6399,
      "mean_token_accuracy": 0.1721988409757614,
      "num_tokens": 65855759.0,
      "step": 32495
    },
    {
      "entropy": 5.985407066345215,
      "epoch": 2.840164292580617,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004196585070477172,
      "loss": 5.5181,
      "mean_token_accuracy": 0.18689307123422622,
      "num_tokens": 65866009.0,
      "step": 32500
    },
    {
      "entropy": 6.065166807174682,
      "epoch": 2.8406012409333217,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041963463851196466,
      "loss": 5.6292,
      "mean_token_accuracy": 0.1797526404261589,
      "num_tokens": 65875707.0,
      "step": 32505
    },
    {
      "entropy": 6.078594064712524,
      "epoch": 2.841038189286026,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00041961076720204403,
      "loss": 5.6472,
      "mean_token_accuracy": 0.1804350048303604,
      "num_tokens": 65884396.0,
      "step": 32510
    },
    {
      "entropy": 6.040422964096069,
      "epoch": 2.841475137638731,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004195868931184131,
      "loss": 5.67,
      "mean_token_accuracy": 0.1688934177160263,
      "num_tokens": 65894278.0,
      "step": 32515
    },
    {
      "entropy": 6.031313896179199,
      "epoch": 2.841912085991436,
      "grad_norm": 1.71875,
      "learning_rate": 0.00041956301626152974,
      "loss": 5.6748,
      "mean_token_accuracy": 0.17584653198719025,
      "num_tokens": 65903863.0,
      "step": 32520
    },
    {
      "entropy": 6.036254405975342,
      "epoch": 2.8423490343441404,
      "grad_norm": 2.015625,
      "learning_rate": 0.00041953913663185207,
      "loss": 5.5313,
      "mean_token_accuracy": 0.17780380249023436,
      "num_tokens": 65913996.0,
      "step": 32525
    },
    {
      "entropy": 5.992419004440308,
      "epoch": 2.8427859826968453,
      "grad_norm": 1.75,
      "learning_rate": 0.000419515254229838,
      "loss": 5.6029,
      "mean_token_accuracy": 0.18221209645271302,
      "num_tokens": 65923457.0,
      "step": 32530
    },
    {
      "entropy": 6.072156476974487,
      "epoch": 2.84322293104955,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00041949136905594555,
      "loss": 5.6651,
      "mean_token_accuracy": 0.16988272070884705,
      "num_tokens": 65933132.0,
      "step": 32535
    },
    {
      "entropy": 6.043057632446289,
      "epoch": 2.8436598794022547,
      "grad_norm": 1.171875,
      "learning_rate": 0.000419467481110633,
      "loss": 5.6306,
      "mean_token_accuracy": 0.17746004462242126,
      "num_tokens": 65942784.0,
      "step": 32540
    },
    {
      "entropy": 6.0294688701629635,
      "epoch": 2.844096827754959,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004194435903943585,
      "loss": 5.5951,
      "mean_token_accuracy": 0.17819080203771592,
      "num_tokens": 65952734.0,
      "step": 32545
    },
    {
      "entropy": 5.976498031616211,
      "epoch": 2.844533776107664,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041941969690758023,
      "loss": 5.5078,
      "mean_token_accuracy": 0.1927015796303749,
      "num_tokens": 65962732.0,
      "step": 32550
    },
    {
      "entropy": 6.057844352722168,
      "epoch": 2.844970724460369,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00041939580065075645,
      "loss": 5.7252,
      "mean_token_accuracy": 0.17301131039857864,
      "num_tokens": 65973287.0,
      "step": 32555
    },
    {
      "entropy": 6.092573404312134,
      "epoch": 2.8454076728130735,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004193719016243456,
      "loss": 5.6235,
      "mean_token_accuracy": 0.1731373593211174,
      "num_tokens": 65982034.0,
      "step": 32560
    },
    {
      "entropy": 6.032264137268067,
      "epoch": 2.845844621165778,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041934799982880594,
      "loss": 5.654,
      "mean_token_accuracy": 0.1795995905995369,
      "num_tokens": 65992612.0,
      "step": 32565
    },
    {
      "entropy": 6.0129289627075195,
      "epoch": 2.846281569518483,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00041932409526459607,
      "loss": 5.6426,
      "mean_token_accuracy": 0.17555613517761232,
      "num_tokens": 66002788.0,
      "step": 32570
    },
    {
      "entropy": 6.076098823547364,
      "epoch": 2.846718517871188,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041930018793217443,
      "loss": 5.6869,
      "mean_token_accuracy": 0.1726114794611931,
      "num_tokens": 66012559.0,
      "step": 32575
    },
    {
      "entropy": 6.104422092437744,
      "epoch": 2.8471554662238923,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004192762778319996,
      "loss": 5.7148,
      "mean_token_accuracy": 0.1641583651304245,
      "num_tokens": 66023230.0,
      "step": 32580
    },
    {
      "entropy": 6.0748124599456785,
      "epoch": 2.847592414576597,
      "grad_norm": 1.828125,
      "learning_rate": 0.00041925236496453006,
      "loss": 5.7485,
      "mean_token_accuracy": 0.17463057041168212,
      "num_tokens": 66033556.0,
      "step": 32585
    },
    {
      "entropy": 6.098298120498657,
      "epoch": 2.8480293629293016,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041922844933022467,
      "loss": 5.699,
      "mean_token_accuracy": 0.16835228800773622,
      "num_tokens": 66042495.0,
      "step": 32590
    },
    {
      "entropy": 6.016588735580444,
      "epoch": 2.8484663112820066,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041920453092954204,
      "loss": 5.6309,
      "mean_token_accuracy": 0.17761041820049286,
      "num_tokens": 66051933.0,
      "step": 32595
    },
    {
      "entropy": 6.012569236755371,
      "epoch": 2.848903259634711,
      "grad_norm": 1.21875,
      "learning_rate": 0.000419180609762941,
      "loss": 5.5881,
      "mean_token_accuracy": 0.17783350646495819,
      "num_tokens": 66060964.0,
      "step": 32600
    },
    {
      "entropy": 6.042345857620239,
      "epoch": 2.849340207987416,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041915668583088016,
      "loss": 5.6191,
      "mean_token_accuracy": 0.17663344889879226,
      "num_tokens": 66070210.0,
      "step": 32605
    },
    {
      "entropy": 6.0645722389221195,
      "epoch": 2.849777156340121,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00041913275913381864,
      "loss": 5.7298,
      "mean_token_accuracy": 0.17203486412763597,
      "num_tokens": 66081010.0,
      "step": 32610
    },
    {
      "entropy": 6.046843147277832,
      "epoch": 2.8502141046928253,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004191088296722153,
      "loss": 5.6244,
      "mean_token_accuracy": 0.17548077255487443,
      "num_tokens": 66090448.0,
      "step": 32615
    },
    {
      "entropy": 6.128679513931274,
      "epoch": 2.85065105304553,
      "grad_norm": 1.15625,
      "learning_rate": 0.000419084897446529,
      "loss": 5.7217,
      "mean_token_accuracy": 0.17269830256700516,
      "num_tokens": 66100357.0,
      "step": 32620
    },
    {
      "entropy": 6.030619049072266,
      "epoch": 2.8510880013982347,
      "grad_norm": 1.15625,
      "learning_rate": 0.000419060962457219,
      "loss": 5.6632,
      "mean_token_accuracy": 0.17374540716409684,
      "num_tokens": 66111068.0,
      "step": 32625
    },
    {
      "entropy": 6.051315593719482,
      "epoch": 2.8515249497509396,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041903702470474414,
      "loss": 5.685,
      "mean_token_accuracy": 0.17553779631853103,
      "num_tokens": 66122185.0,
      "step": 32630
    },
    {
      "entropy": 6.177775049209595,
      "epoch": 2.851961898103644,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004190130841895637,
      "loss": 5.6789,
      "mean_token_accuracy": 0.16932294070720671,
      "num_tokens": 66133192.0,
      "step": 32635
    },
    {
      "entropy": 6.032251024246216,
      "epoch": 2.852398846456349,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004189891409121369,
      "loss": 5.7172,
      "mean_token_accuracy": 0.16905670315027238,
      "num_tokens": 66143381.0,
      "step": 32640
    },
    {
      "entropy": 6.045573091506958,
      "epoch": 2.8528357948090535,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00041896519487292285,
      "loss": 5.6487,
      "mean_token_accuracy": 0.17699022591114044,
      "num_tokens": 66152883.0,
      "step": 32645
    },
    {
      "entropy": 6.107923221588135,
      "epoch": 2.8532727431617584,
      "grad_norm": 1.25,
      "learning_rate": 0.00041894124607238104,
      "loss": 5.5947,
      "mean_token_accuracy": 0.18175519853830338,
      "num_tokens": 66162725.0,
      "step": 32650
    },
    {
      "entropy": 6.087146139144897,
      "epoch": 2.853709691514463,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004189172945109706,
      "loss": 5.6898,
      "mean_token_accuracy": 0.17217792868614196,
      "num_tokens": 66171478.0,
      "step": 32655
    },
    {
      "entropy": 5.947203159332275,
      "epoch": 2.8541466398671678,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004188933401891511,
      "loss": 5.5591,
      "mean_token_accuracy": 0.18228826969861983,
      "num_tokens": 66181044.0,
      "step": 32660
    },
    {
      "entropy": 6.036250638961792,
      "epoch": 2.8545835882198727,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004188693831073818,
      "loss": 5.6834,
      "mean_token_accuracy": 0.17455545514822007,
      "num_tokens": 66190848.0,
      "step": 32665
    },
    {
      "entropy": 6.119058561325073,
      "epoch": 2.855020536572577,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00041884542326612245,
      "loss": 5.6982,
      "mean_token_accuracy": 0.16898895800113678,
      "num_tokens": 66200756.0,
      "step": 32670
    },
    {
      "entropy": 6.110272359848023,
      "epoch": 2.8554574849252816,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041882146066583257,
      "loss": 5.7224,
      "mean_token_accuracy": 0.16818744987249373,
      "num_tokens": 66211463.0,
      "step": 32675
    },
    {
      "entropy": 6.115203428268432,
      "epoch": 2.8558944332779865,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041879749530697165,
      "loss": 5.6941,
      "mean_token_accuracy": 0.16697739511728288,
      "num_tokens": 66222522.0,
      "step": 32680
    },
    {
      "entropy": 6.053583145141602,
      "epoch": 2.8563313816306914,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004187735271899994,
      "loss": 5.7059,
      "mean_token_accuracy": 0.174971304833889,
      "num_tokens": 66232741.0,
      "step": 32685
    },
    {
      "entropy": 6.08095531463623,
      "epoch": 2.856768329983396,
      "grad_norm": 2.40625,
      "learning_rate": 0.00041874955631537555,
      "loss": 5.5995,
      "mean_token_accuracy": 0.17359778434038162,
      "num_tokens": 66242845.0,
      "step": 32690
    },
    {
      "entropy": 6.0156513214111325,
      "epoch": 2.857205278336101,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041872558268355993,
      "loss": 5.6138,
      "mean_token_accuracy": 0.17654035091400147,
      "num_tokens": 66252697.0,
      "step": 32695
    },
    {
      "entropy": 6.017025184631348,
      "epoch": 2.8576422266888053,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00041870160629501224,
      "loss": 5.6882,
      "mean_token_accuracy": 0.17247844338417054,
      "num_tokens": 66263607.0,
      "step": 32700
    },
    {
      "entropy": 6.0680108070373535,
      "epoch": 2.85807917504151,
      "grad_norm": 1.75,
      "learning_rate": 0.00041867762715019257,
      "loss": 5.6029,
      "mean_token_accuracy": 0.17496681064367295,
      "num_tokens": 66274352.0,
      "step": 32705
    },
    {
      "entropy": 6.014532089233398,
      "epoch": 2.8585161233942147,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00041865364524956066,
      "loss": 5.6551,
      "mean_token_accuracy": 0.17715186178684234,
      "num_tokens": 66283972.0,
      "step": 32710
    },
    {
      "entropy": 6.052670621871949,
      "epoch": 2.8589530717469196,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041862966059357665,
      "loss": 5.7167,
      "mean_token_accuracy": 0.17310549616813659,
      "num_tokens": 66295179.0,
      "step": 32715
    },
    {
      "entropy": 6.163036775588989,
      "epoch": 2.859390020099624,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004186056731827003,
      "loss": 5.7185,
      "mean_token_accuracy": 0.16762714982032775,
      "num_tokens": 66306065.0,
      "step": 32720
    },
    {
      "entropy": 6.000307941436768,
      "epoch": 2.859826968452329,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000418581683017392,
      "loss": 5.6236,
      "mean_token_accuracy": 0.17163815796375276,
      "num_tokens": 66315491.0,
      "step": 32725
    },
    {
      "entropy": 6.0191161155700685,
      "epoch": 2.8602639168050334,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004185576900981119,
      "loss": 5.5915,
      "mean_token_accuracy": 0.17778780311346054,
      "num_tokens": 66324327.0,
      "step": 32730
    },
    {
      "entropy": 6.110948991775513,
      "epoch": 2.8607008651577384,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004185336944253199,
      "loss": 5.6189,
      "mean_token_accuracy": 0.17377718538045883,
      "num_tokens": 66334201.0,
      "step": 32735
    },
    {
      "entropy": 6.050872659683227,
      "epoch": 2.8611378135104433,
      "grad_norm": 1.890625,
      "learning_rate": 0.00041850969599947655,
      "loss": 5.6388,
      "mean_token_accuracy": 0.1804061934351921,
      "num_tokens": 66344745.0,
      "step": 32740
    },
    {
      "entropy": 5.935228633880615,
      "epoch": 2.8615747618631477,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041848569482104197,
      "loss": 5.6328,
      "mean_token_accuracy": 0.18335383385419846,
      "num_tokens": 66354815.0,
      "step": 32745
    },
    {
      "entropy": 6.05367259979248,
      "epoch": 2.862011710215852,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004184616908904765,
      "loss": 5.6405,
      "mean_token_accuracy": 0.1759585842490196,
      "num_tokens": 66364744.0,
      "step": 32750
    },
    {
      "entropy": 6.097091150283814,
      "epoch": 2.862448658568557,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004184376842082408,
      "loss": 5.6968,
      "mean_token_accuracy": 0.17521189749240876,
      "num_tokens": 66376363.0,
      "step": 32755
    },
    {
      "entropy": 6.048199939727783,
      "epoch": 2.862885606921262,
      "grad_norm": 1.25,
      "learning_rate": 0.00041841367477479503,
      "loss": 5.6877,
      "mean_token_accuracy": 0.16743492633104323,
      "num_tokens": 66386149.0,
      "step": 32760
    },
    {
      "entropy": 6.0805785179138185,
      "epoch": 2.8633225552739665,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004183896625905999,
      "loss": 5.6021,
      "mean_token_accuracy": 0.17703621089458466,
      "num_tokens": 66396874.0,
      "step": 32765
    },
    {
      "entropy": 5.979263067245483,
      "epoch": 2.8637595036266714,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041836564765611583,
      "loss": 5.6084,
      "mean_token_accuracy": 0.17395420521497726,
      "num_tokens": 66406761.0,
      "step": 32770
    },
    {
      "entropy": 6.140816783905029,
      "epoch": 2.864196451979376,
      "grad_norm": 1.765625,
      "learning_rate": 0.00041834162997180364,
      "loss": 5.7429,
      "mean_token_accuracy": 0.17394705712795258,
      "num_tokens": 66416947.0,
      "step": 32775
    },
    {
      "entropy": 6.0348066806793215,
      "epoch": 2.864633400332081,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004183176095381238,
      "loss": 5.625,
      "mean_token_accuracy": 0.1794386625289917,
      "num_tokens": 66427980.0,
      "step": 32780
    },
    {
      "entropy": 5.9515852451324465,
      "epoch": 2.8650703486847853,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041829358635553707,
      "loss": 5.5522,
      "mean_token_accuracy": 0.17885153144598007,
      "num_tokens": 66437226.0,
      "step": 32785
    },
    {
      "entropy": 6.050860643386841,
      "epoch": 2.86550729703749,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00041826956042450436,
      "loss": 5.7095,
      "mean_token_accuracy": 0.17005944699048997,
      "num_tokens": 66447820.0,
      "step": 32790
    },
    {
      "entropy": 6.045299625396728,
      "epoch": 2.865944245390195,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004182455317454863,
      "loss": 5.6635,
      "mean_token_accuracy": 0.16978426575660704,
      "num_tokens": 66459727.0,
      "step": 32795
    },
    {
      "entropy": 6.029955863952637,
      "epoch": 2.8663811937428996,
      "grad_norm": 1.546875,
      "learning_rate": 0.00041822150031894397,
      "loss": 5.6094,
      "mean_token_accuracy": 0.1740194395184517,
      "num_tokens": 66469743.0,
      "step": 32800
    },
    {
      "entropy": 6.041649961471558,
      "epoch": 2.866818142095604,
      "grad_norm": 1.5,
      "learning_rate": 0.00041819746614533824,
      "loss": 5.7219,
      "mean_token_accuracy": 0.17342590242624284,
      "num_tokens": 66479974.0,
      "step": 32805
    },
    {
      "entropy": 6.049103832244873,
      "epoch": 2.867255090448309,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041817342922513,
      "loss": 5.5851,
      "mean_token_accuracy": 0.17909797430038452,
      "num_tokens": 66489605.0,
      "step": 32810
    },
    {
      "entropy": 6.038738965988159,
      "epoch": 2.867692038801014,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041814938955878035,
      "loss": 5.6723,
      "mean_token_accuracy": 0.17900241017341614,
      "num_tokens": 66499454.0,
      "step": 32815
    },
    {
      "entropy": 6.059465169906616,
      "epoch": 2.8681289871537183,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041812534714675046,
      "loss": 5.7051,
      "mean_token_accuracy": 0.1750141277909279,
      "num_tokens": 66510076.0,
      "step": 32820
    },
    {
      "entropy": 6.132712507247925,
      "epoch": 2.8685659355064232,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041810130198950137,
      "loss": 5.6639,
      "mean_token_accuracy": 0.17342266589403152,
      "num_tokens": 66520659.0,
      "step": 32825
    },
    {
      "entropy": 6.086273670196533,
      "epoch": 2.8690028838591277,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004180772540874943,
      "loss": 5.7311,
      "mean_token_accuracy": 0.17099483013153077,
      "num_tokens": 66531387.0,
      "step": 32830
    },
    {
      "entropy": 6.053447818756103,
      "epoch": 2.8694398322118326,
      "grad_norm": 1.515625,
      "learning_rate": 0.00041805320344119064,
      "loss": 5.6381,
      "mean_token_accuracy": 0.18089621365070344,
      "num_tokens": 66540654.0,
      "step": 32835
    },
    {
      "entropy": 6.043720769882202,
      "epoch": 2.869876780564537,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004180291500510515,
      "loss": 5.6586,
      "mean_token_accuracy": 0.17636438012123107,
      "num_tokens": 66550966.0,
      "step": 32840
    },
    {
      "entropy": 6.124829912185669,
      "epoch": 2.870313728917242,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041800509391753834,
      "loss": 5.7543,
      "mean_token_accuracy": 0.16585330963134765,
      "num_tokens": 66561550.0,
      "step": 32845
    },
    {
      "entropy": 6.042979383468628,
      "epoch": 2.870750677269947,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004179810350411125,
      "loss": 5.6143,
      "mean_token_accuracy": 0.17880491465330123,
      "num_tokens": 66571477.0,
      "step": 32850
    },
    {
      "entropy": 6.034370279312133,
      "epoch": 2.8711876256226514,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004179569734222355,
      "loss": 5.6849,
      "mean_token_accuracy": 0.17219637036323548,
      "num_tokens": 66581976.0,
      "step": 32855
    },
    {
      "entropy": 6.091075611114502,
      "epoch": 2.871624573975356,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041793290906136886,
      "loss": 5.638,
      "mean_token_accuracy": 0.17021238207817077,
      "num_tokens": 66591747.0,
      "step": 32860
    },
    {
      "entropy": 6.058017873764038,
      "epoch": 2.872061522328061,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004179088419589741,
      "loss": 5.6588,
      "mean_token_accuracy": 0.17066488564014434,
      "num_tokens": 66602681.0,
      "step": 32865
    },
    {
      "entropy": 6.064173460006714,
      "epoch": 2.8724984706807657,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041788477211551295,
      "loss": 5.6809,
      "mean_token_accuracy": 0.17534556835889817,
      "num_tokens": 66612887.0,
      "step": 32870
    },
    {
      "entropy": 6.04403133392334,
      "epoch": 2.87293541903347,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041786069953144697,
      "loss": 5.615,
      "mean_token_accuracy": 0.17833898663520814,
      "num_tokens": 66622418.0,
      "step": 32875
    },
    {
      "entropy": 6.089007329940796,
      "epoch": 2.873372367386175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041783662420723786,
      "loss": 5.7167,
      "mean_token_accuracy": 0.1701267257332802,
      "num_tokens": 66633457.0,
      "step": 32880
    },
    {
      "entropy": 6.0887370109558105,
      "epoch": 2.8738093157388795,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004178125461433475,
      "loss": 5.603,
      "mean_token_accuracy": 0.17580579966306686,
      "num_tokens": 66643691.0,
      "step": 32885
    },
    {
      "entropy": 6.003830289840698,
      "epoch": 2.8742462640915845,
      "grad_norm": 1.6875,
      "learning_rate": 0.00041778846534023763,
      "loss": 5.6587,
      "mean_token_accuracy": 0.17320727854967116,
      "num_tokens": 66653745.0,
      "step": 32890
    },
    {
      "entropy": 6.0071581363677975,
      "epoch": 2.874683212444289,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041776438179837026,
      "loss": 5.6355,
      "mean_token_accuracy": 0.1763730227947235,
      "num_tokens": 66664093.0,
      "step": 32895
    },
    {
      "entropy": 6.030020141601563,
      "epoch": 2.875120160796994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004177402955182071,
      "loss": 5.7028,
      "mean_token_accuracy": 0.16880220919847488,
      "num_tokens": 66675204.0,
      "step": 32900
    },
    {
      "entropy": 6.149861288070679,
      "epoch": 2.8755571091496988,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00041771620650021043,
      "loss": 5.6765,
      "mean_token_accuracy": 0.1778062701225281,
      "num_tokens": 66684789.0,
      "step": 32905
    },
    {
      "entropy": 6.124145793914795,
      "epoch": 2.8759940575024032,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041769211474484204,
      "loss": 5.7112,
      "mean_token_accuracy": 0.167207171022892,
      "num_tokens": 66694734.0,
      "step": 32910
    },
    {
      "entropy": 6.0647666454315186,
      "epoch": 2.8764310058551077,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004176680202525641,
      "loss": 5.6713,
      "mean_token_accuracy": 0.16932001262903212,
      "num_tokens": 66705231.0,
      "step": 32915
    },
    {
      "entropy": 6.011656332015991,
      "epoch": 2.8768679542078126,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041764392302383893,
      "loss": 5.6314,
      "mean_token_accuracy": 0.1790419712662697,
      "num_tokens": 66715627.0,
      "step": 32920
    },
    {
      "entropy": 6.0775062084198,
      "epoch": 2.8773049025605175,
      "grad_norm": 1.25,
      "learning_rate": 0.0004176198230591284,
      "loss": 5.6242,
      "mean_token_accuracy": 0.17563146501779556,
      "num_tokens": 66725525.0,
      "step": 32925
    },
    {
      "entropy": 6.05160608291626,
      "epoch": 2.877741850913222,
      "grad_norm": 1.125,
      "learning_rate": 0.000417595720358895,
      "loss": 5.6566,
      "mean_token_accuracy": 0.16896125972270964,
      "num_tokens": 66736352.0,
      "step": 32930
    },
    {
      "entropy": 6.0349678039550785,
      "epoch": 2.878178799265927,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041757161492360096,
      "loss": 5.7222,
      "mean_token_accuracy": 0.16632694751024246,
      "num_tokens": 66745961.0,
      "step": 32935
    },
    {
      "entropy": 6.127788448333741,
      "epoch": 2.8786157476186314,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041754750675370865,
      "loss": 5.5963,
      "mean_token_accuracy": 0.17988073378801345,
      "num_tokens": 66755295.0,
      "step": 32940
    },
    {
      "entropy": 6.031705856323242,
      "epoch": 2.8790526959713363,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004175233958496804,
      "loss": 5.6168,
      "mean_token_accuracy": 0.1716320589184761,
      "num_tokens": 66765089.0,
      "step": 32945
    },
    {
      "entropy": 6.0284192085266115,
      "epoch": 2.8794896443240408,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004174992822119788,
      "loss": 5.6468,
      "mean_token_accuracy": 0.1706136718392372,
      "num_tokens": 66774995.0,
      "step": 32950
    },
    {
      "entropy": 5.996316766738891,
      "epoch": 2.8799265926767457,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004174751658410663,
      "loss": 5.6523,
      "mean_token_accuracy": 0.1783875361084938,
      "num_tokens": 66784584.0,
      "step": 32955
    },
    {
      "entropy": 6.070488309860229,
      "epoch": 2.8803635410294506,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041745104673740545,
      "loss": 5.6394,
      "mean_token_accuracy": 0.1763817235827446,
      "num_tokens": 66794747.0,
      "step": 32960
    },
    {
      "entropy": 6.097298002243042,
      "epoch": 2.880800489382155,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041742692490145895,
      "loss": 5.6443,
      "mean_token_accuracy": 0.1763642594218254,
      "num_tokens": 66804632.0,
      "step": 32965
    },
    {
      "entropy": 6.003846597671509,
      "epoch": 2.8812374377348595,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004174028003336893,
      "loss": 5.601,
      "mean_token_accuracy": 0.1714627668261528,
      "num_tokens": 66814534.0,
      "step": 32970
    },
    {
      "entropy": 6.050198602676391,
      "epoch": 2.8816743860875644,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004173786730345594,
      "loss": 5.7179,
      "mean_token_accuracy": 0.16676483303308487,
      "num_tokens": 66824629.0,
      "step": 32975
    },
    {
      "entropy": 6.09140453338623,
      "epoch": 2.8821113344402693,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004173545430045319,
      "loss": 5.6946,
      "mean_token_accuracy": 0.17067162990570067,
      "num_tokens": 66834345.0,
      "step": 32980
    },
    {
      "entropy": 6.025650835037231,
      "epoch": 2.882548282792974,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041733041024406966,
      "loss": 5.6495,
      "mean_token_accuracy": 0.178168947994709,
      "num_tokens": 66845272.0,
      "step": 32985
    },
    {
      "entropy": 6.103330945968628,
      "epoch": 2.8829852311456787,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004173062747536356,
      "loss": 5.6885,
      "mean_token_accuracy": 0.17485735714435577,
      "num_tokens": 66855022.0,
      "step": 32990
    },
    {
      "entropy": 6.052393579483033,
      "epoch": 2.883422179498383,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004172821365336926,
      "loss": 5.6144,
      "mean_token_accuracy": 0.17296697348356246,
      "num_tokens": 66864895.0,
      "step": 32995
    },
    {
      "entropy": 6.039212608337403,
      "epoch": 2.883859127851088,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004172579955847037,
      "loss": 5.6371,
      "mean_token_accuracy": 0.17909765988588333,
      "num_tokens": 66876134.0,
      "step": 33000
    },
    {
      "epoch": 2.883859127851088,
      "eval_entropy": 5.8332228808983215,
      "eval_loss": 5.701754570007324,
      "eval_mean_token_accuracy": 0.17968540220747428,
      "eval_num_tokens": 66876134.0,
      "eval_runtime": 27.7925,
      "eval_samples_per_second": 1331.8,
      "eval_steps_per_second": 166.484,
      "step": 33000
    },
    {
      "entropy": 6.027366495132446,
      "epoch": 2.8842960762037926,
      "grad_norm": 1.25,
      "learning_rate": 0.0004172338519071318,
      "loss": 5.6037,
      "mean_token_accuracy": 0.17789103388786315,
      "num_tokens": 66886089.0,
      "step": 33005
    },
    {
      "entropy": 5.939702033996582,
      "epoch": 2.8847330245564975,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041720970550144024,
      "loss": 5.5267,
      "mean_token_accuracy": 0.189230278134346,
      "num_tokens": 66896072.0,
      "step": 33010
    },
    {
      "entropy": 6.054059839248657,
      "epoch": 2.8851699729092024,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041718555636809194,
      "loss": 5.7471,
      "mean_token_accuracy": 0.17071056365966797,
      "num_tokens": 66906030.0,
      "step": 33015
    },
    {
      "entropy": 6.076147747039795,
      "epoch": 2.885606921261907,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004171614045075501,
      "loss": 5.5997,
      "mean_token_accuracy": 0.17323019504547119,
      "num_tokens": 66916714.0,
      "step": 33020
    },
    {
      "entropy": 6.053783226013183,
      "epoch": 2.8860438696146113,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004171372499202781,
      "loss": 5.6739,
      "mean_token_accuracy": 0.17005425542593003,
      "num_tokens": 66926570.0,
      "step": 33025
    },
    {
      "entropy": 6.085605096817017,
      "epoch": 2.8864808179673163,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004171130926067391,
      "loss": 5.782,
      "mean_token_accuracy": 0.16721908003091812,
      "num_tokens": 66937565.0,
      "step": 33030
    },
    {
      "entropy": 6.115561008453369,
      "epoch": 2.886917766320021,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004170889325673967,
      "loss": 5.7208,
      "mean_token_accuracy": 0.1728288635611534,
      "num_tokens": 66948363.0,
      "step": 33035
    },
    {
      "entropy": 6.043719387054443,
      "epoch": 2.8873547146727256,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041706476980271395,
      "loss": 5.6036,
      "mean_token_accuracy": 0.17635212242603301,
      "num_tokens": 66958312.0,
      "step": 33040
    },
    {
      "entropy": 6.051319551467896,
      "epoch": 2.8877916630254306,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041704060431315456,
      "loss": 5.5738,
      "mean_token_accuracy": 0.18316741436719894,
      "num_tokens": 66967880.0,
      "step": 33045
    },
    {
      "entropy": 5.981820869445801,
      "epoch": 2.888228611378135,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00041701643609918187,
      "loss": 5.564,
      "mean_token_accuracy": 0.18291160464286804,
      "num_tokens": 66977930.0,
      "step": 33050
    },
    {
      "entropy": 5.999826717376709,
      "epoch": 2.88866555973084,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004169922651612596,
      "loss": 5.719,
      "mean_token_accuracy": 0.1692126378417015,
      "num_tokens": 66987780.0,
      "step": 33055
    },
    {
      "entropy": 6.1167035579681395,
      "epoch": 2.8891025080835444,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041696809149985124,
      "loss": 5.6737,
      "mean_token_accuracy": 0.1721012070775032,
      "num_tokens": 66997910.0,
      "step": 33060
    },
    {
      "entropy": 6.040970277786255,
      "epoch": 2.8895394564362493,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041694391511542053,
      "loss": 5.5891,
      "mean_token_accuracy": 0.18029947876930236,
      "num_tokens": 67007365.0,
      "step": 33065
    },
    {
      "entropy": 5.902463293075561,
      "epoch": 2.889976404788954,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004169197360084311,
      "loss": 5.5621,
      "mean_token_accuracy": 0.18309176415205003,
      "num_tokens": 67016438.0,
      "step": 33070
    },
    {
      "entropy": 5.969402694702149,
      "epoch": 2.8904133531416587,
      "grad_norm": 1.375,
      "learning_rate": 0.00041689555417934675,
      "loss": 5.4835,
      "mean_token_accuracy": 0.19361480921506882,
      "num_tokens": 67026842.0,
      "step": 33075
    },
    {
      "entropy": 6.0829843044281,
      "epoch": 2.890850301494363,
      "grad_norm": 1.71875,
      "learning_rate": 0.00041687136962863144,
      "loss": 5.6949,
      "mean_token_accuracy": 0.17194506675004959,
      "num_tokens": 67037343.0,
      "step": 33080
    },
    {
      "entropy": 6.084178018569946,
      "epoch": 2.891287249847068,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041684718235674887,
      "loss": 5.6171,
      "mean_token_accuracy": 0.1775491639971733,
      "num_tokens": 67047352.0,
      "step": 33085
    },
    {
      "entropy": 6.0668529033660885,
      "epoch": 2.891724198199773,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004168229923641629,
      "loss": 5.7072,
      "mean_token_accuracy": 0.16562965959310533,
      "num_tokens": 67058717.0,
      "step": 33090
    },
    {
      "entropy": 6.043492984771729,
      "epoch": 2.8921611465524775,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004167987996513378,
      "loss": 5.6413,
      "mean_token_accuracy": 0.178192800283432,
      "num_tokens": 67068124.0,
      "step": 33095
    },
    {
      "entropy": 6.027216625213623,
      "epoch": 2.892598094905182,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004167746042187373,
      "loss": 5.6719,
      "mean_token_accuracy": 0.1742849215865135,
      "num_tokens": 67078906.0,
      "step": 33100
    },
    {
      "entropy": 5.960216045379639,
      "epoch": 2.893035043257887,
      "grad_norm": 2.9375,
      "learning_rate": 0.00041675040606682564,
      "loss": 5.6118,
      "mean_token_accuracy": 0.17921141535043716,
      "num_tokens": 67090424.0,
      "step": 33105
    },
    {
      "entropy": 6.09547905921936,
      "epoch": 2.8934719916105918,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041672620519606687,
      "loss": 5.6341,
      "mean_token_accuracy": 0.17610969990491868,
      "num_tokens": 67100311.0,
      "step": 33110
    },
    {
      "entropy": 6.0108537673950195,
      "epoch": 2.8939089399632962,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004167020016069253,
      "loss": 5.6567,
      "mean_token_accuracy": 0.17526133060455323,
      "num_tokens": 67109337.0,
      "step": 33115
    },
    {
      "entropy": 6.040956115722656,
      "epoch": 2.894345888316001,
      "grad_norm": 1.53125,
      "learning_rate": 0.00041667779529986504,
      "loss": 5.6436,
      "mean_token_accuracy": 0.16506059765815734,
      "num_tokens": 67119795.0,
      "step": 33120
    },
    {
      "entropy": 6.015885496139527,
      "epoch": 2.8947828366687056,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041665358627535043,
      "loss": 5.6739,
      "mean_token_accuracy": 0.16913044154644014,
      "num_tokens": 67129444.0,
      "step": 33125
    },
    {
      "entropy": 6.035404396057129,
      "epoch": 2.8952197850214105,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004166293745338458,
      "loss": 5.6907,
      "mean_token_accuracy": 0.17460355758666993,
      "num_tokens": 67138868.0,
      "step": 33130
    },
    {
      "entropy": 6.13363299369812,
      "epoch": 2.895656733374115,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004166051600758156,
      "loss": 5.7109,
      "mean_token_accuracy": 0.1723172515630722,
      "num_tokens": 67149615.0,
      "step": 33135
    },
    {
      "entropy": 6.072695350646972,
      "epoch": 2.89609368172682,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00041658094290172415,
      "loss": 5.6562,
      "mean_token_accuracy": 0.178301340341568,
      "num_tokens": 67159745.0,
      "step": 33140
    },
    {
      "entropy": 6.042237329483032,
      "epoch": 2.896530630079525,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004165567230120362,
      "loss": 5.595,
      "mean_token_accuracy": 0.17272464632987977,
      "num_tokens": 67169900.0,
      "step": 33145
    },
    {
      "entropy": 5.977771759033203,
      "epoch": 2.8969675784322293,
      "grad_norm": 6.375,
      "learning_rate": 0.0004165325004072159,
      "loss": 5.654,
      "mean_token_accuracy": 0.17896738648414612,
      "num_tokens": 67180752.0,
      "step": 33150
    },
    {
      "entropy": 6.0439976215362545,
      "epoch": 2.8974045267849338,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041650827508772826,
      "loss": 5.7064,
      "mean_token_accuracy": 0.16978997737169266,
      "num_tokens": 67191466.0,
      "step": 33155
    },
    {
      "entropy": 6.039007329940796,
      "epoch": 2.8978414751376387,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00041648404705403765,
      "loss": 5.641,
      "mean_token_accuracy": 0.17249351292848586,
      "num_tokens": 67202813.0,
      "step": 33160
    },
    {
      "entropy": 6.056124544143676,
      "epoch": 2.8982784234903436,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00041645981630660894,
      "loss": 5.5521,
      "mean_token_accuracy": 0.17824209928512574,
      "num_tokens": 67213686.0,
      "step": 33165
    },
    {
      "entropy": 6.0360212326049805,
      "epoch": 2.898715371843048,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004164355828459068,
      "loss": 5.6733,
      "mean_token_accuracy": 0.17284219861030578,
      "num_tokens": 67224955.0,
      "step": 33170
    },
    {
      "entropy": 5.995470666885376,
      "epoch": 2.899152320195753,
      "grad_norm": 1.25,
      "learning_rate": 0.000416411346672396,
      "loss": 5.606,
      "mean_token_accuracy": 0.1777358204126358,
      "num_tokens": 67234438.0,
      "step": 33175
    },
    {
      "entropy": 6.093151044845581,
      "epoch": 2.8995892685484574,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041638710778654157,
      "loss": 5.6961,
      "mean_token_accuracy": 0.17026770114898682,
      "num_tokens": 67245151.0,
      "step": 33180
    },
    {
      "entropy": 6.012516260147095,
      "epoch": 2.9000262169011624,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004163628661888083,
      "loss": 5.606,
      "mean_token_accuracy": 0.183434334397316,
      "num_tokens": 67254942.0,
      "step": 33185
    },
    {
      "entropy": 6.060494661331177,
      "epoch": 2.900463165253867,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041633862187966125,
      "loss": 5.6595,
      "mean_token_accuracy": 0.17294344902038575,
      "num_tokens": 67264435.0,
      "step": 33190
    },
    {
      "entropy": 5.992692708969116,
      "epoch": 2.9009001136065717,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004163143748595653,
      "loss": 5.5993,
      "mean_token_accuracy": 0.1799977883696556,
      "num_tokens": 67273304.0,
      "step": 33195
    },
    {
      "entropy": 6.00979380607605,
      "epoch": 2.9013370619592767,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041629012512898556,
      "loss": 5.7036,
      "mean_token_accuracy": 0.17148973047733307,
      "num_tokens": 67283713.0,
      "step": 33200
    },
    {
      "entropy": 6.000548076629639,
      "epoch": 2.901774010311981,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004162658726883873,
      "loss": 5.5249,
      "mean_token_accuracy": 0.185195192694664,
      "num_tokens": 67292420.0,
      "step": 33205
    },
    {
      "entropy": 6.014600324630737,
      "epoch": 2.9022109586646856,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004162416175382354,
      "loss": 5.6115,
      "mean_token_accuracy": 0.17966508567333223,
      "num_tokens": 67302138.0,
      "step": 33210
    },
    {
      "entropy": 6.003279495239258,
      "epoch": 2.9026479070173905,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004162173596789954,
      "loss": 5.6391,
      "mean_token_accuracy": 0.17585935145616532,
      "num_tokens": 67311776.0,
      "step": 33215
    },
    {
      "entropy": 6.098690462112427,
      "epoch": 2.9030848553700954,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004161930991111325,
      "loss": 5.6719,
      "mean_token_accuracy": 0.17564031779766082,
      "num_tokens": 67322296.0,
      "step": 33220
    },
    {
      "entropy": 6.061326694488526,
      "epoch": 2.9035218037228,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004161688358351118,
      "loss": 5.6183,
      "mean_token_accuracy": 0.16970693171024323,
      "num_tokens": 67332173.0,
      "step": 33225
    },
    {
      "entropy": 6.000789928436279,
      "epoch": 2.903958752075505,
      "grad_norm": 1.71875,
      "learning_rate": 0.000416144569851399,
      "loss": 5.6255,
      "mean_token_accuracy": 0.1754552647471428,
      "num_tokens": 67342156.0,
      "step": 33230
    },
    {
      "entropy": 6.044655179977417,
      "epoch": 2.9043957004282093,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041612030116045933,
      "loss": 5.6937,
      "mean_token_accuracy": 0.17269186079502105,
      "num_tokens": 67353208.0,
      "step": 33235
    },
    {
      "entropy": 6.068669319152832,
      "epoch": 2.904832648780914,
      "grad_norm": 1.453125,
      "learning_rate": 0.00041609602976275836,
      "loss": 5.6092,
      "mean_token_accuracy": 0.1840887576341629,
      "num_tokens": 67362577.0,
      "step": 33240
    },
    {
      "entropy": 5.995051050186158,
      "epoch": 2.9052695971336187,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004160717556587616,
      "loss": 5.5816,
      "mean_token_accuracy": 0.18181031942367554,
      "num_tokens": 67371996.0,
      "step": 33245
    },
    {
      "entropy": 5.948962259292602,
      "epoch": 2.9057065454863236,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004160474788489346,
      "loss": 5.5803,
      "mean_token_accuracy": 0.1830452337861061,
      "num_tokens": 67381714.0,
      "step": 33250
    },
    {
      "entropy": 6.027104282379151,
      "epoch": 2.9061434938390285,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041602319933374305,
      "loss": 5.5972,
      "mean_token_accuracy": 0.1836426466703415,
      "num_tokens": 67391551.0,
      "step": 33255
    },
    {
      "entropy": 6.01473913192749,
      "epoch": 2.906580442191733,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004159989171136527,
      "loss": 5.6344,
      "mean_token_accuracy": 0.16958957612514497,
      "num_tokens": 67400816.0,
      "step": 33260
    },
    {
      "entropy": 6.043849515914917,
      "epoch": 2.9070173905444374,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004159746321891292,
      "loss": 5.6385,
      "mean_token_accuracy": 0.17911465764045714,
      "num_tokens": 67410633.0,
      "step": 33265
    },
    {
      "entropy": 6.04818754196167,
      "epoch": 2.9074543388971423,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004159503445606385,
      "loss": 5.6178,
      "mean_token_accuracy": 0.1826648384332657,
      "num_tokens": 67420588.0,
      "step": 33270
    },
    {
      "entropy": 5.990749645233154,
      "epoch": 2.9078912872498472,
      "grad_norm": 1.734375,
      "learning_rate": 0.00041592605422864615,
      "loss": 5.65,
      "mean_token_accuracy": 0.1790078103542328,
      "num_tokens": 67430494.0,
      "step": 33275
    },
    {
      "entropy": 5.9406591892242435,
      "epoch": 2.9083282356025517,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004159017611936183,
      "loss": 5.5596,
      "mean_token_accuracy": 0.19215232878923416,
      "num_tokens": 67440692.0,
      "step": 33280
    },
    {
      "entropy": 6.0291900634765625,
      "epoch": 2.9087651839552566,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041587746545602094,
      "loss": 5.6245,
      "mean_token_accuracy": 0.17778499573469161,
      "num_tokens": 67450936.0,
      "step": 33285
    },
    {
      "entropy": 6.059746837615966,
      "epoch": 2.909202132307961,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004158531670163199,
      "loss": 5.6358,
      "mean_token_accuracy": 0.1802768349647522,
      "num_tokens": 67460997.0,
      "step": 33290
    },
    {
      "entropy": 6.030207729339599,
      "epoch": 2.909639080660666,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004158288658749813,
      "loss": 5.6615,
      "mean_token_accuracy": 0.17749529629945754,
      "num_tokens": 67470685.0,
      "step": 33295
    },
    {
      "entropy": 6.039952993392944,
      "epoch": 2.9100760290133705,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00041580456203247127,
      "loss": 5.6101,
      "mean_token_accuracy": 0.18069436848163606,
      "num_tokens": 67481734.0,
      "step": 33300
    },
    {
      "entropy": 5.936866569519043,
      "epoch": 2.9105129773660754,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000415780255489256,
      "loss": 5.4279,
      "mean_token_accuracy": 0.19863445907831193,
      "num_tokens": 67491290.0,
      "step": 33305
    },
    {
      "entropy": 5.993852376937866,
      "epoch": 2.9109499257187803,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004157559462458016,
      "loss": 5.6759,
      "mean_token_accuracy": 0.17502873092889787,
      "num_tokens": 67501758.0,
      "step": 33310
    },
    {
      "entropy": 6.028465414047242,
      "epoch": 2.9113868740714848,
      "grad_norm": 1.125,
      "learning_rate": 0.00041573163430257435,
      "loss": 5.6449,
      "mean_token_accuracy": 0.17740681022405624,
      "num_tokens": 67511435.0,
      "step": 33315
    },
    {
      "entropy": 6.054773807525635,
      "epoch": 2.9118238224241892,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041570731966004065,
      "loss": 5.6688,
      "mean_token_accuracy": 0.16998017579317093,
      "num_tokens": 67521408.0,
      "step": 33320
    },
    {
      "entropy": 6.060177612304687,
      "epoch": 2.912260770776894,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004156830023186669,
      "loss": 5.701,
      "mean_token_accuracy": 0.17219492942094802,
      "num_tokens": 67531631.0,
      "step": 33325
    },
    {
      "entropy": 6.079078578948975,
      "epoch": 2.912697719129599,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004156586822789194,
      "loss": 5.6759,
      "mean_token_accuracy": 0.16186123192310334,
      "num_tokens": 67542265.0,
      "step": 33330
    },
    {
      "entropy": 6.041330337524414,
      "epoch": 2.9131346674823035,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00041563435954126467,
      "loss": 5.6749,
      "mean_token_accuracy": 0.17760352045297623,
      "num_tokens": 67553664.0,
      "step": 33335
    },
    {
      "entropy": 6.056892347335816,
      "epoch": 2.9135716158350085,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004156100341061693,
      "loss": 5.5969,
      "mean_token_accuracy": 0.17449628114700316,
      "num_tokens": 67564500.0,
      "step": 33340
    },
    {
      "entropy": 5.985874843597412,
      "epoch": 2.914008564187713,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004155857059740997,
      "loss": 5.5658,
      "mean_token_accuracy": 0.18266146034002304,
      "num_tokens": 67573611.0,
      "step": 33345
    },
    {
      "entropy": 5.99328989982605,
      "epoch": 2.914445512540418,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004155613751455226,
      "loss": 5.6162,
      "mean_token_accuracy": 0.18461614102125168,
      "num_tokens": 67583721.0,
      "step": 33350
    },
    {
      "entropy": 6.016164684295655,
      "epoch": 2.9148824608931223,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041553704162090475,
      "loss": 5.6482,
      "mean_token_accuracy": 0.18139889389276503,
      "num_tokens": 67593381.0,
      "step": 33355
    },
    {
      "entropy": 6.063201141357422,
      "epoch": 2.915319409245827,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004155127054007127,
      "loss": 5.6838,
      "mean_token_accuracy": 0.17399442940950394,
      "num_tokens": 67603799.0,
      "step": 33360
    },
    {
      "entropy": 6.109624290466309,
      "epoch": 2.9157563575985317,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041548836648541346,
      "loss": 5.6703,
      "mean_token_accuracy": 0.17248117923736572,
      "num_tokens": 67613543.0,
      "step": 33365
    },
    {
      "entropy": 6.009053039550781,
      "epoch": 2.9161933059512366,
      "grad_norm": 1.484375,
      "learning_rate": 0.00041546402487547363,
      "loss": 5.6886,
      "mean_token_accuracy": 0.17563792765140535,
      "num_tokens": 67623302.0,
      "step": 33370
    },
    {
      "entropy": 6.033738708496093,
      "epoch": 2.916630254303941,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004154396805713602,
      "loss": 5.5489,
      "mean_token_accuracy": 0.1861480087041855,
      "num_tokens": 67633655.0,
      "step": 33375
    },
    {
      "entropy": 6.062474393844605,
      "epoch": 2.917067202656646,
      "grad_norm": 1.40625,
      "learning_rate": 0.00041541533357354013,
      "loss": 5.6793,
      "mean_token_accuracy": 0.17416726797819138,
      "num_tokens": 67644832.0,
      "step": 33380
    },
    {
      "entropy": 6.053424596786499,
      "epoch": 2.917504151009351,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004153909838824804,
      "loss": 5.6342,
      "mean_token_accuracy": 0.1736384630203247,
      "num_tokens": 67655290.0,
      "step": 33385
    },
    {
      "entropy": 6.016648292541504,
      "epoch": 2.9179410993620554,
      "grad_norm": 1.5625,
      "learning_rate": 0.000415366631498648,
      "loss": 5.5524,
      "mean_token_accuracy": 0.18512767255306245,
      "num_tokens": 67664252.0,
      "step": 33390
    },
    {
      "entropy": 6.028489303588867,
      "epoch": 2.9183780477147603,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004153422764225101,
      "loss": 5.6268,
      "mean_token_accuracy": 0.17507093250751496,
      "num_tokens": 67674655.0,
      "step": 33395
    },
    {
      "entropy": 6.023326206207275,
      "epoch": 2.9188149960674648,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004153179186545338,
      "loss": 5.5722,
      "mean_token_accuracy": 0.18026058673858641,
      "num_tokens": 67685126.0,
      "step": 33400
    },
    {
      "entropy": 5.956499910354614,
      "epoch": 2.9192519444201697,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041529355819518626,
      "loss": 5.5622,
      "mean_token_accuracy": 0.17997201383113862,
      "num_tokens": 67694111.0,
      "step": 33405
    },
    {
      "entropy": 6.0326869010925295,
      "epoch": 2.919688892772874,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041526919504493475,
      "loss": 5.642,
      "mean_token_accuracy": 0.1747376635670662,
      "num_tokens": 67704229.0,
      "step": 33410
    },
    {
      "entropy": 6.022142362594605,
      "epoch": 2.920125841125579,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041524482920424666,
      "loss": 5.6508,
      "mean_token_accuracy": 0.17897726148366927,
      "num_tokens": 67714625.0,
      "step": 33415
    },
    {
      "entropy": 6.024442863464356,
      "epoch": 2.9205627894782835,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041522046067358917,
      "loss": 5.6388,
      "mean_token_accuracy": 0.17314197421073912,
      "num_tokens": 67724094.0,
      "step": 33420
    },
    {
      "entropy": 6.0165441036224365,
      "epoch": 2.9209997378309884,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041519608945342985,
      "loss": 5.6283,
      "mean_token_accuracy": 0.17950611263513566,
      "num_tokens": 67733233.0,
      "step": 33425
    },
    {
      "entropy": 6.043569040298462,
      "epoch": 2.921436686183693,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000415171715544236,
      "loss": 5.5683,
      "mean_token_accuracy": 0.18124632388353348,
      "num_tokens": 67742781.0,
      "step": 33430
    },
    {
      "entropy": 6.037564325332641,
      "epoch": 2.921873634536398,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00041514733894647526,
      "loss": 5.6048,
      "mean_token_accuracy": 0.17706320732831954,
      "num_tokens": 67753166.0,
      "step": 33435
    },
    {
      "entropy": 6.032747459411621,
      "epoch": 2.9223105828891027,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004151229596606151,
      "loss": 5.7471,
      "mean_token_accuracy": 0.17215062081813812,
      "num_tokens": 67762599.0,
      "step": 33440
    },
    {
      "entropy": 6.0265021324157715,
      "epoch": 2.922747531241807,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004150985776871231,
      "loss": 5.6164,
      "mean_token_accuracy": 0.17903145551681518,
      "num_tokens": 67772620.0,
      "step": 33445
    },
    {
      "entropy": 6.179292678833008,
      "epoch": 2.9231844795945117,
      "grad_norm": 1.078125,
      "learning_rate": 0.000415074193026467,
      "loss": 5.7342,
      "mean_token_accuracy": 0.16174079179763795,
      "num_tokens": 67782962.0,
      "step": 33450
    },
    {
      "entropy": 6.047802734375,
      "epoch": 2.9236214279472166,
      "grad_norm": 1.5,
      "learning_rate": 0.00041504980567911453,
      "loss": 5.6704,
      "mean_token_accuracy": 0.17389266490936278,
      "num_tokens": 67793168.0,
      "step": 33455
    },
    {
      "entropy": 5.993545341491699,
      "epoch": 2.9240583762999215,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041502541564553336,
      "loss": 5.6754,
      "mean_token_accuracy": 0.1700911670923233,
      "num_tokens": 67803317.0,
      "step": 33460
    },
    {
      "entropy": 6.030530595779419,
      "epoch": 2.924495324652626,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00041500102292619135,
      "loss": 5.5965,
      "mean_token_accuracy": 0.17099803686141968,
      "num_tokens": 67812594.0,
      "step": 33465
    },
    {
      "entropy": 6.062567663192749,
      "epoch": 2.924932273005331,
      "grad_norm": 1.25,
      "learning_rate": 0.00041497662752155637,
      "loss": 5.5813,
      "mean_token_accuracy": 0.18214988112449645,
      "num_tokens": 67822277.0,
      "step": 33470
    },
    {
      "entropy": 6.071072149276733,
      "epoch": 2.9253692213580353,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004149522294320963,
      "loss": 5.6788,
      "mean_token_accuracy": 0.1733776420354843,
      "num_tokens": 67832948.0,
      "step": 33475
    },
    {
      "entropy": 5.987959861755371,
      "epoch": 2.9258061697107403,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041492782865827914,
      "loss": 5.5789,
      "mean_token_accuracy": 0.17807616293430328,
      "num_tokens": 67843965.0,
      "step": 33480
    },
    {
      "entropy": 6.017596292495727,
      "epoch": 2.9262431180634447,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041490342520057293,
      "loss": 5.6616,
      "mean_token_accuracy": 0.173825241625309,
      "num_tokens": 67854079.0,
      "step": 33485
    },
    {
      "entropy": 6.041804265975952,
      "epoch": 2.9266800664161496,
      "grad_norm": 1.734375,
      "learning_rate": 0.00041487901905944575,
      "loss": 5.6221,
      "mean_token_accuracy": 0.17696609497070312,
      "num_tokens": 67865218.0,
      "step": 33490
    },
    {
      "entropy": 6.042144775390625,
      "epoch": 2.9271170147688546,
      "grad_norm": 1.390625,
      "learning_rate": 0.00041485461023536564,
      "loss": 5.5984,
      "mean_token_accuracy": 0.18498313128948213,
      "num_tokens": 67875229.0,
      "step": 33495
    },
    {
      "entropy": 6.037544298171997,
      "epoch": 2.927553963121559,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004148301987288008,
      "loss": 5.7105,
      "mean_token_accuracy": 0.17208757996559143,
      "num_tokens": 67885795.0,
      "step": 33500
    },
    {
      "entropy": 6.0126824378967285,
      "epoch": 2.9279909114742635,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004148057845402195,
      "loss": 5.6175,
      "mean_token_accuracy": 0.18335498869419098,
      "num_tokens": 67896668.0,
      "step": 33505
    },
    {
      "entropy": 6.002299022674561,
      "epoch": 2.9284278598269684,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041478136767009004,
      "loss": 5.5777,
      "mean_token_accuracy": 0.18400725722312927,
      "num_tokens": 67907082.0,
      "step": 33510
    },
    {
      "entropy": 6.054030323028565,
      "epoch": 2.9288648081796733,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004147569481188806,
      "loss": 5.6665,
      "mean_token_accuracy": 0.17607434689998627,
      "num_tokens": 67917517.0,
      "step": 33515
    },
    {
      "entropy": 6.104587125778198,
      "epoch": 2.929301756532378,
      "grad_norm": 1.25,
      "learning_rate": 0.00041473252588705977,
      "loss": 5.6825,
      "mean_token_accuracy": 0.17159879505634307,
      "num_tokens": 67927332.0,
      "step": 33520
    },
    {
      "entropy": 6.079684019088745,
      "epoch": 2.9297387048850827,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041470810097509587,
      "loss": 5.659,
      "mean_token_accuracy": 0.17491605281829833,
      "num_tokens": 67936939.0,
      "step": 33525
    },
    {
      "entropy": 5.981093788146973,
      "epoch": 2.930175653237787,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004146836733834573,
      "loss": 5.6452,
      "mean_token_accuracy": 0.176852448284626,
      "num_tokens": 67947553.0,
      "step": 33530
    },
    {
      "entropy": 6.032007026672363,
      "epoch": 2.930612601590492,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004146592431126128,
      "loss": 5.5876,
      "mean_token_accuracy": 0.17568584829568862,
      "num_tokens": 67958489.0,
      "step": 33535
    },
    {
      "entropy": 6.0206481456756595,
      "epoch": 2.9310495499431966,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004146348101630307,
      "loss": 5.5678,
      "mean_token_accuracy": 0.17535132467746734,
      "num_tokens": 67967305.0,
      "step": 33540
    },
    {
      "entropy": 5.980746078491211,
      "epoch": 2.9314864982959015,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004146103745351799,
      "loss": 5.6096,
      "mean_token_accuracy": 0.17384326010942458,
      "num_tokens": 67977400.0,
      "step": 33545
    },
    {
      "entropy": 6.026413440704346,
      "epoch": 2.9319234466486064,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004145859362295289,
      "loss": 5.6162,
      "mean_token_accuracy": 0.17986689358949662,
      "num_tokens": 67987275.0,
      "step": 33550
    },
    {
      "entropy": 6.160294008255005,
      "epoch": 2.932360395001311,
      "grad_norm": 1.625,
      "learning_rate": 0.0004145614952465465,
      "loss": 5.8096,
      "mean_token_accuracy": 0.16193373054265975,
      "num_tokens": 67999002.0,
      "step": 33555
    },
    {
      "entropy": 6.07983365058899,
      "epoch": 2.9327973433540153,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00041453705158670147,
      "loss": 5.5887,
      "mean_token_accuracy": 0.18542942106723787,
      "num_tokens": 68008504.0,
      "step": 33560
    },
    {
      "entropy": 6.01905632019043,
      "epoch": 2.9332342917067202,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041451260525046267,
      "loss": 5.6195,
      "mean_token_accuracy": 0.17716582715511323,
      "num_tokens": 68019522.0,
      "step": 33565
    },
    {
      "entropy": 6.008382129669189,
      "epoch": 2.933671240059425,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041448815623829907,
      "loss": 5.6628,
      "mean_token_accuracy": 0.17437774389982225,
      "num_tokens": 68029991.0,
      "step": 33570
    },
    {
      "entropy": 6.031062364578247,
      "epoch": 2.9341081884121296,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004144637045506795,
      "loss": 5.6545,
      "mean_token_accuracy": 0.1730638027191162,
      "num_tokens": 68040347.0,
      "step": 33575
    },
    {
      "entropy": 6.077442598342896,
      "epoch": 2.9345451367648345,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000414439250188073,
      "loss": 5.6137,
      "mean_token_accuracy": 0.17840517163276673,
      "num_tokens": 68050932.0,
      "step": 33580
    },
    {
      "entropy": 6.106505250930786,
      "epoch": 2.934982085117539,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004144147931509486,
      "loss": 5.6389,
      "mean_token_accuracy": 0.17545515298843384,
      "num_tokens": 68061104.0,
      "step": 33585
    },
    {
      "entropy": 6.0339583396911625,
      "epoch": 2.935419033470244,
      "grad_norm": 1.25,
      "learning_rate": 0.00041439033343977546,
      "loss": 5.6065,
      "mean_token_accuracy": 0.18027054369449616,
      "num_tokens": 68070374.0,
      "step": 33590
    },
    {
      "entropy": 6.011499643325806,
      "epoch": 2.9358559818229484,
      "grad_norm": 2.0,
      "learning_rate": 0.00041436587105502263,
      "loss": 5.6648,
      "mean_token_accuracy": 0.17461800575256348,
      "num_tokens": 68079996.0,
      "step": 33595
    },
    {
      "entropy": 6.0029052734375,
      "epoch": 2.9362929301756533,
      "grad_norm": 1.125,
      "learning_rate": 0.0004143414059971594,
      "loss": 5.6751,
      "mean_token_accuracy": 0.17315987795591353,
      "num_tokens": 68090322.0,
      "step": 33600
    },
    {
      "entropy": 5.991623258590698,
      "epoch": 2.936729878528358,
      "grad_norm": 1.109375,
      "learning_rate": 0.000414316938266655,
      "loss": 5.5281,
      "mean_token_accuracy": 0.17883714139461518,
      "num_tokens": 68099482.0,
      "step": 33605
    },
    {
      "entropy": 5.984919118881225,
      "epoch": 2.9371668268810627,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004142924678639787,
      "loss": 5.6164,
      "mean_token_accuracy": 0.1812318593263626,
      "num_tokens": 68109852.0,
      "step": 33610
    },
    {
      "entropy": 5.997080707550049,
      "epoch": 2.937603775233767,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041426799478959993,
      "loss": 5.6103,
      "mean_token_accuracy": 0.1750684529542923,
      "num_tokens": 68118721.0,
      "step": 33615
    },
    {
      "entropy": 6.058450222015381,
      "epoch": 2.938040723586472,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004142435190439879,
      "loss": 5.5989,
      "mean_token_accuracy": 0.17681423276662828,
      "num_tokens": 68129080.0,
      "step": 33620
    },
    {
      "entropy": 6.062423992156982,
      "epoch": 2.938477671939177,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041421904062761234,
      "loss": 5.6284,
      "mean_token_accuracy": 0.17721425592899323,
      "num_tokens": 68138154.0,
      "step": 33625
    },
    {
      "entropy": 6.003808164596558,
      "epoch": 2.9389146202918814,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004141945595409426,
      "loss": 5.7247,
      "mean_token_accuracy": 0.1722302407026291,
      "num_tokens": 68148736.0,
      "step": 33630
    },
    {
      "entropy": 6.033364152908325,
      "epoch": 2.9393515686445864,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004141700757844483,
      "loss": 5.5848,
      "mean_token_accuracy": 0.18052540123462676,
      "num_tokens": 68159041.0,
      "step": 33635
    },
    {
      "entropy": 6.043066453933716,
      "epoch": 2.939788516997291,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000414145589358599,
      "loss": 5.6384,
      "mean_token_accuracy": 0.1776345804333687,
      "num_tokens": 68168715.0,
      "step": 33640
    },
    {
      "entropy": 6.002152967453003,
      "epoch": 2.9402254653499957,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041412110026386436,
      "loss": 5.6502,
      "mean_token_accuracy": 0.1788946196436882,
      "num_tokens": 68178863.0,
      "step": 33645
    },
    {
      "entropy": 6.095081567764282,
      "epoch": 2.9406624137027,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004140966085007142,
      "loss": 5.6791,
      "mean_token_accuracy": 0.1720558688044548,
      "num_tokens": 68190131.0,
      "step": 33650
    },
    {
      "entropy": 6.038651037216186,
      "epoch": 2.941099362055405,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00041407211406961815,
      "loss": 5.6162,
      "mean_token_accuracy": 0.17879853993654252,
      "num_tokens": 68200193.0,
      "step": 33655
    },
    {
      "entropy": 6.0346574783325195,
      "epoch": 2.94153631040811,
      "grad_norm": 1.125,
      "learning_rate": 0.00041404761697104605,
      "loss": 5.6136,
      "mean_token_accuracy": 0.17648480832576752,
      "num_tokens": 68210714.0,
      "step": 33660
    },
    {
      "entropy": 6.030304336547852,
      "epoch": 2.9419732587608145,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004140231172054678,
      "loss": 5.6527,
      "mean_token_accuracy": 0.17603821009397508,
      "num_tokens": 68220809.0,
      "step": 33665
    },
    {
      "entropy": 6.061049890518189,
      "epoch": 2.942410207113519,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004139986147733534,
      "loss": 5.6258,
      "mean_token_accuracy": 0.1778973788022995,
      "num_tokens": 68229887.0,
      "step": 33670
    },
    {
      "entropy": 6.0198752880096436,
      "epoch": 2.942847155466224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041397410967517265,
      "loss": 5.6154,
      "mean_token_accuracy": 0.17887613773345948,
      "num_tokens": 68240491.0,
      "step": 33675
    },
    {
      "entropy": 5.989974689483643,
      "epoch": 2.943284103818929,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041394960191139566,
      "loss": 5.706,
      "mean_token_accuracy": 0.17474722415208815,
      "num_tokens": 68250302.0,
      "step": 33680
    },
    {
      "entropy": 5.9828596115112305,
      "epoch": 2.9437210521716333,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004139250914824925,
      "loss": 5.5391,
      "mean_token_accuracy": 0.18197221755981446,
      "num_tokens": 68259245.0,
      "step": 33685
    },
    {
      "entropy": 6.043305683135986,
      "epoch": 2.944158000524338,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041390057838893335,
      "loss": 5.5656,
      "mean_token_accuracy": 0.18558724224567413,
      "num_tokens": 68269096.0,
      "step": 33690
    },
    {
      "entropy": 6.030605602264404,
      "epoch": 2.9445949488770427,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041387606263118825,
      "loss": 5.6402,
      "mean_token_accuracy": 0.17628934979438782,
      "num_tokens": 68278949.0,
      "step": 33695
    },
    {
      "entropy": 5.955866861343384,
      "epoch": 2.9450318972297476,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004138515442097275,
      "loss": 5.5678,
      "mean_token_accuracy": 0.1822138398885727,
      "num_tokens": 68288526.0,
      "step": 33700
    },
    {
      "entropy": 6.046344757080078,
      "epoch": 2.945468845582452,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004138270231250214,
      "loss": 5.5583,
      "mean_token_accuracy": 0.1780635178089142,
      "num_tokens": 68297997.0,
      "step": 33705
    },
    {
      "entropy": 5.981385660171509,
      "epoch": 2.945905793935157,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004138024993775403,
      "loss": 5.6701,
      "mean_token_accuracy": 0.17577019780874253,
      "num_tokens": 68307316.0,
      "step": 33710
    },
    {
      "entropy": 6.0278645038604735,
      "epoch": 2.9463427422878614,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004137779729677544,
      "loss": 5.532,
      "mean_token_accuracy": 0.18519419580698013,
      "num_tokens": 68317124.0,
      "step": 33715
    },
    {
      "entropy": 5.959818744659424,
      "epoch": 2.9467796906405663,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004137534438961344,
      "loss": 5.6246,
      "mean_token_accuracy": 0.17821103036403657,
      "num_tokens": 68326576.0,
      "step": 33720
    },
    {
      "entropy": 6.047134447097778,
      "epoch": 2.947216638993271,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041372891216315055,
      "loss": 5.5838,
      "mean_token_accuracy": 0.18232387453317642,
      "num_tokens": 68336999.0,
      "step": 33725
    },
    {
      "entropy": 6.111956071853638,
      "epoch": 2.9476535873459757,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041370437776927347,
      "loss": 5.6195,
      "mean_token_accuracy": 0.1788083255290985,
      "num_tokens": 68347084.0,
      "step": 33730
    },
    {
      "entropy": 5.899163103103637,
      "epoch": 2.9480905356986806,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004136798407149738,
      "loss": 5.5951,
      "mean_token_accuracy": 0.17718974202871324,
      "num_tokens": 68357136.0,
      "step": 33735
    },
    {
      "entropy": 6.0091673851013185,
      "epoch": 2.948527484051385,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041365530100072215,
      "loss": 5.6904,
      "mean_token_accuracy": 0.17278377413749696,
      "num_tokens": 68367020.0,
      "step": 33740
    },
    {
      "entropy": 6.145753717422485,
      "epoch": 2.9489644324040896,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004136307586269891,
      "loss": 5.6043,
      "mean_token_accuracy": 0.18326887041330336,
      "num_tokens": 68378436.0,
      "step": 33745
    },
    {
      "entropy": 5.98725004196167,
      "epoch": 2.9494013807567945,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041360621359424544,
      "loss": 5.5882,
      "mean_token_accuracy": 0.1740494042634964,
      "num_tokens": 68387777.0,
      "step": 33750
    },
    {
      "entropy": 5.909999799728394,
      "epoch": 2.9498383291094994,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041358166590296205,
      "loss": 5.5921,
      "mean_token_accuracy": 0.17802691459655762,
      "num_tokens": 68397648.0,
      "step": 33755
    },
    {
      "entropy": 6.069326543807984,
      "epoch": 2.950275277462204,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041355711555360965,
      "loss": 5.7317,
      "mean_token_accuracy": 0.1708195447921753,
      "num_tokens": 68409539.0,
      "step": 33760
    },
    {
      "entropy": 6.115047931671143,
      "epoch": 2.9507122258149088,
      "grad_norm": 1.4375,
      "learning_rate": 0.00041353256254665914,
      "loss": 5.6788,
      "mean_token_accuracy": 0.17372955232858658,
      "num_tokens": 68419626.0,
      "step": 33765
    },
    {
      "entropy": 6.034664869308472,
      "epoch": 2.9511491741676132,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041350800688258157,
      "loss": 5.6117,
      "mean_token_accuracy": 0.1841362714767456,
      "num_tokens": 68429289.0,
      "step": 33770
    },
    {
      "entropy": 6.0110687732696535,
      "epoch": 2.951586122520318,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041348344856184785,
      "loss": 5.6912,
      "mean_token_accuracy": 0.1759934291243553,
      "num_tokens": 68440186.0,
      "step": 33775
    },
    {
      "entropy": 6.028170728683472,
      "epoch": 2.9520230708730226,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041345888758492896,
      "loss": 5.5613,
      "mean_token_accuracy": 0.18456137776374817,
      "num_tokens": 68450292.0,
      "step": 33780
    },
    {
      "entropy": 6.030490493774414,
      "epoch": 2.9524600192257275,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041343432395229605,
      "loss": 5.6602,
      "mean_token_accuracy": 0.17034459263086318,
      "num_tokens": 68460419.0,
      "step": 33785
    },
    {
      "entropy": 5.972700834274292,
      "epoch": 2.9528969675784325,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004134097576644203,
      "loss": 5.6162,
      "mean_token_accuracy": 0.17703873217105864,
      "num_tokens": 68470488.0,
      "step": 33790
    },
    {
      "entropy": 6.033439350128174,
      "epoch": 2.953333915931137,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041338518872177286,
      "loss": 5.629,
      "mean_token_accuracy": 0.177373905479908,
      "num_tokens": 68479872.0,
      "step": 33795
    },
    {
      "entropy": 6.080256700515747,
      "epoch": 2.9537708642838414,
      "grad_norm": 1.40625,
      "learning_rate": 0.00041336061712482497,
      "loss": 5.6561,
      "mean_token_accuracy": 0.17455149739980697,
      "num_tokens": 68491086.0,
      "step": 33800
    },
    {
      "entropy": 6.106858348846435,
      "epoch": 2.9542078126365463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041333604287404805,
      "loss": 5.7088,
      "mean_token_accuracy": 0.1666451632976532,
      "num_tokens": 68501136.0,
      "step": 33805
    },
    {
      "entropy": 5.981393432617187,
      "epoch": 2.954644760989251,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004133114659699132,
      "loss": 5.5624,
      "mean_token_accuracy": 0.1829144164919853,
      "num_tokens": 68511579.0,
      "step": 33810
    },
    {
      "entropy": 5.980108022689819,
      "epoch": 2.9550817093419557,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041328688641289205,
      "loss": 5.5944,
      "mean_token_accuracy": 0.18180596977472305,
      "num_tokens": 68521716.0,
      "step": 33815
    },
    {
      "entropy": 6.01460862159729,
      "epoch": 2.9555186576946606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041326230420345584,
      "loss": 5.6413,
      "mean_token_accuracy": 0.17893314808607103,
      "num_tokens": 68531814.0,
      "step": 33820
    },
    {
      "entropy": 5.956286191940308,
      "epoch": 2.955955606047365,
      "grad_norm": 1.125,
      "learning_rate": 0.0004132377193420763,
      "loss": 5.5322,
      "mean_token_accuracy": 0.1851096749305725,
      "num_tokens": 68541388.0,
      "step": 33825
    },
    {
      "entropy": 6.075498390197754,
      "epoch": 2.95639255440007,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041321313182922484,
      "loss": 5.7372,
      "mean_token_accuracy": 0.16740028336644172,
      "num_tokens": 68552259.0,
      "step": 33830
    },
    {
      "entropy": 6.056469058990478,
      "epoch": 2.9568295027527745,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004131885416653731,
      "loss": 5.6118,
      "mean_token_accuracy": 0.1820767715573311,
      "num_tokens": 68562001.0,
      "step": 33835
    },
    {
      "entropy": 6.041748523712158,
      "epoch": 2.9572664511054794,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004131639488509927,
      "loss": 5.6188,
      "mean_token_accuracy": 0.17540107369422914,
      "num_tokens": 68571636.0,
      "step": 33840
    },
    {
      "entropy": 6.062867546081543,
      "epoch": 2.9577033994581843,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004131393533865554,
      "loss": 5.7219,
      "mean_token_accuracy": 0.16862320750951768,
      "num_tokens": 68581448.0,
      "step": 33845
    },
    {
      "entropy": 6.109378480911255,
      "epoch": 2.9581403478108887,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041311475527253284,
      "loss": 5.7338,
      "mean_token_accuracy": 0.16959771513938904,
      "num_tokens": 68591880.0,
      "step": 33850
    },
    {
      "entropy": 6.11752233505249,
      "epoch": 2.958577296163593,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00041309015450939694,
      "loss": 5.7078,
      "mean_token_accuracy": 0.17508587688207627,
      "num_tokens": 68603025.0,
      "step": 33855
    },
    {
      "entropy": 6.049499082565307,
      "epoch": 2.959014244516298,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004130655510976196,
      "loss": 5.7114,
      "mean_token_accuracy": 0.17089732438325883,
      "num_tokens": 68613580.0,
      "step": 33860
    },
    {
      "entropy": 6.0788990497589115,
      "epoch": 2.959451192869003,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00041304094503767256,
      "loss": 5.7165,
      "mean_token_accuracy": 0.17255432903766632,
      "num_tokens": 68624451.0,
      "step": 33865
    },
    {
      "entropy": 6.00931568145752,
      "epoch": 2.9598881412217075,
      "grad_norm": 2.0625,
      "learning_rate": 0.00041301633633002784,
      "loss": 5.627,
      "mean_token_accuracy": 0.17695718258619308,
      "num_tokens": 68634408.0,
      "step": 33870
    },
    {
      "entropy": 6.017003107070923,
      "epoch": 2.9603250895744124,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004129917249751575,
      "loss": 5.5773,
      "mean_token_accuracy": 0.17789889127016068,
      "num_tokens": 68643795.0,
      "step": 33875
    },
    {
      "entropy": 6.011682510375977,
      "epoch": 2.960762037927117,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004129671109735336,
      "loss": 5.5788,
      "mean_token_accuracy": 0.1766352817416191,
      "num_tokens": 68653669.0,
      "step": 33880
    },
    {
      "entropy": 6.068315172195435,
      "epoch": 2.961198986279822,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041294249432562825,
      "loss": 5.6588,
      "mean_token_accuracy": 0.17131030261516572,
      "num_tokens": 68663094.0,
      "step": 33885
    },
    {
      "entropy": 5.976094722747803,
      "epoch": 2.9616359346325263,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00041291787503191347,
      "loss": 5.5977,
      "mean_token_accuracy": 0.1897681325674057,
      "num_tokens": 68673561.0,
      "step": 33890
    },
    {
      "entropy": 6.038009595870972,
      "epoch": 2.962072882985231,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041289325309286167,
      "loss": 5.7092,
      "mean_token_accuracy": 0.16313636898994446,
      "num_tokens": 68683624.0,
      "step": 33895
    },
    {
      "entropy": 6.0880786895751955,
      "epoch": 2.962509831337936,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041286862850894497,
      "loss": 5.6824,
      "mean_token_accuracy": 0.1749862924218178,
      "num_tokens": 68693805.0,
      "step": 33900
    },
    {
      "entropy": 6.089125394821167,
      "epoch": 2.9629467796906406,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004128440012806357,
      "loss": 5.6766,
      "mean_token_accuracy": 0.1756579488515854,
      "num_tokens": 68704020.0,
      "step": 33905
    },
    {
      "entropy": 6.053240823745727,
      "epoch": 2.963383728043345,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00041281937140840634,
      "loss": 5.6219,
      "mean_token_accuracy": 0.17149950712919235,
      "num_tokens": 68714486.0,
      "step": 33910
    },
    {
      "entropy": 6.0109058856964115,
      "epoch": 2.96382067639605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004127947388927292,
      "loss": 5.6007,
      "mean_token_accuracy": 0.1767755478620529,
      "num_tokens": 68724759.0,
      "step": 33915
    },
    {
      "entropy": 5.955769634246826,
      "epoch": 2.964257624748755,
      "grad_norm": 1.125,
      "learning_rate": 0.0004127701037340767,
      "loss": 5.549,
      "mean_token_accuracy": 0.1825248047709465,
      "num_tokens": 68734826.0,
      "step": 33920
    },
    {
      "entropy": 5.995889377593994,
      "epoch": 2.9646945731014593,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041274546593292143,
      "loss": 5.6511,
      "mean_token_accuracy": 0.17646472603082658,
      "num_tokens": 68744786.0,
      "step": 33925
    },
    {
      "entropy": 6.0836046695709225,
      "epoch": 2.9651315214541643,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004127208254897361,
      "loss": 5.6797,
      "mean_token_accuracy": 0.17229376286268233,
      "num_tokens": 68754751.0,
      "step": 33930
    },
    {
      "entropy": 6.120688343048096,
      "epoch": 2.9655684698068687,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041269618240499296,
      "loss": 5.6594,
      "mean_token_accuracy": 0.16892863512039186,
      "num_tokens": 68765652.0,
      "step": 33935
    },
    {
      "entropy": 6.020415449142456,
      "epoch": 2.9660054181595736,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041267153667916506,
      "loss": 5.6015,
      "mean_token_accuracy": 0.18102313727140426,
      "num_tokens": 68775587.0,
      "step": 33940
    },
    {
      "entropy": 6.027443075180054,
      "epoch": 2.966442366512278,
      "grad_norm": 1.578125,
      "learning_rate": 0.00041264688831272483,
      "loss": 5.6377,
      "mean_token_accuracy": 0.1703471153974533,
      "num_tokens": 68786678.0,
      "step": 33945
    },
    {
      "entropy": 5.997250986099243,
      "epoch": 2.966879314864983,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004126222373061452,
      "loss": 5.5046,
      "mean_token_accuracy": 0.1854551464319229,
      "num_tokens": 68796108.0,
      "step": 33950
    },
    {
      "entropy": 5.985467863082886,
      "epoch": 2.967316263217688,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00041259758365989896,
      "loss": 5.5832,
      "mean_token_accuracy": 0.18294304609298706,
      "num_tokens": 68806364.0,
      "step": 33955
    },
    {
      "entropy": 6.035828256607056,
      "epoch": 2.9677532115703924,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004125729273744589,
      "loss": 5.6418,
      "mean_token_accuracy": 0.17198735028505324,
      "num_tokens": 68816099.0,
      "step": 33960
    },
    {
      "entropy": 6.011405944824219,
      "epoch": 2.968190159923097,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041254826845029806,
      "loss": 5.6603,
      "mean_token_accuracy": 0.173758402466774,
      "num_tokens": 68826480.0,
      "step": 33965
    },
    {
      "entropy": 6.033042287826538,
      "epoch": 2.968627108275802,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041252360688788937,
      "loss": 5.5715,
      "mean_token_accuracy": 0.18364618271589278,
      "num_tokens": 68835657.0,
      "step": 33970
    },
    {
      "entropy": 6.083949422836303,
      "epoch": 2.9690640566285067,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004124989426877058,
      "loss": 5.669,
      "mean_token_accuracy": 0.17396994531154633,
      "num_tokens": 68845115.0,
      "step": 33975
    },
    {
      "entropy": 6.0031815528869625,
      "epoch": 2.969501004981211,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004124742758502205,
      "loss": 5.6246,
      "mean_token_accuracy": 0.1793006107211113,
      "num_tokens": 68854562.0,
      "step": 33980
    },
    {
      "entropy": 6.074796676635742,
      "epoch": 2.969937953333916,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004124496063759066,
      "loss": 5.6851,
      "mean_token_accuracy": 0.1744104415178299,
      "num_tokens": 68864787.0,
      "step": 33985
    },
    {
      "entropy": 5.994910001754761,
      "epoch": 2.9703749016866206,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004124249342652371,
      "loss": 5.5934,
      "mean_token_accuracy": 0.17740283012390137,
      "num_tokens": 68874518.0,
      "step": 33990
    },
    {
      "entropy": 6.0602082252502445,
      "epoch": 2.9708118500393255,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041240025951868545,
      "loss": 5.5738,
      "mean_token_accuracy": 0.16913546919822692,
      "num_tokens": 68883765.0,
      "step": 33995
    },
    {
      "entropy": 6.060413408279419,
      "epoch": 2.97124879839203,
      "grad_norm": 1.890625,
      "learning_rate": 0.00041237558213672477,
      "loss": 5.6558,
      "mean_token_accuracy": 0.17794110476970673,
      "num_tokens": 68894191.0,
      "step": 34000
    },
    {
      "entropy": 6.052305889129639,
      "epoch": 2.971685746744735,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00041235090211982846,
      "loss": 5.6165,
      "mean_token_accuracy": 0.1760520949959755,
      "num_tokens": 68905136.0,
      "step": 34005
    },
    {
      "entropy": 6.023360347747802,
      "epoch": 2.9721226950974398,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004123262194684699,
      "loss": 5.6099,
      "mean_token_accuracy": 0.1774646148085594,
      "num_tokens": 68914620.0,
      "step": 34010
    },
    {
      "entropy": 6.00660719871521,
      "epoch": 2.9725596434501442,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041230153418312244,
      "loss": 5.6168,
      "mean_token_accuracy": 0.17525864839553834,
      "num_tokens": 68923781.0,
      "step": 34015
    },
    {
      "entropy": 5.972195672988891,
      "epoch": 2.9729965918028487,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004122768462642596,
      "loss": 5.6737,
      "mean_token_accuracy": 0.17562351524829864,
      "num_tokens": 68934283.0,
      "step": 34020
    },
    {
      "entropy": 6.003650808334351,
      "epoch": 2.9734335401555536,
      "grad_norm": 2.484375,
      "learning_rate": 0.000412252155712355,
      "loss": 5.5834,
      "mean_token_accuracy": 0.18044463992118837,
      "num_tokens": 68943799.0,
      "step": 34025
    },
    {
      "entropy": 6.059888553619385,
      "epoch": 2.9738704885082585,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041222746252788213,
      "loss": 5.6521,
      "mean_token_accuracy": 0.1761580690741539,
      "num_tokens": 68953733.0,
      "step": 34030
    },
    {
      "entropy": 6.003673362731933,
      "epoch": 2.974307436860963,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004122027667113146,
      "loss": 5.6128,
      "mean_token_accuracy": 0.17786848098039626,
      "num_tokens": 68963888.0,
      "step": 34035
    },
    {
      "entropy": 6.101166391372681,
      "epoch": 2.974744385213668,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000412178068263126,
      "loss": 5.7105,
      "mean_token_accuracy": 0.17089378088712692,
      "num_tokens": 68974652.0,
      "step": 34040
    },
    {
      "entropy": 6.0886047840118405,
      "epoch": 2.9751813335663724,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004121533671837903,
      "loss": 5.579,
      "mean_token_accuracy": 0.17601008862257003,
      "num_tokens": 68984583.0,
      "step": 34045
    },
    {
      "entropy": 5.938350963592529,
      "epoch": 2.9756182819190773,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004121286634737811,
      "loss": 5.6874,
      "mean_token_accuracy": 0.16847685724496841,
      "num_tokens": 68995790.0,
      "step": 34050
    },
    {
      "entropy": 6.0816388607025145,
      "epoch": 2.9760552302717818,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004121039571335723,
      "loss": 5.6557,
      "mean_token_accuracy": 0.17275261878967285,
      "num_tokens": 69006695.0,
      "step": 34055
    },
    {
      "entropy": 6.025640535354614,
      "epoch": 2.9764921786244867,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004120792481636378,
      "loss": 5.5792,
      "mean_token_accuracy": 0.18415750563144684,
      "num_tokens": 69017504.0,
      "step": 34060
    },
    {
      "entropy": 6.038791370391846,
      "epoch": 2.976929126977191,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004120545365644515,
      "loss": 5.7234,
      "mean_token_accuracy": 0.16639079600572587,
      "num_tokens": 69026938.0,
      "step": 34065
    },
    {
      "entropy": 6.056998300552368,
      "epoch": 2.977366075329896,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004120298223364872,
      "loss": 5.5999,
      "mean_token_accuracy": 0.1842193067073822,
      "num_tokens": 69037404.0,
      "step": 34070
    },
    {
      "entropy": 6.033103132247925,
      "epoch": 2.9778030236826005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041200510548021935,
      "loss": 5.6052,
      "mean_token_accuracy": 0.18048035055398942,
      "num_tokens": 69047507.0,
      "step": 34075
    },
    {
      "entropy": 5.9982747554779055,
      "epoch": 2.9782399720353054,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041198038599612156,
      "loss": 5.6281,
      "mean_token_accuracy": 0.17524972409009934,
      "num_tokens": 69057626.0,
      "step": 34080
    },
    {
      "entropy": 6.026861906051636,
      "epoch": 2.9786769203880104,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041195566388466836,
      "loss": 5.6371,
      "mean_token_accuracy": 0.17771223485469817,
      "num_tokens": 69068142.0,
      "step": 34085
    },
    {
      "entropy": 6.108320236206055,
      "epoch": 2.979113868740715,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041193093914633364,
      "loss": 5.6607,
      "mean_token_accuracy": 0.17337738573551179,
      "num_tokens": 69077935.0,
      "step": 34090
    },
    {
      "entropy": 6.055626249313354,
      "epoch": 2.9795508170934193,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004119062117815918,
      "loss": 5.6884,
      "mean_token_accuracy": 0.16691234707832336,
      "num_tokens": 69087397.0,
      "step": 34095
    },
    {
      "entropy": 6.068542051315307,
      "epoch": 2.979987765446124,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004118814817909171,
      "loss": 5.7065,
      "mean_token_accuracy": 0.17422695457935333,
      "num_tokens": 69097666.0,
      "step": 34100
    },
    {
      "entropy": 6.058681774139404,
      "epoch": 2.980424713798829,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004118567491747839,
      "loss": 5.6392,
      "mean_token_accuracy": 0.1750301882624626,
      "num_tokens": 69107516.0,
      "step": 34105
    },
    {
      "entropy": 6.062700319290161,
      "epoch": 2.9808616621515336,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004118320139336665,
      "loss": 5.6208,
      "mean_token_accuracy": 0.17223456352949143,
      "num_tokens": 69117462.0,
      "step": 34110
    },
    {
      "entropy": 6.040162992477417,
      "epoch": 2.9812986105042385,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004118072760680393,
      "loss": 5.5904,
      "mean_token_accuracy": 0.18265196233987807,
      "num_tokens": 69128171.0,
      "step": 34115
    },
    {
      "entropy": 5.943731641769409,
      "epoch": 2.981735558856943,
      "grad_norm": 1.453125,
      "learning_rate": 0.00041178253557837697,
      "loss": 5.6305,
      "mean_token_accuracy": 0.17742612659931184,
      "num_tokens": 69138169.0,
      "step": 34120
    },
    {
      "entropy": 6.005683326721192,
      "epoch": 2.982172507209648,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041175779246515384,
      "loss": 5.5768,
      "mean_token_accuracy": 0.18199165612459184,
      "num_tokens": 69147518.0,
      "step": 34125
    },
    {
      "entropy": 5.999617195129394,
      "epoch": 2.9826094555623524,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004117330467288447,
      "loss": 5.6076,
      "mean_token_accuracy": 0.17475373148918152,
      "num_tokens": 69157049.0,
      "step": 34130
    },
    {
      "entropy": 6.005071783065796,
      "epoch": 2.9830464039150573,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041170829836992403,
      "loss": 5.6078,
      "mean_token_accuracy": 0.17864982187747955,
      "num_tokens": 69166502.0,
      "step": 34135
    },
    {
      "entropy": 6.061207580566406,
      "epoch": 2.983483352267762,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004116835473888665,
      "loss": 5.5889,
      "mean_token_accuracy": 0.17917933613061904,
      "num_tokens": 69176566.0,
      "step": 34140
    },
    {
      "entropy": 6.009816026687622,
      "epoch": 2.9839203006204666,
      "grad_norm": 1.125,
      "learning_rate": 0.0004116587937861469,
      "loss": 5.5992,
      "mean_token_accuracy": 0.1743439257144928,
      "num_tokens": 69186677.0,
      "step": 34145
    },
    {
      "entropy": 6.02970404624939,
      "epoch": 2.984357248973171,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004116340375622402,
      "loss": 5.69,
      "mean_token_accuracy": 0.17393888235092164,
      "num_tokens": 69197595.0,
      "step": 34150
    },
    {
      "entropy": 6.010616159439087,
      "epoch": 2.984794197325876,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004116092787176209,
      "loss": 5.6338,
      "mean_token_accuracy": 0.18173471987247466,
      "num_tokens": 69207384.0,
      "step": 34155
    },
    {
      "entropy": 6.017258262634277,
      "epoch": 2.985231145678581,
      "grad_norm": 1.375,
      "learning_rate": 0.00041158451725276405,
      "loss": 5.5325,
      "mean_token_accuracy": 0.1891583576798439,
      "num_tokens": 69217019.0,
      "step": 34160
    },
    {
      "entropy": 6.103298044204712,
      "epoch": 2.9856680940312854,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004115597531681447,
      "loss": 5.814,
      "mean_token_accuracy": 0.15869167000055312,
      "num_tokens": 69229076.0,
      "step": 34165
    },
    {
      "entropy": 6.10277681350708,
      "epoch": 2.9861050423839903,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041153498646423763,
      "loss": 5.7399,
      "mean_token_accuracy": 0.16564396172761917,
      "num_tokens": 69238771.0,
      "step": 34170
    },
    {
      "entropy": 6.067801094055175,
      "epoch": 2.986541990736695,
      "grad_norm": 1.1875,
      "learning_rate": 0.000411510217141518,
      "loss": 5.6311,
      "mean_token_accuracy": 0.17325173318386078,
      "num_tokens": 69248864.0,
      "step": 34175
    },
    {
      "entropy": 6.002050971984863,
      "epoch": 2.9869789390893997,
      "grad_norm": 1.484375,
      "learning_rate": 0.00041148544520046087,
      "loss": 5.7,
      "mean_token_accuracy": 0.1727786272764206,
      "num_tokens": 69259372.0,
      "step": 34180
    },
    {
      "entropy": 6.06704363822937,
      "epoch": 2.987415887442104,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004114606706415415,
      "loss": 5.5885,
      "mean_token_accuracy": 0.17908452749252318,
      "num_tokens": 69270079.0,
      "step": 34185
    },
    {
      "entropy": 6.101355171203613,
      "epoch": 2.987852835794809,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004114358934652349,
      "loss": 5.6324,
      "mean_token_accuracy": 0.17589557617902757,
      "num_tokens": 69279909.0,
      "step": 34190
    },
    {
      "entropy": 5.98056583404541,
      "epoch": 2.988289784147514,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004114111136720163,
      "loss": 5.5638,
      "mean_token_accuracy": 0.1818606287240982,
      "num_tokens": 69289598.0,
      "step": 34195
    },
    {
      "entropy": 6.087423658370971,
      "epoch": 2.9887267325002185,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004113863312623611,
      "loss": 5.6439,
      "mean_token_accuracy": 0.17507508397102356,
      "num_tokens": 69301305.0,
      "step": 34200
    },
    {
      "entropy": 6.012872838973999,
      "epoch": 2.989163680852923,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004113615462367446,
      "loss": 5.6018,
      "mean_token_accuracy": 0.17758129835128783,
      "num_tokens": 69310560.0,
      "step": 34205
    },
    {
      "entropy": 5.957680416107178,
      "epoch": 2.989600629205628,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041133675859564213,
      "loss": 5.6563,
      "mean_token_accuracy": 0.17544558942317962,
      "num_tokens": 69320286.0,
      "step": 34210
    },
    {
      "entropy": 5.984813547134399,
      "epoch": 2.9900375775583328,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041131196833952925,
      "loss": 5.5698,
      "mean_token_accuracy": 0.17566167265176774,
      "num_tokens": 69329895.0,
      "step": 34215
    },
    {
      "entropy": 6.10569200515747,
      "epoch": 2.9904745259110372,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004112871754688814,
      "loss": 5.5945,
      "mean_token_accuracy": 0.1731451079249382,
      "num_tokens": 69340122.0,
      "step": 34220
    },
    {
      "entropy": 6.034338712692261,
      "epoch": 2.990911474263742,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041126237998417414,
      "loss": 5.6107,
      "mean_token_accuracy": 0.17952706217765807,
      "num_tokens": 69349354.0,
      "step": 34225
    },
    {
      "entropy": 5.992438411712646,
      "epoch": 2.9913484226164466,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041123758188588296,
      "loss": 5.7118,
      "mean_token_accuracy": 0.16692153811454774,
      "num_tokens": 69360236.0,
      "step": 34230
    },
    {
      "entropy": 6.103920745849609,
      "epoch": 2.9917853709691515,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041121278117448365,
      "loss": 5.6574,
      "mean_token_accuracy": 0.1772230312228203,
      "num_tokens": 69369362.0,
      "step": 34235
    },
    {
      "entropy": 6.050707149505615,
      "epoch": 2.992222319321856,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00041118797785045174,
      "loss": 5.6407,
      "mean_token_accuracy": 0.17607313245534897,
      "num_tokens": 69379403.0,
      "step": 34240
    },
    {
      "entropy": 5.9888063907623295,
      "epoch": 2.992659267674561,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004111631719142631,
      "loss": 5.6209,
      "mean_token_accuracy": 0.18389790654182434,
      "num_tokens": 69388667.0,
      "step": 34245
    },
    {
      "entropy": 6.02210841178894,
      "epoch": 2.993096216027266,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041113836336639347,
      "loss": 5.6146,
      "mean_token_accuracy": 0.17540943771600723,
      "num_tokens": 69399552.0,
      "step": 34250
    },
    {
      "entropy": 6.036433362960816,
      "epoch": 2.9935331643799703,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004111135522073187,
      "loss": 5.5311,
      "mean_token_accuracy": 0.18175918608903885,
      "num_tokens": 69409323.0,
      "step": 34255
    },
    {
      "entropy": 6.055034875869751,
      "epoch": 2.9939701127326748,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004110887384375147,
      "loss": 5.6412,
      "mean_token_accuracy": 0.17046479880809784,
      "num_tokens": 69418696.0,
      "step": 34260
    },
    {
      "entropy": 6.088893795013428,
      "epoch": 2.9944070610853797,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004110639220574573,
      "loss": 5.7431,
      "mean_token_accuracy": 0.16544662714004515,
      "num_tokens": 69429470.0,
      "step": 34265
    },
    {
      "entropy": 6.002839660644531,
      "epoch": 2.9948440094380846,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041103910306762267,
      "loss": 5.5789,
      "mean_token_accuracy": 0.18127548098564147,
      "num_tokens": 69440636.0,
      "step": 34270
    },
    {
      "entropy": 6.037485980987549,
      "epoch": 2.995280957790789,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00041101428146848667,
      "loss": 5.6643,
      "mean_token_accuracy": 0.17422042340040206,
      "num_tokens": 69451819.0,
      "step": 34275
    },
    {
      "entropy": 6.075433015823364,
      "epoch": 2.995717906143494,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041098945726052556,
      "loss": 5.7271,
      "mean_token_accuracy": 0.17119294553995132,
      "num_tokens": 69462690.0,
      "step": 34280
    },
    {
      "entropy": 6.003259134292603,
      "epoch": 2.9961548544961984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004109646304442154,
      "loss": 5.5713,
      "mean_token_accuracy": 0.1821837082505226,
      "num_tokens": 69473338.0,
      "step": 34285
    },
    {
      "entropy": 5.993269395828247,
      "epoch": 2.9965918028489034,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004109398010200323,
      "loss": 5.6046,
      "mean_token_accuracy": 0.17361137866973878,
      "num_tokens": 69482189.0,
      "step": 34290
    },
    {
      "entropy": 5.922343921661377,
      "epoch": 2.997028751201608,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004109149689884527,
      "loss": 5.5641,
      "mean_token_accuracy": 0.18569830507040025,
      "num_tokens": 69492084.0,
      "step": 34295
    },
    {
      "entropy": 6.048063039779663,
      "epoch": 2.9974656995543127,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041089013434995273,
      "loss": 5.6158,
      "mean_token_accuracy": 0.1759561374783516,
      "num_tokens": 69502886.0,
      "step": 34300
    },
    {
      "entropy": 6.055998516082764,
      "epoch": 2.9979026479070177,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004108652971050088,
      "loss": 5.5617,
      "mean_token_accuracy": 0.1798476070165634,
      "num_tokens": 69512670.0,
      "step": 34305
    },
    {
      "entropy": 5.973301744461059,
      "epoch": 2.998339596259722,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041084045725409726,
      "loss": 5.6493,
      "mean_token_accuracy": 0.1765329048037529,
      "num_tokens": 69522744.0,
      "step": 34310
    },
    {
      "entropy": 5.969941234588623,
      "epoch": 2.9987765446124266,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004108156147976945,
      "loss": 5.5227,
      "mean_token_accuracy": 0.1805964082479477,
      "num_tokens": 69533195.0,
      "step": 34315
    },
    {
      "entropy": 6.006800413131714,
      "epoch": 2.9992134929651315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004107907697362772,
      "loss": 5.6774,
      "mean_token_accuracy": 0.17305036038160324,
      "num_tokens": 69543058.0,
      "step": 34320
    },
    {
      "entropy": 6.096932029724121,
      "epoch": 2.9996504413178364,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004107659220703217,
      "loss": 5.6394,
      "mean_token_accuracy": 0.1777610182762146,
      "num_tokens": 69552428.0,
      "step": 34325
    },
    {
      "entropy": 6.041003942489624,
      "epoch": 3.000087389670541,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004107410718003047,
      "loss": 5.6138,
      "mean_token_accuracy": 0.1777206227183342,
      "num_tokens": 69561735.0,
      "step": 34330
    },
    {
      "entropy": 5.999575233459472,
      "epoch": 3.000524338023246,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004107162189267028,
      "loss": 5.5972,
      "mean_token_accuracy": 0.17347080558538436,
      "num_tokens": 69573165.0,
      "step": 34335
    },
    {
      "entropy": 6.017105197906494,
      "epoch": 3.0009612863759503,
      "grad_norm": 1.53125,
      "learning_rate": 0.00041069136344999273,
      "loss": 5.5566,
      "mean_token_accuracy": 0.1826242685317993,
      "num_tokens": 69583484.0,
      "step": 34340
    },
    {
      "entropy": 5.985267686843872,
      "epoch": 3.001398234728655,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004106665053706512,
      "loss": 5.5883,
      "mean_token_accuracy": 0.17854418605566025,
      "num_tokens": 69593486.0,
      "step": 34345
    },
    {
      "entropy": 6.086971950531006,
      "epoch": 3.0018351830813597,
      "grad_norm": 1.15625,
      "learning_rate": 0.000410641644689155,
      "loss": 5.6849,
      "mean_token_accuracy": 0.17718938887119293,
      "num_tokens": 69604245.0,
      "step": 34350
    },
    {
      "entropy": 6.066306686401367,
      "epoch": 3.0022721314340646,
      "grad_norm": 1.34375,
      "learning_rate": 0.000410616781405981,
      "loss": 5.5687,
      "mean_token_accuracy": 0.1797807976603508,
      "num_tokens": 69614237.0,
      "step": 34355
    },
    {
      "entropy": 6.021577072143555,
      "epoch": 3.002709079786769,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004105919155216061,
      "loss": 5.602,
      "mean_token_accuracy": 0.17868556678295136,
      "num_tokens": 69623695.0,
      "step": 34360
    },
    {
      "entropy": 5.98225769996643,
      "epoch": 3.003146028139474,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041056704703650715,
      "loss": 5.5076,
      "mean_token_accuracy": 0.18588008284568786,
      "num_tokens": 69633299.0,
      "step": 34365
    },
    {
      "entropy": 5.987063264846801,
      "epoch": 3.0035829764921784,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004105421759511612,
      "loss": 5.55,
      "mean_token_accuracy": 0.1767369583249092,
      "num_tokens": 69642853.0,
      "step": 34370
    },
    {
      "entropy": 6.018863105773926,
      "epoch": 3.0040199248448833,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004105173022660453,
      "loss": 5.5884,
      "mean_token_accuracy": 0.17708974927663804,
      "num_tokens": 69652697.0,
      "step": 34375
    },
    {
      "entropy": 6.045815467834473,
      "epoch": 3.0044568731975883,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004104924259816366,
      "loss": 5.6436,
      "mean_token_accuracy": 0.18195427805185319,
      "num_tokens": 69664261.0,
      "step": 34380
    },
    {
      "entropy": 6.002862024307251,
      "epoch": 3.0048938215502927,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004104675470984121,
      "loss": 5.6103,
      "mean_token_accuracy": 0.18574083596467972,
      "num_tokens": 69674819.0,
      "step": 34385
    },
    {
      "entropy": 5.97865252494812,
      "epoch": 3.0053307699029976,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041044266561684914,
      "loss": 5.5729,
      "mean_token_accuracy": 0.17834870666265487,
      "num_tokens": 69685546.0,
      "step": 34390
    },
    {
      "entropy": 5.971510982513427,
      "epoch": 3.005767718255702,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004104177815374248,
      "loss": 5.4968,
      "mean_token_accuracy": 0.18592198342084884,
      "num_tokens": 69695192.0,
      "step": 34395
    },
    {
      "entropy": 6.0017365455627445,
      "epoch": 3.006204666608407,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004103928948606164,
      "loss": 5.5378,
      "mean_token_accuracy": 0.18223846405744554,
      "num_tokens": 69705669.0,
      "step": 34400
    },
    {
      "entropy": 5.961722040176392,
      "epoch": 3.0066416149611115,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004103680055869015,
      "loss": 5.5186,
      "mean_token_accuracy": 0.19233435094356538,
      "num_tokens": 69716274.0,
      "step": 34405
    },
    {
      "entropy": 6.022051572799683,
      "epoch": 3.0070785633138164,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004103431137167572,
      "loss": 5.596,
      "mean_token_accuracy": 0.17904244214296341,
      "num_tokens": 69726438.0,
      "step": 34410
    },
    {
      "entropy": 5.978540372848511,
      "epoch": 3.007515511666521,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041031821925066115,
      "loss": 5.5878,
      "mean_token_accuracy": 0.1789975419640541,
      "num_tokens": 69735467.0,
      "step": 34415
    },
    {
      "entropy": 6.033618211746216,
      "epoch": 3.007952460019226,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041029332218909067,
      "loss": 5.5831,
      "mean_token_accuracy": 0.17822141498327254,
      "num_tokens": 69744886.0,
      "step": 34420
    },
    {
      "entropy": 5.9874529361724855,
      "epoch": 3.0083894083719303,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041026842253252333,
      "loss": 5.58,
      "mean_token_accuracy": 0.1773035630583763,
      "num_tokens": 69754337.0,
      "step": 34425
    },
    {
      "entropy": 5.942834186553955,
      "epoch": 3.008826356724635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004102435202814369,
      "loss": 5.5762,
      "mean_token_accuracy": 0.1810708612203598,
      "num_tokens": 69764660.0,
      "step": 34430
    },
    {
      "entropy": 5.97731466293335,
      "epoch": 3.00926330507734,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004102186154363088,
      "loss": 5.5436,
      "mean_token_accuracy": 0.18099396377801896,
      "num_tokens": 69774339.0,
      "step": 34435
    },
    {
      "entropy": 6.039388132095337,
      "epoch": 3.0097002534300445,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041019370799761685,
      "loss": 5.643,
      "mean_token_accuracy": 0.17715198397636414,
      "num_tokens": 69785272.0,
      "step": 34440
    },
    {
      "entropy": 5.992170572280884,
      "epoch": 3.0101372017827495,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004101687979658386,
      "loss": 5.6792,
      "mean_token_accuracy": 0.17407356202602386,
      "num_tokens": 69796646.0,
      "step": 34445
    },
    {
      "entropy": 5.975277471542358,
      "epoch": 3.010574150135454,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041014388534145213,
      "loss": 5.5178,
      "mean_token_accuracy": 0.18796104788780213,
      "num_tokens": 69807233.0,
      "step": 34450
    },
    {
      "entropy": 6.059723424911499,
      "epoch": 3.011011098488159,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004101189701249351,
      "loss": 5.6107,
      "mean_token_accuracy": 0.17597500681877137,
      "num_tokens": 69818131.0,
      "step": 34455
    },
    {
      "entropy": 6.085000944137573,
      "epoch": 3.0114480468408633,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004100940523167654,
      "loss": 5.6827,
      "mean_token_accuracy": 0.1715848669409752,
      "num_tokens": 69829537.0,
      "step": 34460
    },
    {
      "entropy": 6.022730350494385,
      "epoch": 3.0118849951935682,
      "grad_norm": 1.15625,
      "learning_rate": 0.000410069131917421,
      "loss": 5.6346,
      "mean_token_accuracy": 0.17044018357992172,
      "num_tokens": 69840964.0,
      "step": 34465
    },
    {
      "entropy": 5.983765172958374,
      "epoch": 3.0123219435462727,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041004420892737993,
      "loss": 5.5116,
      "mean_token_accuracy": 0.18149976581335067,
      "num_tokens": 69849919.0,
      "step": 34470
    },
    {
      "entropy": 5.959059762954712,
      "epoch": 3.0127588918989776,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004100192833471201,
      "loss": 5.6064,
      "mean_token_accuracy": 0.18140823394060135,
      "num_tokens": 69859832.0,
      "step": 34475
    },
    {
      "entropy": 6.001055192947388,
      "epoch": 3.013195840251682,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004099943551771196,
      "loss": 5.5234,
      "mean_token_accuracy": 0.17929122745990753,
      "num_tokens": 69869382.0,
      "step": 34480
    },
    {
      "entropy": 6.031747627258301,
      "epoch": 3.013632788604387,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004099694244178568,
      "loss": 5.5554,
      "mean_token_accuracy": 0.181969578564167,
      "num_tokens": 69879282.0,
      "step": 34485
    },
    {
      "entropy": 5.970047235488892,
      "epoch": 3.0140697369570915,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004099444910698096,
      "loss": 5.4981,
      "mean_token_accuracy": 0.19042607843875886,
      "num_tokens": 69889876.0,
      "step": 34490
    },
    {
      "entropy": 6.031424140930175,
      "epoch": 3.0145066853097964,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004099195551334564,
      "loss": 5.6054,
      "mean_token_accuracy": 0.17309515327215194,
      "num_tokens": 69899642.0,
      "step": 34495
    },
    {
      "entropy": 6.007369947433472,
      "epoch": 3.0149436336625013,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00040989461660927536,
      "loss": 5.527,
      "mean_token_accuracy": 0.18264400064945222,
      "num_tokens": 69909877.0,
      "step": 34500
    },
    {
      "entropy": 5.9998448371887205,
      "epoch": 3.0153805820152058,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040986967549774497,
      "loss": 5.5189,
      "mean_token_accuracy": 0.1869829624891281,
      "num_tokens": 69920061.0,
      "step": 34505
    },
    {
      "entropy": 5.924544811248779,
      "epoch": 3.0158175303679107,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004098447317993435,
      "loss": 5.5548,
      "mean_token_accuracy": 0.18452685624361037,
      "num_tokens": 69930739.0,
      "step": 34510
    },
    {
      "entropy": 5.9216938495635985,
      "epoch": 3.016254478720615,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040981978551454947,
      "loss": 5.4775,
      "mean_token_accuracy": 0.18670547902584075,
      "num_tokens": 69939825.0,
      "step": 34515
    },
    {
      "entropy": 6.076263093948365,
      "epoch": 3.01669142707332,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040979483664384123,
      "loss": 5.7444,
      "mean_token_accuracy": 0.16284918040037155,
      "num_tokens": 69949662.0,
      "step": 34520
    },
    {
      "entropy": 6.015770530700683,
      "epoch": 3.0171283754260245,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040976988518769743,
      "loss": 5.5892,
      "mean_token_accuracy": 0.17995876222848892,
      "num_tokens": 69959950.0,
      "step": 34525
    },
    {
      "entropy": 6.013490629196167,
      "epoch": 3.0175653237787294,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040974493114659654,
      "loss": 5.6241,
      "mean_token_accuracy": 0.1744014784693718,
      "num_tokens": 69970451.0,
      "step": 34530
    },
    {
      "entropy": 6.015825843811035,
      "epoch": 3.018002272131434,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004097199745210174,
      "loss": 5.5434,
      "mean_token_accuracy": 0.18257719576358794,
      "num_tokens": 69979940.0,
      "step": 34535
    },
    {
      "entropy": 6.03703203201294,
      "epoch": 3.018439220484139,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040969501531143844,
      "loss": 5.5362,
      "mean_token_accuracy": 0.18633184731006622,
      "num_tokens": 69989507.0,
      "step": 34540
    },
    {
      "entropy": 6.035801124572754,
      "epoch": 3.0188761688368433,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040967005351833863,
      "loss": 5.5728,
      "mean_token_accuracy": 0.1832582801580429,
      "num_tokens": 69999184.0,
      "step": 34545
    },
    {
      "entropy": 6.080097961425781,
      "epoch": 3.019313117189548,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004096450891421965,
      "loss": 5.6514,
      "mean_token_accuracy": 0.17043739110231398,
      "num_tokens": 70009569.0,
      "step": 34550
    },
    {
      "entropy": 6.038575124740601,
      "epoch": 3.019750065542253,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00040962012218349105,
      "loss": 5.5887,
      "mean_token_accuracy": 0.17375398427248,
      "num_tokens": 70019824.0,
      "step": 34555
    },
    {
      "entropy": 5.9379603385925295,
      "epoch": 3.0201870138949576,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004095951526427011,
      "loss": 5.4657,
      "mean_token_accuracy": 0.18627456873655318,
      "num_tokens": 70029043.0,
      "step": 34560
    },
    {
      "entropy": 5.98900203704834,
      "epoch": 3.0206239622476625,
      "grad_norm": 2.0,
      "learning_rate": 0.0004095701805203056,
      "loss": 5.6053,
      "mean_token_accuracy": 0.17522365748882293,
      "num_tokens": 70039794.0,
      "step": 34565
    },
    {
      "entropy": 6.050228118896484,
      "epoch": 3.021060910600367,
      "grad_norm": 1.25,
      "learning_rate": 0.0004095452058167837,
      "loss": 5.6232,
      "mean_token_accuracy": 0.17770951986312866,
      "num_tokens": 70049866.0,
      "step": 34570
    },
    {
      "entropy": 6.016648769378662,
      "epoch": 3.021497858953072,
      "grad_norm": 1.140625,
      "learning_rate": 0.000409520228532614,
      "loss": 5.5249,
      "mean_token_accuracy": 0.1875476434826851,
      "num_tokens": 70059471.0,
      "step": 34575
    },
    {
      "entropy": 5.980247068405151,
      "epoch": 3.0219348073057763,
      "grad_norm": 1.953125,
      "learning_rate": 0.000409495248668276,
      "loss": 5.546,
      "mean_token_accuracy": 0.18432123363018035,
      "num_tokens": 70069003.0,
      "step": 34580
    },
    {
      "entropy": 6.027529954910278,
      "epoch": 3.0223717556584813,
      "grad_norm": 1.25,
      "learning_rate": 0.00040947026622424853,
      "loss": 5.6031,
      "mean_token_accuracy": 0.18105224072933196,
      "num_tokens": 70078650.0,
      "step": 34585
    },
    {
      "entropy": 5.987028121948242,
      "epoch": 3.0228087040111857,
      "grad_norm": 2.0625,
      "learning_rate": 0.00040944528120101097,
      "loss": 5.555,
      "mean_token_accuracy": 0.18091438412666322,
      "num_tokens": 70087361.0,
      "step": 34590
    },
    {
      "entropy": 6.023150014877319,
      "epoch": 3.0232456523638906,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040942029359904245,
      "loss": 5.6075,
      "mean_token_accuracy": 0.18074328154325486,
      "num_tokens": 70097289.0,
      "step": 34595
    },
    {
      "entropy": 6.016015863418579,
      "epoch": 3.023682600716595,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040939530341882225,
      "loss": 5.5568,
      "mean_token_accuracy": 0.18268407732248307,
      "num_tokens": 70108099.0,
      "step": 34600
    },
    {
      "entropy": 6.028542423248291,
      "epoch": 3.0241195490693,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040937031066082974,
      "loss": 5.6329,
      "mean_token_accuracy": 0.17437736690044403,
      "num_tokens": 70119243.0,
      "step": 34605
    },
    {
      "entropy": 5.989737844467163,
      "epoch": 3.0245564974220045,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004093453153255442,
      "loss": 5.5694,
      "mean_token_accuracy": 0.1821526974439621,
      "num_tokens": 70129073.0,
      "step": 34610
    },
    {
      "entropy": 5.988790130615234,
      "epoch": 3.0249934457747094,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004093203174134452,
      "loss": 5.5036,
      "mean_token_accuracy": 0.1900139033794403,
      "num_tokens": 70139778.0,
      "step": 34615
    },
    {
      "entropy": 6.010918474197387,
      "epoch": 3.0254303941274143,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040929531692501206,
      "loss": 5.6765,
      "mean_token_accuracy": 0.17084366530179979,
      "num_tokens": 70151088.0,
      "step": 34620
    },
    {
      "entropy": 5.947905874252319,
      "epoch": 3.025867342480119,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00040927031386072435,
      "loss": 5.5418,
      "mean_token_accuracy": 0.19246915280818938,
      "num_tokens": 70160133.0,
      "step": 34625
    },
    {
      "entropy": 6.038924312591552,
      "epoch": 3.0263042908328237,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00040924530822106177,
      "loss": 5.5508,
      "mean_token_accuracy": 0.17675867229700087,
      "num_tokens": 70171176.0,
      "step": 34630
    },
    {
      "entropy": 6.019088792800903,
      "epoch": 3.026741239185528,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040922030000650376,
      "loss": 5.5888,
      "mean_token_accuracy": 0.17615578025579454,
      "num_tokens": 70181413.0,
      "step": 34635
    },
    {
      "entropy": 6.0128199577331545,
      "epoch": 3.027178187538233,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004091952892175301,
      "loss": 5.6,
      "mean_token_accuracy": 0.17395245432853698,
      "num_tokens": 70190535.0,
      "step": 34640
    },
    {
      "entropy": 5.997662210464478,
      "epoch": 3.0276151358909376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040917027585462044,
      "loss": 5.5484,
      "mean_token_accuracy": 0.18098742961883546,
      "num_tokens": 70199440.0,
      "step": 34645
    },
    {
      "entropy": 5.988158941268921,
      "epoch": 3.0280520842436425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004091452599182546,
      "loss": 5.5304,
      "mean_token_accuracy": 0.18284120708703994,
      "num_tokens": 70209846.0,
      "step": 34650
    },
    {
      "entropy": 6.011117887496948,
      "epoch": 3.028489032596347,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040912024140891235,
      "loss": 5.6092,
      "mean_token_accuracy": 0.1803586572408676,
      "num_tokens": 70219910.0,
      "step": 34655
    },
    {
      "entropy": 6.072006177902222,
      "epoch": 3.028925980949052,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004090952203270737,
      "loss": 5.5833,
      "mean_token_accuracy": 0.1771734654903412,
      "num_tokens": 70230937.0,
      "step": 34660
    },
    {
      "entropy": 6.001955223083496,
      "epoch": 3.0293629293017563,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040907019667321833,
      "loss": 5.6204,
      "mean_token_accuracy": 0.17752413898706437,
      "num_tokens": 70240791.0,
      "step": 34665
    },
    {
      "entropy": 6.0584405899047855,
      "epoch": 3.0297998776544612,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004090451704478264,
      "loss": 5.5777,
      "mean_token_accuracy": 0.18042204082012175,
      "num_tokens": 70251231.0,
      "step": 34670
    },
    {
      "entropy": 6.011781120300293,
      "epoch": 3.030236826007166,
      "grad_norm": 1.515625,
      "learning_rate": 0.00040902014165137787,
      "loss": 5.4964,
      "mean_token_accuracy": 0.18207320272922517,
      "num_tokens": 70260993.0,
      "step": 34675
    },
    {
      "entropy": 6.051492166519165,
      "epoch": 3.0306737743598706,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004089951102843528,
      "loss": 5.6877,
      "mean_token_accuracy": 0.17380574345588684,
      "num_tokens": 70271380.0,
      "step": 34680
    },
    {
      "entropy": 6.0602099895477295,
      "epoch": 3.0311107227125755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040897007634723125,
      "loss": 5.6159,
      "mean_token_accuracy": 0.1790269449353218,
      "num_tokens": 70282614.0,
      "step": 34685
    },
    {
      "entropy": 6.036565446853638,
      "epoch": 3.03154767106528,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004089450398404935,
      "loss": 5.575,
      "mean_token_accuracy": 0.1771117478609085,
      "num_tokens": 70292084.0,
      "step": 34690
    },
    {
      "entropy": 5.9164410591125485,
      "epoch": 3.031984619417985,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040892000076461966,
      "loss": 5.4715,
      "mean_token_accuracy": 0.18444754034280778,
      "num_tokens": 70302205.0,
      "step": 34695
    },
    {
      "entropy": 6.01539044380188,
      "epoch": 3.0324215677706894,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040889495912009,
      "loss": 5.6304,
      "mean_token_accuracy": 0.1788687914609909,
      "num_tokens": 70312264.0,
      "step": 34700
    },
    {
      "entropy": 6.099437284469604,
      "epoch": 3.0328585161233943,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040886991490738495,
      "loss": 5.5365,
      "mean_token_accuracy": 0.17904392927885054,
      "num_tokens": 70321733.0,
      "step": 34705
    },
    {
      "entropy": 5.973316669464111,
      "epoch": 3.0332954644760988,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040884486812698475,
      "loss": 5.6347,
      "mean_token_accuracy": 0.17508405297994614,
      "num_tokens": 70331621.0,
      "step": 34710
    },
    {
      "entropy": 5.9513791561126705,
      "epoch": 3.0337324128288037,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00040881981877936987,
      "loss": 5.597,
      "mean_token_accuracy": 0.17050814628601074,
      "num_tokens": 70342247.0,
      "step": 34715
    },
    {
      "entropy": 5.993042373657227,
      "epoch": 3.034169361181508,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004087947668650207,
      "loss": 5.5571,
      "mean_token_accuracy": 0.18115791380405427,
      "num_tokens": 70352423.0,
      "step": 34720
    },
    {
      "entropy": 6.067420721054077,
      "epoch": 3.034606309534213,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004087697123844178,
      "loss": 5.6405,
      "mean_token_accuracy": 0.1755082607269287,
      "num_tokens": 70362740.0,
      "step": 34725
    },
    {
      "entropy": 5.995874500274658,
      "epoch": 3.035043257886918,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004087446553380417,
      "loss": 5.5039,
      "mean_token_accuracy": 0.18768833130598067,
      "num_tokens": 70372456.0,
      "step": 34730
    },
    {
      "entropy": 5.962320613861084,
      "epoch": 3.0354802062396224,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004087195957263731,
      "loss": 5.5438,
      "mean_token_accuracy": 0.19016212075948716,
      "num_tokens": 70382741.0,
      "step": 34735
    },
    {
      "entropy": 5.993225812911987,
      "epoch": 3.0359171545923274,
      "grad_norm": 2.4375,
      "learning_rate": 0.00040869453354989255,
      "loss": 5.5456,
      "mean_token_accuracy": 0.18189185112714767,
      "num_tokens": 70392712.0,
      "step": 34740
    },
    {
      "entropy": 5.99072847366333,
      "epoch": 3.036354102945032,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040866946880908076,
      "loss": 5.6315,
      "mean_token_accuracy": 0.17863137125968934,
      "num_tokens": 70403655.0,
      "step": 34745
    },
    {
      "entropy": 6.051542282104492,
      "epoch": 3.0367910512977367,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004086444015044186,
      "loss": 5.6638,
      "mean_token_accuracy": 0.16623355001211165,
      "num_tokens": 70413720.0,
      "step": 34750
    },
    {
      "entropy": 6.051432943344116,
      "epoch": 3.037227999650441,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004086193316363867,
      "loss": 5.5912,
      "mean_token_accuracy": 0.17848280370235442,
      "num_tokens": 70423910.0,
      "step": 34755
    },
    {
      "entropy": 6.032374334335327,
      "epoch": 3.037664948003146,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040859425920546604,
      "loss": 5.6016,
      "mean_token_accuracy": 0.17925694286823274,
      "num_tokens": 70434593.0,
      "step": 34760
    },
    {
      "entropy": 6.002400064468384,
      "epoch": 3.0381018963558506,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004085691842121375,
      "loss": 5.5581,
      "mean_token_accuracy": 0.18089874237775802,
      "num_tokens": 70443967.0,
      "step": 34765
    },
    {
      "entropy": 6.04295334815979,
      "epoch": 3.0385388447085555,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000408544106656882,
      "loss": 5.5683,
      "mean_token_accuracy": 0.17681464776396752,
      "num_tokens": 70454098.0,
      "step": 34770
    },
    {
      "entropy": 5.977893114089966,
      "epoch": 3.03897579306126,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040851902654018057,
      "loss": 5.5597,
      "mean_token_accuracy": 0.18141748607158661,
      "num_tokens": 70463946.0,
      "step": 34775
    },
    {
      "entropy": 6.082613706588745,
      "epoch": 3.039412741413965,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040849394386251425,
      "loss": 5.606,
      "mean_token_accuracy": 0.18011785298585892,
      "num_tokens": 70474805.0,
      "step": 34780
    },
    {
      "entropy": 5.9655694484710695,
      "epoch": 3.03984968976667,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004084688586243641,
      "loss": 5.5274,
      "mean_token_accuracy": 0.18064242005348205,
      "num_tokens": 70484585.0,
      "step": 34785
    },
    {
      "entropy": 6.029680299758911,
      "epoch": 3.0402866381193743,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004084437708262113,
      "loss": 5.5786,
      "mean_token_accuracy": 0.1801335632801056,
      "num_tokens": 70494811.0,
      "step": 34790
    },
    {
      "entropy": 6.044291353225708,
      "epoch": 3.040723586472079,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004084186804685371,
      "loss": 5.6314,
      "mean_token_accuracy": 0.17438188791275025,
      "num_tokens": 70505392.0,
      "step": 34795
    },
    {
      "entropy": 5.997835540771485,
      "epoch": 3.0411605348247837,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004083935875518227,
      "loss": 5.535,
      "mean_token_accuracy": 0.18376736342906952,
      "num_tokens": 70515066.0,
      "step": 34800
    },
    {
      "entropy": 6.044647741317749,
      "epoch": 3.0415974831774886,
      "grad_norm": 1.25,
      "learning_rate": 0.00040836849207654937,
      "loss": 5.545,
      "mean_token_accuracy": 0.17683502584695815,
      "num_tokens": 70525100.0,
      "step": 34805
    },
    {
      "entropy": 5.970483589172363,
      "epoch": 3.042034431530193,
      "grad_norm": 1.125,
      "learning_rate": 0.0004083433940431985,
      "loss": 5.5655,
      "mean_token_accuracy": 0.18294728994369508,
      "num_tokens": 70535286.0,
      "step": 34810
    },
    {
      "entropy": 6.048352766036987,
      "epoch": 3.042471379882898,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004083182934522514,
      "loss": 5.6283,
      "mean_token_accuracy": 0.17767963707447051,
      "num_tokens": 70545724.0,
      "step": 34815
    },
    {
      "entropy": 6.018744421005249,
      "epoch": 3.0429083282356024,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040829319030418967,
      "loss": 5.5695,
      "mean_token_accuracy": 0.1801689699292183,
      "num_tokens": 70555179.0,
      "step": 34820
    },
    {
      "entropy": 6.050676012039185,
      "epoch": 3.0433452765883073,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004082680845994946,
      "loss": 5.6095,
      "mean_token_accuracy": 0.17790422290563584,
      "num_tokens": 70564355.0,
      "step": 34825
    },
    {
      "entropy": 5.922778701782226,
      "epoch": 3.043782224941012,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004082429763386479,
      "loss": 5.5132,
      "mean_token_accuracy": 0.19400384724140168,
      "num_tokens": 70575367.0,
      "step": 34830
    },
    {
      "entropy": 5.951603078842163,
      "epoch": 3.0442191732937167,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040821786552213107,
      "loss": 5.5438,
      "mean_token_accuracy": 0.184976789355278,
      "num_tokens": 70585047.0,
      "step": 34835
    },
    {
      "entropy": 6.069112396240234,
      "epoch": 3.044656121646421,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040819275215042586,
      "loss": 5.5819,
      "mean_token_accuracy": 0.17577446699142457,
      "num_tokens": 70594736.0,
      "step": 34840
    },
    {
      "entropy": 6.009849882125854,
      "epoch": 3.045093069999126,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040816763622401376,
      "loss": 5.5961,
      "mean_token_accuracy": 0.17788982540369033,
      "num_tokens": 70604762.0,
      "step": 34845
    },
    {
      "entropy": 6.035097551345825,
      "epoch": 3.045530018351831,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040814251774337666,
      "loss": 5.5915,
      "mean_token_accuracy": 0.18046973943710326,
      "num_tokens": 70614438.0,
      "step": 34850
    },
    {
      "entropy": 6.123410415649414,
      "epoch": 3.0459669667045355,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004081173967089963,
      "loss": 5.6999,
      "mean_token_accuracy": 0.1711602434515953,
      "num_tokens": 70624669.0,
      "step": 34855
    },
    {
      "entropy": 5.989370107650757,
      "epoch": 3.0464039150572404,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040809227312135456,
      "loss": 5.4983,
      "mean_token_accuracy": 0.18273769468069076,
      "num_tokens": 70635082.0,
      "step": 34860
    },
    {
      "entropy": 6.01068902015686,
      "epoch": 3.046840863409945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004080671469809332,
      "loss": 5.5606,
      "mean_token_accuracy": 0.18810865432024002,
      "num_tokens": 70644117.0,
      "step": 34865
    },
    {
      "entropy": 6.022942972183228,
      "epoch": 3.04727781176265,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004080420182882144,
      "loss": 5.6309,
      "mean_token_accuracy": 0.17368691116571428,
      "num_tokens": 70653299.0,
      "step": 34870
    },
    {
      "entropy": 6.059678268432617,
      "epoch": 3.0477147601153542,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040801688704367975,
      "loss": 5.5977,
      "mean_token_accuracy": 0.1774837389588356,
      "num_tokens": 70664477.0,
      "step": 34875
    },
    {
      "entropy": 6.0295356750488285,
      "epoch": 3.048151708468059,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040799175324781164,
      "loss": 5.5928,
      "mean_token_accuracy": 0.17744536995887755,
      "num_tokens": 70674635.0,
      "step": 34880
    },
    {
      "entropy": 5.948680973052978,
      "epoch": 3.0485886568207636,
      "grad_norm": 1.203125,
      "learning_rate": 0.000407966616901092,
      "loss": 5.5392,
      "mean_token_accuracy": 0.18108357042074202,
      "num_tokens": 70684633.0,
      "step": 34885
    },
    {
      "entropy": 6.042013931274414,
      "epoch": 3.0490256051734685,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004079414780040029,
      "loss": 5.6219,
      "mean_token_accuracy": 0.1829304128885269,
      "num_tokens": 70694416.0,
      "step": 34890
    },
    {
      "entropy": 6.026387119293213,
      "epoch": 3.049462553526173,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004079163365570267,
      "loss": 5.5339,
      "mean_token_accuracy": 0.1837512046098709,
      "num_tokens": 70703373.0,
      "step": 34895
    },
    {
      "entropy": 5.990605211257934,
      "epoch": 3.049899501878878,
      "grad_norm": 1.125,
      "learning_rate": 0.0004078911925606455,
      "loss": 5.5473,
      "mean_token_accuracy": 0.1769771993160248,
      "num_tokens": 70713868.0,
      "step": 34900
    },
    {
      "entropy": 6.084040880203247,
      "epoch": 3.050336450231583,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004078660460153415,
      "loss": 5.6932,
      "mean_token_accuracy": 0.16785066500306128,
      "num_tokens": 70724306.0,
      "step": 34905
    },
    {
      "entropy": 6.064218521118164,
      "epoch": 3.0507733985842873,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040784089692159724,
      "loss": 5.6213,
      "mean_token_accuracy": 0.1794444128870964,
      "num_tokens": 70733923.0,
      "step": 34910
    },
    {
      "entropy": 6.0738523483276365,
      "epoch": 3.0512103469369922,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004078157452798949,
      "loss": 5.6504,
      "mean_token_accuracy": 0.17096723169088363,
      "num_tokens": 70744964.0,
      "step": 34915
    },
    {
      "entropy": 5.963043880462647,
      "epoch": 3.0516472952896967,
      "grad_norm": 1.125,
      "learning_rate": 0.00040779059109071706,
      "loss": 5.501,
      "mean_token_accuracy": 0.18752632588148116,
      "num_tokens": 70754639.0,
      "step": 34920
    },
    {
      "entropy": 6.06464638710022,
      "epoch": 3.0520842436424016,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040776543435454597,
      "loss": 5.66,
      "mean_token_accuracy": 0.16895612254738807,
      "num_tokens": 70764810.0,
      "step": 34925
    },
    {
      "entropy": 6.069260025024414,
      "epoch": 3.052521191995106,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004077402750718643,
      "loss": 5.6037,
      "mean_token_accuracy": 0.1746211975812912,
      "num_tokens": 70774698.0,
      "step": 34930
    },
    {
      "entropy": 5.966358423233032,
      "epoch": 3.052958140347811,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004077151132431546,
      "loss": 5.5173,
      "mean_token_accuracy": 0.1875631555914879,
      "num_tokens": 70785522.0,
      "step": 34935
    },
    {
      "entropy": 6.0225691318511965,
      "epoch": 3.0533950887005155,
      "grad_norm": 2.125,
      "learning_rate": 0.0004076899488688996,
      "loss": 5.6324,
      "mean_token_accuracy": 0.17718748599290848,
      "num_tokens": 70797001.0,
      "step": 34940
    },
    {
      "entropy": 5.983132934570312,
      "epoch": 3.0538320370532204,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040766478194958176,
      "loss": 5.6233,
      "mean_token_accuracy": 0.18415720015764236,
      "num_tokens": 70807247.0,
      "step": 34945
    },
    {
      "entropy": 6.070856428146362,
      "epoch": 3.054268985405925,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004076396124856839,
      "loss": 5.6895,
      "mean_token_accuracy": 0.16647714525461196,
      "num_tokens": 70818269.0,
      "step": 34950
    },
    {
      "entropy": 6.084919500350952,
      "epoch": 3.0547059337586298,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004076144404776888,
      "loss": 5.653,
      "mean_token_accuracy": 0.1768209829926491,
      "num_tokens": 70827759.0,
      "step": 34955
    },
    {
      "entropy": 6.025166034698486,
      "epoch": 3.0551428821113342,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004075892659260792,
      "loss": 5.5938,
      "mean_token_accuracy": 0.17456659972667693,
      "num_tokens": 70838296.0,
      "step": 34960
    },
    {
      "entropy": 6.028687334060669,
      "epoch": 3.055579830464039,
      "grad_norm": 1.484375,
      "learning_rate": 0.00040756408883133797,
      "loss": 5.5649,
      "mean_token_accuracy": 0.1772652953863144,
      "num_tokens": 70848364.0,
      "step": 34965
    },
    {
      "entropy": 6.036456537246704,
      "epoch": 3.056016778816744,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004075389091939481,
      "loss": 5.5734,
      "mean_token_accuracy": 0.18475494980812074,
      "num_tokens": 70858845.0,
      "step": 34970
    },
    {
      "entropy": 5.943465089797973,
      "epoch": 3.0564537271694485,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004075137270143925,
      "loss": 5.659,
      "mean_token_accuracy": 0.18570100516080856,
      "num_tokens": 70868976.0,
      "step": 34975
    },
    {
      "entropy": 5.9691332340240475,
      "epoch": 3.0568906755221534,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004074885422931542,
      "loss": 5.5453,
      "mean_token_accuracy": 0.18940455317497254,
      "num_tokens": 70878287.0,
      "step": 34980
    },
    {
      "entropy": 6.02064061164856,
      "epoch": 3.057327623874858,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040746335503071625,
      "loss": 5.5562,
      "mean_token_accuracy": 0.18345831632614135,
      "num_tokens": 70889374.0,
      "step": 34985
    },
    {
      "entropy": 5.987636137008667,
      "epoch": 3.057764572227563,
      "grad_norm": 2.296875,
      "learning_rate": 0.00040743816522756165,
      "loss": 5.5852,
      "mean_token_accuracy": 0.18595873415470124,
      "num_tokens": 70899741.0,
      "step": 34990
    },
    {
      "entropy": 6.032076597213745,
      "epoch": 3.0582015205802673,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040741297288417373,
      "loss": 5.614,
      "mean_token_accuracy": 0.18289846032857895,
      "num_tokens": 70909962.0,
      "step": 34995
    },
    {
      "entropy": 6.058357667922974,
      "epoch": 3.058638468932972,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004073877780010356,
      "loss": 5.5839,
      "mean_token_accuracy": 0.17646660059690475,
      "num_tokens": 70921129.0,
      "step": 35000
    },
    {
      "entropy": 5.9830413341522215,
      "epoch": 3.0590754172856767,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004073625805786305,
      "loss": 5.556,
      "mean_token_accuracy": 0.18149320632219315,
      "num_tokens": 70930492.0,
      "step": 35005
    },
    {
      "entropy": 6.064654016494751,
      "epoch": 3.0595123656383816,
      "grad_norm": 1.125,
      "learning_rate": 0.0004073373806174417,
      "loss": 5.6059,
      "mean_token_accuracy": 0.17789264470338823,
      "num_tokens": 70940667.0,
      "step": 35010
    },
    {
      "entropy": 6.053144454956055,
      "epoch": 3.059949313991086,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004073121781179527,
      "loss": 5.5651,
      "mean_token_accuracy": 0.1815630868077278,
      "num_tokens": 70949815.0,
      "step": 35015
    },
    {
      "entropy": 6.041891813278198,
      "epoch": 3.060386262343791,
      "grad_norm": 1.4375,
      "learning_rate": 0.00040728697308064674,
      "loss": 5.6822,
      "mean_token_accuracy": 0.1678500786423683,
      "num_tokens": 70959931.0,
      "step": 35020
    },
    {
      "entropy": 6.022421264648438,
      "epoch": 3.060823210696496,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004072617655060073,
      "loss": 5.5248,
      "mean_token_accuracy": 0.1797905296087265,
      "num_tokens": 70970204.0,
      "step": 35025
    },
    {
      "entropy": 6.043280410766601,
      "epoch": 3.0612601590492003,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040723655539451787,
      "loss": 5.6696,
      "mean_token_accuracy": 0.17426451444625854,
      "num_tokens": 70981841.0,
      "step": 35030
    },
    {
      "entropy": 6.049639272689819,
      "epoch": 3.0616971074019053,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004072113427466621,
      "loss": 5.58,
      "mean_token_accuracy": 0.1766219988465309,
      "num_tokens": 70992281.0,
      "step": 35035
    },
    {
      "entropy": 5.941154193878174,
      "epoch": 3.0621340557546097,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040718612756292347,
      "loss": 5.5007,
      "mean_token_accuracy": 0.1896835073828697,
      "num_tokens": 71002737.0,
      "step": 35040
    },
    {
      "entropy": 6.007180213928223,
      "epoch": 3.0625710041073146,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004071609098437856,
      "loss": 5.6039,
      "mean_token_accuracy": 0.178285151720047,
      "num_tokens": 71013604.0,
      "step": 35045
    },
    {
      "entropy": 5.957886362075806,
      "epoch": 3.063007952460019,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040713568958973225,
      "loss": 5.5633,
      "mean_token_accuracy": 0.1803284376859665,
      "num_tokens": 71023278.0,
      "step": 35050
    },
    {
      "entropy": 5.958213758468628,
      "epoch": 3.063444900812724,
      "grad_norm": 1.71875,
      "learning_rate": 0.00040711046680124715,
      "loss": 5.6052,
      "mean_token_accuracy": 0.17934967428445817,
      "num_tokens": 71033823.0,
      "step": 35055
    },
    {
      "entropy": 6.033347034454346,
      "epoch": 3.0638818491654285,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004070852414788141,
      "loss": 5.6232,
      "mean_token_accuracy": 0.18049458116292955,
      "num_tokens": 71043802.0,
      "step": 35060
    },
    {
      "entropy": 6.074287986755371,
      "epoch": 3.0643187975181334,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004070600136229168,
      "loss": 5.6467,
      "mean_token_accuracy": 0.174909770488739,
      "num_tokens": 71054152.0,
      "step": 35065
    },
    {
      "entropy": 6.00726203918457,
      "epoch": 3.064755745870838,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040703478323403944,
      "loss": 5.5641,
      "mean_token_accuracy": 0.1803474172949791,
      "num_tokens": 71065346.0,
      "step": 35070
    },
    {
      "entropy": 6.033222293853759,
      "epoch": 3.065192694223543,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040700955031266556,
      "loss": 5.5831,
      "mean_token_accuracy": 0.18103684186935426,
      "num_tokens": 71074855.0,
      "step": 35075
    },
    {
      "entropy": 6.032961702346801,
      "epoch": 3.0656296425762477,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004069843148592794,
      "loss": 5.5738,
      "mean_token_accuracy": 0.18239859044551848,
      "num_tokens": 71084372.0,
      "step": 35080
    },
    {
      "entropy": 6.027803945541382,
      "epoch": 3.066066590928952,
      "grad_norm": 1.8125,
      "learning_rate": 0.000406959076874365,
      "loss": 5.5931,
      "mean_token_accuracy": 0.18481332212686538,
      "num_tokens": 71093365.0,
      "step": 35085
    },
    {
      "entropy": 5.938286685943604,
      "epoch": 3.066503539281657,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040693383635840625,
      "loss": 5.5208,
      "mean_token_accuracy": 0.18387491255998611,
      "num_tokens": 71102748.0,
      "step": 35090
    },
    {
      "entropy": 6.0088907241821286,
      "epoch": 3.0669404876343616,
      "grad_norm": 1.125,
      "learning_rate": 0.0004069085933118874,
      "loss": 5.6149,
      "mean_token_accuracy": 0.17920879274606705,
      "num_tokens": 71113878.0,
      "step": 35095
    },
    {
      "entropy": 6.124137544631958,
      "epoch": 3.0673774359870665,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004068833477352927,
      "loss": 5.6943,
      "mean_token_accuracy": 0.16833848506212234,
      "num_tokens": 71124849.0,
      "step": 35100
    },
    {
      "entropy": 6.032493638992309,
      "epoch": 3.067814384339771,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040685809962910614,
      "loss": 5.5792,
      "mean_token_accuracy": 0.18595270067453384,
      "num_tokens": 71134064.0,
      "step": 35105
    },
    {
      "entropy": 5.8616291046142575,
      "epoch": 3.068251332692476,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040683284899381225,
      "loss": 5.4602,
      "mean_token_accuracy": 0.1832078590989113,
      "num_tokens": 71144270.0,
      "step": 35110
    },
    {
      "entropy": 5.981511926651001,
      "epoch": 3.0686882810451803,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040680759582989516,
      "loss": 5.4901,
      "mean_token_accuracy": 0.18465267419815062,
      "num_tokens": 71153949.0,
      "step": 35115
    },
    {
      "entropy": 6.021999740600586,
      "epoch": 3.0691252293978852,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040678234013783945,
      "loss": 5.6185,
      "mean_token_accuracy": 0.17689593732357026,
      "num_tokens": 71164858.0,
      "step": 35120
    },
    {
      "entropy": 6.017419481277466,
      "epoch": 3.0695621777505897,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004067570819181293,
      "loss": 5.5033,
      "mean_token_accuracy": 0.18397019505500795,
      "num_tokens": 71174832.0,
      "step": 35125
    },
    {
      "entropy": 5.978369855880738,
      "epoch": 3.0699991261032946,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004067318211712492,
      "loss": 5.6049,
      "mean_token_accuracy": 0.17365654557943344,
      "num_tokens": 71185089.0,
      "step": 35130
    },
    {
      "entropy": 6.123804235458374,
      "epoch": 3.070436074455999,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004067065578976839,
      "loss": 5.5994,
      "mean_token_accuracy": 0.1721036195755005,
      "num_tokens": 71195889.0,
      "step": 35135
    },
    {
      "entropy": 6.068632459640503,
      "epoch": 3.070873022808704,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040668129209791773,
      "loss": 5.6389,
      "mean_token_accuracy": 0.1746524006128311,
      "num_tokens": 71205825.0,
      "step": 35140
    },
    {
      "entropy": 5.961380290985107,
      "epoch": 3.071309971161409,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004066560237724354,
      "loss": 5.5081,
      "mean_token_accuracy": 0.18508177995681763,
      "num_tokens": 71215201.0,
      "step": 35145
    },
    {
      "entropy": 5.926025819778443,
      "epoch": 3.0717469195141134,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004066307529217215,
      "loss": 5.4774,
      "mean_token_accuracy": 0.18921613097190856,
      "num_tokens": 71225329.0,
      "step": 35150
    },
    {
      "entropy": 5.9954132556915285,
      "epoch": 3.0721838678668183,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040660547954626076,
      "loss": 5.4687,
      "mean_token_accuracy": 0.189494751393795,
      "num_tokens": 71235301.0,
      "step": 35155
    },
    {
      "entropy": 5.970604658126831,
      "epoch": 3.0726208162195228,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040658020364653803,
      "loss": 5.5752,
      "mean_token_accuracy": 0.17766468971967697,
      "num_tokens": 71244991.0,
      "step": 35160
    },
    {
      "entropy": 5.956370306015015,
      "epoch": 3.0730577645722277,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000406554925223038,
      "loss": 5.552,
      "mean_token_accuracy": 0.18037534803152083,
      "num_tokens": 71255480.0,
      "step": 35165
    },
    {
      "entropy": 6.087510347366333,
      "epoch": 3.073494712924932,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004065296442762456,
      "loss": 5.7029,
      "mean_token_accuracy": 0.17105701118707656,
      "num_tokens": 71265996.0,
      "step": 35170
    },
    {
      "entropy": 6.059965515136719,
      "epoch": 3.073931661277637,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00040650436080664576,
      "loss": 5.595,
      "mean_token_accuracy": 0.17524192929267884,
      "num_tokens": 71277293.0,
      "step": 35175
    },
    {
      "entropy": 5.981881618499756,
      "epoch": 3.0743686096303415,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040647907481472333,
      "loss": 5.5116,
      "mean_token_accuracy": 0.18191390931606294,
      "num_tokens": 71285903.0,
      "step": 35180
    },
    {
      "entropy": 5.930354928970337,
      "epoch": 3.0748055579830464,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004064537863009633,
      "loss": 5.5043,
      "mean_token_accuracy": 0.1911282017827034,
      "num_tokens": 71294979.0,
      "step": 35185
    },
    {
      "entropy": 5.971945476531983,
      "epoch": 3.075242506335751,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040642849526585087,
      "loss": 5.5351,
      "mean_token_accuracy": 0.19061816185712815,
      "num_tokens": 71306223.0,
      "step": 35190
    },
    {
      "entropy": 6.110079288482666,
      "epoch": 3.075679454688456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004064032017098709,
      "loss": 5.672,
      "mean_token_accuracy": 0.1703340381383896,
      "num_tokens": 71316984.0,
      "step": 35195
    },
    {
      "entropy": 6.087320232391358,
      "epoch": 3.0761164030411607,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040637790563350886,
      "loss": 5.6479,
      "mean_token_accuracy": 0.18307924121618271,
      "num_tokens": 71327855.0,
      "step": 35200
    },
    {
      "entropy": 5.970929670333862,
      "epoch": 3.076553351393865,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004063526070372496,
      "loss": 5.5508,
      "mean_token_accuracy": 0.1832497626543045,
      "num_tokens": 71338115.0,
      "step": 35205
    },
    {
      "entropy": 5.950755643844604,
      "epoch": 3.07699029974657,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004063273059215785,
      "loss": 5.5264,
      "mean_token_accuracy": 0.18209238499403,
      "num_tokens": 71348533.0,
      "step": 35210
    },
    {
      "entropy": 5.972426223754883,
      "epoch": 3.0774272480992746,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004063020022869809,
      "loss": 5.3982,
      "mean_token_accuracy": 0.19886647611856462,
      "num_tokens": 71358368.0,
      "step": 35215
    },
    {
      "entropy": 6.018426895141602,
      "epoch": 3.0778641964519795,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040627669613394206,
      "loss": 5.611,
      "mean_token_accuracy": 0.1792762503027916,
      "num_tokens": 71368443.0,
      "step": 35220
    },
    {
      "entropy": 6.015796899795532,
      "epoch": 3.078301144804684,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040625138746294747,
      "loss": 5.6028,
      "mean_token_accuracy": 0.1707278609275818,
      "num_tokens": 71378508.0,
      "step": 35225
    },
    {
      "entropy": 6.02926082611084,
      "epoch": 3.078738093157389,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004062260762744825,
      "loss": 5.5217,
      "mean_token_accuracy": 0.18282075077295304,
      "num_tokens": 71388512.0,
      "step": 35230
    },
    {
      "entropy": 6.00957179069519,
      "epoch": 3.0791750415100934,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040620076256903256,
      "loss": 5.5729,
      "mean_token_accuracy": 0.1786183625459671,
      "num_tokens": 71397908.0,
      "step": 35235
    },
    {
      "entropy": 6.0480729103088375,
      "epoch": 3.0796119898627983,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004061754463470833,
      "loss": 5.6421,
      "mean_token_accuracy": 0.17347114235162736,
      "num_tokens": 71408287.0,
      "step": 35240
    },
    {
      "entropy": 6.099638557434082,
      "epoch": 3.0800489382155027,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004061501276091202,
      "loss": 5.6442,
      "mean_token_accuracy": 0.17382457256317138,
      "num_tokens": 71419227.0,
      "step": 35245
    },
    {
      "entropy": 6.010980892181396,
      "epoch": 3.0804858865682077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000406124806355629,
      "loss": 5.5875,
      "mean_token_accuracy": 0.1825943723320961,
      "num_tokens": 71429398.0,
      "step": 35250
    },
    {
      "entropy": 5.990431213378907,
      "epoch": 3.080922834920912,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040609948258709517,
      "loss": 5.5746,
      "mean_token_accuracy": 0.17889802604913713,
      "num_tokens": 71440185.0,
      "step": 35255
    },
    {
      "entropy": 5.920681619644165,
      "epoch": 3.081359783273617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004060741563040047,
      "loss": 5.52,
      "mean_token_accuracy": 0.1933665692806244,
      "num_tokens": 71451160.0,
      "step": 35260
    },
    {
      "entropy": 6.056257438659668,
      "epoch": 3.081796731626322,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004060488275068432,
      "loss": 5.6235,
      "mean_token_accuracy": 0.1751692295074463,
      "num_tokens": 71461459.0,
      "step": 35265
    },
    {
      "entropy": 5.938502645492553,
      "epoch": 3.0822336799790264,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00040602349619609646,
      "loss": 5.5313,
      "mean_token_accuracy": 0.18195926547050476,
      "num_tokens": 71470806.0,
      "step": 35270
    },
    {
      "entropy": 5.8890634059906,
      "epoch": 3.0826706283317313,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040599816237225044,
      "loss": 5.5377,
      "mean_token_accuracy": 0.18443059772253037,
      "num_tokens": 71480929.0,
      "step": 35275
    },
    {
      "entropy": 6.020742177963257,
      "epoch": 3.083107576684436,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040597282603579103,
      "loss": 5.5692,
      "mean_token_accuracy": 0.183108289539814,
      "num_tokens": 71490360.0,
      "step": 35280
    },
    {
      "entropy": 5.989965200424194,
      "epoch": 3.0835445250371407,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040594748718720424,
      "loss": 5.6112,
      "mean_token_accuracy": 0.1757531315088272,
      "num_tokens": 71500355.0,
      "step": 35285
    },
    {
      "entropy": 5.944901084899902,
      "epoch": 3.083981473389845,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040592214582697595,
      "loss": 5.5544,
      "mean_token_accuracy": 0.18594004958868027,
      "num_tokens": 71509581.0,
      "step": 35290
    },
    {
      "entropy": 6.00521445274353,
      "epoch": 3.08441842174255,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004058968019555923,
      "loss": 5.601,
      "mean_token_accuracy": 0.17420065253973008,
      "num_tokens": 71519967.0,
      "step": 35295
    },
    {
      "entropy": 5.946793842315674,
      "epoch": 3.0848553700952546,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040587145557353943,
      "loss": 5.4997,
      "mean_token_accuracy": 0.18349019438028336,
      "num_tokens": 71530619.0,
      "step": 35300
    },
    {
      "entropy": 6.057626056671142,
      "epoch": 3.0852923184479595,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004058461066813035,
      "loss": 5.6384,
      "mean_token_accuracy": 0.1742366537451744,
      "num_tokens": 71541682.0,
      "step": 35305
    },
    {
      "entropy": 5.998140478134156,
      "epoch": 3.085729266800664,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040582075527937066,
      "loss": 5.5465,
      "mean_token_accuracy": 0.1798931822180748,
      "num_tokens": 71552971.0,
      "step": 35310
    },
    {
      "entropy": 5.963777828216553,
      "epoch": 3.086166215153369,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004057954013682272,
      "loss": 5.5502,
      "mean_token_accuracy": 0.18249205946922303,
      "num_tokens": 71562932.0,
      "step": 35315
    },
    {
      "entropy": 6.03954119682312,
      "epoch": 3.0866031635060738,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040577004494835934,
      "loss": 5.5117,
      "mean_token_accuracy": 0.18287068903446196,
      "num_tokens": 71572662.0,
      "step": 35320
    },
    {
      "entropy": 5.9664195537567135,
      "epoch": 3.0870401118587782,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040574468602025355,
      "loss": 5.5327,
      "mean_token_accuracy": 0.18731914907693864,
      "num_tokens": 71582503.0,
      "step": 35325
    },
    {
      "entropy": 6.015541458129883,
      "epoch": 3.087477060211483,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040571932458439617,
      "loss": 5.601,
      "mean_token_accuracy": 0.17561887800693513,
      "num_tokens": 71592261.0,
      "step": 35330
    },
    {
      "entropy": 6.022887182235718,
      "epoch": 3.0879140085641876,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040569396064127355,
      "loss": 5.551,
      "mean_token_accuracy": 0.18781552016735076,
      "num_tokens": 71602514.0,
      "step": 35335
    },
    {
      "entropy": 6.040487003326416,
      "epoch": 3.0883509569168925,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004056685941913724,
      "loss": 5.5873,
      "mean_token_accuracy": 0.18355743885040282,
      "num_tokens": 71612912.0,
      "step": 35340
    },
    {
      "entropy": 5.931476402282715,
      "epoch": 3.088787905269597,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040564322523517905,
      "loss": 5.4993,
      "mean_token_accuracy": 0.1786837711930275,
      "num_tokens": 71622673.0,
      "step": 35345
    },
    {
      "entropy": 6.013609933853149,
      "epoch": 3.089224853622302,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004056178537731803,
      "loss": 5.572,
      "mean_token_accuracy": 0.1788382738828659,
      "num_tokens": 71633315.0,
      "step": 35350
    },
    {
      "entropy": 6.045373964309692,
      "epoch": 3.0896618019750064,
      "grad_norm": 1.875,
      "learning_rate": 0.0004055924798058626,
      "loss": 5.6198,
      "mean_token_accuracy": 0.175252240896225,
      "num_tokens": 71644159.0,
      "step": 35355
    },
    {
      "entropy": 6.012170839309692,
      "epoch": 3.0900987503277113,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040556710333371275,
      "loss": 5.5568,
      "mean_token_accuracy": 0.1828256815671921,
      "num_tokens": 71654227.0,
      "step": 35360
    },
    {
      "entropy": 6.055656147003174,
      "epoch": 3.0905356986804158,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004055417243572174,
      "loss": 5.5296,
      "mean_token_accuracy": 0.18055397570133208,
      "num_tokens": 71664711.0,
      "step": 35365
    },
    {
      "entropy": 5.9899975776672365,
      "epoch": 3.0909726470331207,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040551634287686343,
      "loss": 5.5736,
      "mean_token_accuracy": 0.18235231041908265,
      "num_tokens": 71674880.0,
      "step": 35370
    },
    {
      "entropy": 6.015809631347656,
      "epoch": 3.0914095953858256,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004054909588931375,
      "loss": 5.6245,
      "mean_token_accuracy": 0.17065771520137787,
      "num_tokens": 71685386.0,
      "step": 35375
    },
    {
      "entropy": 6.03713903427124,
      "epoch": 3.09184654373853,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004054655724065268,
      "loss": 5.6938,
      "mean_token_accuracy": 0.17327403128147126,
      "num_tokens": 71696008.0,
      "step": 35380
    },
    {
      "entropy": 5.994466829299927,
      "epoch": 3.092283492091235,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004054401834175179,
      "loss": 5.542,
      "mean_token_accuracy": 0.17845948189496993,
      "num_tokens": 71705548.0,
      "step": 35385
    },
    {
      "entropy": 5.980149984359741,
      "epoch": 3.0927204404439395,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040541479192659797,
      "loss": 5.5947,
      "mean_token_accuracy": 0.1818003237247467,
      "num_tokens": 71716891.0,
      "step": 35390
    },
    {
      "entropy": 6.0637153625488285,
      "epoch": 3.0931573887966444,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040538939793425394,
      "loss": 5.6506,
      "mean_token_accuracy": 0.16989256888628007,
      "num_tokens": 71727954.0,
      "step": 35395
    },
    {
      "entropy": 6.009288787841797,
      "epoch": 3.093594337149349,
      "grad_norm": 1.015625,
      "learning_rate": 0.000405364001440973,
      "loss": 5.6056,
      "mean_token_accuracy": 0.17797950208187102,
      "num_tokens": 71738282.0,
      "step": 35400
    },
    {
      "entropy": 6.0353960514068605,
      "epoch": 3.0940312855020538,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004053386024472422,
      "loss": 5.6276,
      "mean_token_accuracy": 0.17841200679540634,
      "num_tokens": 71747895.0,
      "step": 35405
    },
    {
      "entropy": 5.974404335021973,
      "epoch": 3.094468233854758,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004053132009535487,
      "loss": 5.552,
      "mean_token_accuracy": 0.1841041177511215,
      "num_tokens": 71758107.0,
      "step": 35410
    },
    {
      "entropy": 6.0719388961792,
      "epoch": 3.094905182207463,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004052877969603797,
      "loss": 5.6628,
      "mean_token_accuracy": 0.17489190697669982,
      "num_tokens": 71767661.0,
      "step": 35415
    },
    {
      "entropy": 6.0678094863891605,
      "epoch": 3.0953421305601676,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004052623904682225,
      "loss": 5.6117,
      "mean_token_accuracy": 0.17866088896989823,
      "num_tokens": 71777908.0,
      "step": 35420
    },
    {
      "entropy": 5.9787046909332275,
      "epoch": 3.0957790789128725,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040523698147756443,
      "loss": 5.559,
      "mean_token_accuracy": 0.1819167599081993,
      "num_tokens": 71786809.0,
      "step": 35425
    },
    {
      "entropy": 5.946514558792114,
      "epoch": 3.0962160272655774,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040521156998889283,
      "loss": 5.5618,
      "mean_token_accuracy": 0.17802517414093016,
      "num_tokens": 71797299.0,
      "step": 35430
    },
    {
      "entropy": 6.072604703903198,
      "epoch": 3.096652975618282,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040518615600269506,
      "loss": 5.5811,
      "mean_token_accuracy": 0.1815621718764305,
      "num_tokens": 71807565.0,
      "step": 35435
    },
    {
      "entropy": 5.96094708442688,
      "epoch": 3.097089923970987,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004051607395194586,
      "loss": 5.5717,
      "mean_token_accuracy": 0.1755717873573303,
      "num_tokens": 71817288.0,
      "step": 35440
    },
    {
      "entropy": 6.071892786026001,
      "epoch": 3.0975268723236913,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040513532053967086,
      "loss": 5.5421,
      "mean_token_accuracy": 0.18054813742637635,
      "num_tokens": 71827371.0,
      "step": 35445
    },
    {
      "entropy": 6.074623060226441,
      "epoch": 3.097963820676396,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040510989906381955,
      "loss": 5.6945,
      "mean_token_accuracy": 0.1760356143116951,
      "num_tokens": 71837872.0,
      "step": 35450
    },
    {
      "entropy": 6.00829701423645,
      "epoch": 3.0984007690291007,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040508447509239224,
      "loss": 5.5376,
      "mean_token_accuracy": 0.18654794991016388,
      "num_tokens": 71848507.0,
      "step": 35455
    },
    {
      "entropy": 6.024956512451172,
      "epoch": 3.0988377173818056,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004050590486258764,
      "loss": 5.6155,
      "mean_token_accuracy": 0.17324639707803727,
      "num_tokens": 71860495.0,
      "step": 35460
    },
    {
      "entropy": 6.0142584323883055,
      "epoch": 3.09927466573451,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00040503361966476005,
      "loss": 5.5414,
      "mean_token_accuracy": 0.1863507881760597,
      "num_tokens": 71870081.0,
      "step": 35465
    },
    {
      "entropy": 6.049063587188721,
      "epoch": 3.099711614087215,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004050081882095307,
      "loss": 5.5226,
      "mean_token_accuracy": 0.18032127022743225,
      "num_tokens": 71880072.0,
      "step": 35470
    },
    {
      "entropy": 5.960941934585572,
      "epoch": 3.1001485624399194,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004049827542606761,
      "loss": 5.5428,
      "mean_token_accuracy": 0.17981524467468263,
      "num_tokens": 71889225.0,
      "step": 35475
    },
    {
      "entropy": 5.9670374393463135,
      "epoch": 3.1005855107926243,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004049573178186842,
      "loss": 5.5549,
      "mean_token_accuracy": 0.18386708348989486,
      "num_tokens": 71898601.0,
      "step": 35480
    },
    {
      "entropy": 6.037700939178467,
      "epoch": 3.101022459145329,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040493187888404283,
      "loss": 5.5917,
      "mean_token_accuracy": 0.17237084954977036,
      "num_tokens": 71909267.0,
      "step": 35485
    },
    {
      "entropy": 6.045227336883545,
      "epoch": 3.1014594074980337,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040490643745724,
      "loss": 5.5979,
      "mean_token_accuracy": 0.17819452434778213,
      "num_tokens": 71919066.0,
      "step": 35490
    },
    {
      "entropy": 5.995789861679077,
      "epoch": 3.1018963558507386,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004048809935387635,
      "loss": 5.5444,
      "mean_token_accuracy": 0.18128197640180588,
      "num_tokens": 71929067.0,
      "step": 35495
    },
    {
      "entropy": 6.018492364883423,
      "epoch": 3.102333304203443,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040485554712910155,
      "loss": 5.4855,
      "mean_token_accuracy": 0.18079111278057097,
      "num_tokens": 71938044.0,
      "step": 35500
    },
    {
      "entropy": 5.96659631729126,
      "epoch": 3.102770252556148,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040483009822874216,
      "loss": 5.5812,
      "mean_token_accuracy": 0.18351407051086427,
      "num_tokens": 71947714.0,
      "step": 35505
    },
    {
      "entropy": 5.999036741256714,
      "epoch": 3.1032072009088525,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004048046468381735,
      "loss": 5.5514,
      "mean_token_accuracy": 0.18085912466049195,
      "num_tokens": 71957660.0,
      "step": 35510
    },
    {
      "entropy": 5.981327295303345,
      "epoch": 3.1036441492615574,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004047791929578837,
      "loss": 5.4992,
      "mean_token_accuracy": 0.1916690707206726,
      "num_tokens": 71967063.0,
      "step": 35515
    },
    {
      "entropy": 5.933385753631592,
      "epoch": 3.104081097614262,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004047537365883609,
      "loss": 5.5847,
      "mean_token_accuracy": 0.18116633892059325,
      "num_tokens": 71977331.0,
      "step": 35520
    },
    {
      "entropy": 5.94909782409668,
      "epoch": 3.104518045966967,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040472827773009346,
      "loss": 5.4783,
      "mean_token_accuracy": 0.19156375527381897,
      "num_tokens": 71987047.0,
      "step": 35525
    },
    {
      "entropy": 6.0328692436218265,
      "epoch": 3.1049549943196713,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040470281638356974,
      "loss": 5.5094,
      "mean_token_accuracy": 0.1803298145532608,
      "num_tokens": 71997057.0,
      "step": 35530
    },
    {
      "entropy": 6.038141870498658,
      "epoch": 3.105391942672376,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000404677352549278,
      "loss": 5.6269,
      "mean_token_accuracy": 0.17800198048353194,
      "num_tokens": 72007092.0,
      "step": 35535
    },
    {
      "entropy": 6.074412536621094,
      "epoch": 3.1058288910250806,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040465188622770663,
      "loss": 5.6498,
      "mean_token_accuracy": 0.18070678263902665,
      "num_tokens": 72016847.0,
      "step": 35540
    },
    {
      "entropy": 5.957003879547119,
      "epoch": 3.1062658393777856,
      "grad_norm": 1.125,
      "learning_rate": 0.0004046264174193443,
      "loss": 5.5236,
      "mean_token_accuracy": 0.18220469504594802,
      "num_tokens": 72026706.0,
      "step": 35545
    },
    {
      "entropy": 6.076756954193115,
      "epoch": 3.1067027877304905,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004046009461246792,
      "loss": 5.68,
      "mean_token_accuracy": 0.17264709621667862,
      "num_tokens": 72036358.0,
      "step": 35550
    },
    {
      "entropy": 6.048372650146485,
      "epoch": 3.107139736083195,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004045754723442001,
      "loss": 5.658,
      "mean_token_accuracy": 0.17355891317129135,
      "num_tokens": 72049156.0,
      "step": 35555
    },
    {
      "entropy": 6.009410810470581,
      "epoch": 3.1075766844359,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004045499960783955,
      "loss": 5.532,
      "mean_token_accuracy": 0.1829036757349968,
      "num_tokens": 72058986.0,
      "step": 35560
    },
    {
      "entropy": 6.035253477096558,
      "epoch": 3.1080136327886043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004045245173277543,
      "loss": 5.5354,
      "mean_token_accuracy": 0.1826351597905159,
      "num_tokens": 72069141.0,
      "step": 35565
    },
    {
      "entropy": 5.935996341705322,
      "epoch": 3.1084505811413092,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040449903609276474,
      "loss": 5.5033,
      "mean_token_accuracy": 0.19583865404129028,
      "num_tokens": 72079009.0,
      "step": 35570
    },
    {
      "entropy": 6.018488645553589,
      "epoch": 3.1088875294940137,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004044735523739159,
      "loss": 5.5606,
      "mean_token_accuracy": 0.17877979576587677,
      "num_tokens": 72089551.0,
      "step": 35575
    },
    {
      "entropy": 6.065915632247925,
      "epoch": 3.1093244778467186,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040444806617169655,
      "loss": 5.5444,
      "mean_token_accuracy": 0.18489671051502227,
      "num_tokens": 72098965.0,
      "step": 35580
    },
    {
      "entropy": 6.01924352645874,
      "epoch": 3.109761426199423,
      "grad_norm": 2.203125,
      "learning_rate": 0.00040442257748659537,
      "loss": 5.5776,
      "mean_token_accuracy": 0.18097543865442275,
      "num_tokens": 72109568.0,
      "step": 35585
    },
    {
      "entropy": 5.952869606018067,
      "epoch": 3.110198374552128,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040439708631910136,
      "loss": 5.5722,
      "mean_token_accuracy": 0.18289681673049926,
      "num_tokens": 72119988.0,
      "step": 35590
    },
    {
      "entropy": 5.967882204055786,
      "epoch": 3.1106353229048325,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004043715926697034,
      "loss": 5.6602,
      "mean_token_accuracy": 0.1808217868208885,
      "num_tokens": 72130515.0,
      "step": 35595
    },
    {
      "entropy": 6.097915363311768,
      "epoch": 3.1110722712575374,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004043460965388906,
      "loss": 5.5797,
      "mean_token_accuracy": 0.17823438793420793,
      "num_tokens": 72140362.0,
      "step": 35600
    },
    {
      "entropy": 6.033664894104004,
      "epoch": 3.111509219610242,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004043205979271518,
      "loss": 5.5626,
      "mean_token_accuracy": 0.17796529978513717,
      "num_tokens": 72150726.0,
      "step": 35605
    },
    {
      "entropy": 6.032419157028198,
      "epoch": 3.1119461679629468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004042950968349762,
      "loss": 5.5533,
      "mean_token_accuracy": 0.18242936581373215,
      "num_tokens": 72160264.0,
      "step": 35610
    },
    {
      "entropy": 5.988433790206909,
      "epoch": 3.1123831163156517,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004042695932628529,
      "loss": 5.5904,
      "mean_token_accuracy": 0.17926383018493652,
      "num_tokens": 72171407.0,
      "step": 35615
    },
    {
      "entropy": 6.059711217880249,
      "epoch": 3.112820064668356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004042440872112711,
      "loss": 5.6181,
      "mean_token_accuracy": 0.17887272089719772,
      "num_tokens": 72182132.0,
      "step": 35620
    },
    {
      "entropy": 6.007093095779419,
      "epoch": 3.113257013021061,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00040421857868071996,
      "loss": 5.5755,
      "mean_token_accuracy": 0.17337146401405334,
      "num_tokens": 72192129.0,
      "step": 35625
    },
    {
      "entropy": 5.95858907699585,
      "epoch": 3.1136939613737655,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040419306767168874,
      "loss": 5.5532,
      "mean_token_accuracy": 0.17818013727664947,
      "num_tokens": 72202053.0,
      "step": 35630
    },
    {
      "entropy": 6.018731355667114,
      "epoch": 3.1141309097264704,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004041675541846668,
      "loss": 5.5254,
      "mean_token_accuracy": 0.1815118357539177,
      "num_tokens": 72211352.0,
      "step": 35635
    },
    {
      "entropy": 5.986498546600342,
      "epoch": 3.114567858079175,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040414203822014343,
      "loss": 5.5345,
      "mean_token_accuracy": 0.18408384621143342,
      "num_tokens": 72220816.0,
      "step": 35640
    },
    {
      "entropy": 6.035413217544556,
      "epoch": 3.11500480643188,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040411651977860817,
      "loss": 5.6422,
      "mean_token_accuracy": 0.17327730655670165,
      "num_tokens": 72231318.0,
      "step": 35645
    },
    {
      "entropy": 5.993533039093018,
      "epoch": 3.1154417547845843,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040409099886055044,
      "loss": 5.564,
      "mean_token_accuracy": 0.17949487119913102,
      "num_tokens": 72241035.0,
      "step": 35650
    },
    {
      "entropy": 6.046349716186524,
      "epoch": 3.115878703137289,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004040654754664596,
      "loss": 5.5054,
      "mean_token_accuracy": 0.18322764337062836,
      "num_tokens": 72249967.0,
      "step": 35655
    },
    {
      "entropy": 6.06565670967102,
      "epoch": 3.1163156514899937,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040403994959682546,
      "loss": 5.6064,
      "mean_token_accuracy": 0.18472241163253783,
      "num_tokens": 72260114.0,
      "step": 35660
    },
    {
      "entropy": 6.034187459945679,
      "epoch": 3.1167525998426986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040401442125213735,
      "loss": 5.5212,
      "mean_token_accuracy": 0.17749065309762954,
      "num_tokens": 72270257.0,
      "step": 35665
    },
    {
      "entropy": 5.958642387390137,
      "epoch": 3.1171895481954035,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040398889043288507,
      "loss": 5.5127,
      "mean_token_accuracy": 0.18483913093805313,
      "num_tokens": 72279628.0,
      "step": 35670
    },
    {
      "entropy": 5.962270307540893,
      "epoch": 3.117626496548108,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004039633571395583,
      "loss": 5.5506,
      "mean_token_accuracy": 0.18161542862653732,
      "num_tokens": 72289493.0,
      "step": 35675
    },
    {
      "entropy": 5.941229629516601,
      "epoch": 3.118063444900813,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040393782137264676,
      "loss": 5.4665,
      "mean_token_accuracy": 0.19091463088989258,
      "num_tokens": 72300007.0,
      "step": 35680
    },
    {
      "entropy": 5.993233680725098,
      "epoch": 3.1185003932535174,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004039122831326402,
      "loss": 5.5418,
      "mean_token_accuracy": 0.18707987517118455,
      "num_tokens": 72310596.0,
      "step": 35685
    },
    {
      "entropy": 6.010622024536133,
      "epoch": 3.1189373416062223,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004038867424200285,
      "loss": 5.6124,
      "mean_token_accuracy": 0.17840465307235717,
      "num_tokens": 72320156.0,
      "step": 35690
    },
    {
      "entropy": 6.053464126586914,
      "epoch": 3.1193742899589267,
      "grad_norm": 1.25,
      "learning_rate": 0.0004038611992353016,
      "loss": 5.5739,
      "mean_token_accuracy": 0.17960516065359117,
      "num_tokens": 72329608.0,
      "step": 35695
    },
    {
      "entropy": 5.9526939868927,
      "epoch": 3.1198112383116317,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004038356535789494,
      "loss": 5.5051,
      "mean_token_accuracy": 0.18552400916814804,
      "num_tokens": 72339565.0,
      "step": 35700
    },
    {
      "entropy": 5.98362021446228,
      "epoch": 3.120248186664336,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040381010545146174,
      "loss": 5.6158,
      "mean_token_accuracy": 0.17679723501205444,
      "num_tokens": 72349566.0,
      "step": 35705
    },
    {
      "entropy": 6.071577692031861,
      "epoch": 3.120685135017041,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004037845548533289,
      "loss": 5.5261,
      "mean_token_accuracy": 0.18255942761898042,
      "num_tokens": 72360366.0,
      "step": 35710
    },
    {
      "entropy": 6.07366189956665,
      "epoch": 3.1211220833697455,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040375900178504066,
      "loss": 5.6803,
      "mean_token_accuracy": 0.17754087299108506,
      "num_tokens": 72369893.0,
      "step": 35715
    },
    {
      "entropy": 5.988529109954834,
      "epoch": 3.1215590317224504,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004037334462470874,
      "loss": 5.5955,
      "mean_token_accuracy": 0.1793469101190567,
      "num_tokens": 72380484.0,
      "step": 35720
    },
    {
      "entropy": 6.011472606658936,
      "epoch": 3.1219959800751553,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004037078882399591,
      "loss": 5.5099,
      "mean_token_accuracy": 0.1845584899187088,
      "num_tokens": 72391088.0,
      "step": 35725
    },
    {
      "entropy": 6.076337194442749,
      "epoch": 3.12243292842786,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004036823277641462,
      "loss": 5.6052,
      "mean_token_accuracy": 0.17147317826747893,
      "num_tokens": 72401253.0,
      "step": 35730
    },
    {
      "entropy": 5.999081087112427,
      "epoch": 3.1228698767805647,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040365676482013877,
      "loss": 5.491,
      "mean_token_accuracy": 0.18742310255765915,
      "num_tokens": 72411235.0,
      "step": 35735
    },
    {
      "entropy": 5.967172574996948,
      "epoch": 3.123306825133269,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004036311994084271,
      "loss": 5.612,
      "mean_token_accuracy": 0.1793233186006546,
      "num_tokens": 72421387.0,
      "step": 35740
    },
    {
      "entropy": 5.979734420776367,
      "epoch": 3.123743773485974,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00040360563152950173,
      "loss": 5.598,
      "mean_token_accuracy": 0.17904335856437684,
      "num_tokens": 72431431.0,
      "step": 35745
    },
    {
      "entropy": 5.989374113082886,
      "epoch": 3.1241807218386786,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004035800611838529,
      "loss": 5.5762,
      "mean_token_accuracy": 0.18506680130958558,
      "num_tokens": 72441286.0,
      "step": 35750
    },
    {
      "entropy": 5.992023086547851,
      "epoch": 3.1246176701913835,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040355448837197115,
      "loss": 5.5899,
      "mean_token_accuracy": 0.18398657739162444,
      "num_tokens": 72451661.0,
      "step": 35755
    },
    {
      "entropy": 6.0224157810211185,
      "epoch": 3.125054618544088,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040352891309434693,
      "loss": 5.5852,
      "mean_token_accuracy": 0.17643724828958512,
      "num_tokens": 72462013.0,
      "step": 35760
    },
    {
      "entropy": 5.983252429962159,
      "epoch": 3.125491566896793,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004035033353514708,
      "loss": 5.5188,
      "mean_token_accuracy": 0.18271221965551376,
      "num_tokens": 72472804.0,
      "step": 35765
    },
    {
      "entropy": 6.013465738296508,
      "epoch": 3.1259285152494973,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00040347775514383344,
      "loss": 5.5684,
      "mean_token_accuracy": 0.180029259622097,
      "num_tokens": 72482893.0,
      "step": 35770
    },
    {
      "entropy": 5.998140048980713,
      "epoch": 3.1263654636022022,
      "grad_norm": 1.53125,
      "learning_rate": 0.00040345217247192536,
      "loss": 5.5513,
      "mean_token_accuracy": 0.18292346894741057,
      "num_tokens": 72492882.0,
      "step": 35775
    },
    {
      "entropy": 5.957479667663574,
      "epoch": 3.126802411954907,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004034265873362373,
      "loss": 5.4679,
      "mean_token_accuracy": 0.1914670631289482,
      "num_tokens": 72502322.0,
      "step": 35780
    },
    {
      "entropy": 6.002833604812622,
      "epoch": 3.1272393603076116,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040340099973726,
      "loss": 5.564,
      "mean_token_accuracy": 0.1871866539120674,
      "num_tokens": 72511401.0,
      "step": 35785
    },
    {
      "entropy": 5.981036376953125,
      "epoch": 3.1276763086603165,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040337540967548427,
      "loss": 5.6053,
      "mean_token_accuracy": 0.18362920880317687,
      "num_tokens": 72522071.0,
      "step": 35790
    },
    {
      "entropy": 6.05577392578125,
      "epoch": 3.128113257013021,
      "grad_norm": 2.21875,
      "learning_rate": 0.00040334981715140086,
      "loss": 5.6931,
      "mean_token_accuracy": 0.17334589809179307,
      "num_tokens": 72532899.0,
      "step": 35795
    },
    {
      "entropy": 6.0867656707763675,
      "epoch": 3.128550205365726,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040332422216550067,
      "loss": 5.6246,
      "mean_token_accuracy": 0.18230101764202117,
      "num_tokens": 72543372.0,
      "step": 35800
    },
    {
      "entropy": 5.969922637939453,
      "epoch": 3.1289871537184304,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004032986247182748,
      "loss": 5.5588,
      "mean_token_accuracy": 0.18846285939216614,
      "num_tokens": 72553014.0,
      "step": 35805
    },
    {
      "entropy": 5.945178461074829,
      "epoch": 3.1294241020711353,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040327302481021394,
      "loss": 5.4917,
      "mean_token_accuracy": 0.19421261996030809,
      "num_tokens": 72562281.0,
      "step": 35810
    },
    {
      "entropy": 6.004241991043091,
      "epoch": 3.1298610504238398,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040324742244180937,
      "loss": 5.5,
      "mean_token_accuracy": 0.18542379885911942,
      "num_tokens": 72571817.0,
      "step": 35815
    },
    {
      "entropy": 5.97601547241211,
      "epoch": 3.1302979987765447,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040322181761355194,
      "loss": 5.5544,
      "mean_token_accuracy": 0.1848520204424858,
      "num_tokens": 72582935.0,
      "step": 35820
    },
    {
      "entropy": 6.010506200790405,
      "epoch": 3.130734947129249,
      "grad_norm": 1.125,
      "learning_rate": 0.0004031962103259329,
      "loss": 5.5499,
      "mean_token_accuracy": 0.182881598174572,
      "num_tokens": 72593280.0,
      "step": 35825
    },
    {
      "entropy": 5.977372741699218,
      "epoch": 3.131171895481954,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004031706005794434,
      "loss": 5.5245,
      "mean_token_accuracy": 0.18052926510572434,
      "num_tokens": 72603193.0,
      "step": 35830
    },
    {
      "entropy": 5.911997556686401,
      "epoch": 3.1316088438346585,
      "grad_norm": 1.125,
      "learning_rate": 0.0004031449883745746,
      "loss": 5.5536,
      "mean_token_accuracy": 0.1885907918214798,
      "num_tokens": 72613363.0,
      "step": 35835
    },
    {
      "entropy": 6.007095098495483,
      "epoch": 3.1320457921873635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040311937371181774,
      "loss": 5.6925,
      "mean_token_accuracy": 0.17509364783763887,
      "num_tokens": 72624151.0,
      "step": 35840
    },
    {
      "entropy": 6.045786714553833,
      "epoch": 3.1324827405400684,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040309375659166414,
      "loss": 5.4895,
      "mean_token_accuracy": 0.18985640108585358,
      "num_tokens": 72633692.0,
      "step": 35845
    },
    {
      "entropy": 5.977776384353637,
      "epoch": 3.132919688892773,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040306813701460525,
      "loss": 5.5602,
      "mean_token_accuracy": 0.18105557858943938,
      "num_tokens": 72644016.0,
      "step": 35850
    },
    {
      "entropy": 5.98924241065979,
      "epoch": 3.1333566372454777,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004030425149811324,
      "loss": 5.5628,
      "mean_token_accuracy": 0.1871321126818657,
      "num_tokens": 72654033.0,
      "step": 35855
    },
    {
      "entropy": 6.052525758743286,
      "epoch": 3.133793585598182,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004030168904917369,
      "loss": 5.5177,
      "mean_token_accuracy": 0.17655303329229355,
      "num_tokens": 72663452.0,
      "step": 35860
    },
    {
      "entropy": 6.016387701034546,
      "epoch": 3.134230533950887,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040299126354691043,
      "loss": 5.6463,
      "mean_token_accuracy": 0.17637302130460739,
      "num_tokens": 72674766.0,
      "step": 35865
    },
    {
      "entropy": 6.022907829284668,
      "epoch": 3.1346674823035916,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040296563414714444,
      "loss": 5.5329,
      "mean_token_accuracy": 0.18141630440950393,
      "num_tokens": 72683816.0,
      "step": 35870
    },
    {
      "entropy": 5.9550511837005615,
      "epoch": 3.1351044306562965,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004029400022929306,
      "loss": 5.4983,
      "mean_token_accuracy": 0.1852658435702324,
      "num_tokens": 72693604.0,
      "step": 35875
    },
    {
      "entropy": 5.954784822463989,
      "epoch": 3.135541379009001,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004029143679847604,
      "loss": 5.5541,
      "mean_token_accuracy": 0.18697417229413987,
      "num_tokens": 72703936.0,
      "step": 35880
    },
    {
      "entropy": 6.042789459228516,
      "epoch": 3.135978327361706,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004028887312231256,
      "loss": 5.5319,
      "mean_token_accuracy": 0.18602230548858642,
      "num_tokens": 72713646.0,
      "step": 35885
    },
    {
      "entropy": 6.076071310043335,
      "epoch": 3.1364152757144104,
      "grad_norm": 1.75,
      "learning_rate": 0.00040286309200851793,
      "loss": 5.5061,
      "mean_token_accuracy": 0.1826412484049797,
      "num_tokens": 72723573.0,
      "step": 35890
    },
    {
      "entropy": 5.990224599838257,
      "epoch": 3.1368522240671153,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004028374503414292,
      "loss": 5.5815,
      "mean_token_accuracy": 0.17931547313928603,
      "num_tokens": 72733805.0,
      "step": 35895
    },
    {
      "entropy": 5.95464448928833,
      "epoch": 3.1372891724198197,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004028118062223512,
      "loss": 5.5503,
      "mean_token_accuracy": 0.18141744136810303,
      "num_tokens": 72744254.0,
      "step": 35900
    },
    {
      "entropy": 6.086490631103516,
      "epoch": 3.1377261207725247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004027861596517757,
      "loss": 5.5971,
      "mean_token_accuracy": 0.17518599927425385,
      "num_tokens": 72754267.0,
      "step": 35905
    },
    {
      "entropy": 5.871933841705323,
      "epoch": 3.1381630691252296,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004027605106301948,
      "loss": 5.4671,
      "mean_token_accuracy": 0.20047194361686707,
      "num_tokens": 72764598.0,
      "step": 35910
    },
    {
      "entropy": 5.968994235992431,
      "epoch": 3.138600017477934,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040273485915810026,
      "loss": 5.5874,
      "mean_token_accuracy": 0.18245422095060349,
      "num_tokens": 72774145.0,
      "step": 35915
    },
    {
      "entropy": 6.083212661743164,
      "epoch": 3.139036965830639,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040270920523598425,
      "loss": 5.6677,
      "mean_token_accuracy": 0.170682954788208,
      "num_tokens": 72783766.0,
      "step": 35920
    },
    {
      "entropy": 6.045496940612793,
      "epoch": 3.1394739141833434,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004026835488643387,
      "loss": 5.5742,
      "mean_token_accuracy": 0.1910937711596489,
      "num_tokens": 72793907.0,
      "step": 35925
    },
    {
      "entropy": 6.062714910507202,
      "epoch": 3.1399108625360483,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040265789004365586,
      "loss": 5.6631,
      "mean_token_accuracy": 0.1721968650817871,
      "num_tokens": 72802927.0,
      "step": 35930
    },
    {
      "entropy": 5.915670394897461,
      "epoch": 3.140347810888753,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004026322287744277,
      "loss": 5.4196,
      "mean_token_accuracy": 0.1975184738636017,
      "num_tokens": 72813387.0,
      "step": 35935
    },
    {
      "entropy": 5.964169025421143,
      "epoch": 3.1407847592414577,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004026065650571466,
      "loss": 5.4896,
      "mean_token_accuracy": 0.18707786202430726,
      "num_tokens": 72824109.0,
      "step": 35940
    },
    {
      "entropy": 5.967561674118042,
      "epoch": 3.141221707594162,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004025808988923047,
      "loss": 5.5177,
      "mean_token_accuracy": 0.1826806604862213,
      "num_tokens": 72834107.0,
      "step": 35945
    },
    {
      "entropy": 6.046197748184204,
      "epoch": 3.141658655946867,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040255523028039433,
      "loss": 5.5495,
      "mean_token_accuracy": 0.1835218533873558,
      "num_tokens": 72843726.0,
      "step": 35950
    },
    {
      "entropy": 6.04792160987854,
      "epoch": 3.1420956042995716,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040252955922190776,
      "loss": 5.5965,
      "mean_token_accuracy": 0.1813612163066864,
      "num_tokens": 72854203.0,
      "step": 35955
    },
    {
      "entropy": 5.9715534210205075,
      "epoch": 3.1425325526522765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040250388571733744,
      "loss": 5.5582,
      "mean_token_accuracy": 0.17638446837663652,
      "num_tokens": 72864689.0,
      "step": 35960
    },
    {
      "entropy": 5.9699578285217285,
      "epoch": 3.1429695010049814,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004024782097671758,
      "loss": 5.576,
      "mean_token_accuracy": 0.18496580570936202,
      "num_tokens": 72875040.0,
      "step": 35965
    },
    {
      "entropy": 5.967297840118408,
      "epoch": 3.143406449357686,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004024525313719153,
      "loss": 5.5431,
      "mean_token_accuracy": 0.18466522693634033,
      "num_tokens": 72885174.0,
      "step": 35970
    },
    {
      "entropy": 5.967678117752075,
      "epoch": 3.143843397710391,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040242685053204844,
      "loss": 5.541,
      "mean_token_accuracy": 0.18879763185977935,
      "num_tokens": 72895344.0,
      "step": 35975
    },
    {
      "entropy": 6.05382924079895,
      "epoch": 3.1442803460630953,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004024011672480678,
      "loss": 5.6282,
      "mean_token_accuracy": 0.18091974258422852,
      "num_tokens": 72905950.0,
      "step": 35980
    },
    {
      "entropy": 6.012133312225342,
      "epoch": 3.1447172944158,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040237548152046603,
      "loss": 5.5405,
      "mean_token_accuracy": 0.18671734184026717,
      "num_tokens": 72916828.0,
      "step": 35985
    },
    {
      "entropy": 5.981035470962524,
      "epoch": 3.1451542427685046,
      "grad_norm": 1.25,
      "learning_rate": 0.00040234979334973577,
      "loss": 5.6198,
      "mean_token_accuracy": 0.1813381940126419,
      "num_tokens": 72926951.0,
      "step": 35990
    },
    {
      "entropy": 5.953158664703369,
      "epoch": 3.1455911911212096,
      "grad_norm": 2.125,
      "learning_rate": 0.0004023241027363698,
      "loss": 5.4885,
      "mean_token_accuracy": 0.18379459381103516,
      "num_tokens": 72937909.0,
      "step": 35995
    },
    {
      "entropy": 6.031801223754883,
      "epoch": 3.146028139473914,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004022984096808608,
      "loss": 5.5919,
      "mean_token_accuracy": 0.1782295361161232,
      "num_tokens": 72949898.0,
      "step": 36000
    },
    {
      "epoch": 3.146028139473914,
      "eval_entropy": 5.737647519349691,
      "eval_loss": 5.635207176208496,
      "eval_mean_token_accuracy": 0.1856179369493968,
      "eval_num_tokens": 72949898.0,
      "eval_runtime": 27.8939,
      "eval_samples_per_second": 1326.958,
      "eval_steps_per_second": 165.879,
      "step": 36000
    },
    {
      "entropy": 5.9885560989379885,
      "epoch": 3.146465087826619,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040227271418370156,
      "loss": 5.5973,
      "mean_token_accuracy": 0.18272272795438765,
      "num_tokens": 72959643.0,
      "step": 36005
    },
    {
      "entropy": 5.948122215270996,
      "epoch": 3.1469020361793234,
      "grad_norm": 1.0625,
      "learning_rate": 0.000402247016245385,
      "loss": 5.5744,
      "mean_token_accuracy": 0.17641631513834,
      "num_tokens": 72969745.0,
      "step": 36010
    },
    {
      "entropy": 6.029787302017212,
      "epoch": 3.1473389845320283,
      "grad_norm": 1.5625,
      "learning_rate": 0.00040222131586640397,
      "loss": 5.5369,
      "mean_token_accuracy": 0.18408095687627793,
      "num_tokens": 72979892.0,
      "step": 36015
    },
    {
      "entropy": 5.947958087921142,
      "epoch": 3.1477759328847332,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040219561304725147,
      "loss": 5.5467,
      "mean_token_accuracy": 0.18572222590446472,
      "num_tokens": 72989526.0,
      "step": 36020
    },
    {
      "entropy": 5.898528528213501,
      "epoch": 3.1482128812374377,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040216990778842046,
      "loss": 5.5795,
      "mean_token_accuracy": 0.18420538902282715,
      "num_tokens": 72999771.0,
      "step": 36025
    },
    {
      "entropy": 6.05102653503418,
      "epoch": 3.1486498295901426,
      "grad_norm": 1.125,
      "learning_rate": 0.00040214420009040394,
      "loss": 5.5898,
      "mean_token_accuracy": 0.17691615372896194,
      "num_tokens": 73011602.0,
      "step": 36030
    },
    {
      "entropy": 6.095385932922364,
      "epoch": 3.149086777942847,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000402118489953695,
      "loss": 5.5855,
      "mean_token_accuracy": 0.1850763440132141,
      "num_tokens": 73021838.0,
      "step": 36035
    },
    {
      "entropy": 6.004723834991455,
      "epoch": 3.149523726295552,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040209277737878683,
      "loss": 5.5663,
      "mean_token_accuracy": 0.1803848534822464,
      "num_tokens": 73032242.0,
      "step": 36040
    },
    {
      "entropy": 5.9963178634643555,
      "epoch": 3.1499606746482565,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004020670623661727,
      "loss": 5.6381,
      "mean_token_accuracy": 0.1746555432677269,
      "num_tokens": 73042903.0,
      "step": 36045
    },
    {
      "entropy": 5.976078844070434,
      "epoch": 3.1503976230009614,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004020413449163457,
      "loss": 5.5209,
      "mean_token_accuracy": 0.18548427969217302,
      "num_tokens": 73054083.0,
      "step": 36050
    },
    {
      "entropy": 6.159701871871948,
      "epoch": 3.150834571353666,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004020156250297991,
      "loss": 5.6173,
      "mean_token_accuracy": 0.1719298005104065,
      "num_tokens": 73064215.0,
      "step": 36055
    },
    {
      "entropy": 5.942703104019165,
      "epoch": 3.1512715197063708,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004019899027070262,
      "loss": 5.5114,
      "mean_token_accuracy": 0.1929589182138443,
      "num_tokens": 73074319.0,
      "step": 36060
    },
    {
      "entropy": 5.939399576187133,
      "epoch": 3.1517084680590752,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004019641779485206,
      "loss": 5.6089,
      "mean_token_accuracy": 0.1782762050628662,
      "num_tokens": 73083972.0,
      "step": 36065
    },
    {
      "entropy": 6.008944416046143,
      "epoch": 3.15214541641178,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040193845075477537,
      "loss": 5.5934,
      "mean_token_accuracy": 0.1797328308224678,
      "num_tokens": 73094495.0,
      "step": 36070
    },
    {
      "entropy": 5.969381380081177,
      "epoch": 3.152582364764485,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004019127211262842,
      "loss": 5.4493,
      "mean_token_accuracy": 0.19161344319581985,
      "num_tokens": 73104355.0,
      "step": 36075
    },
    {
      "entropy": 5.955718469619751,
      "epoch": 3.1530193131171895,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040188698906354055,
      "loss": 5.5609,
      "mean_token_accuracy": 0.17781177908182144,
      "num_tokens": 73115795.0,
      "step": 36080
    },
    {
      "entropy": 5.989797878265381,
      "epoch": 3.1534562614698944,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004018612545670379,
      "loss": 5.4656,
      "mean_token_accuracy": 0.1921807572245598,
      "num_tokens": 73125351.0,
      "step": 36085
    },
    {
      "entropy": 5.938551139831543,
      "epoch": 3.153893209822599,
      "grad_norm": 2.03125,
      "learning_rate": 0.00040183551763727,
      "loss": 5.5395,
      "mean_token_accuracy": 0.18547942787408828,
      "num_tokens": 73135087.0,
      "step": 36090
    },
    {
      "entropy": 5.923488187789917,
      "epoch": 3.154330158175304,
      "grad_norm": 1.734375,
      "learning_rate": 0.00040180977827473035,
      "loss": 5.539,
      "mean_token_accuracy": 0.1757095381617546,
      "num_tokens": 73145382.0,
      "step": 36095
    },
    {
      "entropy": 6.041371583938599,
      "epoch": 3.1547671065280083,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040178403647991276,
      "loss": 5.6572,
      "mean_token_accuracy": 0.1691105082631111,
      "num_tokens": 73155689.0,
      "step": 36100
    },
    {
      "entropy": 6.106725120544434,
      "epoch": 3.155204054880713,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004017582922533109,
      "loss": 5.5902,
      "mean_token_accuracy": 0.17550546526908875,
      "num_tokens": 73165803.0,
      "step": 36105
    },
    {
      "entropy": 6.030052995681762,
      "epoch": 3.1556410032334177,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040173254559541853,
      "loss": 5.5782,
      "mean_token_accuracy": 0.18439080566167831,
      "num_tokens": 73176029.0,
      "step": 36110
    },
    {
      "entropy": 5.999540090560913,
      "epoch": 3.1560779515861226,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004017067965067296,
      "loss": 5.5877,
      "mean_token_accuracy": 0.18250850439071656,
      "num_tokens": 73187708.0,
      "step": 36115
    },
    {
      "entropy": 5.99883451461792,
      "epoch": 3.156514899938827,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004016810449877379,
      "loss": 5.5625,
      "mean_token_accuracy": 0.1773655444383621,
      "num_tokens": 73197458.0,
      "step": 36120
    },
    {
      "entropy": 6.011194276809692,
      "epoch": 3.156951848291532,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004016552910389373,
      "loss": 5.5715,
      "mean_token_accuracy": 0.1802339509129524,
      "num_tokens": 73206931.0,
      "step": 36125
    },
    {
      "entropy": 5.987840461730957,
      "epoch": 3.157388796644237,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040162953466082194,
      "loss": 5.5558,
      "mean_token_accuracy": 0.18421462178230286,
      "num_tokens": 73217048.0,
      "step": 36130
    },
    {
      "entropy": 5.9661789894104,
      "epoch": 3.1578257449969414,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004016037758538857,
      "loss": 5.5573,
      "mean_token_accuracy": 0.1787792772054672,
      "num_tokens": 73226649.0,
      "step": 36135
    },
    {
      "entropy": 5.973288345336914,
      "epoch": 3.1582626933496463,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004015780146186227,
      "loss": 5.4908,
      "mean_token_accuracy": 0.18609416633844375,
      "num_tokens": 73236822.0,
      "step": 36140
    },
    {
      "entropy": 6.08447961807251,
      "epoch": 3.1586996417023507,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004015522509555271,
      "loss": 5.6715,
      "mean_token_accuracy": 0.17554232627153396,
      "num_tokens": 73247078.0,
      "step": 36145
    },
    {
      "entropy": 5.895925760269165,
      "epoch": 3.1591365900550556,
      "grad_norm": 1.046875,
      "learning_rate": 0.000401526484865093,
      "loss": 5.3821,
      "mean_token_accuracy": 0.1980715349316597,
      "num_tokens": 73257001.0,
      "step": 36150
    },
    {
      "entropy": 5.919516706466675,
      "epoch": 3.15957353840776,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004015007163478146,
      "loss": 5.4674,
      "mean_token_accuracy": 0.1858864203095436,
      "num_tokens": 73266922.0,
      "step": 36155
    },
    {
      "entropy": 5.965799760818482,
      "epoch": 3.160010486760465,
      "grad_norm": 0.96875,
      "learning_rate": 0.00040147494540418617,
      "loss": 5.5401,
      "mean_token_accuracy": 0.18400923162698746,
      "num_tokens": 73277858.0,
      "step": 36160
    },
    {
      "entropy": 5.969141674041748,
      "epoch": 3.1604474351131695,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040144917203470203,
      "loss": 5.5564,
      "mean_token_accuracy": 0.18132039159536362,
      "num_tokens": 73287361.0,
      "step": 36165
    },
    {
      "entropy": 5.994181728363037,
      "epoch": 3.1608843834658744,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004014233962398565,
      "loss": 5.4886,
      "mean_token_accuracy": 0.18437778055667878,
      "num_tokens": 73296521.0,
      "step": 36170
    },
    {
      "entropy": 5.922270154953003,
      "epoch": 3.161321331818579,
      "grad_norm": 1.03125,
      "learning_rate": 0.000401397618020144,
      "loss": 5.5308,
      "mean_token_accuracy": 0.18767091482877732,
      "num_tokens": 73306257.0,
      "step": 36175
    },
    {
      "entropy": 5.936605834960938,
      "epoch": 3.161758280171284,
      "grad_norm": 1.0625,
      "learning_rate": 0.000401371837376059,
      "loss": 5.4653,
      "mean_token_accuracy": 0.19087254554033278,
      "num_tokens": 73315939.0,
      "step": 36180
    },
    {
      "entropy": 5.9283942699432375,
      "epoch": 3.1621952285239883,
      "grad_norm": 1.125,
      "learning_rate": 0.0004013460543080959,
      "loss": 5.5157,
      "mean_token_accuracy": 0.1840816169977188,
      "num_tokens": 73326407.0,
      "step": 36185
    },
    {
      "entropy": 6.0025365352630615,
      "epoch": 3.162632176876693,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004013202688167493,
      "loss": 5.5751,
      "mean_token_accuracy": 0.17884917855262755,
      "num_tokens": 73336188.0,
      "step": 36190
    },
    {
      "entropy": 6.071408224105835,
      "epoch": 3.163069125229398,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004012944809025136,
      "loss": 5.5775,
      "mean_token_accuracy": 0.18075229376554489,
      "num_tokens": 73346824.0,
      "step": 36195
    },
    {
      "entropy": 6.002386665344238,
      "epoch": 3.1635060735821026,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040126869056588375,
      "loss": 5.5466,
      "mean_token_accuracy": 0.18149864971637725,
      "num_tokens": 73356688.0,
      "step": 36200
    },
    {
      "entropy": 5.9981669902801515,
      "epoch": 3.1639430219348075,
      "grad_norm": 1.125,
      "learning_rate": 0.0004012428978073543,
      "loss": 5.6307,
      "mean_token_accuracy": 0.17448274195194244,
      "num_tokens": 73366905.0,
      "step": 36205
    },
    {
      "entropy": 6.092702102661133,
      "epoch": 3.164379970287512,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040121710262741974,
      "loss": 5.622,
      "mean_token_accuracy": 0.17430318593978883,
      "num_tokens": 73376870.0,
      "step": 36210
    },
    {
      "entropy": 6.010173225402832,
      "epoch": 3.164816918640217,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004011913050265751,
      "loss": 5.4984,
      "mean_token_accuracy": 0.1814899891614914,
      "num_tokens": 73386153.0,
      "step": 36215
    },
    {
      "entropy": 5.9357123374938965,
      "epoch": 3.1652538669929213,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040116550500531506,
      "loss": 5.5767,
      "mean_token_accuracy": 0.1809008851647377,
      "num_tokens": 73395598.0,
      "step": 36220
    },
    {
      "entropy": 6.017400741577148,
      "epoch": 3.1656908153456262,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040113970256413465,
      "loss": 5.5297,
      "mean_token_accuracy": 0.1786169156432152,
      "num_tokens": 73406883.0,
      "step": 36225
    },
    {
      "entropy": 6.004170608520508,
      "epoch": 3.1661277636983307,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040111389770352846,
      "loss": 5.568,
      "mean_token_accuracy": 0.1880134105682373,
      "num_tokens": 73418504.0,
      "step": 36230
    },
    {
      "entropy": 5.962109375,
      "epoch": 3.1665647120510356,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040108809042399183,
      "loss": 5.5806,
      "mean_token_accuracy": 0.17856378108263016,
      "num_tokens": 73429228.0,
      "step": 36235
    },
    {
      "entropy": 5.995914697647095,
      "epoch": 3.16700166040374,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004010622807260194,
      "loss": 5.5311,
      "mean_token_accuracy": 0.1849777266383171,
      "num_tokens": 73439269.0,
      "step": 36240
    },
    {
      "entropy": 5.985223579406738,
      "epoch": 3.167438608756445,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004010364686101064,
      "loss": 5.5646,
      "mean_token_accuracy": 0.18069496154785156,
      "num_tokens": 73449554.0,
      "step": 36245
    },
    {
      "entropy": 5.912935972213745,
      "epoch": 3.1678755571091495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000401010654076748,
      "loss": 5.5095,
      "mean_token_accuracy": 0.19155547618865967,
      "num_tokens": 73460453.0,
      "step": 36250
    },
    {
      "entropy": 5.992454147338867,
      "epoch": 3.1683125054618544,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004009848371264391,
      "loss": 5.4731,
      "mean_token_accuracy": 0.18648838996887207,
      "num_tokens": 73469905.0,
      "step": 36255
    },
    {
      "entropy": 5.974794387817383,
      "epoch": 3.1687494538145593,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040095901775967507,
      "loss": 5.6233,
      "mean_token_accuracy": 0.1809692367911339,
      "num_tokens": 73480363.0,
      "step": 36260
    },
    {
      "entropy": 5.985497951507568,
      "epoch": 3.1691864021672638,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040093319597695105,
      "loss": 5.5758,
      "mean_token_accuracy": 0.18255967199802398,
      "num_tokens": 73490080.0,
      "step": 36265
    },
    {
      "entropy": 6.107400751113891,
      "epoch": 3.1696233505199687,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004009073717787624,
      "loss": 5.626,
      "mean_token_accuracy": 0.17241345345973969,
      "num_tokens": 73500464.0,
      "step": 36270
    },
    {
      "entropy": 5.972395610809326,
      "epoch": 3.170060298872673,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004008815451656043,
      "loss": 5.5317,
      "mean_token_accuracy": 0.18596585243940353,
      "num_tokens": 73509834.0,
      "step": 36275
    },
    {
      "entropy": 5.951634693145752,
      "epoch": 3.170497247225378,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040085571613797226,
      "loss": 5.6572,
      "mean_token_accuracy": 0.17066467553377151,
      "num_tokens": 73520345.0,
      "step": 36280
    },
    {
      "entropy": 6.036613273620605,
      "epoch": 3.1709341955780825,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004008298846963617,
      "loss": 5.605,
      "mean_token_accuracy": 0.18196238428354264,
      "num_tokens": 73530769.0,
      "step": 36285
    },
    {
      "entropy": 6.053609037399292,
      "epoch": 3.1713711439307875,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004008040508412678,
      "loss": 5.5329,
      "mean_token_accuracy": 0.18155740946531296,
      "num_tokens": 73539989.0,
      "step": 36290
    },
    {
      "entropy": 6.010236024856567,
      "epoch": 3.171808092283492,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004007782145731865,
      "loss": 5.5868,
      "mean_token_accuracy": 0.17643952816724778,
      "num_tokens": 73551344.0,
      "step": 36295
    },
    {
      "entropy": 5.982710075378418,
      "epoch": 3.172245040636197,
      "grad_norm": 1.25,
      "learning_rate": 0.00040075237589261303,
      "loss": 5.4821,
      "mean_token_accuracy": 0.18872709423303605,
      "num_tokens": 73560580.0,
      "step": 36300
    },
    {
      "entropy": 5.955947399139404,
      "epoch": 3.1726819889889013,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004007265348000432,
      "loss": 5.5289,
      "mean_token_accuracy": 0.17655061036348343,
      "num_tokens": 73569500.0,
      "step": 36305
    },
    {
      "entropy": 6.086483144760132,
      "epoch": 3.173118937341606,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040070069129597243,
      "loss": 5.6134,
      "mean_token_accuracy": 0.17376304119825364,
      "num_tokens": 73578658.0,
      "step": 36310
    },
    {
      "entropy": 6.019506931304932,
      "epoch": 3.173555885694311,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040067484538089653,
      "loss": 5.5302,
      "mean_token_accuracy": 0.18526704162359237,
      "num_tokens": 73588842.0,
      "step": 36315
    },
    {
      "entropy": 6.00553150177002,
      "epoch": 3.1739928340470156,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004006489970553114,
      "loss": 5.6044,
      "mean_token_accuracy": 0.1810029089450836,
      "num_tokens": 73598962.0,
      "step": 36320
    },
    {
      "entropy": 6.032311582565308,
      "epoch": 3.1744297823997205,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040062314631971255,
      "loss": 5.6184,
      "mean_token_accuracy": 0.17877379059791565,
      "num_tokens": 73609890.0,
      "step": 36325
    },
    {
      "entropy": 5.973724699020385,
      "epoch": 3.174866730752425,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004005972931745959,
      "loss": 5.5086,
      "mean_token_accuracy": 0.18664756119251252,
      "num_tokens": 73620453.0,
      "step": 36330
    },
    {
      "entropy": 6.027878952026367,
      "epoch": 3.17530367910513,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040057143762045746,
      "loss": 5.5675,
      "mean_token_accuracy": 0.18421215564012527,
      "num_tokens": 73630385.0,
      "step": 36335
    },
    {
      "entropy": 5.992668724060058,
      "epoch": 3.1757406274578344,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040054557965779294,
      "loss": 5.5753,
      "mean_token_accuracy": 0.17832280993461608,
      "num_tokens": 73640081.0,
      "step": 36340
    },
    {
      "entropy": 5.9806873321533205,
      "epoch": 3.1761775758105393,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040051971928709853,
      "loss": 5.5386,
      "mean_token_accuracy": 0.18129930198192595,
      "num_tokens": 73650934.0,
      "step": 36345
    },
    {
      "entropy": 5.977779531478882,
      "epoch": 3.1766145241632437,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040049385650887004,
      "loss": 5.544,
      "mean_token_accuracy": 0.182547889649868,
      "num_tokens": 73661835.0,
      "step": 36350
    },
    {
      "entropy": 6.024565887451172,
      "epoch": 3.1770514725159487,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004004679913236036,
      "loss": 5.6051,
      "mean_token_accuracy": 0.18009078055620192,
      "num_tokens": 73671738.0,
      "step": 36355
    },
    {
      "entropy": 6.015345859527588,
      "epoch": 3.177488420868653,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040044212373179547,
      "loss": 5.5912,
      "mean_token_accuracy": 0.17778275161981583,
      "num_tokens": 73681834.0,
      "step": 36360
    },
    {
      "entropy": 5.967773580551148,
      "epoch": 3.177925369221358,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040041625373394153,
      "loss": 5.523,
      "mean_token_accuracy": 0.18510178327560425,
      "num_tokens": 73692137.0,
      "step": 36365
    },
    {
      "entropy": 6.005325698852539,
      "epoch": 3.178362317574063,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004003903813305383,
      "loss": 5.5934,
      "mean_token_accuracy": 0.17830158323049544,
      "num_tokens": 73702263.0,
      "step": 36370
    },
    {
      "entropy": 5.993245363235474,
      "epoch": 3.1787992659267674,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004003645065220817,
      "loss": 5.5969,
      "mean_token_accuracy": 0.1741851195693016,
      "num_tokens": 73711945.0,
      "step": 36375
    },
    {
      "entropy": 6.130704879760742,
      "epoch": 3.1792362142794723,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004003386293090682,
      "loss": 5.645,
      "mean_token_accuracy": 0.17846589237451554,
      "num_tokens": 73722205.0,
      "step": 36380
    },
    {
      "entropy": 6.064955472946167,
      "epoch": 3.179673162632177,
      "grad_norm": 1.25,
      "learning_rate": 0.0004003127496919942,
      "loss": 5.6066,
      "mean_token_accuracy": 0.1789669007062912,
      "num_tokens": 73732049.0,
      "step": 36385
    },
    {
      "entropy": 5.944052410125733,
      "epoch": 3.1801101109848817,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004002868676713559,
      "loss": 5.5931,
      "mean_token_accuracy": 0.17834636867046355,
      "num_tokens": 73741559.0,
      "step": 36390
    },
    {
      "entropy": 6.05268235206604,
      "epoch": 3.180547059337586,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040026098324764985,
      "loss": 5.6831,
      "mean_token_accuracy": 0.1731008842587471,
      "num_tokens": 73751576.0,
      "step": 36395
    },
    {
      "entropy": 6.077070856094361,
      "epoch": 3.180984007690291,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004002350964213725,
      "loss": 5.6653,
      "mean_token_accuracy": 0.17623185962438584,
      "num_tokens": 73762155.0,
      "step": 36400
    },
    {
      "entropy": 5.965621566772461,
      "epoch": 3.1814209560429956,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004002092071930204,
      "loss": 5.5379,
      "mean_token_accuracy": 0.18139581084251405,
      "num_tokens": 73772427.0,
      "step": 36405
    },
    {
      "entropy": 6.005879974365234,
      "epoch": 3.1818579043957005,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004001833155630901,
      "loss": 5.5176,
      "mean_token_accuracy": 0.19008628129959107,
      "num_tokens": 73782868.0,
      "step": 36410
    },
    {
      "entropy": 5.956051254272461,
      "epoch": 3.182294852748405,
      "grad_norm": 1.0,
      "learning_rate": 0.00040015742153207823,
      "loss": 5.5217,
      "mean_token_accuracy": 0.19041467905044557,
      "num_tokens": 73792480.0,
      "step": 36415
    },
    {
      "entropy": 6.039970874786377,
      "epoch": 3.18273180110111,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004001315251004814,
      "loss": 5.5843,
      "mean_token_accuracy": 0.17648855745792388,
      "num_tokens": 73803264.0,
      "step": 36420
    },
    {
      "entropy": 6.012645435333252,
      "epoch": 3.183168749453815,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040010562626879636,
      "loss": 5.451,
      "mean_token_accuracy": 0.19325510859489442,
      "num_tokens": 73812491.0,
      "step": 36425
    },
    {
      "entropy": 5.934329986572266,
      "epoch": 3.1836056978065193,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040007972503751984,
      "loss": 5.5082,
      "mean_token_accuracy": 0.18342878818511962,
      "num_tokens": 73821672.0,
      "step": 36430
    },
    {
      "entropy": 5.961886978149414,
      "epoch": 3.184042646159224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040005382140714867,
      "loss": 5.5119,
      "mean_token_accuracy": 0.1857929363846779,
      "num_tokens": 73831709.0,
      "step": 36435
    },
    {
      "entropy": 5.981790018081665,
      "epoch": 3.1844795945119286,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004000279153781797,
      "loss": 5.5596,
      "mean_token_accuracy": 0.18043217808008194,
      "num_tokens": 73841121.0,
      "step": 36440
    },
    {
      "entropy": 6.012610149383545,
      "epoch": 3.1849165428646335,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040000200695110977,
      "loss": 5.5991,
      "mean_token_accuracy": 0.1759676530957222,
      "num_tokens": 73851299.0,
      "step": 36445
    },
    {
      "entropy": 6.1186842918396,
      "epoch": 3.185353491217338,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003999760961264359,
      "loss": 5.5989,
      "mean_token_accuracy": 0.1773848533630371,
      "num_tokens": 73862020.0,
      "step": 36450
    },
    {
      "entropy": 5.9682105541229244,
      "epoch": 3.185790439570043,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039995018290465495,
      "loss": 5.5634,
      "mean_token_accuracy": 0.17954340279102327,
      "num_tokens": 73872786.0,
      "step": 36455
    },
    {
      "entropy": 5.950773096084594,
      "epoch": 3.1862273879227474,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003999242672862641,
      "loss": 5.4979,
      "mean_token_accuracy": 0.18288700580596923,
      "num_tokens": 73882303.0,
      "step": 36460
    },
    {
      "entropy": 5.96838755607605,
      "epoch": 3.1866643362754523,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003998983492717602,
      "loss": 5.5073,
      "mean_token_accuracy": 0.18646086156368255,
      "num_tokens": 73892124.0,
      "step": 36465
    },
    {
      "entropy": 5.993217182159424,
      "epoch": 3.187101284628157,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003998724288616406,
      "loss": 5.5231,
      "mean_token_accuracy": 0.1858457460999489,
      "num_tokens": 73901803.0,
      "step": 36470
    },
    {
      "entropy": 6.008458852767944,
      "epoch": 3.1875382329808617,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039984650605640244,
      "loss": 5.4572,
      "mean_token_accuracy": 0.1910540908575058,
      "num_tokens": 73911767.0,
      "step": 36475
    },
    {
      "entropy": 6.003986597061157,
      "epoch": 3.187975181333566,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003998205808565428,
      "loss": 5.6296,
      "mean_token_accuracy": 0.178010693192482,
      "num_tokens": 73921746.0,
      "step": 36480
    },
    {
      "entropy": 5.955504131317139,
      "epoch": 3.188412129686271,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039979465326255904,
      "loss": 5.5198,
      "mean_token_accuracy": 0.1877928912639618,
      "num_tokens": 73932282.0,
      "step": 36485
    },
    {
      "entropy": 6.050834655761719,
      "epoch": 3.188849078038976,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003997687232749484,
      "loss": 5.6195,
      "mean_token_accuracy": 0.1735095664858818,
      "num_tokens": 73942739.0,
      "step": 36490
    },
    {
      "entropy": 6.009697580337525,
      "epoch": 3.1892860263916805,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039974279089420833,
      "loss": 5.5631,
      "mean_token_accuracy": 0.18218289017677308,
      "num_tokens": 73953739.0,
      "step": 36495
    },
    {
      "entropy": 6.005862808227539,
      "epoch": 3.1897229747443854,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039971685612083616,
      "loss": 5.5877,
      "mean_token_accuracy": 0.17704476416110992,
      "num_tokens": 73963808.0,
      "step": 36500
    },
    {
      "entropy": 6.051014947891235,
      "epoch": 3.19015992309709,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039969091895532933,
      "loss": 5.6336,
      "mean_token_accuracy": 0.17785099446773528,
      "num_tokens": 73973083.0,
      "step": 36505
    },
    {
      "entropy": 5.884251260757447,
      "epoch": 3.1905968714497948,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003996649793981853,
      "loss": 5.4641,
      "mean_token_accuracy": 0.19623491019010544,
      "num_tokens": 73983966.0,
      "step": 36510
    },
    {
      "entropy": 6.071758604049682,
      "epoch": 3.1910338198024992,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039963903744990166,
      "loss": 5.6023,
      "mean_token_accuracy": 0.17996898144483567,
      "num_tokens": 73993874.0,
      "step": 36515
    },
    {
      "entropy": 5.9607460498809814,
      "epoch": 3.191470768155204,
      "grad_norm": 1.40625,
      "learning_rate": 0.000399613093110976,
      "loss": 5.5885,
      "mean_token_accuracy": 0.17312668412923812,
      "num_tokens": 74004179.0,
      "step": 36520
    },
    {
      "entropy": 5.981183958053589,
      "epoch": 3.1919077165079086,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003995871463819059,
      "loss": 5.5641,
      "mean_token_accuracy": 0.18378947228193282,
      "num_tokens": 74014941.0,
      "step": 36525
    },
    {
      "entropy": 6.056381702423096,
      "epoch": 3.1923446648606135,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039956119726318907,
      "loss": 5.6473,
      "mean_token_accuracy": 0.1706016778945923,
      "num_tokens": 74026620.0,
      "step": 36530
    },
    {
      "entropy": 6.025672245025635,
      "epoch": 3.192781613213318,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039953524575532317,
      "loss": 5.5889,
      "mean_token_accuracy": 0.17685885578393937,
      "num_tokens": 74036341.0,
      "step": 36535
    },
    {
      "entropy": 6.025006628036499,
      "epoch": 3.193218561566023,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039950929185880604,
      "loss": 5.5431,
      "mean_token_accuracy": 0.18362044990062715,
      "num_tokens": 74046884.0,
      "step": 36540
    },
    {
      "entropy": 5.956268835067749,
      "epoch": 3.1936555099187274,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003994833355741354,
      "loss": 5.5124,
      "mean_token_accuracy": 0.18383831679821014,
      "num_tokens": 74056706.0,
      "step": 36545
    },
    {
      "entropy": 5.7815062522888185,
      "epoch": 3.1940924582714323,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039945737690180917,
      "loss": 5.4164,
      "mean_token_accuracy": 0.2054467022418976,
      "num_tokens": 74067046.0,
      "step": 36550
    },
    {
      "entropy": 6.024000787734986,
      "epoch": 3.194529406624137,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003994314158423252,
      "loss": 5.5287,
      "mean_token_accuracy": 0.1807890295982361,
      "num_tokens": 74076884.0,
      "step": 36555
    },
    {
      "entropy": 5.992914295196533,
      "epoch": 3.1949663549768417,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003994054523961816,
      "loss": 5.5905,
      "mean_token_accuracy": 0.17640287727117537,
      "num_tokens": 74086482.0,
      "step": 36560
    },
    {
      "entropy": 5.9260711669921875,
      "epoch": 3.1954033033295466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039937948656387616,
      "loss": 5.4886,
      "mean_token_accuracy": 0.1871662512421608,
      "num_tokens": 74095916.0,
      "step": 36565
    },
    {
      "entropy": 6.027384042739868,
      "epoch": 3.195840251682251,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039935351834590696,
      "loss": 5.6166,
      "mean_token_accuracy": 0.1797584593296051,
      "num_tokens": 74106054.0,
      "step": 36570
    },
    {
      "entropy": 5.960271787643433,
      "epoch": 3.196277200034956,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039932754774277214,
      "loss": 5.5344,
      "mean_token_accuracy": 0.18956379890441893,
      "num_tokens": 74115641.0,
      "step": 36575
    },
    {
      "entropy": 5.9967128276824955,
      "epoch": 3.1967141483876604,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003993015747549698,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1843174010515213,
      "num_tokens": 74125370.0,
      "step": 36580
    },
    {
      "entropy": 6.009608173370362,
      "epoch": 3.1971510967403654,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039927559938299814,
      "loss": 5.5457,
      "mean_token_accuracy": 0.18659325391054155,
      "num_tokens": 74134959.0,
      "step": 36585
    },
    {
      "entropy": 5.977586889266968,
      "epoch": 3.19758804509307,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003992496216273554,
      "loss": 5.6641,
      "mean_token_accuracy": 0.17171211689710617,
      "num_tokens": 74144939.0,
      "step": 36590
    },
    {
      "entropy": 6.009849166870117,
      "epoch": 3.1980249934457747,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039922364148853973,
      "loss": 5.4527,
      "mean_token_accuracy": 0.19198439866304398,
      "num_tokens": 74154336.0,
      "step": 36595
    },
    {
      "entropy": 5.970107650756836,
      "epoch": 3.198461941798479,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003991976589670496,
      "loss": 5.5607,
      "mean_token_accuracy": 0.17969662696123123,
      "num_tokens": 74164022.0,
      "step": 36600
    },
    {
      "entropy": 6.051518440246582,
      "epoch": 3.198898890151184,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003991716740633832,
      "loss": 5.543,
      "mean_token_accuracy": 0.18018641024827958,
      "num_tokens": 74174186.0,
      "step": 36605
    },
    {
      "entropy": 5.965350723266601,
      "epoch": 3.199335838503889,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003991456867780391,
      "loss": 5.5396,
      "mean_token_accuracy": 0.18494074940681457,
      "num_tokens": 74183913.0,
      "step": 36610
    },
    {
      "entropy": 5.985903692245484,
      "epoch": 3.1997727868565935,
      "grad_norm": 1.125,
      "learning_rate": 0.0003991196971115157,
      "loss": 5.6024,
      "mean_token_accuracy": 0.17910590171813964,
      "num_tokens": 74194307.0,
      "step": 36615
    },
    {
      "entropy": 6.040549087524414,
      "epoch": 3.2002097352092984,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039909370506431146,
      "loss": 5.5645,
      "mean_token_accuracy": 0.17754233926534652,
      "num_tokens": 74204672.0,
      "step": 36620
    },
    {
      "entropy": 6.037330484390258,
      "epoch": 3.200646683562003,
      "grad_norm": 1.0,
      "learning_rate": 0.0003990677106369249,
      "loss": 5.5554,
      "mean_token_accuracy": 0.1815015360713005,
      "num_tokens": 74214022.0,
      "step": 36625
    },
    {
      "entropy": 5.8946398258209225,
      "epoch": 3.201083631914708,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039904171382985465,
      "loss": 5.4911,
      "mean_token_accuracy": 0.18456245809793473,
      "num_tokens": 74223736.0,
      "step": 36630
    },
    {
      "entropy": 5.977530241012573,
      "epoch": 3.2015205802674123,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003990157146435993,
      "loss": 5.553,
      "mean_token_accuracy": 0.18416628688573838,
      "num_tokens": 74234742.0,
      "step": 36635
    },
    {
      "entropy": 6.077485466003418,
      "epoch": 3.201957528620117,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003989897130786577,
      "loss": 5.5375,
      "mean_token_accuracy": 0.1848514512181282,
      "num_tokens": 74244949.0,
      "step": 36640
    },
    {
      "entropy": 6.0302129745483395,
      "epoch": 3.2023944769728216,
      "grad_norm": 1.765625,
      "learning_rate": 0.00039896370913552825,
      "loss": 5.5927,
      "mean_token_accuracy": 0.17878569811582565,
      "num_tokens": 74255280.0,
      "step": 36645
    },
    {
      "entropy": 5.806890249252319,
      "epoch": 3.2028314253255266,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039893770281471005,
      "loss": 5.4369,
      "mean_token_accuracy": 0.19778584986925124,
      "num_tokens": 74265615.0,
      "step": 36650
    },
    {
      "entropy": 5.938663196563721,
      "epoch": 3.203268373678231,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003989116941167016,
      "loss": 5.5729,
      "mean_token_accuracy": 0.18714021295309066,
      "num_tokens": 74275791.0,
      "step": 36655
    },
    {
      "entropy": 6.0826451778411865,
      "epoch": 3.203705322030936,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039888568304200206,
      "loss": 5.5565,
      "mean_token_accuracy": 0.1772422194480896,
      "num_tokens": 74286295.0,
      "step": 36660
    },
    {
      "entropy": 5.95123815536499,
      "epoch": 3.204142270383641,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039885966959111015,
      "loss": 5.4946,
      "mean_token_accuracy": 0.18797433525323867,
      "num_tokens": 74295163.0,
      "step": 36665
    },
    {
      "entropy": 5.92780647277832,
      "epoch": 3.2045792187363453,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003988336537645249,
      "loss": 5.538,
      "mean_token_accuracy": 0.18042973428964615,
      "num_tokens": 74304855.0,
      "step": 36670
    },
    {
      "entropy": 6.024414730072022,
      "epoch": 3.2050161670890502,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003988076355627453,
      "loss": 5.5684,
      "mean_token_accuracy": 0.1783779889345169,
      "num_tokens": 74314584.0,
      "step": 36675
    },
    {
      "entropy": 5.9888636589050295,
      "epoch": 3.2054531154417547,
      "grad_norm": 1.125,
      "learning_rate": 0.0003987816149862703,
      "loss": 5.558,
      "mean_token_accuracy": 0.1820295497775078,
      "num_tokens": 74324992.0,
      "step": 36680
    },
    {
      "entropy": 5.938571834564209,
      "epoch": 3.2058900637944596,
      "grad_norm": 1.125,
      "learning_rate": 0.00039875559203559913,
      "loss": 5.5102,
      "mean_token_accuracy": 0.18283575028181076,
      "num_tokens": 74336180.0,
      "step": 36685
    },
    {
      "entropy": 5.94270076751709,
      "epoch": 3.206327012147164,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003987295667112309,
      "loss": 5.4283,
      "mean_token_accuracy": 0.19356254935264589,
      "num_tokens": 74346073.0,
      "step": 36690
    },
    {
      "entropy": 5.982026147842407,
      "epoch": 3.206763960499869,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003987035390136647,
      "loss": 5.5009,
      "mean_token_accuracy": 0.18755156695842742,
      "num_tokens": 74356470.0,
      "step": 36695
    },
    {
      "entropy": 5.937069034576416,
      "epoch": 3.2072009088525735,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039867750894339974,
      "loss": 5.4892,
      "mean_token_accuracy": 0.18871806114912032,
      "num_tokens": 74367035.0,
      "step": 36700
    },
    {
      "entropy": 5.9436366081237795,
      "epoch": 3.2076378572052784,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003986514765009354,
      "loss": 5.5286,
      "mean_token_accuracy": 0.18713961988687516,
      "num_tokens": 74377264.0,
      "step": 36705
    },
    {
      "entropy": 6.07653112411499,
      "epoch": 3.208074805557983,
      "grad_norm": 1.421875,
      "learning_rate": 0.00039862544168677103,
      "loss": 5.653,
      "mean_token_accuracy": 0.16934690177440642,
      "num_tokens": 74389053.0,
      "step": 36710
    },
    {
      "entropy": 6.057690954208374,
      "epoch": 3.2085117539106878,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003985994045014058,
      "loss": 5.5489,
      "mean_token_accuracy": 0.18691090494394302,
      "num_tokens": 74399546.0,
      "step": 36715
    },
    {
      "entropy": 5.958754253387451,
      "epoch": 3.2089487022633927,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00039857336494533924,
      "loss": 5.5526,
      "mean_token_accuracy": 0.18488425612449647,
      "num_tokens": 74409905.0,
      "step": 36720
    },
    {
      "entropy": 5.924856758117675,
      "epoch": 3.209385650616097,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003985473230190708,
      "loss": 5.5274,
      "mean_token_accuracy": 0.18265988677740097,
      "num_tokens": 74419893.0,
      "step": 36725
    },
    {
      "entropy": 5.9847776889801025,
      "epoch": 3.209822598968802,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003985212787231,
      "loss": 5.527,
      "mean_token_accuracy": 0.1839903861284256,
      "num_tokens": 74430286.0,
      "step": 36730
    },
    {
      "entropy": 5.986248445510864,
      "epoch": 3.2102595473215065,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003984952320579263,
      "loss": 5.499,
      "mean_token_accuracy": 0.18769453167915345,
      "num_tokens": 74439601.0,
      "step": 36735
    },
    {
      "entropy": 6.02476053237915,
      "epoch": 3.2106964956742114,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003984691830240494,
      "loss": 5.56,
      "mean_token_accuracy": 0.17905724793672562,
      "num_tokens": 74448763.0,
      "step": 36740
    },
    {
      "entropy": 6.054367303848267,
      "epoch": 3.211133444026916,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039844313162196876,
      "loss": 5.6589,
      "mean_token_accuracy": 0.17556950598955154,
      "num_tokens": 74458357.0,
      "step": 36745
    },
    {
      "entropy": 5.996430158615112,
      "epoch": 3.211570392379621,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039841707785218426,
      "loss": 5.5197,
      "mean_token_accuracy": 0.1848635897040367,
      "num_tokens": 74468280.0,
      "step": 36750
    },
    {
      "entropy": 5.944570398330688,
      "epoch": 3.2120073407323253,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039839102171519556,
      "loss": 5.4497,
      "mean_token_accuracy": 0.19284634590148925,
      "num_tokens": 74478103.0,
      "step": 36755
    },
    {
      "entropy": 6.000587511062622,
      "epoch": 3.21244428908503,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003983649632115024,
      "loss": 5.4989,
      "mean_token_accuracy": 0.18260581344366072,
      "num_tokens": 74487903.0,
      "step": 36760
    },
    {
      "entropy": 5.929780721664429,
      "epoch": 3.2128812374377347,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003983389023416045,
      "loss": 5.5339,
      "mean_token_accuracy": 0.18577968776226045,
      "num_tokens": 74497998.0,
      "step": 36765
    },
    {
      "entropy": 5.960685348510742,
      "epoch": 3.2133181857904396,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00039831283910600197,
      "loss": 5.5069,
      "mean_token_accuracy": 0.18191025108098985,
      "num_tokens": 74507313.0,
      "step": 36770
    },
    {
      "entropy": 6.029903602600098,
      "epoch": 3.2137551341431445,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003982867735051945,
      "loss": 5.5362,
      "mean_token_accuracy": 0.18157870918512345,
      "num_tokens": 74516983.0,
      "step": 36775
    },
    {
      "entropy": 5.978306865692138,
      "epoch": 3.214192082495849,
      "grad_norm": 2.421875,
      "learning_rate": 0.0003982607055396821,
      "loss": 5.5209,
      "mean_token_accuracy": 0.1828486368060112,
      "num_tokens": 74527061.0,
      "step": 36780
    },
    {
      "entropy": 5.947231388092041,
      "epoch": 3.214629030848554,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039823463520996484,
      "loss": 5.56,
      "mean_token_accuracy": 0.1880180597305298,
      "num_tokens": 74537782.0,
      "step": 36785
    },
    {
      "entropy": 6.043452596664428,
      "epoch": 3.2150659792012584,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003982085625165426,
      "loss": 5.5497,
      "mean_token_accuracy": 0.1862048700451851,
      "num_tokens": 74548082.0,
      "step": 36790
    },
    {
      "entropy": 5.996845436096192,
      "epoch": 3.2155029275539633,
      "grad_norm": 1.625,
      "learning_rate": 0.0003981824874599157,
      "loss": 5.5521,
      "mean_token_accuracy": 0.18207562118768691,
      "num_tokens": 74559286.0,
      "step": 36795
    },
    {
      "entropy": 5.943031644821167,
      "epoch": 3.2159398759066677,
      "grad_norm": 1.015625,
      "learning_rate": 0.000398156410040584,
      "loss": 5.5788,
      "mean_token_accuracy": 0.17828590869903566,
      "num_tokens": 74569306.0,
      "step": 36800
    },
    {
      "entropy": 6.071094989776611,
      "epoch": 3.2163768242593727,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039813033025904793,
      "loss": 5.6242,
      "mean_token_accuracy": 0.17420183271169662,
      "num_tokens": 74579798.0,
      "step": 36805
    },
    {
      "entropy": 5.999210548400879,
      "epoch": 3.216813772612077,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039810424811580756,
      "loss": 5.5453,
      "mean_token_accuracy": 0.18172501772642136,
      "num_tokens": 74590068.0,
      "step": 36810
    },
    {
      "entropy": 5.991153717041016,
      "epoch": 3.217250720964782,
      "grad_norm": 1.6875,
      "learning_rate": 0.00039807816361136323,
      "loss": 5.6162,
      "mean_token_accuracy": 0.17666986286640168,
      "num_tokens": 74600341.0,
      "step": 36815
    },
    {
      "entropy": 5.972793865203857,
      "epoch": 3.2176876693174865,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003980520767462152,
      "loss": 5.5688,
      "mean_token_accuracy": 0.18800324201583862,
      "num_tokens": 74609548.0,
      "step": 36820
    },
    {
      "entropy": 6.047718191146851,
      "epoch": 3.2181246176701914,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003980259875208639,
      "loss": 5.4818,
      "mean_token_accuracy": 0.18655060976743698,
      "num_tokens": 74619590.0,
      "step": 36825
    },
    {
      "entropy": 6.029158258438111,
      "epoch": 3.218561566022896,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003979998959358097,
      "loss": 5.5315,
      "mean_token_accuracy": 0.179734206199646,
      "num_tokens": 74629664.0,
      "step": 36830
    },
    {
      "entropy": 5.948283290863037,
      "epoch": 3.218998514375601,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003979738019915529,
      "loss": 5.5456,
      "mean_token_accuracy": 0.18652368932962418,
      "num_tokens": 74639114.0,
      "step": 36835
    },
    {
      "entropy": 5.9230903625488285,
      "epoch": 3.2194354627283057,
      "grad_norm": 1.25,
      "learning_rate": 0.0003979477056885943,
      "loss": 5.4587,
      "mean_token_accuracy": 0.1948036327958107,
      "num_tokens": 74648730.0,
      "step": 36840
    },
    {
      "entropy": 6.050540971755981,
      "epoch": 3.21987241108101,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039792160702743426,
      "loss": 5.5807,
      "mean_token_accuracy": 0.18067288547754287,
      "num_tokens": 74659102.0,
      "step": 36845
    },
    {
      "entropy": 6.0557756423950195,
      "epoch": 3.220309359433715,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039789550600857336,
      "loss": 5.6627,
      "mean_token_accuracy": 0.16979477405548096,
      "num_tokens": 74669344.0,
      "step": 36850
    },
    {
      "entropy": 6.064534997940063,
      "epoch": 3.2207463077864196,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003978694026325123,
      "loss": 5.5388,
      "mean_token_accuracy": 0.1820995181798935,
      "num_tokens": 74678524.0,
      "step": 36855
    },
    {
      "entropy": 5.9809798240661625,
      "epoch": 3.2211832561391245,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003978432968997517,
      "loss": 5.5813,
      "mean_token_accuracy": 0.18416340798139572,
      "num_tokens": 74688584.0,
      "step": 36860
    },
    {
      "entropy": 6.06178240776062,
      "epoch": 3.221620204491829,
      "grad_norm": 0.9375,
      "learning_rate": 0.00039781718881079227,
      "loss": 5.6122,
      "mean_token_accuracy": 0.1762694612145424,
      "num_tokens": 74699233.0,
      "step": 36865
    },
    {
      "entropy": 6.051771020889282,
      "epoch": 3.222057152844534,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039779107836613475,
      "loss": 5.589,
      "mean_token_accuracy": 0.18165897876024245,
      "num_tokens": 74710144.0,
      "step": 36870
    },
    {
      "entropy": 5.958129119873047,
      "epoch": 3.2224941011972383,
      "grad_norm": 1.25,
      "learning_rate": 0.00039776496556628014,
      "loss": 5.6054,
      "mean_token_accuracy": 0.17930916398763658,
      "num_tokens": 74720289.0,
      "step": 36875
    },
    {
      "entropy": 5.977130889892578,
      "epoch": 3.2229310495499432,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039773885041172907,
      "loss": 5.602,
      "mean_token_accuracy": 0.17276728749275208,
      "num_tokens": 74732395.0,
      "step": 36880
    },
    {
      "entropy": 6.008520030975342,
      "epoch": 3.2233679979026477,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039771273290298254,
      "loss": 5.5082,
      "mean_token_accuracy": 0.19021973758935928,
      "num_tokens": 74741986.0,
      "step": 36885
    },
    {
      "entropy": 5.910481691360474,
      "epoch": 3.2238049462553526,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003976866130405415,
      "loss": 5.4728,
      "mean_token_accuracy": 0.18189530968666076,
      "num_tokens": 74751540.0,
      "step": 36890
    },
    {
      "entropy": 5.948031091690064,
      "epoch": 3.224241894608057,
      "grad_norm": 1.84375,
      "learning_rate": 0.00039766049082490696,
      "loss": 5.5221,
      "mean_token_accuracy": 0.18237309604883195,
      "num_tokens": 74761875.0,
      "step": 36895
    },
    {
      "entropy": 5.958959341049194,
      "epoch": 3.224678842960762,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039763436625657987,
      "loss": 5.4569,
      "mean_token_accuracy": 0.18372741788625718,
      "num_tokens": 74771273.0,
      "step": 36900
    },
    {
      "entropy": 5.92314829826355,
      "epoch": 3.225115791313467,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003976082393360614,
      "loss": 5.4652,
      "mean_token_accuracy": 0.18762468546628952,
      "num_tokens": 74781176.0,
      "step": 36905
    },
    {
      "entropy": 5.982079744338989,
      "epoch": 3.2255527396661714,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039758211006385273,
      "loss": 5.5471,
      "mean_token_accuracy": 0.18563752621412277,
      "num_tokens": 74790731.0,
      "step": 36910
    },
    {
      "entropy": 5.948986530303955,
      "epoch": 3.2259896880188763,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039755597844045486,
      "loss": 5.4673,
      "mean_token_accuracy": 0.18395109474658966,
      "num_tokens": 74799810.0,
      "step": 36915
    },
    {
      "entropy": 5.957611560821533,
      "epoch": 3.226426636371581,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003975298444663692,
      "loss": 5.5753,
      "mean_token_accuracy": 0.1806183412671089,
      "num_tokens": 74809701.0,
      "step": 36920
    },
    {
      "entropy": 6.003874158859253,
      "epoch": 3.2268635847242857,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039750370814209683,
      "loss": 5.5305,
      "mean_token_accuracy": 0.189536252617836,
      "num_tokens": 74820061.0,
      "step": 36925
    },
    {
      "entropy": 6.050349855422974,
      "epoch": 3.22730053307699,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003974775694681392,
      "loss": 5.5376,
      "mean_token_accuracy": 0.1797567218542099,
      "num_tokens": 74830360.0,
      "step": 36930
    },
    {
      "entropy": 6.00196008682251,
      "epoch": 3.227737481429695,
      "grad_norm": 1.0,
      "learning_rate": 0.0003974514284449975,
      "loss": 5.6049,
      "mean_token_accuracy": 0.18216290771961213,
      "num_tokens": 74841855.0,
      "step": 36935
    },
    {
      "entropy": 5.886476659774781,
      "epoch": 3.2281744297823995,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039742528507317337,
      "loss": 5.4822,
      "mean_token_accuracy": 0.1909455120563507,
      "num_tokens": 74851955.0,
      "step": 36940
    },
    {
      "entropy": 5.9932928562164305,
      "epoch": 3.2286113781351045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003973991393531681,
      "loss": 5.5313,
      "mean_token_accuracy": 0.1917917400598526,
      "num_tokens": 74861720.0,
      "step": 36945
    },
    {
      "entropy": 5.9875391006469725,
      "epoch": 3.229048326487809,
      "grad_norm": 1.453125,
      "learning_rate": 0.00039737299128548317,
      "loss": 5.5122,
      "mean_token_accuracy": 0.17768411189317704,
      "num_tokens": 74872136.0,
      "step": 36950
    },
    {
      "entropy": 6.028617858886719,
      "epoch": 3.229485274840514,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003973468408706202,
      "loss": 5.5436,
      "mean_token_accuracy": 0.1787821888923645,
      "num_tokens": 74881683.0,
      "step": 36955
    },
    {
      "entropy": 6.0028105735778805,
      "epoch": 3.2299222231932188,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00039732068810908067,
      "loss": 5.4913,
      "mean_token_accuracy": 0.18766270875930785,
      "num_tokens": 74892175.0,
      "step": 36960
    },
    {
      "entropy": 5.90830888748169,
      "epoch": 3.2303591715459232,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003972945330013662,
      "loss": 5.5171,
      "mean_token_accuracy": 0.18992394655942918,
      "num_tokens": 74903547.0,
      "step": 36965
    },
    {
      "entropy": 5.956855773925781,
      "epoch": 3.230796119898628,
      "grad_norm": 2.265625,
      "learning_rate": 0.00039726837554797856,
      "loss": 5.4912,
      "mean_token_accuracy": 0.18585818260908127,
      "num_tokens": 74912990.0,
      "step": 36970
    },
    {
      "entropy": 5.962987041473388,
      "epoch": 3.2312330682513326,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039724221574941943,
      "loss": 5.5523,
      "mean_token_accuracy": 0.18130647242069245,
      "num_tokens": 74922306.0,
      "step": 36975
    },
    {
      "entropy": 5.943728351593018,
      "epoch": 3.2316700166040375,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039721605360619054,
      "loss": 5.5228,
      "mean_token_accuracy": 0.18330036848783493,
      "num_tokens": 74932207.0,
      "step": 36980
    },
    {
      "entropy": 5.955518770217895,
      "epoch": 3.232106964956742,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039718988911879367,
      "loss": 5.5724,
      "mean_token_accuracy": 0.17981388717889785,
      "num_tokens": 74941933.0,
      "step": 36985
    },
    {
      "entropy": 6.077213001251221,
      "epoch": 3.232543913309447,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039716372228773073,
      "loss": 5.6261,
      "mean_token_accuracy": 0.17711969316005707,
      "num_tokens": 74953418.0,
      "step": 36990
    },
    {
      "entropy": 6.0765844821929935,
      "epoch": 3.2329808616621514,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039713755311350356,
      "loss": 5.5613,
      "mean_token_accuracy": 0.18188949078321456,
      "num_tokens": 74963381.0,
      "step": 36995
    },
    {
      "entropy": 6.007251262664795,
      "epoch": 3.2334178100148563,
      "grad_norm": 1.53125,
      "learning_rate": 0.00039711138159661413,
      "loss": 5.5684,
      "mean_token_accuracy": 0.18688598722219468,
      "num_tokens": 74974072.0,
      "step": 37000
    },
    {
      "entropy": 5.885683679580689,
      "epoch": 3.2338547583675608,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003970852077375644,
      "loss": 5.4332,
      "mean_token_accuracy": 0.19466383010149002,
      "num_tokens": 74984191.0,
      "step": 37005
    },
    {
      "entropy": 5.962734699249268,
      "epoch": 3.2342917067202657,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003970590315368564,
      "loss": 5.5399,
      "mean_token_accuracy": 0.1845573216676712,
      "num_tokens": 74994278.0,
      "step": 37010
    },
    {
      "entropy": 5.927823066711426,
      "epoch": 3.2347286550729706,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003970328529949923,
      "loss": 5.497,
      "mean_token_accuracy": 0.1930128425359726,
      "num_tokens": 75004150.0,
      "step": 37015
    },
    {
      "entropy": 5.9568202018737795,
      "epoch": 3.235165603425675,
      "grad_norm": 1.125,
      "learning_rate": 0.0003970066721124741,
      "loss": 5.4978,
      "mean_token_accuracy": 0.18937959671020507,
      "num_tokens": 75013632.0,
      "step": 37020
    },
    {
      "entropy": 5.980709457397461,
      "epoch": 3.23560255177838,
      "grad_norm": 1.0,
      "learning_rate": 0.0003969804888898039,
      "loss": 5.5819,
      "mean_token_accuracy": 0.1801724150776863,
      "num_tokens": 75023217.0,
      "step": 37025
    },
    {
      "entropy": 5.974165391921997,
      "epoch": 3.2360395001310844,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003969543033274841,
      "loss": 5.5251,
      "mean_token_accuracy": 0.18488100171089172,
      "num_tokens": 75033771.0,
      "step": 37030
    },
    {
      "entropy": 6.046087551116943,
      "epoch": 3.2364764484837893,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039692811542601684,
      "loss": 5.5901,
      "mean_token_accuracy": 0.18345193415880204,
      "num_tokens": 75044151.0,
      "step": 37035
    },
    {
      "entropy": 5.989060401916504,
      "epoch": 3.236913396836494,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003969019251859045,
      "loss": 5.4761,
      "mean_token_accuracy": 0.1980838358402252,
      "num_tokens": 75054129.0,
      "step": 37040
    },
    {
      "entropy": 5.94588623046875,
      "epoch": 3.2373503451891987,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039687573260764933,
      "loss": 5.5023,
      "mean_token_accuracy": 0.18540114313364028,
      "num_tokens": 75063147.0,
      "step": 37045
    },
    {
      "entropy": 5.8931995868682865,
      "epoch": 3.237787293541903,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039684953769175376,
      "loss": 5.5733,
      "mean_token_accuracy": 0.17872073650360107,
      "num_tokens": 75072389.0,
      "step": 37050
    },
    {
      "entropy": 5.9600889682769775,
      "epoch": 3.238224241894608,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003968233404387201,
      "loss": 5.5251,
      "mean_token_accuracy": 0.18668034821748733,
      "num_tokens": 75081253.0,
      "step": 37055
    },
    {
      "entropy": 5.947373914718628,
      "epoch": 3.2386611902473126,
      "grad_norm": 1.03125,
      "learning_rate": 0.000396797140849051,
      "loss": 5.4996,
      "mean_token_accuracy": 0.194967582821846,
      "num_tokens": 75091316.0,
      "step": 37060
    },
    {
      "entropy": 6.045733642578125,
      "epoch": 3.2390981386000175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039677093892324895,
      "loss": 5.6422,
      "mean_token_accuracy": 0.17854313999414445,
      "num_tokens": 75101204.0,
      "step": 37065
    },
    {
      "entropy": 5.925362014770508,
      "epoch": 3.2395350869527224,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003967447346618164,
      "loss": 5.5175,
      "mean_token_accuracy": 0.1793399468064308,
      "num_tokens": 75111524.0,
      "step": 37070
    },
    {
      "entropy": 5.9532005310058596,
      "epoch": 3.239972035305427,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039671852806525613,
      "loss": 5.5024,
      "mean_token_accuracy": 0.18250584751367568,
      "num_tokens": 75120699.0,
      "step": 37075
    },
    {
      "entropy": 6.052551174163819,
      "epoch": 3.240408983658132,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039669231913407065,
      "loss": 5.5356,
      "mean_token_accuracy": 0.17648100852966309,
      "num_tokens": 75130892.0,
      "step": 37080
    },
    {
      "entropy": 6.029707336425782,
      "epoch": 3.2408459320108363,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003966661078687628,
      "loss": 5.6002,
      "mean_token_accuracy": 0.1773662820458412,
      "num_tokens": 75140994.0,
      "step": 37085
    },
    {
      "entropy": 5.977050304412842,
      "epoch": 3.241282880363541,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003966398942698352,
      "loss": 5.622,
      "mean_token_accuracy": 0.18258688151836394,
      "num_tokens": 75151084.0,
      "step": 37090
    },
    {
      "entropy": 6.011880731582641,
      "epoch": 3.2417198287162456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003966136783377907,
      "loss": 5.551,
      "mean_token_accuracy": 0.18201530873775482,
      "num_tokens": 75160815.0,
      "step": 37095
    },
    {
      "entropy": 6.026603126525879,
      "epoch": 3.2421567770689506,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003965874600731321,
      "loss": 5.5305,
      "mean_token_accuracy": 0.18382564038038254,
      "num_tokens": 75170827.0,
      "step": 37100
    },
    {
      "entropy": 6.017712593078613,
      "epoch": 3.242593725421655,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003965612394763623,
      "loss": 5.5562,
      "mean_token_accuracy": 0.18645580410957335,
      "num_tokens": 75180250.0,
      "step": 37105
    },
    {
      "entropy": 5.972575712203979,
      "epoch": 3.24303067377436,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003965350165479842,
      "loss": 5.5606,
      "mean_token_accuracy": 0.1806705802679062,
      "num_tokens": 75190862.0,
      "step": 37110
    },
    {
      "entropy": 5.972254371643066,
      "epoch": 3.2434676221270644,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003965087912885009,
      "loss": 5.5138,
      "mean_token_accuracy": 0.18469755351543427,
      "num_tokens": 75199632.0,
      "step": 37115
    },
    {
      "entropy": 5.935790777206421,
      "epoch": 3.2439045704797693,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039648256369841523,
      "loss": 5.4844,
      "mean_token_accuracy": 0.18407825231552125,
      "num_tokens": 75209626.0,
      "step": 37120
    },
    {
      "entropy": 5.975972604751587,
      "epoch": 3.2443415188324742,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003964563337782305,
      "loss": 5.5861,
      "mean_token_accuracy": 0.1756226673722267,
      "num_tokens": 75220682.0,
      "step": 37125
    },
    {
      "entropy": 6.026590824127197,
      "epoch": 3.2447784671851787,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039643010152844946,
      "loss": 5.5392,
      "mean_token_accuracy": 0.1863638460636139,
      "num_tokens": 75230330.0,
      "step": 37130
    },
    {
      "entropy": 6.043656063079834,
      "epoch": 3.2452154155378836,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039640386694957553,
      "loss": 5.5428,
      "mean_token_accuracy": 0.18312908560037613,
      "num_tokens": 75240485.0,
      "step": 37135
    },
    {
      "entropy": 5.998636960983276,
      "epoch": 3.245652363890588,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039637763004211175,
      "loss": 5.563,
      "mean_token_accuracy": 0.18519103825092315,
      "num_tokens": 75250670.0,
      "step": 37140
    },
    {
      "entropy": 5.96561336517334,
      "epoch": 3.246089312243293,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039635139080656146,
      "loss": 5.5388,
      "mean_token_accuracy": 0.1899004727602005,
      "num_tokens": 75260312.0,
      "step": 37145
    },
    {
      "entropy": 6.001893663406372,
      "epoch": 3.2465262605959975,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00039632514924342794,
      "loss": 5.54,
      "mean_token_accuracy": 0.18231704235076904,
      "num_tokens": 75269978.0,
      "step": 37150
    },
    {
      "entropy": 5.9548797607421875,
      "epoch": 3.2469632089487024,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039629890535321444,
      "loss": 5.5794,
      "mean_token_accuracy": 0.18220483064651488,
      "num_tokens": 75279781.0,
      "step": 37155
    },
    {
      "entropy": 6.073662042617798,
      "epoch": 3.247400157301407,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00039627265913642435,
      "loss": 5.6135,
      "mean_token_accuracy": 0.17593055069446564,
      "num_tokens": 75290352.0,
      "step": 37160
    },
    {
      "entropy": 6.025544118881226,
      "epoch": 3.2478371056541118,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003962464105935612,
      "loss": 5.5311,
      "mean_token_accuracy": 0.18040233105421066,
      "num_tokens": 75300511.0,
      "step": 37165
    },
    {
      "entropy": 5.971381330490113,
      "epoch": 3.2482740540068162,
      "grad_norm": 1.546875,
      "learning_rate": 0.00039622015972512833,
      "loss": 5.5543,
      "mean_token_accuracy": 0.18660055249929428,
      "num_tokens": 75310717.0,
      "step": 37170
    },
    {
      "entropy": 6.013277053833008,
      "epoch": 3.248711002359521,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003961939065316293,
      "loss": 5.5613,
      "mean_token_accuracy": 0.18237764835357667,
      "num_tokens": 75322402.0,
      "step": 37175
    },
    {
      "entropy": 5.946441698074341,
      "epoch": 3.2491479507122256,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039616765101356755,
      "loss": 5.442,
      "mean_token_accuracy": 0.192926062643528,
      "num_tokens": 75331866.0,
      "step": 37180
    },
    {
      "entropy": 5.986715364456177,
      "epoch": 3.2495848990649305,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039614139317144676,
      "loss": 5.4512,
      "mean_token_accuracy": 0.1894882544875145,
      "num_tokens": 75341125.0,
      "step": 37185
    },
    {
      "entropy": 5.929032564163208,
      "epoch": 3.250021847417635,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003961151330057706,
      "loss": 5.5272,
      "mean_token_accuracy": 0.18214289844036102,
      "num_tokens": 75350518.0,
      "step": 37190
    },
    {
      "entropy": 6.011961507797241,
      "epoch": 3.25045879577034,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003960888705170428,
      "loss": 5.5929,
      "mean_token_accuracy": 0.17636508494615555,
      "num_tokens": 75359330.0,
      "step": 37195
    },
    {
      "entropy": 6.082571363449096,
      "epoch": 3.250895744123045,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000396062605705767,
      "loss": 5.6152,
      "mean_token_accuracy": 0.1815860316157341,
      "num_tokens": 75368771.0,
      "step": 37200
    },
    {
      "entropy": 6.038357257843018,
      "epoch": 3.2513326924757493,
      "grad_norm": 1.40625,
      "learning_rate": 0.000396036338572447,
      "loss": 5.5583,
      "mean_token_accuracy": 0.17721727192401887,
      "num_tokens": 75379656.0,
      "step": 37205
    },
    {
      "entropy": 5.950801134109497,
      "epoch": 3.251769640828454,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039601006911758646,
      "loss": 5.5689,
      "mean_token_accuracy": 0.18270267695188522,
      "num_tokens": 75390248.0,
      "step": 37210
    },
    {
      "entropy": 5.991058301925659,
      "epoch": 3.2522065891811587,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003959837973416895,
      "loss": 5.587,
      "mean_token_accuracy": 0.1757294327020645,
      "num_tokens": 75400090.0,
      "step": 37215
    },
    {
      "entropy": 6.03140172958374,
      "epoch": 3.2526435375338636,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039595752324525984,
      "loss": 5.5829,
      "mean_token_accuracy": 0.17802435606718064,
      "num_tokens": 75410380.0,
      "step": 37220
    },
    {
      "entropy": 6.065441608428955,
      "epoch": 3.253080485886568,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003959312468288016,
      "loss": 5.4817,
      "mean_token_accuracy": 0.18557247668504714,
      "num_tokens": 75421265.0,
      "step": 37225
    },
    {
      "entropy": 5.969675731658936,
      "epoch": 3.253517434239273,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003959049680928187,
      "loss": 5.5322,
      "mean_token_accuracy": 0.18600088208913804,
      "num_tokens": 75432049.0,
      "step": 37230
    },
    {
      "entropy": 5.981667613983154,
      "epoch": 3.2539543825919774,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003958786870378151,
      "loss": 5.6087,
      "mean_token_accuracy": 0.18209224939346313,
      "num_tokens": 75441364.0,
      "step": 37235
    },
    {
      "entropy": 6.004035997390747,
      "epoch": 3.2543913309446824,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003958524036642949,
      "loss": 5.5603,
      "mean_token_accuracy": 0.18284882009029388,
      "num_tokens": 75452019.0,
      "step": 37240
    },
    {
      "entropy": 6.021402883529663,
      "epoch": 3.254828279297387,
      "grad_norm": 1.703125,
      "learning_rate": 0.00039582611797276237,
      "loss": 5.5854,
      "mean_token_accuracy": 0.17409219592809677,
      "num_tokens": 75463053.0,
      "step": 37245
    },
    {
      "entropy": 5.944664859771729,
      "epoch": 3.2552652276500917,
      "grad_norm": 1.125,
      "learning_rate": 0.0003957998299637215,
      "loss": 5.5231,
      "mean_token_accuracy": 0.184543976187706,
      "num_tokens": 75472757.0,
      "step": 37250
    },
    {
      "entropy": 5.939795732498169,
      "epoch": 3.2557021760027967,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003957735396376766,
      "loss": 5.4797,
      "mean_token_accuracy": 0.19076758474111558,
      "num_tokens": 75482228.0,
      "step": 37255
    },
    {
      "entropy": 5.971332120895386,
      "epoch": 3.256139124355501,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003957472469951319,
      "loss": 5.5769,
      "mean_token_accuracy": 0.18431263267993928,
      "num_tokens": 75491344.0,
      "step": 37260
    },
    {
      "entropy": 5.972324705123901,
      "epoch": 3.256576072708206,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039572095203659177,
      "loss": 5.6267,
      "mean_token_accuracy": 0.17678564488887788,
      "num_tokens": 75502042.0,
      "step": 37265
    },
    {
      "entropy": 6.008614492416382,
      "epoch": 3.2570130210609105,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003956946547625606,
      "loss": 5.4956,
      "mean_token_accuracy": 0.18680136948823928,
      "num_tokens": 75512116.0,
      "step": 37270
    },
    {
      "entropy": 5.947762537002563,
      "epoch": 3.2574499694136154,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039566835517354266,
      "loss": 5.4755,
      "mean_token_accuracy": 0.19007740020751954,
      "num_tokens": 75521220.0,
      "step": 37275
    },
    {
      "entropy": 6.0225084781646725,
      "epoch": 3.25788691776632,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003956420532700423,
      "loss": 5.5671,
      "mean_token_accuracy": 0.18255791664123536,
      "num_tokens": 75530556.0,
      "step": 37280
    },
    {
      "entropy": 5.973177337646485,
      "epoch": 3.258323866119025,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003956157490525643,
      "loss": 5.5563,
      "mean_token_accuracy": 0.1884783312678337,
      "num_tokens": 75539843.0,
      "step": 37285
    },
    {
      "entropy": 5.947100639343262,
      "epoch": 3.2587608144717293,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000395589442521613,
      "loss": 5.4998,
      "mean_token_accuracy": 0.18855311423540116,
      "num_tokens": 75549945.0,
      "step": 37290
    },
    {
      "entropy": 5.98464527130127,
      "epoch": 3.259197762824434,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039556313367769297,
      "loss": 5.6351,
      "mean_token_accuracy": 0.17890742123126985,
      "num_tokens": 75560808.0,
      "step": 37295
    },
    {
      "entropy": 6.051611375808716,
      "epoch": 3.2596347111771387,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039553682252130886,
      "loss": 5.6043,
      "mean_token_accuracy": 0.1725266233086586,
      "num_tokens": 75571431.0,
      "step": 37300
    },
    {
      "entropy": 6.0201324939727785,
      "epoch": 3.2600716595298436,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039551050905296543,
      "loss": 5.543,
      "mean_token_accuracy": 0.18090534806251526,
      "num_tokens": 75582543.0,
      "step": 37305
    },
    {
      "entropy": 5.941777849197388,
      "epoch": 3.2605086078825485,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039548419327316715,
      "loss": 5.5041,
      "mean_token_accuracy": 0.18689975291490554,
      "num_tokens": 75591951.0,
      "step": 37310
    },
    {
      "entropy": 5.90203709602356,
      "epoch": 3.260945556235253,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000395457875182419,
      "loss": 5.5214,
      "mean_token_accuracy": 0.1859283298254013,
      "num_tokens": 75602107.0,
      "step": 37315
    },
    {
      "entropy": 5.941532135009766,
      "epoch": 3.261382504587958,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039543155478122564,
      "loss": 5.4795,
      "mean_token_accuracy": 0.18971692025661469,
      "num_tokens": 75612449.0,
      "step": 37320
    },
    {
      "entropy": 6.010607051849365,
      "epoch": 3.2618194529406623,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000395405232070092,
      "loss": 5.5865,
      "mean_token_accuracy": 0.17833534479141236,
      "num_tokens": 75622078.0,
      "step": 37325
    },
    {
      "entropy": 6.052184104919434,
      "epoch": 3.2622564012933672,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039537890704952294,
      "loss": 5.6013,
      "mean_token_accuracy": 0.18135499358177185,
      "num_tokens": 75632672.0,
      "step": 37330
    },
    {
      "entropy": 5.97726354598999,
      "epoch": 3.2626933496460717,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039535257972002325,
      "loss": 5.5399,
      "mean_token_accuracy": 0.1855751872062683,
      "num_tokens": 75641633.0,
      "step": 37335
    },
    {
      "entropy": 5.954747009277344,
      "epoch": 3.2631302979987766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039532625008209814,
      "loss": 5.4581,
      "mean_token_accuracy": 0.19156711995601655,
      "num_tokens": 75651109.0,
      "step": 37340
    },
    {
      "entropy": 5.990666151046753,
      "epoch": 3.263567246351481,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003952999181362525,
      "loss": 5.5242,
      "mean_token_accuracy": 0.1863347500562668,
      "num_tokens": 75661176.0,
      "step": 37345
    },
    {
      "entropy": 6.05031418800354,
      "epoch": 3.264004194704186,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039527358388299135,
      "loss": 5.5938,
      "mean_token_accuracy": 0.1746885970234871,
      "num_tokens": 75670838.0,
      "step": 37350
    },
    {
      "entropy": 5.961955451965332,
      "epoch": 3.2644411430568905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003952472473228198,
      "loss": 5.5182,
      "mean_token_accuracy": 0.18525384813547136,
      "num_tokens": 75681079.0,
      "step": 37355
    },
    {
      "entropy": 5.978346586227417,
      "epoch": 3.2648780914095954,
      "grad_norm": 1.09375,
      "learning_rate": 0.000395220908456243,
      "loss": 5.439,
      "mean_token_accuracy": 0.19912093579769136,
      "num_tokens": 75691748.0,
      "step": 37360
    },
    {
      "entropy": 5.973310089111328,
      "epoch": 3.2653150397623003,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003951945672837663,
      "loss": 5.5508,
      "mean_token_accuracy": 0.18436937183141708,
      "num_tokens": 75702751.0,
      "step": 37365
    },
    {
      "entropy": 5.984009599685669,
      "epoch": 3.265751988115005,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039516822380589476,
      "loss": 5.5346,
      "mean_token_accuracy": 0.18848855644464493,
      "num_tokens": 75712107.0,
      "step": 37370
    },
    {
      "entropy": 6.01241545677185,
      "epoch": 3.2661889364677097,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003951418780231338,
      "loss": 5.5289,
      "mean_token_accuracy": 0.18907247930765153,
      "num_tokens": 75721595.0,
      "step": 37375
    },
    {
      "entropy": 5.971722936630249,
      "epoch": 3.266625884820414,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003951155299359886,
      "loss": 5.4742,
      "mean_token_accuracy": 0.18351746797561647,
      "num_tokens": 75731332.0,
      "step": 37380
    },
    {
      "entropy": 5.952707672119141,
      "epoch": 3.267062833173119,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003950891795449646,
      "loss": 5.4724,
      "mean_token_accuracy": 0.18453672975301744,
      "num_tokens": 75740425.0,
      "step": 37385
    },
    {
      "entropy": 5.912100696563721,
      "epoch": 3.2674997815258235,
      "grad_norm": 1.0,
      "learning_rate": 0.0003950628268505672,
      "loss": 5.5401,
      "mean_token_accuracy": 0.18803485482931137,
      "num_tokens": 75750091.0,
      "step": 37390
    },
    {
      "entropy": 6.057372999191284,
      "epoch": 3.2679367298785285,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039503647185330185,
      "loss": 5.6009,
      "mean_token_accuracy": 0.17723310589790345,
      "num_tokens": 75760780.0,
      "step": 37395
    },
    {
      "entropy": 6.023770141601562,
      "epoch": 3.268373678231233,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039501011455367406,
      "loss": 5.5484,
      "mean_token_accuracy": 0.18610524237155915,
      "num_tokens": 75771056.0,
      "step": 37400
    },
    {
      "entropy": 5.949983692169189,
      "epoch": 3.268810626583938,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003949837549521895,
      "loss": 5.4921,
      "mean_token_accuracy": 0.18780603855848313,
      "num_tokens": 75781265.0,
      "step": 37405
    },
    {
      "entropy": 6.001104831695557,
      "epoch": 3.2692475749366423,
      "grad_norm": 2.796875,
      "learning_rate": 0.00039495739304935357,
      "loss": 5.5634,
      "mean_token_accuracy": 0.1833595022559166,
      "num_tokens": 75791911.0,
      "step": 37410
    },
    {
      "entropy": 5.982775115966797,
      "epoch": 3.2696845232893472,
      "grad_norm": 1.328125,
      "learning_rate": 0.000394931028845672,
      "loss": 5.4586,
      "mean_token_accuracy": 0.1858783096075058,
      "num_tokens": 75802085.0,
      "step": 37415
    },
    {
      "entropy": 5.973740100860596,
      "epoch": 3.270121471642052,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003949046623416504,
      "loss": 5.4891,
      "mean_token_accuracy": 0.1897572711110115,
      "num_tokens": 75810740.0,
      "step": 37420
    },
    {
      "entropy": 5.994152021408081,
      "epoch": 3.2705584199947566,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039487829353779464,
      "loss": 5.5051,
      "mean_token_accuracy": 0.18665046542882918,
      "num_tokens": 75820260.0,
      "step": 37425
    },
    {
      "entropy": 5.965489768981934,
      "epoch": 3.2709953683474615,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039485192243461035,
      "loss": 5.5036,
      "mean_token_accuracy": 0.18888915181159974,
      "num_tokens": 75830152.0,
      "step": 37430
    },
    {
      "entropy": 5.913066959381103,
      "epoch": 3.271432316700166,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003948255490326034,
      "loss": 5.4923,
      "mean_token_accuracy": 0.19107874780893325,
      "num_tokens": 75840617.0,
      "step": 37435
    },
    {
      "entropy": 5.94170069694519,
      "epoch": 3.271869265052871,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003947991733322796,
      "loss": 5.468,
      "mean_token_accuracy": 0.1912022814154625,
      "num_tokens": 75850114.0,
      "step": 37440
    },
    {
      "entropy": 5.91039457321167,
      "epoch": 3.2723062134055754,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003947727953341449,
      "loss": 5.5068,
      "mean_token_accuracy": 0.19302290678024292,
      "num_tokens": 75861202.0,
      "step": 37445
    },
    {
      "entropy": 6.030451965332031,
      "epoch": 3.2727431617582803,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039474641503870524,
      "loss": 5.4981,
      "mean_token_accuracy": 0.1770041674375534,
      "num_tokens": 75870613.0,
      "step": 37450
    },
    {
      "entropy": 5.955254793167114,
      "epoch": 3.2731801101109848,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039472003244646656,
      "loss": 5.485,
      "mean_token_accuracy": 0.18486209809780121,
      "num_tokens": 75880550.0,
      "step": 37455
    },
    {
      "entropy": 5.952261352539063,
      "epoch": 3.2736170584636897,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003946936475579349,
      "loss": 5.4711,
      "mean_token_accuracy": 0.19303075820207596,
      "num_tokens": 75890838.0,
      "step": 37460
    },
    {
      "entropy": 5.964072751998901,
      "epoch": 3.274054006816394,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003946672603736164,
      "loss": 5.5853,
      "mean_token_accuracy": 0.18351449966430664,
      "num_tokens": 75900834.0,
      "step": 37465
    },
    {
      "entropy": 5.923960018157959,
      "epoch": 3.274490955169099,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039464087089401714,
      "loss": 5.5669,
      "mean_token_accuracy": 0.18511078953742982,
      "num_tokens": 75911442.0,
      "step": 37470
    },
    {
      "entropy": 6.0329270362854,
      "epoch": 3.274927903521804,
      "grad_norm": 1.125,
      "learning_rate": 0.00039461447911964327,
      "loss": 5.6037,
      "mean_token_accuracy": 0.18044360876083373,
      "num_tokens": 75922824.0,
      "step": 37475
    },
    {
      "entropy": 6.001888084411621,
      "epoch": 3.2753648518745084,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003945880850510011,
      "loss": 5.5355,
      "mean_token_accuracy": 0.1831520289182663,
      "num_tokens": 75932784.0,
      "step": 37480
    },
    {
      "entropy": 5.898534870147705,
      "epoch": 3.275801800227213,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039456168868859663,
      "loss": 5.4694,
      "mean_token_accuracy": 0.1930832177400589,
      "num_tokens": 75942176.0,
      "step": 37485
    },
    {
      "entropy": 5.978522491455078,
      "epoch": 3.276238748579918,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039453529003293647,
      "loss": 5.4914,
      "mean_token_accuracy": 0.18592797517776488,
      "num_tokens": 75952602.0,
      "step": 37490
    },
    {
      "entropy": 5.973237657546997,
      "epoch": 3.2766756969326227,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039450888908452677,
      "loss": 5.4779,
      "mean_token_accuracy": 0.19001152217388154,
      "num_tokens": 75962971.0,
      "step": 37495
    },
    {
      "entropy": 5.9934581279754635,
      "epoch": 3.277112645285327,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003944824858438739,
      "loss": 5.6055,
      "mean_token_accuracy": 0.18329101502895356,
      "num_tokens": 75972751.0,
      "step": 37500
    },
    {
      "entropy": 6.053712368011475,
      "epoch": 3.277549593638032,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003944560803114845,
      "loss": 5.6643,
      "mean_token_accuracy": 0.17827588319778442,
      "num_tokens": 75984653.0,
      "step": 37505
    },
    {
      "entropy": 6.041428661346435,
      "epoch": 3.2779865419907366,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003944296724878647,
      "loss": 5.5426,
      "mean_token_accuracy": 0.18502767533063888,
      "num_tokens": 75994211.0,
      "step": 37510
    },
    {
      "entropy": 5.925850677490234,
      "epoch": 3.2784234903434415,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003944032623735213,
      "loss": 5.466,
      "mean_token_accuracy": 0.18812734186649321,
      "num_tokens": 76003891.0,
      "step": 37515
    },
    {
      "entropy": 5.974750947952271,
      "epoch": 3.278860438696146,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003943768499689607,
      "loss": 5.521,
      "mean_token_accuracy": 0.18439611196517944,
      "num_tokens": 76014625.0,
      "step": 37520
    },
    {
      "entropy": 5.9454652786254885,
      "epoch": 3.279297387048851,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039435043527468963,
      "loss": 5.4932,
      "mean_token_accuracy": 0.1859426811337471,
      "num_tokens": 76024773.0,
      "step": 37525
    },
    {
      "entropy": 5.935676622390747,
      "epoch": 3.2797343354015553,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039432401829121467,
      "loss": 5.4188,
      "mean_token_accuracy": 0.18980349004268646,
      "num_tokens": 76035085.0,
      "step": 37530
    },
    {
      "entropy": 5.960848760604859,
      "epoch": 3.2801712837542603,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003942975990190425,
      "loss": 5.5794,
      "mean_token_accuracy": 0.17655138075351715,
      "num_tokens": 76046055.0,
      "step": 37535
    },
    {
      "entropy": 5.911958122253418,
      "epoch": 3.2806082321069647,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003942711774586799,
      "loss": 5.4287,
      "mean_token_accuracy": 0.1968392997980118,
      "num_tokens": 76055761.0,
      "step": 37540
    },
    {
      "entropy": 6.003742122650147,
      "epoch": 3.2810451804596696,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003942447536106335,
      "loss": 5.6165,
      "mean_token_accuracy": 0.17936030626296998,
      "num_tokens": 76065844.0,
      "step": 37545
    },
    {
      "entropy": 6.016689872741699,
      "epoch": 3.2814821288123746,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00039421832747541034,
      "loss": 5.5501,
      "mean_token_accuracy": 0.18526527136564255,
      "num_tokens": 76076048.0,
      "step": 37550
    },
    {
      "entropy": 5.971002721786499,
      "epoch": 3.281919077165079,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003941918990535172,
      "loss": 5.5045,
      "mean_token_accuracy": 0.1863154649734497,
      "num_tokens": 76086775.0,
      "step": 37555
    },
    {
      "entropy": 5.9566596984863285,
      "epoch": 3.282356025517784,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003941654683454609,
      "loss": 5.5561,
      "mean_token_accuracy": 0.18184247612953186,
      "num_tokens": 76097492.0,
      "step": 37560
    },
    {
      "entropy": 5.954274082183838,
      "epoch": 3.2827929738704884,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039413903535174855,
      "loss": 5.5263,
      "mean_token_accuracy": 0.18120166808366775,
      "num_tokens": 76106709.0,
      "step": 37565
    },
    {
      "entropy": 5.982485961914063,
      "epoch": 3.2832299222231933,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000394112600072887,
      "loss": 5.5107,
      "mean_token_accuracy": 0.18359669297933578,
      "num_tokens": 76117189.0,
      "step": 37570
    },
    {
      "entropy": 6.062857055664063,
      "epoch": 3.283666870575898,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00039408616250938353,
      "loss": 5.5669,
      "mean_token_accuracy": 0.1855427071452141,
      "num_tokens": 76127624.0,
      "step": 37575
    },
    {
      "entropy": 5.9887701034545895,
      "epoch": 3.2841038189286027,
      "grad_norm": 1.671875,
      "learning_rate": 0.00039405972266174494,
      "loss": 5.5429,
      "mean_token_accuracy": 0.18467687517404557,
      "num_tokens": 76138138.0,
      "step": 37580
    },
    {
      "entropy": 5.854605340957642,
      "epoch": 3.284540767281307,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00039403328053047847,
      "loss": 5.421,
      "mean_token_accuracy": 0.19863791465759278,
      "num_tokens": 76148780.0,
      "step": 37585
    },
    {
      "entropy": 5.871038198471069,
      "epoch": 3.284977715634012,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003940068361160913,
      "loss": 5.462,
      "mean_token_accuracy": 0.19543488174676896,
      "num_tokens": 76158468.0,
      "step": 37590
    },
    {
      "entropy": 6.0576495170593265,
      "epoch": 3.2854146639867166,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039398038941909066,
      "loss": 5.6066,
      "mean_token_accuracy": 0.1760564610362053,
      "num_tokens": 76168789.0,
      "step": 37595
    },
    {
      "entropy": 5.989035892486572,
      "epoch": 3.2858516123394215,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003939539404399839,
      "loss": 5.5131,
      "mean_token_accuracy": 0.18428089916706086,
      "num_tokens": 76179355.0,
      "step": 37600
    },
    {
      "entropy": 6.05069670677185,
      "epoch": 3.2862885606921264,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003939274891792781,
      "loss": 5.5719,
      "mean_token_accuracy": 0.1755944237112999,
      "num_tokens": 76189235.0,
      "step": 37605
    },
    {
      "entropy": 5.894771242141724,
      "epoch": 3.286725509044831,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039390103563748086,
      "loss": 5.4835,
      "mean_token_accuracy": 0.19272365868091584,
      "num_tokens": 76200146.0,
      "step": 37610
    },
    {
      "entropy": 5.963190174102783,
      "epoch": 3.2871624573975358,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039387457981509937,
      "loss": 5.5145,
      "mean_token_accuracy": 0.18765036761760712,
      "num_tokens": 76209519.0,
      "step": 37615
    },
    {
      "entropy": 5.980617523193359,
      "epoch": 3.2875994057502402,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003938481217126412,
      "loss": 5.5122,
      "mean_token_accuracy": 0.18685126155614853,
      "num_tokens": 76219316.0,
      "step": 37620
    },
    {
      "entropy": 5.9830162525177,
      "epoch": 3.288036354102945,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003938216613306138,
      "loss": 5.5138,
      "mean_token_accuracy": 0.18763607889413833,
      "num_tokens": 76230707.0,
      "step": 37625
    },
    {
      "entropy": 6.04437780380249,
      "epoch": 3.2884733024556496,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00039379519866952463,
      "loss": 5.5739,
      "mean_token_accuracy": 0.18098372966051102,
      "num_tokens": 76240197.0,
      "step": 37630
    },
    {
      "entropy": 6.015156269073486,
      "epoch": 3.2889102508083545,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039376873372988123,
      "loss": 5.5827,
      "mean_token_accuracy": 0.18167395144701004,
      "num_tokens": 76249292.0,
      "step": 37635
    },
    {
      "entropy": 5.89233078956604,
      "epoch": 3.289347199161059,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039374226651219137,
      "loss": 5.4439,
      "mean_token_accuracy": 0.19256094247102737,
      "num_tokens": 76259598.0,
      "step": 37640
    },
    {
      "entropy": 5.974091720581055,
      "epoch": 3.289784147513764,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003937157970169626,
      "loss": 5.555,
      "mean_token_accuracy": 0.18762751370668412,
      "num_tokens": 76269624.0,
      "step": 37645
    },
    {
      "entropy": 5.998912811279297,
      "epoch": 3.2902210958664684,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0003936893252447026,
      "loss": 5.5677,
      "mean_token_accuracy": 0.18046191930770875,
      "num_tokens": 76280662.0,
      "step": 37650
    },
    {
      "entropy": 6.005195188522339,
      "epoch": 3.2906580442191733,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003936628511959193,
      "loss": 5.5136,
      "mean_token_accuracy": 0.18406639844179154,
      "num_tokens": 76291194.0,
      "step": 37655
    },
    {
      "entropy": 5.936027574539184,
      "epoch": 3.291094992571878,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003936363748711202,
      "loss": 5.5028,
      "mean_token_accuracy": 0.19458998888731002,
      "num_tokens": 76301926.0,
      "step": 37660
    },
    {
      "entropy": 6.052884483337403,
      "epoch": 3.2915319409245827,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003936098962708132,
      "loss": 5.6093,
      "mean_token_accuracy": 0.1788911923766136,
      "num_tokens": 76313126.0,
      "step": 37665
    },
    {
      "entropy": 5.9718669891357425,
      "epoch": 3.2919688892772876,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003935834153955064,
      "loss": 5.4608,
      "mean_token_accuracy": 0.1884316548705101,
      "num_tokens": 76322133.0,
      "step": 37670
    },
    {
      "entropy": 5.867129278182984,
      "epoch": 3.292405837629992,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003935569322457075,
      "loss": 5.44,
      "mean_token_accuracy": 0.19222813099622726,
      "num_tokens": 76332459.0,
      "step": 37675
    },
    {
      "entropy": 5.914612913131714,
      "epoch": 3.292842785982697,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003935304468219245,
      "loss": 5.5396,
      "mean_token_accuracy": 0.18425138741731645,
      "num_tokens": 76342510.0,
      "step": 37680
    },
    {
      "entropy": 6.019817256927491,
      "epoch": 3.2932797343354014,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003935039591246654,
      "loss": 5.5292,
      "mean_token_accuracy": 0.18688387274742127,
      "num_tokens": 76352164.0,
      "step": 37685
    },
    {
      "entropy": 5.9625143051147464,
      "epoch": 3.2937166826881064,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003934774691544383,
      "loss": 5.5224,
      "mean_token_accuracy": 0.1884372815489769,
      "num_tokens": 76362141.0,
      "step": 37690
    },
    {
      "entropy": 5.972754430770874,
      "epoch": 3.294153631040811,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003934509769117513,
      "loss": 5.5775,
      "mean_token_accuracy": 0.18090205788612365,
      "num_tokens": 76372152.0,
      "step": 37695
    },
    {
      "entropy": 6.032066345214844,
      "epoch": 3.2945905793935157,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003934244823971124,
      "loss": 5.5492,
      "mean_token_accuracy": 0.1850225016474724,
      "num_tokens": 76383033.0,
      "step": 37700
    },
    {
      "entropy": 6.029238748550415,
      "epoch": 3.29502752774622,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039339798561102994,
      "loss": 5.5231,
      "mean_token_accuracy": 0.18147324323654174,
      "num_tokens": 76393552.0,
      "step": 37705
    },
    {
      "entropy": 5.900669717788697,
      "epoch": 3.295464476098925,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003933714865540121,
      "loss": 5.4512,
      "mean_token_accuracy": 0.19021745026111603,
      "num_tokens": 76404363.0,
      "step": 37710
    },
    {
      "entropy": 6.010454845428467,
      "epoch": 3.29590142445163,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039334498522656706,
      "loss": 5.506,
      "mean_token_accuracy": 0.18578242510557175,
      "num_tokens": 76413483.0,
      "step": 37715
    },
    {
      "entropy": 5.897186517715454,
      "epoch": 3.2963383728043345,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0003933184816292032,
      "loss": 5.5603,
      "mean_token_accuracy": 0.17831592857837678,
      "num_tokens": 76424786.0,
      "step": 37720
    },
    {
      "entropy": 5.921576690673828,
      "epoch": 3.2967753211570394,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003932919757624289,
      "loss": 5.4465,
      "mean_token_accuracy": 0.19238048642873765,
      "num_tokens": 76434923.0,
      "step": 37725
    },
    {
      "entropy": 6.017120599746704,
      "epoch": 3.297212269509744,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039326546762675254,
      "loss": 5.5074,
      "mean_token_accuracy": 0.18881768137216567,
      "num_tokens": 76444999.0,
      "step": 37730
    },
    {
      "entropy": 5.984810447692871,
      "epoch": 3.297649217862449,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039323895722268243,
      "loss": 5.5076,
      "mean_token_accuracy": 0.18658076226711273,
      "num_tokens": 76454869.0,
      "step": 37735
    },
    {
      "entropy": 5.8828388214111325,
      "epoch": 3.2980861662151533,
      "grad_norm": 1.0,
      "learning_rate": 0.0003932124445507273,
      "loss": 5.5153,
      "mean_token_accuracy": 0.18645517379045487,
      "num_tokens": 76465336.0,
      "step": 37740
    },
    {
      "entropy": 5.97086820602417,
      "epoch": 3.298523114567858,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039318592961139545,
      "loss": 5.5371,
      "mean_token_accuracy": 0.18311750888824463,
      "num_tokens": 76476142.0,
      "step": 37745
    },
    {
      "entropy": 6.0107186794281,
      "epoch": 3.2989600629205627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039315941240519557,
      "loss": 5.5562,
      "mean_token_accuracy": 0.1879253253340721,
      "num_tokens": 76486194.0,
      "step": 37750
    },
    {
      "entropy": 6.014800214767456,
      "epoch": 3.2993970112732676,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003931328929326363,
      "loss": 5.621,
      "mean_token_accuracy": 0.17210501432418823,
      "num_tokens": 76496243.0,
      "step": 37755
    },
    {
      "entropy": 5.9610748291015625,
      "epoch": 3.299833959625972,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00039310637119422607,
      "loss": 5.4987,
      "mean_token_accuracy": 0.1847972497344017,
      "num_tokens": 76506661.0,
      "step": 37760
    },
    {
      "entropy": 6.007007360458374,
      "epoch": 3.300270907978677,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000393079847190474,
      "loss": 5.5409,
      "mean_token_accuracy": 0.18880517929792404,
      "num_tokens": 76516812.0,
      "step": 37765
    },
    {
      "entropy": 5.916446208953857,
      "epoch": 3.300707856331382,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003930533209218884,
      "loss": 5.5621,
      "mean_token_accuracy": 0.18705370873212815,
      "num_tokens": 76527280.0,
      "step": 37770
    },
    {
      "entropy": 5.88518500328064,
      "epoch": 3.3011448046840863,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039302679238897826,
      "loss": 5.4193,
      "mean_token_accuracy": 0.19763472080230712,
      "num_tokens": 76536706.0,
      "step": 37775
    },
    {
      "entropy": 5.9412651538848875,
      "epoch": 3.3015817530367912,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003930002615922525,
      "loss": 5.4384,
      "mean_token_accuracy": 0.18809262216091155,
      "num_tokens": 76546574.0,
      "step": 37780
    },
    {
      "entropy": 5.969975900650025,
      "epoch": 3.3020187013894957,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003929737285322198,
      "loss": 5.5192,
      "mean_token_accuracy": 0.18775646686553954,
      "num_tokens": 76556315.0,
      "step": 37785
    },
    {
      "entropy": 5.939244127273559,
      "epoch": 3.3024556497422006,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039294719320938923,
      "loss": 5.58,
      "mean_token_accuracy": 0.18201784938573837,
      "num_tokens": 76565721.0,
      "step": 37790
    },
    {
      "entropy": 6.033815383911133,
      "epoch": 3.302892598094905,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039292065562426966,
      "loss": 5.5972,
      "mean_token_accuracy": 0.1771220624446869,
      "num_tokens": 76575524.0,
      "step": 37795
    },
    {
      "entropy": 6.029710721969605,
      "epoch": 3.30332954644761,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039289411577737014,
      "loss": 5.568,
      "mean_token_accuracy": 0.17915353924036026,
      "num_tokens": 76586653.0,
      "step": 37800
    },
    {
      "entropy": 5.981989622116089,
      "epoch": 3.3037664948003145,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039286757366919963,
      "loss": 5.5215,
      "mean_token_accuracy": 0.18643736392259597,
      "num_tokens": 76597176.0,
      "step": 37805
    },
    {
      "entropy": 5.953076362609863,
      "epoch": 3.3042034431530194,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003928410293002674,
      "loss": 5.5416,
      "mean_token_accuracy": 0.1856747880578041,
      "num_tokens": 76608010.0,
      "step": 37810
    },
    {
      "entropy": 5.941651630401611,
      "epoch": 3.304640391505724,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003928144826710824,
      "loss": 5.4366,
      "mean_token_accuracy": 0.19055067151784896,
      "num_tokens": 76618029.0,
      "step": 37815
    },
    {
      "entropy": 6.017742729187011,
      "epoch": 3.3050773398584288,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039278793378215394,
      "loss": 5.5641,
      "mean_token_accuracy": 0.18367964923381805,
      "num_tokens": 76629529.0,
      "step": 37820
    },
    {
      "entropy": 6.031916284561158,
      "epoch": 3.3055142882111337,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003927613826339912,
      "loss": 5.5981,
      "mean_token_accuracy": 0.17802149504423143,
      "num_tokens": 76640488.0,
      "step": 37825
    },
    {
      "entropy": 5.932537603378296,
      "epoch": 3.305951236563838,
      "grad_norm": 2.546875,
      "learning_rate": 0.00039273482922710343,
      "loss": 5.3847,
      "mean_token_accuracy": 0.19846706837415695,
      "num_tokens": 76650271.0,
      "step": 37830
    },
    {
      "entropy": 5.929218769073486,
      "epoch": 3.3063881849165426,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003927082735619999,
      "loss": 5.5697,
      "mean_token_accuracy": 0.18165494352579117,
      "num_tokens": 76661413.0,
      "step": 37835
    },
    {
      "entropy": 5.902573204040527,
      "epoch": 3.3068251332692475,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00039268171563919,
      "loss": 5.5315,
      "mean_token_accuracy": 0.1852956861257553,
      "num_tokens": 76671055.0,
      "step": 37840
    },
    {
      "entropy": 6.1291420459747314,
      "epoch": 3.3072620816219525,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003926551554591832,
      "loss": 5.683,
      "mean_token_accuracy": 0.1741972491145134,
      "num_tokens": 76683472.0,
      "step": 37845
    },
    {
      "entropy": 6.017670345306397,
      "epoch": 3.307699029974657,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00039262859302248876,
      "loss": 5.5168,
      "mean_token_accuracy": 0.1849704459309578,
      "num_tokens": 76694641.0,
      "step": 37850
    },
    {
      "entropy": 5.975084066390991,
      "epoch": 3.308135978327362,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003926020283296164,
      "loss": 5.5302,
      "mean_token_accuracy": 0.18112099766731263,
      "num_tokens": 76704737.0,
      "step": 37855
    },
    {
      "entropy": 5.9230053424835205,
      "epoch": 3.3085729266800663,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003925754613810754,
      "loss": 5.4792,
      "mean_token_accuracy": 0.19220271408557893,
      "num_tokens": 76714157.0,
      "step": 37860
    },
    {
      "entropy": 5.975358343124389,
      "epoch": 3.309009875032771,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00039254889217737554,
      "loss": 5.5567,
      "mean_token_accuracy": 0.18144042938947677,
      "num_tokens": 76724488.0,
      "step": 37865
    },
    {
      "entropy": 6.013671255111694,
      "epoch": 3.3094468233854757,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003925223207190263,
      "loss": 5.6096,
      "mean_token_accuracy": 0.18470166027545928,
      "num_tokens": 76734700.0,
      "step": 37870
    },
    {
      "entropy": 5.986474847793579,
      "epoch": 3.3098837717381806,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003924957470065373,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1881905436515808,
      "num_tokens": 76744578.0,
      "step": 37875
    },
    {
      "entropy": 5.958262491226196,
      "epoch": 3.310320720090885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039246917104041843,
      "loss": 5.4926,
      "mean_token_accuracy": 0.19116229116916655,
      "num_tokens": 76754690.0,
      "step": 37880
    },
    {
      "entropy": 6.057399463653565,
      "epoch": 3.31075766844359,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003924425928211792,
      "loss": 5.5856,
      "mean_token_accuracy": 0.17861365377902985,
      "num_tokens": 76765197.0,
      "step": 37885
    },
    {
      "entropy": 6.0500819206237795,
      "epoch": 3.3111946167962945,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039241601234932954,
      "loss": 5.5321,
      "mean_token_accuracy": 0.18472992181777953,
      "num_tokens": 76774985.0,
      "step": 37890
    },
    {
      "entropy": 5.9764433860778805,
      "epoch": 3.3116315651489994,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003923894296253792,
      "loss": 5.5539,
      "mean_token_accuracy": 0.18166446536779404,
      "num_tokens": 76784033.0,
      "step": 37895
    },
    {
      "entropy": 5.994999408721924,
      "epoch": 3.3120685135017043,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003923628446498382,
      "loss": 5.5474,
      "mean_token_accuracy": 0.18434872776269912,
      "num_tokens": 76794826.0,
      "step": 37900
    },
    {
      "entropy": 5.939693117141724,
      "epoch": 3.3125054618544088,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003923362574232163,
      "loss": 5.467,
      "mean_token_accuracy": 0.18657045364379882,
      "num_tokens": 76805429.0,
      "step": 37905
    },
    {
      "entropy": 5.994266700744629,
      "epoch": 3.3129424102071137,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003923096679460234,
      "loss": 5.5315,
      "mean_token_accuracy": 0.1798008441925049,
      "num_tokens": 76814880.0,
      "step": 37910
    },
    {
      "entropy": 5.969972801208496,
      "epoch": 3.313379358559818,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003922830762187697,
      "loss": 5.5336,
      "mean_token_accuracy": 0.18697703778743743,
      "num_tokens": 76824828.0,
      "step": 37915
    },
    {
      "entropy": 6.0087450504302975,
      "epoch": 3.313816306912523,
      "grad_norm": 1.09375,
      "learning_rate": 0.000392256482241965,
      "loss": 5.6588,
      "mean_token_accuracy": 0.1762121245265007,
      "num_tokens": 76835140.0,
      "step": 37920
    },
    {
      "entropy": 5.990776491165161,
      "epoch": 3.3142532552652275,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003922298860161197,
      "loss": 5.5555,
      "mean_token_accuracy": 0.1859504297375679,
      "num_tokens": 76844718.0,
      "step": 37925
    },
    {
      "entropy": 6.057602787017823,
      "epoch": 3.3146902036179324,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039220328754174376,
      "loss": 5.6205,
      "mean_token_accuracy": 0.18189049512147903,
      "num_tokens": 76854738.0,
      "step": 37930
    },
    {
      "entropy": 5.9852907180786135,
      "epoch": 3.315127151970637,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039217668681934727,
      "loss": 5.5584,
      "mean_token_accuracy": 0.182184337079525,
      "num_tokens": 76864939.0,
      "step": 37935
    },
    {
      "entropy": 5.957215976715088,
      "epoch": 3.315564100323342,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003921500838494405,
      "loss": 5.5917,
      "mean_token_accuracy": 0.1788998544216156,
      "num_tokens": 76875895.0,
      "step": 37940
    },
    {
      "entropy": 5.965761709213257,
      "epoch": 3.3160010486760463,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039212347863253386,
      "loss": 5.5178,
      "mean_token_accuracy": 0.18583426028490066,
      "num_tokens": 76885415.0,
      "step": 37945
    },
    {
      "entropy": 5.9386804580688475,
      "epoch": 3.316437997028751,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003920968711691374,
      "loss": 5.5605,
      "mean_token_accuracy": 0.1826233059167862,
      "num_tokens": 76896961.0,
      "step": 37950
    },
    {
      "entropy": 6.014612340927124,
      "epoch": 3.316874945381456,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039207026145976167,
      "loss": 5.5328,
      "mean_token_accuracy": 0.18971752524375915,
      "num_tokens": 76906521.0,
      "step": 37955
    },
    {
      "entropy": 5.957805871963501,
      "epoch": 3.3173118937341606,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039204364950491693,
      "loss": 5.4936,
      "mean_token_accuracy": 0.18350694179534913,
      "num_tokens": 76916971.0,
      "step": 37960
    },
    {
      "entropy": 5.995894718170166,
      "epoch": 3.3177488420868655,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003920170353051137,
      "loss": 5.5663,
      "mean_token_accuracy": 0.18174216896295547,
      "num_tokens": 76928070.0,
      "step": 37965
    },
    {
      "entropy": 5.967554187774658,
      "epoch": 3.31818579043957,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039199041886086233,
      "loss": 5.5726,
      "mean_token_accuracy": 0.18072682917118071,
      "num_tokens": 76938020.0,
      "step": 37970
    },
    {
      "entropy": 5.969951868057251,
      "epoch": 3.318622738792275,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003919638001726735,
      "loss": 5.5424,
      "mean_token_accuracy": 0.18956940174102782,
      "num_tokens": 76949371.0,
      "step": 37975
    },
    {
      "entropy": 6.0409787654876705,
      "epoch": 3.3190596871449793,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003919371792410578,
      "loss": 5.4661,
      "mean_token_accuracy": 0.18844878077507018,
      "num_tokens": 76959120.0,
      "step": 37980
    },
    {
      "entropy": 5.888188409805298,
      "epoch": 3.3194966354976843,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039191055606652554,
      "loss": 5.4915,
      "mean_token_accuracy": 0.19199176430702208,
      "num_tokens": 76969493.0,
      "step": 37985
    },
    {
      "entropy": 5.96273980140686,
      "epoch": 3.3199335838503887,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00039188393064958766,
      "loss": 5.5309,
      "mean_token_accuracy": 0.18319804668426515,
      "num_tokens": 76979732.0,
      "step": 37990
    },
    {
      "entropy": 6.023360013961792,
      "epoch": 3.3203705322030936,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039185730299075473,
      "loss": 5.565,
      "mean_token_accuracy": 0.18229323029518127,
      "num_tokens": 76990354.0,
      "step": 37995
    },
    {
      "entropy": 5.987699174880982,
      "epoch": 3.320807480555798,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039183067309053743,
      "loss": 5.4827,
      "mean_token_accuracy": 0.19170013219118118,
      "num_tokens": 76999469.0,
      "step": 38000
    },
    {
      "entropy": 5.930022239685059,
      "epoch": 3.321244428908503,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003918040409494467,
      "loss": 5.4996,
      "mean_token_accuracy": 0.1894169569015503,
      "num_tokens": 77009499.0,
      "step": 38005
    },
    {
      "entropy": 5.959903383255005,
      "epoch": 3.321681377261208,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039177740656799314,
      "loss": 5.5734,
      "mean_token_accuracy": 0.17187613397836685,
      "num_tokens": 77019426.0,
      "step": 38010
    },
    {
      "entropy": 6.005490922927857,
      "epoch": 3.3221183256139124,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003917507699466879,
      "loss": 5.5078,
      "mean_token_accuracy": 0.18326969146728517,
      "num_tokens": 77028274.0,
      "step": 38015
    },
    {
      "entropy": 5.919297504425049,
      "epoch": 3.3225552739666173,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003917241310860416,
      "loss": 5.4922,
      "mean_token_accuracy": 0.19007878005504608,
      "num_tokens": 77038790.0,
      "step": 38020
    },
    {
      "entropy": 5.985548257827759,
      "epoch": 3.322992222319322,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039169748998656536,
      "loss": 5.5025,
      "mean_token_accuracy": 0.1872544601559639,
      "num_tokens": 77047799.0,
      "step": 38025
    },
    {
      "entropy": 5.952177238464356,
      "epoch": 3.3234291706720267,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003916708466487701,
      "loss": 5.6035,
      "mean_token_accuracy": 0.17548037320375443,
      "num_tokens": 77057080.0,
      "step": 38030
    },
    {
      "entropy": 6.078908014297485,
      "epoch": 3.323866119024731,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039164420107316676,
      "loss": 5.5459,
      "mean_token_accuracy": 0.18255301415920258,
      "num_tokens": 77066689.0,
      "step": 38035
    },
    {
      "entropy": 6.023069047927857,
      "epoch": 3.324303067377436,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003916175532602666,
      "loss": 5.5456,
      "mean_token_accuracy": 0.1829133853316307,
      "num_tokens": 77076366.0,
      "step": 38040
    },
    {
      "entropy": 5.909700965881347,
      "epoch": 3.3247400157301406,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003915909032105808,
      "loss": 5.4842,
      "mean_token_accuracy": 0.1926823765039444,
      "num_tokens": 77085825.0,
      "step": 38045
    },
    {
      "entropy": 6.040801048278809,
      "epoch": 3.3251769640828455,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003915642509246202,
      "loss": 5.4751,
      "mean_token_accuracy": 0.19056106507778167,
      "num_tokens": 77095249.0,
      "step": 38050
    },
    {
      "entropy": 5.8517450332641605,
      "epoch": 3.32561391243555,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00039153759640289627,
      "loss": 5.4564,
      "mean_token_accuracy": 0.18672899305820465,
      "num_tokens": 77105323.0,
      "step": 38055
    },
    {
      "entropy": 5.977421045303345,
      "epoch": 3.326050860788255,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039151093964592007,
      "loss": 5.5683,
      "mean_token_accuracy": 0.1800731122493744,
      "num_tokens": 77116055.0,
      "step": 38060
    },
    {
      "entropy": 5.991630363464355,
      "epoch": 3.3264878091409598,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003914842806542031,
      "loss": 5.5155,
      "mean_token_accuracy": 0.18604100197553636,
      "num_tokens": 77127293.0,
      "step": 38065
    },
    {
      "entropy": 5.982021999359131,
      "epoch": 3.3269247574936642,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003914576194282566,
      "loss": 5.5005,
      "mean_token_accuracy": 0.18136337548494338,
      "num_tokens": 77136872.0,
      "step": 38070
    },
    {
      "entropy": 6.0315269947052,
      "epoch": 3.327361705846369,
      "grad_norm": 1.125,
      "learning_rate": 0.00039143095596859184,
      "loss": 5.5463,
      "mean_token_accuracy": 0.18055547326803206,
      "num_tokens": 77146561.0,
      "step": 38075
    },
    {
      "entropy": 5.93588924407959,
      "epoch": 3.3277986541990736,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039140429027572043,
      "loss": 5.5103,
      "mean_token_accuracy": 0.18905143290758133,
      "num_tokens": 77156043.0,
      "step": 38080
    },
    {
      "entropy": 5.930223226547241,
      "epoch": 3.3282356025517785,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003913776223501537,
      "loss": 5.5193,
      "mean_token_accuracy": 0.18487158566713333,
      "num_tokens": 77165578.0,
      "step": 38085
    },
    {
      "entropy": 5.963954925537109,
      "epoch": 3.328672550904483,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003913509521924032,
      "loss": 5.525,
      "mean_token_accuracy": 0.18389775604009628,
      "num_tokens": 77175105.0,
      "step": 38090
    },
    {
      "entropy": 5.9723851680755615,
      "epoch": 3.329109499257188,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003913242798029805,
      "loss": 5.5644,
      "mean_token_accuracy": 0.17929794937372207,
      "num_tokens": 77185821.0,
      "step": 38095
    },
    {
      "entropy": 5.954179525375366,
      "epoch": 3.3295464476098924,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003912976051823971,
      "loss": 5.4672,
      "mean_token_accuracy": 0.18897943049669266,
      "num_tokens": 77195028.0,
      "step": 38100
    },
    {
      "entropy": 5.97786660194397,
      "epoch": 3.3299833959625973,
      "grad_norm": 0.921875,
      "learning_rate": 0.0003912709283311648,
      "loss": 5.4933,
      "mean_token_accuracy": 0.18305540233850479,
      "num_tokens": 77206317.0,
      "step": 38105
    },
    {
      "entropy": 5.920809364318847,
      "epoch": 3.3304203443153018,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003912442492497951,
      "loss": 5.4527,
      "mean_token_accuracy": 0.18660086244344712,
      "num_tokens": 77216485.0,
      "step": 38110
    },
    {
      "entropy": 5.928443193435669,
      "epoch": 3.3308572926680067,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003912175679387997,
      "loss": 5.495,
      "mean_token_accuracy": 0.1938502237200737,
      "num_tokens": 77226457.0,
      "step": 38115
    },
    {
      "entropy": 5.8998747825622555,
      "epoch": 3.3312942410207116,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003911908843986906,
      "loss": 5.477,
      "mean_token_accuracy": 0.18881182372570038,
      "num_tokens": 77237673.0,
      "step": 38120
    },
    {
      "entropy": 5.955220031738281,
      "epoch": 3.331731189373416,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003911641986299793,
      "loss": 5.4571,
      "mean_token_accuracy": 0.18633945286273956,
      "num_tokens": 77247620.0,
      "step": 38125
    },
    {
      "entropy": 6.0039308071136475,
      "epoch": 3.3321681377261205,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039113751063317795,
      "loss": 5.4985,
      "mean_token_accuracy": 0.18360200971364976,
      "num_tokens": 77257739.0,
      "step": 38130
    },
    {
      "entropy": 5.966236734390259,
      "epoch": 3.3326050860788254,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003911108204087982,
      "loss": 5.548,
      "mean_token_accuracy": 0.1817763105034828,
      "num_tokens": 77268917.0,
      "step": 38135
    },
    {
      "entropy": 5.951000642776489,
      "epoch": 3.3330420344315304,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003910841279573521,
      "loss": 5.5671,
      "mean_token_accuracy": 0.1847999170422554,
      "num_tokens": 77279225.0,
      "step": 38140
    },
    {
      "entropy": 5.965719842910767,
      "epoch": 3.333478982784235,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003910574332793516,
      "loss": 5.5029,
      "mean_token_accuracy": 0.18839203417301179,
      "num_tokens": 77289504.0,
      "step": 38145
    },
    {
      "entropy": 5.973706436157227,
      "epoch": 3.3339159311369397,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039103073637530877,
      "loss": 5.5569,
      "mean_token_accuracy": 0.18513923436403273,
      "num_tokens": 77299663.0,
      "step": 38150
    },
    {
      "entropy": 5.982092094421387,
      "epoch": 3.334352879489644,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039100403724573555,
      "loss": 5.5092,
      "mean_token_accuracy": 0.18891769349575044,
      "num_tokens": 77308755.0,
      "step": 38155
    },
    {
      "entropy": 5.9537755966186525,
      "epoch": 3.334789827842349,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003909773358911441,
      "loss": 5.4492,
      "mean_token_accuracy": 0.19170660823583602,
      "num_tokens": 77318351.0,
      "step": 38160
    },
    {
      "entropy": 5.882534503936768,
      "epoch": 3.3352267761950536,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039095063231204657,
      "loss": 5.3909,
      "mean_token_accuracy": 0.19663702547550202,
      "num_tokens": 77328091.0,
      "step": 38165
    },
    {
      "entropy": 5.903151369094848,
      "epoch": 3.3356637245477585,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003909239265089552,
      "loss": 5.4966,
      "mean_token_accuracy": 0.18616329729557038,
      "num_tokens": 77338209.0,
      "step": 38170
    },
    {
      "entropy": 6.01740870475769,
      "epoch": 3.336100672900463,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039089721848238216,
      "loss": 5.4634,
      "mean_token_accuracy": 0.18466396033763885,
      "num_tokens": 77349811.0,
      "step": 38175
    },
    {
      "entropy": 5.97958927154541,
      "epoch": 3.336537621253168,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003908705082328397,
      "loss": 5.6024,
      "mean_token_accuracy": 0.17686805576086045,
      "num_tokens": 77359995.0,
      "step": 38180
    },
    {
      "entropy": 5.8986376285552975,
      "epoch": 3.3369745696058724,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003908437957608403,
      "loss": 5.446,
      "mean_token_accuracy": 0.19036893397569657,
      "num_tokens": 77371149.0,
      "step": 38185
    },
    {
      "entropy": 5.9387514114379885,
      "epoch": 3.3374115179585773,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003908170810668961,
      "loss": 5.5111,
      "mean_token_accuracy": 0.18421502858400346,
      "num_tokens": 77381107.0,
      "step": 38190
    },
    {
      "entropy": 6.019530487060547,
      "epoch": 3.337848466311282,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039079036415151967,
      "loss": 5.6008,
      "mean_token_accuracy": 0.17353934347629546,
      "num_tokens": 77391982.0,
      "step": 38195
    },
    {
      "entropy": 5.974614238739013,
      "epoch": 3.3382854146639866,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003907636450152232,
      "loss": 5.5135,
      "mean_token_accuracy": 0.18391443639993668,
      "num_tokens": 77403129.0,
      "step": 38200
    },
    {
      "entropy": 5.955299425125122,
      "epoch": 3.3387223630166916,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003907369236585195,
      "loss": 5.5667,
      "mean_token_accuracy": 0.18247307389974593,
      "num_tokens": 77412562.0,
      "step": 38205
    },
    {
      "entropy": 5.960418558120727,
      "epoch": 3.339159311369396,
      "grad_norm": 1.015625,
      "learning_rate": 0.000390710200081921,
      "loss": 5.5304,
      "mean_token_accuracy": 0.18353622406721115,
      "num_tokens": 77422693.0,
      "step": 38210
    },
    {
      "entropy": 5.959579372406006,
      "epoch": 3.339596259722101,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003906834742859402,
      "loss": 5.5066,
      "mean_token_accuracy": 0.1880863353610039,
      "num_tokens": 77432455.0,
      "step": 38215
    },
    {
      "entropy": 5.951659774780273,
      "epoch": 3.3400332080748054,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003906567462710897,
      "loss": 5.5403,
      "mean_token_accuracy": 0.18171169012784957,
      "num_tokens": 77442785.0,
      "step": 38220
    },
    {
      "entropy": 5.979417181015014,
      "epoch": 3.3404701564275103,
      "grad_norm": 1.0,
      "learning_rate": 0.00039063001603788217,
      "loss": 5.5049,
      "mean_token_accuracy": 0.17897011786699296,
      "num_tokens": 77453307.0,
      "step": 38225
    },
    {
      "entropy": 6.014841318130493,
      "epoch": 3.340907104780215,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003906032835868304,
      "loss": 5.6313,
      "mean_token_accuracy": 0.17932258993387223,
      "num_tokens": 77463851.0,
      "step": 38230
    },
    {
      "entropy": 5.994874763488769,
      "epoch": 3.3413440531329197,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039057654891844716,
      "loss": 5.5427,
      "mean_token_accuracy": 0.18095465749502182,
      "num_tokens": 77473391.0,
      "step": 38235
    },
    {
      "entropy": 5.978799772262573,
      "epoch": 3.341781001485624,
      "grad_norm": 1.140625,
      "learning_rate": 0.000390549812033245,
      "loss": 5.5391,
      "mean_token_accuracy": 0.17963542491197587,
      "num_tokens": 77482750.0,
      "step": 38240
    },
    {
      "entropy": 5.917666912078857,
      "epoch": 3.342217949838329,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000390523072931737,
      "loss": 5.4936,
      "mean_token_accuracy": 0.18591111451387404,
      "num_tokens": 77492637.0,
      "step": 38245
    },
    {
      "entropy": 6.015812778472901,
      "epoch": 3.342654898191034,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003904963316144359,
      "loss": 5.5967,
      "mean_token_accuracy": 0.178884157538414,
      "num_tokens": 77502877.0,
      "step": 38250
    },
    {
      "entropy": 5.9637538433074955,
      "epoch": 3.3430918465437385,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039046958808185454,
      "loss": 5.536,
      "mean_token_accuracy": 0.18998703509569168,
      "num_tokens": 77513630.0,
      "step": 38255
    },
    {
      "entropy": 5.941945600509643,
      "epoch": 3.3435287948964434,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039044284233450614,
      "loss": 5.4637,
      "mean_token_accuracy": 0.18986955732107164,
      "num_tokens": 77523801.0,
      "step": 38260
    },
    {
      "entropy": 5.929589653015137,
      "epoch": 3.343965743249148,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039041609437290346,
      "loss": 5.5193,
      "mean_token_accuracy": 0.19104781299829482,
      "num_tokens": 77533985.0,
      "step": 38265
    },
    {
      "entropy": 5.941238307952881,
      "epoch": 3.3444026916018528,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003903893441975596,
      "loss": 5.4464,
      "mean_token_accuracy": 0.1950631856918335,
      "num_tokens": 77543713.0,
      "step": 38270
    },
    {
      "entropy": 5.876354694366455,
      "epoch": 3.3448396399545572,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003903625918089877,
      "loss": 5.5413,
      "mean_token_accuracy": 0.18470236510038376,
      "num_tokens": 77553945.0,
      "step": 38275
    },
    {
      "entropy": 5.886843538284301,
      "epoch": 3.345276588307262,
      "grad_norm": 1.640625,
      "learning_rate": 0.0003903358372077008,
      "loss": 5.4443,
      "mean_token_accuracy": 0.19267356842756272,
      "num_tokens": 77563615.0,
      "step": 38280
    },
    {
      "entropy": 5.985167407989502,
      "epoch": 3.3457135366599666,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003903090803942121,
      "loss": 5.4781,
      "mean_token_accuracy": 0.18713676929473877,
      "num_tokens": 77573886.0,
      "step": 38285
    },
    {
      "entropy": 5.995095443725586,
      "epoch": 3.3461504850126715,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039028232136903483,
      "loss": 5.5534,
      "mean_token_accuracy": 0.18136465400457383,
      "num_tokens": 77585017.0,
      "step": 38290
    },
    {
      "entropy": 5.9383885860443115,
      "epoch": 3.346587433365376,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00039025556013268226,
      "loss": 5.5316,
      "mean_token_accuracy": 0.18761617094278335,
      "num_tokens": 77594605.0,
      "step": 38295
    },
    {
      "entropy": 6.003759431838989,
      "epoch": 3.347024381718081,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039022879668566764,
      "loss": 5.4658,
      "mean_token_accuracy": 0.18539872467517854,
      "num_tokens": 77603262.0,
      "step": 38300
    },
    {
      "entropy": 5.887278318405151,
      "epoch": 3.347461330070786,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003902020310285043,
      "loss": 5.4887,
      "mean_token_accuracy": 0.18342002630233764,
      "num_tokens": 77612963.0,
      "step": 38305
    },
    {
      "entropy": 5.976710939407349,
      "epoch": 3.3478982784234903,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003901752631617056,
      "loss": 5.5047,
      "mean_token_accuracy": 0.1851601257920265,
      "num_tokens": 77623237.0,
      "step": 38310
    },
    {
      "entropy": 5.958795261383057,
      "epoch": 3.348335226776195,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003901484930857851,
      "loss": 5.4683,
      "mean_token_accuracy": 0.19027435630559922,
      "num_tokens": 77632513.0,
      "step": 38315
    },
    {
      "entropy": 5.886477279663086,
      "epoch": 3.3487721751288997,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003901217208012561,
      "loss": 5.4722,
      "mean_token_accuracy": 0.19267069399356843,
      "num_tokens": 77643160.0,
      "step": 38320
    },
    {
      "entropy": 5.914928388595581,
      "epoch": 3.3492091234816046,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039009494630863216,
      "loss": 5.5203,
      "mean_token_accuracy": 0.1848937004804611,
      "num_tokens": 77653311.0,
      "step": 38325
    },
    {
      "entropy": 5.972085428237915,
      "epoch": 3.349646071834309,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039006816960842695,
      "loss": 5.5135,
      "mean_token_accuracy": 0.18732283711433412,
      "num_tokens": 77663142.0,
      "step": 38330
    },
    {
      "entropy": 5.998404884338379,
      "epoch": 3.350083020187014,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003900413907011538,
      "loss": 5.5921,
      "mean_token_accuracy": 0.17813536524772644,
      "num_tokens": 77675143.0,
      "step": 38335
    },
    {
      "entropy": 5.959784460067749,
      "epoch": 3.3505199685397185,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003900146095873266,
      "loss": 5.515,
      "mean_token_accuracy": 0.19208359271287917,
      "num_tokens": 77685464.0,
      "step": 38340
    },
    {
      "entropy": 5.984517431259155,
      "epoch": 3.3509569168924234,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003899878262674588,
      "loss": 5.5267,
      "mean_token_accuracy": 0.18703874796628953,
      "num_tokens": 77695749.0,
      "step": 38345
    },
    {
      "entropy": 5.968956995010376,
      "epoch": 3.351393865245128,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003899610407420643,
      "loss": 5.4584,
      "mean_token_accuracy": 0.18548254370689393,
      "num_tokens": 77705678.0,
      "step": 38350
    },
    {
      "entropy": 6.015661954879761,
      "epoch": 3.3518308135978327,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003899342530116569,
      "loss": 5.5596,
      "mean_token_accuracy": 0.17999957501888275,
      "num_tokens": 77716481.0,
      "step": 38355
    },
    {
      "entropy": 5.95823450088501,
      "epoch": 3.3522677619505377,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003899074630767502,
      "loss": 5.5577,
      "mean_token_accuracy": 0.17870397716760636,
      "num_tokens": 77726977.0,
      "step": 38360
    },
    {
      "entropy": 5.874106454849243,
      "epoch": 3.352704710303242,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038988067093785817,
      "loss": 5.449,
      "mean_token_accuracy": 0.1965733453631401,
      "num_tokens": 77736624.0,
      "step": 38365
    },
    {
      "entropy": 5.947691631317139,
      "epoch": 3.353141658655947,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038985387659549466,
      "loss": 5.5143,
      "mean_token_accuracy": 0.19268915355205535,
      "num_tokens": 77745964.0,
      "step": 38370
    },
    {
      "entropy": 5.98395938873291,
      "epoch": 3.3535786070086515,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038982708005017366,
      "loss": 5.535,
      "mean_token_accuracy": 0.1849605306982994,
      "num_tokens": 77755801.0,
      "step": 38375
    },
    {
      "entropy": 6.002136516571045,
      "epoch": 3.3540155553613564,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003898002813024091,
      "loss": 5.555,
      "mean_token_accuracy": 0.18078541308641433,
      "num_tokens": 77765604.0,
      "step": 38380
    },
    {
      "entropy": 6.048355770111084,
      "epoch": 3.354452503714061,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003897734803527149,
      "loss": 5.5645,
      "mean_token_accuracy": 0.18402449190616607,
      "num_tokens": 77775754.0,
      "step": 38385
    },
    {
      "entropy": 6.064676094055176,
      "epoch": 3.354889452066766,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038974667720160534,
      "loss": 5.6763,
      "mean_token_accuracy": 0.1768420457839966,
      "num_tokens": 77786063.0,
      "step": 38390
    },
    {
      "entropy": 5.915525007247925,
      "epoch": 3.3553264004194703,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038971987184959427,
      "loss": 5.4724,
      "mean_token_accuracy": 0.19271021634340285,
      "num_tokens": 77795217.0,
      "step": 38395
    },
    {
      "entropy": 5.933238697052002,
      "epoch": 3.355763348772175,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000389693064297196,
      "loss": 5.5657,
      "mean_token_accuracy": 0.18376415073871613,
      "num_tokens": 77804637.0,
      "step": 38400
    },
    {
      "entropy": 6.018787288665772,
      "epoch": 3.3562002971248797,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003896662545449247,
      "loss": 5.5639,
      "mean_token_accuracy": 0.1804865062236786,
      "num_tokens": 77816048.0,
      "step": 38405
    },
    {
      "entropy": 5.988553476333618,
      "epoch": 3.3566372454775846,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003896394425932944,
      "loss": 5.4865,
      "mean_token_accuracy": 0.1873774603009224,
      "num_tokens": 77826385.0,
      "step": 38410
    },
    {
      "entropy": 5.914698266983033,
      "epoch": 3.3570741938302895,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038961262844281966,
      "loss": 5.565,
      "mean_token_accuracy": 0.17880983650684357,
      "num_tokens": 77835744.0,
      "step": 38415
    },
    {
      "entropy": 5.966506624221802,
      "epoch": 3.357511142182994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003895858120940146,
      "loss": 5.5318,
      "mean_token_accuracy": 0.18377121388912201,
      "num_tokens": 77845883.0,
      "step": 38420
    },
    {
      "entropy": 5.976754093170166,
      "epoch": 3.357948090535699,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038955899354739364,
      "loss": 5.4959,
      "mean_token_accuracy": 0.18367176800966262,
      "num_tokens": 77856213.0,
      "step": 38425
    },
    {
      "entropy": 5.986442184448242,
      "epoch": 3.3583850388884033,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038953217280347113,
      "loss": 5.5784,
      "mean_token_accuracy": 0.18318309336900712,
      "num_tokens": 77867266.0,
      "step": 38430
    },
    {
      "entropy": 6.022797870635986,
      "epoch": 3.3588219872411083,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003895053498627616,
      "loss": 5.6006,
      "mean_token_accuracy": 0.18470653444528579,
      "num_tokens": 77878358.0,
      "step": 38435
    },
    {
      "entropy": 6.019107723236084,
      "epoch": 3.3592589355938127,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003894785247257794,
      "loss": 5.6129,
      "mean_token_accuracy": 0.18217793107032776,
      "num_tokens": 77889191.0,
      "step": 38440
    },
    {
      "entropy": 5.952113151550293,
      "epoch": 3.3596958839465176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003894516973930392,
      "loss": 5.5533,
      "mean_token_accuracy": 0.1866435319185257,
      "num_tokens": 77898937.0,
      "step": 38445
    },
    {
      "entropy": 5.951045036315918,
      "epoch": 3.360132832299222,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003894248678650553,
      "loss": 5.4506,
      "mean_token_accuracy": 0.18470935374498368,
      "num_tokens": 77907440.0,
      "step": 38450
    },
    {
      "entropy": 5.965227746963501,
      "epoch": 3.360569780651927,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003893980361423426,
      "loss": 5.5215,
      "mean_token_accuracy": 0.18979375809431076,
      "num_tokens": 77917471.0,
      "step": 38455
    },
    {
      "entropy": 5.986183786392212,
      "epoch": 3.3610067290046315,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003893712022254156,
      "loss": 5.4603,
      "mean_token_accuracy": 0.18671342432498933,
      "num_tokens": 77927292.0,
      "step": 38460
    },
    {
      "entropy": 5.9653349876403805,
      "epoch": 3.3614436773573364,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038934436611478895,
      "loss": 5.5832,
      "mean_token_accuracy": 0.18385089784860612,
      "num_tokens": 77937870.0,
      "step": 38465
    },
    {
      "entropy": 5.971484804153443,
      "epoch": 3.3618806257100413,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038931752781097757,
      "loss": 5.504,
      "mean_token_accuracy": 0.18698930442333223,
      "num_tokens": 77947369.0,
      "step": 38470
    },
    {
      "entropy": 5.993114185333252,
      "epoch": 3.362317574062746,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038929068731449604,
      "loss": 5.5796,
      "mean_token_accuracy": 0.18211345821619035,
      "num_tokens": 77958314.0,
      "step": 38475
    },
    {
      "entropy": 6.023623466491699,
      "epoch": 3.3627545224154503,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003892638446258592,
      "loss": 5.6294,
      "mean_token_accuracy": 0.17568876445293427,
      "num_tokens": 77968734.0,
      "step": 38480
    },
    {
      "entropy": 5.979064321517944,
      "epoch": 3.363191470768155,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000389236999745582,
      "loss": 5.5469,
      "mean_token_accuracy": 0.1807507485151291,
      "num_tokens": 77978964.0,
      "step": 38485
    },
    {
      "entropy": 5.990039491653443,
      "epoch": 3.36362841912086,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003892101526741792,
      "loss": 5.4865,
      "mean_token_accuracy": 0.19139208495616913,
      "num_tokens": 77989530.0,
      "step": 38490
    },
    {
      "entropy": 5.980192613601685,
      "epoch": 3.3640653674735645,
      "grad_norm": 1.015625,
      "learning_rate": 0.000389183303412166,
      "loss": 5.4994,
      "mean_token_accuracy": 0.18829589635133742,
      "num_tokens": 77999870.0,
      "step": 38495
    },
    {
      "entropy": 5.963641548156739,
      "epoch": 3.3645023158262695,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038915645196005704,
      "loss": 5.529,
      "mean_token_accuracy": 0.18477287888526917,
      "num_tokens": 78010047.0,
      "step": 38500
    },
    {
      "entropy": 5.945576238632202,
      "epoch": 3.364939264178974,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003891295983183676,
      "loss": 5.4582,
      "mean_token_accuracy": 0.18618826568126678,
      "num_tokens": 78019834.0,
      "step": 38505
    },
    {
      "entropy": 5.960300302505493,
      "epoch": 3.365376212531679,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038910274248761263,
      "loss": 5.5072,
      "mean_token_accuracy": 0.18199365288019181,
      "num_tokens": 78029639.0,
      "step": 38510
    },
    {
      "entropy": 5.952827310562133,
      "epoch": 3.3658131608843833,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003890758844683072,
      "loss": 5.4798,
      "mean_token_accuracy": 0.186307629942894,
      "num_tokens": 78039638.0,
      "step": 38515
    },
    {
      "entropy": 5.929330205917358,
      "epoch": 3.3662501092370882,
      "grad_norm": 1.0,
      "learning_rate": 0.00038904902426096664,
      "loss": 5.5193,
      "mean_token_accuracy": 0.18266571462154388,
      "num_tokens": 78050666.0,
      "step": 38520
    },
    {
      "entropy": 5.928685188293457,
      "epoch": 3.3666870575897927,
      "grad_norm": 2.515625,
      "learning_rate": 0.000389022161866106,
      "loss": 5.4439,
      "mean_token_accuracy": 0.19286171048879625,
      "num_tokens": 78060292.0,
      "step": 38525
    },
    {
      "entropy": 5.920567560195923,
      "epoch": 3.3671240059424976,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038899529728424057,
      "loss": 5.53,
      "mean_token_accuracy": 0.18573201894760133,
      "num_tokens": 78069900.0,
      "step": 38530
    },
    {
      "entropy": 5.904878997802735,
      "epoch": 3.367560954295202,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003889684305158856,
      "loss": 5.4762,
      "mean_token_accuracy": 0.1884980320930481,
      "num_tokens": 78078788.0,
      "step": 38535
    },
    {
      "entropy": 5.951647758483887,
      "epoch": 3.367997902647907,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003889415615615564,
      "loss": 5.546,
      "mean_token_accuracy": 0.18414795845746995,
      "num_tokens": 78087612.0,
      "step": 38540
    },
    {
      "entropy": 5.966780519485473,
      "epoch": 3.368434851000612,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00038891469042176834,
      "loss": 5.4929,
      "mean_token_accuracy": 0.19024239778518676,
      "num_tokens": 78098399.0,
      "step": 38545
    },
    {
      "entropy": 6.000438451766968,
      "epoch": 3.3688717993533164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003888878170970369,
      "loss": 5.5154,
      "mean_token_accuracy": 0.17990904599428176,
      "num_tokens": 78107956.0,
      "step": 38550
    },
    {
      "entropy": 6.011092281341552,
      "epoch": 3.3693087477060213,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038886094158787726,
      "loss": 5.5643,
      "mean_token_accuracy": 0.18139724135398866,
      "num_tokens": 78119613.0,
      "step": 38555
    },
    {
      "entropy": 6.022885847091675,
      "epoch": 3.3697456960587258,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003888340638948052,
      "loss": 5.52,
      "mean_token_accuracy": 0.1892743766307831,
      "num_tokens": 78128986.0,
      "step": 38560
    },
    {
      "entropy": 5.943691921234131,
      "epoch": 3.3701826444114307,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038880718401833623,
      "loss": 5.568,
      "mean_token_accuracy": 0.1855802744626999,
      "num_tokens": 78139085.0,
      "step": 38565
    },
    {
      "entropy": 5.91651029586792,
      "epoch": 3.370619592764135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003887803019589858,
      "loss": 5.4218,
      "mean_token_accuracy": 0.19327341616153718,
      "num_tokens": 78148780.0,
      "step": 38570
    },
    {
      "entropy": 5.9782496929168705,
      "epoch": 3.37105654111684,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003887534177172695,
      "loss": 5.5497,
      "mean_token_accuracy": 0.18823131322860717,
      "num_tokens": 78159021.0,
      "step": 38575
    },
    {
      "entropy": 5.9892467021942135,
      "epoch": 3.3714934894695445,
      "grad_norm": 1.390625,
      "learning_rate": 0.00038872653129370303,
      "loss": 5.5061,
      "mean_token_accuracy": 0.18852974474430084,
      "num_tokens": 78169553.0,
      "step": 38580
    },
    {
      "entropy": 5.904952096939087,
      "epoch": 3.3719304378222494,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038869964268880224,
      "loss": 5.4228,
      "mean_token_accuracy": 0.19424231946468354,
      "num_tokens": 78179631.0,
      "step": 38585
    },
    {
      "entropy": 5.9902119636535645,
      "epoch": 3.372367386174954,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003886727519030826,
      "loss": 5.4686,
      "mean_token_accuracy": 0.18318344950675963,
      "num_tokens": 78189497.0,
      "step": 38590
    },
    {
      "entropy": 5.869361448287964,
      "epoch": 3.372804334527659,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00038864585893706006,
      "loss": 5.4983,
      "mean_token_accuracy": 0.1890581876039505,
      "num_tokens": 78200786.0,
      "step": 38595
    },
    {
      "entropy": 5.952424383163452,
      "epoch": 3.3732412828803637,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038861896379125034,
      "loss": 5.5038,
      "mean_token_accuracy": 0.18911461383104325,
      "num_tokens": 78209783.0,
      "step": 38600
    },
    {
      "entropy": 5.970980978012085,
      "epoch": 3.373678231233068,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003885920664661695,
      "loss": 5.4921,
      "mean_token_accuracy": 0.18921433091163636,
      "num_tokens": 78218988.0,
      "step": 38605
    },
    {
      "entropy": 5.943610620498657,
      "epoch": 3.374115179585773,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003885651669623331,
      "loss": 5.4894,
      "mean_token_accuracy": 0.18931460827589036,
      "num_tokens": 78228545.0,
      "step": 38610
    },
    {
      "entropy": 5.886772394180298,
      "epoch": 3.3745521279384776,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003885382652802574,
      "loss": 5.4595,
      "mean_token_accuracy": 0.19405080378055573,
      "num_tokens": 78238167.0,
      "step": 38615
    },
    {
      "entropy": 5.914488124847412,
      "epoch": 3.3749890762911825,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003885113614204583,
      "loss": 5.4865,
      "mean_token_accuracy": 0.1898217648267746,
      "num_tokens": 78248257.0,
      "step": 38620
    },
    {
      "entropy": 5.993667078018189,
      "epoch": 3.375426024643887,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038848445538345173,
      "loss": 5.5385,
      "mean_token_accuracy": 0.18730249404907226,
      "num_tokens": 78257406.0,
      "step": 38625
    },
    {
      "entropy": 5.916818237304687,
      "epoch": 3.375862972996592,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003884575471697539,
      "loss": 5.4115,
      "mean_token_accuracy": 0.19323202073574067,
      "num_tokens": 78267022.0,
      "step": 38630
    },
    {
      "entropy": 5.916592264175415,
      "epoch": 3.3762999213492964,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038843063677988093,
      "loss": 5.5528,
      "mean_token_accuracy": 0.18108511865139007,
      "num_tokens": 78277584.0,
      "step": 38635
    },
    {
      "entropy": 5.98019118309021,
      "epoch": 3.3767368697020013,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003884037242143488,
      "loss": 5.5314,
      "mean_token_accuracy": 0.1846074193716049,
      "num_tokens": 78288218.0,
      "step": 38640
    },
    {
      "entropy": 5.9998026371002195,
      "epoch": 3.3771738180547057,
      "grad_norm": 1.0,
      "learning_rate": 0.0003883768094736738,
      "loss": 5.5076,
      "mean_token_accuracy": 0.19051734656095504,
      "num_tokens": 78297585.0,
      "step": 38645
    },
    {
      "entropy": 5.908041334152221,
      "epoch": 3.3776107664074106,
      "grad_norm": 1.0,
      "learning_rate": 0.0003883498925583723,
      "loss": 5.503,
      "mean_token_accuracy": 0.18437434881925582,
      "num_tokens": 78308406.0,
      "step": 38650
    },
    {
      "entropy": 5.968244838714599,
      "epoch": 3.3780477147601156,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003883229734689604,
      "loss": 5.4741,
      "mean_token_accuracy": 0.18904346823692322,
      "num_tokens": 78318151.0,
      "step": 38655
    },
    {
      "entropy": 5.9550652503967285,
      "epoch": 3.37848466311282,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003882960522059545,
      "loss": 5.4722,
      "mean_token_accuracy": 0.18757736682891846,
      "num_tokens": 78327316.0,
      "step": 38660
    },
    {
      "entropy": 5.897326993942261,
      "epoch": 3.378921611465525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003882691287698709,
      "loss": 5.4601,
      "mean_token_accuracy": 0.19015639871358872,
      "num_tokens": 78337564.0,
      "step": 38665
    },
    {
      "entropy": 6.00125470161438,
      "epoch": 3.3793585598182294,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003882422031612261,
      "loss": 5.5305,
      "mean_token_accuracy": 0.18385094553232192,
      "num_tokens": 78347796.0,
      "step": 38670
    },
    {
      "entropy": 6.0520414352417,
      "epoch": 3.3797955081709343,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003882152753805366,
      "loss": 5.5904,
      "mean_token_accuracy": 0.17712444365024566,
      "num_tokens": 78357269.0,
      "step": 38675
    },
    {
      "entropy": 5.993684196472168,
      "epoch": 3.380232456523639,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00038818834542831863,
      "loss": 5.4828,
      "mean_token_accuracy": 0.1927453339099884,
      "num_tokens": 78368001.0,
      "step": 38680
    },
    {
      "entropy": 5.894858312606812,
      "epoch": 3.3806694048763437,
      "grad_norm": 1.0625,
      "learning_rate": 0.000388161413305089,
      "loss": 5.4102,
      "mean_token_accuracy": 0.18968380093574524,
      "num_tokens": 78377908.0,
      "step": 38685
    },
    {
      "entropy": 5.936956310272217,
      "epoch": 3.381106353229048,
      "grad_norm": 1.25,
      "learning_rate": 0.0003881344790113641,
      "loss": 5.6043,
      "mean_token_accuracy": 0.182046140730381,
      "num_tokens": 78389450.0,
      "step": 38690
    },
    {
      "entropy": 6.017686080932617,
      "epoch": 3.381543301581753,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038810754254766066,
      "loss": 5.5225,
      "mean_token_accuracy": 0.18112950772047043,
      "num_tokens": 78400397.0,
      "step": 38695
    },
    {
      "entropy": 5.958931636810303,
      "epoch": 3.3819802499344576,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038808060391449525,
      "loss": 5.4421,
      "mean_token_accuracy": 0.1844001978635788,
      "num_tokens": 78410322.0,
      "step": 38700
    },
    {
      "entropy": 5.944100284576416,
      "epoch": 3.3824171982871625,
      "grad_norm": 1.375,
      "learning_rate": 0.0003880536631123847,
      "loss": 5.5374,
      "mean_token_accuracy": 0.17720816731452943,
      "num_tokens": 78420816.0,
      "step": 38705
    },
    {
      "entropy": 5.906044292449951,
      "epoch": 3.3828541466398674,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003880267201418455,
      "loss": 5.5305,
      "mean_token_accuracy": 0.18985736817121507,
      "num_tokens": 78430217.0,
      "step": 38710
    },
    {
      "entropy": 6.017871761322022,
      "epoch": 3.383291094992572,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003879997750033946,
      "loss": 5.5122,
      "mean_token_accuracy": 0.1846362754702568,
      "num_tokens": 78439938.0,
      "step": 38715
    },
    {
      "entropy": 5.991017866134643,
      "epoch": 3.3837280433452768,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003879728276975489,
      "loss": 5.4956,
      "mean_token_accuracy": 0.1838059216737747,
      "num_tokens": 78450805.0,
      "step": 38720
    },
    {
      "entropy": 5.929128694534302,
      "epoch": 3.3841649916979812,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00038794587822482506,
      "loss": 5.495,
      "mean_token_accuracy": 0.18872111439704894,
      "num_tokens": 78460653.0,
      "step": 38725
    },
    {
      "entropy": 5.931749391555786,
      "epoch": 3.384601940050686,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038791892658574016,
      "loss": 5.4756,
      "mean_token_accuracy": 0.18867062777280807,
      "num_tokens": 78471457.0,
      "step": 38730
    },
    {
      "entropy": 6.007035636901856,
      "epoch": 3.3850388884033906,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003878919727808111,
      "loss": 5.534,
      "mean_token_accuracy": 0.18221311569213866,
      "num_tokens": 78480834.0,
      "step": 38735
    },
    {
      "entropy": 5.950442838668823,
      "epoch": 3.3854758367560955,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038786501681055483,
      "loss": 5.5386,
      "mean_token_accuracy": 0.19095512181520463,
      "num_tokens": 78491509.0,
      "step": 38740
    },
    {
      "entropy": 6.022509241104126,
      "epoch": 3.3859127851088,
      "grad_norm": 1.65625,
      "learning_rate": 0.0003878380586754884,
      "loss": 5.6117,
      "mean_token_accuracy": 0.17686131745576858,
      "num_tokens": 78502137.0,
      "step": 38745
    },
    {
      "entropy": 5.941155004501343,
      "epoch": 3.386349733461505,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038781109837612886,
      "loss": 5.4124,
      "mean_token_accuracy": 0.19563591182231904,
      "num_tokens": 78511888.0,
      "step": 38750
    },
    {
      "entropy": 5.9362269878387455,
      "epoch": 3.3867866818142094,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003877841359129934,
      "loss": 5.5305,
      "mean_token_accuracy": 0.1758059561252594,
      "num_tokens": 78521553.0,
      "step": 38755
    },
    {
      "entropy": 5.882699012756348,
      "epoch": 3.3872236301669143,
      "grad_norm": 1.0,
      "learning_rate": 0.000387757171286599,
      "loss": 5.3935,
      "mean_token_accuracy": 0.1932707831263542,
      "num_tokens": 78531290.0,
      "step": 38760
    },
    {
      "entropy": 5.89458966255188,
      "epoch": 3.387660578519619,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003877302044974631,
      "loss": 5.4843,
      "mean_token_accuracy": 0.19044706374406814,
      "num_tokens": 78541200.0,
      "step": 38765
    },
    {
      "entropy": 5.956818342208862,
      "epoch": 3.3880975268723237,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003877032355461027,
      "loss": 5.4929,
      "mean_token_accuracy": 0.1889594778418541,
      "num_tokens": 78551075.0,
      "step": 38770
    },
    {
      "entropy": 5.9628297805786135,
      "epoch": 3.388534475225028,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038767626443303525,
      "loss": 5.523,
      "mean_token_accuracy": 0.18992679566144943,
      "num_tokens": 78560935.0,
      "step": 38775
    },
    {
      "entropy": 5.913304567337036,
      "epoch": 3.388971423577733,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000387649291158778,
      "loss": 5.5055,
      "mean_token_accuracy": 0.187005215883255,
      "num_tokens": 78570856.0,
      "step": 38780
    },
    {
      "entropy": 5.942455196380616,
      "epoch": 3.389408371930438,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038762231572384833,
      "loss": 5.5677,
      "mean_token_accuracy": 0.182282891869545,
      "num_tokens": 78581128.0,
      "step": 38785
    },
    {
      "entropy": 6.003952360153198,
      "epoch": 3.3898453202831424,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003875953381287636,
      "loss": 5.4953,
      "mean_token_accuracy": 0.18612758070230484,
      "num_tokens": 78591131.0,
      "step": 38790
    },
    {
      "entropy": 5.972111558914184,
      "epoch": 3.3902822686358474,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003875683583740413,
      "loss": 5.5479,
      "mean_token_accuracy": 0.18965609222650529,
      "num_tokens": 78600470.0,
      "step": 38795
    },
    {
      "entropy": 5.942761993408203,
      "epoch": 3.390719216988552,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003875413764601989,
      "loss": 5.5579,
      "mean_token_accuracy": 0.17626475393772126,
      "num_tokens": 78610916.0,
      "step": 38800
    },
    {
      "entropy": 5.961607217788696,
      "epoch": 3.3911561653412567,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003875143923877539,
      "loss": 5.4432,
      "mean_token_accuracy": 0.1934823825955391,
      "num_tokens": 78620199.0,
      "step": 38805
    },
    {
      "entropy": 6.045629549026489,
      "epoch": 3.391593113693961,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038748740615722395,
      "loss": 5.5598,
      "mean_token_accuracy": 0.18874942809343337,
      "num_tokens": 78629186.0,
      "step": 38810
    },
    {
      "entropy": 5.9125275135040285,
      "epoch": 3.392030062046666,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038746041776912654,
      "loss": 5.4668,
      "mean_token_accuracy": 0.18310863971710206,
      "num_tokens": 78638786.0,
      "step": 38815
    },
    {
      "entropy": 6.024863338470459,
      "epoch": 3.392467010399371,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00038743342722397946,
      "loss": 5.5684,
      "mean_token_accuracy": 0.18177201896905898,
      "num_tokens": 78647904.0,
      "step": 38820
    },
    {
      "entropy": 5.950965929031372,
      "epoch": 3.3929039587520755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003874064345223003,
      "loss": 5.5097,
      "mean_token_accuracy": 0.18138696402311325,
      "num_tokens": 78657765.0,
      "step": 38825
    },
    {
      "entropy": 5.944734811782837,
      "epoch": 3.39334090710478,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038737943966460684,
      "loss": 5.5059,
      "mean_token_accuracy": 0.1848906621336937,
      "num_tokens": 78667871.0,
      "step": 38830
    },
    {
      "entropy": 5.923687410354614,
      "epoch": 3.393777855457485,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038735244265141676,
      "loss": 5.4659,
      "mean_token_accuracy": 0.18894458413124085,
      "num_tokens": 78677824.0,
      "step": 38835
    },
    {
      "entropy": 6.015665292739868,
      "epoch": 3.39421480381019,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003873254434832481,
      "loss": 5.5681,
      "mean_token_accuracy": 0.18041759729385376,
      "num_tokens": 78687498.0,
      "step": 38840
    },
    {
      "entropy": 6.013678455352784,
      "epoch": 3.3946517521628943,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038729844216061845,
      "loss": 5.5726,
      "mean_token_accuracy": 0.1819591626524925,
      "num_tokens": 78698645.0,
      "step": 38845
    },
    {
      "entropy": 5.94770131111145,
      "epoch": 3.395088700515599,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038727143868404597,
      "loss": 5.3974,
      "mean_token_accuracy": 0.19316928684711457,
      "num_tokens": 78707796.0,
      "step": 38850
    },
    {
      "entropy": 5.927992677688598,
      "epoch": 3.3955256488683037,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003872444330540483,
      "loss": 5.5093,
      "mean_token_accuracy": 0.1863206669688225,
      "num_tokens": 78717838.0,
      "step": 38855
    },
    {
      "entropy": 5.992005014419556,
      "epoch": 3.3959625972210086,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003872174252711437,
      "loss": 5.472,
      "mean_token_accuracy": 0.18658279478549958,
      "num_tokens": 78727318.0,
      "step": 38860
    },
    {
      "entropy": 6.0056188106536865,
      "epoch": 3.396399545573713,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038719041533585,
      "loss": 5.5661,
      "mean_token_accuracy": 0.1778501585125923,
      "num_tokens": 78737446.0,
      "step": 38865
    },
    {
      "entropy": 5.9767311096191404,
      "epoch": 3.396836493926418,
      "grad_norm": 1.125,
      "learning_rate": 0.0003871634032486854,
      "loss": 5.5612,
      "mean_token_accuracy": 0.18276410549879074,
      "num_tokens": 78748557.0,
      "step": 38870
    },
    {
      "entropy": 6.082166147232056,
      "epoch": 3.3972734422791224,
      "grad_norm": 1.015625,
      "learning_rate": 0.000387136389010168,
      "loss": 5.6435,
      "mean_token_accuracy": 0.17860193699598312,
      "num_tokens": 78759482.0,
      "step": 38875
    },
    {
      "entropy": 5.906142330169677,
      "epoch": 3.3977103906318273,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038710937262081583,
      "loss": 5.3791,
      "mean_token_accuracy": 0.19751896411180497,
      "num_tokens": 78768959.0,
      "step": 38880
    },
    {
      "entropy": 5.888628578186035,
      "epoch": 3.398147338984532,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003870823540811472,
      "loss": 5.4846,
      "mean_token_accuracy": 0.19466589391231537,
      "num_tokens": 78779631.0,
      "step": 38885
    },
    {
      "entropy": 5.940888690948486,
      "epoch": 3.3985842873372367,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038705533339168023,
      "loss": 5.5223,
      "mean_token_accuracy": 0.19205092638731003,
      "num_tokens": 78789291.0,
      "step": 38890
    },
    {
      "entropy": 5.959935188293457,
      "epoch": 3.3990212356899416,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003870283105529333,
      "loss": 5.534,
      "mean_token_accuracy": 0.18549549281597139,
      "num_tokens": 78799452.0,
      "step": 38895
    },
    {
      "entropy": 5.993732118606568,
      "epoch": 3.399458184042646,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038700128556542465,
      "loss": 5.5101,
      "mean_token_accuracy": 0.18228067755699157,
      "num_tokens": 78809390.0,
      "step": 38900
    },
    {
      "entropy": 5.9248522281646725,
      "epoch": 3.399895132395351,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003869742584296727,
      "loss": 5.4668,
      "mean_token_accuracy": 0.1888127714395523,
      "num_tokens": 78818979.0,
      "step": 38905
    },
    {
      "entropy": 6.016562652587891,
      "epoch": 3.4003320807480555,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003869472291461957,
      "loss": 5.6009,
      "mean_token_accuracy": 0.18279919326305388,
      "num_tokens": 78829773.0,
      "step": 38910
    },
    {
      "entropy": 6.009703874588013,
      "epoch": 3.4007690291007604,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00038692019771551227,
      "loss": 5.512,
      "mean_token_accuracy": 0.1827257201075554,
      "num_tokens": 78839664.0,
      "step": 38915
    },
    {
      "entropy": 5.969226312637329,
      "epoch": 3.401205977453465,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003868931641381408,
      "loss": 5.4809,
      "mean_token_accuracy": 0.18935419023036956,
      "num_tokens": 78849941.0,
      "step": 38920
    },
    {
      "entropy": 6.0106995582580565,
      "epoch": 3.40164292580617,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038686612841459973,
      "loss": 5.559,
      "mean_token_accuracy": 0.17677475959062577,
      "num_tokens": 78860709.0,
      "step": 38925
    },
    {
      "entropy": 5.99626727104187,
      "epoch": 3.4020798741588743,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003868390905454078,
      "loss": 5.5215,
      "mean_token_accuracy": 0.18292857110500335,
      "num_tokens": 78871001.0,
      "step": 38930
    },
    {
      "entropy": 5.9506909370422365,
      "epoch": 3.402516822511579,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003868120505310835,
      "loss": 5.5291,
      "mean_token_accuracy": 0.1865488812327385,
      "num_tokens": 78881033.0,
      "step": 38935
    },
    {
      "entropy": 5.890879678726196,
      "epoch": 3.4029537708642836,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038678500837214543,
      "loss": 5.4683,
      "mean_token_accuracy": 0.19667085260152817,
      "num_tokens": 78890078.0,
      "step": 38940
    },
    {
      "entropy": 5.962704753875732,
      "epoch": 3.4033907192169885,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038675796406911235,
      "loss": 5.5186,
      "mean_token_accuracy": 0.18521210849285125,
      "num_tokens": 78900209.0,
      "step": 38945
    },
    {
      "entropy": 5.998072719573974,
      "epoch": 3.4038276675696935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038673091762250295,
      "loss": 5.5569,
      "mean_token_accuracy": 0.18486496806144714,
      "num_tokens": 78911431.0,
      "step": 38950
    },
    {
      "entropy": 5.966909027099609,
      "epoch": 3.404264615922398,
      "grad_norm": 1.234375,
      "learning_rate": 0.000386703869032836,
      "loss": 5.5707,
      "mean_token_accuracy": 0.1836704283952713,
      "num_tokens": 78922516.0,
      "step": 38955
    },
    {
      "entropy": 5.959451580047608,
      "epoch": 3.404701564275103,
      "grad_norm": 2.15625,
      "learning_rate": 0.0003866768183006303,
      "loss": 5.4262,
      "mean_token_accuracy": 0.1895005762577057,
      "num_tokens": 78933186.0,
      "step": 38960
    },
    {
      "entropy": 5.851574230194092,
      "epoch": 3.4051385126278073,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003866497654264048,
      "loss": 5.3945,
      "mean_token_accuracy": 0.19766322374343873,
      "num_tokens": 78944602.0,
      "step": 38965
    },
    {
      "entropy": 5.849846124649048,
      "epoch": 3.4055754609805122,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003866227104106782,
      "loss": 5.4258,
      "mean_token_accuracy": 0.2016455665230751,
      "num_tokens": 78954336.0,
      "step": 38970
    },
    {
      "entropy": 5.946090126037598,
      "epoch": 3.4060124093332167,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0003865956532539695,
      "loss": 5.5136,
      "mean_token_accuracy": 0.18951081186532975,
      "num_tokens": 78965306.0,
      "step": 38975
    },
    {
      "entropy": 5.947805118560791,
      "epoch": 3.4064493576859216,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003865685939567977,
      "loss": 5.5169,
      "mean_token_accuracy": 0.1806902140378952,
      "num_tokens": 78975528.0,
      "step": 38980
    },
    {
      "entropy": 6.053284978866577,
      "epoch": 3.406886306038626,
      "grad_norm": 1.0,
      "learning_rate": 0.0003865415325196819,
      "loss": 5.6118,
      "mean_token_accuracy": 0.1772738814353943,
      "num_tokens": 78985402.0,
      "step": 38985
    },
    {
      "entropy": 5.956943130493164,
      "epoch": 3.407323254391331,
      "grad_norm": 0.953125,
      "learning_rate": 0.00038651446894314107,
      "loss": 5.534,
      "mean_token_accuracy": 0.18706098645925523,
      "num_tokens": 78996053.0,
      "step": 38990
    },
    {
      "entropy": 5.989487934112549,
      "epoch": 3.4077602027440355,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003864874032276942,
      "loss": 5.5804,
      "mean_token_accuracy": 0.1801146537065506,
      "num_tokens": 79007645.0,
      "step": 38995
    },
    {
      "entropy": 6.005895805358887,
      "epoch": 3.4081971510967404,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003864603353738605,
      "loss": 5.4608,
      "mean_token_accuracy": 0.19336313605308533,
      "num_tokens": 79018649.0,
      "step": 39000
    },
    {
      "epoch": 3.4081971510967404,
      "eval_entropy": 5.766228501214975,
      "eval_loss": 5.572558879852295,
      "eval_mean_token_accuracy": 0.19181000722702235,
      "eval_num_tokens": 79018649.0,
      "eval_runtime": 28.1599,
      "eval_samples_per_second": 1314.422,
      "eval_steps_per_second": 164.312,
      "step": 39000
    }
  ],
  "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": 1.1825277926976e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}