{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.078563313816307,
  "eval_steps": 3000,
  "global_step": 81000,
  "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
    },
    {
      "entropy": 5.9136803150177,
      "epoch": 3.4086340994494453,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003864332653821592,
      "loss": 5.4143,
      "mean_token_accuracy": 0.19374284893274307,
      "num_tokens": 79028549.0,
      "step": 39005
    },
    {
      "entropy": 5.878360176086426,
      "epoch": 3.4090710478021498,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003864061932531095,
      "loss": 5.4775,
      "mean_token_accuracy": 0.18625412732362748,
      "num_tokens": 79037788.0,
      "step": 39010
    },
    {
      "entropy": 6.004088306427002,
      "epoch": 3.4095079961548547,
      "grad_norm": 1.125,
      "learning_rate": 0.00038637911898723067,
      "loss": 5.5384,
      "mean_token_accuracy": 0.18175956308841706,
      "num_tokens": 79048205.0,
      "step": 39015
    },
    {
      "entropy": 5.903505277633667,
      "epoch": 3.409944944507559,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003863520425850419,
      "loss": 5.4267,
      "mean_token_accuracy": 0.19212424904108047,
      "num_tokens": 79057594.0,
      "step": 39020
    },
    {
      "entropy": 5.916986131668091,
      "epoch": 3.410381892860264,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003863249640470627,
      "loss": 5.51,
      "mean_token_accuracy": 0.18893196880817414,
      "num_tokens": 79066815.0,
      "step": 39025
    },
    {
      "entropy": 5.979528951644897,
      "epoch": 3.4108188412129685,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038629788337381227,
      "loss": 5.4648,
      "mean_token_accuracy": 0.18802562057971955,
      "num_tokens": 79077191.0,
      "step": 39030
    },
    {
      "entropy": 5.9166919708251955,
      "epoch": 3.4112557895656734,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003862708005658102,
      "loss": 5.4992,
      "mean_token_accuracy": 0.1859675884246826,
      "num_tokens": 79086958.0,
      "step": 39035
    },
    {
      "entropy": 5.929417753219605,
      "epoch": 3.411692737918378,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038624371562357573,
      "loss": 5.5396,
      "mean_token_accuracy": 0.18497301489114762,
      "num_tokens": 79098298.0,
      "step": 39040
    },
    {
      "entropy": 5.9282959461212155,
      "epoch": 3.412129686271083,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038621662854762865,
      "loss": 5.4889,
      "mean_token_accuracy": 0.1855600282549858,
      "num_tokens": 79107954.0,
      "step": 39045
    },
    {
      "entropy": 5.972679328918457,
      "epoch": 3.4125666346237873,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003861895393384882,
      "loss": 5.5213,
      "mean_token_accuracy": 0.18564339876174926,
      "num_tokens": 79118062.0,
      "step": 39050
    },
    {
      "entropy": 6.002381420135498,
      "epoch": 3.413003582976492,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038616244799667427,
      "loss": 5.5476,
      "mean_token_accuracy": 0.18666295409202577,
      "num_tokens": 79128358.0,
      "step": 39055
    },
    {
      "entropy": 6.027896213531494,
      "epoch": 3.413440531329197,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003861353545227063,
      "loss": 5.5629,
      "mean_token_accuracy": 0.17849205136299134,
      "num_tokens": 79139882.0,
      "step": 39060
    },
    {
      "entropy": 5.991833066940307,
      "epoch": 3.4138774796819016,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003861082589171039,
      "loss": 5.5506,
      "mean_token_accuracy": 0.18426993638277053,
      "num_tokens": 79150748.0,
      "step": 39065
    },
    {
      "entropy": 5.906841802597046,
      "epoch": 3.4143144280346065,
      "grad_norm": 1.125,
      "learning_rate": 0.00038608116118038705,
      "loss": 5.4779,
      "mean_token_accuracy": 0.18780468553304672,
      "num_tokens": 79160088.0,
      "step": 39070
    },
    {
      "entropy": 5.882471418380737,
      "epoch": 3.414751376387311,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003860540613130752,
      "loss": 5.4182,
      "mean_token_accuracy": 0.20036334693431854,
      "num_tokens": 79169539.0,
      "step": 39075
    },
    {
      "entropy": 5.9963689804077145,
      "epoch": 3.415188324740016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038602695931568834,
      "loss": 5.5778,
      "mean_token_accuracy": 0.1851288601756096,
      "num_tokens": 79179505.0,
      "step": 39080
    },
    {
      "entropy": 5.9651861667633055,
      "epoch": 3.4156252730927203,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038599985518874614,
      "loss": 5.4827,
      "mean_token_accuracy": 0.18729677498340608,
      "num_tokens": 79189791.0,
      "step": 39085
    },
    {
      "entropy": 5.992525386810303,
      "epoch": 3.4160622214454253,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038597274893276855,
      "loss": 5.5143,
      "mean_token_accuracy": 0.18420348316431046,
      "num_tokens": 79199708.0,
      "step": 39090
    },
    {
      "entropy": 5.90216703414917,
      "epoch": 3.4164991697981297,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003859456405482755,
      "loss": 5.5173,
      "mean_token_accuracy": 0.191865935921669,
      "num_tokens": 79209624.0,
      "step": 39095
    },
    {
      "entropy": 5.9052510261535645,
      "epoch": 3.4169361181508346,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000385918530035787,
      "loss": 5.4827,
      "mean_token_accuracy": 0.19078044444322587,
      "num_tokens": 79218873.0,
      "step": 39100
    },
    {
      "entropy": 5.947568941116333,
      "epoch": 3.417373066503539,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00038589141739582295,
      "loss": 5.5141,
      "mean_token_accuracy": 0.18461664468050004,
      "num_tokens": 79228838.0,
      "step": 39105
    },
    {
      "entropy": 5.967162609100342,
      "epoch": 3.417810014856244,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003858643026289034,
      "loss": 5.4815,
      "mean_token_accuracy": 0.18859535604715347,
      "num_tokens": 79238185.0,
      "step": 39110
    },
    {
      "entropy": 6.0124382972717285,
      "epoch": 3.418246963208949,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038583718573554845,
      "loss": 5.5021,
      "mean_token_accuracy": 0.18338433653116226,
      "num_tokens": 79248777.0,
      "step": 39115
    },
    {
      "entropy": 5.890842914581299,
      "epoch": 3.4186839115616534,
      "grad_norm": 1.0,
      "learning_rate": 0.0003858100667162781,
      "loss": 5.4128,
      "mean_token_accuracy": 0.1948304668068886,
      "num_tokens": 79258471.0,
      "step": 39120
    },
    {
      "entropy": 5.814173030853271,
      "epoch": 3.419120859914358,
      "grad_norm": 1.0,
      "learning_rate": 0.0003857829455716127,
      "loss": 5.4154,
      "mean_token_accuracy": 0.19068270921707153,
      "num_tokens": 79268483.0,
      "step": 39125
    },
    {
      "entropy": 5.926399564743042,
      "epoch": 3.419557808267063,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003857558223020724,
      "loss": 5.4955,
      "mean_token_accuracy": 0.18877243399620056,
      "num_tokens": 79278497.0,
      "step": 39130
    },
    {
      "entropy": 5.968505334854126,
      "epoch": 3.4199947566197677,
      "grad_norm": 0.90625,
      "learning_rate": 0.0003857286969081773,
      "loss": 5.4532,
      "mean_token_accuracy": 0.19188061952590943,
      "num_tokens": 79288010.0,
      "step": 39135
    },
    {
      "entropy": 5.933250904083252,
      "epoch": 3.420431704972472,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038570156939044784,
      "loss": 5.5265,
      "mean_token_accuracy": 0.19028003960847856,
      "num_tokens": 79298092.0,
      "step": 39140
    },
    {
      "entropy": 5.973419618606568,
      "epoch": 3.420868653325177,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003856744397494042,
      "loss": 5.5555,
      "mean_token_accuracy": 0.18154251873493193,
      "num_tokens": 79308458.0,
      "step": 39145
    },
    {
      "entropy": 6.0469403743743895,
      "epoch": 3.4213056016778816,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003856473079855668,
      "loss": 5.5665,
      "mean_token_accuracy": 0.1798962026834488,
      "num_tokens": 79318740.0,
      "step": 39150
    },
    {
      "entropy": 5.98318657875061,
      "epoch": 3.4217425500305865,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038562017409945614,
      "loss": 5.5324,
      "mean_token_accuracy": 0.19393400847911835,
      "num_tokens": 79329738.0,
      "step": 39155
    },
    {
      "entropy": 5.965099239349366,
      "epoch": 3.422179498383291,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038559303809159254,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1853828176856041,
      "num_tokens": 79341137.0,
      "step": 39160
    },
    {
      "entropy": 5.951865768432617,
      "epoch": 3.422616446735996,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003855658999624965,
      "loss": 5.516,
      "mean_token_accuracy": 0.1867011532187462,
      "num_tokens": 79351601.0,
      "step": 39165
    },
    {
      "entropy": 5.92706332206726,
      "epoch": 3.4230533950887003,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003855387597126886,
      "loss": 5.4998,
      "mean_token_accuracy": 0.18370723426342012,
      "num_tokens": 79360875.0,
      "step": 39170
    },
    {
      "entropy": 5.962479972839356,
      "epoch": 3.4234903434414052,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038551161734268934,
      "loss": 5.5357,
      "mean_token_accuracy": 0.18513910472393036,
      "num_tokens": 79371033.0,
      "step": 39175
    },
    {
      "entropy": 5.999136686325073,
      "epoch": 3.4239272917941097,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038548447285301934,
      "loss": 5.5349,
      "mean_token_accuracy": 0.19030856043100358,
      "num_tokens": 79381237.0,
      "step": 39180
    },
    {
      "entropy": 5.96987361907959,
      "epoch": 3.4243642401468146,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003854573262441993,
      "loss": 5.5666,
      "mean_token_accuracy": 0.17705174535512924,
      "num_tokens": 79392279.0,
      "step": 39185
    },
    {
      "entropy": 5.951351594924927,
      "epoch": 3.4248011884995195,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00038543017751674986,
      "loss": 5.5126,
      "mean_token_accuracy": 0.18397694379091262,
      "num_tokens": 79401723.0,
      "step": 39190
    },
    {
      "entropy": 5.903625202178955,
      "epoch": 3.425238136852224,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003854030266711917,
      "loss": 5.4346,
      "mean_token_accuracy": 0.20080699175596237,
      "num_tokens": 79413436.0,
      "step": 39195
    },
    {
      "entropy": 5.929594850540161,
      "epoch": 3.425675085204929,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00038537587370804567,
      "loss": 5.5196,
      "mean_token_accuracy": 0.1864091783761978,
      "num_tokens": 79423923.0,
      "step": 39200
    },
    {
      "entropy": 5.970825481414795,
      "epoch": 3.4261120335576334,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003853487186278326,
      "loss": 5.4942,
      "mean_token_accuracy": 0.18616314530372619,
      "num_tokens": 79433992.0,
      "step": 39205
    },
    {
      "entropy": 5.954268312454223,
      "epoch": 3.4265489819103383,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00038532156143107323,
      "loss": 5.5352,
      "mean_token_accuracy": 0.18586039990186692,
      "num_tokens": 79444564.0,
      "step": 39210
    },
    {
      "entropy": 5.971515274047851,
      "epoch": 3.4269859302630428,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003852944021182885,
      "loss": 5.5055,
      "mean_token_accuracy": 0.18767796456813812,
      "num_tokens": 79454985.0,
      "step": 39215
    },
    {
      "entropy": 5.945884847640992,
      "epoch": 3.4274228786157477,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003852672406899993,
      "loss": 5.5307,
      "mean_token_accuracy": 0.18681458085775376,
      "num_tokens": 79464959.0,
      "step": 39220
    },
    {
      "entropy": 5.962272071838379,
      "epoch": 3.427859826968452,
      "grad_norm": 0.96875,
      "learning_rate": 0.00038524007714672674,
      "loss": 5.4928,
      "mean_token_accuracy": 0.1882188156247139,
      "num_tokens": 79474679.0,
      "step": 39225
    },
    {
      "entropy": 5.944274377822876,
      "epoch": 3.428296775321157,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003852129114889916,
      "loss": 5.538,
      "mean_token_accuracy": 0.1854730799794197,
      "num_tokens": 79485908.0,
      "step": 39230
    },
    {
      "entropy": 5.941825532913208,
      "epoch": 3.4287337236738615,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003851857437173152,
      "loss": 5.5368,
      "mean_token_accuracy": 0.18915131241083144,
      "num_tokens": 79496209.0,
      "step": 39235
    },
    {
      "entropy": 5.982985162734986,
      "epoch": 3.4291706720265664,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003851585738322184,
      "loss": 5.481,
      "mean_token_accuracy": 0.19352952837944032,
      "num_tokens": 79507351.0,
      "step": 39240
    },
    {
      "entropy": 5.884948492050171,
      "epoch": 3.4296076203792714,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038513140183422256,
      "loss": 5.4755,
      "mean_token_accuracy": 0.18797972947359085,
      "num_tokens": 79517517.0,
      "step": 39245
    },
    {
      "entropy": 5.903098154067993,
      "epoch": 3.430044568731976,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003851042277238486,
      "loss": 5.4506,
      "mean_token_accuracy": 0.19118238240480423,
      "num_tokens": 79528616.0,
      "step": 39250
    },
    {
      "entropy": 5.973155689239502,
      "epoch": 3.4304815170846807,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003850770515016179,
      "loss": 5.5048,
      "mean_token_accuracy": 0.18516432940959932,
      "num_tokens": 79538424.0,
      "step": 39255
    },
    {
      "entropy": 5.989159440994262,
      "epoch": 3.430918465437385,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003850498731680517,
      "loss": 5.5536,
      "mean_token_accuracy": 0.17919782400131226,
      "num_tokens": 79548717.0,
      "step": 39260
    },
    {
      "entropy": 5.973058223724365,
      "epoch": 3.43135541379009,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038502269272367123,
      "loss": 5.4283,
      "mean_token_accuracy": 0.19093876332044601,
      "num_tokens": 79558998.0,
      "step": 39265
    },
    {
      "entropy": 5.986805486679077,
      "epoch": 3.4317923621427946,
      "grad_norm": 1.25,
      "learning_rate": 0.00038499551016899793,
      "loss": 5.5804,
      "mean_token_accuracy": 0.17861207574605942,
      "num_tokens": 79569424.0,
      "step": 39270
    },
    {
      "entropy": 5.959056663513183,
      "epoch": 3.4322293104954995,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038496832550455307,
      "loss": 5.4848,
      "mean_token_accuracy": 0.18593389093875884,
      "num_tokens": 79578670.0,
      "step": 39275
    },
    {
      "entropy": 5.961947679519653,
      "epoch": 3.432666258848204,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038494113873085805,
      "loss": 5.5765,
      "mean_token_accuracy": 0.1827679842710495,
      "num_tokens": 79588883.0,
      "step": 39280
    },
    {
      "entropy": 5.910299396514892,
      "epoch": 3.433103207200909,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038491394984843446,
      "loss": 5.4777,
      "mean_token_accuracy": 0.19332474917173387,
      "num_tokens": 79598158.0,
      "step": 39285
    },
    {
      "entropy": 5.882826662063598,
      "epoch": 3.4335401555536134,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038488675885780366,
      "loss": 5.4389,
      "mean_token_accuracy": 0.19381902366876602,
      "num_tokens": 79608699.0,
      "step": 39290
    },
    {
      "entropy": 5.982617616653442,
      "epoch": 3.4339771039063183,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038485956575948726,
      "loss": 5.4655,
      "mean_token_accuracy": 0.19178287088871002,
      "num_tokens": 79618628.0,
      "step": 39295
    },
    {
      "entropy": 5.968356132507324,
      "epoch": 3.434414052259023,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003848323705540069,
      "loss": 5.4925,
      "mean_token_accuracy": 0.18413717299699783,
      "num_tokens": 79628836.0,
      "step": 39300
    },
    {
      "entropy": 5.921951484680176,
      "epoch": 3.4348510006117277,
      "grad_norm": 1.03125,
      "learning_rate": 0.000384805173241884,
      "loss": 5.4483,
      "mean_token_accuracy": 0.18805354982614517,
      "num_tokens": 79638306.0,
      "step": 39305
    },
    {
      "entropy": 5.89973521232605,
      "epoch": 3.4352879489644326,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038477797382364037,
      "loss": 5.5381,
      "mean_token_accuracy": 0.18413039147853852,
      "num_tokens": 79647955.0,
      "step": 39310
    },
    {
      "entropy": 5.9747837543487545,
      "epoch": 3.435724897317137,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038475077229979774,
      "loss": 5.5291,
      "mean_token_accuracy": 0.184048992395401,
      "num_tokens": 79658701.0,
      "step": 39315
    },
    {
      "entropy": 5.9560527324676515,
      "epoch": 3.436161845669842,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038472356867087776,
      "loss": 5.5598,
      "mean_token_accuracy": 0.1804173320531845,
      "num_tokens": 79669436.0,
      "step": 39320
    },
    {
      "entropy": 5.993048143386841,
      "epoch": 3.4365987940225464,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003846963629374022,
      "loss": 5.5403,
      "mean_token_accuracy": 0.18015642315149308,
      "num_tokens": 79679350.0,
      "step": 39325
    },
    {
      "entropy": 5.957166337966919,
      "epoch": 3.4370357423752513,
      "grad_norm": 1.0,
      "learning_rate": 0.00038466915509989297,
      "loss": 5.4483,
      "mean_token_accuracy": 0.19811357259750367,
      "num_tokens": 79689762.0,
      "step": 39330
    },
    {
      "entropy": 5.93925986289978,
      "epoch": 3.437472690727956,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038464194515887187,
      "loss": 5.4361,
      "mean_token_accuracy": 0.18857277929782867,
      "num_tokens": 79699594.0,
      "step": 39335
    },
    {
      "entropy": 5.928225421905518,
      "epoch": 3.4379096390806607,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003846147331148608,
      "loss": 5.5602,
      "mean_token_accuracy": 0.18473443984985352,
      "num_tokens": 79710403.0,
      "step": 39340
    },
    {
      "entropy": 5.8484124660491945,
      "epoch": 3.438346587433365,
      "grad_norm": 1.578125,
      "learning_rate": 0.00038458751896838176,
      "loss": 5.3593,
      "mean_token_accuracy": 0.19820756167173387,
      "num_tokens": 79721291.0,
      "step": 39345
    },
    {
      "entropy": 5.940815353393555,
      "epoch": 3.43878353578607,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003845603027199566,
      "loss": 5.5395,
      "mean_token_accuracy": 0.1806758910417557,
      "num_tokens": 79732505.0,
      "step": 39350
    },
    {
      "entropy": 6.012080526351928,
      "epoch": 3.439220484138775,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038453308437010747,
      "loss": 5.5465,
      "mean_token_accuracy": 0.18621233701705933,
      "num_tokens": 79743707.0,
      "step": 39355
    },
    {
      "entropy": 6.021396923065185,
      "epoch": 3.4396574324914795,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038450586391935635,
      "loss": 5.5138,
      "mean_token_accuracy": 0.187666654586792,
      "num_tokens": 79754669.0,
      "step": 39360
    },
    {
      "entropy": 5.95346736907959,
      "epoch": 3.4400943808441844,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003844786413682254,
      "loss": 5.5114,
      "mean_token_accuracy": 0.19639771729707717,
      "num_tokens": 79764444.0,
      "step": 39365
    },
    {
      "entropy": 5.961709117889404,
      "epoch": 3.440531329196889,
      "grad_norm": 1.0,
      "learning_rate": 0.0003844514167172368,
      "loss": 5.5688,
      "mean_token_accuracy": 0.17575672268867493,
      "num_tokens": 79774942.0,
      "step": 39370
    },
    {
      "entropy": 6.00725154876709,
      "epoch": 3.440968277549594,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003844241899669126,
      "loss": 5.532,
      "mean_token_accuracy": 0.17570304721593857,
      "num_tokens": 79785710.0,
      "step": 39375
    },
    {
      "entropy": 6.006101894378662,
      "epoch": 3.4414052259022982,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003843969611177751,
      "loss": 5.5667,
      "mean_token_accuracy": 0.1801757276058197,
      "num_tokens": 79795841.0,
      "step": 39380
    },
    {
      "entropy": 5.911998701095581,
      "epoch": 3.441842174255003,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003843697301703466,
      "loss": 5.4387,
      "mean_token_accuracy": 0.18906803131103517,
      "num_tokens": 79805400.0,
      "step": 39385
    },
    {
      "entropy": 5.8997245788574215,
      "epoch": 3.4422791226077076,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003843424971251494,
      "loss": 5.5322,
      "mean_token_accuracy": 0.188840813934803,
      "num_tokens": 79816212.0,
      "step": 39390
    },
    {
      "entropy": 5.946767854690552,
      "epoch": 3.4427160709604125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003843152619827058,
      "loss": 5.5006,
      "mean_token_accuracy": 0.18363959193229676,
      "num_tokens": 79827125.0,
      "step": 39395
    },
    {
      "entropy": 6.020533418655395,
      "epoch": 3.443153019313117,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038428802474353816,
      "loss": 5.5371,
      "mean_token_accuracy": 0.18311394304037093,
      "num_tokens": 79837436.0,
      "step": 39400
    },
    {
      "entropy": 6.026792049407959,
      "epoch": 3.443589967665822,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003842607854081688,
      "loss": 5.5577,
      "mean_token_accuracy": 0.18353070616722106,
      "num_tokens": 79847297.0,
      "step": 39405
    },
    {
      "entropy": 5.955293464660644,
      "epoch": 3.444026916018527,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003842335439771205,
      "loss": 5.4449,
      "mean_token_accuracy": 0.1916890859603882,
      "num_tokens": 79856321.0,
      "step": 39410
    },
    {
      "entropy": 5.9059679985046385,
      "epoch": 3.4444638643712313,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038420630045091556,
      "loss": 5.4595,
      "mean_token_accuracy": 0.1941339612007141,
      "num_tokens": 79866244.0,
      "step": 39415
    },
    {
      "entropy": 5.972674798965454,
      "epoch": 3.4449008127239362,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003841790548300765,
      "loss": 5.5653,
      "mean_token_accuracy": 0.17892435044050217,
      "num_tokens": 79876900.0,
      "step": 39420
    },
    {
      "entropy": 5.998918104171753,
      "epoch": 3.4453377610766407,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038415180711512604,
      "loss": 5.5284,
      "mean_token_accuracy": 0.18519793301820756,
      "num_tokens": 79887864.0,
      "step": 39425
    },
    {
      "entropy": 5.9664875030517575,
      "epoch": 3.4457747094293456,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003841245573065867,
      "loss": 5.4906,
      "mean_token_accuracy": 0.18583797067403793,
      "num_tokens": 79897625.0,
      "step": 39430
    },
    {
      "entropy": 5.971397972106933,
      "epoch": 3.44621165778205,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003840973054049812,
      "loss": 5.5381,
      "mean_token_accuracy": 0.18468852788209916,
      "num_tokens": 79907742.0,
      "step": 39435
    },
    {
      "entropy": 5.940431404113769,
      "epoch": 3.446648606134755,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038407005141083216,
      "loss": 5.4561,
      "mean_token_accuracy": 0.1888837218284607,
      "num_tokens": 79917974.0,
      "step": 39440
    },
    {
      "entropy": 5.930742025375366,
      "epoch": 3.4470855544874595,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003840427953246624,
      "loss": 5.4871,
      "mean_token_accuracy": 0.1888188198208809,
      "num_tokens": 79928089.0,
      "step": 39445
    },
    {
      "entropy": 5.844485282897949,
      "epoch": 3.4475225028401644,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00038401553714699466,
      "loss": 5.4324,
      "mean_token_accuracy": 0.1932738959789276,
      "num_tokens": 79938396.0,
      "step": 39450
    },
    {
      "entropy": 5.992274951934815,
      "epoch": 3.447959451192869,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003839882768783518,
      "loss": 5.5075,
      "mean_token_accuracy": 0.18602386713027955,
      "num_tokens": 79948157.0,
      "step": 39455
    },
    {
      "entropy": 6.022170495986939,
      "epoch": 3.4483963995455738,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003839610145192567,
      "loss": 5.5147,
      "mean_token_accuracy": 0.18482689410448075,
      "num_tokens": 79958653.0,
      "step": 39460
    },
    {
      "entropy": 5.877518224716186,
      "epoch": 3.4488333478982787,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003839337500702323,
      "loss": 5.5063,
      "mean_token_accuracy": 0.18799806535243987,
      "num_tokens": 79969180.0,
      "step": 39465
    },
    {
      "entropy": 5.97568941116333,
      "epoch": 3.449270296250983,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038390648353180145,
      "loss": 5.5843,
      "mean_token_accuracy": 0.1838316097855568,
      "num_tokens": 79978467.0,
      "step": 39470
    },
    {
      "entropy": 5.98711142539978,
      "epoch": 3.4497072446036876,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003838792149044871,
      "loss": 5.4666,
      "mean_token_accuracy": 0.18559375703334807,
      "num_tokens": 79989071.0,
      "step": 39475
    },
    {
      "entropy": 5.9041845321655275,
      "epoch": 3.4501441929563925,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038385194418881247,
      "loss": 5.4883,
      "mean_token_accuracy": 0.18924214988946914,
      "num_tokens": 79999476.0,
      "step": 39480
    },
    {
      "entropy": 5.884802007675171,
      "epoch": 3.4505811413090974,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038382467138530055,
      "loss": 5.4115,
      "mean_token_accuracy": 0.19958099275827407,
      "num_tokens": 80009986.0,
      "step": 39485
    },
    {
      "entropy": 5.9658205032348635,
      "epoch": 3.451018089661802,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038379739649447425,
      "loss": 5.6043,
      "mean_token_accuracy": 0.18145477920770645,
      "num_tokens": 80020916.0,
      "step": 39490
    },
    {
      "entropy": 5.9212377071380615,
      "epoch": 3.451455038014507,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000383770119516857,
      "loss": 5.4556,
      "mean_token_accuracy": 0.1905773863196373,
      "num_tokens": 80030769.0,
      "step": 39495
    },
    {
      "entropy": 5.968823671340942,
      "epoch": 3.4518919863672113,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003837428404529718,
      "loss": 5.5191,
      "mean_token_accuracy": 0.185211181640625,
      "num_tokens": 80040838.0,
      "step": 39500
    },
    {
      "entropy": 5.972611141204834,
      "epoch": 3.452328934719916,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00038371555930334194,
      "loss": 5.5084,
      "mean_token_accuracy": 0.1872662916779518,
      "num_tokens": 80050600.0,
      "step": 39505
    },
    {
      "entropy": 5.930003499984741,
      "epoch": 3.4527658830726207,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003836882760684907,
      "loss": 5.5089,
      "mean_token_accuracy": 0.18630160093307496,
      "num_tokens": 80060826.0,
      "step": 39510
    },
    {
      "entropy": 6.027190351486206,
      "epoch": 3.4532028314253256,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003836609907489414,
      "loss": 5.5572,
      "mean_token_accuracy": 0.18571127206087112,
      "num_tokens": 80070213.0,
      "step": 39515
    },
    {
      "entropy": 5.930661964416504,
      "epoch": 3.45363977977803,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003836337033452173,
      "loss": 5.4521,
      "mean_token_accuracy": 0.19041337966918945,
      "num_tokens": 80080197.0,
      "step": 39520
    },
    {
      "entropy": 5.987533855438232,
      "epoch": 3.454076728130735,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003836064138578419,
      "loss": 5.5507,
      "mean_token_accuracy": 0.18110116571187973,
      "num_tokens": 80089504.0,
      "step": 39525
    },
    {
      "entropy": 5.953745889663696,
      "epoch": 3.4545136764834394,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003835791222873386,
      "loss": 5.4889,
      "mean_token_accuracy": 0.19154423028230666,
      "num_tokens": 80099621.0,
      "step": 39530
    },
    {
      "entropy": 5.889128541946411,
      "epoch": 3.4549506248361443,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038355182863423077,
      "loss": 5.458,
      "mean_token_accuracy": 0.1898203507065773,
      "num_tokens": 80109880.0,
      "step": 39535
    },
    {
      "entropy": 5.896798849105835,
      "epoch": 3.4553875731888493,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00038352453289904206,
      "loss": 5.4723,
      "mean_token_accuracy": 0.19423249512910842,
      "num_tokens": 80119236.0,
      "step": 39540
    },
    {
      "entropy": 5.931785345077515,
      "epoch": 3.4558245215415537,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038349723508229587,
      "loss": 5.472,
      "mean_token_accuracy": 0.188851797580719,
      "num_tokens": 80129598.0,
      "step": 39545
    },
    {
      "entropy": 6.021877861022949,
      "epoch": 3.4562614698942586,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038346993518451587,
      "loss": 5.4923,
      "mean_token_accuracy": 0.17710081338882447,
      "num_tokens": 80140050.0,
      "step": 39550
    },
    {
      "entropy": 5.930074405670166,
      "epoch": 3.456698418246963,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038344263320622574,
      "loss": 5.4666,
      "mean_token_accuracy": 0.19554754495620727,
      "num_tokens": 80150101.0,
      "step": 39555
    },
    {
      "entropy": 5.960126495361328,
      "epoch": 3.457135366599668,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038341532914794904,
      "loss": 5.5,
      "mean_token_accuracy": 0.17717424184083938,
      "num_tokens": 80159416.0,
      "step": 39560
    },
    {
      "entropy": 5.8606713771820065,
      "epoch": 3.4575723149523725,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003833880230102095,
      "loss": 5.4791,
      "mean_token_accuracy": 0.1890251576900482,
      "num_tokens": 80168979.0,
      "step": 39565
    },
    {
      "entropy": 5.953228378295899,
      "epoch": 3.4580092633050774,
      "grad_norm": 1.453125,
      "learning_rate": 0.00038336071479353086,
      "loss": 5.4814,
      "mean_token_accuracy": 0.1910671904683113,
      "num_tokens": 80179328.0,
      "step": 39570
    },
    {
      "entropy": 5.979998207092285,
      "epoch": 3.458446211657782,
      "grad_norm": 1.546875,
      "learning_rate": 0.00038333340449843694,
      "loss": 5.4836,
      "mean_token_accuracy": 0.19075314849615096,
      "num_tokens": 80189151.0,
      "step": 39575
    },
    {
      "entropy": 5.959800481796265,
      "epoch": 3.458883160010487,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003833060921254517,
      "loss": 5.5163,
      "mean_token_accuracy": 0.1913122445344925,
      "num_tokens": 80199622.0,
      "step": 39580
    },
    {
      "entropy": 6.007354974746704,
      "epoch": 3.4593201083631913,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038327877767509866,
      "loss": 5.5052,
      "mean_token_accuracy": 0.18507008105516434,
      "num_tokens": 80209335.0,
      "step": 39585
    },
    {
      "entropy": 5.940871620178223,
      "epoch": 3.459757056715896,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000383251461147902,
      "loss": 5.5154,
      "mean_token_accuracy": 0.18745961487293245,
      "num_tokens": 80220336.0,
      "step": 39590
    },
    {
      "entropy": 5.9677550315856935,
      "epoch": 3.460194005068601,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00038322414254438557,
      "loss": 5.5288,
      "mean_token_accuracy": 0.17964121103286743,
      "num_tokens": 80231459.0,
      "step": 39595
    },
    {
      "entropy": 5.971529483795166,
      "epoch": 3.4606309534213056,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003831968218650735,
      "loss": 5.5333,
      "mean_token_accuracy": 0.18677197992801667,
      "num_tokens": 80240997.0,
      "step": 39600
    },
    {
      "entropy": 5.944520378112793,
      "epoch": 3.4610679017740105,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038316949911048953,
      "loss": 5.5093,
      "mean_token_accuracy": 0.18764936476945876,
      "num_tokens": 80251663.0,
      "step": 39605
    },
    {
      "entropy": 5.963939046859741,
      "epoch": 3.461504850126715,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00038314217428115797,
      "loss": 5.5416,
      "mean_token_accuracy": 0.18328649997711183,
      "num_tokens": 80261059.0,
      "step": 39610
    },
    {
      "entropy": 5.889348077774048,
      "epoch": 3.46194179847942,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003831148473776028,
      "loss": 5.4355,
      "mean_token_accuracy": 0.19725095629692077,
      "num_tokens": 80271781.0,
      "step": 39615
    },
    {
      "entropy": 5.889503335952758,
      "epoch": 3.4623787468321243,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00038308751840034826,
      "loss": 5.5311,
      "mean_token_accuracy": 0.1860017642378807,
      "num_tokens": 80281996.0,
      "step": 39620
    },
    {
      "entropy": 5.954548645019531,
      "epoch": 3.4628156951848292,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003830601873499185,
      "loss": 5.4841,
      "mean_token_accuracy": 0.19376535564661027,
      "num_tokens": 80293181.0,
      "step": 39625
    },
    {
      "entropy": 6.000442600250244,
      "epoch": 3.4632526435375337,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003830328542268377,
      "loss": 5.5101,
      "mean_token_accuracy": 0.18664814829826354,
      "num_tokens": 80303697.0,
      "step": 39630
    },
    {
      "entropy": 5.908944225311279,
      "epoch": 3.4636895918902386,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038300551903163013,
      "loss": 5.4175,
      "mean_token_accuracy": 0.19292390197515488,
      "num_tokens": 80313104.0,
      "step": 39635
    },
    {
      "entropy": 5.8731605052948,
      "epoch": 3.464126540242943,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038297818176482006,
      "loss": 5.4226,
      "mean_token_accuracy": 0.20196904093027115,
      "num_tokens": 80323168.0,
      "step": 39640
    },
    {
      "entropy": 5.913434028625488,
      "epoch": 3.464563488595648,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000382950842426932,
      "loss": 5.5121,
      "mean_token_accuracy": 0.18499234467744827,
      "num_tokens": 80333064.0,
      "step": 39645
    },
    {
      "entropy": 5.9113822937011715,
      "epoch": 3.465000436948353,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038292350101849017,
      "loss": 5.4093,
      "mean_token_accuracy": 0.19471413791179656,
      "num_tokens": 80342827.0,
      "step": 39650
    },
    {
      "entropy": 5.980172824859619,
      "epoch": 3.4654373853010574,
      "grad_norm": 1.0,
      "learning_rate": 0.000382896157540019,
      "loss": 5.5252,
      "mean_token_accuracy": 0.18582326918840408,
      "num_tokens": 80352732.0,
      "step": 39655
    },
    {
      "entropy": 5.991584062576294,
      "epoch": 3.4658743336537623,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000382868811992043,
      "loss": 5.5445,
      "mean_token_accuracy": 0.1846605509519577,
      "num_tokens": 80362427.0,
      "step": 39660
    },
    {
      "entropy": 6.003034067153931,
      "epoch": 3.4663112820064668,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003828414643750867,
      "loss": 5.478,
      "mean_token_accuracy": 0.1924673706293106,
      "num_tokens": 80372719.0,
      "step": 39665
    },
    {
      "entropy": 5.947220373153686,
      "epoch": 3.4667482303591717,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038281411468967457,
      "loss": 5.4581,
      "mean_token_accuracy": 0.1853904038667679,
      "num_tokens": 80381753.0,
      "step": 39670
    },
    {
      "entropy": 5.870768594741821,
      "epoch": 3.467185178711876,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003827867629363312,
      "loss": 5.5072,
      "mean_token_accuracy": 0.19165723025798798,
      "num_tokens": 80393401.0,
      "step": 39675
    },
    {
      "entropy": 5.997729635238647,
      "epoch": 3.467622127064581,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038275940911558135,
      "loss": 5.538,
      "mean_token_accuracy": 0.18565505594015122,
      "num_tokens": 80403194.0,
      "step": 39680
    },
    {
      "entropy": 5.964574909210205,
      "epoch": 3.4680590754172855,
      "grad_norm": 1.0,
      "learning_rate": 0.0003827320532279495,
      "loss": 5.4664,
      "mean_token_accuracy": 0.18503087311983107,
      "num_tokens": 80413283.0,
      "step": 39685
    },
    {
      "entropy": 5.973863697052002,
      "epoch": 3.4684960237699904,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038270469527396035,
      "loss": 5.5193,
      "mean_token_accuracy": 0.18426886200904846,
      "num_tokens": 80423976.0,
      "step": 39690
    },
    {
      "entropy": 5.889187574386597,
      "epoch": 3.468932972122695,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038267733525413884,
      "loss": 5.4917,
      "mean_token_accuracy": 0.19000918716192244,
      "num_tokens": 80434017.0,
      "step": 39695
    },
    {
      "entropy": 5.889046716690063,
      "epoch": 3.4693699204754,
      "grad_norm": 1.25,
      "learning_rate": 0.0003826499731690095,
      "loss": 5.4727,
      "mean_token_accuracy": 0.18914036452770233,
      "num_tokens": 80444289.0,
      "step": 39700
    },
    {
      "entropy": 6.00470118522644,
      "epoch": 3.4698068688281047,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003826226090190974,
      "loss": 5.5266,
      "mean_token_accuracy": 0.18401649296283723,
      "num_tokens": 80453420.0,
      "step": 39705
    },
    {
      "entropy": 6.01183819770813,
      "epoch": 3.470243817180809,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038259524280492716,
      "loss": 5.4987,
      "mean_token_accuracy": 0.18922465443611144,
      "num_tokens": 80462977.0,
      "step": 39710
    },
    {
      "entropy": 5.86129059791565,
      "epoch": 3.470680765533514,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038256787452702377,
      "loss": 5.3814,
      "mean_token_accuracy": 0.19053756594657897,
      "num_tokens": 80471806.0,
      "step": 39715
    },
    {
      "entropy": 5.886973810195923,
      "epoch": 3.4711177138862186,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038254050418591227,
      "loss": 5.4832,
      "mean_token_accuracy": 0.18491990864276886,
      "num_tokens": 80481510.0,
      "step": 39720
    },
    {
      "entropy": 5.997308206558228,
      "epoch": 3.4715546622389235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003825131317821175,
      "loss": 5.4926,
      "mean_token_accuracy": 0.18454622179269792,
      "num_tokens": 80491670.0,
      "step": 39725
    },
    {
      "entropy": 5.996175956726074,
      "epoch": 3.471991610591628,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00038248575731616454,
      "loss": 5.5933,
      "mean_token_accuracy": 0.18248986303806305,
      "num_tokens": 80501518.0,
      "step": 39730
    },
    {
      "entropy": 5.923818969726563,
      "epoch": 3.472428558944333,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00038245838078857837,
      "loss": 5.4828,
      "mean_token_accuracy": 0.1897611752152443,
      "num_tokens": 80511943.0,
      "step": 39735
    },
    {
      "entropy": 5.9161123752594,
      "epoch": 3.4728655072970374,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003824310021998841,
      "loss": 5.4488,
      "mean_token_accuracy": 0.1873147264122963,
      "num_tokens": 80523114.0,
      "step": 39740
    },
    {
      "entropy": 6.023334646224976,
      "epoch": 3.4733024556497423,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003824036215506071,
      "loss": 5.5747,
      "mean_token_accuracy": 0.17720255255699158,
      "num_tokens": 80534802.0,
      "step": 39745
    },
    {
      "entropy": 6.027471971511841,
      "epoch": 3.4737394040024467,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00038237623884127225,
      "loss": 5.485,
      "mean_token_accuracy": 0.1865319788455963,
      "num_tokens": 80544367.0,
      "step": 39750
    },
    {
      "entropy": 5.8966583728790285,
      "epoch": 3.4741763523551517,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038234885407240494,
      "loss": 5.5611,
      "mean_token_accuracy": 0.186593696475029,
      "num_tokens": 80554841.0,
      "step": 39755
    },
    {
      "entropy": 6.0054792881011965,
      "epoch": 3.4746133007078566,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003823214672445302,
      "loss": 5.522,
      "mean_token_accuracy": 0.1811122104525566,
      "num_tokens": 80565076.0,
      "step": 39760
    },
    {
      "entropy": 5.9919229507446286,
      "epoch": 3.475050249060561,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038229407835817366,
      "loss": 5.5643,
      "mean_token_accuracy": 0.1828638345003128,
      "num_tokens": 80575410.0,
      "step": 39765
    },
    {
      "entropy": 5.910904598236084,
      "epoch": 3.4754871974132655,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038226668741386046,
      "loss": 5.5011,
      "mean_token_accuracy": 0.18667433112859727,
      "num_tokens": 80585400.0,
      "step": 39770
    },
    {
      "entropy": 6.021160507202149,
      "epoch": 3.4759241457659704,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003822392944121159,
      "loss": 5.4871,
      "mean_token_accuracy": 0.1866036534309387,
      "num_tokens": 80595429.0,
      "step": 39775
    },
    {
      "entropy": 5.943646764755249,
      "epoch": 3.4763610941186753,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003822118993534655,
      "loss": 5.4757,
      "mean_token_accuracy": 0.19040271192789077,
      "num_tokens": 80605060.0,
      "step": 39780
    },
    {
      "entropy": 5.896329069137574,
      "epoch": 3.47679804247138,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003821845022384348,
      "loss": 5.4019,
      "mean_token_accuracy": 0.19198458939790725,
      "num_tokens": 80614927.0,
      "step": 39785
    },
    {
      "entropy": 5.874010229110718,
      "epoch": 3.4772349908240847,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038215710306754906,
      "loss": 5.4794,
      "mean_token_accuracy": 0.19215115308761596,
      "num_tokens": 80624748.0,
      "step": 39790
    },
    {
      "entropy": 6.000562524795532,
      "epoch": 3.477671939176789,
      "grad_norm": 0.96875,
      "learning_rate": 0.00038212970184133395,
      "loss": 5.5821,
      "mean_token_accuracy": 0.1775862380862236,
      "num_tokens": 80635221.0,
      "step": 39795
    },
    {
      "entropy": 5.971331071853638,
      "epoch": 3.478108887529494,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038210229856031503,
      "loss": 5.5445,
      "mean_token_accuracy": 0.18697671741247177,
      "num_tokens": 80645268.0,
      "step": 39800
    },
    {
      "entropy": 6.099808645248413,
      "epoch": 3.4785458358821986,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003820748932250179,
      "loss": 5.5567,
      "mean_token_accuracy": 0.17615439295768737,
      "num_tokens": 80655209.0,
      "step": 39805
    },
    {
      "entropy": 5.951766014099121,
      "epoch": 3.4789827842349035,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00038204748583596835,
      "loss": 5.5455,
      "mean_token_accuracy": 0.18733263909816741,
      "num_tokens": 80665652.0,
      "step": 39810
    },
    {
      "entropy": 5.898327159881592,
      "epoch": 3.4794197325876084,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003820200763936917,
      "loss": 5.5077,
      "mean_token_accuracy": 0.19022778123617173,
      "num_tokens": 80675430.0,
      "step": 39815
    },
    {
      "entropy": 5.981950473785401,
      "epoch": 3.479856680940313,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003819926648987141,
      "loss": 5.4977,
      "mean_token_accuracy": 0.1847696214914322,
      "num_tokens": 80685997.0,
      "step": 39820
    },
    {
      "entropy": 5.988628149032593,
      "epoch": 3.4802936292930173,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038196525135156094,
      "loss": 5.5283,
      "mean_token_accuracy": 0.1807534620165825,
      "num_tokens": 80697470.0,
      "step": 39825
    },
    {
      "entropy": 5.969229888916016,
      "epoch": 3.4807305776457222,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003819378357527583,
      "loss": 5.4818,
      "mean_token_accuracy": 0.18614667057991027,
      "num_tokens": 80708810.0,
      "step": 39830
    },
    {
      "entropy": 5.895774412155151,
      "epoch": 3.481167525998427,
      "grad_norm": 1.0625,
      "learning_rate": 0.000381910418102832,
      "loss": 5.3941,
      "mean_token_accuracy": 0.20226566940546037,
      "num_tokens": 80718036.0,
      "step": 39835
    },
    {
      "entropy": 5.93081316947937,
      "epoch": 3.4816044743511316,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003818829984023078,
      "loss": 5.5307,
      "mean_token_accuracy": 0.18763178884983062,
      "num_tokens": 80728247.0,
      "step": 39840
    },
    {
      "entropy": 5.934335374832154,
      "epoch": 3.4820414227038365,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038185557665171174,
      "loss": 5.5032,
      "mean_token_accuracy": 0.1888561338186264,
      "num_tokens": 80738873.0,
      "step": 39845
    },
    {
      "entropy": 5.993309354782104,
      "epoch": 3.482478371056541,
      "grad_norm": 0.96875,
      "learning_rate": 0.00038182815285156964,
      "loss": 5.5266,
      "mean_token_accuracy": 0.18190221339464188,
      "num_tokens": 80748734.0,
      "step": 39850
    },
    {
      "entropy": 5.948193073272705,
      "epoch": 3.482915319409246,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003818007270024076,
      "loss": 5.4789,
      "mean_token_accuracy": 0.193984317779541,
      "num_tokens": 80759256.0,
      "step": 39855
    },
    {
      "entropy": 5.933566045761109,
      "epoch": 3.4833522677619504,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003817732991047518,
      "loss": 5.4589,
      "mean_token_accuracy": 0.19573962092399597,
      "num_tokens": 80769670.0,
      "step": 39860
    },
    {
      "entropy": 5.886237764358521,
      "epoch": 3.4837892161146553,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038174586915912804,
      "loss": 5.4181,
      "mean_token_accuracy": 0.19436275064945222,
      "num_tokens": 80778937.0,
      "step": 39865
    },
    {
      "entropy": 5.913418483734131,
      "epoch": 3.4842261644673598,
      "grad_norm": 1.0,
      "learning_rate": 0.00038171843716606254,
      "loss": 5.4722,
      "mean_token_accuracy": 0.18823930472135544,
      "num_tokens": 80789329.0,
      "step": 39870
    },
    {
      "entropy": 5.910842084884644,
      "epoch": 3.4846631128200647,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00038169100312608156,
      "loss": 5.4778,
      "mean_token_accuracy": 0.18749107271432877,
      "num_tokens": 80798943.0,
      "step": 39875
    },
    {
      "entropy": 6.030255222320557,
      "epoch": 3.485100061172769,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003816635670397112,
      "loss": 5.553,
      "mean_token_accuracy": 0.18140195906162263,
      "num_tokens": 80808915.0,
      "step": 39880
    },
    {
      "entropy": 5.941267824172973,
      "epoch": 3.485537009525474,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00038163612890747777,
      "loss": 5.4886,
      "mean_token_accuracy": 0.19299552589654922,
      "num_tokens": 80820162.0,
      "step": 39885
    },
    {
      "entropy": 5.847454690933228,
      "epoch": 3.485973957878179,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003816086887299075,
      "loss": 5.4087,
      "mean_token_accuracy": 0.2058575764298439,
      "num_tokens": 80830179.0,
      "step": 39890
    },
    {
      "entropy": 5.979230642318726,
      "epoch": 3.4864109062308835,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003815812465075267,
      "loss": 5.4815,
      "mean_token_accuracy": 0.19046678394079208,
      "num_tokens": 80839924.0,
      "step": 39895
    },
    {
      "entropy": 5.955972719192505,
      "epoch": 3.4868478545835884,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038155380224086177,
      "loss": 5.4838,
      "mean_token_accuracy": 0.18916407823562623,
      "num_tokens": 80849594.0,
      "step": 39900
    },
    {
      "entropy": 5.894352865219116,
      "epoch": 3.487284802936293,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003815263559304391,
      "loss": 5.4572,
      "mean_token_accuracy": 0.19026477187871932,
      "num_tokens": 80859706.0,
      "step": 39905
    },
    {
      "entropy": 5.978844738006591,
      "epoch": 3.4877217512889978,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038149890757678507,
      "loss": 5.5486,
      "mean_token_accuracy": 0.18155749440193175,
      "num_tokens": 80869121.0,
      "step": 39910
    },
    {
      "entropy": 5.944537258148193,
      "epoch": 3.488158699641702,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038147145718042626,
      "loss": 5.5382,
      "mean_token_accuracy": 0.18572770655155182,
      "num_tokens": 80879347.0,
      "step": 39915
    },
    {
      "entropy": 5.929362058639526,
      "epoch": 3.488595647994407,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000381444004741889,
      "loss": 5.468,
      "mean_token_accuracy": 0.19095506072044371,
      "num_tokens": 80889819.0,
      "step": 39920
    },
    {
      "entropy": 6.020323991775513,
      "epoch": 3.4890325963471116,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003814165502617,
      "loss": 5.5318,
      "mean_token_accuracy": 0.18735142052173615,
      "num_tokens": 80900576.0,
      "step": 39925
    },
    {
      "entropy": 5.992763996124268,
      "epoch": 3.4894695446998165,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003813890937403858,
      "loss": 5.4691,
      "mean_token_accuracy": 0.19424747079610824,
      "num_tokens": 80910966.0,
      "step": 39930
    },
    {
      "entropy": 5.853727340698242,
      "epoch": 3.489906493052521,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038136163517847306,
      "loss": 5.4681,
      "mean_token_accuracy": 0.18829023092985153,
      "num_tokens": 80920533.0,
      "step": 39935
    },
    {
      "entropy": 5.958118581771851,
      "epoch": 3.490343441405226,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038133417457648835,
      "loss": 5.5127,
      "mean_token_accuracy": 0.18434954881668092,
      "num_tokens": 80930989.0,
      "step": 39940
    },
    {
      "entropy": 6.04156494140625,
      "epoch": 3.490780389757931,
      "grad_norm": 1.0,
      "learning_rate": 0.00038130671193495845,
      "loss": 5.5265,
      "mean_token_accuracy": 0.18358989506959916,
      "num_tokens": 80942283.0,
      "step": 39945
    },
    {
      "entropy": 5.976937580108642,
      "epoch": 3.4912173381106353,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038127924725441024,
      "loss": 5.4675,
      "mean_token_accuracy": 0.19112753719091416,
      "num_tokens": 80953258.0,
      "step": 39950
    },
    {
      "entropy": 5.868747758865356,
      "epoch": 3.49165428646334,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003812517805353703,
      "loss": 5.4328,
      "mean_token_accuracy": 0.1959875449538231,
      "num_tokens": 80963748.0,
      "step": 39955
    },
    {
      "entropy": 5.901561689376831,
      "epoch": 3.4920912348160447,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0003812243117783656,
      "loss": 5.4951,
      "mean_token_accuracy": 0.18237485885620117,
      "num_tokens": 80973677.0,
      "step": 39960
    },
    {
      "entropy": 5.982271480560303,
      "epoch": 3.4925281831687496,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038119684098392284,
      "loss": 5.515,
      "mean_token_accuracy": 0.1852603182196617,
      "num_tokens": 80984021.0,
      "step": 39965
    },
    {
      "entropy": 5.902640151977539,
      "epoch": 3.492965131521454,
      "grad_norm": 1.046875,
      "learning_rate": 0.000381169368152569,
      "loss": 5.3387,
      "mean_token_accuracy": 0.19653447568416596,
      "num_tokens": 80993498.0,
      "step": 39970
    },
    {
      "entropy": 5.948106956481934,
      "epoch": 3.493402079874159,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038114189328483124,
      "loss": 5.5325,
      "mean_token_accuracy": 0.1844900816679001,
      "num_tokens": 81002693.0,
      "step": 39975
    },
    {
      "entropy": 5.972675371170044,
      "epoch": 3.4938390282268634,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003811144163812362,
      "loss": 5.5514,
      "mean_token_accuracy": 0.18631932884454727,
      "num_tokens": 81012715.0,
      "step": 39980
    },
    {
      "entropy": 5.976652145385742,
      "epoch": 3.4942759765795683,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038108693744231105,
      "loss": 5.4776,
      "mean_token_accuracy": 0.19059130549430847,
      "num_tokens": 81023779.0,
      "step": 39985
    },
    {
      "entropy": 5.882216691970825,
      "epoch": 3.494712924932273,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003810594564685829,
      "loss": 5.4447,
      "mean_token_accuracy": 0.20329279452562332,
      "num_tokens": 81035103.0,
      "step": 39990
    },
    {
      "entropy": 5.941090536117554,
      "epoch": 3.4951498732849777,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003810319734605787,
      "loss": 5.5155,
      "mean_token_accuracy": 0.1885618433356285,
      "num_tokens": 81045011.0,
      "step": 39995
    },
    {
      "entropy": 5.950661087036133,
      "epoch": 3.4955868216376826,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003810044884188258,
      "loss": 5.4959,
      "mean_token_accuracy": 0.19259555041790008,
      "num_tokens": 81055841.0,
      "step": 40000
    },
    {
      "entropy": 5.943976640701294,
      "epoch": 3.496023769990387,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003809770013438512,
      "loss": 5.5153,
      "mean_token_accuracy": 0.18445902913808823,
      "num_tokens": 81067270.0,
      "step": 40005
    },
    {
      "entropy": 5.989645862579346,
      "epoch": 3.496460718343092,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038094951223618224,
      "loss": 5.4919,
      "mean_token_accuracy": 0.19012177288532256,
      "num_tokens": 81076963.0,
      "step": 40010
    },
    {
      "entropy": 5.899583625793457,
      "epoch": 3.4968976666957965,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038092202109634603,
      "loss": 5.4532,
      "mean_token_accuracy": 0.19133660495281218,
      "num_tokens": 81086146.0,
      "step": 40015
    },
    {
      "entropy": 5.943194770812989,
      "epoch": 3.4973346150485014,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038089452792487005,
      "loss": 5.4787,
      "mean_token_accuracy": 0.19070324897766114,
      "num_tokens": 81095924.0,
      "step": 40020
    },
    {
      "entropy": 5.964000606536866,
      "epoch": 3.497771563401206,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003808670327222815,
      "loss": 5.491,
      "mean_token_accuracy": 0.18184123635292054,
      "num_tokens": 81106396.0,
      "step": 40025
    },
    {
      "entropy": 5.934715509414673,
      "epoch": 3.498208511753911,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038083953548910776,
      "loss": 5.4483,
      "mean_token_accuracy": 0.18978586941957473,
      "num_tokens": 81117098.0,
      "step": 40030
    },
    {
      "entropy": 5.883282232284546,
      "epoch": 3.4986454601066153,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00038081203622587637,
      "loss": 5.4509,
      "mean_token_accuracy": 0.18765627294778825,
      "num_tokens": 81127335.0,
      "step": 40035
    },
    {
      "entropy": 5.944164085388183,
      "epoch": 3.49908240845932,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038078453493311463,
      "loss": 5.4707,
      "mean_token_accuracy": 0.1845555156469345,
      "num_tokens": 81136608.0,
      "step": 40040
    },
    {
      "entropy": 6.0007248878479,
      "epoch": 3.4995193568120246,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038075703161135,
      "loss": 5.5709,
      "mean_token_accuracy": 0.18138879388570786,
      "num_tokens": 81146690.0,
      "step": 40045
    },
    {
      "entropy": 5.9351154327392575,
      "epoch": 3.4999563051647296,
      "grad_norm": 1.125,
      "learning_rate": 0.00038072952626111016,
      "loss": 5.4166,
      "mean_token_accuracy": 0.19303133934736252,
      "num_tokens": 81155914.0,
      "step": 40050
    },
    {
      "entropy": 5.931235218048096,
      "epoch": 3.5003932535174345,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003807020188829226,
      "loss": 5.439,
      "mean_token_accuracy": 0.1971483588218689,
      "num_tokens": 81166537.0,
      "step": 40055
    },
    {
      "entropy": 5.8882499694824215,
      "epoch": 3.500830201870139,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003806745094773149,
      "loss": 5.4522,
      "mean_token_accuracy": 0.19397936314344405,
      "num_tokens": 81176328.0,
      "step": 40060
    },
    {
      "entropy": 5.89340991973877,
      "epoch": 3.5012671502228434,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038064699804481476,
      "loss": 5.5047,
      "mean_token_accuracy": 0.18643472194671631,
      "num_tokens": 81186345.0,
      "step": 40065
    },
    {
      "entropy": 5.949583959579468,
      "epoch": 3.5017040985755483,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003806194845859499,
      "loss": 5.4949,
      "mean_token_accuracy": 0.18693253993988038,
      "num_tokens": 81196683.0,
      "step": 40070
    },
    {
      "entropy": 5.953500366210937,
      "epoch": 3.5021410469282532,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003805919691012479,
      "loss": 5.4671,
      "mean_token_accuracy": 0.18892274498939515,
      "num_tokens": 81206381.0,
      "step": 40075
    },
    {
      "entropy": 5.964162540435791,
      "epoch": 3.5025779952809577,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038056445159123664,
      "loss": 5.5629,
      "mean_token_accuracy": 0.18394907861948012,
      "num_tokens": 81217056.0,
      "step": 40080
    },
    {
      "entropy": 5.930300521850586,
      "epoch": 3.5030149436336626,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003805369320564438,
      "loss": 5.4727,
      "mean_token_accuracy": 0.19204898178577423,
      "num_tokens": 81227803.0,
      "step": 40085
    },
    {
      "entropy": 5.96289644241333,
      "epoch": 3.503451891986367,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003805094104973974,
      "loss": 5.455,
      "mean_token_accuracy": 0.19251654297113419,
      "num_tokens": 81237680.0,
      "step": 40090
    },
    {
      "entropy": 5.921104907989502,
      "epoch": 3.503888840339072,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003804818869146252,
      "loss": 5.4991,
      "mean_token_accuracy": 0.18901841044425965,
      "num_tokens": 81247624.0,
      "step": 40095
    },
    {
      "entropy": 5.98858904838562,
      "epoch": 3.5043257886917765,
      "grad_norm": 1.4375,
      "learning_rate": 0.00038045436130865515,
      "loss": 5.4733,
      "mean_token_accuracy": 0.19484759122133255,
      "num_tokens": 81257932.0,
      "step": 40100
    },
    {
      "entropy": 5.991704082489013,
      "epoch": 3.5047627370444814,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003804268336800151,
      "loss": 5.514,
      "mean_token_accuracy": 0.18803609758615494,
      "num_tokens": 81268595.0,
      "step": 40105
    },
    {
      "entropy": 5.910090494155884,
      "epoch": 3.5051996853971863,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038039930402923323,
      "loss": 5.4439,
      "mean_token_accuracy": 0.19512676894664766,
      "num_tokens": 81279252.0,
      "step": 40110
    },
    {
      "entropy": 5.89531307220459,
      "epoch": 3.5056366337498908,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038037177235683746,
      "loss": 5.4369,
      "mean_token_accuracy": 0.1957525759935379,
      "num_tokens": 81289897.0,
      "step": 40115
    },
    {
      "entropy": 5.92204213142395,
      "epoch": 3.5060735821025952,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038034423866335586,
      "loss": 5.4812,
      "mean_token_accuracy": 0.1897186741232872,
      "num_tokens": 81300297.0,
      "step": 40120
    },
    {
      "entropy": 5.961403274536133,
      "epoch": 3.5065105304553,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003803167029493166,
      "loss": 5.4982,
      "mean_token_accuracy": 0.18114887923002243,
      "num_tokens": 81309628.0,
      "step": 40125
    },
    {
      "entropy": 5.8885503768920895,
      "epoch": 3.506947478808005,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00038028916521524775,
      "loss": 5.4824,
      "mean_token_accuracy": 0.19177746325731276,
      "num_tokens": 81320511.0,
      "step": 40130
    },
    {
      "entropy": 5.945067119598389,
      "epoch": 3.5073844271607095,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003802616254616776,
      "loss": 5.4854,
      "mean_token_accuracy": 0.1883799612522125,
      "num_tokens": 81330982.0,
      "step": 40135
    },
    {
      "entropy": 5.99360842704773,
      "epoch": 3.5078213755134144,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003802340836891343,
      "loss": 5.4724,
      "mean_token_accuracy": 0.18857266008853912,
      "num_tokens": 81339955.0,
      "step": 40140
    },
    {
      "entropy": 5.883594989776611,
      "epoch": 3.508258323866119,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038020653989814606,
      "loss": 5.4919,
      "mean_token_accuracy": 0.19125393480062486,
      "num_tokens": 81349151.0,
      "step": 40145
    },
    {
      "entropy": 5.9442140579223635,
      "epoch": 3.508695272218824,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003801789940892413,
      "loss": 5.4069,
      "mean_token_accuracy": 0.1945842146873474,
      "num_tokens": 81359563.0,
      "step": 40150
    },
    {
      "entropy": 5.980524015426636,
      "epoch": 3.5091322205715283,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003801514462629483,
      "loss": 5.5287,
      "mean_token_accuracy": 0.18654235750436782,
      "num_tokens": 81369669.0,
      "step": 40155
    },
    {
      "entropy": 5.952213096618652,
      "epoch": 3.509569168924233,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038012389641979555,
      "loss": 5.5613,
      "mean_token_accuracy": 0.18006304949522017,
      "num_tokens": 81380322.0,
      "step": 40160
    },
    {
      "entropy": 5.985963249206543,
      "epoch": 3.510006117276938,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003800963445603114,
      "loss": 5.5019,
      "mean_token_accuracy": 0.18881215751171113,
      "num_tokens": 81389871.0,
      "step": 40165
    },
    {
      "entropy": 5.846839570999146,
      "epoch": 3.5104430656296426,
      "grad_norm": 1.0,
      "learning_rate": 0.0003800687906850242,
      "loss": 5.35,
      "mean_token_accuracy": 0.2025637999176979,
      "num_tokens": 81399684.0,
      "step": 40170
    },
    {
      "entropy": 5.935965919494629,
      "epoch": 3.510880013982347,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003800412347944626,
      "loss": 5.4709,
      "mean_token_accuracy": 0.19143253564834595,
      "num_tokens": 81409638.0,
      "step": 40175
    },
    {
      "entropy": 6.007100582122803,
      "epoch": 3.511316962335052,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003800136768891549,
      "loss": 5.4911,
      "mean_token_accuracy": 0.18640716075897218,
      "num_tokens": 81419963.0,
      "step": 40180
    },
    {
      "entropy": 5.931604385375977,
      "epoch": 3.511753910687757,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037998611696963,
      "loss": 5.4435,
      "mean_token_accuracy": 0.1886257454752922,
      "num_tokens": 81429747.0,
      "step": 40185
    },
    {
      "entropy": 5.900136947631836,
      "epoch": 3.5121908590404614,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003799585550364164,
      "loss": 5.42,
      "mean_token_accuracy": 0.18748452216386796,
      "num_tokens": 81439012.0,
      "step": 40190
    },
    {
      "entropy": 5.945207023620606,
      "epoch": 3.5126278073931663,
      "grad_norm": 1.25,
      "learning_rate": 0.0003799309910900427,
      "loss": 5.5305,
      "mean_token_accuracy": 0.18736913055181503,
      "num_tokens": 81449713.0,
      "step": 40195
    },
    {
      "entropy": 5.8812848091125485,
      "epoch": 3.5130647557458707,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003799034251310376,
      "loss": 5.4121,
      "mean_token_accuracy": 0.19404484927654267,
      "num_tokens": 81459326.0,
      "step": 40200
    },
    {
      "entropy": 5.963953733444214,
      "epoch": 3.5135017040985757,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00037987585715992974,
      "loss": 5.4517,
      "mean_token_accuracy": 0.19227051883935928,
      "num_tokens": 81469419.0,
      "step": 40205
    },
    {
      "entropy": 5.898402833938599,
      "epoch": 3.51393865245128,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037984828717724814,
      "loss": 5.4658,
      "mean_token_accuracy": 0.19659966677427293,
      "num_tokens": 81478494.0,
      "step": 40210
    },
    {
      "entropy": 5.879290914535522,
      "epoch": 3.514375600803985,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037982071518352135,
      "loss": 5.4527,
      "mean_token_accuracy": 0.19141457825899125,
      "num_tokens": 81488080.0,
      "step": 40215
    },
    {
      "entropy": 5.966784858703614,
      "epoch": 3.51481254915669,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003797931411792783,
      "loss": 5.5079,
      "mean_token_accuracy": 0.19330233931541443,
      "num_tokens": 81498075.0,
      "step": 40220
    },
    {
      "entropy": 5.9826396942138675,
      "epoch": 3.5152494975093944,
      "grad_norm": 1.0,
      "learning_rate": 0.00037976556516504796,
      "loss": 5.4667,
      "mean_token_accuracy": 0.18891668915748597,
      "num_tokens": 81508407.0,
      "step": 40225
    },
    {
      "entropy": 5.966143321990967,
      "epoch": 3.515686445862099,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00037973798714135916,
      "loss": 5.4904,
      "mean_token_accuracy": 0.17935618311166762,
      "num_tokens": 81518009.0,
      "step": 40230
    },
    {
      "entropy": 5.896805477142334,
      "epoch": 3.516123394214804,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003797104071087409,
      "loss": 5.4804,
      "mean_token_accuracy": 0.19606601148843766,
      "num_tokens": 81527468.0,
      "step": 40235
    },
    {
      "entropy": 5.942311716079712,
      "epoch": 3.5165603425675087,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003796828250677221,
      "loss": 5.4826,
      "mean_token_accuracy": 0.19071678221225738,
      "num_tokens": 81538133.0,
      "step": 40240
    },
    {
      "entropy": 5.877294301986694,
      "epoch": 3.516997290920213,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037965524101883196,
      "loss": 5.411,
      "mean_token_accuracy": 0.19409576207399368,
      "num_tokens": 81548051.0,
      "step": 40245
    },
    {
      "entropy": 5.961598014831543,
      "epoch": 3.517434239272918,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003796276549625994,
      "loss": 5.533,
      "mean_token_accuracy": 0.1862824559211731,
      "num_tokens": 81557984.0,
      "step": 40250
    },
    {
      "entropy": 5.942479085922241,
      "epoch": 3.5178711876256226,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037960006689955356,
      "loss": 5.4985,
      "mean_token_accuracy": 0.19271458834409713,
      "num_tokens": 81568587.0,
      "step": 40255
    },
    {
      "entropy": 5.932956314086914,
      "epoch": 3.5183081359783275,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037957247683022375,
      "loss": 5.4549,
      "mean_token_accuracy": 0.1876130536198616,
      "num_tokens": 81578085.0,
      "step": 40260
    },
    {
      "entropy": 5.990452480316162,
      "epoch": 3.518745084331032,
      "grad_norm": 1.0,
      "learning_rate": 0.0003795448847551389,
      "loss": 5.5694,
      "mean_token_accuracy": 0.17866235822439194,
      "num_tokens": 81588037.0,
      "step": 40265
    },
    {
      "entropy": 5.909860467910766,
      "epoch": 3.519182032683737,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037951729067482845,
      "loss": 5.4669,
      "mean_token_accuracy": 0.19376726895570756,
      "num_tokens": 81598364.0,
      "step": 40270
    },
    {
      "entropy": 5.944582796096801,
      "epoch": 3.5196189810364413,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037948969458982153,
      "loss": 5.439,
      "mean_token_accuracy": 0.1859223186969757,
      "num_tokens": 81609428.0,
      "step": 40275
    },
    {
      "entropy": 5.8735426902771,
      "epoch": 3.5200559293891462,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037946209650064754,
      "loss": 5.5051,
      "mean_token_accuracy": 0.17904915511608124,
      "num_tokens": 81619605.0,
      "step": 40280
    },
    {
      "entropy": 5.964547061920166,
      "epoch": 3.5204928777418507,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003794344964078359,
      "loss": 5.5222,
      "mean_token_accuracy": 0.17997415959835053,
      "num_tokens": 81628987.0,
      "step": 40285
    },
    {
      "entropy": 5.950854158401489,
      "epoch": 3.5209298260945556,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003794068943119159,
      "loss": 5.4267,
      "mean_token_accuracy": 0.19505785554647445,
      "num_tokens": 81637628.0,
      "step": 40290
    },
    {
      "entropy": 5.835343933105468,
      "epoch": 3.5213667744472605,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003793792902134168,
      "loss": 5.417,
      "mean_token_accuracy": 0.19572878926992415,
      "num_tokens": 81647470.0,
      "step": 40295
    },
    {
      "entropy": 5.923657321929932,
      "epoch": 3.521803722799965,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003793516841128683,
      "loss": 5.4575,
      "mean_token_accuracy": 0.18619214445352555,
      "num_tokens": 81657882.0,
      "step": 40300
    },
    {
      "entropy": 5.952400064468383,
      "epoch": 3.5222406711526695,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037932407601079987,
      "loss": 5.5159,
      "mean_token_accuracy": 0.1868539869785309,
      "num_tokens": 81668506.0,
      "step": 40305
    },
    {
      "entropy": 5.9416587352752686,
      "epoch": 3.5226776195053744,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037929646590774093,
      "loss": 5.4852,
      "mean_token_accuracy": 0.18741605579853057,
      "num_tokens": 81678590.0,
      "step": 40310
    },
    {
      "entropy": 5.911036682128906,
      "epoch": 3.5231145678580793,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003792688538042211,
      "loss": 5.4905,
      "mean_token_accuracy": 0.2022777244448662,
      "num_tokens": 81689625.0,
      "step": 40315
    },
    {
      "entropy": 5.923299980163574,
      "epoch": 3.5235515162107838,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003792412397007701,
      "loss": 5.5432,
      "mean_token_accuracy": 0.18853494971990586,
      "num_tokens": 81699251.0,
      "step": 40320
    },
    {
      "entropy": 5.9302112579345705,
      "epoch": 3.5239884645634887,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037921362359791743,
      "loss": 5.4274,
      "mean_token_accuracy": 0.1960911899805069,
      "num_tokens": 81708851.0,
      "step": 40325
    },
    {
      "entropy": 5.946701717376709,
      "epoch": 3.524425412916193,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003791860054961929,
      "loss": 5.4933,
      "mean_token_accuracy": 0.18160222768783568,
      "num_tokens": 81718399.0,
      "step": 40330
    },
    {
      "entropy": 5.975591611862183,
      "epoch": 3.524862361268898,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037915838539612624,
      "loss": 5.5438,
      "mean_token_accuracy": 0.1905321478843689,
      "num_tokens": 81728546.0,
      "step": 40335
    },
    {
      "entropy": 5.980654621124268,
      "epoch": 3.5252993096216025,
      "grad_norm": 1.078125,
      "learning_rate": 0.000379130763298247,
      "loss": 5.4735,
      "mean_token_accuracy": 0.18807657659053803,
      "num_tokens": 81737830.0,
      "step": 40340
    },
    {
      "entropy": 5.870275068283081,
      "epoch": 3.5257362579743075,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003791031392030853,
      "loss": 5.4131,
      "mean_token_accuracy": 0.19063063114881515,
      "num_tokens": 81747534.0,
      "step": 40345
    },
    {
      "entropy": 5.914513826370239,
      "epoch": 3.5261732063270124,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037907551311117083,
      "loss": 5.511,
      "mean_token_accuracy": 0.186151947081089,
      "num_tokens": 81758411.0,
      "step": 40350
    },
    {
      "entropy": 5.9332915306091305,
      "epoch": 3.526610154679717,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003790478850230335,
      "loss": 5.4481,
      "mean_token_accuracy": 0.1876710206270218,
      "num_tokens": 81768441.0,
      "step": 40355
    },
    {
      "entropy": 5.9574120998382565,
      "epoch": 3.5270471030324213,
      "grad_norm": 1.0,
      "learning_rate": 0.00037902025493920314,
      "loss": 5.5234,
      "mean_token_accuracy": 0.1871328517794609,
      "num_tokens": 81778899.0,
      "step": 40360
    },
    {
      "entropy": 5.957236814498901,
      "epoch": 3.527484051385126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037899262286020984,
      "loss": 5.4597,
      "mean_token_accuracy": 0.19337200224399567,
      "num_tokens": 81788643.0,
      "step": 40365
    },
    {
      "entropy": 5.9043220520019535,
      "epoch": 3.527920999737831,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037896498878658345,
      "loss": 5.4935,
      "mean_token_accuracy": 0.1878335252404213,
      "num_tokens": 81798781.0,
      "step": 40370
    },
    {
      "entropy": 5.94243016242981,
      "epoch": 3.5283579480905356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003789373527188542,
      "loss": 5.4411,
      "mean_token_accuracy": 0.19148629009723664,
      "num_tokens": 81808402.0,
      "step": 40375
    },
    {
      "entropy": 5.975637292861938,
      "epoch": 3.5287948964432405,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003789097146575521,
      "loss": 5.5157,
      "mean_token_accuracy": 0.18719488382339478,
      "num_tokens": 81819285.0,
      "step": 40380
    },
    {
      "entropy": 5.916110849380493,
      "epoch": 3.529231844795945,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003788820746032071,
      "loss": 5.4152,
      "mean_token_accuracy": 0.1923023521900177,
      "num_tokens": 81829357.0,
      "step": 40385
    },
    {
      "entropy": 5.920270109176636,
      "epoch": 3.52966879314865,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037885443255634955,
      "loss": 5.4185,
      "mean_token_accuracy": 0.19819316267967224,
      "num_tokens": 81838547.0,
      "step": 40390
    },
    {
      "entropy": 5.963284730911255,
      "epoch": 3.5301057415013544,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003788267885175095,
      "loss": 5.4963,
      "mean_token_accuracy": 0.18620895296335221,
      "num_tokens": 81849286.0,
      "step": 40395
    },
    {
      "entropy": 6.059962940216065,
      "epoch": 3.5305426898540593,
      "grad_norm": 1.0,
      "learning_rate": 0.00037879914248721737,
      "loss": 5.5537,
      "mean_token_accuracy": 0.17624777555465698,
      "num_tokens": 81859599.0,
      "step": 40400
    },
    {
      "entropy": 5.8900707244873045,
      "epoch": 3.530979638206764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003787714944660032,
      "loss": 5.4624,
      "mean_token_accuracy": 0.19681985825300216,
      "num_tokens": 81869497.0,
      "step": 40405
    },
    {
      "entropy": 5.9241663932800295,
      "epoch": 3.5314165865594687,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037874384445439743,
      "loss": 5.4184,
      "mean_token_accuracy": 0.19525830447673798,
      "num_tokens": 81878502.0,
      "step": 40410
    },
    {
      "entropy": 5.922545289993286,
      "epoch": 3.531853534912173,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003787161924529303,
      "loss": 5.3901,
      "mean_token_accuracy": 0.19668898731470108,
      "num_tokens": 81888748.0,
      "step": 40415
    },
    {
      "entropy": 5.878271818161011,
      "epoch": 3.532290483264878,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003786885384621323,
      "loss": 5.4055,
      "mean_token_accuracy": 0.196788389980793,
      "num_tokens": 81897705.0,
      "step": 40420
    },
    {
      "entropy": 5.882609272003174,
      "epoch": 3.532727431617583,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003786608824825338,
      "loss": 5.5187,
      "mean_token_accuracy": 0.1896447792649269,
      "num_tokens": 81909294.0,
      "step": 40425
    },
    {
      "entropy": 6.038519239425659,
      "epoch": 3.5331643799702874,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037863322451466523,
      "loss": 5.5571,
      "mean_token_accuracy": 0.18184809535741805,
      "num_tokens": 81919191.0,
      "step": 40430
    },
    {
      "entropy": 5.952717876434326,
      "epoch": 3.5336013283229923,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003786055645590571,
      "loss": 5.4245,
      "mean_token_accuracy": 0.1958484798669815,
      "num_tokens": 81928660.0,
      "step": 40435
    },
    {
      "entropy": 5.903357744216919,
      "epoch": 3.534038276675697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037857790261624006,
      "loss": 5.4691,
      "mean_token_accuracy": 0.19591008871793747,
      "num_tokens": 81938626.0,
      "step": 40440
    },
    {
      "entropy": 5.913920927047729,
      "epoch": 3.5344752250284017,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003785502386867445,
      "loss": 5.4598,
      "mean_token_accuracy": 0.1979872167110443,
      "num_tokens": 81948304.0,
      "step": 40445
    },
    {
      "entropy": 5.8771076679229735,
      "epoch": 3.534912173381106,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037852257277110106,
      "loss": 5.4633,
      "mean_token_accuracy": 0.19808624684810638,
      "num_tokens": 81958455.0,
      "step": 40450
    },
    {
      "entropy": 5.873592042922974,
      "epoch": 3.535349121733811,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003784949048698405,
      "loss": 5.3569,
      "mean_token_accuracy": 0.1999485895037651,
      "num_tokens": 81967809.0,
      "step": 40455
    },
    {
      "entropy": 5.8993659019470215,
      "epoch": 3.535786070086516,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003784672349834934,
      "loss": 5.4385,
      "mean_token_accuracy": 0.19616811722517014,
      "num_tokens": 81979801.0,
      "step": 40460
    },
    {
      "entropy": 5.900163650512695,
      "epoch": 3.5362230184392205,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037843956311259047,
      "loss": 5.4703,
      "mean_token_accuracy": 0.19287526458501816,
      "num_tokens": 81989591.0,
      "step": 40465
    },
    {
      "entropy": 5.980801200866699,
      "epoch": 3.536659966791925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003784118892576626,
      "loss": 5.4355,
      "mean_token_accuracy": 0.19210323989391326,
      "num_tokens": 81999470.0,
      "step": 40470
    },
    {
      "entropy": 5.861103153228759,
      "epoch": 3.53709691514463,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037838421341924045,
      "loss": 5.4446,
      "mean_token_accuracy": 0.19496857821941377,
      "num_tokens": 82010307.0,
      "step": 40475
    },
    {
      "entropy": 5.901847171783447,
      "epoch": 3.537533863497335,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037835653559785496,
      "loss": 5.3879,
      "mean_token_accuracy": 0.19360740929841996,
      "num_tokens": 82019997.0,
      "step": 40480
    },
    {
      "entropy": 5.930353212356567,
      "epoch": 3.5379708118500393,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003783288557940369,
      "loss": 5.3941,
      "mean_token_accuracy": 0.1929010882973671,
      "num_tokens": 82030328.0,
      "step": 40485
    },
    {
      "entropy": 5.97603211402893,
      "epoch": 3.538407760202744,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003783011740083173,
      "loss": 5.542,
      "mean_token_accuracy": 0.18188936561346053,
      "num_tokens": 82041153.0,
      "step": 40490
    },
    {
      "entropy": 5.847078037261963,
      "epoch": 3.5388447085554486,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003782734902412271,
      "loss": 5.3335,
      "mean_token_accuracy": 0.21046337038278579,
      "num_tokens": 82050962.0,
      "step": 40495
    },
    {
      "entropy": 5.898296165466308,
      "epoch": 3.5392816569081536,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037824580449329716,
      "loss": 5.4711,
      "mean_token_accuracy": 0.19494690448045732,
      "num_tokens": 82061033.0,
      "step": 40500
    },
    {
      "entropy": 5.945635271072388,
      "epoch": 3.539718605260858,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037821811676505855,
      "loss": 5.4563,
      "mean_token_accuracy": 0.19284846484661103,
      "num_tokens": 82070352.0,
      "step": 40505
    },
    {
      "entropy": 5.948750829696655,
      "epoch": 3.540155553613563,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037819042705704244,
      "loss": 5.4706,
      "mean_token_accuracy": 0.1902967557311058,
      "num_tokens": 82080608.0,
      "step": 40510
    },
    {
      "entropy": 5.944639492034912,
      "epoch": 3.540592501966268,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00037816273536977986,
      "loss": 5.417,
      "mean_token_accuracy": 0.1946392759680748,
      "num_tokens": 82089935.0,
      "step": 40515
    },
    {
      "entropy": 5.887181377410888,
      "epoch": 3.5410294503189723,
      "grad_norm": 1.046875,
      "learning_rate": 0.000378135041703802,
      "loss": 5.4688,
      "mean_token_accuracy": 0.19431037455797195,
      "num_tokens": 82100983.0,
      "step": 40520
    },
    {
      "entropy": 5.963826656341553,
      "epoch": 3.541466398671677,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037810734605963996,
      "loss": 5.5426,
      "mean_token_accuracy": 0.18254973143339157,
      "num_tokens": 82110791.0,
      "step": 40525
    },
    {
      "entropy": 5.937514734268189,
      "epoch": 3.5419033470243817,
      "grad_norm": 3.265625,
      "learning_rate": 0.000378079648437825,
      "loss": 5.4317,
      "mean_token_accuracy": 0.188762266933918,
      "num_tokens": 82120330.0,
      "step": 40530
    },
    {
      "entropy": 5.99147663116455,
      "epoch": 3.5423402953770866,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00037805194883888834,
      "loss": 5.4944,
      "mean_token_accuracy": 0.18798459470272064,
      "num_tokens": 82130589.0,
      "step": 40535
    },
    {
      "entropy": 5.925175952911377,
      "epoch": 3.542777243729791,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003780242472633614,
      "loss": 5.4422,
      "mean_token_accuracy": 0.19583311229944228,
      "num_tokens": 82141421.0,
      "step": 40540
    },
    {
      "entropy": 5.924081230163575,
      "epoch": 3.543214192082496,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003779965437117753,
      "loss": 5.4614,
      "mean_token_accuracy": 0.18963005095720292,
      "num_tokens": 82150688.0,
      "step": 40545
    },
    {
      "entropy": 6.027154397964478,
      "epoch": 3.5436511404352005,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037796883818466155,
      "loss": 5.5305,
      "mean_token_accuracy": 0.18523377180099487,
      "num_tokens": 82161850.0,
      "step": 40550
    },
    {
      "entropy": 5.995859956741333,
      "epoch": 3.5440880887879054,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037794113068255155,
      "loss": 5.464,
      "mean_token_accuracy": 0.18804397732019423,
      "num_tokens": 82171492.0,
      "step": 40555
    },
    {
      "entropy": 5.931247568130493,
      "epoch": 3.54452503714061,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037791342120597674,
      "loss": 5.526,
      "mean_token_accuracy": 0.18564552962779998,
      "num_tokens": 82182469.0,
      "step": 40560
    },
    {
      "entropy": 5.943647861480713,
      "epoch": 3.5449619854933148,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037788570975546854,
      "loss": 5.4405,
      "mean_token_accuracy": 0.193599109351635,
      "num_tokens": 82192395.0,
      "step": 40565
    },
    {
      "entropy": 5.929768180847168,
      "epoch": 3.5453989338460197,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037785799633155854,
      "loss": 5.4701,
      "mean_token_accuracy": 0.18377660512924193,
      "num_tokens": 82201555.0,
      "step": 40570
    },
    {
      "entropy": 5.911764907836914,
      "epoch": 3.545835882198724,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003778302809347782,
      "loss": 5.4694,
      "mean_token_accuracy": 0.18775930255651474,
      "num_tokens": 82211013.0,
      "step": 40575
    },
    {
      "entropy": 5.974308776855469,
      "epoch": 3.5462728305514286,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003778025635656593,
      "loss": 5.4853,
      "mean_token_accuracy": 0.1854271411895752,
      "num_tokens": 82221317.0,
      "step": 40580
    },
    {
      "entropy": 5.922763347625732,
      "epoch": 3.5467097789041335,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003777748442247333,
      "loss": 5.4303,
      "mean_token_accuracy": 0.1933995693922043,
      "num_tokens": 82230837.0,
      "step": 40585
    },
    {
      "entropy": 5.946191263198853,
      "epoch": 3.5471467272568384,
      "grad_norm": 1.125,
      "learning_rate": 0.000377747122912532,
      "loss": 5.516,
      "mean_token_accuracy": 0.1882798820734024,
      "num_tokens": 82241034.0,
      "step": 40590
    },
    {
      "entropy": 5.921898365020752,
      "epoch": 3.547583675609543,
      "grad_norm": 1.140625,
      "learning_rate": 0.000377719399629587,
      "loss": 5.4867,
      "mean_token_accuracy": 0.19025015085935593,
      "num_tokens": 82251047.0,
      "step": 40595
    },
    {
      "entropy": 6.029739141464233,
      "epoch": 3.548020623962248,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003776916743764301,
      "loss": 5.4699,
      "mean_token_accuracy": 0.1869089871644974,
      "num_tokens": 82261215.0,
      "step": 40600
    },
    {
      "entropy": 5.94800968170166,
      "epoch": 3.5484575723149523,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003776639471535931,
      "loss": 5.4866,
      "mean_token_accuracy": 0.18749197274446489,
      "num_tokens": 82271527.0,
      "step": 40605
    },
    {
      "entropy": 5.974728631973266,
      "epoch": 3.548894520667657,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003776362179616077,
      "loss": 5.6001,
      "mean_token_accuracy": 0.1776071935892105,
      "num_tokens": 82282021.0,
      "step": 40610
    },
    {
      "entropy": 5.962168216705322,
      "epoch": 3.5493314690203617,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037760848680100595,
      "loss": 5.5354,
      "mean_token_accuracy": 0.18077332079410552,
      "num_tokens": 82291279.0,
      "step": 40615
    },
    {
      "entropy": 5.9127342224121096,
      "epoch": 3.5497684173730666,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003775807536723196,
      "loss": 5.3898,
      "mean_token_accuracy": 0.19661909639835357,
      "num_tokens": 82300756.0,
      "step": 40620
    },
    {
      "entropy": 5.9492823600769045,
      "epoch": 3.550205365725771,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003775530185760807,
      "loss": 5.479,
      "mean_token_accuracy": 0.18991570621728898,
      "num_tokens": 82309896.0,
      "step": 40625
    },
    {
      "entropy": 5.880956983566284,
      "epoch": 3.550642314078476,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003775252815128211,
      "loss": 5.3395,
      "mean_token_accuracy": 0.20227806121110917,
      "num_tokens": 82319646.0,
      "step": 40630
    },
    {
      "entropy": 5.918704652786255,
      "epoch": 3.5510792624311804,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003774975424830729,
      "loss": 5.5213,
      "mean_token_accuracy": 0.18501628786325455,
      "num_tokens": 82329492.0,
      "step": 40635
    },
    {
      "entropy": 5.904903221130371,
      "epoch": 3.5515162107838854,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003774698014873682,
      "loss": 5.4435,
      "mean_token_accuracy": 0.19441934376955033,
      "num_tokens": 82340339.0,
      "step": 40640
    },
    {
      "entropy": 5.928556680679321,
      "epoch": 3.5519531591365903,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037744205852623895,
      "loss": 5.4775,
      "mean_token_accuracy": 0.1892494186758995,
      "num_tokens": 82351337.0,
      "step": 40645
    },
    {
      "entropy": 5.894093561172485,
      "epoch": 3.5523901074892947,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003774143136002173,
      "loss": 5.4573,
      "mean_token_accuracy": 0.19510486274957656,
      "num_tokens": 82361616.0,
      "step": 40650
    },
    {
      "entropy": 5.926919460296631,
      "epoch": 3.552827055841999,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037738656670983553,
      "loss": 5.4256,
      "mean_token_accuracy": 0.1846161350607872,
      "num_tokens": 82371359.0,
      "step": 40655
    },
    {
      "entropy": 5.869516038894654,
      "epoch": 3.553264004194704,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003773588178556258,
      "loss": 5.354,
      "mean_token_accuracy": 0.19409862756729127,
      "num_tokens": 82381575.0,
      "step": 40660
    },
    {
      "entropy": 5.819342947006225,
      "epoch": 3.553700952547409,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003773310670381202,
      "loss": 5.4455,
      "mean_token_accuracy": 0.1939442291855812,
      "num_tokens": 82392175.0,
      "step": 40665
    },
    {
      "entropy": 5.9649927616119385,
      "epoch": 3.5541379009001135,
      "grad_norm": 1.78125,
      "learning_rate": 0.00037730331425785114,
      "loss": 5.5196,
      "mean_token_accuracy": 0.18499539643526078,
      "num_tokens": 82402570.0,
      "step": 40670
    },
    {
      "entropy": 6.016451120376587,
      "epoch": 3.5545748492528184,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00037727555951535093,
      "loss": 5.4893,
      "mean_token_accuracy": 0.18650226593017577,
      "num_tokens": 82413433.0,
      "step": 40675
    },
    {
      "entropy": 5.982560300827027,
      "epoch": 3.555011797605523,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003772478028111519,
      "loss": 5.5031,
      "mean_token_accuracy": 0.18238288164138794,
      "num_tokens": 82423144.0,
      "step": 40680
    },
    {
      "entropy": 5.867653465270996,
      "epoch": 3.555448745958228,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003772200441457864,
      "loss": 5.4073,
      "mean_token_accuracy": 0.19295885860919954,
      "num_tokens": 82433096.0,
      "step": 40685
    },
    {
      "entropy": 5.82521619796753,
      "epoch": 3.5558856943109323,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003771922835197869,
      "loss": 5.3909,
      "mean_token_accuracy": 0.19852253645658494,
      "num_tokens": 82444261.0,
      "step": 40690
    },
    {
      "entropy": 6.023891448974609,
      "epoch": 3.556322642663637,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003771645209336859,
      "loss": 5.5411,
      "mean_token_accuracy": 0.18298961520195006,
      "num_tokens": 82454442.0,
      "step": 40695
    },
    {
      "entropy": 5.944293594360351,
      "epoch": 3.556759591016342,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003771367563880158,
      "loss": 5.507,
      "mean_token_accuracy": 0.1860708087682724,
      "num_tokens": 82464589.0,
      "step": 40700
    },
    {
      "entropy": 5.905562686920166,
      "epoch": 3.5571965393690466,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037710898988330924,
      "loss": 5.5182,
      "mean_token_accuracy": 0.18731743842363358,
      "num_tokens": 82475624.0,
      "step": 40705
    },
    {
      "entropy": 5.971003437042237,
      "epoch": 3.557633487721751,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003770812214200987,
      "loss": 5.4873,
      "mean_token_accuracy": 0.19163680225610732,
      "num_tokens": 82485142.0,
      "step": 40710
    },
    {
      "entropy": 5.984836673736572,
      "epoch": 3.558070436074456,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037705345099891684,
      "loss": 5.4997,
      "mean_token_accuracy": 0.1853805124759674,
      "num_tokens": 82495165.0,
      "step": 40715
    },
    {
      "entropy": 5.842480564117432,
      "epoch": 3.558507384427161,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003770256786202964,
      "loss": 5.4027,
      "mean_token_accuracy": 0.1854572296142578,
      "num_tokens": 82505867.0,
      "step": 40720
    },
    {
      "entropy": 5.9822674751281735,
      "epoch": 3.5589443327798653,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00037699790428476996,
      "loss": 5.5572,
      "mean_token_accuracy": 0.1821570798754692,
      "num_tokens": 82517416.0,
      "step": 40725
    },
    {
      "entropy": 6.001047563552857,
      "epoch": 3.5593812811325702,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003769701279928702,
      "loss": 5.4673,
      "mean_token_accuracy": 0.19775970727205278,
      "num_tokens": 82527192.0,
      "step": 40730
    },
    {
      "entropy": 5.965149402618408,
      "epoch": 3.5598182294852747,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037694234974512995,
      "loss": 5.4829,
      "mean_token_accuracy": 0.18871565014123917,
      "num_tokens": 82536600.0,
      "step": 40735
    },
    {
      "entropy": 5.955157613754272,
      "epoch": 3.5602551778379796,
      "grad_norm": 1.0,
      "learning_rate": 0.00037691456954208207,
      "loss": 5.5363,
      "mean_token_accuracy": 0.18813802897930146,
      "num_tokens": 82546899.0,
      "step": 40740
    },
    {
      "entropy": 5.930088806152344,
      "epoch": 3.560692126190684,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003768867873842593,
      "loss": 5.497,
      "mean_token_accuracy": 0.1875934347510338,
      "num_tokens": 82557884.0,
      "step": 40745
    },
    {
      "entropy": 5.943565130233765,
      "epoch": 3.561129074543389,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003768590032721946,
      "loss": 5.369,
      "mean_token_accuracy": 0.20208689868450164,
      "num_tokens": 82567726.0,
      "step": 40750
    },
    {
      "entropy": 5.96208176612854,
      "epoch": 3.561566022896094,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037683121720642087,
      "loss": 5.5393,
      "mean_token_accuracy": 0.18691059499979018,
      "num_tokens": 82577741.0,
      "step": 40755
    },
    {
      "entropy": 5.980183935165405,
      "epoch": 3.5620029712487984,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037680342918747103,
      "loss": 5.5418,
      "mean_token_accuracy": 0.17907738387584687,
      "num_tokens": 82588591.0,
      "step": 40760
    },
    {
      "entropy": 5.913513135910034,
      "epoch": 3.562439919601503,
      "grad_norm": 1.0625,
      "learning_rate": 0.000376775639215878,
      "loss": 5.4309,
      "mean_token_accuracy": 0.18866890519857407,
      "num_tokens": 82598276.0,
      "step": 40765
    },
    {
      "entropy": 5.87406358718872,
      "epoch": 3.5628768679542078,
      "grad_norm": 1.0,
      "learning_rate": 0.00037674784729217493,
      "loss": 5.4473,
      "mean_token_accuracy": 0.1920888066291809,
      "num_tokens": 82608715.0,
      "step": 40770
    },
    {
      "entropy": 5.953827667236328,
      "epoch": 3.5633138163069127,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037672005341689486,
      "loss": 5.5114,
      "mean_token_accuracy": 0.19147929400205613,
      "num_tokens": 82618854.0,
      "step": 40775
    },
    {
      "entropy": 5.936001110076904,
      "epoch": 3.563750764659617,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003766922575905708,
      "loss": 5.419,
      "mean_token_accuracy": 0.19719327688217164,
      "num_tokens": 82628391.0,
      "step": 40780
    },
    {
      "entropy": 5.963489961624146,
      "epoch": 3.564187713012322,
      "grad_norm": 1.0625,
      "learning_rate": 0.000376664459813736,
      "loss": 5.5811,
      "mean_token_accuracy": 0.18241357654333115,
      "num_tokens": 82639560.0,
      "step": 40785
    },
    {
      "entropy": 5.883200740814209,
      "epoch": 3.5646246613650265,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037663666008692356,
      "loss": 5.4647,
      "mean_token_accuracy": 0.1904116153717041,
      "num_tokens": 82650252.0,
      "step": 40790
    },
    {
      "entropy": 6.005696821212768,
      "epoch": 3.5650616097177315,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003766088584106667,
      "loss": 5.4895,
      "mean_token_accuracy": 0.18857313990592955,
      "num_tokens": 82660035.0,
      "step": 40795
    },
    {
      "entropy": 5.883221626281738,
      "epoch": 3.565498558070436,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037658105478549875,
      "loss": 5.4117,
      "mean_token_accuracy": 0.19707635045051575,
      "num_tokens": 82669435.0,
      "step": 40800
    },
    {
      "entropy": 5.9752771854400635,
      "epoch": 3.565935506423141,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037655324921195295,
      "loss": 5.5115,
      "mean_token_accuracy": 0.18276509940624236,
      "num_tokens": 82679314.0,
      "step": 40805
    },
    {
      "entropy": 5.89200177192688,
      "epoch": 3.5663724547758457,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003765254416905626,
      "loss": 5.4184,
      "mean_token_accuracy": 0.1926673471927643,
      "num_tokens": 82690140.0,
      "step": 40810
    },
    {
      "entropy": 5.920170259475708,
      "epoch": 3.56680940312855,
      "grad_norm": 1.078125,
      "learning_rate": 0.000376497632221861,
      "loss": 5.465,
      "mean_token_accuracy": 0.19145489484071732,
      "num_tokens": 82699210.0,
      "step": 40815
    },
    {
      "entropy": 6.02099838256836,
      "epoch": 3.5672463514812547,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003764698208063818,
      "loss": 5.5809,
      "mean_token_accuracy": 0.18186419010162352,
      "num_tokens": 82710398.0,
      "step": 40820
    },
    {
      "entropy": 6.051115322113037,
      "epoch": 3.5676832998339596,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003764420074446581,
      "loss": 5.4577,
      "mean_token_accuracy": 0.18711840212345124,
      "num_tokens": 82721036.0,
      "step": 40825
    },
    {
      "entropy": 5.981624937057495,
      "epoch": 3.5681202481866645,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003764141921372236,
      "loss": 5.5239,
      "mean_token_accuracy": 0.18489535599946977,
      "num_tokens": 82730884.0,
      "step": 40830
    },
    {
      "entropy": 5.922293615341187,
      "epoch": 3.568557196539369,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003763863748846118,
      "loss": 5.433,
      "mean_token_accuracy": 0.1935209169983864,
      "num_tokens": 82740561.0,
      "step": 40835
    },
    {
      "entropy": 5.97333345413208,
      "epoch": 3.568994144892074,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037635855568735613,
      "loss": 5.552,
      "mean_token_accuracy": 0.1791262447834015,
      "num_tokens": 82751064.0,
      "step": 40840
    },
    {
      "entropy": 5.96770133972168,
      "epoch": 3.5694310932447784,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003763307345459903,
      "loss": 5.4161,
      "mean_token_accuracy": 0.18864326626062394,
      "num_tokens": 82760502.0,
      "step": 40845
    },
    {
      "entropy": 5.888483381271362,
      "epoch": 3.5698680415974833,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003763029114610479,
      "loss": 5.4916,
      "mean_token_accuracy": 0.19035072326660157,
      "num_tokens": 82771477.0,
      "step": 40850
    },
    {
      "entropy": 5.915207195281982,
      "epoch": 3.5703049899501877,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0003762750864330625,
      "loss": 5.4523,
      "mean_token_accuracy": 0.1937395691871643,
      "num_tokens": 82781572.0,
      "step": 40855
    },
    {
      "entropy": 5.999121236801147,
      "epoch": 3.5707419383028927,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037624725946256793,
      "loss": 5.3997,
      "mean_token_accuracy": 0.19266846925020217,
      "num_tokens": 82790217.0,
      "step": 40860
    },
    {
      "entropy": 5.953083610534668,
      "epoch": 3.5711788866555976,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00037621943055009786,
      "loss": 5.481,
      "mean_token_accuracy": 0.1919805720448494,
      "num_tokens": 82800566.0,
      "step": 40865
    },
    {
      "entropy": 5.929382991790772,
      "epoch": 3.571615835008302,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003761915996961861,
      "loss": 5.4813,
      "mean_token_accuracy": 0.1936257392168045,
      "num_tokens": 82810188.0,
      "step": 40870
    },
    {
      "entropy": 5.937492561340332,
      "epoch": 3.5720527833610065,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037616376690136646,
      "loss": 5.4815,
      "mean_token_accuracy": 0.18919043242931366,
      "num_tokens": 82819509.0,
      "step": 40875
    },
    {
      "entropy": 5.97958812713623,
      "epoch": 3.5724897317137114,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003761359321661727,
      "loss": 5.4913,
      "mean_token_accuracy": 0.1818997785449028,
      "num_tokens": 82830172.0,
      "step": 40880
    },
    {
      "entropy": 5.964294910430908,
      "epoch": 3.5729266800664163,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037610809549113876,
      "loss": 5.4767,
      "mean_token_accuracy": 0.19275061339139937,
      "num_tokens": 82839245.0,
      "step": 40885
    },
    {
      "entropy": 5.932531404495239,
      "epoch": 3.573363628419121,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003760802568767986,
      "loss": 5.5256,
      "mean_token_accuracy": 0.18480328768491744,
      "num_tokens": 82850479.0,
      "step": 40890
    },
    {
      "entropy": 5.935771322250366,
      "epoch": 3.5738005767718257,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003760524163236862,
      "loss": 5.4709,
      "mean_token_accuracy": 0.18439532071352005,
      "num_tokens": 82859816.0,
      "step": 40895
    },
    {
      "entropy": 5.924276876449585,
      "epoch": 3.57423752512453,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003760245738323354,
      "loss": 5.5323,
      "mean_token_accuracy": 0.1829732909798622,
      "num_tokens": 82870124.0,
      "step": 40900
    },
    {
      "entropy": 5.9985740184783936,
      "epoch": 3.574674473477235,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003759967294032803,
      "loss": 5.485,
      "mean_token_accuracy": 0.1856289893388748,
      "num_tokens": 82880721.0,
      "step": 40905
    },
    {
      "entropy": 5.983310174942017,
      "epoch": 3.5751114218299396,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003759688830370551,
      "loss": 5.5042,
      "mean_token_accuracy": 0.18530227690935136,
      "num_tokens": 82890580.0,
      "step": 40910
    },
    {
      "entropy": 5.962758779525757,
      "epoch": 3.5755483701826445,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037594103473419384,
      "loss": 5.6294,
      "mean_token_accuracy": 0.1770484894514084,
      "num_tokens": 82902465.0,
      "step": 40915
    },
    {
      "entropy": 5.91344313621521,
      "epoch": 3.575985318535349,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037591318449523063,
      "loss": 5.4321,
      "mean_token_accuracy": 0.19187518060207367,
      "num_tokens": 82912665.0,
      "step": 40920
    },
    {
      "entropy": 5.877542638778687,
      "epoch": 3.576422266888054,
      "grad_norm": 1.125,
      "learning_rate": 0.00037588533232069967,
      "loss": 5.3809,
      "mean_token_accuracy": 0.19355624318122863,
      "num_tokens": 82921727.0,
      "step": 40925
    },
    {
      "entropy": 5.95458083152771,
      "epoch": 3.5768592152407583,
      "grad_norm": 1.125,
      "learning_rate": 0.0003758574782111351,
      "loss": 5.4633,
      "mean_token_accuracy": 0.18867013603448868,
      "num_tokens": 82932449.0,
      "step": 40930
    },
    {
      "entropy": 5.885551071166992,
      "epoch": 3.5772961635934633,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003758296221670713,
      "loss": 5.3752,
      "mean_token_accuracy": 0.19630831927061082,
      "num_tokens": 82941939.0,
      "step": 40935
    },
    {
      "entropy": 5.8166179180145265,
      "epoch": 3.577733111946168,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003758017641890425,
      "loss": 5.3896,
      "mean_token_accuracy": 0.20088217556476592,
      "num_tokens": 82951958.0,
      "step": 40940
    },
    {
      "entropy": 5.830600118637085,
      "epoch": 3.5781700602988726,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003757739042775831,
      "loss": 5.4535,
      "mean_token_accuracy": 0.20630926191806792,
      "num_tokens": 82963014.0,
      "step": 40945
    },
    {
      "entropy": 5.919361877441406,
      "epoch": 3.5786070086515775,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003757460424332273,
      "loss": 5.4595,
      "mean_token_accuracy": 0.1890733554959297,
      "num_tokens": 82973081.0,
      "step": 40950
    },
    {
      "entropy": 5.97399935722351,
      "epoch": 3.579043957004282,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003757181786565097,
      "loss": 5.4863,
      "mean_token_accuracy": 0.18600644767284394,
      "num_tokens": 82982835.0,
      "step": 40955
    },
    {
      "entropy": 5.925924587249756,
      "epoch": 3.579480905356987,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003756903129479646,
      "loss": 5.4723,
      "mean_token_accuracy": 0.1893023356795311,
      "num_tokens": 82993275.0,
      "step": 40960
    },
    {
      "entropy": 5.981524515151977,
      "epoch": 3.5799178537096914,
      "grad_norm": 1.5,
      "learning_rate": 0.0003756624453081266,
      "loss": 5.5251,
      "mean_token_accuracy": 0.18468817323446274,
      "num_tokens": 83003844.0,
      "step": 40965
    },
    {
      "entropy": 5.964415788650513,
      "epoch": 3.5803548020623963,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003756345757375302,
      "loss": 5.4828,
      "mean_token_accuracy": 0.18560860902071,
      "num_tokens": 83014248.0,
      "step": 40970
    },
    {
      "entropy": 5.87408766746521,
      "epoch": 3.580791750415101,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003756067042367098,
      "loss": 5.4223,
      "mean_token_accuracy": 0.1927127093076706,
      "num_tokens": 83023967.0,
      "step": 40975
    },
    {
      "entropy": 5.962201499938965,
      "epoch": 3.5812286987678057,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003755788308062001,
      "loss": 5.5089,
      "mean_token_accuracy": 0.18721868246793746,
      "num_tokens": 83034167.0,
      "step": 40980
    },
    {
      "entropy": 5.879263162612915,
      "epoch": 3.58166564712051,
      "grad_norm": 0.984375,
      "learning_rate": 0.00037555095544653574,
      "loss": 5.4503,
      "mean_token_accuracy": 0.1943529650568962,
      "num_tokens": 83043822.0,
      "step": 40985
    },
    {
      "entropy": 5.881018018722534,
      "epoch": 3.582102595473215,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037552307815825136,
      "loss": 5.4765,
      "mean_token_accuracy": 0.19400246143341066,
      "num_tokens": 83054152.0,
      "step": 40990
    },
    {
      "entropy": 5.962343454360962,
      "epoch": 3.58253954382592,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00037549519894188174,
      "loss": 5.4341,
      "mean_token_accuracy": 0.1932268813252449,
      "num_tokens": 83064553.0,
      "step": 40995
    },
    {
      "entropy": 5.880416536331177,
      "epoch": 3.5829764921786245,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003754673177979614,
      "loss": 5.4522,
      "mean_token_accuracy": 0.1952180430293083,
      "num_tokens": 83073423.0,
      "step": 41000
    },
    {
      "entropy": 5.917631196975708,
      "epoch": 3.583413440531329,
      "grad_norm": 0.890625,
      "learning_rate": 0.00037543943472702536,
      "loss": 5.469,
      "mean_token_accuracy": 0.19100075960159302,
      "num_tokens": 83083774.0,
      "step": 41005
    },
    {
      "entropy": 5.840063667297363,
      "epoch": 3.583850388884034,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003754115497296084,
      "loss": 5.3593,
      "mean_token_accuracy": 0.1994456559419632,
      "num_tokens": 83092647.0,
      "step": 41010
    },
    {
      "entropy": 5.945784997940064,
      "epoch": 3.5842873372367388,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037538366280624513,
      "loss": 5.6233,
      "mean_token_accuracy": 0.18174557983875275,
      "num_tokens": 83104070.0,
      "step": 41015
    },
    {
      "entropy": 5.971376895904541,
      "epoch": 3.5847242855894432,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003753557739574707,
      "loss": 5.4457,
      "mean_token_accuracy": 0.19475086480379106,
      "num_tokens": 83113747.0,
      "step": 41020
    },
    {
      "entropy": 5.957550048828125,
      "epoch": 3.585161233942148,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037532788318382,
      "loss": 5.5035,
      "mean_token_accuracy": 0.180975079536438,
      "num_tokens": 83123678.0,
      "step": 41025
    },
    {
      "entropy": 6.002565670013428,
      "epoch": 3.5855981822948526,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003752999904858278,
      "loss": 5.4997,
      "mean_token_accuracy": 0.1856356829404831,
      "num_tokens": 83133232.0,
      "step": 41030
    },
    {
      "entropy": 6.005835819244385,
      "epoch": 3.5860351306475575,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037527209586402934,
      "loss": 5.5256,
      "mean_token_accuracy": 0.1826336622238159,
      "num_tokens": 83144351.0,
      "step": 41035
    },
    {
      "entropy": 5.945231199264526,
      "epoch": 3.586472079000262,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003752441993189595,
      "loss": 5.5081,
      "mean_token_accuracy": 0.18576153218746186,
      "num_tokens": 83155116.0,
      "step": 41040
    },
    {
      "entropy": 5.933318948745727,
      "epoch": 3.586909027352967,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003752163008511534,
      "loss": 5.5187,
      "mean_token_accuracy": 0.18729946613311768,
      "num_tokens": 83165227.0,
      "step": 41045
    },
    {
      "entropy": 6.007360887527466,
      "epoch": 3.587345975705672,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003751884004611461,
      "loss": 5.5258,
      "mean_token_accuracy": 0.18826237618923186,
      "num_tokens": 83175235.0,
      "step": 41050
    },
    {
      "entropy": 5.9488507270812985,
      "epoch": 3.5877829240583763,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037516049814947277,
      "loss": 5.4566,
      "mean_token_accuracy": 0.18739547580480576,
      "num_tokens": 83185373.0,
      "step": 41055
    },
    {
      "entropy": 6.043288803100586,
      "epoch": 3.5882198724110808,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003751325939166687,
      "loss": 5.517,
      "mean_token_accuracy": 0.18270884156227113,
      "num_tokens": 83195046.0,
      "step": 41060
    },
    {
      "entropy": 5.931273031234741,
      "epoch": 3.5886568207637857,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037510468776326894,
      "loss": 5.4866,
      "mean_token_accuracy": 0.1932094246149063,
      "num_tokens": 83203817.0,
      "step": 41065
    },
    {
      "entropy": 6.0013782501220705,
      "epoch": 3.5890937691164906,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003750767796898088,
      "loss": 5.5474,
      "mean_token_accuracy": 0.18637098073959352,
      "num_tokens": 83215096.0,
      "step": 41070
    },
    {
      "entropy": 5.941671371459961,
      "epoch": 3.589530717469195,
      "grad_norm": 1.0,
      "learning_rate": 0.0003750488696968236,
      "loss": 5.5681,
      "mean_token_accuracy": 0.18878596276044846,
      "num_tokens": 83225547.0,
      "step": 41075
    },
    {
      "entropy": 5.927292776107788,
      "epoch": 3.5899676658219,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037502095778484866,
      "loss": 5.4376,
      "mean_token_accuracy": 0.19225132912397386,
      "num_tokens": 83235658.0,
      "step": 41080
    },
    {
      "entropy": 5.900456380844116,
      "epoch": 3.5904046141746044,
      "grad_norm": 1.0,
      "learning_rate": 0.00037499304395441943,
      "loss": 5.4455,
      "mean_token_accuracy": 0.18540063947439195,
      "num_tokens": 83244819.0,
      "step": 41085
    },
    {
      "entropy": 5.902797412872315,
      "epoch": 3.5908415625273093,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003749651282060711,
      "loss": 5.4144,
      "mean_token_accuracy": 0.205031418800354,
      "num_tokens": 83255729.0,
      "step": 41090
    },
    {
      "entropy": 5.982131481170654,
      "epoch": 3.591278510880014,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003749372105403392,
      "loss": 5.5006,
      "mean_token_accuracy": 0.19162437915802003,
      "num_tokens": 83265687.0,
      "step": 41095
    },
    {
      "entropy": 5.934624052047729,
      "epoch": 3.5917154592327187,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037490929095775934,
      "loss": 5.4568,
      "mean_token_accuracy": 0.19049558639526368,
      "num_tokens": 83275558.0,
      "step": 41100
    },
    {
      "entropy": 5.934456872940063,
      "epoch": 3.5921524075854236,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037488136945886685,
      "loss": 5.4298,
      "mean_token_accuracy": 0.1957937002182007,
      "num_tokens": 83284744.0,
      "step": 41105
    },
    {
      "entropy": 5.925251722335815,
      "epoch": 3.592589355938128,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003748534460441975,
      "loss": 5.4717,
      "mean_token_accuracy": 0.1892337054014206,
      "num_tokens": 83294372.0,
      "step": 41110
    },
    {
      "entropy": 5.939765167236328,
      "epoch": 3.5930263042908326,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037482552071428654,
      "loss": 5.4344,
      "mean_token_accuracy": 0.1897716000676155,
      "num_tokens": 83305378.0,
      "step": 41115
    },
    {
      "entropy": 5.990040349960327,
      "epoch": 3.5934632526435375,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037479759346966985,
      "loss": 5.5469,
      "mean_token_accuracy": 0.18399833589792253,
      "num_tokens": 83315073.0,
      "step": 41120
    },
    {
      "entropy": 5.867813301086426,
      "epoch": 3.5939002009962424,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037476966431088305,
      "loss": 5.3693,
      "mean_token_accuracy": 0.20023904889822006,
      "num_tokens": 83324940.0,
      "step": 41125
    },
    {
      "entropy": 5.947489213943482,
      "epoch": 3.594337149348947,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003747417332384618,
      "loss": 5.534,
      "mean_token_accuracy": 0.18330475240945815,
      "num_tokens": 83335678.0,
      "step": 41130
    },
    {
      "entropy": 5.882610559463501,
      "epoch": 3.594774097701652,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00037471380025294186,
      "loss": 5.4053,
      "mean_token_accuracy": 0.199093821644783,
      "num_tokens": 83345682.0,
      "step": 41135
    },
    {
      "entropy": 5.860930490493774,
      "epoch": 3.5952110460543563,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000374685865354859,
      "loss": 5.4572,
      "mean_token_accuracy": 0.1970256060361862,
      "num_tokens": 83354934.0,
      "step": 41140
    },
    {
      "entropy": 5.909845399856567,
      "epoch": 3.595647994407061,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000374657928544749,
      "loss": 5.5174,
      "mean_token_accuracy": 0.18545733839273454,
      "num_tokens": 83365647.0,
      "step": 41145
    },
    {
      "entropy": 5.934750938415528,
      "epoch": 3.5960849427597656,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003746299898231477,
      "loss": 5.4722,
      "mean_token_accuracy": 0.19324229806661605,
      "num_tokens": 83376040.0,
      "step": 41150
    },
    {
      "entropy": 5.941199588775635,
      "epoch": 3.5965218911124706,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003746020491905909,
      "loss": 5.4354,
      "mean_token_accuracy": 0.19336924850940704,
      "num_tokens": 83386097.0,
      "step": 41155
    },
    {
      "entropy": 5.975954675674439,
      "epoch": 3.5969588394651755,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037457410664761475,
      "loss": 5.5875,
      "mean_token_accuracy": 0.1785598248243332,
      "num_tokens": 83395555.0,
      "step": 41160
    },
    {
      "entropy": 5.954074430465698,
      "epoch": 3.59739578781788,
      "grad_norm": 1.84375,
      "learning_rate": 0.000374546162194755,
      "loss": 5.451,
      "mean_token_accuracy": 0.19354101717472078,
      "num_tokens": 83406195.0,
      "step": 41165
    },
    {
      "entropy": 5.95368447303772,
      "epoch": 3.5978327361705844,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037451821583254764,
      "loss": 5.4215,
      "mean_token_accuracy": 0.19668751657009126,
      "num_tokens": 83415254.0,
      "step": 41170
    },
    {
      "entropy": 5.883230972290039,
      "epoch": 3.5982696845232893,
      "grad_norm": 1.0,
      "learning_rate": 0.00037449026756152876,
      "loss": 5.4317,
      "mean_token_accuracy": 0.19202427119016646,
      "num_tokens": 83425112.0,
      "step": 41175
    },
    {
      "entropy": 5.897903823852539,
      "epoch": 3.5987066328759942,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037446231738223453,
      "loss": 5.5444,
      "mean_token_accuracy": 0.17671126276254653,
      "num_tokens": 83436185.0,
      "step": 41180
    },
    {
      "entropy": 6.032114744186401,
      "epoch": 3.5991435812286987,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003744343652952008,
      "loss": 5.4885,
      "mean_token_accuracy": 0.18356257528066636,
      "num_tokens": 83446540.0,
      "step": 41185
    },
    {
      "entropy": 5.965924644470215,
      "epoch": 3.5995805295814036,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037440641130096386,
      "loss": 5.4204,
      "mean_token_accuracy": 0.19413010478019715,
      "num_tokens": 83457247.0,
      "step": 41190
    },
    {
      "entropy": 5.931784963607788,
      "epoch": 3.600017477934108,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003743784554000599,
      "loss": 5.4716,
      "mean_token_accuracy": 0.1944369986653328,
      "num_tokens": 83468132.0,
      "step": 41195
    },
    {
      "entropy": 5.870323657989502,
      "epoch": 3.600454426286813,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003743504975930252,
      "loss": 5.4417,
      "mean_token_accuracy": 0.1931237742304802,
      "num_tokens": 83477824.0,
      "step": 41200
    },
    {
      "entropy": 5.968573045730591,
      "epoch": 3.6008913746395175,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003743225378803958,
      "loss": 5.4617,
      "mean_token_accuracy": 0.1863288551568985,
      "num_tokens": 83487352.0,
      "step": 41205
    },
    {
      "entropy": 5.934669065475464,
      "epoch": 3.6013283229922224,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003742945762627081,
      "loss": 5.4678,
      "mean_token_accuracy": 0.19376259446144103,
      "num_tokens": 83497957.0,
      "step": 41210
    },
    {
      "entropy": 5.883463621139526,
      "epoch": 3.6017652713449273,
      "grad_norm": 1.0,
      "learning_rate": 0.0003742666127404983,
      "loss": 5.3995,
      "mean_token_accuracy": 0.19603385925292968,
      "num_tokens": 83508213.0,
      "step": 41215
    },
    {
      "entropy": 5.8932736873626705,
      "epoch": 3.6022022196976318,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003742386473143029,
      "loss": 5.4474,
      "mean_token_accuracy": 0.19694723188877106,
      "num_tokens": 83518694.0,
      "step": 41220
    },
    {
      "entropy": 5.909800577163696,
      "epoch": 3.6026391680503362,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037421067998465835,
      "loss": 5.4131,
      "mean_token_accuracy": 0.19463701993227006,
      "num_tokens": 83529077.0,
      "step": 41225
    },
    {
      "entropy": 5.889842748641968,
      "epoch": 3.603076116403041,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00037418271075210087,
      "loss": 5.4459,
      "mean_token_accuracy": 0.1935655727982521,
      "num_tokens": 83539502.0,
      "step": 41230
    },
    {
      "entropy": 5.884242820739746,
      "epoch": 3.603513064755746,
      "grad_norm": 1.046875,
      "learning_rate": 0.000374154739617167,
      "loss": 5.4122,
      "mean_token_accuracy": 0.1970929741859436,
      "num_tokens": 83549073.0,
      "step": 41235
    },
    {
      "entropy": 5.9176383972167965,
      "epoch": 3.6039500131084505,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003741267665803934,
      "loss": 5.5035,
      "mean_token_accuracy": 0.1885184556245804,
      "num_tokens": 83560404.0,
      "step": 41240
    },
    {
      "entropy": 5.888156080245972,
      "epoch": 3.6043869614611554,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003740987916423164,
      "loss": 5.4091,
      "mean_token_accuracy": 0.19507601261138915,
      "num_tokens": 83570421.0,
      "step": 41245
    },
    {
      "entropy": 5.990869808197021,
      "epoch": 3.60482390981386,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037407081480347263,
      "loss": 5.4889,
      "mean_token_accuracy": 0.18496885597705842,
      "num_tokens": 83581653.0,
      "step": 41250
    },
    {
      "entropy": 5.942560005187988,
      "epoch": 3.605260858166565,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003740428360643987,
      "loss": 5.4096,
      "mean_token_accuracy": 0.19206136763095855,
      "num_tokens": 83592190.0,
      "step": 41255
    },
    {
      "entropy": 5.8955525875091555,
      "epoch": 3.6056978065192693,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003740148554256313,
      "loss": 5.4525,
      "mean_token_accuracy": 0.1829906553030014,
      "num_tokens": 83602495.0,
      "step": 41260
    },
    {
      "entropy": 5.902038955688477,
      "epoch": 3.606134754871974,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003739868728877071,
      "loss": 5.4462,
      "mean_token_accuracy": 0.18991309106349946,
      "num_tokens": 83612772.0,
      "step": 41265
    },
    {
      "entropy": 5.946122407913208,
      "epoch": 3.6065717032246787,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003739588884511628,
      "loss": 5.5055,
      "mean_token_accuracy": 0.18983375132083893,
      "num_tokens": 83623472.0,
      "step": 41270
    },
    {
      "entropy": 5.937967348098755,
      "epoch": 3.6070086515773836,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003739309021165352,
      "loss": 5.428,
      "mean_token_accuracy": 0.19197224229574203,
      "num_tokens": 83633783.0,
      "step": 41275
    },
    {
      "entropy": 5.926026201248169,
      "epoch": 3.607445599930088,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037390291388436105,
      "loss": 5.5025,
      "mean_token_accuracy": 0.1869748830795288,
      "num_tokens": 83643594.0,
      "step": 41280
    },
    {
      "entropy": 5.881234073638916,
      "epoch": 3.607882548282793,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003738749237551772,
      "loss": 5.3583,
      "mean_token_accuracy": 0.20453914552927016,
      "num_tokens": 83653405.0,
      "step": 41285
    },
    {
      "entropy": 5.87143611907959,
      "epoch": 3.608319496635498,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037384693172952047,
      "loss": 5.3961,
      "mean_token_accuracy": 0.19100710302591323,
      "num_tokens": 83663233.0,
      "step": 41290
    },
    {
      "entropy": 5.9552816390991214,
      "epoch": 3.6087564449882024,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003738189378079278,
      "loss": 5.4705,
      "mean_token_accuracy": 0.19362252205610275,
      "num_tokens": 83672899.0,
      "step": 41295
    },
    {
      "entropy": 5.98354549407959,
      "epoch": 3.609193393340907,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037379094199093603,
      "loss": 5.5236,
      "mean_token_accuracy": 0.19043870121240616,
      "num_tokens": 83683542.0,
      "step": 41300
    },
    {
      "entropy": 5.985205364227295,
      "epoch": 3.6096303416936117,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037376294427908237,
      "loss": 5.5329,
      "mean_token_accuracy": 0.18794909864664078,
      "num_tokens": 83693806.0,
      "step": 41305
    },
    {
      "entropy": 5.920180606842041,
      "epoch": 3.6100672900463167,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003737349446729036,
      "loss": 5.3981,
      "mean_token_accuracy": 0.19541337639093398,
      "num_tokens": 83704293.0,
      "step": 41310
    },
    {
      "entropy": 5.903774356842041,
      "epoch": 3.610504238399021,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003737069431729369,
      "loss": 5.403,
      "mean_token_accuracy": 0.19861648231744766,
      "num_tokens": 83714321.0,
      "step": 41315
    },
    {
      "entropy": 5.861458778381348,
      "epoch": 3.610941186751726,
      "grad_norm": 0.984375,
      "learning_rate": 0.00037367893977971933,
      "loss": 5.3771,
      "mean_token_accuracy": 0.1996009513735771,
      "num_tokens": 83723815.0,
      "step": 41320
    },
    {
      "entropy": 5.923236608505249,
      "epoch": 3.6113781351044305,
      "grad_norm": 1.0,
      "learning_rate": 0.00037365093449378796,
      "loss": 5.5033,
      "mean_token_accuracy": 0.18932937681674958,
      "num_tokens": 83734036.0,
      "step": 41325
    },
    {
      "entropy": 6.0050877094268795,
      "epoch": 3.6118150834571354,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037362292731567997,
      "loss": 5.474,
      "mean_token_accuracy": 0.18354139626026153,
      "num_tokens": 83744343.0,
      "step": 41330
    },
    {
      "entropy": 6.044431114196778,
      "epoch": 3.61225203180984,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003735949182459325,
      "loss": 5.5312,
      "mean_token_accuracy": 0.18349913954734803,
      "num_tokens": 83754898.0,
      "step": 41335
    },
    {
      "entropy": 5.9997721195220945,
      "epoch": 3.612688980162545,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003735669072850829,
      "loss": 5.4926,
      "mean_token_accuracy": 0.1817541927099228,
      "num_tokens": 83765964.0,
      "step": 41340
    },
    {
      "entropy": 5.8920708179473875,
      "epoch": 3.6131259285152497,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037353889443366843,
      "loss": 5.4105,
      "mean_token_accuracy": 0.18965615928173066,
      "num_tokens": 83775920.0,
      "step": 41345
    },
    {
      "entropy": 5.902997827529907,
      "epoch": 3.613562876867954,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037351087969222626,
      "loss": 5.3744,
      "mean_token_accuracy": 0.19906380027532578,
      "num_tokens": 83784371.0,
      "step": 41350
    },
    {
      "entropy": 5.886451625823975,
      "epoch": 3.6139998252206587,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003734828630612938,
      "loss": 5.4688,
      "mean_token_accuracy": 0.18839700967073442,
      "num_tokens": 83794618.0,
      "step": 41355
    },
    {
      "entropy": 5.912632179260254,
      "epoch": 3.6144367735733636,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003734548445414084,
      "loss": 5.4894,
      "mean_token_accuracy": 0.19167644530534744,
      "num_tokens": 83805070.0,
      "step": 41360
    },
    {
      "entropy": 5.950252294540405,
      "epoch": 3.6148737219260685,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037342682413310746,
      "loss": 5.5392,
      "mean_token_accuracy": 0.18360855281352997,
      "num_tokens": 83816257.0,
      "step": 41365
    },
    {
      "entropy": 5.944398355484009,
      "epoch": 3.615310670278773,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037339880183692855,
      "loss": 5.4849,
      "mean_token_accuracy": 0.18411986529827118,
      "num_tokens": 83826432.0,
      "step": 41370
    },
    {
      "entropy": 5.997402000427246,
      "epoch": 3.615747618631478,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00037337077765340905,
      "loss": 5.4933,
      "mean_token_accuracy": 0.18906636536121368,
      "num_tokens": 83836746.0,
      "step": 41375
    },
    {
      "entropy": 5.840465307235718,
      "epoch": 3.6161845669841823,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037334275158308643,
      "loss": 5.4128,
      "mean_token_accuracy": 0.19255604296922685,
      "num_tokens": 83845783.0,
      "step": 41380
    },
    {
      "entropy": 5.892651319503784,
      "epoch": 3.6166215153368872,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003733147236264983,
      "loss": 5.4437,
      "mean_token_accuracy": 0.18940605819225312,
      "num_tokens": 83856981.0,
      "step": 41385
    },
    {
      "entropy": 5.9856078147888185,
      "epoch": 3.6170584636895917,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037328669378418236,
      "loss": 5.5112,
      "mean_token_accuracy": 0.1835736870765686,
      "num_tokens": 83867873.0,
      "step": 41390
    },
    {
      "entropy": 5.869127988815308,
      "epoch": 3.6174954120422966,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003732586620566761,
      "loss": 5.4379,
      "mean_token_accuracy": 0.18606919795274734,
      "num_tokens": 83877946.0,
      "step": 41395
    },
    {
      "entropy": 5.882523727416992,
      "epoch": 3.6179323603950015,
      "grad_norm": 1.0,
      "learning_rate": 0.00037323062844451713,
      "loss": 5.4458,
      "mean_token_accuracy": 0.19028151780366898,
      "num_tokens": 83888318.0,
      "step": 41400
    },
    {
      "entropy": 5.881806087493897,
      "epoch": 3.618369308747706,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003732025929482433,
      "loss": 5.4149,
      "mean_token_accuracy": 0.19352113008499144,
      "num_tokens": 83899092.0,
      "step": 41405
    },
    {
      "entropy": 5.8758608341217045,
      "epoch": 3.6188062571004105,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003731745555683922,
      "loss": 5.4219,
      "mean_token_accuracy": 0.1988394945859909,
      "num_tokens": 83908257.0,
      "step": 41410
    },
    {
      "entropy": 5.9207604885101315,
      "epoch": 3.6192432054531154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037314651630550186,
      "loss": 5.4911,
      "mean_token_accuracy": 0.1882939875125885,
      "num_tokens": 83918821.0,
      "step": 41415
    },
    {
      "entropy": 5.899106121063232,
      "epoch": 3.6196801538058203,
      "grad_norm": 1.0,
      "learning_rate": 0.00037311847516010975,
      "loss": 5.4006,
      "mean_token_accuracy": 0.1989538073539734,
      "num_tokens": 83928122.0,
      "step": 41420
    },
    {
      "entropy": 5.957376003265381,
      "epoch": 3.620117102158525,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003730904321327539,
      "loss": 5.5329,
      "mean_token_accuracy": 0.18695502430200578,
      "num_tokens": 83939904.0,
      "step": 41425
    },
    {
      "entropy": 5.945994806289673,
      "epoch": 3.6205540505112297,
      "grad_norm": 1.0,
      "learning_rate": 0.0003730623872239722,
      "loss": 5.4442,
      "mean_token_accuracy": 0.19105636030435563,
      "num_tokens": 83950631.0,
      "step": 41430
    },
    {
      "entropy": 5.96798300743103,
      "epoch": 3.620990998863934,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003730343404343025,
      "loss": 5.5254,
      "mean_token_accuracy": 0.18972617089748384,
      "num_tokens": 83960860.0,
      "step": 41435
    },
    {
      "entropy": 6.005668973922729,
      "epoch": 3.621427947216639,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003730062917642828,
      "loss": 5.5316,
      "mean_token_accuracy": 0.1814400389790535,
      "num_tokens": 83971182.0,
      "step": 41440
    },
    {
      "entropy": 5.8473522663116455,
      "epoch": 3.6218648955693435,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003729782412144511,
      "loss": 5.3824,
      "mean_token_accuracy": 0.19535730332136153,
      "num_tokens": 83981576.0,
      "step": 41445
    },
    {
      "entropy": 5.9085280418396,
      "epoch": 3.6223018439220485,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003729501887853453,
      "loss": 5.4546,
      "mean_token_accuracy": 0.1891619086265564,
      "num_tokens": 83991526.0,
      "step": 41450
    },
    {
      "entropy": 5.951790380477905,
      "epoch": 3.6227387922747534,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003729221344775036,
      "loss": 5.5116,
      "mean_token_accuracy": 0.1856861487030983,
      "num_tokens": 84001979.0,
      "step": 41455
    },
    {
      "entropy": 5.842079114913941,
      "epoch": 3.623175740627458,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003728940782914641,
      "loss": 5.3598,
      "mean_token_accuracy": 0.19658925533294677,
      "num_tokens": 84011977.0,
      "step": 41460
    },
    {
      "entropy": 5.951859045028686,
      "epoch": 3.6236126889801623,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00037286602022776485,
      "loss": 5.4798,
      "mean_token_accuracy": 0.18848826438188554,
      "num_tokens": 84021826.0,
      "step": 41465
    },
    {
      "entropy": 5.934786367416382,
      "epoch": 3.6240496373328672,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037283796028694403,
      "loss": 5.395,
      "mean_token_accuracy": 0.19940334558486938,
      "num_tokens": 84032180.0,
      "step": 41470
    },
    {
      "entropy": 5.937854337692261,
      "epoch": 3.624486585685572,
      "grad_norm": 1.375,
      "learning_rate": 0.00037280989846953986,
      "loss": 5.5521,
      "mean_token_accuracy": 0.18956880867481232,
      "num_tokens": 84043698.0,
      "step": 41475
    },
    {
      "entropy": 5.881349277496338,
      "epoch": 3.6249235340382766,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003727818347760906,
      "loss": 5.382,
      "mean_token_accuracy": 0.19405100494623184,
      "num_tokens": 84053323.0,
      "step": 41480
    },
    {
      "entropy": 6.01643648147583,
      "epoch": 3.6253604823909815,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037275376920713454,
      "loss": 5.6119,
      "mean_token_accuracy": 0.18215524405241013,
      "num_tokens": 84064092.0,
      "step": 41485
    },
    {
      "entropy": 5.95266809463501,
      "epoch": 3.625797430743686,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037272570176320995,
      "loss": 5.4569,
      "mean_token_accuracy": 0.19142162650823594,
      "num_tokens": 84073661.0,
      "step": 41490
    },
    {
      "entropy": 5.866907835006714,
      "epoch": 3.626234379096391,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003726976324448552,
      "loss": 5.4901,
      "mean_token_accuracy": 0.1938299611210823,
      "num_tokens": 84083199.0,
      "step": 41495
    },
    {
      "entropy": 5.921707391738892,
      "epoch": 3.6266713274490954,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003726695612526086,
      "loss": 5.4815,
      "mean_token_accuracy": 0.19076877534389497,
      "num_tokens": 84093600.0,
      "step": 41500
    },
    {
      "entropy": 6.000950002670288,
      "epoch": 3.6271082758018003,
      "grad_norm": 1.125,
      "learning_rate": 0.0003726414881870087,
      "loss": 5.4948,
      "mean_token_accuracy": 0.1828370660543442,
      "num_tokens": 84103147.0,
      "step": 41505
    },
    {
      "entropy": 5.822010612487793,
      "epoch": 3.627545224154505,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037261341324859386,
      "loss": 5.3905,
      "mean_token_accuracy": 0.195612595975399,
      "num_tokens": 84113719.0,
      "step": 41510
    },
    {
      "entropy": 5.918060207366944,
      "epoch": 3.6279821725072097,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037258533643790264,
      "loss": 5.4031,
      "mean_token_accuracy": 0.18751690536737442,
      "num_tokens": 84123563.0,
      "step": 41515
    },
    {
      "entropy": 5.930522537231445,
      "epoch": 3.628419120859914,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00037255725775547357,
      "loss": 5.496,
      "mean_token_accuracy": 0.18826301395893097,
      "num_tokens": 84134753.0,
      "step": 41520
    },
    {
      "entropy": 5.952950191497803,
      "epoch": 3.628856069212619,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037252917720184507,
      "loss": 5.4774,
      "mean_token_accuracy": 0.18203016519546508,
      "num_tokens": 84146549.0,
      "step": 41525
    },
    {
      "entropy": 6.005272006988525,
      "epoch": 3.629293017565324,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003725010947775559,
      "loss": 5.5521,
      "mean_token_accuracy": 0.19079733341932298,
      "num_tokens": 84157069.0,
      "step": 41530
    },
    {
      "entropy": 5.86670413017273,
      "epoch": 3.6297299659180284,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037247301048314464,
      "loss": 5.4707,
      "mean_token_accuracy": 0.19242150336503983,
      "num_tokens": 84167516.0,
      "step": 41535
    },
    {
      "entropy": 5.941293811798095,
      "epoch": 3.6301669142707333,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037244492431914996,
      "loss": 5.4882,
      "mean_token_accuracy": 0.1907168760895729,
      "num_tokens": 84178539.0,
      "step": 41540
    },
    {
      "entropy": 6.020019245147705,
      "epoch": 3.630603862623438,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00037241683628611055,
      "loss": 5.4624,
      "mean_token_accuracy": 0.1945735141634941,
      "num_tokens": 84188605.0,
      "step": 41545
    },
    {
      "entropy": 5.886772537231446,
      "epoch": 3.6310408109761427,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037238874638456524,
      "loss": 5.4126,
      "mean_token_accuracy": 0.19243836998939515,
      "num_tokens": 84197478.0,
      "step": 41550
    },
    {
      "entropy": 5.8983986377716064,
      "epoch": 3.631477759328847,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003723606546150527,
      "loss": 5.4848,
      "mean_token_accuracy": 0.18764981478452683,
      "num_tokens": 84207455.0,
      "step": 41555
    },
    {
      "entropy": 5.904276514053345,
      "epoch": 3.631914707681552,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003723325609781118,
      "loss": 5.3533,
      "mean_token_accuracy": 0.20268296748399733,
      "num_tokens": 84217336.0,
      "step": 41560
    },
    {
      "entropy": 5.897464370727539,
      "epoch": 3.632351656034257,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037230446547428136,
      "loss": 5.4258,
      "mean_token_accuracy": 0.19280888140201569,
      "num_tokens": 84227474.0,
      "step": 41565
    },
    {
      "entropy": 5.901781797409058,
      "epoch": 3.6327886043869615,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003722763681041003,
      "loss": 5.4981,
      "mean_token_accuracy": 0.18925123214721679,
      "num_tokens": 84237382.0,
      "step": 41570
    },
    {
      "entropy": 5.993843841552734,
      "epoch": 3.633225552739666,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003722482688681075,
      "loss": 5.4712,
      "mean_token_accuracy": 0.18432635366916655,
      "num_tokens": 84247816.0,
      "step": 41575
    },
    {
      "entropy": 5.931415939331055,
      "epoch": 3.633662501092371,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000372220167766842,
      "loss": 5.4333,
      "mean_token_accuracy": 0.1885937735438347,
      "num_tokens": 84257056.0,
      "step": 41580
    },
    {
      "entropy": 5.959470462799072,
      "epoch": 3.634099449445076,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037219206480084273,
      "loss": 5.585,
      "mean_token_accuracy": 0.18412486761808394,
      "num_tokens": 84267499.0,
      "step": 41585
    },
    {
      "entropy": 5.944855260848999,
      "epoch": 3.6345363977977803,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003721639599706487,
      "loss": 5.4154,
      "mean_token_accuracy": 0.19018856287002564,
      "num_tokens": 84277548.0,
      "step": 41590
    },
    {
      "entropy": 5.912780094146728,
      "epoch": 3.634973346150485,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037213585327679904,
      "loss": 5.4188,
      "mean_token_accuracy": 0.18770815432071686,
      "num_tokens": 84287461.0,
      "step": 41595
    },
    {
      "entropy": 5.958068323135376,
      "epoch": 3.6354102945031896,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00037210774471983283,
      "loss": 5.4187,
      "mean_token_accuracy": 0.19268838167190552,
      "num_tokens": 84297878.0,
      "step": 41600
    },
    {
      "entropy": 5.858879661560058,
      "epoch": 3.6358472428558946,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037207963430028915,
      "loss": 5.3764,
      "mean_token_accuracy": 0.19897860437631607,
      "num_tokens": 84307850.0,
      "step": 41605
    },
    {
      "entropy": 5.904374790191651,
      "epoch": 3.636284191208599,
      "grad_norm": 1.0,
      "learning_rate": 0.0003720515220187072,
      "loss": 5.5024,
      "mean_token_accuracy": 0.18910576552152633,
      "num_tokens": 84318111.0,
      "step": 41610
    },
    {
      "entropy": 5.851055002212524,
      "epoch": 3.636721139561304,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003720234078756262,
      "loss": 5.3197,
      "mean_token_accuracy": 0.20448584854602814,
      "num_tokens": 84327118.0,
      "step": 41615
    },
    {
      "entropy": 5.949374961853027,
      "epoch": 3.6371580879140084,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003719952918715854,
      "loss": 5.4964,
      "mean_token_accuracy": 0.18217568546533586,
      "num_tokens": 84338428.0,
      "step": 41620
    },
    {
      "entropy": 5.990686988830566,
      "epoch": 3.6375950362667133,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037196717400712415,
      "loss": 5.5447,
      "mean_token_accuracy": 0.18294394314289092,
      "num_tokens": 84349117.0,
      "step": 41625
    },
    {
      "entropy": 5.945403242111206,
      "epoch": 3.638031984619418,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003719390542827816,
      "loss": 5.5105,
      "mean_token_accuracy": 0.19130094796419145,
      "num_tokens": 84358807.0,
      "step": 41630
    },
    {
      "entropy": 5.860381174087524,
      "epoch": 3.6384689329721227,
      "grad_norm": 1.125,
      "learning_rate": 0.0003719109326990973,
      "loss": 5.4751,
      "mean_token_accuracy": 0.19419276863336563,
      "num_tokens": 84368970.0,
      "step": 41635
    },
    {
      "entropy": 5.923798751831055,
      "epoch": 3.6389058813248276,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037188280925661043,
      "loss": 5.3772,
      "mean_token_accuracy": 0.1980660781264305,
      "num_tokens": 84378475.0,
      "step": 41640
    },
    {
      "entropy": 5.893909692764282,
      "epoch": 3.639342829677532,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003718546839558606,
      "loss": 5.3974,
      "mean_token_accuracy": 0.19552736282348632,
      "num_tokens": 84387832.0,
      "step": 41645
    },
    {
      "entropy": 5.884138774871826,
      "epoch": 3.6397797780302366,
      "grad_norm": 0.953125,
      "learning_rate": 0.00037182655679738713,
      "loss": 5.393,
      "mean_token_accuracy": 0.19220607429742814,
      "num_tokens": 84398316.0,
      "step": 41650
    },
    {
      "entropy": 5.911664819717407,
      "epoch": 3.6402167263829415,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003717984277817296,
      "loss": 5.3839,
      "mean_token_accuracy": 0.19041362255811692,
      "num_tokens": 84408624.0,
      "step": 41655
    },
    {
      "entropy": 5.9454771995544435,
      "epoch": 3.6406536747356464,
      "grad_norm": 0.96875,
      "learning_rate": 0.00037177029690942744,
      "loss": 5.3761,
      "mean_token_accuracy": 0.19908568412065505,
      "num_tokens": 84417422.0,
      "step": 41660
    },
    {
      "entropy": 5.928072118759156,
      "epoch": 3.641090623088351,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003717421641810202,
      "loss": 5.5025,
      "mean_token_accuracy": 0.18428249210119246,
      "num_tokens": 84427857.0,
      "step": 41665
    },
    {
      "entropy": 5.9882227897644045,
      "epoch": 3.6415275714410558,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00037171402959704767,
      "loss": 5.4933,
      "mean_token_accuracy": 0.17820259779691697,
      "num_tokens": 84438468.0,
      "step": 41670
    },
    {
      "entropy": 5.891583633422852,
      "epoch": 3.6419645197937602,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037168589315804944,
      "loss": 5.4808,
      "mean_token_accuracy": 0.19083086848258973,
      "num_tokens": 84449371.0,
      "step": 41675
    },
    {
      "entropy": 5.9215624809265135,
      "epoch": 3.642401468146465,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003716577548645651,
      "loss": 5.4127,
      "mean_token_accuracy": 0.1911706581711769,
      "num_tokens": 84458569.0,
      "step": 41680
    },
    {
      "entropy": 5.89416995048523,
      "epoch": 3.6428384164991696,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003716296147171343,
      "loss": 5.5274,
      "mean_token_accuracy": 0.19328239411115647,
      "num_tokens": 84469244.0,
      "step": 41685
    },
    {
      "entropy": 5.866381502151489,
      "epoch": 3.6432753648518745,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003716014727162969,
      "loss": 5.4073,
      "mean_token_accuracy": 0.20789740234613419,
      "num_tokens": 84478902.0,
      "step": 41690
    },
    {
      "entropy": 5.859259748458863,
      "epoch": 3.6437123132045794,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037157332886259253,
      "loss": 5.4035,
      "mean_token_accuracy": 0.1918851152062416,
      "num_tokens": 84489166.0,
      "step": 41695
    },
    {
      "entropy": 5.924052572250366,
      "epoch": 3.644149261557284,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003715451831565613,
      "loss": 5.4021,
      "mean_token_accuracy": 0.1943131923675537,
      "num_tokens": 84499093.0,
      "step": 41700
    },
    {
      "entropy": 5.918253660202026,
      "epoch": 3.6445862099099884,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037151703559874273,
      "loss": 5.4081,
      "mean_token_accuracy": 0.19291305541992188,
      "num_tokens": 84508898.0,
      "step": 41705
    },
    {
      "entropy": 5.976589822769165,
      "epoch": 3.6450231582626933,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037148888618967677,
      "loss": 5.5215,
      "mean_token_accuracy": 0.18440228551626206,
      "num_tokens": 84518277.0,
      "step": 41710
    },
    {
      "entropy": 5.955251932144165,
      "epoch": 3.645460106615398,
      "grad_norm": 1.0,
      "learning_rate": 0.0003714607349299035,
      "loss": 5.4773,
      "mean_token_accuracy": 0.18420992493629457,
      "num_tokens": 84527882.0,
      "step": 41715
    },
    {
      "entropy": 5.908655452728271,
      "epoch": 3.6458970549681027,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037143258181996286,
      "loss": 5.4505,
      "mean_token_accuracy": 0.19474500566720962,
      "num_tokens": 84538065.0,
      "step": 41720
    },
    {
      "entropy": 5.913551759719849,
      "epoch": 3.6463340033208076,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003714044268603947,
      "loss": 5.4607,
      "mean_token_accuracy": 0.18937838226556777,
      "num_tokens": 84547923.0,
      "step": 41725
    },
    {
      "entropy": 5.917511796951294,
      "epoch": 3.646770951673512,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003713762700517392,
      "loss": 5.467,
      "mean_token_accuracy": 0.18840005099773408,
      "num_tokens": 84558725.0,
      "step": 41730
    },
    {
      "entropy": 5.947612142562866,
      "epoch": 3.647207900026217,
      "grad_norm": 1.25,
      "learning_rate": 0.00037134811139453625,
      "loss": 5.4477,
      "mean_token_accuracy": 0.1976452186703682,
      "num_tokens": 84568945.0,
      "step": 41735
    },
    {
      "entropy": 5.958585739135742,
      "epoch": 3.6476448483789214,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000371319950889326,
      "loss": 5.5199,
      "mean_token_accuracy": 0.1876250460743904,
      "num_tokens": 84579230.0,
      "step": 41740
    },
    {
      "entropy": 5.99686861038208,
      "epoch": 3.6480817967316264,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003712917885366487,
      "loss": 5.4705,
      "mean_token_accuracy": 0.19220881760120392,
      "num_tokens": 84588966.0,
      "step": 41745
    },
    {
      "entropy": 5.89012975692749,
      "epoch": 3.6485187450843313,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003712636243370444,
      "loss": 5.3968,
      "mean_token_accuracy": 0.1970161259174347,
      "num_tokens": 84598631.0,
      "step": 41750
    },
    {
      "entropy": 5.929517126083374,
      "epoch": 3.6489556934370357,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037123545829105336,
      "loss": 5.4445,
      "mean_token_accuracy": 0.18724633604288102,
      "num_tokens": 84609849.0,
      "step": 41755
    },
    {
      "entropy": 5.920911455154419,
      "epoch": 3.64939264178974,
      "grad_norm": 1.59375,
      "learning_rate": 0.0003712072903992159,
      "loss": 5.4592,
      "mean_token_accuracy": 0.1896432727575302,
      "num_tokens": 84619959.0,
      "step": 41760
    },
    {
      "entropy": 5.974271678924561,
      "epoch": 3.649829590142445,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003711791206620721,
      "loss": 5.4281,
      "mean_token_accuracy": 0.1903488516807556,
      "num_tokens": 84630224.0,
      "step": 41765
    },
    {
      "entropy": 5.931581592559814,
      "epoch": 3.65026653849515,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003711509490801624,
      "loss": 5.4722,
      "mean_token_accuracy": 0.18778379261493683,
      "num_tokens": 84641595.0,
      "step": 41770
    },
    {
      "entropy": 5.845051956176758,
      "epoch": 3.6507034868478545,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000371122775654027,
      "loss": 5.4089,
      "mean_token_accuracy": 0.1995820477604866,
      "num_tokens": 84652406.0,
      "step": 41775
    },
    {
      "entropy": 5.857023525238037,
      "epoch": 3.6511404352005594,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003710946003842065,
      "loss": 5.3489,
      "mean_token_accuracy": 0.2027664452791214,
      "num_tokens": 84662319.0,
      "step": 41780
    },
    {
      "entropy": 5.922950792312622,
      "epoch": 3.651577383553264,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003710664232712412,
      "loss": 5.4395,
      "mean_token_accuracy": 0.1886270672082901,
      "num_tokens": 84673086.0,
      "step": 41785
    },
    {
      "entropy": 5.926275110244751,
      "epoch": 3.652014331905969,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003710382443156716,
      "loss": 5.5017,
      "mean_token_accuracy": 0.18800483644008636,
      "num_tokens": 84684536.0,
      "step": 41790
    },
    {
      "entropy": 5.926619958877564,
      "epoch": 3.6524512802586733,
      "grad_norm": 3.234375,
      "learning_rate": 0.0003710100635180381,
      "loss": 5.4578,
      "mean_token_accuracy": 0.19131042063236237,
      "num_tokens": 84694975.0,
      "step": 41795
    },
    {
      "entropy": 5.867151975631714,
      "epoch": 3.652888228611378,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003709818808788813,
      "loss": 5.4312,
      "mean_token_accuracy": 0.1886235237121582,
      "num_tokens": 84704290.0,
      "step": 41800
    },
    {
      "entropy": 6.00506625175476,
      "epoch": 3.653325176964083,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003709536963987417,
      "loss": 5.4431,
      "mean_token_accuracy": 0.1880548283457756,
      "num_tokens": 84714280.0,
      "step": 41805
    },
    {
      "entropy": 6.004946756362915,
      "epoch": 3.6537621253167876,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037092551007815997,
      "loss": 5.4878,
      "mean_token_accuracy": 0.19028424769639968,
      "num_tokens": 84725127.0,
      "step": 41810
    },
    {
      "entropy": 5.925186014175415,
      "epoch": 3.654199073669492,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003708973219176767,
      "loss": 5.4838,
      "mean_token_accuracy": 0.18538293987512589,
      "num_tokens": 84736385.0,
      "step": 41815
    },
    {
      "entropy": 5.83893928527832,
      "epoch": 3.654636022022197,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00037086913191783247,
      "loss": 5.4092,
      "mean_token_accuracy": 0.20295058637857438,
      "num_tokens": 84746489.0,
      "step": 41820
    },
    {
      "entropy": 5.91980767250061,
      "epoch": 3.655072970374902,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003708409400791681,
      "loss": 5.4576,
      "mean_token_accuracy": 0.1889411225914955,
      "num_tokens": 84756743.0,
      "step": 41825
    },
    {
      "entropy": 5.93027629852295,
      "epoch": 3.6555099187276063,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037081274640222425,
      "loss": 5.445,
      "mean_token_accuracy": 0.1924266740679741,
      "num_tokens": 84767647.0,
      "step": 41830
    },
    {
      "entropy": 5.979290866851807,
      "epoch": 3.6559468670803112,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003707845508875417,
      "loss": 5.5045,
      "mean_token_accuracy": 0.18424327820539474,
      "num_tokens": 84778122.0,
      "step": 41835
    },
    {
      "entropy": 5.952229595184326,
      "epoch": 3.6563838154330157,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003707563535356614,
      "loss": 5.4856,
      "mean_token_accuracy": 0.1963601142168045,
      "num_tokens": 84787300.0,
      "step": 41840
    },
    {
      "entropy": 5.935810232162476,
      "epoch": 3.6568207637857206,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003707281543471239,
      "loss": 5.4593,
      "mean_token_accuracy": 0.18379078209400176,
      "num_tokens": 84798271.0,
      "step": 41845
    },
    {
      "entropy": 5.974690723419189,
      "epoch": 3.657257712138425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003706999533224704,
      "loss": 5.5402,
      "mean_token_accuracy": 0.1843202605843544,
      "num_tokens": 84810377.0,
      "step": 41850
    },
    {
      "entropy": 6.008256816864014,
      "epoch": 3.65769466049113,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003706717504622415,
      "loss": 5.4243,
      "mean_token_accuracy": 0.18911335319280625,
      "num_tokens": 84821319.0,
      "step": 41855
    },
    {
      "entropy": 5.979615831375122,
      "epoch": 3.658131608843835,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003706435457669784,
      "loss": 5.4614,
      "mean_token_accuracy": 0.18784730434417723,
      "num_tokens": 84831558.0,
      "step": 41860
    },
    {
      "entropy": 5.92698483467102,
      "epoch": 3.6585685571965394,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037061533923722195,
      "loss": 5.431,
      "mean_token_accuracy": 0.19208526164293288,
      "num_tokens": 84840983.0,
      "step": 41865
    },
    {
      "entropy": 5.822942638397217,
      "epoch": 3.659005505549244,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00037058713087351317,
      "loss": 5.3725,
      "mean_token_accuracy": 0.20096295773983003,
      "num_tokens": 84851304.0,
      "step": 41870
    },
    {
      "entropy": 5.855408191680908,
      "epoch": 3.6594424539019488,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037055892067639314,
      "loss": 5.4324,
      "mean_token_accuracy": 0.19096617102622987,
      "num_tokens": 84862849.0,
      "step": 41875
    },
    {
      "entropy": 5.930584144592285,
      "epoch": 3.6598794022546537,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037053070864640294,
      "loss": 5.4155,
      "mean_token_accuracy": 0.19786760509014129,
      "num_tokens": 84873591.0,
      "step": 41880
    },
    {
      "entropy": 5.903568506240845,
      "epoch": 3.660316350607358,
      "grad_norm": 1.0,
      "learning_rate": 0.00037050249478408364,
      "loss": 5.4145,
      "mean_token_accuracy": 0.1988359808921814,
      "num_tokens": 84883701.0,
      "step": 41885
    },
    {
      "entropy": 5.906601858139038,
      "epoch": 3.660753298960063,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003704742790899765,
      "loss": 5.4401,
      "mean_token_accuracy": 0.18935108184814453,
      "num_tokens": 84893842.0,
      "step": 41890
    },
    {
      "entropy": 5.970252180099488,
      "epoch": 3.6611902473127675,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003704460615646227,
      "loss": 5.518,
      "mean_token_accuracy": 0.1827940672636032,
      "num_tokens": 84903797.0,
      "step": 41895
    },
    {
      "entropy": 5.992389726638794,
      "epoch": 3.6616271956654725,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037041784220856336,
      "loss": 5.5532,
      "mean_token_accuracy": 0.18797175139188765,
      "num_tokens": 84915459.0,
      "step": 41900
    },
    {
      "entropy": 5.952640628814697,
      "epoch": 3.662064144018177,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003703896210223399,
      "loss": 5.4818,
      "mean_token_accuracy": 0.19145339876413345,
      "num_tokens": 84926067.0,
      "step": 41905
    },
    {
      "entropy": 5.967639636993408,
      "epoch": 3.662501092370882,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037036139800649336,
      "loss": 5.4549,
      "mean_token_accuracy": 0.19673602133989335,
      "num_tokens": 84936612.0,
      "step": 41910
    },
    {
      "entropy": 5.819804430007935,
      "epoch": 3.6629380407235863,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003703331731615654,
      "loss": 5.3718,
      "mean_token_accuracy": 0.2018071308732033,
      "num_tokens": 84946620.0,
      "step": 41915
    },
    {
      "entropy": 5.924941873550415,
      "epoch": 3.663374989076291,
      "grad_norm": 1.125,
      "learning_rate": 0.00037030494648809716,
      "loss": 5.413,
      "mean_token_accuracy": 0.19372545629739762,
      "num_tokens": 84955890.0,
      "step": 41920
    },
    {
      "entropy": 5.8054906845092775,
      "epoch": 3.6638119374289957,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003702767179866301,
      "loss": 5.371,
      "mean_token_accuracy": 0.19689476788043975,
      "num_tokens": 84964694.0,
      "step": 41925
    },
    {
      "entropy": 5.919905328750611,
      "epoch": 3.6642488857817006,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003702484876577057,
      "loss": 5.4235,
      "mean_token_accuracy": 0.19658099114894867,
      "num_tokens": 84974203.0,
      "step": 41930
    },
    {
      "entropy": 5.869400215148926,
      "epoch": 3.6646858341344055,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037022025550186537,
      "loss": 5.3784,
      "mean_token_accuracy": 0.19840268343687056,
      "num_tokens": 84984326.0,
      "step": 41935
    },
    {
      "entropy": 5.874671077728271,
      "epoch": 3.66512278248711,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003701920215196506,
      "loss": 5.3932,
      "mean_token_accuracy": 0.1906021624803543,
      "num_tokens": 84993284.0,
      "step": 41940
    },
    {
      "entropy": 5.903240823745728,
      "epoch": 3.665559730839815,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003701637857116031,
      "loss": 5.4105,
      "mean_token_accuracy": 0.19280242621898652,
      "num_tokens": 85003239.0,
      "step": 41945
    },
    {
      "entropy": 5.893546199798584,
      "epoch": 3.6659966791925194,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003701355480782642,
      "loss": 5.4526,
      "mean_token_accuracy": 0.19236384630203246,
      "num_tokens": 85015493.0,
      "step": 41950
    },
    {
      "entropy": 5.9068550109863285,
      "epoch": 3.6664336275452243,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00037010730862017575,
      "loss": 5.4142,
      "mean_token_accuracy": 0.19293072819709778,
      "num_tokens": 85025677.0,
      "step": 41955
    },
    {
      "entropy": 5.943179368972778,
      "epoch": 3.6668705758979288,
      "grad_norm": 1.0,
      "learning_rate": 0.00037007906733787913,
      "loss": 5.4299,
      "mean_token_accuracy": 0.1949151188135147,
      "num_tokens": 85034908.0,
      "step": 41960
    },
    {
      "entropy": 5.907729148864746,
      "epoch": 3.6673075242506337,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003700508242319163,
      "loss": 5.4732,
      "mean_token_accuracy": 0.18238222897052764,
      "num_tokens": 85045416.0,
      "step": 41965
    },
    {
      "entropy": 5.972193765640259,
      "epoch": 3.667744472603338,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037002257930282887,
      "loss": 5.5393,
      "mean_token_accuracy": 0.18461430370807647,
      "num_tokens": 85056259.0,
      "step": 41970
    },
    {
      "entropy": 5.93310956954956,
      "epoch": 3.668181420956043,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003699943325511585,
      "loss": 5.4341,
      "mean_token_accuracy": 0.19385594129562378,
      "num_tokens": 85065579.0,
      "step": 41975
    },
    {
      "entropy": 5.880173063278198,
      "epoch": 3.6686183693087475,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003699660839774471,
      "loss": 5.4379,
      "mean_token_accuracy": 0.19013077467679979,
      "num_tokens": 85075084.0,
      "step": 41980
    },
    {
      "entropy": 5.895682382583618,
      "epoch": 3.6690553176614524,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003699378335822364,
      "loss": 5.4259,
      "mean_token_accuracy": 0.19826584607362746,
      "num_tokens": 85084536.0,
      "step": 41985
    },
    {
      "entropy": 5.94668664932251,
      "epoch": 3.6694922660141573,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003699095813660683,
      "loss": 5.4266,
      "mean_token_accuracy": 0.1954363241791725,
      "num_tokens": 85094076.0,
      "step": 41990
    },
    {
      "entropy": 5.852444839477539,
      "epoch": 3.669929214366862,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003698813273294848,
      "loss": 5.4016,
      "mean_token_accuracy": 0.19081756621599197,
      "num_tokens": 85103784.0,
      "step": 41995
    },
    {
      "entropy": 5.866221141815186,
      "epoch": 3.6703661627195663,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036985307147302765,
      "loss": 5.4491,
      "mean_token_accuracy": 0.19622035026550294,
      "num_tokens": 85114066.0,
      "step": 42000
    },
    {
      "epoch": 3.6703661627195663,
      "eval_entropy": 5.654896732022805,
      "eval_loss": 5.508319854736328,
      "eval_mean_token_accuracy": 0.19630870594205674,
      "eval_num_tokens": 85114066.0,
      "eval_runtime": 27.8664,
      "eval_samples_per_second": 1328.265,
      "eval_steps_per_second": 166.042,
      "step": 42000
    },
    {
      "entropy": 5.911890316009521,
      "epoch": 3.670803111072271,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000369824813797239,
      "loss": 5.4806,
      "mean_token_accuracy": 0.1864310920238495,
      "num_tokens": 85123881.0,
      "step": 42005
    },
    {
      "entropy": 5.9796747207641605,
      "epoch": 3.671240059424976,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003697965543026606,
      "loss": 5.472,
      "mean_token_accuracy": 0.18657437413930894,
      "num_tokens": 85135159.0,
      "step": 42010
    },
    {
      "entropy": 5.996543455123901,
      "epoch": 3.6716770077776806,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003697682929898347,
      "loss": 5.4896,
      "mean_token_accuracy": 0.18718259036540985,
      "num_tokens": 85144718.0,
      "step": 42015
    },
    {
      "entropy": 5.8118593215942385,
      "epoch": 3.6721139561303855,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036974002985930315,
      "loss": 5.3951,
      "mean_token_accuracy": 0.20125190168619156,
      "num_tokens": 85154094.0,
      "step": 42020
    },
    {
      "entropy": 5.933946561813355,
      "epoch": 3.67255090448309,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003697117649116083,
      "loss": 5.4102,
      "mean_token_accuracy": 0.18801676481962204,
      "num_tokens": 85163636.0,
      "step": 42025
    },
    {
      "entropy": 5.932662010192871,
      "epoch": 3.672987852835795,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003696834981472921,
      "loss": 5.4242,
      "mean_token_accuracy": 0.19269352406263351,
      "num_tokens": 85172977.0,
      "step": 42030
    },
    {
      "entropy": 5.853267240524292,
      "epoch": 3.6734248011884993,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00036965522956689683,
      "loss": 5.4382,
      "mean_token_accuracy": 0.19787138402462007,
      "num_tokens": 85182956.0,
      "step": 42035
    },
    {
      "entropy": 5.9798057079315186,
      "epoch": 3.6738617495412043,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003696269591709647,
      "loss": 5.5132,
      "mean_token_accuracy": 0.18304951936006547,
      "num_tokens": 85193317.0,
      "step": 42040
    },
    {
      "entropy": 5.993457651138305,
      "epoch": 3.674298697893909,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036959868696003785,
      "loss": 5.496,
      "mean_token_accuracy": 0.189279206097126,
      "num_tokens": 85203724.0,
      "step": 42045
    },
    {
      "entropy": 5.907876014709473,
      "epoch": 3.6747356462466136,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036957041293465864,
      "loss": 5.4499,
      "mean_token_accuracy": 0.1902102917432785,
      "num_tokens": 85213409.0,
      "step": 42050
    },
    {
      "entropy": 5.954780960083008,
      "epoch": 3.675172594599318,
      "grad_norm": 1.25,
      "learning_rate": 0.0003695421370953694,
      "loss": 5.4872,
      "mean_token_accuracy": 0.18711258322000504,
      "num_tokens": 85223975.0,
      "step": 42055
    },
    {
      "entropy": 5.931151866912842,
      "epoch": 3.675609542952023,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00036951385944271233,
      "loss": 5.4616,
      "mean_token_accuracy": 0.1926542639732361,
      "num_tokens": 85234522.0,
      "step": 42060
    },
    {
      "entropy": 5.9049842834472654,
      "epoch": 3.676046491304728,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036948557997722994,
      "loss": 5.4482,
      "mean_token_accuracy": 0.18880468159914016,
      "num_tokens": 85243449.0,
      "step": 42065
    },
    {
      "entropy": 5.906725978851318,
      "epoch": 3.6764834396574324,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036945729869946464,
      "loss": 5.4331,
      "mean_token_accuracy": 0.1931474819779396,
      "num_tokens": 85254328.0,
      "step": 42070
    },
    {
      "entropy": 5.931981086730957,
      "epoch": 3.6769203880101373,
      "grad_norm": 1.0,
      "learning_rate": 0.00036942901560995886,
      "loss": 5.4388,
      "mean_token_accuracy": 0.19089249074459075,
      "num_tokens": 85264827.0,
      "step": 42075
    },
    {
      "entropy": 5.918890857696534,
      "epoch": 3.677357336362842,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003694007307092551,
      "loss": 5.3812,
      "mean_token_accuracy": 0.19295211881399155,
      "num_tokens": 85274564.0,
      "step": 42080
    },
    {
      "entropy": 5.8755590438842775,
      "epoch": 3.6777942847155467,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003693724439978958,
      "loss": 5.3649,
      "mean_token_accuracy": 0.2007196307182312,
      "num_tokens": 85284086.0,
      "step": 42085
    },
    {
      "entropy": 5.976910924911499,
      "epoch": 3.678231233068251,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003693441554764236,
      "loss": 5.5227,
      "mean_token_accuracy": 0.18946213871240616,
      "num_tokens": 85294145.0,
      "step": 42090
    },
    {
      "entropy": 5.778555536270142,
      "epoch": 3.678668181420956,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003693158651453811,
      "loss": 5.3173,
      "mean_token_accuracy": 0.21025940477848054,
      "num_tokens": 85304645.0,
      "step": 42095
    },
    {
      "entropy": 5.935131788253784,
      "epoch": 3.679105129773661,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036928757300531083,
      "loss": 5.3471,
      "mean_token_accuracy": 0.1931897297501564,
      "num_tokens": 85313844.0,
      "step": 42100
    },
    {
      "entropy": 5.842485189437866,
      "epoch": 3.6795420781263655,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036925927905675555,
      "loss": 5.4688,
      "mean_token_accuracy": 0.19632472842931747,
      "num_tokens": 85322660.0,
      "step": 42105
    },
    {
      "entropy": 5.981348514556885,
      "epoch": 3.67997902647907,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003692309833002579,
      "loss": 5.497,
      "mean_token_accuracy": 0.18751912266016008,
      "num_tokens": 85334053.0,
      "step": 42110
    },
    {
      "entropy": 6.005953693389893,
      "epoch": 3.680415974831775,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00036920268573636055,
      "loss": 5.3579,
      "mean_token_accuracy": 0.20009447038173675,
      "num_tokens": 85343890.0,
      "step": 42115
    },
    {
      "entropy": 5.816577053070068,
      "epoch": 3.6808529231844798,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036917438636560645,
      "loss": 5.3734,
      "mean_token_accuracy": 0.20261071622371674,
      "num_tokens": 85353325.0,
      "step": 42120
    },
    {
      "entropy": 5.917014455795288,
      "epoch": 3.6812898715371842,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003691460851885382,
      "loss": 5.5349,
      "mean_token_accuracy": 0.18357240706682204,
      "num_tokens": 85363072.0,
      "step": 42125
    },
    {
      "entropy": 5.916262149810791,
      "epoch": 3.681726819889889,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003691177822056987,
      "loss": 5.3607,
      "mean_token_accuracy": 0.19507942199707032,
      "num_tokens": 85372709.0,
      "step": 42130
    },
    {
      "entropy": 5.984115743637085,
      "epoch": 3.6821637682425936,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036908947741763084,
      "loss": 5.4984,
      "mean_token_accuracy": 0.19691523313522338,
      "num_tokens": 85383143.0,
      "step": 42135
    },
    {
      "entropy": 5.866680860519409,
      "epoch": 3.6826007165952985,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003690611708248774,
      "loss": 5.4244,
      "mean_token_accuracy": 0.1970758780837059,
      "num_tokens": 85393245.0,
      "step": 42140
    },
    {
      "entropy": 5.877350664138794,
      "epoch": 3.683037664948003,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036903286242798155,
      "loss": 5.4083,
      "mean_token_accuracy": 0.18888608664274215,
      "num_tokens": 85404031.0,
      "step": 42145
    },
    {
      "entropy": 5.931664896011353,
      "epoch": 3.683474613300708,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0003690045522274861,
      "loss": 5.4939,
      "mean_token_accuracy": 0.1898234724998474,
      "num_tokens": 85414561.0,
      "step": 42150
    },
    {
      "entropy": 5.996125841140747,
      "epoch": 3.683911561653413,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036897624022393403,
      "loss": 5.4979,
      "mean_token_accuracy": 0.18886565864086152,
      "num_tokens": 85424755.0,
      "step": 42155
    },
    {
      "entropy": 5.86908369064331,
      "epoch": 3.6843485100061173,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036894792641786853,
      "loss": 5.4614,
      "mean_token_accuracy": 0.19180241823196412,
      "num_tokens": 85434086.0,
      "step": 42160
    },
    {
      "entropy": 5.853403615951538,
      "epoch": 3.6847854583588218,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036891961080983245,
      "loss": 5.4441,
      "mean_token_accuracy": 0.19145480394363404,
      "num_tokens": 85443540.0,
      "step": 42165
    },
    {
      "entropy": 5.93033881187439,
      "epoch": 3.6852224067115267,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003688912934003691,
      "loss": 5.4307,
      "mean_token_accuracy": 0.20013788789510728,
      "num_tokens": 85452813.0,
      "step": 42170
    },
    {
      "entropy": 5.951689624786377,
      "epoch": 3.6856593550642316,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036886297419002153,
      "loss": 5.4703,
      "mean_token_accuracy": 0.18749553114175796,
      "num_tokens": 85461838.0,
      "step": 42175
    },
    {
      "entropy": 5.974275207519531,
      "epoch": 3.686096303416936,
      "grad_norm": 1.0,
      "learning_rate": 0.0003688346531793329,
      "loss": 5.5057,
      "mean_token_accuracy": 0.1843563050031662,
      "num_tokens": 85472272.0,
      "step": 42180
    },
    {
      "entropy": 5.965197229385376,
      "epoch": 3.686533251769641,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003688063303688465,
      "loss": 5.4387,
      "mean_token_accuracy": 0.192031392455101,
      "num_tokens": 85482663.0,
      "step": 42185
    },
    {
      "entropy": 5.929974889755249,
      "epoch": 3.6869702001223454,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003687780057591055,
      "loss": 5.4491,
      "mean_token_accuracy": 0.1883937031030655,
      "num_tokens": 85492697.0,
      "step": 42190
    },
    {
      "entropy": 5.885043907165527,
      "epoch": 3.6874071484750504,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003687496793506532,
      "loss": 5.3908,
      "mean_token_accuracy": 0.19594082832336426,
      "num_tokens": 85501785.0,
      "step": 42195
    },
    {
      "entropy": 5.881900978088379,
      "epoch": 3.687844096827755,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036872135114403293,
      "loss": 5.4005,
      "mean_token_accuracy": 0.19474317133426666,
      "num_tokens": 85512493.0,
      "step": 42200
    },
    {
      "entropy": 5.93018741607666,
      "epoch": 3.6882810451804597,
      "grad_norm": 1.109375,
      "learning_rate": 0.000368693021139788,
      "loss": 5.4326,
      "mean_token_accuracy": 0.1938090741634369,
      "num_tokens": 85521118.0,
      "step": 42205
    },
    {
      "entropy": 5.98661994934082,
      "epoch": 3.6887179935331647,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003686646893384619,
      "loss": 5.5322,
      "mean_token_accuracy": 0.1847444400191307,
      "num_tokens": 85531048.0,
      "step": 42210
    },
    {
      "entropy": 5.979235506057739,
      "epoch": 3.689154941885869,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000368636355740598,
      "loss": 5.4119,
      "mean_token_accuracy": 0.19163281768560408,
      "num_tokens": 85541526.0,
      "step": 42215
    },
    {
      "entropy": 5.890831851959229,
      "epoch": 3.6895918902385736,
      "grad_norm": 1.0,
      "learning_rate": 0.0003686080203467397,
      "loss": 5.4048,
      "mean_token_accuracy": 0.19334670156240463,
      "num_tokens": 85551588.0,
      "step": 42220
    },
    {
      "entropy": 5.840831232070923,
      "epoch": 3.6900288385912785,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003685796831574305,
      "loss": 5.422,
      "mean_token_accuracy": 0.19821756035089494,
      "num_tokens": 85560775.0,
      "step": 42225
    },
    {
      "entropy": 5.870162487030029,
      "epoch": 3.6904657869439834,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003685513441732139,
      "loss": 5.4301,
      "mean_token_accuracy": 0.18566914051771163,
      "num_tokens": 85571495.0,
      "step": 42230
    },
    {
      "entropy": 5.981692886352539,
      "epoch": 3.690902735296688,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003685230033946336,
      "loss": 5.4451,
      "mean_token_accuracy": 0.1897352531552315,
      "num_tokens": 85583782.0,
      "step": 42235
    },
    {
      "entropy": 5.9583306312561035,
      "epoch": 3.691339683649393,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000368494660822233,
      "loss": 5.5253,
      "mean_token_accuracy": 0.19136299043893815,
      "num_tokens": 85594342.0,
      "step": 42240
    },
    {
      "entropy": 5.948502540588379,
      "epoch": 3.6917766320020973,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003684663164565558,
      "loss": 5.424,
      "mean_token_accuracy": 0.19319071024656295,
      "num_tokens": 85604829.0,
      "step": 42245
    },
    {
      "entropy": 5.922760915756226,
      "epoch": 3.692213580354802,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036843797029814575,
      "loss": 5.3945,
      "mean_token_accuracy": 0.19605109095573425,
      "num_tokens": 85614530.0,
      "step": 42250
    },
    {
      "entropy": 5.928645467758178,
      "epoch": 3.6926505287075067,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003684096223475464,
      "loss": 5.4348,
      "mean_token_accuracy": 0.18955239802598953,
      "num_tokens": 85624110.0,
      "step": 42255
    },
    {
      "entropy": 5.95276517868042,
      "epoch": 3.6930874770602116,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003683812726053017,
      "loss": 5.4661,
      "mean_token_accuracy": 0.19224573224782943,
      "num_tokens": 85634345.0,
      "step": 42260
    },
    {
      "entropy": 5.916877317428589,
      "epoch": 3.693524425412916,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003683529210719551,
      "loss": 5.4138,
      "mean_token_accuracy": 0.19560324102640153,
      "num_tokens": 85644813.0,
      "step": 42265
    },
    {
      "entropy": 5.890698385238648,
      "epoch": 3.693961373765621,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003683245677480506,
      "loss": 5.495,
      "mean_token_accuracy": 0.18704817295074463,
      "num_tokens": 85655848.0,
      "step": 42270
    },
    {
      "entropy": 5.903331327438354,
      "epoch": 3.6943983221183254,
      "grad_norm": 1.015625,
      "learning_rate": 0.000368296212634132,
      "loss": 5.3806,
      "mean_token_accuracy": 0.19502607136964797,
      "num_tokens": 85664672.0,
      "step": 42275
    },
    {
      "entropy": 5.944244718551635,
      "epoch": 3.6948352704710303,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003682678557307432,
      "loss": 5.4899,
      "mean_token_accuracy": 0.1830936402082443,
      "num_tokens": 85674179.0,
      "step": 42280
    },
    {
      "entropy": 5.970790100097656,
      "epoch": 3.6952722188237352,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000368239497038428,
      "loss": 5.4676,
      "mean_token_accuracy": 0.1885622501373291,
      "num_tokens": 85684244.0,
      "step": 42285
    },
    {
      "entropy": 5.971868801116943,
      "epoch": 3.6957091671764397,
      "grad_norm": 1.0,
      "learning_rate": 0.0003682111365577304,
      "loss": 5.5193,
      "mean_token_accuracy": 0.1885785534977913,
      "num_tokens": 85694998.0,
      "step": 42290
    },
    {
      "entropy": 5.8738871097564695,
      "epoch": 3.696146115529144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036818277428919433,
      "loss": 5.468,
      "mean_token_accuracy": 0.18441897481679917,
      "num_tokens": 85705083.0,
      "step": 42295
    },
    {
      "entropy": 5.894248008728027,
      "epoch": 3.696583063881849,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003681544102333638,
      "loss": 5.4496,
      "mean_token_accuracy": 0.19350251108407973,
      "num_tokens": 85714442.0,
      "step": 42300
    },
    {
      "entropy": 5.9606194496154785,
      "epoch": 3.697020012234554,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036812604439078297,
      "loss": 5.3795,
      "mean_token_accuracy": 0.19949280321598054,
      "num_tokens": 85724124.0,
      "step": 42305
    },
    {
      "entropy": 5.956612205505371,
      "epoch": 3.6974569605872585,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036809767676199585,
      "loss": 5.4406,
      "mean_token_accuracy": 0.19412619024515151,
      "num_tokens": 85734271.0,
      "step": 42310
    },
    {
      "entropy": 5.932268714904785,
      "epoch": 3.6978939089399634,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036806930734754634,
      "loss": 5.4741,
      "mean_token_accuracy": 0.19044507741928102,
      "num_tokens": 85744008.0,
      "step": 42315
    },
    {
      "entropy": 5.920311975479126,
      "epoch": 3.698330857292668,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00036804093614797883,
      "loss": 5.4452,
      "mean_token_accuracy": 0.1930151641368866,
      "num_tokens": 85755377.0,
      "step": 42320
    },
    {
      "entropy": 5.93400993347168,
      "epoch": 3.6987678056453728,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003680125631638375,
      "loss": 5.2977,
      "mean_token_accuracy": 0.2021140217781067,
      "num_tokens": 85764640.0,
      "step": 42325
    },
    {
      "entropy": 5.903770303726196,
      "epoch": 3.6992047539980772,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036798418839566633,
      "loss": 5.3981,
      "mean_token_accuracy": 0.20418150424957277,
      "num_tokens": 85774282.0,
      "step": 42330
    },
    {
      "entropy": 5.855151033401489,
      "epoch": 3.699641702350782,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036795581184400977,
      "loss": 5.3869,
      "mean_token_accuracy": 0.19739913195371628,
      "num_tokens": 85783936.0,
      "step": 42335
    },
    {
      "entropy": 5.967626237869263,
      "epoch": 3.700078650703487,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000367927433509412,
      "loss": 5.4913,
      "mean_token_accuracy": 0.1839989498257637,
      "num_tokens": 85794427.0,
      "step": 42340
    },
    {
      "entropy": 5.970624780654907,
      "epoch": 3.7005155990561915,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036789905339241737,
      "loss": 5.4993,
      "mean_token_accuracy": 0.18474035412073136,
      "num_tokens": 85804367.0,
      "step": 42345
    },
    {
      "entropy": 5.904102182388305,
      "epoch": 3.700952547408896,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003678706714935702,
      "loss": 5.3364,
      "mean_token_accuracy": 0.2005745366215706,
      "num_tokens": 85814393.0,
      "step": 42350
    },
    {
      "entropy": 5.894415187835693,
      "epoch": 3.701389495761601,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003678422878134149,
      "loss": 5.4163,
      "mean_token_accuracy": 0.191349658370018,
      "num_tokens": 85824979.0,
      "step": 42355
    },
    {
      "entropy": 5.94775767326355,
      "epoch": 3.701826444114306,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036781390235249593,
      "loss": 5.4772,
      "mean_token_accuracy": 0.18948538601398468,
      "num_tokens": 85835181.0,
      "step": 42360
    },
    {
      "entropy": 5.909161138534546,
      "epoch": 3.7022633924670103,
      "grad_norm": 1.0,
      "learning_rate": 0.0003677855151113576,
      "loss": 5.4465,
      "mean_token_accuracy": 0.19004557579755782,
      "num_tokens": 85844720.0,
      "step": 42365
    },
    {
      "entropy": 5.994647645950318,
      "epoch": 3.702700340819715,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003677571260905445,
      "loss": 5.5178,
      "mean_token_accuracy": 0.18564348220825194,
      "num_tokens": 85854085.0,
      "step": 42370
    },
    {
      "entropy": 5.931602954864502,
      "epoch": 3.7031372891724197,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003677287352906012,
      "loss": 5.4689,
      "mean_token_accuracy": 0.19130588918924332,
      "num_tokens": 85864675.0,
      "step": 42375
    },
    {
      "entropy": 5.900839948654175,
      "epoch": 3.7035742375251246,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036770034271207204,
      "loss": 5.3996,
      "mean_token_accuracy": 0.1938207432627678,
      "num_tokens": 85875122.0,
      "step": 42380
    },
    {
      "entropy": 5.924886083602905,
      "epoch": 3.704011185877829,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036767194835550185,
      "loss": 5.359,
      "mean_token_accuracy": 0.19552288949489594,
      "num_tokens": 85884485.0,
      "step": 42385
    },
    {
      "entropy": 5.865861892700195,
      "epoch": 3.704448134230534,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000367643552221435,
      "loss": 5.444,
      "mean_token_accuracy": 0.19244116097688674,
      "num_tokens": 85895033.0,
      "step": 42390
    },
    {
      "entropy": 5.883698320388794,
      "epoch": 3.704885082583239,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036761515431041634,
      "loss": 5.3978,
      "mean_token_accuracy": 0.19432517588138581,
      "num_tokens": 85904882.0,
      "step": 42395
    },
    {
      "entropy": 5.943054580688477,
      "epoch": 3.7053220309359434,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036758675462299044,
      "loss": 5.4305,
      "mean_token_accuracy": 0.1929578587412834,
      "num_tokens": 85914173.0,
      "step": 42400
    },
    {
      "entropy": 5.886132526397705,
      "epoch": 3.705758979288648,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003675583531597022,
      "loss": 5.4885,
      "mean_token_accuracy": 0.18367073088884353,
      "num_tokens": 85923682.0,
      "step": 42405
    },
    {
      "entropy": 5.941156721115112,
      "epoch": 3.7061959276413527,
      "grad_norm": 1.0,
      "learning_rate": 0.00036752994992109617,
      "loss": 5.4206,
      "mean_token_accuracy": 0.18746291548013688,
      "num_tokens": 85934061.0,
      "step": 42410
    },
    {
      "entropy": 5.961226606369019,
      "epoch": 3.7066328759940577,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003675015449077172,
      "loss": 5.4448,
      "mean_token_accuracy": 0.18821489363908767,
      "num_tokens": 85943568.0,
      "step": 42415
    },
    {
      "entropy": 5.90615758895874,
      "epoch": 3.707069824346762,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036747313812011007,
      "loss": 5.4198,
      "mean_token_accuracy": 0.1964723765850067,
      "num_tokens": 85953998.0,
      "step": 42420
    },
    {
      "entropy": 5.868173694610595,
      "epoch": 3.707506772699467,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003674447295588198,
      "loss": 5.4048,
      "mean_token_accuracy": 0.19807036817073823,
      "num_tokens": 85963520.0,
      "step": 42425
    },
    {
      "entropy": 5.8748046398162845,
      "epoch": 3.7079437210521715,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003674163192243912,
      "loss": 5.408,
      "mean_token_accuracy": 0.18714888393878937,
      "num_tokens": 85974142.0,
      "step": 42430
    },
    {
      "entropy": 5.930280017852783,
      "epoch": 3.7083806694048764,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036738790711736907,
      "loss": 5.4499,
      "mean_token_accuracy": 0.19388801902532576,
      "num_tokens": 85984724.0,
      "step": 42435
    },
    {
      "entropy": 5.871231269836426,
      "epoch": 3.708817617757581,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003673594932382985,
      "loss": 5.342,
      "mean_token_accuracy": 0.20339301824569703,
      "num_tokens": 85995475.0,
      "step": 42440
    },
    {
      "entropy": 5.849819803237915,
      "epoch": 3.709254566110286,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003673310775877244,
      "loss": 5.4263,
      "mean_token_accuracy": 0.1988450691103935,
      "num_tokens": 86005672.0,
      "step": 42445
    },
    {
      "entropy": 5.965377855300903,
      "epoch": 3.7096915144629907,
      "grad_norm": 1.0625,
      "learning_rate": 0.000367302660166192,
      "loss": 5.4971,
      "mean_token_accuracy": 0.18445746153593062,
      "num_tokens": 86016918.0,
      "step": 42450
    },
    {
      "entropy": 5.932121992111206,
      "epoch": 3.710128462815695,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036727424097424615,
      "loss": 5.4953,
      "mean_token_accuracy": 0.18743828386068345,
      "num_tokens": 86029345.0,
      "step": 42455
    },
    {
      "entropy": 5.952695035934449,
      "epoch": 3.7105654111683997,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036724582001243196,
      "loss": 5.4195,
      "mean_token_accuracy": 0.19243264198303223,
      "num_tokens": 86039368.0,
      "step": 42460
    },
    {
      "entropy": 5.932540559768677,
      "epoch": 3.7110023595211046,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003672173972812946,
      "loss": 5.475,
      "mean_token_accuracy": 0.18383533209562303,
      "num_tokens": 86048220.0,
      "step": 42465
    },
    {
      "entropy": 6.023512268066407,
      "epoch": 3.7114393078738095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036718897278137933,
      "loss": 5.4997,
      "mean_token_accuracy": 0.18133290559053422,
      "num_tokens": 86058316.0,
      "step": 42470
    },
    {
      "entropy": 5.896181678771972,
      "epoch": 3.711876256226514,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003671605465132312,
      "loss": 5.3818,
      "mean_token_accuracy": 0.19256990700960158,
      "num_tokens": 86068995.0,
      "step": 42475
    },
    {
      "entropy": 5.908245515823364,
      "epoch": 3.712313204579219,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036713211847739553,
      "loss": 5.4043,
      "mean_token_accuracy": 0.19750444442033768,
      "num_tokens": 86080604.0,
      "step": 42480
    },
    {
      "entropy": 5.922627401351929,
      "epoch": 3.7127501529319233,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00036710368867441743,
      "loss": 5.4596,
      "mean_token_accuracy": 0.1921766445040703,
      "num_tokens": 86092636.0,
      "step": 42485
    },
    {
      "entropy": 5.867632818222046,
      "epoch": 3.7131871012846283,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003670752571048425,
      "loss": 5.4335,
      "mean_token_accuracy": 0.19019600003957748,
      "num_tokens": 86101454.0,
      "step": 42490
    },
    {
      "entropy": 5.8966573715209964,
      "epoch": 3.7136240496373327,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036704682376921576,
      "loss": 5.4488,
      "mean_token_accuracy": 0.1962893009185791,
      "num_tokens": 86111460.0,
      "step": 42495
    },
    {
      "entropy": 5.948596572875976,
      "epoch": 3.7140609979900376,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003670183886680828,
      "loss": 5.4431,
      "mean_token_accuracy": 0.19356634020805358,
      "num_tokens": 86121347.0,
      "step": 42500
    },
    {
      "entropy": 5.952212142944336,
      "epoch": 3.7144979463427426,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036698995180198886,
      "loss": 5.5358,
      "mean_token_accuracy": 0.1869075819849968,
      "num_tokens": 86132084.0,
      "step": 42505
    },
    {
      "entropy": 5.865876531600952,
      "epoch": 3.714934894695447,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00036696151317147944,
      "loss": 5.372,
      "mean_token_accuracy": 0.2082224503159523,
      "num_tokens": 86141300.0,
      "step": 42510
    },
    {
      "entropy": 5.899409055709839,
      "epoch": 3.7153718430481515,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036693307277710005,
      "loss": 5.4347,
      "mean_token_accuracy": 0.19410257637500763,
      "num_tokens": 86151640.0,
      "step": 42515
    },
    {
      "entropy": 5.882134103775025,
      "epoch": 3.7158087914008564,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00036690463061939615,
      "loss": 5.3147,
      "mean_token_accuracy": 0.2031505823135376,
      "num_tokens": 86162442.0,
      "step": 42520
    },
    {
      "entropy": 5.901881980895996,
      "epoch": 3.7162457397535613,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003668761866989132,
      "loss": 5.445,
      "mean_token_accuracy": 0.18998920321464538,
      "num_tokens": 86172824.0,
      "step": 42525
    },
    {
      "entropy": 5.877415323257447,
      "epoch": 3.716682688106266,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003668477410161969,
      "loss": 5.3692,
      "mean_token_accuracy": 0.20185719579458236,
      "num_tokens": 86182829.0,
      "step": 42530
    },
    {
      "entropy": 5.933149957656861,
      "epoch": 3.7171196364589707,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00036681929357179285,
      "loss": 5.4993,
      "mean_token_accuracy": 0.18828742802143097,
      "num_tokens": 86193656.0,
      "step": 42535
    },
    {
      "entropy": 5.879664039611816,
      "epoch": 3.717556584811675,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003667908443662466,
      "loss": 5.3917,
      "mean_token_accuracy": 0.20074671357870102,
      "num_tokens": 86203025.0,
      "step": 42540
    },
    {
      "entropy": 5.847734355926514,
      "epoch": 3.71799353316438,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036676239340010386,
      "loss": 5.3954,
      "mean_token_accuracy": 0.19720099866390228,
      "num_tokens": 86213029.0,
      "step": 42545
    },
    {
      "entropy": 6.030997276306152,
      "epoch": 3.7184304815170846,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003667339406739103,
      "loss": 5.4655,
      "mean_token_accuracy": 0.1814518004655838,
      "num_tokens": 86222725.0,
      "step": 42550
    },
    {
      "entropy": 5.962336635589599,
      "epoch": 3.7188674298697895,
      "grad_norm": 1.0,
      "learning_rate": 0.0003667054861882117,
      "loss": 5.5009,
      "mean_token_accuracy": 0.18163650035858153,
      "num_tokens": 86232514.0,
      "step": 42555
    },
    {
      "entropy": 5.95518307685852,
      "epoch": 3.7193043782224944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036667702994355386,
      "loss": 5.5537,
      "mean_token_accuracy": 0.18270711302757264,
      "num_tokens": 86242840.0,
      "step": 42560
    },
    {
      "entropy": 5.985308504104614,
      "epoch": 3.719741326575199,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003666485719404825,
      "loss": 5.5575,
      "mean_token_accuracy": 0.18096640110015869,
      "num_tokens": 86253221.0,
      "step": 42565
    },
    {
      "entropy": 5.951711845397949,
      "epoch": 3.7201782749279033,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036662011217954355,
      "loss": 5.4536,
      "mean_token_accuracy": 0.18554124236106873,
      "num_tokens": 86263154.0,
      "step": 42570
    },
    {
      "entropy": 5.933514976501465,
      "epoch": 3.7206152232806082,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003665916506612828,
      "loss": 5.508,
      "mean_token_accuracy": 0.18439760208129882,
      "num_tokens": 86273041.0,
      "step": 42575
    },
    {
      "entropy": 5.93213963508606,
      "epoch": 3.721052171633313,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036656318738624626,
      "loss": 5.4247,
      "mean_token_accuracy": 0.19609395563602447,
      "num_tokens": 86284669.0,
      "step": 42580
    },
    {
      "entropy": 5.858678197860717,
      "epoch": 3.7214891199860176,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003665347223549797,
      "loss": 5.3131,
      "mean_token_accuracy": 0.2044554904103279,
      "num_tokens": 86294186.0,
      "step": 42585
    },
    {
      "entropy": 5.817522573471069,
      "epoch": 3.7219260683387225,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003665062555680293,
      "loss": 5.3796,
      "mean_token_accuracy": 0.18936630785465242,
      "num_tokens": 86304055.0,
      "step": 42590
    },
    {
      "entropy": 5.964726209640503,
      "epoch": 3.722363016691427,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036647778702594087,
      "loss": 5.4456,
      "mean_token_accuracy": 0.19115789085626603,
      "num_tokens": 86314258.0,
      "step": 42595
    },
    {
      "entropy": 5.911658668518067,
      "epoch": 3.722799965044132,
      "grad_norm": 1.0,
      "learning_rate": 0.00036644931672926063,
      "loss": 5.4022,
      "mean_token_accuracy": 0.19470895677804947,
      "num_tokens": 86324829.0,
      "step": 42600
    },
    {
      "entropy": 5.901606321334839,
      "epoch": 3.7232369133968364,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003664208446785346,
      "loss": 5.4548,
      "mean_token_accuracy": 0.19256600886583328,
      "num_tokens": 86334901.0,
      "step": 42605
    },
    {
      "entropy": 5.9032268047332765,
      "epoch": 3.7236738617495413,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036639237087430875,
      "loss": 5.4473,
      "mean_token_accuracy": 0.19273791462183,
      "num_tokens": 86345798.0,
      "step": 42610
    },
    {
      "entropy": 6.018934297561645,
      "epoch": 3.7241108101022458,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003663638953171295,
      "loss": 5.5292,
      "mean_token_accuracy": 0.18341776728630066,
      "num_tokens": 86355676.0,
      "step": 42615
    },
    {
      "entropy": 5.958837175369263,
      "epoch": 3.7245477584549507,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036633541800754277,
      "loss": 5.4561,
      "mean_token_accuracy": 0.19337234944105147,
      "num_tokens": 86366590.0,
      "step": 42620
    },
    {
      "entropy": 5.938102102279663,
      "epoch": 3.724984706807655,
      "grad_norm": 1.0,
      "learning_rate": 0.00036630693894609487,
      "loss": 5.4836,
      "mean_token_accuracy": 0.19142054468393327,
      "num_tokens": 86376495.0,
      "step": 42625
    },
    {
      "entropy": 5.85388674736023,
      "epoch": 3.72542165516036,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00036627845813333204,
      "loss": 5.4309,
      "mean_token_accuracy": 0.1887502819299698,
      "num_tokens": 86386990.0,
      "step": 42630
    },
    {
      "entropy": 5.967373275756836,
      "epoch": 3.725858603513065,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003662499755698006,
      "loss": 5.4247,
      "mean_token_accuracy": 0.1909652382135391,
      "num_tokens": 86396989.0,
      "step": 42635
    },
    {
      "entropy": 5.8928934097290036,
      "epoch": 3.7262955518657694,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036622149125604676,
      "loss": 5.395,
      "mean_token_accuracy": 0.19819358438253404,
      "num_tokens": 86407870.0,
      "step": 42640
    },
    {
      "entropy": 5.863723087310791,
      "epoch": 3.726732500218474,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003661930051926169,
      "loss": 5.372,
      "mean_token_accuracy": 0.19847846329212188,
      "num_tokens": 86417831.0,
      "step": 42645
    },
    {
      "entropy": 5.91892523765564,
      "epoch": 3.727169448571179,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003661645173800574,
      "loss": 5.5377,
      "mean_token_accuracy": 0.18936415761709213,
      "num_tokens": 86428186.0,
      "step": 42650
    },
    {
      "entropy": 5.925449466705322,
      "epoch": 3.7276063969238837,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036613602781891474,
      "loss": 5.4697,
      "mean_token_accuracy": 0.19778748750686645,
      "num_tokens": 86438592.0,
      "step": 42655
    },
    {
      "entropy": 5.895026445388794,
      "epoch": 3.728043345276588,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036610753650973536,
      "loss": 5.3515,
      "mean_token_accuracy": 0.19888793379068376,
      "num_tokens": 86448488.0,
      "step": 42660
    },
    {
      "entropy": 5.912296772003174,
      "epoch": 3.728480293629293,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003660790434530656,
      "loss": 5.407,
      "mean_token_accuracy": 0.19775228649377824,
      "num_tokens": 86459524.0,
      "step": 42665
    },
    {
      "entropy": 5.907283687591553,
      "epoch": 3.7289172419819976,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036605054864945213,
      "loss": 5.2941,
      "mean_token_accuracy": 0.2060931444168091,
      "num_tokens": 86470264.0,
      "step": 42670
    },
    {
      "entropy": 5.883208990097046,
      "epoch": 3.7293541903347025,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003660220520994413,
      "loss": 5.4344,
      "mean_token_accuracy": 0.189316962659359,
      "num_tokens": 86481155.0,
      "step": 42675
    },
    {
      "entropy": 5.883836460113526,
      "epoch": 3.729791138687407,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036599355380358004,
      "loss": 5.4688,
      "mean_token_accuracy": 0.18482399880886077,
      "num_tokens": 86491666.0,
      "step": 42680
    },
    {
      "entropy": 5.964666509628296,
      "epoch": 3.730228087040112,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036596505376241455,
      "loss": 5.5037,
      "mean_token_accuracy": 0.1893462359905243,
      "num_tokens": 86501953.0,
      "step": 42685
    },
    {
      "entropy": 5.879721927642822,
      "epoch": 3.730665035392817,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003659365519764917,
      "loss": 5.3563,
      "mean_token_accuracy": 0.19941780269145964,
      "num_tokens": 86512658.0,
      "step": 42690
    },
    {
      "entropy": 5.900302457809448,
      "epoch": 3.7311019837455213,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036590804844635817,
      "loss": 5.5325,
      "mean_token_accuracy": 0.1848825216293335,
      "num_tokens": 86522856.0,
      "step": 42695
    },
    {
      "entropy": 5.996601533889771,
      "epoch": 3.7315389320982257,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036587954317256064,
      "loss": 5.4686,
      "mean_token_accuracy": 0.18941646218299865,
      "num_tokens": 86533650.0,
      "step": 42700
    },
    {
      "entropy": 5.999467992782593,
      "epoch": 3.7319758804509306,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003658510361556458,
      "loss": 5.4947,
      "mean_token_accuracy": 0.18570114970207213,
      "num_tokens": 86543801.0,
      "step": 42705
    },
    {
      "entropy": 5.9217194557189945,
      "epoch": 3.7324128288036356,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003658225273961605,
      "loss": 5.4477,
      "mean_token_accuracy": 0.18771475553512573,
      "num_tokens": 86553637.0,
      "step": 42710
    },
    {
      "entropy": 5.9264250755310055,
      "epoch": 3.73284977715634,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003657940168946516,
      "loss": 5.5028,
      "mean_token_accuracy": 0.1861940547823906,
      "num_tokens": 86563531.0,
      "step": 42715
    },
    {
      "entropy": 5.828302049636841,
      "epoch": 3.733286725509045,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003657655046516658,
      "loss": 5.3826,
      "mean_token_accuracy": 0.2056463450193405,
      "num_tokens": 86573675.0,
      "step": 42720
    },
    {
      "entropy": 5.893008470535278,
      "epoch": 3.7337236738617494,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036573699066775006,
      "loss": 5.4003,
      "mean_token_accuracy": 0.19269172549247743,
      "num_tokens": 86583817.0,
      "step": 42725
    },
    {
      "entropy": 5.889606857299805,
      "epoch": 3.7341606222144543,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00036570847494345134,
      "loss": 5.4495,
      "mean_token_accuracy": 0.1896290212869644,
      "num_tokens": 86593170.0,
      "step": 42730
    },
    {
      "entropy": 5.940661907196045,
      "epoch": 3.734597570567159,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003656799574793166,
      "loss": 5.4629,
      "mean_token_accuracy": 0.1871857687830925,
      "num_tokens": 86603224.0,
      "step": 42735
    },
    {
      "entropy": 5.917451667785644,
      "epoch": 3.7350345189198637,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00036565143827589267,
      "loss": 5.3815,
      "mean_token_accuracy": 0.1990641325712204,
      "num_tokens": 86613442.0,
      "step": 42740
    },
    {
      "entropy": 5.91563081741333,
      "epoch": 3.7354714672725686,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003656229173337266,
      "loss": 5.3459,
      "mean_token_accuracy": 0.19914075285196303,
      "num_tokens": 86623459.0,
      "step": 42745
    },
    {
      "entropy": 5.783817481994629,
      "epoch": 3.735908415625273,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003655943946533656,
      "loss": 5.327,
      "mean_token_accuracy": 0.20438577681779863,
      "num_tokens": 86633407.0,
      "step": 42750
    },
    {
      "entropy": 5.8399436473846436,
      "epoch": 3.7363453639779776,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003655658702353566,
      "loss": 5.4558,
      "mean_token_accuracy": 0.18788781017065048,
      "num_tokens": 86643902.0,
      "step": 42755
    },
    {
      "entropy": 5.92161979675293,
      "epoch": 3.7367823123306825,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003655373440802467,
      "loss": 5.4069,
      "mean_token_accuracy": 0.19572288542985916,
      "num_tokens": 86653871.0,
      "step": 42760
    },
    {
      "entropy": 5.940517234802246,
      "epoch": 3.7372192606833874,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003655088161885831,
      "loss": 5.3769,
      "mean_token_accuracy": 0.20251520425081254,
      "num_tokens": 86663890.0,
      "step": 42765
    },
    {
      "entropy": 5.843564081192016,
      "epoch": 3.737656209036092,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003654802865609131,
      "loss": 5.3757,
      "mean_token_accuracy": 0.1990557387471199,
      "num_tokens": 86672954.0,
      "step": 42770
    },
    {
      "entropy": 5.8747272968292235,
      "epoch": 3.7380931573887968,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003654517551977837,
      "loss": 5.4722,
      "mean_token_accuracy": 0.18863226920366288,
      "num_tokens": 86682432.0,
      "step": 42775
    },
    {
      "entropy": 5.951723575592041,
      "epoch": 3.7385301057415012,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00036542322209974227,
      "loss": 5.397,
      "mean_token_accuracy": 0.19676092118024827,
      "num_tokens": 86691832.0,
      "step": 42780
    },
    {
      "entropy": 5.95592474937439,
      "epoch": 3.738967054094206,
      "grad_norm": 1.0,
      "learning_rate": 0.000365394687267336,
      "loss": 5.5149,
      "mean_token_accuracy": 0.1890360891819,
      "num_tokens": 86703284.0,
      "step": 42785
    },
    {
      "entropy": 5.965306520462036,
      "epoch": 3.7394040024469106,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036536615070111234,
      "loss": 5.4886,
      "mean_token_accuracy": 0.18968641459941865,
      "num_tokens": 86713619.0,
      "step": 42790
    },
    {
      "entropy": 5.918927001953125,
      "epoch": 3.7398409507996155,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003653376124016185,
      "loss": 5.4047,
      "mean_token_accuracy": 0.191445092856884,
      "num_tokens": 86723802.0,
      "step": 42795
    },
    {
      "entropy": 5.927799558639526,
      "epoch": 3.7402778991523205,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000365309072369402,
      "loss": 5.4255,
      "mean_token_accuracy": 0.19689894914627076,
      "num_tokens": 86733855.0,
      "step": 42800
    },
    {
      "entropy": 5.879811382293701,
      "epoch": 3.740714847505025,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036528053060501,
      "loss": 5.4176,
      "mean_token_accuracy": 0.2019481658935547,
      "num_tokens": 86742969.0,
      "step": 42805
    },
    {
      "entropy": 5.917296314239502,
      "epoch": 3.7411517958577294,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003652519871089903,
      "loss": 5.4573,
      "mean_token_accuracy": 0.1898327201604843,
      "num_tokens": 86753749.0,
      "step": 42810
    },
    {
      "entropy": 5.999043321609497,
      "epoch": 3.7415887442104343,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003652234418818901,
      "loss": 5.4541,
      "mean_token_accuracy": 0.18782103061676025,
      "num_tokens": 86764287.0,
      "step": 42815
    },
    {
      "entropy": 5.970267057418823,
      "epoch": 3.742025692563139,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00036519489492425704,
      "loss": 5.5083,
      "mean_token_accuracy": 0.18926939815282823,
      "num_tokens": 86774711.0,
      "step": 42820
    },
    {
      "entropy": 5.793001508712768,
      "epoch": 3.7424626409158437,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003651663462366387,
      "loss": 5.3787,
      "mean_token_accuracy": 0.20388678908348085,
      "num_tokens": 86784551.0,
      "step": 42825
    },
    {
      "entropy": 5.796547317504883,
      "epoch": 3.7428995892685486,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003651377958195825,
      "loss": 5.2897,
      "mean_token_accuracy": 0.2058465823531151,
      "num_tokens": 86794258.0,
      "step": 42830
    },
    {
      "entropy": 5.910200738906861,
      "epoch": 3.743336537621253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003651092436736362,
      "loss": 5.4033,
      "mean_token_accuracy": 0.19020977914333342,
      "num_tokens": 86804332.0,
      "step": 42835
    },
    {
      "entropy": 5.910720729827881,
      "epoch": 3.743773485973958,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003650806897993473,
      "loss": 5.3519,
      "mean_token_accuracy": 0.20196821242570878,
      "num_tokens": 86814593.0,
      "step": 42840
    },
    {
      "entropy": 5.879820060729981,
      "epoch": 3.7442104343266625,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036505213419726364,
      "loss": 5.4959,
      "mean_token_accuracy": 0.1879050463438034,
      "num_tokens": 86824489.0,
      "step": 42845
    },
    {
      "entropy": 5.887066411972046,
      "epoch": 3.7446473826793674,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036502357686793284,
      "loss": 5.4688,
      "mean_token_accuracy": 0.1946820229291916,
      "num_tokens": 86835052.0,
      "step": 42850
    },
    {
      "entropy": 5.931369924545288,
      "epoch": 3.7450843310320723,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003649950178119027,
      "loss": 5.3914,
      "mean_token_accuracy": 0.19636888056993484,
      "num_tokens": 86845183.0,
      "step": 42855
    },
    {
      "entropy": 5.873373413085938,
      "epoch": 3.7455212793847767,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003649664570297209,
      "loss": 5.4041,
      "mean_token_accuracy": 0.19685003608465196,
      "num_tokens": 86854752.0,
      "step": 42860
    },
    {
      "entropy": 5.916598892211914,
      "epoch": 3.745958227737481,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00036493789452193533,
      "loss": 5.4301,
      "mean_token_accuracy": 0.19138217121362686,
      "num_tokens": 86865876.0,
      "step": 42865
    },
    {
      "entropy": 5.901046323776245,
      "epoch": 3.746395176090186,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036490933028909383,
      "loss": 5.449,
      "mean_token_accuracy": 0.19704513549804686,
      "num_tokens": 86875626.0,
      "step": 42870
    },
    {
      "entropy": 5.889851379394531,
      "epoch": 3.746832124442891,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036488076433174426,
      "loss": 5.3654,
      "mean_token_accuracy": 0.19650930911302567,
      "num_tokens": 86884944.0,
      "step": 42875
    },
    {
      "entropy": 5.905115461349487,
      "epoch": 3.7472690727955955,
      "grad_norm": 1.0,
      "learning_rate": 0.00036485219665043463,
      "loss": 5.4495,
      "mean_token_accuracy": 0.19548297971487044,
      "num_tokens": 86894570.0,
      "step": 42880
    },
    {
      "entropy": 5.8744447231292725,
      "epoch": 3.7477060211483004,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003648236272457127,
      "loss": 5.4193,
      "mean_token_accuracy": 0.19204108715057372,
      "num_tokens": 86904093.0,
      "step": 42885
    },
    {
      "entropy": 5.942668485641479,
      "epoch": 3.748142969501005,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003647950561181265,
      "loss": 5.4309,
      "mean_token_accuracy": 0.19227448850870132,
      "num_tokens": 86915019.0,
      "step": 42890
    },
    {
      "entropy": 5.956768751144409,
      "epoch": 3.74857991785371,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036476648326822403,
      "loss": 5.377,
      "mean_token_accuracy": 0.19739483743906022,
      "num_tokens": 86924695.0,
      "step": 42895
    },
    {
      "entropy": 5.888198566436768,
      "epoch": 3.7490168662064143,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003647379086965535,
      "loss": 5.3702,
      "mean_token_accuracy": 0.20147186666727065,
      "num_tokens": 86935030.0,
      "step": 42900
    },
    {
      "entropy": 5.890108680725097,
      "epoch": 3.749453814559119,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036470933240366276,
      "loss": 5.4909,
      "mean_token_accuracy": 0.19190214425325394,
      "num_tokens": 86945871.0,
      "step": 42905
    },
    {
      "entropy": 5.948291349411011,
      "epoch": 3.7498907629118237,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036468075439010007,
      "loss": 5.6191,
      "mean_token_accuracy": 0.19309444725513458,
      "num_tokens": 86958855.0,
      "step": 42910
    },
    {
      "entropy": 5.974273157119751,
      "epoch": 3.7503277112645286,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036465217465641355,
      "loss": 5.4176,
      "mean_token_accuracy": 0.19525895416736602,
      "num_tokens": 86969071.0,
      "step": 42915
    },
    {
      "entropy": 5.903648138046265,
      "epoch": 3.750764659617233,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036462359320315127,
      "loss": 5.4199,
      "mean_token_accuracy": 0.1887261003255844,
      "num_tokens": 86978985.0,
      "step": 42920
    },
    {
      "entropy": 5.844113731384278,
      "epoch": 3.751201607969938,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003645950100308616,
      "loss": 5.3133,
      "mean_token_accuracy": 0.20576980113983154,
      "num_tokens": 86988527.0,
      "step": 42925
    },
    {
      "entropy": 5.934939765930176,
      "epoch": 3.751638556322643,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003645664251400926,
      "loss": 5.42,
      "mean_token_accuracy": 0.19080697298049926,
      "num_tokens": 86998218.0,
      "step": 42930
    },
    {
      "entropy": 5.818993139266968,
      "epoch": 3.7520755046753473,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036453783853139266,
      "loss": 5.278,
      "mean_token_accuracy": 0.20691969990730286,
      "num_tokens": 87007147.0,
      "step": 42935
    },
    {
      "entropy": 5.832198858261108,
      "epoch": 3.752512453028052,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003645092502053101,
      "loss": 5.355,
      "mean_token_accuracy": 0.20087713450193406,
      "num_tokens": 87016440.0,
      "step": 42940
    },
    {
      "entropy": 5.853746366500855,
      "epoch": 3.7529494013807567,
      "grad_norm": 0.90625,
      "learning_rate": 0.00036448066016239316,
      "loss": 5.3255,
      "mean_token_accuracy": 0.2026684910058975,
      "num_tokens": 87027969.0,
      "step": 42945
    },
    {
      "entropy": 6.006126785278321,
      "epoch": 3.7533863497334616,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003644520684031902,
      "loss": 5.4215,
      "mean_token_accuracy": 0.1916422054171562,
      "num_tokens": 87038403.0,
      "step": 42950
    },
    {
      "entropy": 5.906960582733154,
      "epoch": 3.753823298086166,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00036442347492824983,
      "loss": 5.4417,
      "mean_token_accuracy": 0.19221145510673524,
      "num_tokens": 87047484.0,
      "step": 42955
    },
    {
      "entropy": 5.924467945098877,
      "epoch": 3.754260246438871,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003643948797381203,
      "loss": 5.4055,
      "mean_token_accuracy": 0.19060832411050796,
      "num_tokens": 87058151.0,
      "step": 42960
    },
    {
      "entropy": 5.866067552566529,
      "epoch": 3.7546971947915755,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003643662828333502,
      "loss": 5.351,
      "mean_token_accuracy": 0.1997813105583191,
      "num_tokens": 87068256.0,
      "step": 42965
    },
    {
      "entropy": 5.978467416763306,
      "epoch": 3.7551341431442804,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003643376842144879,
      "loss": 5.5149,
      "mean_token_accuracy": 0.1828669711947441,
      "num_tokens": 87078346.0,
      "step": 42970
    },
    {
      "entropy": 5.920680236816406,
      "epoch": 3.755571091496985,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036430908388208205,
      "loss": 5.3904,
      "mean_token_accuracy": 0.19058413058519363,
      "num_tokens": 87088086.0,
      "step": 42975
    },
    {
      "entropy": 5.868765354156494,
      "epoch": 3.75600803984969,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003642804818366811,
      "loss": 5.4097,
      "mean_token_accuracy": 0.19857707172632216,
      "num_tokens": 87098358.0,
      "step": 42980
    },
    {
      "entropy": 5.937406873703003,
      "epoch": 3.7564449882023947,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036425187807883377,
      "loss": 5.5029,
      "mean_token_accuracy": 0.18803760707378386,
      "num_tokens": 87108970.0,
      "step": 42985
    },
    {
      "entropy": 5.938066005706787,
      "epoch": 3.756881936555099,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036422327260908873,
      "loss": 5.4195,
      "mean_token_accuracy": 0.20046785175800325,
      "num_tokens": 87117950.0,
      "step": 42990
    },
    {
      "entropy": 6.002581405639648,
      "epoch": 3.7573188849078036,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003641946654279945,
      "loss": 5.533,
      "mean_token_accuracy": 0.1845380112528801,
      "num_tokens": 87129177.0,
      "step": 42995
    },
    {
      "entropy": 5.95547080039978,
      "epoch": 3.7577558332605085,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003641660565360997,
      "loss": 5.4848,
      "mean_token_accuracy": 0.19041720181703567,
      "num_tokens": 87140418.0,
      "step": 43000
    },
    {
      "entropy": 5.948586320877075,
      "epoch": 3.7581927816132135,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036413744593395345,
      "loss": 5.425,
      "mean_token_accuracy": 0.18987047523260117,
      "num_tokens": 87151624.0,
      "step": 43005
    },
    {
      "entropy": 5.9694970607757565,
      "epoch": 3.758629729965918,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036410883362210416,
      "loss": 5.4484,
      "mean_token_accuracy": 0.18838099688291549,
      "num_tokens": 87160880.0,
      "step": 43010
    },
    {
      "entropy": 5.873884296417236,
      "epoch": 3.759066678318623,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036408021960110084,
      "loss": 5.3724,
      "mean_token_accuracy": 0.19528738111257554,
      "num_tokens": 87170673.0,
      "step": 43015
    },
    {
      "entropy": 5.879766607284546,
      "epoch": 3.7595036266713273,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003640516038714921,
      "loss": 5.4121,
      "mean_token_accuracy": 0.1954762876033783,
      "num_tokens": 87180751.0,
      "step": 43020
    },
    {
      "entropy": 5.9471070766448975,
      "epoch": 3.7599405750240322,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036402298643382696,
      "loss": 5.4175,
      "mean_token_accuracy": 0.19677240401506424,
      "num_tokens": 87191501.0,
      "step": 43025
    },
    {
      "entropy": 5.909118032455444,
      "epoch": 3.7603775233767367,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003639943672886544,
      "loss": 5.4382,
      "mean_token_accuracy": 0.19463686048984527,
      "num_tokens": 87201382.0,
      "step": 43030
    },
    {
      "entropy": 5.888462162017822,
      "epoch": 3.7608144717294416,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00036396574643652316,
      "loss": 5.4369,
      "mean_token_accuracy": 0.1896348774433136,
      "num_tokens": 87212454.0,
      "step": 43035
    },
    {
      "entropy": 5.939186763763428,
      "epoch": 3.7612514200821465,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036393712387798236,
      "loss": 5.4198,
      "mean_token_accuracy": 0.19365792274475097,
      "num_tokens": 87222439.0,
      "step": 43040
    },
    {
      "entropy": 5.861933755874634,
      "epoch": 3.761688368434851,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036390849961358084,
      "loss": 5.3739,
      "mean_token_accuracy": 0.19429756700992584,
      "num_tokens": 87232620.0,
      "step": 43045
    },
    {
      "entropy": 5.910324668884277,
      "epoch": 3.7621253167875555,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003638798736438677,
      "loss": 5.4148,
      "mean_token_accuracy": 0.19386186003684996,
      "num_tokens": 87242031.0,
      "step": 43050
    },
    {
      "entropy": 5.86150484085083,
      "epoch": 3.7625622651402604,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036385124596939204,
      "loss": 5.2773,
      "mean_token_accuracy": 0.20481530874967574,
      "num_tokens": 87251113.0,
      "step": 43055
    },
    {
      "entropy": 5.872992372512817,
      "epoch": 3.7629992134929653,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000363822616590703,
      "loss": 5.3949,
      "mean_token_accuracy": 0.19227845668792726,
      "num_tokens": 87261255.0,
      "step": 43060
    },
    {
      "entropy": 5.988427686691284,
      "epoch": 3.7634361618456698,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003637939855083495,
      "loss": 5.5498,
      "mean_token_accuracy": 0.1812896892428398,
      "num_tokens": 87271677.0,
      "step": 43065
    },
    {
      "entropy": 5.926836633682251,
      "epoch": 3.7638731101983747,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003637653527228809,
      "loss": 5.5111,
      "mean_token_accuracy": 0.18751197308301926,
      "num_tokens": 87280417.0,
      "step": 43070
    },
    {
      "entropy": 5.95164155960083,
      "epoch": 3.764310058551079,
      "grad_norm": 2.140625,
      "learning_rate": 0.00036373671823484636,
      "loss": 5.5043,
      "mean_token_accuracy": 0.18737008273601533,
      "num_tokens": 87291069.0,
      "step": 43075
    },
    {
      "entropy": 6.015288686752319,
      "epoch": 3.764747006903784,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036370808204479497,
      "loss": 5.5323,
      "mean_token_accuracy": 0.1851954996585846,
      "num_tokens": 87301586.0,
      "step": 43080
    },
    {
      "entropy": 5.992779731750488,
      "epoch": 3.7651839552564885,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003636794441532761,
      "loss": 5.4359,
      "mean_token_accuracy": 0.19685456901788712,
      "num_tokens": 87312292.0,
      "step": 43085
    },
    {
      "entropy": 5.904217338562011,
      "epoch": 3.7656209036091934,
      "grad_norm": 1.0,
      "learning_rate": 0.00036365080456083915,
      "loss": 5.4184,
      "mean_token_accuracy": 0.19216244369745256,
      "num_tokens": 87321896.0,
      "step": 43090
    },
    {
      "entropy": 5.887899589538574,
      "epoch": 3.7660578519618984,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036362216326803323,
      "loss": 5.3974,
      "mean_token_accuracy": 0.1950958862900734,
      "num_tokens": 87331615.0,
      "step": 43095
    },
    {
      "entropy": 5.909921026229858,
      "epoch": 3.766494800314603,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003635935202754077,
      "loss": 5.4478,
      "mean_token_accuracy": 0.1890431299805641,
      "num_tokens": 87341345.0,
      "step": 43100
    },
    {
      "entropy": 5.9406520366668705,
      "epoch": 3.7669317486673073,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036356487558351214,
      "loss": 5.4337,
      "mean_token_accuracy": 0.18952849209308625,
      "num_tokens": 87351015.0,
      "step": 43105
    },
    {
      "entropy": 5.866679859161377,
      "epoch": 3.767368697020012,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036353622919289586,
      "loss": 5.4227,
      "mean_token_accuracy": 0.20029188990592955,
      "num_tokens": 87360976.0,
      "step": 43110
    },
    {
      "entropy": 5.936081218719482,
      "epoch": 3.767805645372717,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036350758110410825,
      "loss": 5.4089,
      "mean_token_accuracy": 0.19459358304739,
      "num_tokens": 87370863.0,
      "step": 43115
    },
    {
      "entropy": 5.9089884757995605,
      "epoch": 3.7682425937254216,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036347893131769894,
      "loss": 5.4162,
      "mean_token_accuracy": 0.1891456663608551,
      "num_tokens": 87380091.0,
      "step": 43120
    },
    {
      "entropy": 5.822175645828247,
      "epoch": 3.7686795420781265,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00036345027983421724,
      "loss": 5.4014,
      "mean_token_accuracy": 0.19889146238565444,
      "num_tokens": 87391005.0,
      "step": 43125
    },
    {
      "entropy": 5.889281129837036,
      "epoch": 3.769116490430831,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003634216266542129,
      "loss": 5.3692,
      "mean_token_accuracy": 0.20346253663301467,
      "num_tokens": 87400586.0,
      "step": 43130
    },
    {
      "entropy": 5.940411710739136,
      "epoch": 3.769553438783536,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036339297177823544,
      "loss": 5.4125,
      "mean_token_accuracy": 0.1995253011584282,
      "num_tokens": 87411054.0,
      "step": 43135
    },
    {
      "entropy": 5.8849711418151855,
      "epoch": 3.7699903871362404,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003633643152068344,
      "loss": 5.4034,
      "mean_token_accuracy": 0.19533893465995789,
      "num_tokens": 87421413.0,
      "step": 43140
    },
    {
      "entropy": 5.934936094284057,
      "epoch": 3.7704273354889453,
      "grad_norm": 0.90625,
      "learning_rate": 0.00036333565694055957,
      "loss": 5.3926,
      "mean_token_accuracy": 0.195073102414608,
      "num_tokens": 87431592.0,
      "step": 43145
    },
    {
      "entropy": 5.875665187835693,
      "epoch": 3.77086428384165,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036330699697996043,
      "loss": 5.4263,
      "mean_token_accuracy": 0.19048491567373277,
      "num_tokens": 87441708.0,
      "step": 43150
    },
    {
      "entropy": 5.8875476837158205,
      "epoch": 3.7713012321943546,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036327833532558704,
      "loss": 5.4378,
      "mean_token_accuracy": 0.1920551672577858,
      "num_tokens": 87452083.0,
      "step": 43155
    },
    {
      "entropy": 5.901424884796143,
      "epoch": 3.771738180547059,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003632496719779888,
      "loss": 5.418,
      "mean_token_accuracy": 0.19393349140882493,
      "num_tokens": 87460743.0,
      "step": 43160
    },
    {
      "entropy": 5.979290342330932,
      "epoch": 3.772175128899764,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036322100693771554,
      "loss": 5.456,
      "mean_token_accuracy": 0.1892918676137924,
      "num_tokens": 87470865.0,
      "step": 43165
    },
    {
      "entropy": 5.9434301376342775,
      "epoch": 3.772612077252469,
      "grad_norm": 1.0,
      "learning_rate": 0.00036319234020531723,
      "loss": 5.4694,
      "mean_token_accuracy": 0.19146504551172255,
      "num_tokens": 87482063.0,
      "step": 43170
    },
    {
      "entropy": 5.969300413131714,
      "epoch": 3.7730490256051734,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003631636717813435,
      "loss": 5.4212,
      "mean_token_accuracy": 0.19309413135051728,
      "num_tokens": 87492284.0,
      "step": 43175
    },
    {
      "entropy": 5.945356178283691,
      "epoch": 3.7734859739578783,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003631350016663445,
      "loss": 5.4681,
      "mean_token_accuracy": 0.19212356060743332,
      "num_tokens": 87501586.0,
      "step": 43180
    },
    {
      "entropy": 5.925312900543213,
      "epoch": 3.773922922310583,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036310632986086993,
      "loss": 5.4873,
      "mean_token_accuracy": 0.18715905994176865,
      "num_tokens": 87510622.0,
      "step": 43185
    },
    {
      "entropy": 5.921862030029297,
      "epoch": 3.7743598706632877,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003630776563654698,
      "loss": 5.4628,
      "mean_token_accuracy": 0.1861961677670479,
      "num_tokens": 87519480.0,
      "step": 43190
    },
    {
      "entropy": 6.001962661743164,
      "epoch": 3.774796819015992,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000363048981180694,
      "loss": 5.4382,
      "mean_token_accuracy": 0.19083696603775024,
      "num_tokens": 87529560.0,
      "step": 43195
    },
    {
      "entropy": 5.988526058197022,
      "epoch": 3.775233767368697,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036302030430709263,
      "loss": 5.5156,
      "mean_token_accuracy": 0.18459491282701493,
      "num_tokens": 87539969.0,
      "step": 43200
    },
    {
      "entropy": 5.897686433792114,
      "epoch": 3.775670715721402,
      "grad_norm": 0.96875,
      "learning_rate": 0.00036299162574521585,
      "loss": 5.4512,
      "mean_token_accuracy": 0.192854842543602,
      "num_tokens": 87550883.0,
      "step": 43205
    },
    {
      "entropy": 5.885497617721557,
      "epoch": 3.7761076640741065,
      "grad_norm": 0.953125,
      "learning_rate": 0.00036296294549561343,
      "loss": 5.392,
      "mean_token_accuracy": 0.19765715450048446,
      "num_tokens": 87560569.0,
      "step": 43210
    },
    {
      "entropy": 5.941313695907593,
      "epoch": 3.776544612426811,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003629342635588356,
      "loss": 5.4905,
      "mean_token_accuracy": 0.19240507781505584,
      "num_tokens": 87570881.0,
      "step": 43215
    },
    {
      "entropy": 5.929141902923584,
      "epoch": 3.776981560779516,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00036290557993543253,
      "loss": 5.406,
      "mean_token_accuracy": 0.21352732479572295,
      "num_tokens": 87581334.0,
      "step": 43220
    },
    {
      "entropy": 5.917771100997925,
      "epoch": 3.7774185091322208,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036287689462595447,
      "loss": 5.4181,
      "mean_token_accuracy": 0.19395218193531036,
      "num_tokens": 87591554.0,
      "step": 43225
    },
    {
      "entropy": 5.8879955291748045,
      "epoch": 3.7778554574849252,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003628482076309515,
      "loss": 5.3545,
      "mean_token_accuracy": 0.20039117485284805,
      "num_tokens": 87600484.0,
      "step": 43230
    },
    {
      "entropy": 5.908120203018188,
      "epoch": 3.77829240583763,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036281951895097384,
      "loss": 5.4595,
      "mean_token_accuracy": 0.19211444407701492,
      "num_tokens": 87610632.0,
      "step": 43235
    },
    {
      "entropy": 5.834910821914673,
      "epoch": 3.7787293541903346,
      "grad_norm": 0.921875,
      "learning_rate": 0.00036279082858657174,
      "loss": 5.2753,
      "mean_token_accuracy": 0.21081586927175522,
      "num_tokens": 87620835.0,
      "step": 43240
    },
    {
      "entropy": 5.953351259231567,
      "epoch": 3.7791663025430395,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003627621365382956,
      "loss": 5.4718,
      "mean_token_accuracy": 0.19022012650966644,
      "num_tokens": 87630437.0,
      "step": 43245
    },
    {
      "entropy": 5.961935043334961,
      "epoch": 3.779603250895744,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036273344280669573,
      "loss": 5.4839,
      "mean_token_accuracy": 0.18532403111457824,
      "num_tokens": 87640293.0,
      "step": 43250
    },
    {
      "entropy": 5.919696712493897,
      "epoch": 3.780040199248449,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036270474739232244,
      "loss": 5.3981,
      "mean_token_accuracy": 0.19227583408355714,
      "num_tokens": 87651026.0,
      "step": 43255
    },
    {
      "entropy": 5.942612028121948,
      "epoch": 3.7804771476011534,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036267605029572617,
      "loss": 5.4119,
      "mean_token_accuracy": 0.19427523016929626,
      "num_tokens": 87660957.0,
      "step": 43260
    },
    {
      "entropy": 5.921995449066162,
      "epoch": 3.7809140959538583,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003626473515174572,
      "loss": 5.3972,
      "mean_token_accuracy": 0.2024674728512764,
      "num_tokens": 87671311.0,
      "step": 43265
    },
    {
      "entropy": 5.867823123931885,
      "epoch": 3.7813510443065628,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003626186510580661,
      "loss": 5.3475,
      "mean_token_accuracy": 0.20032797008752823,
      "num_tokens": 87681217.0,
      "step": 43270
    },
    {
      "entropy": 5.89544997215271,
      "epoch": 3.7817879926592677,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036258994891810357,
      "loss": 5.4094,
      "mean_token_accuracy": 0.19850323647260665,
      "num_tokens": 87691014.0,
      "step": 43275
    },
    {
      "entropy": 5.918487691879273,
      "epoch": 3.7822249410119726,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003625612450981197,
      "loss": 5.4856,
      "mean_token_accuracy": 0.18676695078611374,
      "num_tokens": 87701125.0,
      "step": 43280
    },
    {
      "entropy": 5.85307354927063,
      "epoch": 3.782661889364677,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036253253959866533,
      "loss": 5.2648,
      "mean_token_accuracy": 0.20867226123809815,
      "num_tokens": 87711451.0,
      "step": 43285
    },
    {
      "entropy": 5.922895526885986,
      "epoch": 3.7830988377173815,
      "grad_norm": 1.0625,
      "learning_rate": 0.000362503832420291,
      "loss": 5.4755,
      "mean_token_accuracy": 0.18938506543636321,
      "num_tokens": 87722753.0,
      "step": 43290
    },
    {
      "entropy": 5.860615062713623,
      "epoch": 3.7835357860700864,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003624751235635473,
      "loss": 5.3968,
      "mean_token_accuracy": 0.1965814858675003,
      "num_tokens": 87732771.0,
      "step": 43295
    },
    {
      "entropy": 5.9386481761932375,
      "epoch": 3.7839727344227914,
      "grad_norm": 1.03125,
      "learning_rate": 0.000362446413028985,
      "loss": 5.4943,
      "mean_token_accuracy": 0.18603513687849044,
      "num_tokens": 87742925.0,
      "step": 43300
    },
    {
      "entropy": 5.892086553573608,
      "epoch": 3.784409682775496,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036241770081715455,
      "loss": 5.4219,
      "mean_token_accuracy": 0.1958035185933113,
      "num_tokens": 87752700.0,
      "step": 43305
    },
    {
      "entropy": 5.882717514038086,
      "epoch": 3.7848466311282007,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003623889869286067,
      "loss": 5.3788,
      "mean_token_accuracy": 0.1945655256509781,
      "num_tokens": 87764054.0,
      "step": 43310
    },
    {
      "entropy": 5.891716861724854,
      "epoch": 3.785283579480905,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003623602713638925,
      "loss": 5.3732,
      "mean_token_accuracy": 0.20025490820407868,
      "num_tokens": 87774861.0,
      "step": 43315
    },
    {
      "entropy": 5.896680212020874,
      "epoch": 3.78572052783361,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036233155412356244,
      "loss": 5.3816,
      "mean_token_accuracy": 0.19582664519548415,
      "num_tokens": 87785322.0,
      "step": 43320
    },
    {
      "entropy": 5.90290789604187,
      "epoch": 3.7861574761863146,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036230283520816744,
      "loss": 5.4057,
      "mean_token_accuracy": 0.199970942735672,
      "num_tokens": 87795357.0,
      "step": 43325
    },
    {
      "entropy": 5.853468513488769,
      "epoch": 3.7865944245390195,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003622741146182583,
      "loss": 5.4193,
      "mean_token_accuracy": 0.20048105120658874,
      "num_tokens": 87805642.0,
      "step": 43330
    },
    {
      "entropy": 5.844743394851685,
      "epoch": 3.7870313728917244,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003622453923543858,
      "loss": 5.3651,
      "mean_token_accuracy": 0.20203441083431245,
      "num_tokens": 87815700.0,
      "step": 43335
    },
    {
      "entropy": 5.911895799636841,
      "epoch": 3.787468321244429,
      "grad_norm": 1.03125,
      "learning_rate": 0.000362216668417101,
      "loss": 5.4372,
      "mean_token_accuracy": 0.18815533965826034,
      "num_tokens": 87826121.0,
      "step": 43340
    },
    {
      "entropy": 5.87885594367981,
      "epoch": 3.7879052695971334,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036218794280695494,
      "loss": 5.4014,
      "mean_token_accuracy": 0.19243661910295487,
      "num_tokens": 87838152.0,
      "step": 43345
    },
    {
      "entropy": 5.971854019165039,
      "epoch": 3.7883422179498383,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003621592155244984,
      "loss": 5.4537,
      "mean_token_accuracy": 0.19093335419893265,
      "num_tokens": 87848657.0,
      "step": 43350
    },
    {
      "entropy": 5.947968149185181,
      "epoch": 3.788779166302543,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036213048657028233,
      "loss": 5.5229,
      "mean_token_accuracy": 0.1820896789431572,
      "num_tokens": 87858603.0,
      "step": 43355
    },
    {
      "entropy": 5.911829566955566,
      "epoch": 3.7892161146552477,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036210175594485796,
      "loss": 5.4186,
      "mean_token_accuracy": 0.19829764366149902,
      "num_tokens": 87868091.0,
      "step": 43360
    },
    {
      "entropy": 5.921424627304077,
      "epoch": 3.7896530630079526,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036207302364877627,
      "loss": 5.3699,
      "mean_token_accuracy": 0.19592949151992797,
      "num_tokens": 87878978.0,
      "step": 43365
    },
    {
      "entropy": 5.909151554107666,
      "epoch": 3.790090011360657,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036204428968258836,
      "loss": 5.4487,
      "mean_token_accuracy": 0.18738884031772612,
      "num_tokens": 87888547.0,
      "step": 43370
    },
    {
      "entropy": 5.837736701965332,
      "epoch": 3.790526959713362,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036201555404684533,
      "loss": 5.3919,
      "mean_token_accuracy": 0.19842382222414018,
      "num_tokens": 87898395.0,
      "step": 43375
    },
    {
      "entropy": 5.855496406555176,
      "epoch": 3.7909639080660664,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003619868167420984,
      "loss": 5.4192,
      "mean_token_accuracy": 0.1922559157013893,
      "num_tokens": 87909127.0,
      "step": 43380
    },
    {
      "entropy": 5.983651638031006,
      "epoch": 3.7914008564187713,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036195807776889864,
      "loss": 5.4699,
      "mean_token_accuracy": 0.18771733045578004,
      "num_tokens": 87919554.0,
      "step": 43385
    },
    {
      "entropy": 5.896377897262573,
      "epoch": 3.7918378047714763,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036192933712779756,
      "loss": 5.404,
      "mean_token_accuracy": 0.1894320547580719,
      "num_tokens": 87930728.0,
      "step": 43390
    },
    {
      "entropy": 5.798383045196533,
      "epoch": 3.7922747531241807,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036190059481934614,
      "loss": 5.3174,
      "mean_token_accuracy": 0.20578703582286834,
      "num_tokens": 87939928.0,
      "step": 43395
    },
    {
      "entropy": 5.894020748138428,
      "epoch": 3.792711701476885,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003618718508440958,
      "loss": 5.4557,
      "mean_token_accuracy": 0.18863651007413865,
      "num_tokens": 87949978.0,
      "step": 43400
    },
    {
      "entropy": 5.984597635269165,
      "epoch": 3.79314864982959,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003618431052025978,
      "loss": 5.5078,
      "mean_token_accuracy": 0.18831330686807632,
      "num_tokens": 87960157.0,
      "step": 43405
    },
    {
      "entropy": 5.927347230911255,
      "epoch": 3.793585598182295,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003618143578954036,
      "loss": 5.3648,
      "mean_token_accuracy": 0.2018377438187599,
      "num_tokens": 87971302.0,
      "step": 43410
    },
    {
      "entropy": 5.898335266113281,
      "epoch": 3.7940225465349995,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003617856089230645,
      "loss": 5.335,
      "mean_token_accuracy": 0.20519006997346878,
      "num_tokens": 87980829.0,
      "step": 43415
    },
    {
      "entropy": 5.785174417495727,
      "epoch": 3.7944594948877044,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036175685828613197,
      "loss": 5.3474,
      "mean_token_accuracy": 0.19919979572296143,
      "num_tokens": 87991003.0,
      "step": 43420
    },
    {
      "entropy": 5.864071893692016,
      "epoch": 3.794896443240409,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003617281059851574,
      "loss": 5.4033,
      "mean_token_accuracy": 0.19412774741649627,
      "num_tokens": 88001475.0,
      "step": 43425
    },
    {
      "entropy": 5.96853666305542,
      "epoch": 3.795333391593114,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0003616993520206923,
      "loss": 5.5168,
      "mean_token_accuracy": 0.18684436678886412,
      "num_tokens": 88012714.0,
      "step": 43430
    },
    {
      "entropy": 5.886967134475708,
      "epoch": 3.7957703399458183,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036167059639328824,
      "loss": 5.4578,
      "mean_token_accuracy": 0.19643976539373398,
      "num_tokens": 88023195.0,
      "step": 43435
    },
    {
      "entropy": 5.905713176727295,
      "epoch": 3.796207288298523,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036164183910349676,
      "loss": 5.3842,
      "mean_token_accuracy": 0.1992896318435669,
      "num_tokens": 88032977.0,
      "step": 43440
    },
    {
      "entropy": 5.914722013473511,
      "epoch": 3.796644236651228,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003616130801518694,
      "loss": 5.385,
      "mean_token_accuracy": 0.1942313864827156,
      "num_tokens": 88041761.0,
      "step": 43445
    },
    {
      "entropy": 5.840455007553101,
      "epoch": 3.7970811850039325,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003615843195389577,
      "loss": 5.3685,
      "mean_token_accuracy": 0.20195493549108506,
      "num_tokens": 88050709.0,
      "step": 43450
    },
    {
      "entropy": 5.942530870437622,
      "epoch": 3.797518133356637,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036155555726531343,
      "loss": 5.447,
      "mean_token_accuracy": 0.19061025977134705,
      "num_tokens": 88061381.0,
      "step": 43455
    },
    {
      "entropy": 5.973727893829346,
      "epoch": 3.797955081709342,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036152679333148816,
      "loss": 5.4476,
      "mean_token_accuracy": 0.19355710744857788,
      "num_tokens": 88070926.0,
      "step": 43460
    },
    {
      "entropy": 6.005800056457519,
      "epoch": 3.798392030062047,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00036149802773803374,
      "loss": 5.5136,
      "mean_token_accuracy": 0.18990040123462676,
      "num_tokens": 88081240.0,
      "step": 43465
    },
    {
      "entropy": 5.992464160919189,
      "epoch": 3.7988289784147513,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003614692604855018,
      "loss": 5.5406,
      "mean_token_accuracy": 0.18083547204732894,
      "num_tokens": 88092623.0,
      "step": 43470
    },
    {
      "entropy": 5.901081466674805,
      "epoch": 3.7992659267674562,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003614404915744441,
      "loss": 5.3871,
      "mean_token_accuracy": 0.19527530521154404,
      "num_tokens": 88102532.0,
      "step": 43475
    },
    {
      "entropy": 5.952623128890991,
      "epoch": 3.7997028751201607,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036141172100541245,
      "loss": 5.5169,
      "mean_token_accuracy": 0.18413096219301223,
      "num_tokens": 88112210.0,
      "step": 43480
    },
    {
      "entropy": 5.954039144515991,
      "epoch": 3.8001398234728656,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00036138294877895876,
      "loss": 5.4288,
      "mean_token_accuracy": 0.19421872198581697,
      "num_tokens": 88121800.0,
      "step": 43485
    },
    {
      "entropy": 5.915898418426513,
      "epoch": 3.80057677182557,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003613541748956349,
      "loss": 5.4066,
      "mean_token_accuracy": 0.19343435764312744,
      "num_tokens": 88131820.0,
      "step": 43490
    },
    {
      "entropy": 5.919663000106811,
      "epoch": 3.801013720178275,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003613253993559926,
      "loss": 5.4507,
      "mean_token_accuracy": 0.19484248161315917,
      "num_tokens": 88141799.0,
      "step": 43495
    },
    {
      "entropy": 5.89514570236206,
      "epoch": 3.80145066853098,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036129662216058395,
      "loss": 5.3797,
      "mean_token_accuracy": 0.19923484176397324,
      "num_tokens": 88151860.0,
      "step": 43500
    },
    {
      "entropy": 5.874196910858155,
      "epoch": 3.8018876168836844,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003612678433099609,
      "loss": 5.3519,
      "mean_token_accuracy": 0.19873390048742295,
      "num_tokens": 88160580.0,
      "step": 43505
    },
    {
      "entropy": 5.877199697494507,
      "epoch": 3.802324565236389,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036123906280467543,
      "loss": 5.4532,
      "mean_token_accuracy": 0.19226219058036803,
      "num_tokens": 88170140.0,
      "step": 43510
    },
    {
      "entropy": 5.9152257442474365,
      "epoch": 3.8027615135890938,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003612102806452795,
      "loss": 5.4918,
      "mean_token_accuracy": 0.19500266760587692,
      "num_tokens": 88180288.0,
      "step": 43515
    },
    {
      "entropy": 5.905852556228638,
      "epoch": 3.8031984619417987,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003611814968323252,
      "loss": 5.4187,
      "mean_token_accuracy": 0.19348752051591872,
      "num_tokens": 88190046.0,
      "step": 43520
    },
    {
      "entropy": 5.897337245941162,
      "epoch": 3.803635410294503,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003611527113663647,
      "loss": 5.336,
      "mean_token_accuracy": 0.20172018110752105,
      "num_tokens": 88199321.0,
      "step": 43525
    },
    {
      "entropy": 5.879335880279541,
      "epoch": 3.804072358647208,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036112392424794996,
      "loss": 5.4589,
      "mean_token_accuracy": 0.18430978506803514,
      "num_tokens": 88209277.0,
      "step": 43530
    },
    {
      "entropy": 5.881268167495728,
      "epoch": 3.8045093069999125,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003610951354776333,
      "loss": 5.484,
      "mean_token_accuracy": 0.1867685005068779,
      "num_tokens": 88219664.0,
      "step": 43535
    },
    {
      "entropy": 5.9292457580566404,
      "epoch": 3.8049462553526174,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00036106634505596693,
      "loss": 5.4458,
      "mean_token_accuracy": 0.19215033948421478,
      "num_tokens": 88230058.0,
      "step": 43540
    },
    {
      "entropy": 5.916960716247559,
      "epoch": 3.805383203705322,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00036103755298350277,
      "loss": 5.3802,
      "mean_token_accuracy": 0.19747157841920854,
      "num_tokens": 88240861.0,
      "step": 43545
    },
    {
      "entropy": 5.94283275604248,
      "epoch": 3.805820152058027,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036100875926079335,
      "loss": 5.4583,
      "mean_token_accuracy": 0.18547501862049104,
      "num_tokens": 88250833.0,
      "step": 43550
    },
    {
      "entropy": 5.929120349884033,
      "epoch": 3.8062571004107313,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003609799638883908,
      "loss": 5.3686,
      "mean_token_accuracy": 0.1926803931593895,
      "num_tokens": 88260209.0,
      "step": 43555
    },
    {
      "entropy": 5.9820168018341064,
      "epoch": 3.806694048763436,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003609511668668476,
      "loss": 5.4738,
      "mean_token_accuracy": 0.183423013985157,
      "num_tokens": 88269961.0,
      "step": 43560
    },
    {
      "entropy": 5.869065237045288,
      "epoch": 3.8071309971161407,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036092236819671586,
      "loss": 5.4368,
      "mean_token_accuracy": 0.1895422101020813,
      "num_tokens": 88279916.0,
      "step": 43565
    },
    {
      "entropy": 5.891442584991455,
      "epoch": 3.8075679454688456,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036089356787854816,
      "loss": 5.4812,
      "mean_token_accuracy": 0.1933950513601303,
      "num_tokens": 88290459.0,
      "step": 43570
    },
    {
      "entropy": 5.92170786857605,
      "epoch": 3.8080048938215505,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003608647659128968,
      "loss": 5.4062,
      "mean_token_accuracy": 0.19604710340499878,
      "num_tokens": 88300536.0,
      "step": 43575
    },
    {
      "entropy": 5.977129030227661,
      "epoch": 3.808441842174255,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036083596230031425,
      "loss": 5.4744,
      "mean_token_accuracy": 0.19229586720466613,
      "num_tokens": 88310654.0,
      "step": 43580
    },
    {
      "entropy": 5.873619556427002,
      "epoch": 3.80887879052696,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003608071570413529,
      "loss": 5.3827,
      "mean_token_accuracy": 0.20626062601804734,
      "num_tokens": 88319831.0,
      "step": 43585
    },
    {
      "entropy": 5.961403751373291,
      "epoch": 3.8093157388796643,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003607783501365653,
      "loss": 5.566,
      "mean_token_accuracy": 0.1782939001917839,
      "num_tokens": 88330931.0,
      "step": 43590
    },
    {
      "entropy": 5.948372936248779,
      "epoch": 3.8097526872323693,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003607495415865041,
      "loss": 5.42,
      "mean_token_accuracy": 0.19727415591478348,
      "num_tokens": 88341699.0,
      "step": 43595
    },
    {
      "entropy": 5.896967029571533,
      "epoch": 3.8101896355850737,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036072073139172174,
      "loss": 5.4446,
      "mean_token_accuracy": 0.19495040327310562,
      "num_tokens": 88351375.0,
      "step": 43600
    },
    {
      "entropy": 5.887977600097656,
      "epoch": 3.8106265839377786,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036069191955277085,
      "loss": 5.4382,
      "mean_token_accuracy": 0.1954813241958618,
      "num_tokens": 88361865.0,
      "step": 43605
    },
    {
      "entropy": 5.928629779815674,
      "epoch": 3.811063532290483,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000360663106070204,
      "loss": 5.3923,
      "mean_token_accuracy": 0.19607558697462082,
      "num_tokens": 88371276.0,
      "step": 43610
    },
    {
      "entropy": 5.913670349121094,
      "epoch": 3.811500480643188,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003606342909445739,
      "loss": 5.3844,
      "mean_token_accuracy": 0.19770100861787795,
      "num_tokens": 88382098.0,
      "step": 43615
    },
    {
      "entropy": 5.982509326934815,
      "epoch": 3.8119374289958925,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003606054741764332,
      "loss": 5.3972,
      "mean_token_accuracy": 0.19875625073909758,
      "num_tokens": 88391488.0,
      "step": 43620
    },
    {
      "entropy": 5.933840322494507,
      "epoch": 3.8123743773485974,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036057665576633466,
      "loss": 5.427,
      "mean_token_accuracy": 0.19782939851284026,
      "num_tokens": 88401636.0,
      "step": 43625
    },
    {
      "entropy": 5.80811333656311,
      "epoch": 3.8128113257013023,
      "grad_norm": 1.203125,
      "learning_rate": 0.000360547835714831,
      "loss": 5.3602,
      "mean_token_accuracy": 0.20167906135320662,
      "num_tokens": 88412434.0,
      "step": 43630
    },
    {
      "entropy": 5.901935291290283,
      "epoch": 3.813248274054007,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003605190140224751,
      "loss": 5.4475,
      "mean_token_accuracy": 0.18439230024814607,
      "num_tokens": 88421995.0,
      "step": 43635
    },
    {
      "entropy": 5.898743581771851,
      "epoch": 3.8136852224067113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036049019068981966,
      "loss": 5.3453,
      "mean_token_accuracy": 0.20850538611412048,
      "num_tokens": 88431691.0,
      "step": 43640
    },
    {
      "entropy": 5.868617534637451,
      "epoch": 3.814122170759416,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003604613657174175,
      "loss": 5.4182,
      "mean_token_accuracy": 0.19855592548847198,
      "num_tokens": 88441284.0,
      "step": 43645
    },
    {
      "entropy": 5.831306028366089,
      "epoch": 3.814559119112121,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003604325391058216,
      "loss": 5.413,
      "mean_token_accuracy": 0.19837024211883544,
      "num_tokens": 88450674.0,
      "step": 43650
    },
    {
      "entropy": 5.938483190536499,
      "epoch": 3.8149960674648256,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003604037108555849,
      "loss": 5.4614,
      "mean_token_accuracy": 0.1860383003950119,
      "num_tokens": 88461082.0,
      "step": 43655
    },
    {
      "entropy": 5.965808963775634,
      "epoch": 3.8154330158175305,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003603748809672602,
      "loss": 5.5071,
      "mean_token_accuracy": 0.1821983814239502,
      "num_tokens": 88471229.0,
      "step": 43660
    },
    {
      "entropy": 5.979167699813843,
      "epoch": 3.815869964170235,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036034604944140057,
      "loss": 5.4969,
      "mean_token_accuracy": 0.18615787625312805,
      "num_tokens": 88481269.0,
      "step": 43665
    },
    {
      "entropy": 5.9335733413696286,
      "epoch": 3.81630691252294,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00036031721627855905,
      "loss": 5.3546,
      "mean_token_accuracy": 0.19928823709487914,
      "num_tokens": 88490754.0,
      "step": 43670
    },
    {
      "entropy": 5.838159132003784,
      "epoch": 3.8167438608756443,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00036028838147928854,
      "loss": 5.4077,
      "mean_token_accuracy": 0.20193978101015092,
      "num_tokens": 88500487.0,
      "step": 43675
    },
    {
      "entropy": 5.866075897216797,
      "epoch": 3.8171808092283492,
      "grad_norm": 1.0,
      "learning_rate": 0.00036025954504414223,
      "loss": 5.3505,
      "mean_token_accuracy": 0.20295317769050597,
      "num_tokens": 88510857.0,
      "step": 43680
    },
    {
      "entropy": 5.913027763366699,
      "epoch": 3.817617757581054,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036023070697367314,
      "loss": 5.4179,
      "mean_token_accuracy": 0.19526449590921402,
      "num_tokens": 88522551.0,
      "step": 43685
    },
    {
      "entropy": 5.928721904754639,
      "epoch": 3.8180547059337586,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036020186726843447,
      "loss": 5.5276,
      "mean_token_accuracy": 0.19067790359258652,
      "num_tokens": 88534069.0,
      "step": 43690
    },
    {
      "entropy": 5.904715442657471,
      "epoch": 3.818491654286463,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003601730259289793,
      "loss": 5.3284,
      "mean_token_accuracy": 0.1983220174908638,
      "num_tokens": 88544912.0,
      "step": 43695
    },
    {
      "entropy": 5.800616216659546,
      "epoch": 3.818928602639168,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003601441829558609,
      "loss": 5.3116,
      "mean_token_accuracy": 0.2168743208050728,
      "num_tokens": 88554445.0,
      "step": 43700
    },
    {
      "entropy": 5.859938287734986,
      "epoch": 3.819365550991873,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003601153383496324,
      "loss": 5.4304,
      "mean_token_accuracy": 0.19463014155626296,
      "num_tokens": 88563589.0,
      "step": 43705
    },
    {
      "entropy": 5.933822059631348,
      "epoch": 3.8198024993445774,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036008649211084716,
      "loss": 5.5117,
      "mean_token_accuracy": 0.18995406478643417,
      "num_tokens": 88574072.0,
      "step": 43710
    },
    {
      "entropy": 5.909377479553223,
      "epoch": 3.8202394476972823,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003600576442400584,
      "loss": 5.4248,
      "mean_token_accuracy": 0.19159702211618423,
      "num_tokens": 88584225.0,
      "step": 43715
    },
    {
      "entropy": 5.911449813842774,
      "epoch": 3.8206763960499868,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003600287947378195,
      "loss": 5.3558,
      "mean_token_accuracy": 0.2016768604516983,
      "num_tokens": 88594364.0,
      "step": 43720
    },
    {
      "entropy": 5.962161922454834,
      "epoch": 3.8211133444026917,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003599999436046837,
      "loss": 5.4824,
      "mean_token_accuracy": 0.18723613619804383,
      "num_tokens": 88605678.0,
      "step": 43725
    },
    {
      "entropy": 5.966552019119263,
      "epoch": 3.821550292755396,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00035997109084120446,
      "loss": 5.4008,
      "mean_token_accuracy": 0.19456966370344161,
      "num_tokens": 88615824.0,
      "step": 43730
    },
    {
      "entropy": 5.900209140777588,
      "epoch": 3.821987241108101,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035994223644793516,
      "loss": 5.44,
      "mean_token_accuracy": 0.19807826727628708,
      "num_tokens": 88625939.0,
      "step": 43735
    },
    {
      "entropy": 5.904083204269409,
      "epoch": 3.822424189460806,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003599133804254293,
      "loss": 5.385,
      "mean_token_accuracy": 0.1960999310016632,
      "num_tokens": 88635453.0,
      "step": 43740
    },
    {
      "entropy": 5.909459209442138,
      "epoch": 3.8228611378135104,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003598845227742402,
      "loss": 5.3382,
      "mean_token_accuracy": 0.19729672074317933,
      "num_tokens": 88646190.0,
      "step": 43745
    },
    {
      "entropy": 5.859546184539795,
      "epoch": 3.823298086166215,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035985566349492156,
      "loss": 5.4048,
      "mean_token_accuracy": 0.1969807267189026,
      "num_tokens": 88655957.0,
      "step": 43750
    },
    {
      "entropy": 5.940549707412719,
      "epoch": 3.82373503451892,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035982680258802685,
      "loss": 5.391,
      "mean_token_accuracy": 0.19436269104480744,
      "num_tokens": 88665884.0,
      "step": 43755
    },
    {
      "entropy": 5.8976654529571535,
      "epoch": 3.8241719828716247,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003597979400541096,
      "loss": 5.3981,
      "mean_token_accuracy": 0.1972922444343567,
      "num_tokens": 88677226.0,
      "step": 43760
    },
    {
      "entropy": 5.88382477760315,
      "epoch": 3.824608931224329,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003597690758937234,
      "loss": 5.3912,
      "mean_token_accuracy": 0.20261323899030687,
      "num_tokens": 88687447.0,
      "step": 43765
    },
    {
      "entropy": 5.877028512954712,
      "epoch": 3.825045879577034,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00035974021010742196,
      "loss": 5.3887,
      "mean_token_accuracy": 0.1952529013156891,
      "num_tokens": 88696540.0,
      "step": 43770
    },
    {
      "entropy": 5.932751703262329,
      "epoch": 3.8254828279297386,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00035971134269575884,
      "loss": 5.4018,
      "mean_token_accuracy": 0.19180818051099777,
      "num_tokens": 88707945.0,
      "step": 43775
    },
    {
      "entropy": 6.009298372268677,
      "epoch": 3.8259197762824435,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035968247365928774,
      "loss": 5.4985,
      "mean_token_accuracy": 0.18754257261753082,
      "num_tokens": 88717398.0,
      "step": 43780
    },
    {
      "entropy": 5.93410587310791,
      "epoch": 3.826356724635148,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035965360299856247,
      "loss": 5.3618,
      "mean_token_accuracy": 0.19156017154455185,
      "num_tokens": 88727030.0,
      "step": 43785
    },
    {
      "entropy": 5.879191064834595,
      "epoch": 3.826793672987853,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003596247307141368,
      "loss": 5.3447,
      "mean_token_accuracy": 0.19981604516506196,
      "num_tokens": 88737413.0,
      "step": 43790
    },
    {
      "entropy": 5.910060548782349,
      "epoch": 3.827230621340558,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003595958568065644,
      "loss": 5.4175,
      "mean_token_accuracy": 0.19691664576530457,
      "num_tokens": 88747577.0,
      "step": 43795
    },
    {
      "entropy": 5.906639623641968,
      "epoch": 3.8276675696932623,
      "grad_norm": 1.125,
      "learning_rate": 0.0003595669812763991,
      "loss": 5.4022,
      "mean_token_accuracy": 0.18878866583108903,
      "num_tokens": 88757023.0,
      "step": 43800
    },
    {
      "entropy": 5.857735919952392,
      "epoch": 3.8281045180459667,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003595381041241948,
      "loss": 5.3842,
      "mean_token_accuracy": 0.20242370069026946,
      "num_tokens": 88767200.0,
      "step": 43805
    },
    {
      "entropy": 5.859346342086792,
      "epoch": 3.8285414663986717,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035950922535050534,
      "loss": 5.438,
      "mean_token_accuracy": 0.19636034369468688,
      "num_tokens": 88777448.0,
      "step": 43810
    },
    {
      "entropy": 5.881763648986817,
      "epoch": 3.8289784147513766,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035948034495588473,
      "loss": 5.334,
      "mean_token_accuracy": 0.20539896935224533,
      "num_tokens": 88787477.0,
      "step": 43815
    },
    {
      "entropy": 5.966263723373413,
      "epoch": 3.829415363104081,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003594514629408868,
      "loss": 5.4734,
      "mean_token_accuracy": 0.19006636142730712,
      "num_tokens": 88796971.0,
      "step": 43820
    },
    {
      "entropy": 5.844316673278809,
      "epoch": 3.829852311456786,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003594225793060656,
      "loss": 5.3772,
      "mean_token_accuracy": 0.2016273096203804,
      "num_tokens": 88807960.0,
      "step": 43825
    },
    {
      "entropy": 5.920288801193237,
      "epoch": 3.8302892598094904,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000359393694051975,
      "loss": 5.5101,
      "mean_token_accuracy": 0.18387694507837296,
      "num_tokens": 88818874.0,
      "step": 43830
    },
    {
      "entropy": 5.935523414611817,
      "epoch": 3.8307262081621953,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003593648071791692,
      "loss": 5.3943,
      "mean_token_accuracy": 0.1930634304881096,
      "num_tokens": 88829216.0,
      "step": 43835
    },
    {
      "entropy": 5.992374086380005,
      "epoch": 3.8311631565149,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035933591868820213,
      "loss": 5.4742,
      "mean_token_accuracy": 0.18819574862718583,
      "num_tokens": 88839261.0,
      "step": 43840
    },
    {
      "entropy": 5.907525157928466,
      "epoch": 3.8316001048676047,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035930702857962804,
      "loss": 5.3712,
      "mean_token_accuracy": 0.2023780733346939,
      "num_tokens": 88849180.0,
      "step": 43845
    },
    {
      "entropy": 5.810440015792847,
      "epoch": 3.8320370532203096,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003592781368540009,
      "loss": 5.4616,
      "mean_token_accuracy": 0.1978433236479759,
      "num_tokens": 88859696.0,
      "step": 43850
    },
    {
      "entropy": 6.01371922492981,
      "epoch": 3.832474001573014,
      "grad_norm": 1.0,
      "learning_rate": 0.00035924924351187496,
      "loss": 5.4816,
      "mean_token_accuracy": 0.19249367862939834,
      "num_tokens": 88869932.0,
      "step": 43855
    },
    {
      "entropy": 5.928285598754883,
      "epoch": 3.8329109499257186,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035922034855380436,
      "loss": 5.4298,
      "mean_token_accuracy": 0.19881137907505037,
      "num_tokens": 88880946.0,
      "step": 43860
    },
    {
      "entropy": 5.896613359451294,
      "epoch": 3.8333478982784235,
      "grad_norm": 1.0,
      "learning_rate": 0.00035919145198034337,
      "loss": 5.4502,
      "mean_token_accuracy": 0.1946649894118309,
      "num_tokens": 88890409.0,
      "step": 43865
    },
    {
      "entropy": 5.933578014373779,
      "epoch": 3.8337848466311284,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035916255379204613,
      "loss": 5.5317,
      "mean_token_accuracy": 0.18428455740213395,
      "num_tokens": 88900872.0,
      "step": 43870
    },
    {
      "entropy": 5.959998607635498,
      "epoch": 3.834221794983833,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003591336539894671,
      "loss": 5.369,
      "mean_token_accuracy": 0.19661230742931365,
      "num_tokens": 88909999.0,
      "step": 43875
    },
    {
      "entropy": 5.868185758590698,
      "epoch": 3.834658743336538,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003591047525731605,
      "loss": 5.366,
      "mean_token_accuracy": 0.20150145292282104,
      "num_tokens": 88920219.0,
      "step": 43880
    },
    {
      "entropy": 5.876050615310669,
      "epoch": 3.8350956916892422,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035907584954368067,
      "loss": 5.467,
      "mean_token_accuracy": 0.1918674439191818,
      "num_tokens": 88930115.0,
      "step": 43885
    },
    {
      "entropy": 5.969121980667114,
      "epoch": 3.835532640041947,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035904694490158196,
      "loss": 5.4495,
      "mean_token_accuracy": 0.19210486114025116,
      "num_tokens": 88940114.0,
      "step": 43890
    },
    {
      "entropy": 5.965101432800293,
      "epoch": 3.8359695883946516,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003590180386474188,
      "loss": 5.5322,
      "mean_token_accuracy": 0.18212701827287675,
      "num_tokens": 88951955.0,
      "step": 43895
    },
    {
      "entropy": 5.892612504959106,
      "epoch": 3.8364065367473565,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003589891307817457,
      "loss": 5.3531,
      "mean_token_accuracy": 0.19576888382434846,
      "num_tokens": 88961677.0,
      "step": 43900
    },
    {
      "entropy": 5.925663566589355,
      "epoch": 3.836843485100061,
      "grad_norm": 1.0,
      "learning_rate": 0.000358960221305117,
      "loss": 5.401,
      "mean_token_accuracy": 0.1931746318936348,
      "num_tokens": 88971888.0,
      "step": 43905
    },
    {
      "entropy": 5.9529585361480715,
      "epoch": 3.837280433452766,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003589313102180873,
      "loss": 5.4778,
      "mean_token_accuracy": 0.19147867262363433,
      "num_tokens": 88980681.0,
      "step": 43910
    },
    {
      "entropy": 5.893923759460449,
      "epoch": 3.8377173818054704,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003589023975212111,
      "loss": 5.4898,
      "mean_token_accuracy": 0.1911829099059105,
      "num_tokens": 88991410.0,
      "step": 43915
    },
    {
      "entropy": 5.860382080078125,
      "epoch": 3.8381543301581753,
      "grad_norm": 0.953125,
      "learning_rate": 0.00035887348321504285,
      "loss": 5.3521,
      "mean_token_accuracy": 0.20334338396787643,
      "num_tokens": 89001743.0,
      "step": 43920
    },
    {
      "entropy": 5.840610074996948,
      "epoch": 3.8385912785108802,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035884456730013735,
      "loss": 5.3182,
      "mean_token_accuracy": 0.20434251576662063,
      "num_tokens": 89011138.0,
      "step": 43925
    },
    {
      "entropy": 5.893445682525635,
      "epoch": 3.8390282268635847,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035881564977704913,
      "loss": 5.4272,
      "mean_token_accuracy": 0.19393557608127593,
      "num_tokens": 89021282.0,
      "step": 43930
    },
    {
      "entropy": 5.903754186630249,
      "epoch": 3.839465175216289,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035878673064633276,
      "loss": 5.4738,
      "mean_token_accuracy": 0.1921437293291092,
      "num_tokens": 89032656.0,
      "step": 43935
    },
    {
      "entropy": 5.98931622505188,
      "epoch": 3.839902123568994,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035875780990854303,
      "loss": 5.4947,
      "mean_token_accuracy": 0.18393321186304093,
      "num_tokens": 89042727.0,
      "step": 43940
    },
    {
      "entropy": 5.895911121368409,
      "epoch": 3.840339071921699,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035872888756423456,
      "loss": 5.4179,
      "mean_token_accuracy": 0.1910663977265358,
      "num_tokens": 89052111.0,
      "step": 43945
    },
    {
      "entropy": 5.963227462768555,
      "epoch": 3.8407760202744035,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003586999636139622,
      "loss": 5.4611,
      "mean_token_accuracy": 0.18995196521282195,
      "num_tokens": 89062423.0,
      "step": 43950
    },
    {
      "entropy": 5.921625661849975,
      "epoch": 3.8412129686271084,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035867103805828067,
      "loss": 5.4393,
      "mean_token_accuracy": 0.19732069224119186,
      "num_tokens": 89072457.0,
      "step": 43955
    },
    {
      "entropy": 5.933097076416016,
      "epoch": 3.841649916979813,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035864211089774487,
      "loss": 5.4355,
      "mean_token_accuracy": 0.1910003662109375,
      "num_tokens": 89081919.0,
      "step": 43960
    },
    {
      "entropy": 5.884066963195801,
      "epoch": 3.8420868653325178,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003586131821329094,
      "loss": 5.3945,
      "mean_token_accuracy": 0.19405332654714585,
      "num_tokens": 89092177.0,
      "step": 43965
    },
    {
      "entropy": 5.900504636764526,
      "epoch": 3.8425238136852222,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003585842517643294,
      "loss": 5.4408,
      "mean_token_accuracy": 0.1978064224123955,
      "num_tokens": 89102721.0,
      "step": 43970
    },
    {
      "entropy": 5.948583650588989,
      "epoch": 3.842960762037927,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035855531979255955,
      "loss": 5.3818,
      "mean_token_accuracy": 0.20267827808856964,
      "num_tokens": 89111841.0,
      "step": 43975
    },
    {
      "entropy": 5.892032337188721,
      "epoch": 3.843397710390632,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035852638621815497,
      "loss": 5.4021,
      "mean_token_accuracy": 0.19322735220193862,
      "num_tokens": 89122551.0,
      "step": 43980
    },
    {
      "entropy": 5.953138303756714,
      "epoch": 3.8438346587433365,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003584974510416706,
      "loss": 5.4968,
      "mean_token_accuracy": 0.19272820204496383,
      "num_tokens": 89133388.0,
      "step": 43985
    },
    {
      "entropy": 5.984113311767578,
      "epoch": 3.844271607096041,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00035846851426366125,
      "loss": 5.4401,
      "mean_token_accuracy": 0.19388158321380616,
      "num_tokens": 89144435.0,
      "step": 43990
    },
    {
      "entropy": 5.917956733703614,
      "epoch": 3.844708555448746,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035843957588468207,
      "loss": 5.4328,
      "mean_token_accuracy": 0.19866345822811127,
      "num_tokens": 89154509.0,
      "step": 43995
    },
    {
      "entropy": 5.931872034072876,
      "epoch": 3.845145503801451,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003584106359052881,
      "loss": 5.5036,
      "mean_token_accuracy": 0.19154937118291854,
      "num_tokens": 89165352.0,
      "step": 44000
    },
    {
      "entropy": 5.928022575378418,
      "epoch": 3.8455824521541553,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003583816943260345,
      "loss": 5.41,
      "mean_token_accuracy": 0.19140318483114244,
      "num_tokens": 89175185.0,
      "step": 44005
    },
    {
      "entropy": 5.935611343383789,
      "epoch": 3.84601940050686,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003583527511474763,
      "loss": 5.3796,
      "mean_token_accuracy": 0.19612576812505722,
      "num_tokens": 89183864.0,
      "step": 44010
    },
    {
      "entropy": 5.916254329681396,
      "epoch": 3.8464563488595647,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035832380637016846,
      "loss": 5.4569,
      "mean_token_accuracy": 0.19166044443845748,
      "num_tokens": 89194011.0,
      "step": 44015
    },
    {
      "entropy": 5.866076135635376,
      "epoch": 3.8468932972122696,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035829485999466657,
      "loss": 5.4056,
      "mean_token_accuracy": 0.1961204305291176,
      "num_tokens": 89204213.0,
      "step": 44020
    },
    {
      "entropy": 5.860246658325195,
      "epoch": 3.847330245564974,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003582659120215256,
      "loss": 5.3554,
      "mean_token_accuracy": 0.20389500707387925,
      "num_tokens": 89214160.0,
      "step": 44025
    },
    {
      "entropy": 5.850160217285156,
      "epoch": 3.847767193917679,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035823696245130073,
      "loss": 5.3635,
      "mean_token_accuracy": 0.19781050384044646,
      "num_tokens": 89223399.0,
      "step": 44030
    },
    {
      "entropy": 5.919699573516846,
      "epoch": 3.848204142270384,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035820801128454735,
      "loss": 5.3995,
      "mean_token_accuracy": 0.19895650297403336,
      "num_tokens": 89232703.0,
      "step": 44035
    },
    {
      "entropy": 5.949589681625366,
      "epoch": 3.8486410906230883,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003581790585218207,
      "loss": 5.4632,
      "mean_token_accuracy": 0.1892237901687622,
      "num_tokens": 89243302.0,
      "step": 44040
    },
    {
      "entropy": 5.9215504169464115,
      "epoch": 3.849078038975793,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003581501041636761,
      "loss": 5.4481,
      "mean_token_accuracy": 0.1902448281645775,
      "num_tokens": 89253340.0,
      "step": 44045
    },
    {
      "entropy": 5.9023181915283205,
      "epoch": 3.8495149873284977,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035812114821066903,
      "loss": 5.3337,
      "mean_token_accuracy": 0.20005703419446946,
      "num_tokens": 89263362.0,
      "step": 44050
    },
    {
      "entropy": 5.828938293457031,
      "epoch": 3.8499519356812026,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003580921906633547,
      "loss": 5.3141,
      "mean_token_accuracy": 0.2066015362739563,
      "num_tokens": 89275386.0,
      "step": 44055
    },
    {
      "entropy": 5.929120016098023,
      "epoch": 3.850388884033907,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003580632315222886,
      "loss": 5.4475,
      "mean_token_accuracy": 0.18868503272533416,
      "num_tokens": 89285325.0,
      "step": 44060
    },
    {
      "entropy": 5.887651157379151,
      "epoch": 3.850825832386612,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003580342707880262,
      "loss": 5.3385,
      "mean_token_accuracy": 0.19747605174779892,
      "num_tokens": 89295850.0,
      "step": 44065
    },
    {
      "entropy": 5.829508447647095,
      "epoch": 3.8512627807393165,
      "grad_norm": 1.03125,
      "learning_rate": 0.000358005308461123,
      "loss": 5.3381,
      "mean_token_accuracy": 0.20930792540311813,
      "num_tokens": 89305403.0,
      "step": 44070
    },
    {
      "entropy": 5.8227519512176515,
      "epoch": 3.8516997290920214,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00035797634454213454,
      "loss": 5.2784,
      "mean_token_accuracy": 0.2056650251150131,
      "num_tokens": 89314551.0,
      "step": 44075
    },
    {
      "entropy": 5.913565826416016,
      "epoch": 3.852136677444726,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003579473790316162,
      "loss": 5.443,
      "mean_token_accuracy": 0.19597991853952407,
      "num_tokens": 89323872.0,
      "step": 44080
    },
    {
      "entropy": 5.916910886764526,
      "epoch": 3.852573625797431,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035791841193012373,
      "loss": 5.3994,
      "mean_token_accuracy": 0.20075231939554214,
      "num_tokens": 89335114.0,
      "step": 44085
    },
    {
      "entropy": 5.915490770339966,
      "epoch": 3.8530105741501357,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003578894432382127,
      "loss": 5.3622,
      "mean_token_accuracy": 0.19695742130279542,
      "num_tokens": 89344382.0,
      "step": 44090
    },
    {
      "entropy": 5.895838069915771,
      "epoch": 3.85344752250284,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003578604729564387,
      "loss": 5.433,
      "mean_token_accuracy": 0.18854695856571196,
      "num_tokens": 89355327.0,
      "step": 44095
    },
    {
      "entropy": 5.924550294876099,
      "epoch": 3.8538844708555446,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035783150108535737,
      "loss": 5.4133,
      "mean_token_accuracy": 0.20365233570337296,
      "num_tokens": 89365361.0,
      "step": 44100
    },
    {
      "entropy": 5.9336809635162355,
      "epoch": 3.8543214192082496,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035780252762552444,
      "loss": 5.4363,
      "mean_token_accuracy": 0.19488749504089356,
      "num_tokens": 89375097.0,
      "step": 44105
    },
    {
      "entropy": 5.900425672531128,
      "epoch": 3.8547583675609545,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035777355257749573,
      "loss": 5.3541,
      "mean_token_accuracy": 0.20245643109083175,
      "num_tokens": 89384934.0,
      "step": 44110
    },
    {
      "entropy": 5.847417879104614,
      "epoch": 3.855195315913659,
      "grad_norm": 1.0,
      "learning_rate": 0.00035774457594182685,
      "loss": 5.2979,
      "mean_token_accuracy": 0.21017009466886521,
      "num_tokens": 89395493.0,
      "step": 44115
    },
    {
      "entropy": 5.805315256118774,
      "epoch": 3.855632264266364,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003577155977190736,
      "loss": 5.352,
      "mean_token_accuracy": 0.20991660952568053,
      "num_tokens": 89405057.0,
      "step": 44120
    },
    {
      "entropy": 5.828327274322509,
      "epoch": 3.8560692126190683,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00035768661790979193,
      "loss": 5.3465,
      "mean_token_accuracy": 0.20296231061220169,
      "num_tokens": 89415328.0,
      "step": 44125
    },
    {
      "entropy": 5.872111892700195,
      "epoch": 3.8565061609717732,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003576576365145376,
      "loss": 5.3862,
      "mean_token_accuracy": 0.20191487520933152,
      "num_tokens": 89425138.0,
      "step": 44130
    },
    {
      "entropy": 5.8834113597869875,
      "epoch": 3.8569431093244777,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035762865353386637,
      "loss": 5.3745,
      "mean_token_accuracy": 0.19966418743133546,
      "num_tokens": 89434972.0,
      "step": 44135
    },
    {
      "entropy": 5.858321619033814,
      "epoch": 3.8573800576771826,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003575996689683344,
      "loss": 5.4315,
      "mean_token_accuracy": 0.19627464413642884,
      "num_tokens": 89445661.0,
      "step": 44140
    },
    {
      "entropy": 5.88446536064148,
      "epoch": 3.8578170060298875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035757068281849747,
      "loss": 5.463,
      "mean_token_accuracy": 0.19144458323717117,
      "num_tokens": 89455970.0,
      "step": 44145
    },
    {
      "entropy": 5.921284151077271,
      "epoch": 3.858253954382592,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003575416950849115,
      "loss": 5.4315,
      "mean_token_accuracy": 0.19433552175760269,
      "num_tokens": 89466076.0,
      "step": 44150
    },
    {
      "entropy": 5.929720401763916,
      "epoch": 3.8586909027352965,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003575127057681326,
      "loss": 5.4043,
      "mean_token_accuracy": 0.19791417419910431,
      "num_tokens": 89475525.0,
      "step": 44155
    },
    {
      "entropy": 5.906278657913208,
      "epoch": 3.8591278510880014,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003574837148687168,
      "loss": 5.4691,
      "mean_token_accuracy": 0.19297334998846055,
      "num_tokens": 89485913.0,
      "step": 44160
    },
    {
      "entropy": 5.885423469543457,
      "epoch": 3.8595647994407063,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035745472238722013,
      "loss": 5.4029,
      "mean_token_accuracy": 0.1910183236002922,
      "num_tokens": 89496321.0,
      "step": 44165
    },
    {
      "entropy": 5.900277996063233,
      "epoch": 3.8600017477934108,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00035742572832419867,
      "loss": 5.4191,
      "mean_token_accuracy": 0.19573243111371993,
      "num_tokens": 89506335.0,
      "step": 44170
    },
    {
      "entropy": 5.91270170211792,
      "epoch": 3.8604386961461157,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035739673268020856,
      "loss": 5.3848,
      "mean_token_accuracy": 0.19344454556703566,
      "num_tokens": 89515683.0,
      "step": 44175
    },
    {
      "entropy": 5.83485689163208,
      "epoch": 3.86087564449882,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035736773545580594,
      "loss": 5.4356,
      "mean_token_accuracy": 0.19792935848236085,
      "num_tokens": 89525797.0,
      "step": 44180
    },
    {
      "entropy": 5.944460201263428,
      "epoch": 3.861312592851525,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035733873665154696,
      "loss": 5.4484,
      "mean_token_accuracy": 0.18916877806186677,
      "num_tokens": 89535312.0,
      "step": 44185
    },
    {
      "entropy": 5.774892568588257,
      "epoch": 3.8617495412042295,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003573097362679879,
      "loss": 5.267,
      "mean_token_accuracy": 0.21586034595966339,
      "num_tokens": 89545498.0,
      "step": 44190
    },
    {
      "entropy": 5.906522798538208,
      "epoch": 3.8621864895569344,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003572807343056849,
      "loss": 5.3186,
      "mean_token_accuracy": 0.20077519565820695,
      "num_tokens": 89555421.0,
      "step": 44195
    },
    {
      "entropy": 5.933138418197632,
      "epoch": 3.8626234379096394,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003572517307651944,
      "loss": 5.3936,
      "mean_token_accuracy": 0.1971675470471382,
      "num_tokens": 89565549.0,
      "step": 44200
    },
    {
      "entropy": 5.854747247695923,
      "epoch": 3.863060386262344,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003572227256470725,
      "loss": 5.3822,
      "mean_token_accuracy": 0.19547425508499144,
      "num_tokens": 89575106.0,
      "step": 44205
    },
    {
      "entropy": 5.894582891464234,
      "epoch": 3.8634973346150483,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035719371895187574,
      "loss": 5.3687,
      "mean_token_accuracy": 0.1961258217692375,
      "num_tokens": 89584000.0,
      "step": 44210
    },
    {
      "entropy": 5.8374265193939205,
      "epoch": 3.863934282967753,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003571647106801604,
      "loss": 5.3912,
      "mean_token_accuracy": 0.20089780390262604,
      "num_tokens": 89594557.0,
      "step": 44215
    },
    {
      "entropy": 5.8918084621429445,
      "epoch": 3.864371231320458,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003571357008324828,
      "loss": 5.3623,
      "mean_token_accuracy": 0.1985595554113388,
      "num_tokens": 89603710.0,
      "step": 44220
    },
    {
      "entropy": 5.915180730819702,
      "epoch": 3.8648081796731626,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035710668940939935,
      "loss": 5.4636,
      "mean_token_accuracy": 0.1883682519197464,
      "num_tokens": 89614880.0,
      "step": 44225
    },
    {
      "entropy": 5.9208259105682375,
      "epoch": 3.8652451280258675,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035707767641146663,
      "loss": 5.4471,
      "mean_token_accuracy": 0.19458792954683304,
      "num_tokens": 89624377.0,
      "step": 44230
    },
    {
      "entropy": 5.867324018478394,
      "epoch": 3.865682076378572,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003570486618392411,
      "loss": 5.4204,
      "mean_token_accuracy": 0.19106827080249786,
      "num_tokens": 89634260.0,
      "step": 44235
    },
    {
      "entropy": 5.901328611373901,
      "epoch": 3.866119024731277,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035701964569327926,
      "loss": 5.3884,
      "mean_token_accuracy": 0.18839158564805986,
      "num_tokens": 89643642.0,
      "step": 44240
    },
    {
      "entropy": 5.911831378936768,
      "epoch": 3.8665559730839814,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035699062797413754,
      "loss": 5.4656,
      "mean_token_accuracy": 0.19155731350183486,
      "num_tokens": 89653461.0,
      "step": 44245
    },
    {
      "entropy": 5.878284978866577,
      "epoch": 3.8669929214366863,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003569616086823727,
      "loss": 5.4507,
      "mean_token_accuracy": 0.19077811539173126,
      "num_tokens": 89663108.0,
      "step": 44250
    },
    {
      "entropy": 5.952291584014892,
      "epoch": 3.8674298697893907,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003569325878185412,
      "loss": 5.4537,
      "mean_token_accuracy": 0.1865413561463356,
      "num_tokens": 89673682.0,
      "step": 44255
    },
    {
      "entropy": 5.960775661468506,
      "epoch": 3.8678668181420957,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003569035653831998,
      "loss": 5.4748,
      "mean_token_accuracy": 0.19211483150720596,
      "num_tokens": 89684795.0,
      "step": 44260
    },
    {
      "entropy": 5.931886291503906,
      "epoch": 3.8683037664948,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035687454137690514,
      "loss": 5.4397,
      "mean_token_accuracy": 0.19255033731460572,
      "num_tokens": 89695134.0,
      "step": 44265
    },
    {
      "entropy": 5.874860334396362,
      "epoch": 3.868740714847505,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035684551580021374,
      "loss": 5.4239,
      "mean_token_accuracy": 0.19852781891822815,
      "num_tokens": 89705087.0,
      "step": 44270
    },
    {
      "entropy": 5.913855504989624,
      "epoch": 3.86917766320021,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035681648865368255,
      "loss": 5.3406,
      "mean_token_accuracy": 0.19587456434965134,
      "num_tokens": 89714920.0,
      "step": 44275
    },
    {
      "entropy": 5.895542430877685,
      "epoch": 3.8696146115529144,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035678745993786813,
      "loss": 5.3496,
      "mean_token_accuracy": 0.19863004088401795,
      "num_tokens": 89724022.0,
      "step": 44280
    },
    {
      "entropy": 5.865768480300903,
      "epoch": 3.870051559905619,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003567584296533275,
      "loss": 5.3957,
      "mean_token_accuracy": 0.1953692242503166,
      "num_tokens": 89734179.0,
      "step": 44285
    },
    {
      "entropy": 5.897410345077515,
      "epoch": 3.870488508258324,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003567293978006173,
      "loss": 5.408,
      "mean_token_accuracy": 0.19688791483640672,
      "num_tokens": 89743810.0,
      "step": 44290
    },
    {
      "entropy": 5.903408575057983,
      "epoch": 3.8709254566110287,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003567003643802944,
      "loss": 5.406,
      "mean_token_accuracy": 0.20316313952207565,
      "num_tokens": 89753578.0,
      "step": 44295
    },
    {
      "entropy": 5.857881546020508,
      "epoch": 3.871362404963733,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035667132939291575,
      "loss": 5.36,
      "mean_token_accuracy": 0.19675516933202744,
      "num_tokens": 89763569.0,
      "step": 44300
    },
    {
      "entropy": 5.903402376174927,
      "epoch": 3.871799353316438,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003566422928390381,
      "loss": 5.4166,
      "mean_token_accuracy": 0.19141010493040084,
      "num_tokens": 89772759.0,
      "step": 44305
    },
    {
      "entropy": 5.811553812026977,
      "epoch": 3.8722363016691426,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003566132547192187,
      "loss": 5.3315,
      "mean_token_accuracy": 0.20415816456079483,
      "num_tokens": 89782499.0,
      "step": 44310
    },
    {
      "entropy": 5.846604919433593,
      "epoch": 3.8726732500218475,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035658421503401415,
      "loss": 5.412,
      "mean_token_accuracy": 0.19625313729047775,
      "num_tokens": 89792831.0,
      "step": 44315
    },
    {
      "entropy": 5.904768466949463,
      "epoch": 3.873110198374552,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003565551737839816,
      "loss": 5.4105,
      "mean_token_accuracy": 0.19617018550634385,
      "num_tokens": 89802411.0,
      "step": 44320
    },
    {
      "entropy": 5.888664054870605,
      "epoch": 3.873547146727257,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003565261309696781,
      "loss": 5.4344,
      "mean_token_accuracy": 0.19393368661403657,
      "num_tokens": 89812157.0,
      "step": 44325
    },
    {
      "entropy": 5.894292831420898,
      "epoch": 3.8739840950799618,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003564970865916608,
      "loss": 5.3186,
      "mean_token_accuracy": 0.20673375427722931,
      "num_tokens": 89821849.0,
      "step": 44330
    },
    {
      "entropy": 5.894961166381836,
      "epoch": 3.8744210434326662,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035646804065048653,
      "loss": 5.3786,
      "mean_token_accuracy": 0.1936422750353813,
      "num_tokens": 89830993.0,
      "step": 44335
    },
    {
      "entropy": 5.83001217842102,
      "epoch": 3.8748579917853707,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035643899314671265,
      "loss": 5.3907,
      "mean_token_accuracy": 0.19812514036893844,
      "num_tokens": 89841418.0,
      "step": 44340
    },
    {
      "entropy": 5.880626201629639,
      "epoch": 3.8752949401380756,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035640994408089616,
      "loss": 5.3725,
      "mean_token_accuracy": 0.2064749002456665,
      "num_tokens": 89851994.0,
      "step": 44345
    },
    {
      "entropy": 5.893804931640625,
      "epoch": 3.8757318884907805,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035638089345359426,
      "loss": 5.4067,
      "mean_token_accuracy": 0.20185903757810592,
      "num_tokens": 89861127.0,
      "step": 44350
    },
    {
      "entropy": 5.929302835464478,
      "epoch": 3.876168836843485,
      "grad_norm": 0.953125,
      "learning_rate": 0.00035635184126536426,
      "loss": 5.4349,
      "mean_token_accuracy": 0.192827944457531,
      "num_tokens": 89872346.0,
      "step": 44355
    },
    {
      "entropy": 5.930033206939697,
      "epoch": 3.87660578519619,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003563227875167633,
      "loss": 5.4472,
      "mean_token_accuracy": 0.19110507369041443,
      "num_tokens": 89882223.0,
      "step": 44360
    },
    {
      "entropy": 5.908868598937988,
      "epoch": 3.8770427335488944,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035629373220834857,
      "loss": 5.4242,
      "mean_token_accuracy": 0.19173272848129272,
      "num_tokens": 89892305.0,
      "step": 44365
    },
    {
      "entropy": 5.8597620010375975,
      "epoch": 3.8774796819015993,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003562646753406775,
      "loss": 5.3661,
      "mean_token_accuracy": 0.2007285550236702,
      "num_tokens": 89901547.0,
      "step": 44370
    },
    {
      "entropy": 5.881566095352173,
      "epoch": 3.8779166302543038,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00035623561691430734,
      "loss": 5.4022,
      "mean_token_accuracy": 0.20030203014612197,
      "num_tokens": 89912686.0,
      "step": 44375
    },
    {
      "entropy": 5.960371017456055,
      "epoch": 3.8783535786070087,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003562065569297956,
      "loss": 5.4029,
      "mean_token_accuracy": 0.19637266546487808,
      "num_tokens": 89921728.0,
      "step": 44380
    },
    {
      "entropy": 5.854409456253052,
      "epoch": 3.8787905269597136,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003561774953876995,
      "loss": 5.3553,
      "mean_token_accuracy": 0.20317064672708512,
      "num_tokens": 89931741.0,
      "step": 44385
    },
    {
      "entropy": 5.870904445648193,
      "epoch": 3.879227475312418,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035614843228857644,
      "loss": 5.3819,
      "mean_token_accuracy": 0.19607710242271423,
      "num_tokens": 89941323.0,
      "step": 44390
    },
    {
      "entropy": 5.876072216033935,
      "epoch": 3.8796644236651225,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000356119367632984,
      "loss": 5.4503,
      "mean_token_accuracy": 0.1908787429332733,
      "num_tokens": 89950763.0,
      "step": 44395
    },
    {
      "entropy": 5.901320219039917,
      "epoch": 3.8801013720178275,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003560903014214795,
      "loss": 5.4855,
      "mean_token_accuracy": 0.18721729815006255,
      "num_tokens": 89961371.0,
      "step": 44400
    },
    {
      "entropy": 6.001392221450805,
      "epoch": 3.8805383203705324,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003560612336546206,
      "loss": 5.4884,
      "mean_token_accuracy": 0.19011058956384658,
      "num_tokens": 89971880.0,
      "step": 44405
    },
    {
      "entropy": 5.891507482528686,
      "epoch": 3.880975268723237,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003560321643329648,
      "loss": 5.3418,
      "mean_token_accuracy": 0.20379550755023956,
      "num_tokens": 89982040.0,
      "step": 44410
    },
    {
      "entropy": 5.937879800796509,
      "epoch": 3.8814122170759418,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003560030934570696,
      "loss": 5.4534,
      "mean_token_accuracy": 0.18961456120014192,
      "num_tokens": 89992517.0,
      "step": 44415
    },
    {
      "entropy": 5.881943893432617,
      "epoch": 3.881849165428646,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003559740210274925,
      "loss": 5.3743,
      "mean_token_accuracy": 0.20267096310853958,
      "num_tokens": 90002432.0,
      "step": 44420
    },
    {
      "entropy": 5.861389970779419,
      "epoch": 3.882286113781351,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003559449470447914,
      "loss": 5.295,
      "mean_token_accuracy": 0.20783372223377228,
      "num_tokens": 90012556.0,
      "step": 44425
    },
    {
      "entropy": 5.911662340164185,
      "epoch": 3.8827230621340556,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003559158715095238,
      "loss": 5.4677,
      "mean_token_accuracy": 0.19401186853647232,
      "num_tokens": 90022513.0,
      "step": 44430
    },
    {
      "entropy": 5.855313062667847,
      "epoch": 3.8831600104867605,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003558867944222473,
      "loss": 5.384,
      "mean_token_accuracy": 0.1973993167281151,
      "num_tokens": 90032822.0,
      "step": 44435
    },
    {
      "entropy": 5.906007528305054,
      "epoch": 3.8835969588394654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003558577157835197,
      "loss": 5.3645,
      "mean_token_accuracy": 0.19436005651950836,
      "num_tokens": 90042685.0,
      "step": 44440
    },
    {
      "entropy": 5.885476160049438,
      "epoch": 3.88403390719217,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003558286355938988,
      "loss": 5.4156,
      "mean_token_accuracy": 0.1993445560336113,
      "num_tokens": 90053118.0,
      "step": 44445
    },
    {
      "entropy": 5.83148307800293,
      "epoch": 3.8844708555448744,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003557995538539423,
      "loss": 5.388,
      "mean_token_accuracy": 0.19477609544992447,
      "num_tokens": 90063041.0,
      "step": 44450
    },
    {
      "entropy": 5.9598979473114015,
      "epoch": 3.8849078038975793,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000355770470564208,
      "loss": 5.4342,
      "mean_token_accuracy": 0.19398504346609116,
      "num_tokens": 90072995.0,
      "step": 44455
    },
    {
      "entropy": 5.940928411483765,
      "epoch": 3.885344752250284,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003557413857252538,
      "loss": 5.3982,
      "mean_token_accuracy": 0.19431102722883226,
      "num_tokens": 90084369.0,
      "step": 44460
    },
    {
      "entropy": 5.9170760154724125,
      "epoch": 3.8857817006029887,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003557122993376374,
      "loss": 5.4527,
      "mean_token_accuracy": 0.1860503539443016,
      "num_tokens": 90093207.0,
      "step": 44465
    },
    {
      "entropy": 5.8353697776794435,
      "epoch": 3.8862186489556936,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003556832114019169,
      "loss": 5.3861,
      "mean_token_accuracy": 0.19955874383449554,
      "num_tokens": 90102452.0,
      "step": 44470
    },
    {
      "entropy": 5.904739665985107,
      "epoch": 3.886655597308398,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035565412191865015,
      "loss": 5.3722,
      "mean_token_accuracy": 0.1921822026371956,
      "num_tokens": 90112200.0,
      "step": 44475
    },
    {
      "entropy": 5.876715660095215,
      "epoch": 3.887092545661103,
      "grad_norm": 1.25,
      "learning_rate": 0.00035562503088839496,
      "loss": 5.4337,
      "mean_token_accuracy": 0.19126422256231307,
      "num_tokens": 90122627.0,
      "step": 44480
    },
    {
      "entropy": 5.920163536071778,
      "epoch": 3.8875294940138074,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003555959383117096,
      "loss": 5.3941,
      "mean_token_accuracy": 0.19140960276126862,
      "num_tokens": 90131303.0,
      "step": 44485
    },
    {
      "entropy": 5.901561737060547,
      "epoch": 3.8879664423665123,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003555668441891518,
      "loss": 5.3553,
      "mean_token_accuracy": 0.20260828882455825,
      "num_tokens": 90140904.0,
      "step": 44490
    },
    {
      "entropy": 5.919363164901734,
      "epoch": 3.8884033907192173,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003555377485212797,
      "loss": 5.361,
      "mean_token_accuracy": 0.19501033872365953,
      "num_tokens": 90150729.0,
      "step": 44495
    },
    {
      "entropy": 5.964505195617676,
      "epoch": 3.8888403390719217,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003555086513086514,
      "loss": 5.4548,
      "mean_token_accuracy": 0.1850856989622116,
      "num_tokens": 90160832.0,
      "step": 44500
    },
    {
      "entropy": 5.886656808853149,
      "epoch": 3.889277287424626,
      "grad_norm": 1.0,
      "learning_rate": 0.000355479552551825,
      "loss": 5.4079,
      "mean_token_accuracy": 0.19760433137416838,
      "num_tokens": 90171606.0,
      "step": 44505
    },
    {
      "entropy": 5.827731513977051,
      "epoch": 3.889714235777331,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035545045225135864,
      "loss": 5.3683,
      "mean_token_accuracy": 0.2021147608757019,
      "num_tokens": 90181902.0,
      "step": 44510
    },
    {
      "entropy": 5.901977252960205,
      "epoch": 3.890151184130036,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003554213504078104,
      "loss": 5.4541,
      "mean_token_accuracy": 0.19443643242120742,
      "num_tokens": 90192380.0,
      "step": 44515
    },
    {
      "entropy": 5.875019121170044,
      "epoch": 3.8905881324827405,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003553922470217385,
      "loss": 5.3468,
      "mean_token_accuracy": 0.20852369964122772,
      "num_tokens": 90202469.0,
      "step": 44520
    },
    {
      "entropy": 5.959057092666626,
      "epoch": 3.8910250808354454,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035536314209370126,
      "loss": 5.3839,
      "mean_token_accuracy": 0.2009423404932022,
      "num_tokens": 90212706.0,
      "step": 44525
    },
    {
      "entropy": 5.92743501663208,
      "epoch": 3.89146202918815,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035533403562425676,
      "loss": 5.4022,
      "mean_token_accuracy": 0.19334080666303635,
      "num_tokens": 90222456.0,
      "step": 44530
    },
    {
      "entropy": 5.843095922470093,
      "epoch": 3.891898977540855,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035530492761396337,
      "loss": 5.3607,
      "mean_token_accuracy": 0.19889068305492402,
      "num_tokens": 90232960.0,
      "step": 44535
    },
    {
      "entropy": 5.854761028289795,
      "epoch": 3.8923359258935593,
      "grad_norm": 0.953125,
      "learning_rate": 0.00035527581806337946,
      "loss": 5.3688,
      "mean_token_accuracy": 0.19155070930719376,
      "num_tokens": 90242407.0,
      "step": 44540
    },
    {
      "entropy": 5.919047117233276,
      "epoch": 3.892772874246264,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035524670697306324,
      "loss": 5.3729,
      "mean_token_accuracy": 0.19688211530447006,
      "num_tokens": 90252211.0,
      "step": 44545
    },
    {
      "entropy": 5.903820180892945,
      "epoch": 3.8932098225989686,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0003552175943435731,
      "loss": 5.4052,
      "mean_token_accuracy": 0.19445603787899018,
      "num_tokens": 90262047.0,
      "step": 44550
    },
    {
      "entropy": 5.82642674446106,
      "epoch": 3.8936467709516736,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035518848017546757,
      "loss": 5.3962,
      "mean_token_accuracy": 0.19092909693717958,
      "num_tokens": 90272870.0,
      "step": 44555
    },
    {
      "entropy": 5.895777082443237,
      "epoch": 3.894083719304378,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003551593644693049,
      "loss": 5.4802,
      "mean_token_accuracy": 0.19158830493688583,
      "num_tokens": 90282581.0,
      "step": 44560
    },
    {
      "entropy": 5.9294861316680905,
      "epoch": 3.894520667657083,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035513024722564365,
      "loss": 5.4251,
      "mean_token_accuracy": 0.20192052125930787,
      "num_tokens": 90294100.0,
      "step": 44565
    },
    {
      "entropy": 5.932930994033813,
      "epoch": 3.894957616009788,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035510112844504223,
      "loss": 5.3897,
      "mean_token_accuracy": 0.20383006930351258,
      "num_tokens": 90305106.0,
      "step": 44570
    },
    {
      "entropy": 5.973030996322632,
      "epoch": 3.8953945643624923,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003550720081280592,
      "loss": 5.4096,
      "mean_token_accuracy": 0.19578337222337722,
      "num_tokens": 90315246.0,
      "step": 44575
    },
    {
      "entropy": 5.877092361450195,
      "epoch": 3.895831512715197,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035504288627525313,
      "loss": 5.4195,
      "mean_token_accuracy": 0.19539878964424134,
      "num_tokens": 90324312.0,
      "step": 44580
    },
    {
      "entropy": 5.881743717193603,
      "epoch": 3.8962684610679017,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003550137628871825,
      "loss": 5.3498,
      "mean_token_accuracy": 0.20534342676401138,
      "num_tokens": 90333639.0,
      "step": 44585
    },
    {
      "entropy": 5.806100130081177,
      "epoch": 3.8967054094206066,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00035498463796440605,
      "loss": 5.3112,
      "mean_token_accuracy": 0.2072661742568016,
      "num_tokens": 90343332.0,
      "step": 44590
    },
    {
      "entropy": 5.87686128616333,
      "epoch": 3.897142357773311,
      "grad_norm": 0.984375,
      "learning_rate": 0.00035495551150748226,
      "loss": 5.3571,
      "mean_token_accuracy": 0.19967255592346192,
      "num_tokens": 90355337.0,
      "step": 44595
    },
    {
      "entropy": 5.953037691116333,
      "epoch": 3.897579306126016,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035492638351697,
      "loss": 5.4946,
      "mean_token_accuracy": 0.19123624563217162,
      "num_tokens": 90367126.0,
      "step": 44600
    },
    {
      "entropy": 5.9010782718658445,
      "epoch": 3.8980162544787205,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003548972539934277,
      "loss": 5.4142,
      "mean_token_accuracy": 0.20144130140542985,
      "num_tokens": 90377408.0,
      "step": 44605
    },
    {
      "entropy": 5.933610582351685,
      "epoch": 3.8984532028314254,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003548681229374142,
      "loss": 5.3872,
      "mean_token_accuracy": 0.19909258633852006,
      "num_tokens": 90388451.0,
      "step": 44610
    },
    {
      "entropy": 5.903042840957641,
      "epoch": 3.89889015118413,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0003548389903494883,
      "loss": 5.4081,
      "mean_token_accuracy": 0.1935234099626541,
      "num_tokens": 90399673.0,
      "step": 44615
    },
    {
      "entropy": 5.857484006881714,
      "epoch": 3.8993270995368348,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00035480985623020873,
      "loss": 5.3395,
      "mean_token_accuracy": 0.19866019636392593,
      "num_tokens": 90409567.0,
      "step": 44620
    },
    {
      "entropy": 5.8931983470916744,
      "epoch": 3.8997640478895397,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003547807205801342,
      "loss": 5.4743,
      "mean_token_accuracy": 0.1907406285405159,
      "num_tokens": 90419904.0,
      "step": 44625
    },
    {
      "entropy": 5.8973533630371096,
      "epoch": 3.900200996242244,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003547515833998237,
      "loss": 5.3584,
      "mean_token_accuracy": 0.19675642251968384,
      "num_tokens": 90431445.0,
      "step": 44630
    },
    {
      "entropy": 5.9105010509490965,
      "epoch": 3.9006379445949486,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003547224446898361,
      "loss": 5.3889,
      "mean_token_accuracy": 0.1956273928284645,
      "num_tokens": 90440933.0,
      "step": 44635
    },
    {
      "entropy": 5.857160329818726,
      "epoch": 3.9010748929476535,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035469330445073025,
      "loss": 5.4123,
      "mean_token_accuracy": 0.19746977239847183,
      "num_tokens": 90451145.0,
      "step": 44640
    },
    {
      "entropy": 5.935052299499512,
      "epoch": 3.9015118413003584,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000354664162683065,
      "loss": 5.4727,
      "mean_token_accuracy": 0.18881838619709015,
      "num_tokens": 90460355.0,
      "step": 44645
    },
    {
      "entropy": 5.9447314739227295,
      "epoch": 3.901948789653063,
      "grad_norm": 1.125,
      "learning_rate": 0.00035463501938739936,
      "loss": 5.3996,
      "mean_token_accuracy": 0.19681197851896287,
      "num_tokens": 90471162.0,
      "step": 44650
    },
    {
      "entropy": 5.916376972198487,
      "epoch": 3.902385738005768,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003546058745642923,
      "loss": 5.4386,
      "mean_token_accuracy": 0.19259335696697236,
      "num_tokens": 90481807.0,
      "step": 44655
    },
    {
      "entropy": 5.881879711151123,
      "epoch": 3.9028226863584723,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035457672821430293,
      "loss": 5.3237,
      "mean_token_accuracy": 0.20340973138809204,
      "num_tokens": 90491478.0,
      "step": 44660
    },
    {
      "entropy": 5.777308750152588,
      "epoch": 3.903259634711177,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003545475803379902,
      "loss": 5.2783,
      "mean_token_accuracy": 0.21051336377859114,
      "num_tokens": 90501231.0,
      "step": 44665
    },
    {
      "entropy": 5.864694976806641,
      "epoch": 3.9036965830638817,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00035451843093591316,
      "loss": 5.4476,
      "mean_token_accuracy": 0.1942499101161957,
      "num_tokens": 90512021.0,
      "step": 44670
    },
    {
      "entropy": 5.958668565750122,
      "epoch": 3.9041335314165866,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035448928000863096,
      "loss": 5.4962,
      "mean_token_accuracy": 0.19284610003232955,
      "num_tokens": 90522210.0,
      "step": 44675
    },
    {
      "entropy": 5.983887958526611,
      "epoch": 3.9045704797692915,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003544601275567027,
      "loss": 5.4532,
      "mean_token_accuracy": 0.19624726474285126,
      "num_tokens": 90533024.0,
      "step": 44680
    },
    {
      "entropy": 5.835081005096436,
      "epoch": 3.905007428121996,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003544309735806876,
      "loss": 5.3792,
      "mean_token_accuracy": 0.20761643648147582,
      "num_tokens": 90541965.0,
      "step": 44685
    },
    {
      "entropy": 5.907034587860108,
      "epoch": 3.9054443764747004,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003544018180811448,
      "loss": 5.3564,
      "mean_token_accuracy": 0.19303468018770217,
      "num_tokens": 90552167.0,
      "step": 44690
    },
    {
      "entropy": 5.929994297027588,
      "epoch": 3.9058813248274054,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003543726610586335,
      "loss": 5.3493,
      "mean_token_accuracy": 0.20090316981077194,
      "num_tokens": 90562383.0,
      "step": 44695
    },
    {
      "entropy": 5.947171831130982,
      "epoch": 3.9063182731801103,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00035434350251371295,
      "loss": 5.4007,
      "mean_token_accuracy": 0.19753718078136445,
      "num_tokens": 90572609.0,
      "step": 44700
    },
    {
      "entropy": 5.819735145568847,
      "epoch": 3.9067552215328147,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003543143424469425,
      "loss": 5.4006,
      "mean_token_accuracy": 0.20001446455717087,
      "num_tokens": 90582319.0,
      "step": 44705
    },
    {
      "entropy": 5.890216207504272,
      "epoch": 3.9071921698855197,
      "grad_norm": 1.453125,
      "learning_rate": 0.00035428518085888134,
      "loss": 5.4401,
      "mean_token_accuracy": 0.19456266164779662,
      "num_tokens": 90592493.0,
      "step": 44710
    },
    {
      "entropy": 6.006965446472168,
      "epoch": 3.907629118238224,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035425601775008895,
      "loss": 5.5018,
      "mean_token_accuracy": 0.1908114656805992,
      "num_tokens": 90603504.0,
      "step": 44715
    },
    {
      "entropy": 5.969165468215943,
      "epoch": 3.908066066590929,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003542268531211245,
      "loss": 5.4664,
      "mean_token_accuracy": 0.19029301404953003,
      "num_tokens": 90613957.0,
      "step": 44720
    },
    {
      "entropy": 5.875598192214966,
      "epoch": 3.9085030149436335,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003541976869725476,
      "loss": 5.3778,
      "mean_token_accuracy": 0.20019626170396804,
      "num_tokens": 90623986.0,
      "step": 44725
    },
    {
      "entropy": 5.912237882614136,
      "epoch": 3.9089399632963384,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003541685193049175,
      "loss": 5.416,
      "mean_token_accuracy": 0.1952621355652809,
      "num_tokens": 90633879.0,
      "step": 44730
    },
    {
      "entropy": 5.88684983253479,
      "epoch": 3.9093769116490433,
      "grad_norm": 1.125,
      "learning_rate": 0.0003541393501187937,
      "loss": 5.4332,
      "mean_token_accuracy": 0.1875145986676216,
      "num_tokens": 90644031.0,
      "step": 44735
    },
    {
      "entropy": 5.953464984893799,
      "epoch": 3.909813860001748,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035411017941473576,
      "loss": 5.43,
      "mean_token_accuracy": 0.19766127020120622,
      "num_tokens": 90654550.0,
      "step": 44740
    },
    {
      "entropy": 5.920039415359497,
      "epoch": 3.9102508083544523,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035408100719330304,
      "loss": 5.4362,
      "mean_token_accuracy": 0.18870879262685775,
      "num_tokens": 90664117.0,
      "step": 44745
    },
    {
      "entropy": 5.932317686080933,
      "epoch": 3.910687756707157,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003540518334550552,
      "loss": 5.4367,
      "mean_token_accuracy": 0.18734760731458663,
      "num_tokens": 90674400.0,
      "step": 44750
    },
    {
      "entropy": 5.946831321716308,
      "epoch": 3.911124705059862,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035402265820055185,
      "loss": 5.4296,
      "mean_token_accuracy": 0.1949465572834015,
      "num_tokens": 90684759.0,
      "step": 44755
    },
    {
      "entropy": 6.0108497619628904,
      "epoch": 3.9115616534125666,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035399348143035244,
      "loss": 5.4684,
      "mean_token_accuracy": 0.1930246815085411,
      "num_tokens": 90694346.0,
      "step": 44760
    },
    {
      "entropy": 5.897690010070801,
      "epoch": 3.9119986017652715,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035396430314501665,
      "loss": 5.4247,
      "mean_token_accuracy": 0.19297244101762773,
      "num_tokens": 90704239.0,
      "step": 44765
    },
    {
      "entropy": 5.887103605270386,
      "epoch": 3.912435550117976,
      "grad_norm": 1.0,
      "learning_rate": 0.0003539351233451042,
      "loss": 5.3811,
      "mean_token_accuracy": 0.1948162645101547,
      "num_tokens": 90714240.0,
      "step": 44770
    },
    {
      "entropy": 5.914733743667602,
      "epoch": 3.912872498470681,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003539059420311747,
      "loss": 5.4409,
      "mean_token_accuracy": 0.18728754818439483,
      "num_tokens": 90724575.0,
      "step": 44775
    },
    {
      "entropy": 5.957843542098999,
      "epoch": 3.9133094468233853,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003538767592037879,
      "loss": 5.4203,
      "mean_token_accuracy": 0.19232597351074218,
      "num_tokens": 90735217.0,
      "step": 44780
    },
    {
      "entropy": 5.924194049835205,
      "epoch": 3.9137463951760902,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003538475748635035,
      "loss": 5.4016,
      "mean_token_accuracy": 0.19663604646921157,
      "num_tokens": 90745437.0,
      "step": 44785
    },
    {
      "entropy": 5.909990882873535,
      "epoch": 3.914183343528795,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003538183890108813,
      "loss": 5.4027,
      "mean_token_accuracy": 0.19582452178001403,
      "num_tokens": 90755286.0,
      "step": 44790
    },
    {
      "entropy": 5.897569084167481,
      "epoch": 3.9146202918814996,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003537892016464811,
      "loss": 5.4479,
      "mean_token_accuracy": 0.19278308153152465,
      "num_tokens": 90764734.0,
      "step": 44795
    },
    {
      "entropy": 5.957988309860229,
      "epoch": 3.915057240234204,
      "grad_norm": 1.0,
      "learning_rate": 0.00035376001277086276,
      "loss": 5.4152,
      "mean_token_accuracy": 0.19532177746295928,
      "num_tokens": 90774569.0,
      "step": 44800
    },
    {
      "entropy": 5.972067546844483,
      "epoch": 3.915494188586909,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003537308223845861,
      "loss": 5.478,
      "mean_token_accuracy": 0.18349899351596832,
      "num_tokens": 90785029.0,
      "step": 44805
    },
    {
      "entropy": 5.992567300796509,
      "epoch": 3.915931136939614,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035370163048821095,
      "loss": 5.4821,
      "mean_token_accuracy": 0.18715584874153138,
      "num_tokens": 90795382.0,
      "step": 44810
    },
    {
      "entropy": 5.83109335899353,
      "epoch": 3.9163680852923184,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035367243708229736,
      "loss": 5.3772,
      "mean_token_accuracy": 0.20326193869113923,
      "num_tokens": 90805354.0,
      "step": 44815
    },
    {
      "entropy": 5.8477123260498045,
      "epoch": 3.9168050336450233,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00035364324216740516,
      "loss": 5.4495,
      "mean_token_accuracy": 0.19077080637216567,
      "num_tokens": 90815967.0,
      "step": 44820
    },
    {
      "entropy": 5.983317899703979,
      "epoch": 3.9172419819977278,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00035361404574409436,
      "loss": 5.4156,
      "mean_token_accuracy": 0.19253168106079102,
      "num_tokens": 90826717.0,
      "step": 44825
    },
    {
      "entropy": 5.933637046813965,
      "epoch": 3.9176789303504327,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000353584847812925,
      "loss": 5.3408,
      "mean_token_accuracy": 0.20027564764022826,
      "num_tokens": 90836592.0,
      "step": 44830
    },
    {
      "entropy": 5.891003799438477,
      "epoch": 3.918115878703137,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035355564837445707,
      "loss": 5.443,
      "mean_token_accuracy": 0.19197567105293273,
      "num_tokens": 90847323.0,
      "step": 44835
    },
    {
      "entropy": 5.949301433563233,
      "epoch": 3.918552827055842,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00035352644742925066,
      "loss": 5.5108,
      "mean_token_accuracy": 0.18487870693206787,
      "num_tokens": 90858471.0,
      "step": 44840
    },
    {
      "entropy": 5.945021057128907,
      "epoch": 3.918989775408547,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035349724497786577,
      "loss": 5.4532,
      "mean_token_accuracy": 0.19256106913089752,
      "num_tokens": 90868983.0,
      "step": 44845
    },
    {
      "entropy": 5.919777870178223,
      "epoch": 3.9194267237612515,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00035346804102086266,
      "loss": 5.4104,
      "mean_token_accuracy": 0.19391283690929412,
      "num_tokens": 90879904.0,
      "step": 44850
    },
    {
      "entropy": 5.960743522644043,
      "epoch": 3.919863672113956,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035343883555880143,
      "loss": 5.4454,
      "mean_token_accuracy": 0.19002049863338472,
      "num_tokens": 90889923.0,
      "step": 44855
    },
    {
      "entropy": 5.936284399032592,
      "epoch": 3.920300620466661,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035340962859224217,
      "loss": 5.4004,
      "mean_token_accuracy": 0.19004858136177064,
      "num_tokens": 90899991.0,
      "step": 44860
    },
    {
      "entropy": 5.859996175765991,
      "epoch": 3.9207375688193657,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003533804201217452,
      "loss": 5.3604,
      "mean_token_accuracy": 0.20074986964464187,
      "num_tokens": 90910359.0,
      "step": 44865
    },
    {
      "entropy": 5.898529720306397,
      "epoch": 3.92117451717207,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00035335121014787063,
      "loss": 5.3689,
      "mean_token_accuracy": 0.19886500835418702,
      "num_tokens": 90920033.0,
      "step": 44870
    },
    {
      "entropy": 5.961621427536011,
      "epoch": 3.921611465524775,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003533219986711789,
      "loss": 5.4464,
      "mean_token_accuracy": 0.19810355603694915,
      "num_tokens": 90929301.0,
      "step": 44875
    },
    {
      "entropy": 5.871027326583862,
      "epoch": 3.9220484138774796,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003532927856922301,
      "loss": 5.3318,
      "mean_token_accuracy": 0.2036784678697586,
      "num_tokens": 90939197.0,
      "step": 44880
    },
    {
      "entropy": 5.955079412460327,
      "epoch": 3.9224853622301845,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035326357121158463,
      "loss": 5.431,
      "mean_token_accuracy": 0.19515102952718735,
      "num_tokens": 90949013.0,
      "step": 44885
    },
    {
      "entropy": 5.9039952754974365,
      "epoch": 3.922922310582889,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003532343552298029,
      "loss": 5.4174,
      "mean_token_accuracy": 0.19589525163173677,
      "num_tokens": 90960010.0,
      "step": 44890
    },
    {
      "entropy": 5.836123275756836,
      "epoch": 3.923359258935594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003532051377474453,
      "loss": 5.4336,
      "mean_token_accuracy": 0.1949218690395355,
      "num_tokens": 90969768.0,
      "step": 44895
    },
    {
      "entropy": 5.890427970886231,
      "epoch": 3.9237962072882984,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003531759187650721,
      "loss": 5.4026,
      "mean_token_accuracy": 0.19858113527297974,
      "num_tokens": 90980051.0,
      "step": 44900
    },
    {
      "entropy": 5.9140105724334715,
      "epoch": 3.9242331556410033,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003531466982832438,
      "loss": 5.3832,
      "mean_token_accuracy": 0.2008470892906189,
      "num_tokens": 90990144.0,
      "step": 44905
    },
    {
      "entropy": 5.855803728103638,
      "epoch": 3.9246701039937077,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003531174763025209,
      "loss": 5.2995,
      "mean_token_accuracy": 0.20294807702302933,
      "num_tokens": 91000353.0,
      "step": 44910
    },
    {
      "entropy": 5.935808515548706,
      "epoch": 3.9251070523464127,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003530882528234639,
      "loss": 5.3984,
      "mean_token_accuracy": 0.1995573952794075,
      "num_tokens": 91010819.0,
      "step": 44915
    },
    {
      "entropy": 5.929035425186157,
      "epoch": 3.9255440006991176,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035305902784663333,
      "loss": 5.4418,
      "mean_token_accuracy": 0.18555504977703094,
      "num_tokens": 91020454.0,
      "step": 44920
    },
    {
      "entropy": 5.92770528793335,
      "epoch": 3.925980949051822,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035302980137258964,
      "loss": 5.493,
      "mean_token_accuracy": 0.18776628226041794,
      "num_tokens": 91030710.0,
      "step": 44925
    },
    {
      "entropy": 5.905685043334961,
      "epoch": 3.9264178974045265,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003530005734018935,
      "loss": 5.4027,
      "mean_token_accuracy": 0.19938456267118454,
      "num_tokens": 91039907.0,
      "step": 44930
    },
    {
      "entropy": 5.941298198699951,
      "epoch": 3.9268548457572314,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035297134393510543,
      "loss": 5.419,
      "mean_token_accuracy": 0.19428904354572296,
      "num_tokens": 91049377.0,
      "step": 44935
    },
    {
      "entropy": 5.895479583740235,
      "epoch": 3.9272917941099363,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003529421129727862,
      "loss": 5.4145,
      "mean_token_accuracy": 0.18851068019866943,
      "num_tokens": 91060341.0,
      "step": 44940
    },
    {
      "entropy": 5.932016229629516,
      "epoch": 3.927728742462641,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035291288051549644,
      "loss": 5.3872,
      "mean_token_accuracy": 0.19998223334550858,
      "num_tokens": 91069737.0,
      "step": 44945
    },
    {
      "entropy": 5.912922143936157,
      "epoch": 3.9281656908153457,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035288364656379666,
      "loss": 5.4254,
      "mean_token_accuracy": 0.1968264803290367,
      "num_tokens": 91079290.0,
      "step": 44950
    },
    {
      "entropy": 5.876323461532593,
      "epoch": 3.92860263916805,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00035285441111824784,
      "loss": 5.4053,
      "mean_token_accuracy": 0.19612946659326552,
      "num_tokens": 91090316.0,
      "step": 44955
    },
    {
      "entropy": 5.876632022857666,
      "epoch": 3.929039587520755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035282517417941056,
      "loss": 5.3919,
      "mean_token_accuracy": 0.19415882974863052,
      "num_tokens": 91100521.0,
      "step": 44960
    },
    {
      "entropy": 5.919164037704467,
      "epoch": 3.9294765358734596,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003527959357478457,
      "loss": 5.4284,
      "mean_token_accuracy": 0.19379307329654694,
      "num_tokens": 91110207.0,
      "step": 44965
    },
    {
      "entropy": 5.947861433029175,
      "epoch": 3.9299134842261645,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000352766695824114,
      "loss": 5.4335,
      "mean_token_accuracy": 0.18637275099754333,
      "num_tokens": 91120993.0,
      "step": 44970
    },
    {
      "entropy": 5.877969264984131,
      "epoch": 3.9303504325788694,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003527374544087764,
      "loss": 5.3592,
      "mean_token_accuracy": 0.20176708400249482,
      "num_tokens": 91131769.0,
      "step": 44975
    },
    {
      "entropy": 5.916926002502441,
      "epoch": 3.930787380931574,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003527082115023936,
      "loss": 5.5141,
      "mean_token_accuracy": 0.18704470992088318,
      "num_tokens": 91141958.0,
      "step": 44980
    },
    {
      "entropy": 5.942882394790649,
      "epoch": 3.9312243292842783,
      "grad_norm": 0.984375,
      "learning_rate": 0.00035267896710552667,
      "loss": 5.4002,
      "mean_token_accuracy": 0.19619398415088654,
      "num_tokens": 91151725.0,
      "step": 44985
    },
    {
      "entropy": 5.88894567489624,
      "epoch": 3.9316612776369833,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035264972121873633,
      "loss": 5.3923,
      "mean_token_accuracy": 0.19439311772584916,
      "num_tokens": 91161756.0,
      "step": 44990
    },
    {
      "entropy": 5.924255514144898,
      "epoch": 3.932098225989688,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035262047384258376,
      "loss": 5.5052,
      "mean_token_accuracy": 0.18601775765419007,
      "num_tokens": 91172160.0,
      "step": 44995
    },
    {
      "entropy": 5.97391562461853,
      "epoch": 3.9325351743423926,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003525912249776298,
      "loss": 5.4721,
      "mean_token_accuracy": 0.18815799653530121,
      "num_tokens": 91183030.0,
      "step": 45000
    },
    {
      "epoch": 3.9325351743423926,
      "eval_entropy": 5.714525097720639,
      "eval_loss": 5.456930160522461,
      "eval_mean_token_accuracy": 0.20080244920621274,
      "eval_num_tokens": 91183030.0,
      "eval_runtime": 27.8803,
      "eval_samples_per_second": 1327.604,
      "eval_steps_per_second": 165.959,
      "step": 45000
    },
    {
      "entropy": 5.856861925125122,
      "epoch": 3.9329721226950975,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035256197462443545,
      "loss": 5.3235,
      "mean_token_accuracy": 0.21145770251750945,
      "num_tokens": 91192590.0,
      "step": 45005
    },
    {
      "entropy": 5.848509502410889,
      "epoch": 3.933409071047802,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003525327227835618,
      "loss": 5.3559,
      "mean_token_accuracy": 0.19886096119880675,
      "num_tokens": 91202596.0,
      "step": 45010
    },
    {
      "entropy": 5.897536182403565,
      "epoch": 3.933846019400507,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035250346945556995,
      "loss": 5.4129,
      "mean_token_accuracy": 0.19064736515283584,
      "num_tokens": 91211922.0,
      "step": 45015
    },
    {
      "entropy": 5.914111137390137,
      "epoch": 3.9342829677532114,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003524742146410209,
      "loss": 5.4228,
      "mean_token_accuracy": 0.19340216666460036,
      "num_tokens": 91223325.0,
      "step": 45020
    },
    {
      "entropy": 5.9041918277740475,
      "epoch": 3.9347199161059163,
      "grad_norm": 1.0,
      "learning_rate": 0.0003524449583404758,
      "loss": 5.3943,
      "mean_token_accuracy": 0.20255510210990907,
      "num_tokens": 91233404.0,
      "step": 45025
    },
    {
      "entropy": 5.884339427947998,
      "epoch": 3.9351568644586212,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035241570055449584,
      "loss": 5.4396,
      "mean_token_accuracy": 0.19498061537742614,
      "num_tokens": 91243816.0,
      "step": 45030
    },
    {
      "entropy": 5.9774833679199215,
      "epoch": 3.9355938128113257,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00035238644128364217,
      "loss": 5.4388,
      "mean_token_accuracy": 0.1975722149014473,
      "num_tokens": 91254023.0,
      "step": 45035
    },
    {
      "entropy": 5.893186902999878,
      "epoch": 3.93603076116403,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00035235718052847604,
      "loss": 5.4112,
      "mean_token_accuracy": 0.2014889433979988,
      "num_tokens": 91264151.0,
      "step": 45040
    },
    {
      "entropy": 5.926684379577637,
      "epoch": 3.936467709516735,
      "grad_norm": 1.953125,
      "learning_rate": 0.00035232791828955864,
      "loss": 5.3961,
      "mean_token_accuracy": 0.19751551300287246,
      "num_tokens": 91274088.0,
      "step": 45045
    },
    {
      "entropy": 5.821940326690674,
      "epoch": 3.93690465786944,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003522986545674512,
      "loss": 5.3536,
      "mean_token_accuracy": 0.20581043660640716,
      "num_tokens": 91284461.0,
      "step": 45050
    },
    {
      "entropy": 5.933639287948608,
      "epoch": 3.9373416062221445,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003522693893627151,
      "loss": 5.4249,
      "mean_token_accuracy": 0.19229496717453004,
      "num_tokens": 91293942.0,
      "step": 45055
    },
    {
      "entropy": 5.932270288467407,
      "epoch": 3.9377785545748494,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003522401226759116,
      "loss": 5.4601,
      "mean_token_accuracy": 0.1923312246799469,
      "num_tokens": 91305322.0,
      "step": 45060
    },
    {
      "entropy": 5.9419896602630615,
      "epoch": 3.938215502927554,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003522108545076021,
      "loss": 5.4652,
      "mean_token_accuracy": 0.18815737068653107,
      "num_tokens": 91314947.0,
      "step": 45065
    },
    {
      "entropy": 5.9350687026977536,
      "epoch": 3.9386524512802588,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000352181584858348,
      "loss": 5.3964,
      "mean_token_accuracy": 0.19327891916036605,
      "num_tokens": 91324229.0,
      "step": 45070
    },
    {
      "entropy": 5.910287952423095,
      "epoch": 3.9390893996329632,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035215231372871054,
      "loss": 5.4394,
      "mean_token_accuracy": 0.1883604794740677,
      "num_tokens": 91334012.0,
      "step": 45075
    },
    {
      "entropy": 5.836820411682129,
      "epoch": 3.939526347985668,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003521230411192513,
      "loss": 5.3406,
      "mean_token_accuracy": 0.20667302012443542,
      "num_tokens": 91344067.0,
      "step": 45080
    },
    {
      "entropy": 5.841457414627075,
      "epoch": 3.939963296338373,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003520937670305318,
      "loss": 5.3466,
      "mean_token_accuracy": 0.19653775095939635,
      "num_tokens": 91353832.0,
      "step": 45085
    },
    {
      "entropy": 5.879140758514405,
      "epoch": 3.9404002446910775,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035206449146311345,
      "loss": 5.3172,
      "mean_token_accuracy": 0.20255308598279953,
      "num_tokens": 91362790.0,
      "step": 45090
    },
    {
      "entropy": 5.893137073516845,
      "epoch": 3.940837193043782,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003520352144175577,
      "loss": 5.4089,
      "mean_token_accuracy": 0.1980736345052719,
      "num_tokens": 91372116.0,
      "step": 45095
    },
    {
      "entropy": 5.893943500518799,
      "epoch": 3.941274141396487,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003520059358944262,
      "loss": 5.3403,
      "mean_token_accuracy": 0.19978930950164794,
      "num_tokens": 91382295.0,
      "step": 45100
    },
    {
      "entropy": 5.85927381515503,
      "epoch": 3.941711089749192,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035197665589428045,
      "loss": 5.3864,
      "mean_token_accuracy": 0.19951989352703095,
      "num_tokens": 91391568.0,
      "step": 45105
    },
    {
      "entropy": 5.845892381668091,
      "epoch": 3.9421480381018963,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035194737441768227,
      "loss": 5.3534,
      "mean_token_accuracy": 0.20159758627414703,
      "num_tokens": 91400254.0,
      "step": 45110
    },
    {
      "entropy": 5.921351146697998,
      "epoch": 3.942584986454601,
      "grad_norm": 1.015625,
      "learning_rate": 0.000351918091465193,
      "loss": 5.4286,
      "mean_token_accuracy": 0.18730096518993378,
      "num_tokens": 91409923.0,
      "step": 45115
    },
    {
      "entropy": 5.945035409927368,
      "epoch": 3.9430219348073057,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035188880703737444,
      "loss": 5.4223,
      "mean_token_accuracy": 0.19581856578588486,
      "num_tokens": 91420090.0,
      "step": 45120
    },
    {
      "entropy": 5.915663766860962,
      "epoch": 3.9434588831600106,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035185952113478826,
      "loss": 5.483,
      "mean_token_accuracy": 0.18733440190553666,
      "num_tokens": 91429884.0,
      "step": 45125
    },
    {
      "entropy": 5.855544900894165,
      "epoch": 3.943895831512715,
      "grad_norm": 1.625,
      "learning_rate": 0.0003518302337579962,
      "loss": 5.3905,
      "mean_token_accuracy": 0.18863130807876588,
      "num_tokens": 91439517.0,
      "step": 45130
    },
    {
      "entropy": 5.998754072189331,
      "epoch": 3.94433277986542,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035180094490756006,
      "loss": 5.5354,
      "mean_token_accuracy": 0.1870455339550972,
      "num_tokens": 91449673.0,
      "step": 45135
    },
    {
      "entropy": 5.950795459747314,
      "epoch": 3.944769728218125,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003517716545840415,
      "loss": 5.4705,
      "mean_token_accuracy": 0.1900292932987213,
      "num_tokens": 91460163.0,
      "step": 45140
    },
    {
      "entropy": 5.867746639251709,
      "epoch": 3.9452066765708294,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003517423627880024,
      "loss": 5.4565,
      "mean_token_accuracy": 0.19499676078557968,
      "num_tokens": 91470735.0,
      "step": 45145
    },
    {
      "entropy": 5.869766569137573,
      "epoch": 3.945643624923534,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003517130695200045,
      "loss": 5.3544,
      "mean_token_accuracy": 0.20284609198570253,
      "num_tokens": 91480791.0,
      "step": 45150
    },
    {
      "entropy": 5.89587664604187,
      "epoch": 3.9460805732762387,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035168377478060983,
      "loss": 5.3015,
      "mean_token_accuracy": 0.2029074251651764,
      "num_tokens": 91490061.0,
      "step": 45155
    },
    {
      "entropy": 5.873909425735474,
      "epoch": 3.9465175216289436,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003516544785703801,
      "loss": 5.3504,
      "mean_token_accuracy": 0.20058971792459487,
      "num_tokens": 91500185.0,
      "step": 45160
    },
    {
      "entropy": 5.9348961353302006,
      "epoch": 3.946954469981648,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003516251808898774,
      "loss": 5.4071,
      "mean_token_accuracy": 0.19020168334245682,
      "num_tokens": 91510028.0,
      "step": 45165
    },
    {
      "entropy": 5.840736627578735,
      "epoch": 3.947391418334353,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035159588173966344,
      "loss": 5.3019,
      "mean_token_accuracy": 0.21169986128807067,
      "num_tokens": 91519333.0,
      "step": 45170
    },
    {
      "entropy": 5.847917652130127,
      "epoch": 3.9478283666870575,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035156658112030036,
      "loss": 5.3209,
      "mean_token_accuracy": 0.1972884029150009,
      "num_tokens": 91528878.0,
      "step": 45175
    },
    {
      "entropy": 5.949649000167847,
      "epoch": 3.9482653150397624,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003515372790323502,
      "loss": 5.3918,
      "mean_token_accuracy": 0.1928794041275978,
      "num_tokens": 91537817.0,
      "step": 45180
    },
    {
      "entropy": 5.90645318031311,
      "epoch": 3.948702263392467,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003515079754763749,
      "loss": 5.3534,
      "mean_token_accuracy": 0.20168910026550294,
      "num_tokens": 91548645.0,
      "step": 45185
    },
    {
      "entropy": 5.830988025665283,
      "epoch": 3.949139211745172,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035147867045293656,
      "loss": 5.3865,
      "mean_token_accuracy": 0.19998094737529754,
      "num_tokens": 91558695.0,
      "step": 45190
    },
    {
      "entropy": 5.872022294998169,
      "epoch": 3.9495761600978763,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00035144936396259725,
      "loss": 5.4548,
      "mean_token_accuracy": 0.1957237094640732,
      "num_tokens": 91568711.0,
      "step": 45195
    },
    {
      "entropy": 5.882783222198486,
      "epoch": 3.950013108450581,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035142005600591905,
      "loss": 5.4311,
      "mean_token_accuracy": 0.1983308508992195,
      "num_tokens": 91578507.0,
      "step": 45200
    },
    {
      "entropy": 5.866210126876831,
      "epoch": 3.9504500568032856,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003513907465834642,
      "loss": 5.3293,
      "mean_token_accuracy": 0.20727061182260514,
      "num_tokens": 91588418.0,
      "step": 45205
    },
    {
      "entropy": 5.913521862030029,
      "epoch": 3.9508870051559906,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003513614356957947,
      "loss": 5.3662,
      "mean_token_accuracy": 0.19549959450960158,
      "num_tokens": 91597568.0,
      "step": 45210
    },
    {
      "entropy": 5.945180177688599,
      "epoch": 3.9513239535086955,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003513321233434729,
      "loss": 5.49,
      "mean_token_accuracy": 0.1870751664042473,
      "num_tokens": 91607585.0,
      "step": 45215
    },
    {
      "entropy": 5.953971719741821,
      "epoch": 3.9517609018614,
      "grad_norm": 0.921875,
      "learning_rate": 0.00035130280952706095,
      "loss": 5.4618,
      "mean_token_accuracy": 0.19185817390680313,
      "num_tokens": 91618593.0,
      "step": 45220
    },
    {
      "entropy": 5.923290491104126,
      "epoch": 3.952197850214105,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003512734942471212,
      "loss": 5.4918,
      "mean_token_accuracy": 0.18958788961172104,
      "num_tokens": 91628788.0,
      "step": 45225
    },
    {
      "entropy": 5.92666597366333,
      "epoch": 3.9526347985668093,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003512441775042158,
      "loss": 5.4153,
      "mean_token_accuracy": 0.19251566380262375,
      "num_tokens": 91638340.0,
      "step": 45230
    },
    {
      "entropy": 5.998218441009522,
      "epoch": 3.9530717469195142,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003512148592989072,
      "loss": 5.5132,
      "mean_token_accuracy": 0.18861630111932753,
      "num_tokens": 91649510.0,
      "step": 45235
    },
    {
      "entropy": 5.876001453399658,
      "epoch": 3.9535086952722187,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003511855396317576,
      "loss": 5.3915,
      "mean_token_accuracy": 0.20259582102298737,
      "num_tokens": 91659751.0,
      "step": 45240
    },
    {
      "entropy": 5.892048501968384,
      "epoch": 3.9539456436249236,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003511562185033294,
      "loss": 5.3813,
      "mean_token_accuracy": 0.1966933622956276,
      "num_tokens": 91669395.0,
      "step": 45245
    },
    {
      "entropy": 5.824493026733398,
      "epoch": 3.954382591977628,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035112689591418507,
      "loss": 5.2855,
      "mean_token_accuracy": 0.20424778163433074,
      "num_tokens": 91679844.0,
      "step": 45250
    },
    {
      "entropy": 5.92021484375,
      "epoch": 3.954819540330333,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035109757186488696,
      "loss": 5.5094,
      "mean_token_accuracy": 0.1930283173918724,
      "num_tokens": 91690868.0,
      "step": 45255
    },
    {
      "entropy": 5.872564888000488,
      "epoch": 3.9552564886830375,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035106824635599757,
      "loss": 5.3237,
      "mean_token_accuracy": 0.20029356479644775,
      "num_tokens": 91701785.0,
      "step": 45260
    },
    {
      "entropy": 5.9163237571716305,
      "epoch": 3.9556934370357424,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003510389193880793,
      "loss": 5.3983,
      "mean_token_accuracy": 0.19644566178321837,
      "num_tokens": 91711461.0,
      "step": 45265
    },
    {
      "entropy": 5.8223350048065186,
      "epoch": 3.9561303853884473,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035100959096169473,
      "loss": 5.3684,
      "mean_token_accuracy": 0.19786085486412047,
      "num_tokens": 91720590.0,
      "step": 45270
    },
    {
      "entropy": 5.906509685516357,
      "epoch": 3.9565673337411518,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003509802610774064,
      "loss": 5.3927,
      "mean_token_accuracy": 0.19792972803115844,
      "num_tokens": 91729798.0,
      "step": 45275
    },
    {
      "entropy": 5.930422830581665,
      "epoch": 3.9570042820938562,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003509509297357769,
      "loss": 5.409,
      "mean_token_accuracy": 0.18921822011470796,
      "num_tokens": 91740498.0,
      "step": 45280
    },
    {
      "entropy": 5.9351170539855955,
      "epoch": 3.957441230446561,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003509215969373687,
      "loss": 5.4205,
      "mean_token_accuracy": 0.1988382950425148,
      "num_tokens": 91750659.0,
      "step": 45285
    },
    {
      "entropy": 5.838568687438965,
      "epoch": 3.957878178799266,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035089226268274453,
      "loss": 5.3641,
      "mean_token_accuracy": 0.1983174726366997,
      "num_tokens": 91760426.0,
      "step": 45290
    },
    {
      "entropy": 5.891507768630982,
      "epoch": 3.9583151271519705,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035086292697246696,
      "loss": 5.3695,
      "mean_token_accuracy": 0.20216688960790635,
      "num_tokens": 91769339.0,
      "step": 45295
    },
    {
      "entropy": 5.925682020187378,
      "epoch": 3.9587520755046754,
      "grad_norm": 1.640625,
      "learning_rate": 0.00035083358980709877,
      "loss": 5.4445,
      "mean_token_accuracy": 0.19372064024209976,
      "num_tokens": 91779326.0,
      "step": 45300
    },
    {
      "entropy": 5.981617784500122,
      "epoch": 3.95918902385738,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003508042511872026,
      "loss": 5.3703,
      "mean_token_accuracy": 0.19445003420114518,
      "num_tokens": 91789626.0,
      "step": 45305
    },
    {
      "entropy": 5.842026138305664,
      "epoch": 3.959625972210085,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000350774911113341,
      "loss": 5.3687,
      "mean_token_accuracy": 0.19797832071781157,
      "num_tokens": 91799903.0,
      "step": 45310
    },
    {
      "entropy": 5.85169038772583,
      "epoch": 3.9600629205627893,
      "grad_norm": 1.203125,
      "learning_rate": 0.000350745569586077,
      "loss": 5.3527,
      "mean_token_accuracy": 0.2027351215481758,
      "num_tokens": 91810192.0,
      "step": 45315
    },
    {
      "entropy": 5.868295907974243,
      "epoch": 3.960499868915494,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003507162266059733,
      "loss": 5.3679,
      "mean_token_accuracy": 0.1956479847431183,
      "num_tokens": 91819213.0,
      "step": 45320
    },
    {
      "entropy": 5.9513258934021,
      "epoch": 3.960936817268199,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035068688217359276,
      "loss": 5.469,
      "mean_token_accuracy": 0.19250486940145492,
      "num_tokens": 91829163.0,
      "step": 45325
    },
    {
      "entropy": 5.935011386871338,
      "epoch": 3.9613737656209036,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003506575362894981,
      "loss": 5.4031,
      "mean_token_accuracy": 0.2023719921708107,
      "num_tokens": 91838886.0,
      "step": 45330
    },
    {
      "entropy": 5.847595643997193,
      "epoch": 3.961810713973608,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035062818895425217,
      "loss": 5.3313,
      "mean_token_accuracy": 0.19789474755525588,
      "num_tokens": 91848279.0,
      "step": 45335
    },
    {
      "entropy": 5.822739028930664,
      "epoch": 3.962247662326313,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003505988401684181,
      "loss": 5.3829,
      "mean_token_accuracy": 0.1943417012691498,
      "num_tokens": 91858560.0,
      "step": 45340
    },
    {
      "entropy": 5.955821990966797,
      "epoch": 3.962684610679018,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003505694899325586,
      "loss": 5.3914,
      "mean_token_accuracy": 0.18793023973703385,
      "num_tokens": 91868381.0,
      "step": 45345
    },
    {
      "entropy": 5.95186414718628,
      "epoch": 3.9631215590317224,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003505401382472366,
      "loss": 5.4934,
      "mean_token_accuracy": 0.18841551542282103,
      "num_tokens": 91879512.0,
      "step": 45350
    },
    {
      "entropy": 5.86898980140686,
      "epoch": 3.9635585073844273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035051078511301524,
      "loss": 5.343,
      "mean_token_accuracy": 0.19920626580715178,
      "num_tokens": 91888884.0,
      "step": 45355
    },
    {
      "entropy": 5.805492639541626,
      "epoch": 3.9639954557371317,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003504814305304575,
      "loss": 5.3277,
      "mean_token_accuracy": 0.20825402736663817,
      "num_tokens": 91899029.0,
      "step": 45360
    },
    {
      "entropy": 5.935414981842041,
      "epoch": 3.9644324040898367,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00035045207450012635,
      "loss": 5.4676,
      "mean_token_accuracy": 0.19235997647047043,
      "num_tokens": 91909216.0,
      "step": 45365
    },
    {
      "entropy": 5.986013793945313,
      "epoch": 3.964869352442541,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035042271702258483,
      "loss": 5.443,
      "mean_token_accuracy": 0.18744683861732483,
      "num_tokens": 91920175.0,
      "step": 45370
    },
    {
      "entropy": 5.899414539337158,
      "epoch": 3.965306300795246,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003503933580983961,
      "loss": 5.3477,
      "mean_token_accuracy": 0.1940680682659149,
      "num_tokens": 91929916.0,
      "step": 45375
    },
    {
      "entropy": 5.8558125495910645,
      "epoch": 3.965743249147951,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00035036399772812323,
      "loss": 5.3299,
      "mean_token_accuracy": 0.19864291101694107,
      "num_tokens": 91939063.0,
      "step": 45380
    },
    {
      "entropy": 5.838268661499024,
      "epoch": 3.9661801975006554,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035033463591232946,
      "loss": 5.3946,
      "mean_token_accuracy": 0.19905180484056473,
      "num_tokens": 91949336.0,
      "step": 45385
    },
    {
      "entropy": 5.857119226455689,
      "epoch": 3.96661714585336,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003503052726515778,
      "loss": 5.3506,
      "mean_token_accuracy": 0.19814094305038452,
      "num_tokens": 91958930.0,
      "step": 45390
    },
    {
      "entropy": 5.923067378997803,
      "epoch": 3.967054094206065,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003502759079464316,
      "loss": 5.4171,
      "mean_token_accuracy": 0.19627493470907212,
      "num_tokens": 91970761.0,
      "step": 45395
    },
    {
      "entropy": 5.902187395095825,
      "epoch": 3.9674910425587697,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003502465417974541,
      "loss": 5.4622,
      "mean_token_accuracy": 0.185200235247612,
      "num_tokens": 91981280.0,
      "step": 45400
    },
    {
      "entropy": 5.889195299148559,
      "epoch": 3.967927990911474,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003502171742052084,
      "loss": 5.3267,
      "mean_token_accuracy": 0.19798691868782042,
      "num_tokens": 91990816.0,
      "step": 45405
    },
    {
      "entropy": 5.977925872802734,
      "epoch": 3.968364939264179,
      "grad_norm": 1.125,
      "learning_rate": 0.00035018780517025795,
      "loss": 5.4277,
      "mean_token_accuracy": 0.19384509027004243,
      "num_tokens": 92000183.0,
      "step": 45410
    },
    {
      "entropy": 5.886274242401123,
      "epoch": 3.9688018876168836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035015843469316595,
      "loss": 5.4322,
      "mean_token_accuracy": 0.194427227973938,
      "num_tokens": 92010597.0,
      "step": 45415
    },
    {
      "entropy": 5.8966835021972654,
      "epoch": 3.9692388359695885,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003501290627744959,
      "loss": 5.4526,
      "mean_token_accuracy": 0.19310102313756944,
      "num_tokens": 92021059.0,
      "step": 45420
    },
    {
      "entropy": 5.912199831008911,
      "epoch": 3.969675784322293,
      "grad_norm": 1.0,
      "learning_rate": 0.0003500996894148109,
      "loss": 5.4004,
      "mean_token_accuracy": 0.20462530106306076,
      "num_tokens": 92030942.0,
      "step": 45425
    },
    {
      "entropy": 5.872058868408203,
      "epoch": 3.970112732674998,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035007031461467455,
      "loss": 5.3284,
      "mean_token_accuracy": 0.20279690474271775,
      "num_tokens": 92039821.0,
      "step": 45430
    },
    {
      "entropy": 5.922363996505737,
      "epoch": 3.970549681027703,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003500409383746503,
      "loss": 5.4185,
      "mean_token_accuracy": 0.194613716006279,
      "num_tokens": 92049775.0,
      "step": 45435
    },
    {
      "entropy": 5.893516159057617,
      "epoch": 3.9709866293804073,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003500115606953015,
      "loss": 5.3603,
      "mean_token_accuracy": 0.1973370522260666,
      "num_tokens": 92059832.0,
      "step": 45440
    },
    {
      "entropy": 5.897378921508789,
      "epoch": 3.9714235777331117,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00034998218157719163,
      "loss": 5.3753,
      "mean_token_accuracy": 0.19498716443777084,
      "num_tokens": 92069842.0,
      "step": 45445
    },
    {
      "entropy": 5.845685577392578,
      "epoch": 3.9718605260858166,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003499528010208842,
      "loss": 5.4358,
      "mean_token_accuracy": 0.19403252005577087,
      "num_tokens": 92079488.0,
      "step": 45450
    },
    {
      "entropy": 5.8873741149902346,
      "epoch": 3.9722974744385215,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034992341902694284,
      "loss": 5.3939,
      "mean_token_accuracy": 0.20299759209156037,
      "num_tokens": 92088747.0,
      "step": 45455
    },
    {
      "entropy": 5.894366216659546,
      "epoch": 3.972734422791226,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034989403559593105,
      "loss": 5.3616,
      "mean_token_accuracy": 0.2014786571264267,
      "num_tokens": 92099092.0,
      "step": 45460
    },
    {
      "entropy": 5.833978843688965,
      "epoch": 3.973171371143931,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034986465072841237,
      "loss": 5.4478,
      "mean_token_accuracy": 0.19718173146247864,
      "num_tokens": 92108463.0,
      "step": 45465
    },
    {
      "entropy": 5.870119285583496,
      "epoch": 3.9736083194966354,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003498352644249506,
      "loss": 5.3515,
      "mean_token_accuracy": 0.20543995797634124,
      "num_tokens": 92117768.0,
      "step": 45470
    },
    {
      "entropy": 5.899443292617798,
      "epoch": 3.9740452678493403,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003498058766861091,
      "loss": 5.3821,
      "mean_token_accuracy": 0.1990060418844223,
      "num_tokens": 92127487.0,
      "step": 45475
    },
    {
      "entropy": 5.850205564498902,
      "epoch": 3.974482216202045,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003497764875124517,
      "loss": 5.2287,
      "mean_token_accuracy": 0.20016420334577562,
      "num_tokens": 92138154.0,
      "step": 45480
    },
    {
      "entropy": 5.893602991104126,
      "epoch": 3.9749191645547497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003497470969045421,
      "loss": 5.374,
      "mean_token_accuracy": 0.1926581457257271,
      "num_tokens": 92147591.0,
      "step": 45485
    },
    {
      "entropy": 5.856500673294067,
      "epoch": 3.9753561129074546,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034971770486294404,
      "loss": 5.3613,
      "mean_token_accuracy": 0.20456109941005707,
      "num_tokens": 92157621.0,
      "step": 45490
    },
    {
      "entropy": 5.924800062179566,
      "epoch": 3.975793061260159,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003496883113882212,
      "loss": 5.4828,
      "mean_token_accuracy": 0.18996369391679763,
      "num_tokens": 92168722.0,
      "step": 45495
    },
    {
      "entropy": 5.919185400009155,
      "epoch": 3.9762300096128635,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003496589164809374,
      "loss": 5.4452,
      "mean_token_accuracy": 0.20030816048383712,
      "num_tokens": 92179319.0,
      "step": 45500
    },
    {
      "entropy": 5.979982852935791,
      "epoch": 3.9766669579655685,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003496295201416565,
      "loss": 5.458,
      "mean_token_accuracy": 0.19684595763683319,
      "num_tokens": 92191307.0,
      "step": 45505
    },
    {
      "entropy": 5.928024959564209,
      "epoch": 3.9771039063182734,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034960012237094236,
      "loss": 5.4111,
      "mean_token_accuracy": 0.1924868866801262,
      "num_tokens": 92201055.0,
      "step": 45510
    },
    {
      "entropy": 5.8837579727172855,
      "epoch": 3.977540854670978,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00034957072316935865,
      "loss": 5.4185,
      "mean_token_accuracy": 0.1944488987326622,
      "num_tokens": 92211044.0,
      "step": 45515
    },
    {
      "entropy": 5.914003467559814,
      "epoch": 3.9779778030236828,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003495413225374695,
      "loss": 5.4096,
      "mean_token_accuracy": 0.19797960370779039,
      "num_tokens": 92221350.0,
      "step": 45520
    },
    {
      "entropy": 5.876520490646362,
      "epoch": 3.9784147513763872,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034951192047583867,
      "loss": 5.3434,
      "mean_token_accuracy": 0.20093470215797424,
      "num_tokens": 92230900.0,
      "step": 45525
    },
    {
      "entropy": 5.905517339706421,
      "epoch": 3.978851699729092,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034948251698503017,
      "loss": 5.4517,
      "mean_token_accuracy": 0.18545383363962173,
      "num_tokens": 92241715.0,
      "step": 45530
    },
    {
      "entropy": 5.973772478103638,
      "epoch": 3.9792886480817966,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000349453112065608,
      "loss": 5.4881,
      "mean_token_accuracy": 0.18691618442535402,
      "num_tokens": 92252007.0,
      "step": 45535
    },
    {
      "entropy": 5.94263710975647,
      "epoch": 3.9797255964345015,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003494237057181361,
      "loss": 5.4792,
      "mean_token_accuracy": 0.19344232827425004,
      "num_tokens": 92264377.0,
      "step": 45540
    },
    {
      "entropy": 5.980426454544068,
      "epoch": 3.980162544787206,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003493942979431786,
      "loss": 5.4028,
      "mean_token_accuracy": 0.1997259333729744,
      "num_tokens": 92275339.0,
      "step": 45545
    },
    {
      "entropy": 5.942210865020752,
      "epoch": 3.980599493139911,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034936488874129937,
      "loss": 5.3674,
      "mean_token_accuracy": 0.19414448887109756,
      "num_tokens": 92285923.0,
      "step": 45550
    },
    {
      "entropy": 5.884775829315186,
      "epoch": 3.9810364414926154,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003493354781130626,
      "loss": 5.4379,
      "mean_token_accuracy": 0.1933260217308998,
      "num_tokens": 92295148.0,
      "step": 45555
    },
    {
      "entropy": 5.970304155349732,
      "epoch": 3.9814733898453203,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034930606605903256,
      "loss": 5.3895,
      "mean_token_accuracy": 0.1937657430768013,
      "num_tokens": 92306074.0,
      "step": 45560
    },
    {
      "entropy": 5.8546144485473635,
      "epoch": 3.981910338198025,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003492766525797731,
      "loss": 5.3463,
      "mean_token_accuracy": 0.19955486804246902,
      "num_tokens": 92315772.0,
      "step": 45565
    },
    {
      "entropy": 5.927604246139526,
      "epoch": 3.9823472865507297,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034924723767584847,
      "loss": 5.3926,
      "mean_token_accuracy": 0.19862044155597686,
      "num_tokens": 92326637.0,
      "step": 45570
    },
    {
      "entropy": 5.927302694320678,
      "epoch": 3.982784234903434,
      "grad_norm": 0.96875,
      "learning_rate": 0.00034921782134782306,
      "loss": 5.3994,
      "mean_token_accuracy": 0.1979145035147667,
      "num_tokens": 92337436.0,
      "step": 45575
    },
    {
      "entropy": 5.873194599151612,
      "epoch": 3.983221183256139,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00034918840359626085,
      "loss": 5.4094,
      "mean_token_accuracy": 0.19034309685230255,
      "num_tokens": 92348548.0,
      "step": 45580
    },
    {
      "entropy": 5.874364042282105,
      "epoch": 3.983658131608844,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003491589844217263,
      "loss": 5.3808,
      "mean_token_accuracy": 0.19462193101644515,
      "num_tokens": 92359315.0,
      "step": 45585
    },
    {
      "entropy": 5.906214714050293,
      "epoch": 3.9840950799615484,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003491295638247835,
      "loss": 5.4474,
      "mean_token_accuracy": 0.19549164324998855,
      "num_tokens": 92369815.0,
      "step": 45590
    },
    {
      "entropy": 5.97185206413269,
      "epoch": 3.9845320283142533,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003491001418059968,
      "loss": 5.4341,
      "mean_token_accuracy": 0.19152670055627824,
      "num_tokens": 92380658.0,
      "step": 45595
    },
    {
      "entropy": 5.951324415206909,
      "epoch": 3.984968976666958,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003490707183659306,
      "loss": 5.4412,
      "mean_token_accuracy": 0.19267913848161697,
      "num_tokens": 92389808.0,
      "step": 45600
    },
    {
      "entropy": 5.900061368942261,
      "epoch": 3.9854059250196627,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00034904129350514913,
      "loss": 5.3914,
      "mean_token_accuracy": 0.19080897122621537,
      "num_tokens": 92399951.0,
      "step": 45605
    },
    {
      "entropy": 5.854861450195313,
      "epoch": 3.985842873372367,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003490118672242169,
      "loss": 5.4175,
      "mean_token_accuracy": 0.19605818688869475,
      "num_tokens": 92411157.0,
      "step": 45610
    },
    {
      "entropy": 5.940423059463501,
      "epoch": 3.986279821725072,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034898243952369827,
      "loss": 5.322,
      "mean_token_accuracy": 0.1995625913143158,
      "num_tokens": 92421250.0,
      "step": 45615
    },
    {
      "entropy": 5.954822778701782,
      "epoch": 3.986716770077777,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003489530104041577,
      "loss": 5.5245,
      "mean_token_accuracy": 0.1892366662621498,
      "num_tokens": 92431664.0,
      "step": 45620
    },
    {
      "entropy": 5.926576852798462,
      "epoch": 3.9871537184304815,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003489235798661597,
      "loss": 5.4419,
      "mean_token_accuracy": 0.19788597673177719,
      "num_tokens": 92442362.0,
      "step": 45625
    },
    {
      "entropy": 5.9582435131073,
      "epoch": 3.987590666783186,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034889414791026865,
      "loss": 5.4199,
      "mean_token_accuracy": 0.19707939326763152,
      "num_tokens": 92452526.0,
      "step": 45630
    },
    {
      "entropy": 5.938048458099365,
      "epoch": 3.988027615135891,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034886471453704905,
      "loss": 5.465,
      "mean_token_accuracy": 0.18697733134031297,
      "num_tokens": 92463418.0,
      "step": 45635
    },
    {
      "entropy": 5.905625915527343,
      "epoch": 3.988464563488596,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034883527974706566,
      "loss": 5.3493,
      "mean_token_accuracy": 0.19733744412660598,
      "num_tokens": 92473814.0,
      "step": 45640
    },
    {
      "entropy": 5.880172824859619,
      "epoch": 3.9889015118413003,
      "grad_norm": 0.96875,
      "learning_rate": 0.00034880584354088275,
      "loss": 5.4228,
      "mean_token_accuracy": 0.2028358221054077,
      "num_tokens": 92484234.0,
      "step": 45645
    },
    {
      "entropy": 5.920565271377564,
      "epoch": 3.989338460194005,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034877640591906526,
      "loss": 5.4932,
      "mean_token_accuracy": 0.18296458423137665,
      "num_tokens": 92494746.0,
      "step": 45650
    },
    {
      "entropy": 5.925007629394531,
      "epoch": 3.9897754085467096,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034874696688217755,
      "loss": 5.4131,
      "mean_token_accuracy": 0.1959605932235718,
      "num_tokens": 92505890.0,
      "step": 45655
    },
    {
      "entropy": 5.926703596115113,
      "epoch": 3.9902123568994146,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00034871752643078434,
      "loss": 5.4174,
      "mean_token_accuracy": 0.1930103987455368,
      "num_tokens": 92517593.0,
      "step": 45660
    },
    {
      "entropy": 5.963824510574341,
      "epoch": 3.990649305252119,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003486880845654504,
      "loss": 5.3675,
      "mean_token_accuracy": 0.18779600262641907,
      "num_tokens": 92528426.0,
      "step": 45665
    },
    {
      "entropy": 5.884647703170776,
      "epoch": 3.991086253604824,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034865864128674035,
      "loss": 5.3885,
      "mean_token_accuracy": 0.19629524648189545,
      "num_tokens": 92538946.0,
      "step": 45670
    },
    {
      "entropy": 5.853464078903198,
      "epoch": 3.991523201957529,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003486291965952191,
      "loss": 5.4225,
      "mean_token_accuracy": 0.19450048059225084,
      "num_tokens": 92548772.0,
      "step": 45675
    },
    {
      "entropy": 5.997995758056641,
      "epoch": 3.9919601503102333,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003485997504914511,
      "loss": 5.5218,
      "mean_token_accuracy": 0.1922060430049896,
      "num_tokens": 92560412.0,
      "step": 45680
    },
    {
      "entropy": 5.8912672996521,
      "epoch": 3.992397098662938,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034857030297600136,
      "loss": 5.3396,
      "mean_token_accuracy": 0.20055245459079743,
      "num_tokens": 92570936.0,
      "step": 45685
    },
    {
      "entropy": 5.89818844795227,
      "epoch": 3.9928340470156427,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003485408540494347,
      "loss": 5.3859,
      "mean_token_accuracy": 0.19555462151765823,
      "num_tokens": 92581431.0,
      "step": 45690
    },
    {
      "entropy": 5.951638841629029,
      "epoch": 3.9932709953683476,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003485114037123159,
      "loss": 5.4351,
      "mean_token_accuracy": 0.1981665700674057,
      "num_tokens": 92590787.0,
      "step": 45695
    },
    {
      "entropy": 5.868102025985718,
      "epoch": 3.993707943721052,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003484819519652099,
      "loss": 5.3452,
      "mean_token_accuracy": 0.20087268203496933,
      "num_tokens": 92600473.0,
      "step": 45700
    },
    {
      "entropy": 5.971733474731446,
      "epoch": 3.994144892073757,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003484524988086815,
      "loss": 5.4673,
      "mean_token_accuracy": 0.19307559132575988,
      "num_tokens": 92611201.0,
      "step": 45705
    },
    {
      "entropy": 5.962908601760864,
      "epoch": 3.9945818404264615,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034842304424329576,
      "loss": 5.3871,
      "mean_token_accuracy": 0.19291191250085832,
      "num_tokens": 92620836.0,
      "step": 45710
    },
    {
      "entropy": 5.928276109695434,
      "epoch": 3.9950187887791664,
      "grad_norm": 1.0,
      "learning_rate": 0.0003483935882696175,
      "loss": 5.434,
      "mean_token_accuracy": 0.19243661016225816,
      "num_tokens": 92630372.0,
      "step": 45715
    },
    {
      "entropy": 5.806492567062378,
      "epoch": 3.995455737131871,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003483641308882117,
      "loss": 5.3203,
      "mean_token_accuracy": 0.20301499962806702,
      "num_tokens": 92640281.0,
      "step": 45720
    },
    {
      "entropy": 5.9393548488616945,
      "epoch": 3.9958926854845758,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003483346720996435,
      "loss": 5.3701,
      "mean_token_accuracy": 0.19606555849313737,
      "num_tokens": 92650505.0,
      "step": 45725
    },
    {
      "entropy": 5.900836420059204,
      "epoch": 3.9963296338372807,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003483052119044779,
      "loss": 5.4024,
      "mean_token_accuracy": 0.19653053283691407,
      "num_tokens": 92660240.0,
      "step": 45730
    },
    {
      "entropy": 5.845693206787109,
      "epoch": 3.996766582189985,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003482757503032799,
      "loss": 5.3665,
      "mean_token_accuracy": 0.20014799982309342,
      "num_tokens": 92670491.0,
      "step": 45735
    },
    {
      "entropy": 5.949145317077637,
      "epoch": 3.9972035305426896,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003482462872966146,
      "loss": 5.391,
      "mean_token_accuracy": 0.19419850558042526,
      "num_tokens": 92680660.0,
      "step": 45740
    },
    {
      "entropy": 5.839917469024658,
      "epoch": 3.9976404788953945,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034821682288504704,
      "loss": 5.3101,
      "mean_token_accuracy": 0.2049190416932106,
      "num_tokens": 92690750.0,
      "step": 45745
    },
    {
      "entropy": 5.865982341766357,
      "epoch": 3.9980774272480994,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003481873570691426,
      "loss": 5.3109,
      "mean_token_accuracy": 0.20184877514839172,
      "num_tokens": 92702003.0,
      "step": 45750
    },
    {
      "entropy": 5.912837219238281,
      "epoch": 3.998514375600804,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00034815788984946623,
      "loss": 5.3691,
      "mean_token_accuracy": 0.20490467250347139,
      "num_tokens": 92712480.0,
      "step": 45755
    },
    {
      "entropy": 5.825678968429566,
      "epoch": 3.998951323953509,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003481284212265832,
      "loss": 5.4575,
      "mean_token_accuracy": 0.19069256633520126,
      "num_tokens": 92723623.0,
      "step": 45760
    },
    {
      "entropy": 5.9282526016235355,
      "epoch": 3.9993882723062133,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003480989512010587,
      "loss": 5.3469,
      "mean_token_accuracy": 0.19962022304534913,
      "num_tokens": 92732778.0,
      "step": 45765
    },
    {
      "entropy": 5.967877054214478,
      "epoch": 3.999825220658918,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003480694797734581,
      "loss": 5.4286,
      "mean_token_accuracy": 0.19657745361328124,
      "num_tokens": 92742985.0,
      "step": 45770
    },
    {
      "entropy": 5.9746898174285885,
      "epoch": 4.000262169011623,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034804000694434655,
      "loss": 5.4006,
      "mean_token_accuracy": 0.19472036212682725,
      "num_tokens": 92752370.0,
      "step": 45775
    },
    {
      "entropy": 5.910995054244995,
      "epoch": 4.000699117364327,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003480105327142894,
      "loss": 5.4104,
      "mean_token_accuracy": 0.199201563000679,
      "num_tokens": 92763083.0,
      "step": 45780
    },
    {
      "entropy": 5.877367305755615,
      "epoch": 4.0011360657170325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034798105708385194,
      "loss": 5.3097,
      "mean_token_accuracy": 0.20722163915634156,
      "num_tokens": 92772492.0,
      "step": 45785
    },
    {
      "entropy": 5.888737535476684,
      "epoch": 4.001573014069737,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00034795158005359954,
      "loss": 5.4163,
      "mean_token_accuracy": 0.19261021614074708,
      "num_tokens": 92783260.0,
      "step": 45790
    },
    {
      "entropy": 5.872777223587036,
      "epoch": 4.0020099624224414,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034792210162409767,
      "loss": 5.3356,
      "mean_token_accuracy": 0.1999247908592224,
      "num_tokens": 92794269.0,
      "step": 45795
    },
    {
      "entropy": 5.934380722045899,
      "epoch": 4.002446910775147,
      "grad_norm": 1.0,
      "learning_rate": 0.00034789262179591167,
      "loss": 5.2976,
      "mean_token_accuracy": 0.19984249472618104,
      "num_tokens": 92804373.0,
      "step": 45800
    },
    {
      "entropy": 5.912981796264648,
      "epoch": 4.002883859127851,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000347863140569607,
      "loss": 5.3505,
      "mean_token_accuracy": 0.20037549883127212,
      "num_tokens": 92815626.0,
      "step": 45805
    },
    {
      "entropy": 5.9438752174377445,
      "epoch": 4.003320807480556,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034783365794574907,
      "loss": 5.4259,
      "mean_token_accuracy": 0.18797479718923568,
      "num_tokens": 92825976.0,
      "step": 45810
    },
    {
      "entropy": 5.923715829849243,
      "epoch": 4.00375775583326,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003478041739249035,
      "loss": 5.3509,
      "mean_token_accuracy": 0.19776637554168702,
      "num_tokens": 92835677.0,
      "step": 45815
    },
    {
      "entropy": 5.906292867660523,
      "epoch": 4.004194704185966,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00034777468850763564,
      "loss": 5.4181,
      "mean_token_accuracy": 0.19508682638406755,
      "num_tokens": 92846050.0,
      "step": 45820
    },
    {
      "entropy": 5.82356538772583,
      "epoch": 4.00463165253867,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00034774520169451113,
      "loss": 5.2309,
      "mean_token_accuracy": 0.20817957818508148,
      "num_tokens": 92856016.0,
      "step": 45825
    },
    {
      "entropy": 5.871140003204346,
      "epoch": 4.0050686008913745,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003477157134860955,
      "loss": 5.3942,
      "mean_token_accuracy": 0.19829757660627365,
      "num_tokens": 92866444.0,
      "step": 45830
    },
    {
      "entropy": 5.842125654220581,
      "epoch": 4.005505549244079,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034768622388295443,
      "loss": 5.2846,
      "mean_token_accuracy": 0.20969851613044738,
      "num_tokens": 92877329.0,
      "step": 45835
    },
    {
      "entropy": 5.925069046020508,
      "epoch": 4.005942497596784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003476567328856536,
      "loss": 5.3641,
      "mean_token_accuracy": 0.19596222192049026,
      "num_tokens": 92886988.0,
      "step": 45840
    },
    {
      "entropy": 5.897440576553345,
      "epoch": 4.006379445949489,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003476272404947584,
      "loss": 5.3176,
      "mean_token_accuracy": 0.19893451482057573,
      "num_tokens": 92896953.0,
      "step": 45845
    },
    {
      "entropy": 5.948982620239258,
      "epoch": 4.006816394302193,
      "grad_norm": 1.53125,
      "learning_rate": 0.00034759774671083477,
      "loss": 5.4333,
      "mean_token_accuracy": 0.19304893016815186,
      "num_tokens": 92906938.0,
      "step": 45850
    },
    {
      "entropy": 5.85231122970581,
      "epoch": 4.007253342654899,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034756825153444825,
      "loss": 5.2712,
      "mean_token_accuracy": 0.20821541100740432,
      "num_tokens": 92916160.0,
      "step": 45855
    },
    {
      "entropy": 5.877899980545044,
      "epoch": 4.007690291007603,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034753875496616466,
      "loss": 5.3108,
      "mean_token_accuracy": 0.20138412117958068,
      "num_tokens": 92925610.0,
      "step": 45860
    },
    {
      "entropy": 5.928073596954346,
      "epoch": 4.008127239360308,
      "grad_norm": 1.0,
      "learning_rate": 0.00034750925700654985,
      "loss": 5.3626,
      "mean_token_accuracy": 0.1931956022977829,
      "num_tokens": 92935281.0,
      "step": 45865
    },
    {
      "entropy": 5.934695148468018,
      "epoch": 4.008564187713012,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034747975765616934,
      "loss": 5.4189,
      "mean_token_accuracy": 0.19523862153291702,
      "num_tokens": 92945258.0,
      "step": 45870
    },
    {
      "entropy": 5.886038303375244,
      "epoch": 4.009001136065717,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003474502569155892,
      "loss": 5.2972,
      "mean_token_accuracy": 0.20146115571260453,
      "num_tokens": 92955118.0,
      "step": 45875
    },
    {
      "entropy": 5.884511184692383,
      "epoch": 4.009438084418422,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00034742075478537513,
      "loss": 5.3153,
      "mean_token_accuracy": 0.20599410831928253,
      "num_tokens": 92964579.0,
      "step": 45880
    },
    {
      "entropy": 5.935675621032715,
      "epoch": 4.009875032771126,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000347391251266093,
      "loss": 5.3968,
      "mean_token_accuracy": 0.19378586262464523,
      "num_tokens": 92974479.0,
      "step": 45885
    },
    {
      "entropy": 5.891547441482544,
      "epoch": 4.010311981123831,
      "grad_norm": 1.0,
      "learning_rate": 0.0003473617463583088,
      "loss": 5.3217,
      "mean_token_accuracy": 0.19537819921970367,
      "num_tokens": 92985392.0,
      "step": 45890
    },
    {
      "entropy": 5.947485113143921,
      "epoch": 4.010748929476536,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00034733224006258835,
      "loss": 5.4155,
      "mean_token_accuracy": 0.18935210704803468,
      "num_tokens": 92995742.0,
      "step": 45895
    },
    {
      "entropy": 5.906776857376099,
      "epoch": 4.011185877829241,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034730273237949767,
      "loss": 5.3421,
      "mean_token_accuracy": 0.20044804215431214,
      "num_tokens": 93006411.0,
      "step": 45900
    },
    {
      "entropy": 5.898072052001953,
      "epoch": 4.011622826181945,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034727322330960263,
      "loss": 5.3976,
      "mean_token_accuracy": 0.2040848284959793,
      "num_tokens": 93017369.0,
      "step": 45905
    },
    {
      "entropy": 5.855786466598511,
      "epoch": 4.01205977453465,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034724371285346935,
      "loss": 5.2619,
      "mean_token_accuracy": 0.20628174245357514,
      "num_tokens": 93027989.0,
      "step": 45910
    },
    {
      "entropy": 5.900625085830688,
      "epoch": 4.012496722887355,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00034721420101166384,
      "loss": 5.3412,
      "mean_token_accuracy": 0.20394391417503357,
      "num_tokens": 93037717.0,
      "step": 45915
    },
    {
      "entropy": 5.897967290878296,
      "epoch": 4.012933671240059,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003471846877847521,
      "loss": 5.3318,
      "mean_token_accuracy": 0.2000599607825279,
      "num_tokens": 93046879.0,
      "step": 45920
    },
    {
      "entropy": 5.9355920314788815,
      "epoch": 4.013370619592764,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034715517317330017,
      "loss": 5.4118,
      "mean_token_accuracy": 0.19045657366514207,
      "num_tokens": 93058178.0,
      "step": 45925
    },
    {
      "entropy": 5.975266790390014,
      "epoch": 4.013807567945469,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003471256571778742,
      "loss": 5.3395,
      "mean_token_accuracy": 0.19724259674549102,
      "num_tokens": 93068001.0,
      "step": 45930
    },
    {
      "entropy": 5.859313249588013,
      "epoch": 4.014244516298174,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034709613979904037,
      "loss": 5.3173,
      "mean_token_accuracy": 0.20448849648237227,
      "num_tokens": 93077527.0,
      "step": 45935
    },
    {
      "entropy": 5.854251146316528,
      "epoch": 4.014681464650878,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00034706662103736486,
      "loss": 5.3785,
      "mean_token_accuracy": 0.19564447849988936,
      "num_tokens": 93087242.0,
      "step": 45940
    },
    {
      "entropy": 5.9467082023620605,
      "epoch": 4.015118413003583,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034703710089341365,
      "loss": 5.4199,
      "mean_token_accuracy": 0.2005511924624443,
      "num_tokens": 93097553.0,
      "step": 45945
    },
    {
      "entropy": 5.948889112472534,
      "epoch": 4.015555361356288,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003470075793677532,
      "loss": 5.326,
      "mean_token_accuracy": 0.19995069354772568,
      "num_tokens": 93107433.0,
      "step": 45950
    },
    {
      "entropy": 5.837522077560425,
      "epoch": 4.0159923097089925,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003469780564609496,
      "loss": 5.2656,
      "mean_token_accuracy": 0.2033279210329056,
      "num_tokens": 93117643.0,
      "step": 45955
    },
    {
      "entropy": 5.8433003425598145,
      "epoch": 4.016429258061697,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034694853217356924,
      "loss": 5.3734,
      "mean_token_accuracy": 0.20086101740598677,
      "num_tokens": 93128134.0,
      "step": 45960
    },
    {
      "entropy": 5.919556760787964,
      "epoch": 4.016866206414401,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034691900650617825,
      "loss": 5.3101,
      "mean_token_accuracy": 0.20460173338651658,
      "num_tokens": 93138194.0,
      "step": 45965
    },
    {
      "entropy": 5.916480350494385,
      "epoch": 4.017303154767107,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000346889479459343,
      "loss": 5.3406,
      "mean_token_accuracy": 0.2013170078396797,
      "num_tokens": 93148015.0,
      "step": 45970
    },
    {
      "entropy": 5.820306730270386,
      "epoch": 4.017740103119811,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003468599510336299,
      "loss": 5.2946,
      "mean_token_accuracy": 0.1997912496328354,
      "num_tokens": 93157512.0,
      "step": 45975
    },
    {
      "entropy": 5.83763279914856,
      "epoch": 4.018177051472516,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034683042122960535,
      "loss": 5.3721,
      "mean_token_accuracy": 0.20284077525138855,
      "num_tokens": 93168863.0,
      "step": 45980
    },
    {
      "entropy": 5.900974225997925,
      "epoch": 4.018613999825221,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034680089004783564,
      "loss": 5.411,
      "mean_token_accuracy": 0.19926753789186477,
      "num_tokens": 93179375.0,
      "step": 45985
    },
    {
      "entropy": 5.959545516967774,
      "epoch": 4.0190509481779255,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003467713574888873,
      "loss": 5.3424,
      "mean_token_accuracy": 0.1982802003622055,
      "num_tokens": 93189517.0,
      "step": 45990
    },
    {
      "entropy": 5.890942192077636,
      "epoch": 4.01948789653063,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034674182355332663,
      "loss": 5.3545,
      "mean_token_accuracy": 0.19641682952642442,
      "num_tokens": 93199712.0,
      "step": 45995
    },
    {
      "entropy": 5.893698453903198,
      "epoch": 4.0199248448833345,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003467122882417202,
      "loss": 5.4119,
      "mean_token_accuracy": 0.19668287932872772,
      "num_tokens": 93209976.0,
      "step": 46000
    },
    {
      "entropy": 5.9336036205291744,
      "epoch": 4.02036179323604,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003466827515546345,
      "loss": 5.4177,
      "mean_token_accuracy": 0.19069361984729766,
      "num_tokens": 93220003.0,
      "step": 46005
    },
    {
      "entropy": 5.948287916183472,
      "epoch": 4.020798741588744,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034665321349263605,
      "loss": 5.3613,
      "mean_token_accuracy": 0.19859496504068375,
      "num_tokens": 93230220.0,
      "step": 46010
    },
    {
      "entropy": 5.895063781738282,
      "epoch": 4.021235689941449,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003466236740562915,
      "loss": 5.3495,
      "mean_token_accuracy": 0.19659701883792877,
      "num_tokens": 93241041.0,
      "step": 46015
    },
    {
      "entropy": 5.858892917633057,
      "epoch": 4.021672638294153,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003465941332461673,
      "loss": 5.2613,
      "mean_token_accuracy": 0.20488227903842926,
      "num_tokens": 93251200.0,
      "step": 46020
    },
    {
      "entropy": 5.897190237045288,
      "epoch": 4.022109586646859,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034656459106283013,
      "loss": 5.3944,
      "mean_token_accuracy": 0.1932107001543045,
      "num_tokens": 93260564.0,
      "step": 46025
    },
    {
      "entropy": 5.88404688835144,
      "epoch": 4.022546534999563,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003465350475068466,
      "loss": 5.2814,
      "mean_token_accuracy": 0.2102183312177658,
      "num_tokens": 93270101.0,
      "step": 46030
    },
    {
      "entropy": 5.898939085006714,
      "epoch": 4.0229834833522675,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003465055025787833,
      "loss": 5.3706,
      "mean_token_accuracy": 0.20464972853660585,
      "num_tokens": 93280551.0,
      "step": 46035
    },
    {
      "entropy": 5.87431788444519,
      "epoch": 4.023420431704973,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003464759562792071,
      "loss": 5.3053,
      "mean_token_accuracy": 0.20405578166246413,
      "num_tokens": 93289658.0,
      "step": 46040
    },
    {
      "entropy": 5.878075742721558,
      "epoch": 4.023857380057677,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003464464086086845,
      "loss": 5.3979,
      "mean_token_accuracy": 0.19749684631824493,
      "num_tokens": 93300402.0,
      "step": 46045
    },
    {
      "entropy": 5.874830436706543,
      "epoch": 4.024294328410382,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034641685956778246,
      "loss": 5.2847,
      "mean_token_accuracy": 0.20414580255746842,
      "num_tokens": 93310254.0,
      "step": 46050
    },
    {
      "entropy": 5.830680990219117,
      "epoch": 4.024731276763086,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003463873091570676,
      "loss": 5.3133,
      "mean_token_accuracy": 0.19528478682041167,
      "num_tokens": 93319745.0,
      "step": 46055
    },
    {
      "entropy": 5.821776628494263,
      "epoch": 4.025168225115792,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034635775737710666,
      "loss": 5.3143,
      "mean_token_accuracy": 0.20326584875583648,
      "num_tokens": 93329218.0,
      "step": 46060
    },
    {
      "entropy": 5.913371276855469,
      "epoch": 4.025605173468496,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003463282042284667,
      "loss": 5.3792,
      "mean_token_accuracy": 0.19741162061691284,
      "num_tokens": 93339140.0,
      "step": 46065
    },
    {
      "entropy": 5.927411603927612,
      "epoch": 4.026042121821201,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034629864971171426,
      "loss": 5.354,
      "mean_token_accuracy": 0.19751718789339065,
      "num_tokens": 93348325.0,
      "step": 46070
    },
    {
      "entropy": 5.928796195983887,
      "epoch": 4.026479070173905,
      "grad_norm": 1.0,
      "learning_rate": 0.0003462690938274164,
      "loss": 5.3954,
      "mean_token_accuracy": 0.19625378996133805,
      "num_tokens": 93359050.0,
      "step": 46075
    },
    {
      "entropy": 5.897868537902832,
      "epoch": 4.02691601852661,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034623953657614,
      "loss": 5.3306,
      "mean_token_accuracy": 0.2019113779067993,
      "num_tokens": 93369518.0,
      "step": 46080
    },
    {
      "entropy": 5.8287006378173825,
      "epoch": 4.027352966879315,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034620997795845195,
      "loss": 5.2673,
      "mean_token_accuracy": 0.21014174669981003,
      "num_tokens": 93380922.0,
      "step": 46085
    },
    {
      "entropy": 5.895564985275269,
      "epoch": 4.027789915232019,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003461804179749192,
      "loss": 5.3246,
      "mean_token_accuracy": 0.19876219779253007,
      "num_tokens": 93391196.0,
      "step": 46090
    },
    {
      "entropy": 5.934750080108643,
      "epoch": 4.028226863584725,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034615085662610874,
      "loss": 5.4068,
      "mean_token_accuracy": 0.19060669988393783,
      "num_tokens": 93401923.0,
      "step": 46095
    },
    {
      "entropy": 5.917505073547363,
      "epoch": 4.028663811937429,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003461212939125876,
      "loss": 5.3411,
      "mean_token_accuracy": 0.19830858707427979,
      "num_tokens": 93411940.0,
      "step": 46100
    },
    {
      "entropy": 5.959245586395264,
      "epoch": 4.029100760290134,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003460917298349227,
      "loss": 5.4323,
      "mean_token_accuracy": 0.19169917553663254,
      "num_tokens": 93422509.0,
      "step": 46105
    },
    {
      "entropy": 5.896391725540161,
      "epoch": 4.029537708642838,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003460621643936812,
      "loss": 5.3745,
      "mean_token_accuracy": 0.1982796683907509,
      "num_tokens": 93432426.0,
      "step": 46110
    },
    {
      "entropy": 5.874666357040406,
      "epoch": 4.0299746569955435,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034603259758943017,
      "loss": 5.3217,
      "mean_token_accuracy": 0.2066669464111328,
      "num_tokens": 93442572.0,
      "step": 46115
    },
    {
      "entropy": 5.920871448516846,
      "epoch": 4.030411605348248,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003460030294227367,
      "loss": 5.3054,
      "mean_token_accuracy": 0.20015025585889817,
      "num_tokens": 93452848.0,
      "step": 46120
    },
    {
      "entropy": 5.902816820144653,
      "epoch": 4.030848553700952,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000345973459894168,
      "loss": 5.3399,
      "mean_token_accuracy": 0.1975506380200386,
      "num_tokens": 93462793.0,
      "step": 46125
    },
    {
      "entropy": 5.8750526905059814,
      "epoch": 4.031285502053657,
      "grad_norm": 1.015625,
      "learning_rate": 0.000345943889004291,
      "loss": 5.3783,
      "mean_token_accuracy": 0.20123843997716903,
      "num_tokens": 93472634.0,
      "step": 46130
    },
    {
      "entropy": 5.857318687438965,
      "epoch": 4.031722450406362,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034591431675367305,
      "loss": 5.2914,
      "mean_token_accuracy": 0.20555586963891984,
      "num_tokens": 93482299.0,
      "step": 46135
    },
    {
      "entropy": 5.868207311630249,
      "epoch": 4.032159398759067,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034588474314288135,
      "loss": 5.3422,
      "mean_token_accuracy": 0.205709744989872,
      "num_tokens": 93492675.0,
      "step": 46140
    },
    {
      "entropy": 5.873388147354126,
      "epoch": 4.032596347111771,
      "grad_norm": 1.125,
      "learning_rate": 0.00034585516817248326,
      "loss": 5.2774,
      "mean_token_accuracy": 0.2023654580116272,
      "num_tokens": 93502256.0,
      "step": 46145
    },
    {
      "entropy": 5.89687066078186,
      "epoch": 4.0330332954644765,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034582559184304583,
      "loss": 5.4191,
      "mean_token_accuracy": 0.19607858657836913,
      "num_tokens": 93512282.0,
      "step": 46150
    },
    {
      "entropy": 5.951145124435425,
      "epoch": 4.033470243817181,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003457960141551365,
      "loss": 5.4174,
      "mean_token_accuracy": 0.19541409313678743,
      "num_tokens": 93521916.0,
      "step": 46155
    },
    {
      "entropy": 5.943620443344116,
      "epoch": 4.0339071921698855,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003457664351093224,
      "loss": 5.3444,
      "mean_token_accuracy": 0.1960930734872818,
      "num_tokens": 93531126.0,
      "step": 46160
    },
    {
      "entropy": 5.826351690292358,
      "epoch": 4.03434414052259,
      "grad_norm": 1.015625,
      "learning_rate": 0.000345736854706171,
      "loss": 5.2651,
      "mean_token_accuracy": 0.2078768491744995,
      "num_tokens": 93540368.0,
      "step": 46165
    },
    {
      "entropy": 5.8137506484985355,
      "epoch": 4.034781088875295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003457072729462498,
      "loss": 5.2663,
      "mean_token_accuracy": 0.2030715137720108,
      "num_tokens": 93550213.0,
      "step": 46170
    },
    {
      "entropy": 5.873391103744507,
      "epoch": 4.035218037228,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034567768983012593,
      "loss": 5.3333,
      "mean_token_accuracy": 0.1998935654759407,
      "num_tokens": 93560673.0,
      "step": 46175
    },
    {
      "entropy": 5.877386856079101,
      "epoch": 4.035654985580704,
      "grad_norm": 1.03125,
      "learning_rate": 0.000345648105358367,
      "loss": 5.2885,
      "mean_token_accuracy": 0.2047559455037117,
      "num_tokens": 93571546.0,
      "step": 46180
    },
    {
      "entropy": 5.926168966293335,
      "epoch": 4.036091933933409,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034561851953154036,
      "loss": 5.3475,
      "mean_token_accuracy": 0.19303033053874968,
      "num_tokens": 93581440.0,
      "step": 46185
    },
    {
      "entropy": 5.844431924819946,
      "epoch": 4.036528882286114,
      "grad_norm": 1.0,
      "learning_rate": 0.0003455889323502135,
      "loss": 5.2649,
      "mean_token_accuracy": 0.20608422607183458,
      "num_tokens": 93591686.0,
      "step": 46190
    },
    {
      "entropy": 5.864978456497193,
      "epoch": 4.0369658306388185,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034555934381495394,
      "loss": 5.3805,
      "mean_token_accuracy": 0.20283995866775512,
      "num_tokens": 93602043.0,
      "step": 46195
    },
    {
      "entropy": 5.921495151519776,
      "epoch": 4.037402778991523,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034552975392632926,
      "loss": 5.3339,
      "mean_token_accuracy": 0.20307662934064866,
      "num_tokens": 93611639.0,
      "step": 46200
    },
    {
      "entropy": 5.899835681915283,
      "epoch": 4.0378397273442275,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034550016268490683,
      "loss": 5.3857,
      "mean_token_accuracy": 0.19549704045057298,
      "num_tokens": 93622236.0,
      "step": 46205
    },
    {
      "entropy": 5.839473581314087,
      "epoch": 4.038276675696933,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003454705700912544,
      "loss": 5.2466,
      "mean_token_accuracy": 0.20805730074644088,
      "num_tokens": 93632150.0,
      "step": 46210
    },
    {
      "entropy": 5.929097700119018,
      "epoch": 4.038713624049637,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003454409761459395,
      "loss": 5.4035,
      "mean_token_accuracy": 0.19491021186113358,
      "num_tokens": 93644843.0,
      "step": 46215
    },
    {
      "entropy": 5.902873849868774,
      "epoch": 4.039150572402342,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003454113808495298,
      "loss": 5.3649,
      "mean_token_accuracy": 0.19568041265010833,
      "num_tokens": 93654633.0,
      "step": 46220
    },
    {
      "entropy": 5.935441255569458,
      "epoch": 4.039587520755047,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00034538178420259286,
      "loss": 5.3512,
      "mean_token_accuracy": 0.1979626014828682,
      "num_tokens": 93664158.0,
      "step": 46225
    },
    {
      "entropy": 5.858552885055542,
      "epoch": 4.040024469107752,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003453521862056965,
      "loss": 5.3071,
      "mean_token_accuracy": 0.1939936563372612,
      "num_tokens": 93673845.0,
      "step": 46230
    },
    {
      "entropy": 5.884959411621094,
      "epoch": 4.040461417460456,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034532258685940825,
      "loss": 5.3164,
      "mean_token_accuracy": 0.19987592697143555,
      "num_tokens": 93683871.0,
      "step": 46235
    },
    {
      "entropy": 5.8635780811309814,
      "epoch": 4.0408983658131605,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000345292986164296,
      "loss": 5.3713,
      "mean_token_accuracy": 0.19488137513399123,
      "num_tokens": 93693652.0,
      "step": 46240
    },
    {
      "entropy": 5.938617801666259,
      "epoch": 4.041335314165866,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034526338412092745,
      "loss": 5.4024,
      "mean_token_accuracy": 0.19557632952928544,
      "num_tokens": 93702903.0,
      "step": 46245
    },
    {
      "entropy": 5.86034893989563,
      "epoch": 4.04177226251857,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003452337807298703,
      "loss": 5.2709,
      "mean_token_accuracy": 0.20482251942157745,
      "num_tokens": 93712328.0,
      "step": 46250
    },
    {
      "entropy": 5.8469236373901365,
      "epoch": 4.042209210871275,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003452041759916925,
      "loss": 5.3231,
      "mean_token_accuracy": 0.20355653911828994,
      "num_tokens": 93722399.0,
      "step": 46255
    },
    {
      "entropy": 5.924372434616089,
      "epoch": 4.042646159223979,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003451745699069618,
      "loss": 5.3803,
      "mean_token_accuracy": 0.197206911444664,
      "num_tokens": 93732034.0,
      "step": 46260
    },
    {
      "entropy": 5.852776145935058,
      "epoch": 4.043083107576685,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003451449624762461,
      "loss": 5.2516,
      "mean_token_accuracy": 0.20923105180263518,
      "num_tokens": 93741086.0,
      "step": 46265
    },
    {
      "entropy": 5.831730747222901,
      "epoch": 4.043520055929389,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003451153537001132,
      "loss": 5.2732,
      "mean_token_accuracy": 0.20397974848747252,
      "num_tokens": 93750297.0,
      "step": 46270
    },
    {
      "entropy": 5.873866033554077,
      "epoch": 4.043957004282094,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003450857435791311,
      "loss": 5.2909,
      "mean_token_accuracy": 0.20049781650304793,
      "num_tokens": 93760288.0,
      "step": 46275
    },
    {
      "entropy": 5.850627994537353,
      "epoch": 4.044393952634799,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034505613211386775,
      "loss": 5.2971,
      "mean_token_accuracy": 0.19995682388544084,
      "num_tokens": 93770025.0,
      "step": 46280
    },
    {
      "entropy": 5.848507308959961,
      "epoch": 4.044830900987503,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034502651930489104,
      "loss": 5.2669,
      "mean_token_accuracy": 0.20401553362607955,
      "num_tokens": 93779659.0,
      "step": 46285
    },
    {
      "entropy": 5.897688388824463,
      "epoch": 4.045267849340208,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034499690515276906,
      "loss": 5.3205,
      "mean_token_accuracy": 0.20058860629796982,
      "num_tokens": 93790820.0,
      "step": 46290
    },
    {
      "entropy": 5.786454439163208,
      "epoch": 4.045704797692912,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003449672896580697,
      "loss": 5.3291,
      "mean_token_accuracy": 0.20210572481155395,
      "num_tokens": 93801703.0,
      "step": 46295
    },
    {
      "entropy": 5.843166017532349,
      "epoch": 4.046141746045618,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003449376728213611,
      "loss": 5.323,
      "mean_token_accuracy": 0.20214481204748153,
      "num_tokens": 93812743.0,
      "step": 46300
    },
    {
      "entropy": 5.899038696289063,
      "epoch": 4.046578694398322,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034490805464321123,
      "loss": 5.3541,
      "mean_token_accuracy": 0.20496538430452346,
      "num_tokens": 93822825.0,
      "step": 46305
    },
    {
      "entropy": 5.914792728424072,
      "epoch": 4.047015642751027,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034487843512418827,
      "loss": 5.3457,
      "mean_token_accuracy": 0.19384820014238358,
      "num_tokens": 93832650.0,
      "step": 46310
    },
    {
      "entropy": 5.830446100234985,
      "epoch": 4.047452591103731,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034484881426486043,
      "loss": 5.3104,
      "mean_token_accuracy": 0.20409911572933198,
      "num_tokens": 93842689.0,
      "step": 46315
    },
    {
      "entropy": 5.855859613418579,
      "epoch": 4.0478895394564365,
      "grad_norm": 1.453125,
      "learning_rate": 0.00034481919206579546,
      "loss": 5.3059,
      "mean_token_accuracy": 0.20569882541894913,
      "num_tokens": 93852464.0,
      "step": 46320
    },
    {
      "entropy": 5.8478423118591305,
      "epoch": 4.048326487809141,
      "grad_norm": 1.0,
      "learning_rate": 0.000344789568527562,
      "loss": 5.2913,
      "mean_token_accuracy": 0.20767795592546462,
      "num_tokens": 93862241.0,
      "step": 46325
    },
    {
      "entropy": 5.921223878860474,
      "epoch": 4.048763436161845,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000344759943650728,
      "loss": 5.3714,
      "mean_token_accuracy": 0.19283149987459183,
      "num_tokens": 93872620.0,
      "step": 46330
    },
    {
      "entropy": 5.8357336044311525,
      "epoch": 4.049200384514551,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003447303174358618,
      "loss": 5.2609,
      "mean_token_accuracy": 0.21214157938957215,
      "num_tokens": 93882536.0,
      "step": 46335
    },
    {
      "entropy": 5.915112543106079,
      "epoch": 4.049637332867255,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034470068988353146,
      "loss": 5.2928,
      "mean_token_accuracy": 0.20617303997278214,
      "num_tokens": 93892072.0,
      "step": 46340
    },
    {
      "entropy": 5.87952151298523,
      "epoch": 4.05007428121996,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034467106099430537,
      "loss": 5.4214,
      "mean_token_accuracy": 0.19473652094602584,
      "num_tokens": 93902230.0,
      "step": 46345
    },
    {
      "entropy": 5.851502370834351,
      "epoch": 4.050511229572664,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034464143076875186,
      "loss": 5.3533,
      "mean_token_accuracy": 0.20144754648208618,
      "num_tokens": 93911617.0,
      "step": 46350
    },
    {
      "entropy": 5.8133186340332035,
      "epoch": 4.0509481779253695,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034461179920743925,
      "loss": 5.1569,
      "mean_token_accuracy": 0.21333660334348678,
      "num_tokens": 93922131.0,
      "step": 46355
    },
    {
      "entropy": 5.870093965530396,
      "epoch": 4.051385126278074,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034458216631093585,
      "loss": 5.3664,
      "mean_token_accuracy": 0.1904413253068924,
      "num_tokens": 93932186.0,
      "step": 46360
    },
    {
      "entropy": 5.922551488876342,
      "epoch": 4.0518220746307785,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034455253207981004,
      "loss": 5.4066,
      "mean_token_accuracy": 0.1952519088983536,
      "num_tokens": 93942925.0,
      "step": 46365
    },
    {
      "entropy": 5.884352159500122,
      "epoch": 4.052259022983483,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034452289651463016,
      "loss": 5.3701,
      "mean_token_accuracy": 0.1985226169228554,
      "num_tokens": 93954179.0,
      "step": 46370
    },
    {
      "entropy": 5.945185995101928,
      "epoch": 4.052695971336188,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034449325961596476,
      "loss": 5.401,
      "mean_token_accuracy": 0.19775014519691467,
      "num_tokens": 93964374.0,
      "step": 46375
    },
    {
      "entropy": 5.899231433868408,
      "epoch": 4.053132919688893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003444636213843822,
      "loss": 5.3036,
      "mean_token_accuracy": 0.19631452560424806,
      "num_tokens": 93974321.0,
      "step": 46380
    },
    {
      "entropy": 5.884684371948242,
      "epoch": 4.053569868041597,
      "grad_norm": 1.015625,
      "learning_rate": 0.000344433981820451,
      "loss": 5.4105,
      "mean_token_accuracy": 0.197217221558094,
      "num_tokens": 93984614.0,
      "step": 46385
    },
    {
      "entropy": 5.978163623809815,
      "epoch": 4.054006816394303,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003444043409247396,
      "loss": 5.4253,
      "mean_token_accuracy": 0.19106272161006926,
      "num_tokens": 93994797.0,
      "step": 46390
    },
    {
      "entropy": 5.928600168228149,
      "epoch": 4.054443764747007,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003443746986978166,
      "loss": 5.3707,
      "mean_token_accuracy": 0.1960243195295334,
      "num_tokens": 94005383.0,
      "step": 46395
    },
    {
      "entropy": 5.799351501464844,
      "epoch": 4.0548807130997115,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034434505514025057,
      "loss": 5.2324,
      "mean_token_accuracy": 0.21277956366539003,
      "num_tokens": 94013951.0,
      "step": 46400
    },
    {
      "entropy": 5.903974866867065,
      "epoch": 4.055317661452416,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034431541025260997,
      "loss": 5.3851,
      "mean_token_accuracy": 0.20241405665874482,
      "num_tokens": 94026044.0,
      "step": 46405
    },
    {
      "entropy": 5.883876323699951,
      "epoch": 4.055754609805121,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034428576403546353,
      "loss": 5.3559,
      "mean_token_accuracy": 0.19931021183729172,
      "num_tokens": 94035543.0,
      "step": 46410
    },
    {
      "entropy": 5.845014810562134,
      "epoch": 4.056191558157826,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003442561164893798,
      "loss": 5.3474,
      "mean_token_accuracy": 0.1984977751970291,
      "num_tokens": 94044480.0,
      "step": 46415
    },
    {
      "entropy": 5.869383811950684,
      "epoch": 4.05662850651053,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00034422646761492746,
      "loss": 5.312,
      "mean_token_accuracy": 0.20735324770212174,
      "num_tokens": 94055233.0,
      "step": 46420
    },
    {
      "entropy": 5.913100290298462,
      "epoch": 4.057065454863235,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003441968174126752,
      "loss": 5.3655,
      "mean_token_accuracy": 0.19757342338562012,
      "num_tokens": 94067182.0,
      "step": 46425
    },
    {
      "entropy": 5.846365976333618,
      "epoch": 4.05750240321594,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003441671658831918,
      "loss": 5.3208,
      "mean_token_accuracy": 0.20521768629550935,
      "num_tokens": 94076799.0,
      "step": 46430
    },
    {
      "entropy": 5.838161039352417,
      "epoch": 4.057939351568645,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003441375130270458,
      "loss": 5.271,
      "mean_token_accuracy": 0.20199516117572786,
      "num_tokens": 94086829.0,
      "step": 46435
    },
    {
      "entropy": 5.837695837020874,
      "epoch": 4.058376299921349,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034410785884480616,
      "loss": 5.3104,
      "mean_token_accuracy": 0.20726627558469773,
      "num_tokens": 94097124.0,
      "step": 46440
    },
    {
      "entropy": 5.888335800170898,
      "epoch": 4.058813248274054,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034407820333704153,
      "loss": 5.3195,
      "mean_token_accuracy": 0.19878603219985963,
      "num_tokens": 94107714.0,
      "step": 46445
    },
    {
      "entropy": 5.847216844558716,
      "epoch": 4.059250196626759,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003440485465043208,
      "loss": 5.3738,
      "mean_token_accuracy": 0.19780069291591645,
      "num_tokens": 94118139.0,
      "step": 46450
    },
    {
      "entropy": 5.888811683654785,
      "epoch": 4.059687144979463,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034401888834721277,
      "loss": 5.3405,
      "mean_token_accuracy": 0.2022705301642418,
      "num_tokens": 94128452.0,
      "step": 46455
    },
    {
      "entropy": 5.882267093658447,
      "epoch": 4.060124093332168,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003439892288662863,
      "loss": 5.332,
      "mean_token_accuracy": 0.2011525109410286,
      "num_tokens": 94138195.0,
      "step": 46460
    },
    {
      "entropy": 5.888330745697021,
      "epoch": 4.060561041684873,
      "grad_norm": 0.96875,
      "learning_rate": 0.00034395956806211023,
      "loss": 5.3769,
      "mean_token_accuracy": 0.19217440038919448,
      "num_tokens": 94149261.0,
      "step": 46465
    },
    {
      "entropy": 5.9448529243469235,
      "epoch": 4.060997990037578,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003439299059352535,
      "loss": 5.3602,
      "mean_token_accuracy": 0.19649983793497086,
      "num_tokens": 94159067.0,
      "step": 46470
    },
    {
      "entropy": 5.881113958358765,
      "epoch": 4.061434938390282,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003439002424862851,
      "loss": 5.3556,
      "mean_token_accuracy": 0.19797480404376983,
      "num_tokens": 94169463.0,
      "step": 46475
    },
    {
      "entropy": 5.854767560958862,
      "epoch": 4.061871886742987,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003438705777157739,
      "loss": 5.3473,
      "mean_token_accuracy": 0.2020406812429428,
      "num_tokens": 94179325.0,
      "step": 46480
    },
    {
      "entropy": 5.891619491577148,
      "epoch": 4.062308835095692,
      "grad_norm": 1.09375,
      "learning_rate": 0.000343840911624289,
      "loss": 5.3536,
      "mean_token_accuracy": 0.19985452592372893,
      "num_tokens": 94189428.0,
      "step": 46485
    },
    {
      "entropy": 5.853329038619995,
      "epoch": 4.062745783448396,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003438112442123993,
      "loss": 5.3055,
      "mean_token_accuracy": 0.20807835906744004,
      "num_tokens": 94198995.0,
      "step": 46490
    },
    {
      "entropy": 5.869337701797486,
      "epoch": 4.063182731801101,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003437815754806738,
      "loss": 5.3178,
      "mean_token_accuracy": 0.19682986289262772,
      "num_tokens": 94209456.0,
      "step": 46495
    },
    {
      "entropy": 5.851643228530884,
      "epoch": 4.063619680153806,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034375190542968175,
      "loss": 5.3396,
      "mean_token_accuracy": 0.2055053874850273,
      "num_tokens": 94220390.0,
      "step": 46500
    },
    {
      "entropy": 5.916332340240478,
      "epoch": 4.064056628506511,
      "grad_norm": 1.515625,
      "learning_rate": 0.00034372223405999206,
      "loss": 5.3883,
      "mean_token_accuracy": 0.19841344505548478,
      "num_tokens": 94230904.0,
      "step": 46505
    },
    {
      "entropy": 5.921893215179443,
      "epoch": 4.064493576859215,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000343692561372174,
      "loss": 5.3272,
      "mean_token_accuracy": 0.20374177694320678,
      "num_tokens": 94240817.0,
      "step": 46510
    },
    {
      "entropy": 5.85297122001648,
      "epoch": 4.06493052521192,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034366288736679653,
      "loss": 5.3996,
      "mean_token_accuracy": 0.19898498803377151,
      "num_tokens": 94251972.0,
      "step": 46515
    },
    {
      "entropy": 5.893028259277344,
      "epoch": 4.065367473564625,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003436332120444289,
      "loss": 5.3657,
      "mean_token_accuracy": 0.20222037136554719,
      "num_tokens": 94261450.0,
      "step": 46520
    },
    {
      "entropy": 5.9252753257751465,
      "epoch": 4.0658044219173295,
      "grad_norm": 0.9375,
      "learning_rate": 0.00034360353540564045,
      "loss": 5.3463,
      "mean_token_accuracy": 0.1952725499868393,
      "num_tokens": 94271174.0,
      "step": 46525
    },
    {
      "entropy": 5.883075809478759,
      "epoch": 4.066241370270034,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034357385745100013,
      "loss": 5.277,
      "mean_token_accuracy": 0.2116043448448181,
      "num_tokens": 94281609.0,
      "step": 46530
    },
    {
      "entropy": 5.870708417892456,
      "epoch": 4.066678318622738,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034354417818107725,
      "loss": 5.3428,
      "mean_token_accuracy": 0.19418742507696152,
      "num_tokens": 94292704.0,
      "step": 46535
    },
    {
      "entropy": 5.822545051574707,
      "epoch": 4.067115266975444,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00034351449759644116,
      "loss": 5.2922,
      "mean_token_accuracy": 0.201900851726532,
      "num_tokens": 94302467.0,
      "step": 46540
    },
    {
      "entropy": 5.863503646850586,
      "epoch": 4.067552215328148,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003434848156976612,
      "loss": 5.2963,
      "mean_token_accuracy": 0.20416991263628007,
      "num_tokens": 94312365.0,
      "step": 46545
    },
    {
      "entropy": 5.840181732177735,
      "epoch": 4.067989163680853,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034345513248530647,
      "loss": 5.3179,
      "mean_token_accuracy": 0.20043861567974092,
      "num_tokens": 94323168.0,
      "step": 46550
    },
    {
      "entropy": 5.883028316497803,
      "epoch": 4.068426112033558,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003434254479599464,
      "loss": 5.3686,
      "mean_token_accuracy": 0.19989408254623414,
      "num_tokens": 94333901.0,
      "step": 46555
    },
    {
      "entropy": 5.879370641708374,
      "epoch": 4.0688630603862626,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003433957621221504,
      "loss": 5.3163,
      "mean_token_accuracy": 0.2041911542415619,
      "num_tokens": 94344565.0,
      "step": 46560
    },
    {
      "entropy": 5.876277494430542,
      "epoch": 4.069300008738967,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003433660749724879,
      "loss": 5.3254,
      "mean_token_accuracy": 0.20133761465549468,
      "num_tokens": 94354330.0,
      "step": 46565
    },
    {
      "entropy": 5.860617923736572,
      "epoch": 4.0697369570916715,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034333638651152826,
      "loss": 5.3672,
      "mean_token_accuracy": 0.1964808836579323,
      "num_tokens": 94364845.0,
      "step": 46570
    },
    {
      "entropy": 5.918743515014649,
      "epoch": 4.070173905444377,
      "grad_norm": 1.125,
      "learning_rate": 0.00034330669673984093,
      "loss": 5.3464,
      "mean_token_accuracy": 0.20326881110668182,
      "num_tokens": 94374932.0,
      "step": 46575
    },
    {
      "entropy": 5.842147922515869,
      "epoch": 4.070610853797081,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034327700565799535,
      "loss": 5.3564,
      "mean_token_accuracy": 0.20504407882690429,
      "num_tokens": 94385431.0,
      "step": 46580
    },
    {
      "entropy": 5.951453113555909,
      "epoch": 4.071047802149786,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034324731326656094,
      "loss": 5.4019,
      "mean_token_accuracy": 0.19197386503219604,
      "num_tokens": 94395727.0,
      "step": 46585
    },
    {
      "entropy": 5.876151514053345,
      "epoch": 4.07148475050249,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034321761956610734,
      "loss": 5.3617,
      "mean_token_accuracy": 0.19975212216377258,
      "num_tokens": 94406562.0,
      "step": 46590
    },
    {
      "entropy": 5.897050142288208,
      "epoch": 4.071921698855196,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034318792455720413,
      "loss": 5.3198,
      "mean_token_accuracy": 0.20486714243888854,
      "num_tokens": 94416462.0,
      "step": 46595
    },
    {
      "entropy": 5.901244211196899,
      "epoch": 4.0723586472079,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034315822824042074,
      "loss": 5.3021,
      "mean_token_accuracy": 0.20789612531661988,
      "num_tokens": 94426913.0,
      "step": 46600
    },
    {
      "entropy": 5.872449207305908,
      "epoch": 4.0727955955606046,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034312853061632686,
      "loss": 5.3048,
      "mean_token_accuracy": 0.2051673263311386,
      "num_tokens": 94436622.0,
      "step": 46605
    },
    {
      "entropy": 5.887452745437622,
      "epoch": 4.073232543913309,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000343098831685492,
      "loss": 5.3463,
      "mean_token_accuracy": 0.19785500764846803,
      "num_tokens": 94447027.0,
      "step": 46610
    },
    {
      "entropy": 5.93142728805542,
      "epoch": 4.073669492266014,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003430691314484859,
      "loss": 5.3456,
      "mean_token_accuracy": 0.19738933593034744,
      "num_tokens": 94456085.0,
      "step": 46615
    },
    {
      "entropy": 5.882568264007569,
      "epoch": 4.074106440618719,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003430394299058782,
      "loss": 5.4021,
      "mean_token_accuracy": 0.1968570739030838,
      "num_tokens": 94466497.0,
      "step": 46620
    },
    {
      "entropy": 5.881888580322266,
      "epoch": 4.074543388971423,
      "grad_norm": 1.0,
      "learning_rate": 0.00034300972705823856,
      "loss": 5.3709,
      "mean_token_accuracy": 0.1917018175125122,
      "num_tokens": 94477392.0,
      "step": 46625
    },
    {
      "entropy": 5.887857341766358,
      "epoch": 4.074980337324129,
      "grad_norm": 1.0,
      "learning_rate": 0.00034298002290613673,
      "loss": 5.3234,
      "mean_token_accuracy": 0.2083071693778038,
      "num_tokens": 94486620.0,
      "step": 46630
    },
    {
      "entropy": 5.831424856185913,
      "epoch": 4.075417285676833,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003429503174501424,
      "loss": 5.2502,
      "mean_token_accuracy": 0.2082109719514847,
      "num_tokens": 94497241.0,
      "step": 46635
    },
    {
      "entropy": 5.8773942470550535,
      "epoch": 4.075854234029538,
      "grad_norm": 1.125,
      "learning_rate": 0.0003429206106908254,
      "loss": 5.3306,
      "mean_token_accuracy": 0.20478457361459732,
      "num_tokens": 94505885.0,
      "step": 46640
    },
    {
      "entropy": 5.852011060714721,
      "epoch": 4.076291182382242,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034289090262875553,
      "loss": 5.3138,
      "mean_token_accuracy": 0.20375413149595262,
      "num_tokens": 94515985.0,
      "step": 46645
    },
    {
      "entropy": 5.9064921855926515,
      "epoch": 4.076728130734947,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003428611932645027,
      "loss": 5.382,
      "mean_token_accuracy": 0.1922437220811844,
      "num_tokens": 94526321.0,
      "step": 46650
    },
    {
      "entropy": 5.878824377059937,
      "epoch": 4.077165079087652,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003428314825986364,
      "loss": 5.3069,
      "mean_token_accuracy": 0.20097217857837676,
      "num_tokens": 94536366.0,
      "step": 46655
    },
    {
      "entropy": 5.873091506958008,
      "epoch": 4.077602027440356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003428017706317268,
      "loss": 5.338,
      "mean_token_accuracy": 0.2062996968626976,
      "num_tokens": 94546749.0,
      "step": 46660
    },
    {
      "entropy": 5.869973421096802,
      "epoch": 4.078038975793061,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034277205736434367,
      "loss": 5.356,
      "mean_token_accuracy": 0.2031790629029274,
      "num_tokens": 94558013.0,
      "step": 46665
    },
    {
      "entropy": 5.862815523147583,
      "epoch": 4.078475924145766,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000342742342797057,
      "loss": 5.3021,
      "mean_token_accuracy": 0.20117239654064178,
      "num_tokens": 94567417.0,
      "step": 46670
    },
    {
      "entropy": 5.810282754898071,
      "epoch": 4.078912872498471,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003427126269304367,
      "loss": 5.256,
      "mean_token_accuracy": 0.21113632172346114,
      "num_tokens": 94577598.0,
      "step": 46675
    },
    {
      "entropy": 5.87547664642334,
      "epoch": 4.079349820851175,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003426829097650527,
      "loss": 5.36,
      "mean_token_accuracy": 0.20151544511318206,
      "num_tokens": 94587260.0,
      "step": 46680
    },
    {
      "entropy": 5.915590763092041,
      "epoch": 4.0797867692038805,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034265319130147515,
      "loss": 5.3401,
      "mean_token_accuracy": 0.1970328986644745,
      "num_tokens": 94598266.0,
      "step": 46685
    },
    {
      "entropy": 5.859765100479126,
      "epoch": 4.080223717556585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003426234715402738,
      "loss": 5.3411,
      "mean_token_accuracy": 0.20558356046676635,
      "num_tokens": 94608468.0,
      "step": 46690
    },
    {
      "entropy": 5.892136573791504,
      "epoch": 4.080660665909289,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003425937504820189,
      "loss": 5.3529,
      "mean_token_accuracy": 0.2101845383644104,
      "num_tokens": 94618109.0,
      "step": 46695
    },
    {
      "entropy": 5.898291826248169,
      "epoch": 4.081097614261994,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034256402812728046,
      "loss": 5.4188,
      "mean_token_accuracy": 0.19385626465082167,
      "num_tokens": 94628763.0,
      "step": 46700
    },
    {
      "entropy": 5.881983470916748,
      "epoch": 4.081534562614699,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003425343044766285,
      "loss": 5.3972,
      "mean_token_accuracy": 0.19878243356943132,
      "num_tokens": 94638607.0,
      "step": 46705
    },
    {
      "entropy": 5.924099636077881,
      "epoch": 4.081971510967404,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003425045795306333,
      "loss": 5.3405,
      "mean_token_accuracy": 0.20219461023807525,
      "num_tokens": 94649691.0,
      "step": 46710
    },
    {
      "entropy": 5.879487991333008,
      "epoch": 4.082408459320108,
      "grad_norm": 0.953125,
      "learning_rate": 0.00034247485328986477,
      "loss": 5.3278,
      "mean_token_accuracy": 0.20214079171419144,
      "num_tokens": 94659754.0,
      "step": 46715
    },
    {
      "entropy": 5.893657398223877,
      "epoch": 4.082845407672813,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034244512575489324,
      "loss": 5.4261,
      "mean_token_accuracy": 0.19000620245933533,
      "num_tokens": 94670170.0,
      "step": 46720
    },
    {
      "entropy": 5.832329130172729,
      "epoch": 4.083282356025518,
      "grad_norm": 0.96875,
      "learning_rate": 0.00034241539692628887,
      "loss": 5.2315,
      "mean_token_accuracy": 0.21227014660835267,
      "num_tokens": 94680615.0,
      "step": 46725
    },
    {
      "entropy": 5.885263252258301,
      "epoch": 4.0837193043782225,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034238566680462184,
      "loss": 5.3447,
      "mean_token_accuracy": 0.19911225289106368,
      "num_tokens": 94690166.0,
      "step": 46730
    },
    {
      "entropy": 5.8907571792602536,
      "epoch": 4.084156252730927,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034235593539046243,
      "loss": 5.3448,
      "mean_token_accuracy": 0.20132976770401,
      "num_tokens": 94699953.0,
      "step": 46735
    },
    {
      "entropy": 5.818948698043823,
      "epoch": 4.084593201083632,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003423262026843809,
      "loss": 5.2222,
      "mean_token_accuracy": 0.21172589510679246,
      "num_tokens": 94710030.0,
      "step": 46740
    },
    {
      "entropy": 5.8260256290435795,
      "epoch": 4.085030149436337,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00034229646868694745,
      "loss": 5.3259,
      "mean_token_accuracy": 0.2011576309800148,
      "num_tokens": 94719777.0,
      "step": 46745
    },
    {
      "entropy": 5.8955153465271,
      "epoch": 4.085467097789041,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003422667333987326,
      "loss": 5.3601,
      "mean_token_accuracy": 0.20124827921390534,
      "num_tokens": 94729742.0,
      "step": 46750
    },
    {
      "entropy": 5.969067001342774,
      "epoch": 4.085904046141746,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034223699682030647,
      "loss": 5.4374,
      "mean_token_accuracy": 0.1924482360482216,
      "num_tokens": 94739685.0,
      "step": 46755
    },
    {
      "entropy": 5.770703125,
      "epoch": 4.086340994494451,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003422072589522395,
      "loss": 5.1778,
      "mean_token_accuracy": 0.22076952010393142,
      "num_tokens": 94749133.0,
      "step": 46760
    },
    {
      "entropy": 5.838165712356568,
      "epoch": 4.086777942847156,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003421775197951021,
      "loss": 5.3778,
      "mean_token_accuracy": 0.19820411503314972,
      "num_tokens": 94758918.0,
      "step": 46765
    },
    {
      "entropy": 5.884162712097168,
      "epoch": 4.08721489119986,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034214777934946474,
      "loss": 5.3387,
      "mean_token_accuracy": 0.200135301053524,
      "num_tokens": 94769879.0,
      "step": 46770
    },
    {
      "entropy": 5.925144863128662,
      "epoch": 4.0876518395525645,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003421180376158977,
      "loss": 5.3246,
      "mean_token_accuracy": 0.20330150723457335,
      "num_tokens": 94778668.0,
      "step": 46775
    },
    {
      "entropy": 5.903689575195313,
      "epoch": 4.08808878790527,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003420882945949715,
      "loss": 5.4062,
      "mean_token_accuracy": 0.1948095440864563,
      "num_tokens": 94789388.0,
      "step": 46780
    },
    {
      "entropy": 5.811237478256226,
      "epoch": 4.088525736257974,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003420585502872568,
      "loss": 5.2861,
      "mean_token_accuracy": 0.21395501643419265,
      "num_tokens": 94799467.0,
      "step": 46785
    },
    {
      "entropy": 5.8358618259429935,
      "epoch": 4.088962684610679,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003420288046933239,
      "loss": 5.3139,
      "mean_token_accuracy": 0.20591604858636856,
      "num_tokens": 94809050.0,
      "step": 46790
    },
    {
      "entropy": 5.861971426010132,
      "epoch": 4.089399632963384,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034199905781374327,
      "loss": 5.2882,
      "mean_token_accuracy": 0.20573018491268158,
      "num_tokens": 94818069.0,
      "step": 46795
    },
    {
      "entropy": 5.867584133148194,
      "epoch": 4.089836581316089,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034196930964908574,
      "loss": 5.3289,
      "mean_token_accuracy": 0.19527383595705033,
      "num_tokens": 94828288.0,
      "step": 46800
    },
    {
      "entropy": 5.893059349060058,
      "epoch": 4.090273529668793,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034193956019992175,
      "loss": 5.3476,
      "mean_token_accuracy": 0.20024259388446808,
      "num_tokens": 94839291.0,
      "step": 46805
    },
    {
      "entropy": 5.903534030914306,
      "epoch": 4.090710478021498,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003419098094668219,
      "loss": 5.2944,
      "mean_token_accuracy": 0.20794882476329804,
      "num_tokens": 94848582.0,
      "step": 46810
    },
    {
      "entropy": 5.890750694274902,
      "epoch": 4.091147426374203,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003418800574503569,
      "loss": 5.3542,
      "mean_token_accuracy": 0.2034612253308296,
      "num_tokens": 94859484.0,
      "step": 46815
    },
    {
      "entropy": 5.84281735420227,
      "epoch": 4.091584374726907,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00034185030415109725,
      "loss": 5.2882,
      "mean_token_accuracy": 0.20523710399866105,
      "num_tokens": 94869581.0,
      "step": 46820
    },
    {
      "entropy": 5.883015251159668,
      "epoch": 4.092021323079612,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003418205495696138,
      "loss": 5.3103,
      "mean_token_accuracy": 0.20397306829690934,
      "num_tokens": 94880151.0,
      "step": 46825
    },
    {
      "entropy": 5.771477842330933,
      "epoch": 4.092458271432316,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00034179079370647713,
      "loss": 5.2013,
      "mean_token_accuracy": 0.21790216863155365,
      "num_tokens": 94889302.0,
      "step": 46830
    },
    {
      "entropy": 5.901140642166138,
      "epoch": 4.092895219785022,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034176103656225815,
      "loss": 5.3744,
      "mean_token_accuracy": 0.19705670773983003,
      "num_tokens": 94900122.0,
      "step": 46835
    },
    {
      "entropy": 5.889307451248169,
      "epoch": 4.093332168137726,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003417312781375274,
      "loss": 5.3297,
      "mean_token_accuracy": 0.203213694691658,
      "num_tokens": 94909815.0,
      "step": 46840
    },
    {
      "entropy": 5.868927001953125,
      "epoch": 4.093769116490431,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034170151843285586,
      "loss": 5.3221,
      "mean_token_accuracy": 0.20666043162345887,
      "num_tokens": 94919103.0,
      "step": 46845
    },
    {
      "entropy": 5.866910696029663,
      "epoch": 4.094206064843135,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034167175744881416,
      "loss": 5.2752,
      "mean_token_accuracy": 0.20427079051733016,
      "num_tokens": 94928611.0,
      "step": 46850
    },
    {
      "entropy": 5.879385375976563,
      "epoch": 4.0946430131958405,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003416419951859733,
      "loss": 5.3295,
      "mean_token_accuracy": 0.20072062760591508,
      "num_tokens": 94938178.0,
      "step": 46855
    },
    {
      "entropy": 5.904694652557373,
      "epoch": 4.095079961548545,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034161223164490403,
      "loss": 5.354,
      "mean_token_accuracy": 0.19516662508249283,
      "num_tokens": 94947981.0,
      "step": 46860
    },
    {
      "entropy": 5.8362181186676025,
      "epoch": 4.095516909901249,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034158246682617716,
      "loss": 5.3534,
      "mean_token_accuracy": 0.19702255725860596,
      "num_tokens": 94958171.0,
      "step": 46865
    },
    {
      "entropy": 5.864217329025268,
      "epoch": 4.095953858253955,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003415527007303638,
      "loss": 5.2718,
      "mean_token_accuracy": 0.20484157055616378,
      "num_tokens": 94968262.0,
      "step": 46870
    },
    {
      "entropy": 5.85967378616333,
      "epoch": 4.096390806606659,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034152293335803477,
      "loss": 5.3118,
      "mean_token_accuracy": 0.205554236471653,
      "num_tokens": 94977871.0,
      "step": 46875
    },
    {
      "entropy": 5.884320592880249,
      "epoch": 4.096827754959364,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003414931647097609,
      "loss": 5.3803,
      "mean_token_accuracy": 0.202503103017807,
      "num_tokens": 94988072.0,
      "step": 46880
    },
    {
      "entropy": 5.878733825683594,
      "epoch": 4.097264703312068,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034146339478611343,
      "loss": 5.2952,
      "mean_token_accuracy": 0.20480831116437911,
      "num_tokens": 94998411.0,
      "step": 46885
    },
    {
      "entropy": 5.941124057769775,
      "epoch": 4.0977016516647735,
      "grad_norm": 1.125,
      "learning_rate": 0.0003414336235876631,
      "loss": 5.3788,
      "mean_token_accuracy": 0.19599110931158065,
      "num_tokens": 95008338.0,
      "step": 46890
    },
    {
      "entropy": 5.841588973999023,
      "epoch": 4.098138600017478,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034140385111498117,
      "loss": 5.3104,
      "mean_token_accuracy": 0.19103525429964066,
      "num_tokens": 95019167.0,
      "step": 46895
    },
    {
      "entropy": 5.9116740226745605,
      "epoch": 4.0985755483701825,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034137407736863854,
      "loss": 5.3346,
      "mean_token_accuracy": 0.19935947507619858,
      "num_tokens": 95029219.0,
      "step": 46900
    },
    {
      "entropy": 5.8516401767730715,
      "epoch": 4.099012496722887,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003413443023492063,
      "loss": 5.3081,
      "mean_token_accuracy": 0.20662545561790466,
      "num_tokens": 95038430.0,
      "step": 46905
    },
    {
      "entropy": 5.8591073513031,
      "epoch": 4.099449445075592,
      "grad_norm": 1.0,
      "learning_rate": 0.0003413145260572556,
      "loss": 5.3163,
      "mean_token_accuracy": 0.20300370901823045,
      "num_tokens": 95048444.0,
      "step": 46910
    },
    {
      "entropy": 5.929624557495117,
      "epoch": 4.099886393428297,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034128474849335754,
      "loss": 5.4281,
      "mean_token_accuracy": 0.20337164998054505,
      "num_tokens": 95059427.0,
      "step": 46915
    },
    {
      "entropy": 5.858738565444947,
      "epoch": 4.100323341781001,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034125496965808325,
      "loss": 5.2888,
      "mean_token_accuracy": 0.20251798033714294,
      "num_tokens": 95068306.0,
      "step": 46920
    },
    {
      "entropy": 5.920610618591309,
      "epoch": 4.100760290133707,
      "grad_norm": 1.28125,
      "learning_rate": 0.000341225189552004,
      "loss": 5.4141,
      "mean_token_accuracy": 0.19806679338216782,
      "num_tokens": 95079314.0,
      "step": 46925
    },
    {
      "entropy": 5.907250547409058,
      "epoch": 4.101197238486411,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003411954081756908,
      "loss": 5.338,
      "mean_token_accuracy": 0.19691471755504608,
      "num_tokens": 95089985.0,
      "step": 46930
    },
    {
      "entropy": 5.88381814956665,
      "epoch": 4.1016341868391155,
      "grad_norm": 1.125,
      "learning_rate": 0.00034116562552971504,
      "loss": 5.3454,
      "mean_token_accuracy": 0.19796474874019623,
      "num_tokens": 95099694.0,
      "step": 46935
    },
    {
      "entropy": 5.874949884414673,
      "epoch": 4.10207113519182,
      "grad_norm": 1.0,
      "learning_rate": 0.00034113584161464793,
      "loss": 5.3464,
      "mean_token_accuracy": 0.20518598407506944,
      "num_tokens": 95110585.0,
      "step": 46940
    },
    {
      "entropy": 5.801007175445557,
      "epoch": 4.102508083544525,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003411060564310607,
      "loss": 5.2921,
      "mean_token_accuracy": 0.20809395015239715,
      "num_tokens": 95120003.0,
      "step": 46945
    },
    {
      "entropy": 5.8596831321716305,
      "epoch": 4.10294503189723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034107626997952474,
      "loss": 5.2804,
      "mean_token_accuracy": 0.20262310951948165,
      "num_tokens": 95130697.0,
      "step": 46950
    },
    {
      "entropy": 5.980203437805176,
      "epoch": 4.103381980249934,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003410464822606112,
      "loss": 5.454,
      "mean_token_accuracy": 0.19381851851940154,
      "num_tokens": 95141044.0,
      "step": 46955
    },
    {
      "entropy": 5.9332774639129635,
      "epoch": 4.103818928602639,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003410166932748915,
      "loss": 5.3744,
      "mean_token_accuracy": 0.188279826939106,
      "num_tokens": 95151046.0,
      "step": 46960
    },
    {
      "entropy": 5.873033714294434,
      "epoch": 4.104255876955344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003409869030229371,
      "loss": 5.3348,
      "mean_token_accuracy": 0.20399755090475083,
      "num_tokens": 95160743.0,
      "step": 46965
    },
    {
      "entropy": 5.815719938278198,
      "epoch": 4.104692825308049,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003409571115053193,
      "loss": 5.2637,
      "mean_token_accuracy": 0.2012331083416939,
      "num_tokens": 95170437.0,
      "step": 46970
    },
    {
      "entropy": 5.863739919662476,
      "epoch": 4.105129773660753,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034092731872260965,
      "loss": 5.3572,
      "mean_token_accuracy": 0.20006777197122574,
      "num_tokens": 95179734.0,
      "step": 46975
    },
    {
      "entropy": 5.922918081283569,
      "epoch": 4.105566722013458,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003408975246753793,
      "loss": 5.3743,
      "mean_token_accuracy": 0.20012272894382477,
      "num_tokens": 95190401.0,
      "step": 46980
    },
    {
      "entropy": 6.0009290218353275,
      "epoch": 4.106003670366163,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003408677293642,
      "loss": 5.4333,
      "mean_token_accuracy": 0.187059485912323,
      "num_tokens": 95200629.0,
      "step": 46985
    },
    {
      "entropy": 5.88988242149353,
      "epoch": 4.106440618718867,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003408379327896431,
      "loss": 5.3415,
      "mean_token_accuracy": 0.20062706470489503,
      "num_tokens": 95210858.0,
      "step": 46990
    },
    {
      "entropy": 5.883242845535278,
      "epoch": 4.106877567071572,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034080813495228026,
      "loss": 5.3841,
      "mean_token_accuracy": 0.19256871491670607,
      "num_tokens": 95221531.0,
      "step": 46995
    },
    {
      "entropy": 5.867733716964722,
      "epoch": 4.107314515424277,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00034077833585268276,
      "loss": 5.2891,
      "mean_token_accuracy": 0.2036368355154991,
      "num_tokens": 95231689.0,
      "step": 47000
    },
    {
      "entropy": 5.9319250106811525,
      "epoch": 4.107751463776982,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034074853549142235,
      "loss": 5.3574,
      "mean_token_accuracy": 0.19827849566936492,
      "num_tokens": 95242462.0,
      "step": 47005
    },
    {
      "entropy": 5.835190010070801,
      "epoch": 4.108188412129686,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003407187338690706,
      "loss": 5.251,
      "mean_token_accuracy": 0.20338184237480164,
      "num_tokens": 95252063.0,
      "step": 47010
    },
    {
      "entropy": 5.848442125320434,
      "epoch": 4.108625360482391,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034068893098619906,
      "loss": 5.3338,
      "mean_token_accuracy": 0.1977618530392647,
      "num_tokens": 95262047.0,
      "step": 47015
    },
    {
      "entropy": 5.75152816772461,
      "epoch": 4.109062308835096,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034065912684337954,
      "loss": 5.1974,
      "mean_token_accuracy": 0.21271516531705856,
      "num_tokens": 95272117.0,
      "step": 47020
    },
    {
      "entropy": 5.971253252029419,
      "epoch": 4.1094992571878,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034062932144118336,
      "loss": 5.4365,
      "mean_token_accuracy": 0.19263856261968612,
      "num_tokens": 95281477.0,
      "step": 47025
    },
    {
      "entropy": 5.933200979232788,
      "epoch": 4.109936205540505,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003405995147801825,
      "loss": 5.3323,
      "mean_token_accuracy": 0.20112330466508865,
      "num_tokens": 95291140.0,
      "step": 47030
    },
    {
      "entropy": 5.8839984893798825,
      "epoch": 4.11037315389321,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034056970686094856,
      "loss": 5.3292,
      "mean_token_accuracy": 0.20666404962539672,
      "num_tokens": 95301256.0,
      "step": 47035
    },
    {
      "entropy": 5.912568092346191,
      "epoch": 4.110810102245915,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034053989768405334,
      "loss": 5.3984,
      "mean_token_accuracy": 0.1896122857928276,
      "num_tokens": 95311232.0,
      "step": 47040
    },
    {
      "entropy": 5.881692790985108,
      "epoch": 4.111247050598619,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003405100872500685,
      "loss": 5.2836,
      "mean_token_accuracy": 0.20524548441171647,
      "num_tokens": 95321578.0,
      "step": 47045
    },
    {
      "entropy": 5.875609588623047,
      "epoch": 4.111683998951324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003404802755595658,
      "loss": 5.3449,
      "mean_token_accuracy": 0.19478630721569062,
      "num_tokens": 95330837.0,
      "step": 47050
    },
    {
      "entropy": 5.864798164367675,
      "epoch": 4.112120947304029,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003404504626131172,
      "loss": 5.2761,
      "mean_token_accuracy": 0.20500503927469255,
      "num_tokens": 95341778.0,
      "step": 47055
    },
    {
      "entropy": 5.909754896163941,
      "epoch": 4.1125578956567335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003404206484112944,
      "loss": 5.3562,
      "mean_token_accuracy": 0.1984216868877411,
      "num_tokens": 95353629.0,
      "step": 47060
    },
    {
      "entropy": 5.877660751342773,
      "epoch": 4.112994844009438,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034039083295466925,
      "loss": 5.4149,
      "mean_token_accuracy": 0.19461863189935685,
      "num_tokens": 95364778.0,
      "step": 47065
    },
    {
      "entropy": 5.880101728439331,
      "epoch": 4.113431792362142,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003403610162438137,
      "loss": 5.3258,
      "mean_token_accuracy": 0.19713475108146666,
      "num_tokens": 95375354.0,
      "step": 47070
    },
    {
      "entropy": 5.887013053894043,
      "epoch": 4.113868740714848,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034033119827929964,
      "loss": 5.3377,
      "mean_token_accuracy": 0.20283833295106887,
      "num_tokens": 95385864.0,
      "step": 47075
    },
    {
      "entropy": 5.857278537750244,
      "epoch": 4.114305689067552,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003403013790616989,
      "loss": 5.3148,
      "mean_token_accuracy": 0.20587510466575623,
      "num_tokens": 95397443.0,
      "step": 47080
    },
    {
      "entropy": 5.878548336029053,
      "epoch": 4.114742637420257,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003402715585915835,
      "loss": 5.3712,
      "mean_token_accuracy": 0.19959881156682968,
      "num_tokens": 95407322.0,
      "step": 47085
    },
    {
      "entropy": 5.895947551727295,
      "epoch": 4.115179585772962,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003402417368695255,
      "loss": 5.3304,
      "mean_token_accuracy": 0.2032568484544754,
      "num_tokens": 95417088.0,
      "step": 47090
    },
    {
      "entropy": 5.844367933273316,
      "epoch": 4.1156165341256665,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003402119138960967,
      "loss": 5.2949,
      "mean_token_accuracy": 0.21037444323301316,
      "num_tokens": 95426610.0,
      "step": 47095
    },
    {
      "entropy": 5.887573289871216,
      "epoch": 4.116053482478371,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003401820896718692,
      "loss": 5.3677,
      "mean_token_accuracy": 0.20192893743515014,
      "num_tokens": 95436780.0,
      "step": 47100
    },
    {
      "entropy": 5.903043651580811,
      "epoch": 4.1164904308310755,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003401522641974151,
      "loss": 5.3197,
      "mean_token_accuracy": 0.20247368663549423,
      "num_tokens": 95446143.0,
      "step": 47105
    },
    {
      "entropy": 5.876263904571533,
      "epoch": 4.116927379183781,
      "grad_norm": 1.0,
      "learning_rate": 0.00034012243747330643,
      "loss": 5.297,
      "mean_token_accuracy": 0.20424282252788545,
      "num_tokens": 95456609.0,
      "step": 47110
    },
    {
      "entropy": 5.807846736907959,
      "epoch": 4.117364327536485,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003400926095001153,
      "loss": 5.3136,
      "mean_token_accuracy": 0.19955260455608367,
      "num_tokens": 95466788.0,
      "step": 47115
    },
    {
      "entropy": 5.853653860092163,
      "epoch": 4.11780127588919,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034006278027841385,
      "loss": 5.3165,
      "mean_token_accuracy": 0.2048737734556198,
      "num_tokens": 95476765.0,
      "step": 47120
    },
    {
      "entropy": 5.91916823387146,
      "epoch": 4.118238224241894,
      "grad_norm": 1.0,
      "learning_rate": 0.00034003294980877415,
      "loss": 5.3573,
      "mean_token_accuracy": 0.20222630202770234,
      "num_tokens": 95486604.0,
      "step": 47125
    },
    {
      "entropy": 5.841415929794311,
      "epoch": 4.1186751725946,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034000311809176837,
      "loss": 5.2418,
      "mean_token_accuracy": 0.21322133988142014,
      "num_tokens": 95496258.0,
      "step": 47130
    },
    {
      "entropy": 5.877437257766724,
      "epoch": 4.119112120947304,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033997328512796863,
      "loss": 5.3184,
      "mean_token_accuracy": 0.19965238720178605,
      "num_tokens": 95506493.0,
      "step": 47135
    },
    {
      "entropy": 5.838683938980102,
      "epoch": 4.1195490693000085,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003399434509179475,
      "loss": 5.3123,
      "mean_token_accuracy": 0.19967586994171144,
      "num_tokens": 95516374.0,
      "step": 47140
    },
    {
      "entropy": 5.860969686508179,
      "epoch": 4.119986017652714,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00033991361546227675,
      "loss": 5.3674,
      "mean_token_accuracy": 0.2020518362522125,
      "num_tokens": 95525958.0,
      "step": 47145
    },
    {
      "entropy": 5.8670971393585205,
      "epoch": 4.120422966005418,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033988377876152887,
      "loss": 5.2586,
      "mean_token_accuracy": 0.21389333605766297,
      "num_tokens": 95535703.0,
      "step": 47150
    },
    {
      "entropy": 5.842655849456787,
      "epoch": 4.120859914358123,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033985394081627605,
      "loss": 5.2869,
      "mean_token_accuracy": 0.2022027224302292,
      "num_tokens": 95545821.0,
      "step": 47155
    },
    {
      "entropy": 5.879371643066406,
      "epoch": 4.121296862710827,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003398241016270909,
      "loss": 5.2844,
      "mean_token_accuracy": 0.20309916287660598,
      "num_tokens": 95556095.0,
      "step": 47160
    },
    {
      "entropy": 5.9966073513031,
      "epoch": 4.121733811063533,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003397942611945453,
      "loss": 5.4152,
      "mean_token_accuracy": 0.19127755612134933,
      "num_tokens": 95566695.0,
      "step": 47165
    },
    {
      "entropy": 5.86349048614502,
      "epoch": 4.122170759416237,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003397644195192119,
      "loss": 5.3344,
      "mean_token_accuracy": 0.20212558805942535,
      "num_tokens": 95576525.0,
      "step": 47170
    },
    {
      "entropy": 5.850326061248779,
      "epoch": 4.122607707768942,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033973457660166297,
      "loss": 5.296,
      "mean_token_accuracy": 0.1965040385723114,
      "num_tokens": 95586121.0,
      "step": 47175
    },
    {
      "entropy": 5.877172374725342,
      "epoch": 4.123044656121646,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003397047324424709,
      "loss": 5.3394,
      "mean_token_accuracy": 0.20124042332172393,
      "num_tokens": 95595694.0,
      "step": 47180
    },
    {
      "entropy": 5.862946224212647,
      "epoch": 4.123481604474351,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033967488704220823,
      "loss": 5.2814,
      "mean_token_accuracy": 0.2010890379548073,
      "num_tokens": 95607114.0,
      "step": 47185
    },
    {
      "entropy": 5.911751794815063,
      "epoch": 4.123918552827056,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033964504040144734,
      "loss": 5.3205,
      "mean_token_accuracy": 0.20417612493038179,
      "num_tokens": 95617476.0,
      "step": 47190
    },
    {
      "entropy": 5.9157829761505125,
      "epoch": 4.12435550117976,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003396151925207606,
      "loss": 5.3835,
      "mean_token_accuracy": 0.20311143696308137,
      "num_tokens": 95627205.0,
      "step": 47195
    },
    {
      "entropy": 5.8823463916778564,
      "epoch": 4.124792449532466,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003395853434007207,
      "loss": 5.3343,
      "mean_token_accuracy": 0.20480790883302688,
      "num_tokens": 95637798.0,
      "step": 47200
    },
    {
      "entropy": 5.840808916091919,
      "epoch": 4.12522939788517,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033955549304190007,
      "loss": 5.2607,
      "mean_token_accuracy": 0.20677317529916764,
      "num_tokens": 95647615.0,
      "step": 47205
    },
    {
      "entropy": 5.896752500534058,
      "epoch": 4.125666346237875,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003395256414448712,
      "loss": 5.2943,
      "mean_token_accuracy": 0.20176058262586594,
      "num_tokens": 95657019.0,
      "step": 47210
    },
    {
      "entropy": 5.871463298797607,
      "epoch": 4.126103294590579,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033949578861020675,
      "loss": 5.287,
      "mean_token_accuracy": 0.20661165416240693,
      "num_tokens": 95667378.0,
      "step": 47215
    },
    {
      "entropy": 5.8376024723052975,
      "epoch": 4.1265402429432845,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033946593453847927,
      "loss": 5.2822,
      "mean_token_accuracy": 0.20320467799901962,
      "num_tokens": 95678078.0,
      "step": 47220
    },
    {
      "entropy": 5.850117254257202,
      "epoch": 4.126977191295989,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003394360792302614,
      "loss": 5.3587,
      "mean_token_accuracy": 0.20312895029783248,
      "num_tokens": 95687450.0,
      "step": 47225
    },
    {
      "entropy": 5.904253387451172,
      "epoch": 4.127414139648693,
      "grad_norm": 0.984375,
      "learning_rate": 0.00033940622268612583,
      "loss": 5.3823,
      "mean_token_accuracy": 0.196742045879364,
      "num_tokens": 95698594.0,
      "step": 47230
    },
    {
      "entropy": 5.894360876083374,
      "epoch": 4.127851088001398,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033937636490664506,
      "loss": 5.3194,
      "mean_token_accuracy": 0.1993970289826393,
      "num_tokens": 95708708.0,
      "step": 47235
    },
    {
      "entropy": 5.881155204772949,
      "epoch": 4.128288036354103,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003393465058923919,
      "loss": 5.3241,
      "mean_token_accuracy": 0.2025511831045151,
      "num_tokens": 95718367.0,
      "step": 47240
    },
    {
      "entropy": 5.893716430664062,
      "epoch": 4.128724984706808,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033931664564393907,
      "loss": 5.3784,
      "mean_token_accuracy": 0.1895007833838463,
      "num_tokens": 95728371.0,
      "step": 47245
    },
    {
      "entropy": 5.906233406066894,
      "epoch": 4.129161933059512,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00033928678416185926,
      "loss": 5.3404,
      "mean_token_accuracy": 0.19858996719121932,
      "num_tokens": 95738170.0,
      "step": 47250
    },
    {
      "entropy": 5.766774606704712,
      "epoch": 4.129598881412217,
      "grad_norm": 0.984375,
      "learning_rate": 0.00033925692144672526,
      "loss": 5.2061,
      "mean_token_accuracy": 0.21778236329555511,
      "num_tokens": 95748088.0,
      "step": 47255
    },
    {
      "entropy": 5.87698221206665,
      "epoch": 4.130035829764922,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033922705749910975,
      "loss": 5.298,
      "mean_token_accuracy": 0.20208362489938736,
      "num_tokens": 95758033.0,
      "step": 47260
    },
    {
      "entropy": 5.917874670028686,
      "epoch": 4.1304727781176265,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003391971923195858,
      "loss": 5.4466,
      "mean_token_accuracy": 0.19108215272426604,
      "num_tokens": 95768261.0,
      "step": 47265
    },
    {
      "entropy": 5.886673450469971,
      "epoch": 4.130909726470331,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003391673259087259,
      "loss": 5.2723,
      "mean_token_accuracy": 0.20505752712488173,
      "num_tokens": 95777739.0,
      "step": 47270
    },
    {
      "entropy": 5.887985515594482,
      "epoch": 4.131346674823036,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003391374582671031,
      "loss": 5.3304,
      "mean_token_accuracy": 0.20253219902515412,
      "num_tokens": 95787683.0,
      "step": 47275
    },
    {
      "entropy": 5.940604019165039,
      "epoch": 4.131783623175741,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003391075893952903,
      "loss": 5.3945,
      "mean_token_accuracy": 0.19714035242795944,
      "num_tokens": 95798411.0,
      "step": 47280
    },
    {
      "entropy": 5.85595703125,
      "epoch": 4.132220571528445,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003390777192938603,
      "loss": 5.2934,
      "mean_token_accuracy": 0.20917783677577972,
      "num_tokens": 95808869.0,
      "step": 47285
    },
    {
      "entropy": 5.908448362350464,
      "epoch": 4.13265751988115,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00033904784796338607,
      "loss": 5.3152,
      "mean_token_accuracy": 0.20005894005298613,
      "num_tokens": 95818808.0,
      "step": 47290
    },
    {
      "entropy": 5.893925189971924,
      "epoch": 4.133094468233855,
      "grad_norm": 1.0,
      "learning_rate": 0.0003390179754044406,
      "loss": 5.3365,
      "mean_token_accuracy": 0.20133532732725143,
      "num_tokens": 95828314.0,
      "step": 47295
    },
    {
      "entropy": 5.857647657394409,
      "epoch": 4.1335314165865595,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003389881016175968,
      "loss": 5.3353,
      "mean_token_accuracy": 0.20339627861976622,
      "num_tokens": 95837727.0,
      "step": 47300
    },
    {
      "entropy": 5.901446866989136,
      "epoch": 4.133968364939264,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033895822660342763,
      "loss": 5.3397,
      "mean_token_accuracy": 0.20083816945552826,
      "num_tokens": 95847871.0,
      "step": 47305
    },
    {
      "entropy": 5.849871492385864,
      "epoch": 4.1344053132919685,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033892835036250616,
      "loss": 5.3461,
      "mean_token_accuracy": 0.19984279721975326,
      "num_tokens": 95858147.0,
      "step": 47310
    },
    {
      "entropy": 5.857506847381591,
      "epoch": 4.134842261644674,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033889847289540547,
      "loss": 5.2753,
      "mean_token_accuracy": 0.20022546350955964,
      "num_tokens": 95867588.0,
      "step": 47315
    },
    {
      "entropy": 5.8900096893310545,
      "epoch": 4.135279209997378,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033886859420269864,
      "loss": 5.3865,
      "mean_token_accuracy": 0.19426560252904893,
      "num_tokens": 95878193.0,
      "step": 47320
    },
    {
      "entropy": 5.889259672164917,
      "epoch": 4.135716158350083,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033883871428495865,
      "loss": 5.3948,
      "mean_token_accuracy": 0.19463175386190415,
      "num_tokens": 95888945.0,
      "step": 47325
    },
    {
      "entropy": 5.918506145477295,
      "epoch": 4.136153106702788,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003388088331427586,
      "loss": 5.3335,
      "mean_token_accuracy": 0.20567142218351364,
      "num_tokens": 95898476.0,
      "step": 47330
    },
    {
      "entropy": 5.858529281616211,
      "epoch": 4.136590055055493,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003387789507766718,
      "loss": 5.295,
      "mean_token_accuracy": 0.20930738747119904,
      "num_tokens": 95907781.0,
      "step": 47335
    },
    {
      "entropy": 5.9076252460479735,
      "epoch": 4.137027003408197,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003387490671872712,
      "loss": 5.3786,
      "mean_token_accuracy": 0.1912257418036461,
      "num_tokens": 95917558.0,
      "step": 47340
    },
    {
      "entropy": 5.907480525970459,
      "epoch": 4.1374639517609015,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033871918237513016,
      "loss": 5.3521,
      "mean_token_accuracy": 0.19590117782354355,
      "num_tokens": 95927862.0,
      "step": 47345
    },
    {
      "entropy": 5.829501914978027,
      "epoch": 4.137900900113607,
      "grad_norm": 1.453125,
      "learning_rate": 0.00033868929634082187,
      "loss": 5.303,
      "mean_token_accuracy": 0.2092109352350235,
      "num_tokens": 95937848.0,
      "step": 47350
    },
    {
      "entropy": 5.865637922286988,
      "epoch": 4.138337848466311,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003386594090849194,
      "loss": 5.377,
      "mean_token_accuracy": 0.20340058356523513,
      "num_tokens": 95948013.0,
      "step": 47355
    },
    {
      "entropy": 5.852880811691284,
      "epoch": 4.138774796819016,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003386295206079961,
      "loss": 5.2559,
      "mean_token_accuracy": 0.2077300876379013,
      "num_tokens": 95957483.0,
      "step": 47360
    },
    {
      "entropy": 5.835203075408936,
      "epoch": 4.13921174517172,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033859963091062537,
      "loss": 5.2671,
      "mean_token_accuracy": 0.20875943452119827,
      "num_tokens": 95966940.0,
      "step": 47365
    },
    {
      "entropy": 5.877726888656616,
      "epoch": 4.139648693524426,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033856973999338026,
      "loss": 5.2881,
      "mean_token_accuracy": 0.19689729511737825,
      "num_tokens": 95977634.0,
      "step": 47370
    },
    {
      "entropy": 5.898179721832276,
      "epoch": 4.14008564187713,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033853984785683433,
      "loss": 5.3683,
      "mean_token_accuracy": 0.20153592377901078,
      "num_tokens": 95987114.0,
      "step": 47375
    },
    {
      "entropy": 5.890613317489624,
      "epoch": 4.140522590229835,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003385099545015608,
      "loss": 5.3344,
      "mean_token_accuracy": 0.20555747747421266,
      "num_tokens": 95997051.0,
      "step": 47380
    },
    {
      "entropy": 5.918476581573486,
      "epoch": 4.14095953858254,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033848005992813304,
      "loss": 5.3361,
      "mean_token_accuracy": 0.19861093163490295,
      "num_tokens": 96007160.0,
      "step": 47385
    },
    {
      "entropy": 5.839268970489502,
      "epoch": 4.141396486935244,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003384501641371244,
      "loss": 5.2652,
      "mean_token_accuracy": 0.20745977908372878,
      "num_tokens": 96017448.0,
      "step": 47390
    },
    {
      "entropy": 5.830434417724609,
      "epoch": 4.141833435287949,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003384202671291085,
      "loss": 5.2748,
      "mean_token_accuracy": 0.2078140065073967,
      "num_tokens": 96027061.0,
      "step": 47395
    },
    {
      "entropy": 5.840597343444824,
      "epoch": 4.142270383640653,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033839036890465855,
      "loss": 5.3201,
      "mean_token_accuracy": 0.20458337366580964,
      "num_tokens": 96036843.0,
      "step": 47400
    },
    {
      "entropy": 5.8440831184387205,
      "epoch": 4.142707331993359,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033836046946434815,
      "loss": 5.2988,
      "mean_token_accuracy": 0.20377635806798935,
      "num_tokens": 96047370.0,
      "step": 47405
    },
    {
      "entropy": 5.864636421203613,
      "epoch": 4.143144280346063,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003383305688087507,
      "loss": 5.2999,
      "mean_token_accuracy": 0.20056938678026198,
      "num_tokens": 96056759.0,
      "step": 47410
    },
    {
      "entropy": 5.847242832183838,
      "epoch": 4.143581228698768,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003383006669384398,
      "loss": 5.2823,
      "mean_token_accuracy": 0.2062377229332924,
      "num_tokens": 96067327.0,
      "step": 47415
    },
    {
      "entropy": 5.819560718536377,
      "epoch": 4.144018177051472,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033827076385398896,
      "loss": 5.2405,
      "mean_token_accuracy": 0.20920485258102417,
      "num_tokens": 96076818.0,
      "step": 47420
    },
    {
      "entropy": 5.950828981399536,
      "epoch": 4.1444551254041775,
      "grad_norm": 1.0,
      "learning_rate": 0.0003382408595559717,
      "loss": 5.4011,
      "mean_token_accuracy": 0.19113951921463013,
      "num_tokens": 96087490.0,
      "step": 47425
    },
    {
      "entropy": 5.944339227676392,
      "epoch": 4.144892073756882,
      "grad_norm": 1.0,
      "learning_rate": 0.0003382109540449616,
      "loss": 5.4111,
      "mean_token_accuracy": 0.19448081254959107,
      "num_tokens": 96097042.0,
      "step": 47430
    },
    {
      "entropy": 5.8994996547698975,
      "epoch": 4.145329022109586,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003381810473215323,
      "loss": 5.3601,
      "mean_token_accuracy": 0.19544356763362886,
      "num_tokens": 96107809.0,
      "step": 47435
    },
    {
      "entropy": 5.900193929672241,
      "epoch": 4.145765970462292,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003381511393862574,
      "loss": 5.3814,
      "mean_token_accuracy": 0.20445215851068496,
      "num_tokens": 96118731.0,
      "step": 47440
    },
    {
      "entropy": 5.838494873046875,
      "epoch": 4.146202918814996,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003381212302397105,
      "loss": 5.3273,
      "mean_token_accuracy": 0.20475432723760606,
      "num_tokens": 96128832.0,
      "step": 47445
    },
    {
      "entropy": 5.903419828414917,
      "epoch": 4.146639867167701,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033809131988246546,
      "loss": 5.3337,
      "mean_token_accuracy": 0.20365653336048126,
      "num_tokens": 96139537.0,
      "step": 47450
    },
    {
      "entropy": 5.902101802825928,
      "epoch": 4.147076815520405,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033806140831509567,
      "loss": 5.3453,
      "mean_token_accuracy": 0.20372333079576493,
      "num_tokens": 96150022.0,
      "step": 47455
    },
    {
      "entropy": 5.834050178527832,
      "epoch": 4.1475137638731105,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033803149553817514,
      "loss": 5.2333,
      "mean_token_accuracy": 0.2092055633664131,
      "num_tokens": 96160485.0,
      "step": 47460
    },
    {
      "entropy": 5.800683164596558,
      "epoch": 4.147950712225815,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0003380015815522775,
      "loss": 5.2377,
      "mean_token_accuracy": 0.2094476878643036,
      "num_tokens": 96171068.0,
      "step": 47465
    },
    {
      "entropy": 5.891834354400634,
      "epoch": 4.1483876605785195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003379716663579765,
      "loss": 5.3554,
      "mean_token_accuracy": 0.20165085047483444,
      "num_tokens": 96181448.0,
      "step": 47470
    },
    {
      "entropy": 5.88234314918518,
      "epoch": 4.148824608931224,
      "grad_norm": 1.25,
      "learning_rate": 0.00033794174995584593,
      "loss": 5.3601,
      "mean_token_accuracy": 0.20160307139158248,
      "num_tokens": 96191463.0,
      "step": 47475
    },
    {
      "entropy": 5.872269964218139,
      "epoch": 4.149261557283929,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033791183234645954,
      "loss": 5.2958,
      "mean_token_accuracy": 0.20195502638816834,
      "num_tokens": 96201559.0,
      "step": 47480
    },
    {
      "entropy": 5.984166288375855,
      "epoch": 4.149698505636634,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00033788191353039136,
      "loss": 5.4211,
      "mean_token_accuracy": 0.19229927510023118,
      "num_tokens": 96213065.0,
      "step": 47485
    },
    {
      "entropy": 5.876550102233887,
      "epoch": 4.150135453989338,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003378519935082151,
      "loss": 5.3379,
      "mean_token_accuracy": 0.19716313481330872,
      "num_tokens": 96222952.0,
      "step": 47490
    },
    {
      "entropy": 5.912907218933105,
      "epoch": 4.150572402342043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033782207228050465,
      "loss": 5.3938,
      "mean_token_accuracy": 0.197342087328434,
      "num_tokens": 96232745.0,
      "step": 47495
    },
    {
      "entropy": 5.8667919635772705,
      "epoch": 4.151009350694748,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033779214984783396,
      "loss": 5.2649,
      "mean_token_accuracy": 0.20930628329515458,
      "num_tokens": 96243371.0,
      "step": 47500
    },
    {
      "entropy": 5.901024293899536,
      "epoch": 4.1514462990474525,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003377622262107769,
      "loss": 5.3635,
      "mean_token_accuracy": 0.19612971544265748,
      "num_tokens": 96254019.0,
      "step": 47505
    },
    {
      "entropy": 5.896693468093872,
      "epoch": 4.151883247400157,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003377323013699075,
      "loss": 5.304,
      "mean_token_accuracy": 0.20362394005060197,
      "num_tokens": 96264146.0,
      "step": 47510
    },
    {
      "entropy": 5.894803047180176,
      "epoch": 4.152320195752862,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003377023753257997,
      "loss": 5.3831,
      "mean_token_accuracy": 0.19856895506381989,
      "num_tokens": 96274831.0,
      "step": 47515
    },
    {
      "entropy": 5.855694055557251,
      "epoch": 4.152757144105567,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003376724480790274,
      "loss": 5.3233,
      "mean_token_accuracy": 0.20189658552408218,
      "num_tokens": 96285692.0,
      "step": 47520
    },
    {
      "entropy": 5.83526644706726,
      "epoch": 4.153194092458271,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003376425196301647,
      "loss": 5.2467,
      "mean_token_accuracy": 0.21078075468540192,
      "num_tokens": 96295896.0,
      "step": 47525
    },
    {
      "entropy": 5.88382363319397,
      "epoch": 4.153631040810976,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033761258997978567,
      "loss": 5.3062,
      "mean_token_accuracy": 0.20608583390712737,
      "num_tokens": 96306287.0,
      "step": 47530
    },
    {
      "entropy": 5.786502313613892,
      "epoch": 4.154067989163681,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00033758265912846444,
      "loss": 5.2564,
      "mean_token_accuracy": 0.2004923105239868,
      "num_tokens": 96316296.0,
      "step": 47535
    },
    {
      "entropy": 5.858787584304809,
      "epoch": 4.154504937516386,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000337552727076775,
      "loss": 5.3245,
      "mean_token_accuracy": 0.1978861451148987,
      "num_tokens": 96325713.0,
      "step": 47540
    },
    {
      "entropy": 5.906105184555054,
      "epoch": 4.15494188586909,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003375227938252914,
      "loss": 5.3668,
      "mean_token_accuracy": 0.20023330450057983,
      "num_tokens": 96335391.0,
      "step": 47545
    },
    {
      "entropy": 5.9276868343353275,
      "epoch": 4.1553788342217945,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003374928593745878,
      "loss": 5.4861,
      "mean_token_accuracy": 0.1852387383580208,
      "num_tokens": 96345919.0,
      "step": 47550
    },
    {
      "entropy": 5.855410814285278,
      "epoch": 4.1558157825745,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033746292372523857,
      "loss": 5.3463,
      "mean_token_accuracy": 0.20082978457212447,
      "num_tokens": 96356296.0,
      "step": 47555
    },
    {
      "entropy": 5.890903520584106,
      "epoch": 4.156252730927204,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033743298687781767,
      "loss": 5.3539,
      "mean_token_accuracy": 0.19817492961883545,
      "num_tokens": 96366780.0,
      "step": 47560
    },
    {
      "entropy": 5.945981693267822,
      "epoch": 4.156689679279909,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003374030488328993,
      "loss": 5.3236,
      "mean_token_accuracy": 0.20230563133955,
      "num_tokens": 96376665.0,
      "step": 47565
    },
    {
      "entropy": 5.863989973068238,
      "epoch": 4.157126627632614,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003373731095910578,
      "loss": 5.1899,
      "mean_token_accuracy": 0.21299503445625306,
      "num_tokens": 96386737.0,
      "step": 47570
    },
    {
      "entropy": 5.828361082077026,
      "epoch": 4.157563575985319,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033734316915286737,
      "loss": 5.3205,
      "mean_token_accuracy": 0.20713807940483092,
      "num_tokens": 96397022.0,
      "step": 47575
    },
    {
      "entropy": 5.791719388961792,
      "epoch": 4.158000524338023,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003373132275189023,
      "loss": 5.3009,
      "mean_token_accuracy": 0.20447458922863007,
      "num_tokens": 96406751.0,
      "step": 47580
    },
    {
      "entropy": 5.880269479751587,
      "epoch": 4.158437472690728,
      "grad_norm": 0.96875,
      "learning_rate": 0.00033728328468973697,
      "loss": 5.3231,
      "mean_token_accuracy": 0.20616746842861175,
      "num_tokens": 96416987.0,
      "step": 47585
    },
    {
      "entropy": 5.867292976379394,
      "epoch": 4.158874421043433,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003372533406659455,
      "loss": 5.2833,
      "mean_token_accuracy": 0.2064158499240875,
      "num_tokens": 96427180.0,
      "step": 47590
    },
    {
      "entropy": 5.868499898910523,
      "epoch": 4.159311369396137,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003372233954481023,
      "loss": 5.3485,
      "mean_token_accuracy": 0.1979213610291481,
      "num_tokens": 96437220.0,
      "step": 47595
    },
    {
      "entropy": 5.83452000617981,
      "epoch": 4.159748317748842,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033719344903678183,
      "loss": 5.2956,
      "mean_token_accuracy": 0.20352723002433776,
      "num_tokens": 96447569.0,
      "step": 47600
    },
    {
      "entropy": 5.833372259140015,
      "epoch": 4.160185266101546,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003371635014325585,
      "loss": 5.3015,
      "mean_token_accuracy": 0.20701609253883363,
      "num_tokens": 96457128.0,
      "step": 47605
    },
    {
      "entropy": 5.851713466644287,
      "epoch": 4.160622214454252,
      "grad_norm": 0.984375,
      "learning_rate": 0.00033713355263600655,
      "loss": 5.309,
      "mean_token_accuracy": 0.20172144174575807,
      "num_tokens": 96466678.0,
      "step": 47610
    },
    {
      "entropy": 5.867303991317749,
      "epoch": 4.161059162806956,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033710360264770046,
      "loss": 5.3868,
      "mean_token_accuracy": 0.1932101771235466,
      "num_tokens": 96477506.0,
      "step": 47615
    },
    {
      "entropy": 5.918394565582275,
      "epoch": 4.161496111159661,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003370736514682148,
      "loss": 5.3851,
      "mean_token_accuracy": 0.1919255793094635,
      "num_tokens": 96487811.0,
      "step": 47620
    },
    {
      "entropy": 5.942216205596924,
      "epoch": 4.161933059512366,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033704369909812393,
      "loss": 5.3187,
      "mean_token_accuracy": 0.1990232288837433,
      "num_tokens": 96497363.0,
      "step": 47625
    },
    {
      "entropy": 5.857340335845947,
      "epoch": 4.1623700078650705,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003370137455380024,
      "loss": 5.3038,
      "mean_token_accuracy": 0.201563261449337,
      "num_tokens": 96507390.0,
      "step": 47630
    },
    {
      "entropy": 5.996372985839844,
      "epoch": 4.162806956217775,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033698379078842477,
      "loss": 5.4989,
      "mean_token_accuracy": 0.18546444028615952,
      "num_tokens": 96518705.0,
      "step": 47635
    },
    {
      "entropy": 5.889231872558594,
      "epoch": 4.163243904570479,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033695383484996555,
      "loss": 5.3696,
      "mean_token_accuracy": 0.19694978147745132,
      "num_tokens": 96529296.0,
      "step": 47640
    },
    {
      "entropy": 5.882459831237793,
      "epoch": 4.163680852923185,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033692387772319934,
      "loss": 5.2653,
      "mean_token_accuracy": 0.20857447236776352,
      "num_tokens": 96538959.0,
      "step": 47645
    },
    {
      "entropy": 5.901858949661255,
      "epoch": 4.164117801275889,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003368939194087006,
      "loss": 5.375,
      "mean_token_accuracy": 0.20465677380561828,
      "num_tokens": 96548893.0,
      "step": 47650
    },
    {
      "entropy": 5.900760793685913,
      "epoch": 4.164554749628594,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003368639599070442,
      "loss": 5.289,
      "mean_token_accuracy": 0.20314228981733323,
      "num_tokens": 96557390.0,
      "step": 47655
    },
    {
      "entropy": 5.8724846839904785,
      "epoch": 4.164991697981298,
      "grad_norm": 1.0,
      "learning_rate": 0.0003368339992188045,
      "loss": 5.35,
      "mean_token_accuracy": 0.19563743323087693,
      "num_tokens": 96568010.0,
      "step": 47660
    },
    {
      "entropy": 5.880712509155273,
      "epoch": 4.165428646334004,
      "grad_norm": 1.0,
      "learning_rate": 0.0003368040373445564,
      "loss": 5.336,
      "mean_token_accuracy": 0.2017896816134453,
      "num_tokens": 96578636.0,
      "step": 47665
    },
    {
      "entropy": 5.916008567810058,
      "epoch": 4.165865594686708,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033677407428487445,
      "loss": 5.3962,
      "mean_token_accuracy": 0.18853337615728377,
      "num_tokens": 96589172.0,
      "step": 47670
    },
    {
      "entropy": 5.905094289779663,
      "epoch": 4.1663025430394125,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033674411004033333,
      "loss": 5.4116,
      "mean_token_accuracy": 0.1930481120944023,
      "num_tokens": 96599473.0,
      "step": 47675
    },
    {
      "entropy": 5.913483476638794,
      "epoch": 4.166739491392118,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003367141446115079,
      "loss": 5.4086,
      "mean_token_accuracy": 0.1979569748044014,
      "num_tokens": 96609215.0,
      "step": 47680
    },
    {
      "entropy": 5.8274126052856445,
      "epoch": 4.167176439744822,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003366841779989728,
      "loss": 5.3185,
      "mean_token_accuracy": 0.19643504917621613,
      "num_tokens": 96619742.0,
      "step": 47685
    },
    {
      "entropy": 5.90936803817749,
      "epoch": 4.167613388097527,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003366542102033028,
      "loss": 5.4124,
      "mean_token_accuracy": 0.19528215378522873,
      "num_tokens": 96629899.0,
      "step": 47690
    },
    {
      "entropy": 5.893862009048462,
      "epoch": 4.168050336450231,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003366242412250728,
      "loss": 5.2989,
      "mean_token_accuracy": 0.20050649344921112,
      "num_tokens": 96640308.0,
      "step": 47695
    },
    {
      "entropy": 5.924433851242066,
      "epoch": 4.168487284802937,
      "grad_norm": 1.125,
      "learning_rate": 0.0003365942710648576,
      "loss": 5.3561,
      "mean_token_accuracy": 0.19665995389223098,
      "num_tokens": 96649929.0,
      "step": 47700
    },
    {
      "entropy": 5.8591498851776125,
      "epoch": 4.168924233155641,
      "grad_norm": 1.03125,
      "learning_rate": 0.000336564299723232,
      "loss": 5.2736,
      "mean_token_accuracy": 0.20995769053697586,
      "num_tokens": 96659794.0,
      "step": 47705
    },
    {
      "entropy": 5.9448975086212155,
      "epoch": 4.169361181508346,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033653432720077075,
      "loss": 5.4393,
      "mean_token_accuracy": 0.19186945855617524,
      "num_tokens": 96670452.0,
      "step": 47710
    },
    {
      "entropy": 5.863938093185425,
      "epoch": 4.16979812986105,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000336504353498049,
      "loss": 5.2893,
      "mean_token_accuracy": 0.2075365364551544,
      "num_tokens": 96679926.0,
      "step": 47715
    },
    {
      "entropy": 5.89009313583374,
      "epoch": 4.170235078213755,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00033647437861564146,
      "loss": 5.3814,
      "mean_token_accuracy": 0.20330951362848282,
      "num_tokens": 96690089.0,
      "step": 47720
    },
    {
      "entropy": 5.958780241012573,
      "epoch": 4.17067202656646,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003364444025541231,
      "loss": 5.4491,
      "mean_token_accuracy": 0.19009437412023544,
      "num_tokens": 96700991.0,
      "step": 47725
    },
    {
      "entropy": 5.954361867904663,
      "epoch": 4.171108974919164,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033641442531406895,
      "loss": 5.4377,
      "mean_token_accuracy": 0.1924477681517601,
      "num_tokens": 96711077.0,
      "step": 47730
    },
    {
      "entropy": 5.926263618469238,
      "epoch": 4.17154592327187,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033638444689605386,
      "loss": 5.3327,
      "mean_token_accuracy": 0.19312449395656586,
      "num_tokens": 96721590.0,
      "step": 47735
    },
    {
      "entropy": 5.937473344802856,
      "epoch": 4.171982871624574,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000336354467300653,
      "loss": 5.3479,
      "mean_token_accuracy": 0.20519819855690002,
      "num_tokens": 96732385.0,
      "step": 47740
    },
    {
      "entropy": 5.877824592590332,
      "epoch": 4.172419819977279,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033632448652844133,
      "loss": 5.3759,
      "mean_token_accuracy": 0.20398329496383666,
      "num_tokens": 96742305.0,
      "step": 47745
    },
    {
      "entropy": 5.953591823577881,
      "epoch": 4.172856768329983,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033629450457999384,
      "loss": 5.4916,
      "mean_token_accuracy": 0.18597214221954345,
      "num_tokens": 96753573.0,
      "step": 47750
    },
    {
      "entropy": 5.89206829071045,
      "epoch": 4.1732937166826884,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003362645214558856,
      "loss": 5.2188,
      "mean_token_accuracy": 0.2099027931690216,
      "num_tokens": 96763647.0,
      "step": 47755
    },
    {
      "entropy": 5.892661809921265,
      "epoch": 4.173730665035393,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003362345371566918,
      "loss": 5.3678,
      "mean_token_accuracy": 0.19963608980178832,
      "num_tokens": 96772966.0,
      "step": 47760
    },
    {
      "entropy": 5.9108916282653805,
      "epoch": 4.174167613388097,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003362045516829874,
      "loss": 5.3403,
      "mean_token_accuracy": 0.20149747282266617,
      "num_tokens": 96783465.0,
      "step": 47765
    },
    {
      "entropy": 5.863003349304199,
      "epoch": 4.174604561740802,
      "grad_norm": 0.96875,
      "learning_rate": 0.00033617456503534773,
      "loss": 5.304,
      "mean_token_accuracy": 0.2030820518732071,
      "num_tokens": 96793460.0,
      "step": 47770
    },
    {
      "entropy": 5.933013296127319,
      "epoch": 4.175041510093507,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033614457721434774,
      "loss": 5.3993,
      "mean_token_accuracy": 0.19492313712835313,
      "num_tokens": 96803699.0,
      "step": 47775
    },
    {
      "entropy": 5.963900566101074,
      "epoch": 4.175478458446212,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003361145882205628,
      "loss": 5.3912,
      "mean_token_accuracy": 0.19774900525808334,
      "num_tokens": 96814605.0,
      "step": 47780
    },
    {
      "entropy": 5.898263311386108,
      "epoch": 4.175915406798916,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00033608459805456806,
      "loss": 5.3361,
      "mean_token_accuracy": 0.198864409327507,
      "num_tokens": 96824737.0,
      "step": 47785
    },
    {
      "entropy": 5.897314786911011,
      "epoch": 4.1763523551516215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003360546067169386,
      "loss": 5.3588,
      "mean_token_accuracy": 0.20089674293994902,
      "num_tokens": 96834771.0,
      "step": 47790
    },
    {
      "entropy": 5.933713293075561,
      "epoch": 4.176789303504326,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033602461420824994,
      "loss": 5.3685,
      "mean_token_accuracy": 0.19582572877407073,
      "num_tokens": 96844106.0,
      "step": 47795
    },
    {
      "entropy": 5.885606479644776,
      "epoch": 4.1772262518570304,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033599462052907707,
      "loss": 5.2726,
      "mean_token_accuracy": 0.20541524440050124,
      "num_tokens": 96854296.0,
      "step": 47800
    },
    {
      "entropy": 5.817772769927979,
      "epoch": 4.177663200209735,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033596462567999546,
      "loss": 5.2934,
      "mean_token_accuracy": 0.20880126655101777,
      "num_tokens": 96864351.0,
      "step": 47805
    },
    {
      "entropy": 5.879916143417359,
      "epoch": 4.17810014856244,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003359346296615804,
      "loss": 5.3807,
      "mean_token_accuracy": 0.19877593070268632,
      "num_tokens": 96874789.0,
      "step": 47810
    },
    {
      "entropy": 5.868492221832275,
      "epoch": 4.178537096915145,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003359046324744072,
      "loss": 5.279,
      "mean_token_accuracy": 0.20785239040851594,
      "num_tokens": 96885141.0,
      "step": 47815
    },
    {
      "entropy": 5.918722248077392,
      "epoch": 4.178974045267849,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033587463411905127,
      "loss": 5.3003,
      "mean_token_accuracy": 0.20660100132226944,
      "num_tokens": 96894556.0,
      "step": 47820
    },
    {
      "entropy": 5.795675277709961,
      "epoch": 4.179410993620554,
      "grad_norm": 1.0625,
      "learning_rate": 0.000335844634596088,
      "loss": 5.2354,
      "mean_token_accuracy": 0.20894389152526854,
      "num_tokens": 96904355.0,
      "step": 47825
    },
    {
      "entropy": 5.819770574569702,
      "epoch": 4.179847941973259,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033581463390609264,
      "loss": 5.2997,
      "mean_token_accuracy": 0.21064936518669128,
      "num_tokens": 96913866.0,
      "step": 47830
    },
    {
      "entropy": 5.895600700378418,
      "epoch": 4.1802848903259635,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003357846320496408,
      "loss": 5.3306,
      "mean_token_accuracy": 0.20274561643600464,
      "num_tokens": 96924142.0,
      "step": 47835
    },
    {
      "entropy": 5.915532159805298,
      "epoch": 4.180721838678668,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00033575462902730785,
      "loss": 5.2855,
      "mean_token_accuracy": 0.2078742042183876,
      "num_tokens": 96934197.0,
      "step": 47840
    },
    {
      "entropy": 5.8682044506072994,
      "epoch": 4.181158787031373,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033572462483966934,
      "loss": 5.3227,
      "mean_token_accuracy": 0.20370448976755143,
      "num_tokens": 96942932.0,
      "step": 47845
    },
    {
      "entropy": 5.851399230957031,
      "epoch": 4.181595735384078,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033569461948730064,
      "loss": 5.273,
      "mean_token_accuracy": 0.20616296976804732,
      "num_tokens": 96951859.0,
      "step": 47850
    },
    {
      "entropy": 5.818422555923462,
      "epoch": 4.182032683736782,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033566461297077733,
      "loss": 5.3258,
      "mean_token_accuracy": 0.20329700857400895,
      "num_tokens": 96962206.0,
      "step": 47855
    },
    {
      "entropy": 5.859417963027954,
      "epoch": 4.182469632089487,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033563460529067497,
      "loss": 5.3554,
      "mean_token_accuracy": 0.20537771433591842,
      "num_tokens": 96971302.0,
      "step": 47860
    },
    {
      "entropy": 5.799052429199219,
      "epoch": 4.182906580442192,
      "grad_norm": 1.953125,
      "learning_rate": 0.0003356045964475691,
      "loss": 5.2519,
      "mean_token_accuracy": 0.21143151074647903,
      "num_tokens": 96981446.0,
      "step": 47865
    },
    {
      "entropy": 5.913480567932129,
      "epoch": 4.183343528794897,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003355745864420354,
      "loss": 5.3712,
      "mean_token_accuracy": 0.19744871258735658,
      "num_tokens": 96990383.0,
      "step": 47870
    },
    {
      "entropy": 5.823591661453247,
      "epoch": 4.183780477147601,
      "grad_norm": 0.96875,
      "learning_rate": 0.00033554457527464927,
      "loss": 5.3302,
      "mean_token_accuracy": 0.20430076867341995,
      "num_tokens": 97001497.0,
      "step": 47875
    },
    {
      "entropy": 5.9353498935699465,
      "epoch": 4.1842174255003055,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003355145629459865,
      "loss": 5.3653,
      "mean_token_accuracy": 0.205584292113781,
      "num_tokens": 97011824.0,
      "step": 47880
    },
    {
      "entropy": 5.895792627334595,
      "epoch": 4.184654373853011,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003354845494566227,
      "loss": 5.3262,
      "mean_token_accuracy": 0.19732819944620134,
      "num_tokens": 97022882.0,
      "step": 47885
    },
    {
      "entropy": 5.815241861343384,
      "epoch": 4.185091322205715,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003354545348071337,
      "loss": 5.2258,
      "mean_token_accuracy": 0.21131937801837922,
      "num_tokens": 97032576.0,
      "step": 47890
    },
    {
      "entropy": 5.820606374740601,
      "epoch": 4.18552827055842,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033542451899809485,
      "loss": 5.3231,
      "mean_token_accuracy": 0.20006154626607894,
      "num_tokens": 97042602.0,
      "step": 47895
    },
    {
      "entropy": 5.850113344192505,
      "epoch": 4.185965218911125,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003353945020300821,
      "loss": 5.2947,
      "mean_token_accuracy": 0.21088721305131913,
      "num_tokens": 97052669.0,
      "step": 47900
    },
    {
      "entropy": 5.847073650360107,
      "epoch": 4.18640216726383,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033536448390367114,
      "loss": 5.2572,
      "mean_token_accuracy": 0.20829116851091384,
      "num_tokens": 97062068.0,
      "step": 47905
    },
    {
      "entropy": 5.897817230224609,
      "epoch": 4.186839115616534,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033533446461943777,
      "loss": 5.3091,
      "mean_token_accuracy": 0.20513905882835387,
      "num_tokens": 97071214.0,
      "step": 47910
    },
    {
      "entropy": 5.8459508419036865,
      "epoch": 4.187276063969239,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003353044441779578,
      "loss": 5.3705,
      "mean_token_accuracy": 0.1946142330765724,
      "num_tokens": 97080974.0,
      "step": 47915
    },
    {
      "entropy": 5.995391368865967,
      "epoch": 4.187713012321944,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033527442257980696,
      "loss": 5.4173,
      "mean_token_accuracy": 0.1917559638619423,
      "num_tokens": 97092507.0,
      "step": 47920
    },
    {
      "entropy": 5.937330102920532,
      "epoch": 4.188149960674648,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033524439982556106,
      "loss": 5.3972,
      "mean_token_accuracy": 0.19287612289190292,
      "num_tokens": 97101342.0,
      "step": 47925
    },
    {
      "entropy": 5.901267385482788,
      "epoch": 4.188586909027353,
      "grad_norm": 1.109375,
      "learning_rate": 0.000335214375915796,
      "loss": 5.3123,
      "mean_token_accuracy": 0.19860694110393523,
      "num_tokens": 97112174.0,
      "step": 47930
    },
    {
      "entropy": 5.866397047042847,
      "epoch": 4.189023857380057,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003351843508510878,
      "loss": 5.3306,
      "mean_token_accuracy": 0.19977763444185256,
      "num_tokens": 97122557.0,
      "step": 47935
    },
    {
      "entropy": 5.88834958076477,
      "epoch": 4.189460805732763,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033515432463201213,
      "loss": 5.313,
      "mean_token_accuracy": 0.20316292196512223,
      "num_tokens": 97132565.0,
      "step": 47940
    },
    {
      "entropy": 5.896894741058349,
      "epoch": 4.189897754085467,
      "grad_norm": 1.109375,
      "learning_rate": 0.000335124297259145,
      "loss": 5.2886,
      "mean_token_accuracy": 0.2020077094435692,
      "num_tokens": 97141367.0,
      "step": 47945
    },
    {
      "entropy": 5.908653688430786,
      "epoch": 4.190334702438172,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003350942687330623,
      "loss": 5.3445,
      "mean_token_accuracy": 0.19908449947834014,
      "num_tokens": 97151859.0,
      "step": 47950
    },
    {
      "entropy": 5.887585067749024,
      "epoch": 4.190771650790876,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033506423905434,
      "loss": 5.3982,
      "mean_token_accuracy": 0.19361565113067628,
      "num_tokens": 97161670.0,
      "step": 47955
    },
    {
      "entropy": 5.827013254165649,
      "epoch": 4.1912085991435815,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003350342082235543,
      "loss": 5.2632,
      "mean_token_accuracy": 0.21326120793819428,
      "num_tokens": 97171089.0,
      "step": 47960
    },
    {
      "entropy": 5.913342046737671,
      "epoch": 4.191645547496286,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00033500417624128087,
      "loss": 5.4095,
      "mean_token_accuracy": 0.19842861145734786,
      "num_tokens": 97181934.0,
      "step": 47965
    },
    {
      "entropy": 5.896605396270752,
      "epoch": 4.19208249584899,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000334974143108096,
      "loss": 5.319,
      "mean_token_accuracy": 0.201340451836586,
      "num_tokens": 97191150.0,
      "step": 47970
    },
    {
      "entropy": 5.866413545608521,
      "epoch": 4.192519444201696,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003349441088245756,
      "loss": 5.3721,
      "mean_token_accuracy": 0.19977397173643113,
      "num_tokens": 97202015.0,
      "step": 47975
    },
    {
      "entropy": 5.844359683990478,
      "epoch": 4.1929563925544,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003349140733912957,
      "loss": 5.3227,
      "mean_token_accuracy": 0.20608343780040742,
      "num_tokens": 97211418.0,
      "step": 47980
    },
    {
      "entropy": 5.940219879150391,
      "epoch": 4.193393340907105,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003348840368088326,
      "loss": 5.4579,
      "mean_token_accuracy": 0.1901096671819687,
      "num_tokens": 97222557.0,
      "step": 47985
    },
    {
      "entropy": 5.928141164779663,
      "epoch": 4.193830289259809,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003348539990777623,
      "loss": 5.4464,
      "mean_token_accuracy": 0.1972998321056366,
      "num_tokens": 97231966.0,
      "step": 47990
    },
    {
      "entropy": 5.89643611907959,
      "epoch": 4.1942672376125145,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003348239601986608,
      "loss": 5.303,
      "mean_token_accuracy": 0.20060738772153855,
      "num_tokens": 97241843.0,
      "step": 47995
    },
    {
      "entropy": 5.821544075012207,
      "epoch": 4.194704185965219,
      "grad_norm": 1.0,
      "learning_rate": 0.0003347939201721045,
      "loss": 5.2924,
      "mean_token_accuracy": 0.21063555926084518,
      "num_tokens": 97251987.0,
      "step": 48000
    },
    {
      "epoch": 4.194704185965219,
      "eval_entropy": 5.598088897328307,
      "eval_loss": 5.417087078094482,
      "eval_mean_token_accuracy": 0.20532225384850958,
      "eval_num_tokens": 97251987.0,
      "eval_runtime": 27.5696,
      "eval_samples_per_second": 1342.566,
      "eval_steps_per_second": 167.83,
      "step": 48000
    },
    {
      "entropy": 5.850762891769409,
      "epoch": 4.1951411343179235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033476387899866944,
      "loss": 5.3346,
      "mean_token_accuracy": 0.19923352301120759,
      "num_tokens": 97262489.0,
      "step": 48005
    },
    {
      "entropy": 5.95567946434021,
      "epoch": 4.195578082670628,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033473383667893185,
      "loss": 5.3788,
      "mean_token_accuracy": 0.20131803750991822,
      "num_tokens": 97272380.0,
      "step": 48010
    },
    {
      "entropy": 5.838804388046265,
      "epoch": 4.196015031023333,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033470379321346805,
      "loss": 5.2925,
      "mean_token_accuracy": 0.20357140600681306,
      "num_tokens": 97282313.0,
      "step": 48015
    },
    {
      "entropy": 5.924552869796753,
      "epoch": 4.196451979376038,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033467374860285414,
      "loss": 5.3941,
      "mean_token_accuracy": 0.19246214032173156,
      "num_tokens": 97292720.0,
      "step": 48020
    },
    {
      "entropy": 5.846190881729126,
      "epoch": 4.196888927728742,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033464370284766654,
      "loss": 5.3085,
      "mean_token_accuracy": 0.20225357711315156,
      "num_tokens": 97301988.0,
      "step": 48025
    },
    {
      "entropy": 5.851840877532959,
      "epoch": 4.197325876081448,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00033461365594848134,
      "loss": 5.3407,
      "mean_token_accuracy": 0.2027359202504158,
      "num_tokens": 97312594.0,
      "step": 48030
    },
    {
      "entropy": 5.949756097793579,
      "epoch": 4.197762824434152,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003345836079058751,
      "loss": 5.3825,
      "mean_token_accuracy": 0.19345725029706956,
      "num_tokens": 97321727.0,
      "step": 48035
    },
    {
      "entropy": 5.811826896667481,
      "epoch": 4.1981997727868565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000334553558720424,
      "loss": 5.1688,
      "mean_token_accuracy": 0.2174295589327812,
      "num_tokens": 97331180.0,
      "step": 48040
    },
    {
      "entropy": 5.7675461769104,
      "epoch": 4.198636721139561,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003345235083927044,
      "loss": 5.1738,
      "mean_token_accuracy": 0.21600193679332733,
      "num_tokens": 97340432.0,
      "step": 48045
    },
    {
      "entropy": 5.8243505477905275,
      "epoch": 4.199073669492266,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033449345692329276,
      "loss": 5.2493,
      "mean_token_accuracy": 0.19866276830434798,
      "num_tokens": 97350493.0,
      "step": 48050
    },
    {
      "entropy": 5.832851314544678,
      "epoch": 4.199510617844971,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003344634043127655,
      "loss": 5.3032,
      "mean_token_accuracy": 0.2115929290652275,
      "num_tokens": 97360484.0,
      "step": 48055
    },
    {
      "entropy": 5.865173721313477,
      "epoch": 4.199947566197675,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003344333505616989,
      "loss": 5.3112,
      "mean_token_accuracy": 0.20933400094509125,
      "num_tokens": 97371575.0,
      "step": 48060
    },
    {
      "entropy": 5.792748355865479,
      "epoch": 4.20038451455038,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003344032956706695,
      "loss": 5.2162,
      "mean_token_accuracy": 0.2095935583114624,
      "num_tokens": 97381806.0,
      "step": 48065
    },
    {
      "entropy": 5.8758069515228275,
      "epoch": 4.200821462903085,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033437323964025377,
      "loss": 5.2637,
      "mean_token_accuracy": 0.20853301435708999,
      "num_tokens": 97391679.0,
      "step": 48070
    },
    {
      "entropy": 5.915641450881958,
      "epoch": 4.20125841125579,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033434318247102816,
      "loss": 5.3965,
      "mean_token_accuracy": 0.1935171216726303,
      "num_tokens": 97401683.0,
      "step": 48075
    },
    {
      "entropy": 5.799290418624878,
      "epoch": 4.201695359608494,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003343131241635693,
      "loss": 5.2822,
      "mean_token_accuracy": 0.20661372542381287,
      "num_tokens": 97411650.0,
      "step": 48080
    },
    {
      "entropy": 5.896517133712768,
      "epoch": 4.202132307961199,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003342830647184536,
      "loss": 5.3462,
      "mean_token_accuracy": 0.20113369524478913,
      "num_tokens": 97421246.0,
      "step": 48085
    },
    {
      "entropy": 5.902156972885132,
      "epoch": 4.202569256313904,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003342530041362576,
      "loss": 5.3902,
      "mean_token_accuracy": 0.19583599120378495,
      "num_tokens": 97431287.0,
      "step": 48090
    },
    {
      "entropy": 5.819298934936524,
      "epoch": 4.203006204666608,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033422294241755796,
      "loss": 5.2024,
      "mean_token_accuracy": 0.21194703727960587,
      "num_tokens": 97442283.0,
      "step": 48095
    },
    {
      "entropy": 5.898147678375244,
      "epoch": 4.203443153019313,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003341928795629313,
      "loss": 5.2709,
      "mean_token_accuracy": 0.2046196609735489,
      "num_tokens": 97451993.0,
      "step": 48100
    },
    {
      "entropy": 5.8616852283477785,
      "epoch": 4.203880101372018,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003341628155729542,
      "loss": 5.3082,
      "mean_token_accuracy": 0.20471231639385223,
      "num_tokens": 97462441.0,
      "step": 48105
    },
    {
      "entropy": 5.862246942520142,
      "epoch": 4.204317049724723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033413275044820326,
      "loss": 5.2726,
      "mean_token_accuracy": 0.20068398714065552,
      "num_tokens": 97472827.0,
      "step": 48110
    },
    {
      "entropy": 5.946442031860352,
      "epoch": 4.204753998077427,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033410268418925513,
      "loss": 5.4213,
      "mean_token_accuracy": 0.19320209622383117,
      "num_tokens": 97483696.0,
      "step": 48115
    },
    {
      "entropy": 5.7990114212036135,
      "epoch": 4.205190946430132,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003340726167966867,
      "loss": 5.2659,
      "mean_token_accuracy": 0.2147834673523903,
      "num_tokens": 97493846.0,
      "step": 48120
    },
    {
      "entropy": 5.9185731410980225,
      "epoch": 4.205627894782837,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003340425482710743,
      "loss": 5.3182,
      "mean_token_accuracy": 0.20571119338274002,
      "num_tokens": 97504216.0,
      "step": 48125
    },
    {
      "entropy": 5.914893722534179,
      "epoch": 4.206064843135541,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003340124786129951,
      "loss": 5.4144,
      "mean_token_accuracy": 0.19135391861200332,
      "num_tokens": 97515464.0,
      "step": 48130
    },
    {
      "entropy": 5.872373437881469,
      "epoch": 4.206501791488246,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033398240782302554,
      "loss": 5.3787,
      "mean_token_accuracy": 0.19809639006853103,
      "num_tokens": 97525702.0,
      "step": 48135
    },
    {
      "entropy": 5.9076403141021725,
      "epoch": 4.20693873984095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003339523359017425,
      "loss": 5.3597,
      "mean_token_accuracy": 0.20461036562919616,
      "num_tokens": 97536712.0,
      "step": 48140
    },
    {
      "entropy": 5.938782024383545,
      "epoch": 4.207375688193656,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003339222628497228,
      "loss": 5.3293,
      "mean_token_accuracy": 0.19695979207754136,
      "num_tokens": 97544976.0,
      "step": 48145
    },
    {
      "entropy": 5.893097877502441,
      "epoch": 4.20781263654636,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003338921886675432,
      "loss": 5.396,
      "mean_token_accuracy": 0.19563318490982057,
      "num_tokens": 97555140.0,
      "step": 48150
    },
    {
      "entropy": 5.856651735305786,
      "epoch": 4.208249584899065,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033386211335578053,
      "loss": 5.2609,
      "mean_token_accuracy": 0.20025182962417604,
      "num_tokens": 97564578.0,
      "step": 48155
    },
    {
      "entropy": 5.88273515701294,
      "epoch": 4.20868653325177,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003338320369150117,
      "loss": 5.3206,
      "mean_token_accuracy": 0.20185039788484574,
      "num_tokens": 97574585.0,
      "step": 48160
    },
    {
      "entropy": 5.8298290252685545,
      "epoch": 4.2091234816044745,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033380195934581356,
      "loss": 5.2841,
      "mean_token_accuracy": 0.21033735275268556,
      "num_tokens": 97586133.0,
      "step": 48165
    },
    {
      "entropy": 5.8654993057250975,
      "epoch": 4.209560429957179,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003337718806487631,
      "loss": 5.302,
      "mean_token_accuracy": 0.20126153379678727,
      "num_tokens": 97595940.0,
      "step": 48170
    },
    {
      "entropy": 5.783565950393677,
      "epoch": 4.209997378309883,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003337418008244371,
      "loss": 5.2582,
      "mean_token_accuracy": 0.21370138376951217,
      "num_tokens": 97605656.0,
      "step": 48175
    },
    {
      "entropy": 5.916346311569214,
      "epoch": 4.210434326662589,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033371171987341266,
      "loss": 5.4104,
      "mean_token_accuracy": 0.1956710085272789,
      "num_tokens": 97615256.0,
      "step": 48180
    },
    {
      "entropy": 5.926674318313599,
      "epoch": 4.210871275015293,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033368163779626654,
      "loss": 5.3495,
      "mean_token_accuracy": 0.20422871708869933,
      "num_tokens": 97624820.0,
      "step": 48185
    },
    {
      "entropy": 5.92693076133728,
      "epoch": 4.211308223367998,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00033365155459357595,
      "loss": 5.3717,
      "mean_token_accuracy": 0.2004949226975441,
      "num_tokens": 97636221.0,
      "step": 48190
    },
    {
      "entropy": 5.852913761138916,
      "epoch": 4.211745171720702,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003336214702659179,
      "loss": 5.3378,
      "mean_token_accuracy": 0.20078338980674743,
      "num_tokens": 97646460.0,
      "step": 48195
    },
    {
      "entropy": 5.8378208637237545,
      "epoch": 4.2121821200734075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033359138481386914,
      "loss": 5.2893,
      "mean_token_accuracy": 0.1995704397559166,
      "num_tokens": 97655762.0,
      "step": 48200
    },
    {
      "entropy": 5.830496549606323,
      "epoch": 4.212619068426112,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000333561298238007,
      "loss": 5.2784,
      "mean_token_accuracy": 0.21104613095521926,
      "num_tokens": 97665916.0,
      "step": 48205
    },
    {
      "entropy": 5.935834741592407,
      "epoch": 4.2130560167788165,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033353121053890844,
      "loss": 5.3698,
      "mean_token_accuracy": 0.19706534147262572,
      "num_tokens": 97675529.0,
      "step": 48210
    },
    {
      "entropy": 5.839515924453735,
      "epoch": 4.213492965131522,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033350112171715067,
      "loss": 5.2844,
      "mean_token_accuracy": 0.2027289777994156,
      "num_tokens": 97684860.0,
      "step": 48215
    },
    {
      "entropy": 5.849429655075073,
      "epoch": 4.213929913484226,
      "grad_norm": 1.640625,
      "learning_rate": 0.00033347103177331066,
      "loss": 5.3089,
      "mean_token_accuracy": 0.2044660210609436,
      "num_tokens": 97695271.0,
      "step": 48220
    },
    {
      "entropy": 5.91301007270813,
      "epoch": 4.214366861836931,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003334409407079657,
      "loss": 5.2963,
      "mean_token_accuracy": 0.2054068326950073,
      "num_tokens": 97706346.0,
      "step": 48225
    },
    {
      "entropy": 5.895269536972046,
      "epoch": 4.214803810189635,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003334108485216928,
      "loss": 5.2925,
      "mean_token_accuracy": 0.2005167692899704,
      "num_tokens": 97716667.0,
      "step": 48230
    },
    {
      "entropy": 5.881251001358033,
      "epoch": 4.215240758542341,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003333807552150693,
      "loss": 5.3451,
      "mean_token_accuracy": 0.19798324555158614,
      "num_tokens": 97725868.0,
      "step": 48235
    },
    {
      "entropy": 5.883131504058838,
      "epoch": 4.215677706895045,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003333506607886722,
      "loss": 5.2867,
      "mean_token_accuracy": 0.20115641206502916,
      "num_tokens": 97735308.0,
      "step": 48240
    },
    {
      "entropy": 5.896927547454834,
      "epoch": 4.2161146552477495,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000333320565243079,
      "loss": 5.3153,
      "mean_token_accuracy": 0.20228130221366883,
      "num_tokens": 97745345.0,
      "step": 48245
    },
    {
      "entropy": 5.8472494125366214,
      "epoch": 4.216551603600454,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003332904685788667,
      "loss": 5.3427,
      "mean_token_accuracy": 0.19805748760700226,
      "num_tokens": 97754858.0,
      "step": 48250
    },
    {
      "entropy": 5.816593742370605,
      "epoch": 4.216988551953159,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003332603707966127,
      "loss": 5.2969,
      "mean_token_accuracy": 0.20286710411310196,
      "num_tokens": 97764681.0,
      "step": 48255
    },
    {
      "entropy": 5.871351432800293,
      "epoch": 4.217425500305864,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003332302718968942,
      "loss": 5.3279,
      "mean_token_accuracy": 0.20053336173295974,
      "num_tokens": 97775051.0,
      "step": 48260
    },
    {
      "entropy": 5.894848203659057,
      "epoch": 4.217862448658568,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003332001718802886,
      "loss": 5.3309,
      "mean_token_accuracy": 0.20393735617399217,
      "num_tokens": 97787263.0,
      "step": 48265
    },
    {
      "entropy": 6.001870775222779,
      "epoch": 4.218299397011274,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003331700707473733,
      "loss": 5.373,
      "mean_token_accuracy": 0.19807576090097428,
      "num_tokens": 97797507.0,
      "step": 48270
    },
    {
      "entropy": 5.882831859588623,
      "epoch": 4.218736345363978,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033313996849872543,
      "loss": 5.3301,
      "mean_token_accuracy": 0.19969818741083145,
      "num_tokens": 97807304.0,
      "step": 48275
    },
    {
      "entropy": 5.772719621658325,
      "epoch": 4.219173293716683,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003331098651349225,
      "loss": 5.2814,
      "mean_token_accuracy": 0.2127803549170494,
      "num_tokens": 97817552.0,
      "step": 48280
    },
    {
      "entropy": 5.859962129592896,
      "epoch": 4.219610242069387,
      "grad_norm": 1.0,
      "learning_rate": 0.000333079760656542,
      "loss": 5.2828,
      "mean_token_accuracy": 0.20624052882194518,
      "num_tokens": 97827635.0,
      "step": 48285
    },
    {
      "entropy": 5.908206844329834,
      "epoch": 4.220047190422092,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033304965506416126,
      "loss": 5.3795,
      "mean_token_accuracy": 0.19610296487808226,
      "num_tokens": 97836363.0,
      "step": 48290
    },
    {
      "entropy": 5.8404566764831545,
      "epoch": 4.220484138774797,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003330195483583577,
      "loss": 5.1982,
      "mean_token_accuracy": 0.21312407553195953,
      "num_tokens": 97846877.0,
      "step": 48295
    },
    {
      "entropy": 5.879316568374634,
      "epoch": 4.220921087127501,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003329894405397088,
      "loss": 5.3965,
      "mean_token_accuracy": 0.19960717856884003,
      "num_tokens": 97856502.0,
      "step": 48300
    },
    {
      "entropy": 5.872004461288452,
      "epoch": 4.221358035480206,
      "grad_norm": 1.046875,
      "learning_rate": 0.000332959331608792,
      "loss": 5.36,
      "mean_token_accuracy": 0.19810020327568054,
      "num_tokens": 97866213.0,
      "step": 48305
    },
    {
      "entropy": 5.9527599811553955,
      "epoch": 4.221794983832911,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00033292922156618487,
      "loss": 5.4543,
      "mean_token_accuracy": 0.19255317151546478,
      "num_tokens": 97877074.0,
      "step": 48310
    },
    {
      "entropy": 5.831930780410767,
      "epoch": 4.222231932185616,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000332899110412465,
      "loss": 5.2137,
      "mean_token_accuracy": 0.21243291199207306,
      "num_tokens": 97887090.0,
      "step": 48315
    },
    {
      "entropy": 5.906671953201294,
      "epoch": 4.22266888053832,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003328689981482097,
      "loss": 5.3859,
      "mean_token_accuracy": 0.19337212592363356,
      "num_tokens": 97897390.0,
      "step": 48320
    },
    {
      "entropy": 5.883644342422485,
      "epoch": 4.2231058288910255,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033283888477399676,
      "loss": 5.2205,
      "mean_token_accuracy": 0.20391543209552765,
      "num_tokens": 97907126.0,
      "step": 48325
    },
    {
      "entropy": 5.8627832412719725,
      "epoch": 4.22354277724373,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00033280877029040376,
      "loss": 5.344,
      "mean_token_accuracy": 0.20326261073350907,
      "num_tokens": 97916680.0,
      "step": 48330
    },
    {
      "entropy": 5.843525791168213,
      "epoch": 4.223979725596434,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003327786546980082,
      "loss": 5.3077,
      "mean_token_accuracy": 0.2017510086297989,
      "num_tokens": 97926833.0,
      "step": 48335
    },
    {
      "entropy": 5.839425086975098,
      "epoch": 4.224416673949139,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033274853799738775,
      "loss": 5.2878,
      "mean_token_accuracy": 0.2002599745988846,
      "num_tokens": 97936040.0,
      "step": 48340
    },
    {
      "entropy": 5.817354440689087,
      "epoch": 4.224853622301844,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003327184201891201,
      "loss": 5.2738,
      "mean_token_accuracy": 0.20006984025239943,
      "num_tokens": 97946799.0,
      "step": 48345
    },
    {
      "entropy": 5.853382873535156,
      "epoch": 4.225290570654549,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000332688301273783,
      "loss": 5.2946,
      "mean_token_accuracy": 0.20768300741910933,
      "num_tokens": 97956070.0,
      "step": 48350
    },
    {
      "entropy": 5.892854022979736,
      "epoch": 4.225727519007253,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00033265818125195393,
      "loss": 5.4472,
      "mean_token_accuracy": 0.19196804165840148,
      "num_tokens": 97966088.0,
      "step": 48355
    },
    {
      "entropy": 5.943794393539429,
      "epoch": 4.226164467359958,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033262806012421075,
      "loss": 5.3759,
      "mean_token_accuracy": 0.19822562038898467,
      "num_tokens": 97977394.0,
      "step": 48360
    },
    {
      "entropy": 5.9112647533416744,
      "epoch": 4.226601415712663,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003325979378911312,
      "loss": 5.3149,
      "mean_token_accuracy": 0.20133551210165024,
      "num_tokens": 97987117.0,
      "step": 48365
    },
    {
      "entropy": 5.846525621414185,
      "epoch": 4.2270383640653675,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000332567814553293,
      "loss": 5.3596,
      "mean_token_accuracy": 0.20142561048269272,
      "num_tokens": 97999127.0,
      "step": 48370
    },
    {
      "entropy": 5.910014200210571,
      "epoch": 4.227475312418072,
      "grad_norm": 1.09375,
      "learning_rate": 0.000332537690111274,
      "loss": 5.3773,
      "mean_token_accuracy": 0.20383962392807006,
      "num_tokens": 98008260.0,
      "step": 48375
    },
    {
      "entropy": 5.872779464721679,
      "epoch": 4.227912260770777,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003325075645656519,
      "loss": 5.2837,
      "mean_token_accuracy": 0.20994213819503785,
      "num_tokens": 98017494.0,
      "step": 48380
    },
    {
      "entropy": 5.8145115852355955,
      "epoch": 4.228349209123482,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033247743791700466,
      "loss": 5.2289,
      "mean_token_accuracy": 0.2154061585664749,
      "num_tokens": 98028873.0,
      "step": 48385
    },
    {
      "entropy": 5.895682144165039,
      "epoch": 4.228786157476186,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033244731016591,
      "loss": 5.3315,
      "mean_token_accuracy": 0.20677212178707122,
      "num_tokens": 98039178.0,
      "step": 48390
    },
    {
      "entropy": 5.858576822280884,
      "epoch": 4.229223105828891,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003324171813129457,
      "loss": 5.3399,
      "mean_token_accuracy": 0.20379546731710435,
      "num_tokens": 98048040.0,
      "step": 48395
    },
    {
      "entropy": 5.816562557220459,
      "epoch": 4.229660054181596,
      "grad_norm": 0.96875,
      "learning_rate": 0.00033238705135868994,
      "loss": 5.2739,
      "mean_token_accuracy": 0.21242996901273728,
      "num_tokens": 98057648.0,
      "step": 48400
    },
    {
      "entropy": 5.882165431976318,
      "epoch": 4.2300970025343005,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033235692030372045,
      "loss": 5.2987,
      "mean_token_accuracy": 0.20165475755929946,
      "num_tokens": 98067509.0,
      "step": 48405
    },
    {
      "entropy": 5.95192551612854,
      "epoch": 4.230533950887005,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003323267881486151,
      "loss": 5.4354,
      "mean_token_accuracy": 0.19868575036525726,
      "num_tokens": 98077946.0,
      "step": 48410
    },
    {
      "entropy": 5.875981378555298,
      "epoch": 4.2309708992397095,
      "grad_norm": 1.0625,
      "learning_rate": 0.000332296654893952,
      "loss": 5.3863,
      "mean_token_accuracy": 0.20235239416360856,
      "num_tokens": 98088046.0,
      "step": 48415
    },
    {
      "entropy": 5.783696603775025,
      "epoch": 4.231407847592415,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000332266520540309,
      "loss": 5.2509,
      "mean_token_accuracy": 0.20794788748025894,
      "num_tokens": 98097501.0,
      "step": 48420
    },
    {
      "entropy": 5.865647220611573,
      "epoch": 4.231844795945119,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003322363850882641,
      "loss": 5.3184,
      "mean_token_accuracy": 0.20951129943132402,
      "num_tokens": 98109193.0,
      "step": 48425
    },
    {
      "entropy": 5.776251983642578,
      "epoch": 4.232281744297824,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003322062485383953,
      "loss": 5.1932,
      "mean_token_accuracy": 0.21387260109186174,
      "num_tokens": 98118700.0,
      "step": 48430
    },
    {
      "entropy": 5.8993021011352536,
      "epoch": 4.232718692650529,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00033217611089128086,
      "loss": 5.3287,
      "mean_token_accuracy": 0.19897877871990205,
      "num_tokens": 98128913.0,
      "step": 48435
    },
    {
      "entropy": 5.902312135696411,
      "epoch": 4.233155641003234,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00033214597214749846,
      "loss": 5.381,
      "mean_token_accuracy": 0.19520273953676223,
      "num_tokens": 98139714.0,
      "step": 48440
    },
    {
      "entropy": 5.864932680130005,
      "epoch": 4.233592589355938,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033211583230762637,
      "loss": 5.2788,
      "mean_token_accuracy": 0.19971830844879152,
      "num_tokens": 98149289.0,
      "step": 48445
    },
    {
      "entropy": 5.75885043144226,
      "epoch": 4.2340295377086425,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033208569137224274,
      "loss": 5.178,
      "mean_token_accuracy": 0.21833908855915068,
      "num_tokens": 98158370.0,
      "step": 48450
    },
    {
      "entropy": 5.868388223648071,
      "epoch": 4.234466486061348,
      "grad_norm": 1.125,
      "learning_rate": 0.0003320555493419257,
      "loss": 5.3506,
      "mean_token_accuracy": 0.1986581102013588,
      "num_tokens": 98167798.0,
      "step": 48455
    },
    {
      "entropy": 5.852273321151733,
      "epoch": 4.234903434414052,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003320254062172533,
      "loss": 5.3581,
      "mean_token_accuracy": 0.19816964715719224,
      "num_tokens": 98178206.0,
      "step": 48460
    },
    {
      "entropy": 5.90721435546875,
      "epoch": 4.235340382766757,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033199526199880375,
      "loss": 5.367,
      "mean_token_accuracy": 0.19997959285974504,
      "num_tokens": 98188529.0,
      "step": 48465
    },
    {
      "entropy": 5.87248854637146,
      "epoch": 4.235777331119461,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033196511668715515,
      "loss": 5.3431,
      "mean_token_accuracy": 0.20591501742601395,
      "num_tokens": 98199578.0,
      "step": 48470
    },
    {
      "entropy": 5.87351360321045,
      "epoch": 4.236214279472167,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033193497028288576,
      "loss": 5.2754,
      "mean_token_accuracy": 0.2018424853682518,
      "num_tokens": 98209439.0,
      "step": 48475
    },
    {
      "entropy": 5.891393947601318,
      "epoch": 4.236651227824871,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033190482278657384,
      "loss": 5.3414,
      "mean_token_accuracy": 0.2008736938238144,
      "num_tokens": 98218877.0,
      "step": 48480
    },
    {
      "entropy": 5.861390113830566,
      "epoch": 4.237088176177576,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003318746741987976,
      "loss": 5.3,
      "mean_token_accuracy": 0.19800802171230317,
      "num_tokens": 98228719.0,
      "step": 48485
    },
    {
      "entropy": 5.8480997562408445,
      "epoch": 4.237525124530281,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003318445245201353,
      "loss": 5.2425,
      "mean_token_accuracy": 0.20556473433971406,
      "num_tokens": 98238299.0,
      "step": 48490
    },
    {
      "entropy": 5.88794732093811,
      "epoch": 4.237962072882985,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033181437375116526,
      "loss": 5.2921,
      "mean_token_accuracy": 0.20263565629720687,
      "num_tokens": 98248193.0,
      "step": 48495
    },
    {
      "entropy": 5.917046785354614,
      "epoch": 4.23839902123569,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003317842218924658,
      "loss": 5.4207,
      "mean_token_accuracy": 0.19450930804014205,
      "num_tokens": 98258495.0,
      "step": 48500
    },
    {
      "entropy": 5.897523593902588,
      "epoch": 4.238835969588394,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003317540689446151,
      "loss": 5.3265,
      "mean_token_accuracy": 0.20675169676542282,
      "num_tokens": 98268265.0,
      "step": 48505
    },
    {
      "entropy": 5.825579833984375,
      "epoch": 4.2392729179411,
      "grad_norm": 0.96875,
      "learning_rate": 0.00033172391490819166,
      "loss": 5.2811,
      "mean_token_accuracy": 0.2019793748855591,
      "num_tokens": 98278324.0,
      "step": 48510
    },
    {
      "entropy": 5.93631944656372,
      "epoch": 4.239709866293804,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003316937597837738,
      "loss": 5.3845,
      "mean_token_accuracy": 0.1971125304698944,
      "num_tokens": 98288576.0,
      "step": 48515
    },
    {
      "entropy": 5.88647723197937,
      "epoch": 4.240146814646509,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033166360357193993,
      "loss": 5.3097,
      "mean_token_accuracy": 0.2036693960428238,
      "num_tokens": 98298435.0,
      "step": 48520
    },
    {
      "entropy": 5.849487352371216,
      "epoch": 4.240583762999213,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003316334462732685,
      "loss": 5.2655,
      "mean_token_accuracy": 0.21101195216178895,
      "num_tokens": 98308099.0,
      "step": 48525
    },
    {
      "entropy": 5.842737245559692,
      "epoch": 4.2410207113519185,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003316032878883378,
      "loss": 5.2969,
      "mean_token_accuracy": 0.20614964365959168,
      "num_tokens": 98318252.0,
      "step": 48530
    },
    {
      "entropy": 5.848527908325195,
      "epoch": 4.241457659704623,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003315731284177265,
      "loss": 5.2743,
      "mean_token_accuracy": 0.20765596777200698,
      "num_tokens": 98328851.0,
      "step": 48535
    },
    {
      "entropy": 5.831151056289673,
      "epoch": 4.241894608057327,
      "grad_norm": 1.5625,
      "learning_rate": 0.00033154296786201286,
      "loss": 5.3036,
      "mean_token_accuracy": 0.20155992209911347,
      "num_tokens": 98340099.0,
      "step": 48540
    },
    {
      "entropy": 5.873061609268189,
      "epoch": 4.242331556410033,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003315128062217755,
      "loss": 5.3718,
      "mean_token_accuracy": 0.19599780142307283,
      "num_tokens": 98350919.0,
      "step": 48545
    },
    {
      "entropy": 5.980481433868408,
      "epoch": 4.242768504762737,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003314826434975929,
      "loss": 5.3712,
      "mean_token_accuracy": 0.19736094772815704,
      "num_tokens": 98360808.0,
      "step": 48550
    },
    {
      "entropy": 5.960790205001831,
      "epoch": 4.243205453115442,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033145247969004353,
      "loss": 5.4353,
      "mean_token_accuracy": 0.19676820933818817,
      "num_tokens": 98370620.0,
      "step": 48555
    },
    {
      "entropy": 5.900077152252197,
      "epoch": 4.243642401468146,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033142231479970616,
      "loss": 5.4448,
      "mean_token_accuracy": 0.19120727479457855,
      "num_tokens": 98379964.0,
      "step": 48560
    },
    {
      "entropy": 5.880962800979614,
      "epoch": 4.2440793498208516,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003313921488271591,
      "loss": 5.298,
      "mean_token_accuracy": 0.21167589128017425,
      "num_tokens": 98389744.0,
      "step": 48565
    },
    {
      "entropy": 5.857263803482056,
      "epoch": 4.244516298173556,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003313619817729812,
      "loss": 5.332,
      "mean_token_accuracy": 0.202119842171669,
      "num_tokens": 98399804.0,
      "step": 48570
    },
    {
      "entropy": 5.9203102588653564,
      "epoch": 4.2449532465262605,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033133181363775085,
      "loss": 5.3741,
      "mean_token_accuracy": 0.19776804596185685,
      "num_tokens": 98410064.0,
      "step": 48575
    },
    {
      "entropy": 5.855394554138184,
      "epoch": 4.245390194878965,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003313016444220468,
      "loss": 5.2509,
      "mean_token_accuracy": 0.20696392506361008,
      "num_tokens": 98421502.0,
      "step": 48580
    },
    {
      "entropy": 5.86695704460144,
      "epoch": 4.24582714323167,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033127147412644774,
      "loss": 5.3083,
      "mean_token_accuracy": 0.20802740305662154,
      "num_tokens": 98431495.0,
      "step": 48585
    },
    {
      "entropy": 5.862976217269898,
      "epoch": 4.246264091584375,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033124130275153236,
      "loss": 5.4126,
      "mean_token_accuracy": 0.20131744146347047,
      "num_tokens": 98442008.0,
      "step": 48590
    },
    {
      "entropy": 5.903386068344116,
      "epoch": 4.246701039937079,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033121113029787924,
      "loss": 5.372,
      "mean_token_accuracy": 0.19954962581396102,
      "num_tokens": 98452134.0,
      "step": 48595
    },
    {
      "entropy": 5.91618127822876,
      "epoch": 4.247137988289784,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033118095676606724,
      "loss": 5.3891,
      "mean_token_accuracy": 0.1928120419383049,
      "num_tokens": 98462713.0,
      "step": 48600
    },
    {
      "entropy": 5.88513822555542,
      "epoch": 4.247574936642489,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003311507821566751,
      "loss": 5.3094,
      "mean_token_accuracy": 0.2045045256614685,
      "num_tokens": 98473017.0,
      "step": 48605
    },
    {
      "entropy": 5.8368837356567385,
      "epoch": 4.2480118849951936,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00033112060647028145,
      "loss": 5.2561,
      "mean_token_accuracy": 0.20845529437065125,
      "num_tokens": 98482436.0,
      "step": 48610
    },
    {
      "entropy": 5.934141969680786,
      "epoch": 4.248448833347898,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033109042970746517,
      "loss": 5.3768,
      "mean_token_accuracy": 0.1955185726284981,
      "num_tokens": 98493289.0,
      "step": 48615
    },
    {
      "entropy": 5.956830978393555,
      "epoch": 4.248885781700603,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033106025186880505,
      "loss": 5.3624,
      "mean_token_accuracy": 0.2009049355983734,
      "num_tokens": 98503847.0,
      "step": 48620
    },
    {
      "entropy": 5.8857663631439205,
      "epoch": 4.249322730053308,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00033103007295488,
      "loss": 5.3583,
      "mean_token_accuracy": 0.20252579748630523,
      "num_tokens": 98514242.0,
      "step": 48625
    },
    {
      "entropy": 5.826037263870239,
      "epoch": 4.249759678406012,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003309998929662687,
      "loss": 5.2305,
      "mean_token_accuracy": 0.2093748092651367,
      "num_tokens": 98523323.0,
      "step": 48630
    },
    {
      "entropy": 5.832148218154908,
      "epoch": 4.250196626758717,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003309697119035502,
      "loss": 5.2545,
      "mean_token_accuracy": 0.21305914223194122,
      "num_tokens": 98532870.0,
      "step": 48635
    },
    {
      "entropy": 5.858319520950317,
      "epoch": 4.250633575111422,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033093952976730325,
      "loss": 5.3638,
      "mean_token_accuracy": 0.1981108993291855,
      "num_tokens": 98542795.0,
      "step": 48640
    },
    {
      "entropy": 5.817055892944336,
      "epoch": 4.251070523464127,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003309093465581068,
      "loss": 5.2351,
      "mean_token_accuracy": 0.20924772918224335,
      "num_tokens": 98552627.0,
      "step": 48645
    },
    {
      "entropy": 5.865168762207031,
      "epoch": 4.251507471816831,
      "grad_norm": 1.0,
      "learning_rate": 0.0003308791622765398,
      "loss": 5.3036,
      "mean_token_accuracy": 0.20411608666181563,
      "num_tokens": 98562562.0,
      "step": 48650
    },
    {
      "entropy": 5.871278858184814,
      "epoch": 4.2519444201695356,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003308489769231811,
      "loss": 5.2918,
      "mean_token_accuracy": 0.20351467579603194,
      "num_tokens": 98572516.0,
      "step": 48655
    },
    {
      "entropy": 5.846841621398926,
      "epoch": 4.252381368522241,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003308187904986099,
      "loss": 5.3617,
      "mean_token_accuracy": 0.20290690362453462,
      "num_tokens": 98583725.0,
      "step": 48660
    },
    {
      "entropy": 5.897746515274048,
      "epoch": 4.252818316874945,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033078860300340496,
      "loss": 5.3522,
      "mean_token_accuracy": 0.20317383110523224,
      "num_tokens": 98595644.0,
      "step": 48665
    },
    {
      "entropy": 5.910842561721802,
      "epoch": 4.25325526522765,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033075841443814543,
      "loss": 5.2753,
      "mean_token_accuracy": 0.21148763746023178,
      "num_tokens": 98605206.0,
      "step": 48670
    },
    {
      "entropy": 5.859741830825806,
      "epoch": 4.253692213580355,
      "grad_norm": 0.984375,
      "learning_rate": 0.00033072822480341023,
      "loss": 5.3242,
      "mean_token_accuracy": 0.1966260403394699,
      "num_tokens": 98616129.0,
      "step": 48675
    },
    {
      "entropy": 5.839339828491211,
      "epoch": 4.25412916193306,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003306980340997786,
      "loss": 5.3833,
      "mean_token_accuracy": 0.1994802787899971,
      "num_tokens": 98626710.0,
      "step": 48680
    },
    {
      "entropy": 5.87006573677063,
      "epoch": 4.254566110285764,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003306678423278294,
      "loss": 5.1984,
      "mean_token_accuracy": 0.21861279606819153,
      "num_tokens": 98636458.0,
      "step": 48685
    },
    {
      "entropy": 5.894209480285644,
      "epoch": 4.255003058638469,
      "grad_norm": 0.984375,
      "learning_rate": 0.00033063764948814187,
      "loss": 5.3472,
      "mean_token_accuracy": 0.20095023214817048,
      "num_tokens": 98647033.0,
      "step": 48690
    },
    {
      "entropy": 5.81710991859436,
      "epoch": 4.255440006991174,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000330607455581295,
      "loss": 5.3349,
      "mean_token_accuracy": 0.20184794217348098,
      "num_tokens": 98657939.0,
      "step": 48695
    },
    {
      "entropy": 5.917481565475464,
      "epoch": 4.255876955343878,
      "grad_norm": 1.25,
      "learning_rate": 0.00033057726060786794,
      "loss": 5.3655,
      "mean_token_accuracy": 0.19937579035758973,
      "num_tokens": 98667453.0,
      "step": 48700
    },
    {
      "entropy": 5.895499610900879,
      "epoch": 4.256313903696583,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033054706456844005,
      "loss": 5.3631,
      "mean_token_accuracy": 0.19479933977127076,
      "num_tokens": 98678951.0,
      "step": 48705
    },
    {
      "entropy": 5.860972929000854,
      "epoch": 4.256750852049287,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033051686746359026,
      "loss": 5.2401,
      "mean_token_accuracy": 0.2055353119969368,
      "num_tokens": 98689427.0,
      "step": 48710
    },
    {
      "entropy": 5.858133363723755,
      "epoch": 4.257187800401993,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003304866692938979,
      "loss": 5.3252,
      "mean_token_accuracy": 0.19898565709590912,
      "num_tokens": 98699671.0,
      "step": 48715
    },
    {
      "entropy": 5.929150533676148,
      "epoch": 4.257624748754697,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033045647005994216,
      "loss": 5.3423,
      "mean_token_accuracy": 0.19821014553308486,
      "num_tokens": 98709357.0,
      "step": 48720
    },
    {
      "entropy": 5.893078756332398,
      "epoch": 4.258061697107402,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003304262697623022,
      "loss": 5.3851,
      "mean_token_accuracy": 0.20158203691244125,
      "num_tokens": 98720144.0,
      "step": 48725
    },
    {
      "entropy": 5.937139844894409,
      "epoch": 4.258498645460106,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033039606840155737,
      "loss": 5.3607,
      "mean_token_accuracy": 0.1951176792383194,
      "num_tokens": 98731125.0,
      "step": 48730
    },
    {
      "entropy": 5.867974805831909,
      "epoch": 4.2589355938128115,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033036586597828696,
      "loss": 5.3242,
      "mean_token_accuracy": 0.2108667254447937,
      "num_tokens": 98741745.0,
      "step": 48735
    },
    {
      "entropy": 5.88498330116272,
      "epoch": 4.259372542165516,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033033566249307024,
      "loss": 5.357,
      "mean_token_accuracy": 0.1995963677763939,
      "num_tokens": 98750970.0,
      "step": 48740
    },
    {
      "entropy": 5.849187517166138,
      "epoch": 4.25980949051822,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003303054579464864,
      "loss": 5.3476,
      "mean_token_accuracy": 0.19668541848659515,
      "num_tokens": 98761473.0,
      "step": 48745
    },
    {
      "entropy": 5.960723400115967,
      "epoch": 4.260246438870926,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033027525233911504,
      "loss": 5.3792,
      "mean_token_accuracy": 0.19977012872695923,
      "num_tokens": 98771265.0,
      "step": 48750
    },
    {
      "entropy": 5.8739485263824465,
      "epoch": 4.26068338722363,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003302450456715353,
      "loss": 5.3253,
      "mean_token_accuracy": 0.20419128239154816,
      "num_tokens": 98781297.0,
      "step": 48755
    },
    {
      "entropy": 5.778141784667969,
      "epoch": 4.261120335576335,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033021483794432667,
      "loss": 5.1973,
      "mean_token_accuracy": 0.21133701205253602,
      "num_tokens": 98790804.0,
      "step": 48760
    },
    {
      "entropy": 5.875114154815674,
      "epoch": 4.261557283929039,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003301846291580685,
      "loss": 5.3623,
      "mean_token_accuracy": 0.1963008537888527,
      "num_tokens": 98800788.0,
      "step": 48765
    },
    {
      "entropy": 5.90458312034607,
      "epoch": 4.261994232281745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003301544193133402,
      "loss": 5.4055,
      "mean_token_accuracy": 0.20340532809495926,
      "num_tokens": 98811627.0,
      "step": 48770
    },
    {
      "entropy": 5.9146112442016605,
      "epoch": 4.262431180634449,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033012420841072135,
      "loss": 5.3729,
      "mean_token_accuracy": 0.19529794305562972,
      "num_tokens": 98822206.0,
      "step": 48775
    },
    {
      "entropy": 5.904829120635986,
      "epoch": 4.2628681289871535,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033009399645079113,
      "loss": 5.3623,
      "mean_token_accuracy": 0.20064504891633989,
      "num_tokens": 98832394.0,
      "step": 48780
    },
    {
      "entropy": 5.880752420425415,
      "epoch": 4.263305077339858,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003300637834341292,
      "loss": 5.2985,
      "mean_token_accuracy": 0.20619406700134277,
      "num_tokens": 98842357.0,
      "step": 48785
    },
    {
      "entropy": 5.904808235168457,
      "epoch": 4.263742025692563,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033003356936131523,
      "loss": 5.2579,
      "mean_token_accuracy": 0.20347899198532104,
      "num_tokens": 98852376.0,
      "step": 48790
    },
    {
      "entropy": 5.856093168258667,
      "epoch": 4.264178974045268,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003300033542329284,
      "loss": 5.355,
      "mean_token_accuracy": 0.1976470962166786,
      "num_tokens": 98862473.0,
      "step": 48795
    },
    {
      "entropy": 5.898614501953125,
      "epoch": 4.264615922397972,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003299731380495484,
      "loss": 5.3749,
      "mean_token_accuracy": 0.1988769516348839,
      "num_tokens": 98872915.0,
      "step": 48800
    },
    {
      "entropy": 5.878763580322266,
      "epoch": 4.265052870750678,
      "grad_norm": 1.046875,
      "learning_rate": 0.000329942920811755,
      "loss": 5.3755,
      "mean_token_accuracy": 0.19511038511991502,
      "num_tokens": 98883863.0,
      "step": 48805
    },
    {
      "entropy": 5.890568780899048,
      "epoch": 4.265489819103382,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003299127025201274,
      "loss": 5.3072,
      "mean_token_accuracy": 0.20140629261732101,
      "num_tokens": 98893507.0,
      "step": 48810
    },
    {
      "entropy": 5.920782899856567,
      "epoch": 4.265926767456087,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032988248317524546,
      "loss": 5.3381,
      "mean_token_accuracy": 0.20151386111974717,
      "num_tokens": 98904109.0,
      "step": 48815
    },
    {
      "entropy": 5.838116550445557,
      "epoch": 4.266363715808791,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003298522627776887,
      "loss": 5.3195,
      "mean_token_accuracy": 0.20057135075330734,
      "num_tokens": 98914017.0,
      "step": 48820
    },
    {
      "entropy": 5.8591595649719235,
      "epoch": 4.266800664161496,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003298220413280368,
      "loss": 5.2803,
      "mean_token_accuracy": 0.20336962789297103,
      "num_tokens": 98924096.0,
      "step": 48825
    },
    {
      "entropy": 5.9071635723114015,
      "epoch": 4.267237612514201,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003297918188268695,
      "loss": 5.3664,
      "mean_token_accuracy": 0.19962265640497207,
      "num_tokens": 98934359.0,
      "step": 48830
    },
    {
      "entropy": 5.912214708328247,
      "epoch": 4.267674560866905,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032976159527476636,
      "loss": 5.3486,
      "mean_token_accuracy": 0.20604225099086762,
      "num_tokens": 98944557.0,
      "step": 48835
    },
    {
      "entropy": 5.923732089996338,
      "epoch": 4.26811150921961,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003297313706723071,
      "loss": 5.3399,
      "mean_token_accuracy": 0.20009952485561372,
      "num_tokens": 98954066.0,
      "step": 48840
    },
    {
      "entropy": 5.794389772415161,
      "epoch": 4.268548457572315,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003297011450200715,
      "loss": 5.2094,
      "mean_token_accuracy": 0.21308225244283677,
      "num_tokens": 98963789.0,
      "step": 48845
    },
    {
      "entropy": 5.830374813079834,
      "epoch": 4.26898540592502,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003296709183186393,
      "loss": 5.2751,
      "mean_token_accuracy": 0.20975541919469834,
      "num_tokens": 98973441.0,
      "step": 48850
    },
    {
      "entropy": 5.8941224098205565,
      "epoch": 4.269422354277724,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003296406905685902,
      "loss": 5.3042,
      "mean_token_accuracy": 0.20482458621263505,
      "num_tokens": 98983362.0,
      "step": 48855
    },
    {
      "entropy": 5.885170316696167,
      "epoch": 4.2698593026304295,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032961046177050403,
      "loss": 5.2892,
      "mean_token_accuracy": 0.2078208237886429,
      "num_tokens": 98993174.0,
      "step": 48860
    },
    {
      "entropy": 5.888829517364502,
      "epoch": 4.270296250983134,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003295802319249606,
      "loss": 5.3406,
      "mean_token_accuracy": 0.20515812039375306,
      "num_tokens": 99002092.0,
      "step": 48865
    },
    {
      "entropy": 5.880951690673828,
      "epoch": 4.270733199335838,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003295500010325397,
      "loss": 5.3624,
      "mean_token_accuracy": 0.19848008304834366,
      "num_tokens": 99012818.0,
      "step": 48870
    },
    {
      "entropy": 5.858562278747558,
      "epoch": 4.271170147688543,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003295197690938212,
      "loss": 5.2771,
      "mean_token_accuracy": 0.20458371937274933,
      "num_tokens": 99022988.0,
      "step": 48875
    },
    {
      "entropy": 5.861973810195923,
      "epoch": 4.271607096041248,
      "grad_norm": 1.046875,
      "learning_rate": 0.000329489536109385,
      "loss": 5.3487,
      "mean_token_accuracy": 0.20248837769031525,
      "num_tokens": 99032884.0,
      "step": 48880
    },
    {
      "entropy": 5.9179604053497314,
      "epoch": 4.272044044393953,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003294593020798109,
      "loss": 5.3436,
      "mean_token_accuracy": 0.19615542888641357,
      "num_tokens": 99042274.0,
      "step": 48885
    },
    {
      "entropy": 5.818048000335693,
      "epoch": 4.272480992746657,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00032942906700567884,
      "loss": 5.2599,
      "mean_token_accuracy": 0.21356312930583954,
      "num_tokens": 99052385.0,
      "step": 48890
    },
    {
      "entropy": 5.803498125076294,
      "epoch": 4.272917941099362,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032939883088756877,
      "loss": 5.2441,
      "mean_token_accuracy": 0.21347789019346236,
      "num_tokens": 99062589.0,
      "step": 48895
    },
    {
      "entropy": 5.921853876113891,
      "epoch": 4.273354889452067,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00032936859372606063,
      "loss": 5.3349,
      "mean_token_accuracy": 0.1943282812833786,
      "num_tokens": 99072794.0,
      "step": 48900
    },
    {
      "entropy": 5.897803831100464,
      "epoch": 4.2737918378047715,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032933835552173443,
      "loss": 5.3414,
      "mean_token_accuracy": 0.1996895983815193,
      "num_tokens": 99083759.0,
      "step": 48905
    },
    {
      "entropy": 5.916192960739136,
      "epoch": 4.274228786157476,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032930811627517,
      "loss": 5.4105,
      "mean_token_accuracy": 0.19999715983867644,
      "num_tokens": 99094056.0,
      "step": 48910
    },
    {
      "entropy": 5.927651262283325,
      "epoch": 4.274665734510181,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003292778759869475,
      "loss": 5.323,
      "mean_token_accuracy": 0.20022533982992172,
      "num_tokens": 99104287.0,
      "step": 48915
    },
    {
      "entropy": 5.863661050796509,
      "epoch": 4.275102682862886,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032924763465764676,
      "loss": 5.2755,
      "mean_token_accuracy": 0.20603124648332596,
      "num_tokens": 99114322.0,
      "step": 48920
    },
    {
      "entropy": 5.837005376815796,
      "epoch": 4.27553963121559,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003292173922878481,
      "loss": 5.2807,
      "mean_token_accuracy": 0.20596246868371965,
      "num_tokens": 99124145.0,
      "step": 48925
    },
    {
      "entropy": 5.743163824081421,
      "epoch": 4.275976579568295,
      "grad_norm": 1.578125,
      "learning_rate": 0.0003291871488781315,
      "loss": 5.1897,
      "mean_token_accuracy": 0.21872081607580185,
      "num_tokens": 99135388.0,
      "step": 48930
    },
    {
      "entropy": 5.794740915298462,
      "epoch": 4.276413527921,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003291569044290769,
      "loss": 5.2801,
      "mean_token_accuracy": 0.20730279237031937,
      "num_tokens": 99145615.0,
      "step": 48935
    },
    {
      "entropy": 5.879789543151856,
      "epoch": 4.2768504762737045,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003291266589412645,
      "loss": 5.381,
      "mean_token_accuracy": 0.19914275407791138,
      "num_tokens": 99155805.0,
      "step": 48940
    },
    {
      "entropy": 5.967088603973389,
      "epoch": 4.277287424626409,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003290964124152744,
      "loss": 5.3626,
      "mean_token_accuracy": 0.19947272688150405,
      "num_tokens": 99165124.0,
      "step": 48945
    },
    {
      "entropy": 5.891361999511719,
      "epoch": 4.2777243729791135,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032906616485168686,
      "loss": 5.3505,
      "mean_token_accuracy": 0.19912199079990386,
      "num_tokens": 99175724.0,
      "step": 48950
    },
    {
      "entropy": 5.841943979263306,
      "epoch": 4.278161321331819,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003290359162510819,
      "loss": 5.3182,
      "mean_token_accuracy": 0.20281362980604173,
      "num_tokens": 99184972.0,
      "step": 48955
    },
    {
      "entropy": 5.883037662506103,
      "epoch": 4.278598269684523,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032900566661403974,
      "loss": 5.3241,
      "mean_token_accuracy": 0.20175267308950423,
      "num_tokens": 99194646.0,
      "step": 48960
    },
    {
      "entropy": 5.796527147293091,
      "epoch": 4.279035218037228,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032897541594114054,
      "loss": 5.2748,
      "mean_token_accuracy": 0.2076807752251625,
      "num_tokens": 99203778.0,
      "step": 48965
    },
    {
      "entropy": 5.870899343490601,
      "epoch": 4.279472166389933,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032894516423296466,
      "loss": 5.3021,
      "mean_token_accuracy": 0.20758675634860993,
      "num_tokens": 99213365.0,
      "step": 48970
    },
    {
      "entropy": 5.9473775863647464,
      "epoch": 4.279909114742638,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032891491149009225,
      "loss": 5.3382,
      "mean_token_accuracy": 0.19951007664203643,
      "num_tokens": 99223967.0,
      "step": 48975
    },
    {
      "entropy": 5.861318969726563,
      "epoch": 4.280346063095342,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003288846577131035,
      "loss": 5.3121,
      "mean_token_accuracy": 0.20584678500890732,
      "num_tokens": 99234115.0,
      "step": 48980
    },
    {
      "entropy": 5.973064851760864,
      "epoch": 4.2807830114480465,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003288544029025789,
      "loss": 5.3744,
      "mean_token_accuracy": 0.20105247050523758,
      "num_tokens": 99244405.0,
      "step": 48985
    },
    {
      "entropy": 5.911421155929565,
      "epoch": 4.281219959800752,
      "grad_norm": 1.0,
      "learning_rate": 0.00032882414705909854,
      "loss": 5.3919,
      "mean_token_accuracy": 0.1954948276281357,
      "num_tokens": 99256042.0,
      "step": 48990
    },
    {
      "entropy": 5.911912059783935,
      "epoch": 4.281656908153456,
      "grad_norm": 1.0,
      "learning_rate": 0.0003287938901832429,
      "loss": 5.3057,
      "mean_token_accuracy": 0.20736243724822997,
      "num_tokens": 99266319.0,
      "step": 48995
    },
    {
      "entropy": 5.916321277618408,
      "epoch": 4.282093856506161,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003287636322755923,
      "loss": 5.2848,
      "mean_token_accuracy": 0.20136554539203644,
      "num_tokens": 99277062.0,
      "step": 49000
    },
    {
      "entropy": 5.840919160842896,
      "epoch": 4.282530804858865,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00032873337333672693,
      "loss": 5.3078,
      "mean_token_accuracy": 0.20138345807790756,
      "num_tokens": 99287130.0,
      "step": 49005
    },
    {
      "entropy": 5.756134462356568,
      "epoch": 4.282967753211571,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003287031133672274,
      "loss": 5.2186,
      "mean_token_accuracy": 0.21784138679504395,
      "num_tokens": 99297721.0,
      "step": 49010
    },
    {
      "entropy": 5.902948522567749,
      "epoch": 4.283404701564275,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032867285236767396,
      "loss": 5.3017,
      "mean_token_accuracy": 0.20407512933015823,
      "num_tokens": 99307743.0,
      "step": 49015
    },
    {
      "entropy": 5.858404874801636,
      "epoch": 4.28384164991698,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003286425903386472,
      "loss": 5.2865,
      "mean_token_accuracy": 0.2057699367403984,
      "num_tokens": 99317520.0,
      "step": 49020
    },
    {
      "entropy": 5.862469053268432,
      "epoch": 4.284278598269685,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003286123272807273,
      "loss": 5.3421,
      "mean_token_accuracy": 0.19716619700193405,
      "num_tokens": 99327768.0,
      "step": 49025
    },
    {
      "entropy": 5.860457944869995,
      "epoch": 4.284715546622389,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032858206319449494,
      "loss": 5.3658,
      "mean_token_accuracy": 0.19141852855682373,
      "num_tokens": 99337742.0,
      "step": 49030
    },
    {
      "entropy": 5.830192375183105,
      "epoch": 4.285152494975094,
      "grad_norm": 1.0,
      "learning_rate": 0.00032855179808053054,
      "loss": 5.299,
      "mean_token_accuracy": 0.20335220992565156,
      "num_tokens": 99347069.0,
      "step": 49035
    },
    {
      "entropy": 5.898511219024658,
      "epoch": 4.285589443327798,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003285215319394146,
      "loss": 5.3366,
      "mean_token_accuracy": 0.20390888899564744,
      "num_tokens": 99357801.0,
      "step": 49040
    },
    {
      "entropy": 5.97201247215271,
      "epoch": 4.286026391680504,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003284912647717276,
      "loss": 5.3997,
      "mean_token_accuracy": 0.19750027358531952,
      "num_tokens": 99368451.0,
      "step": 49045
    },
    {
      "entropy": 5.951112461090088,
      "epoch": 4.286463340033208,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003284609965780501,
      "loss": 5.3558,
      "mean_token_accuracy": 0.20013727396726608,
      "num_tokens": 99378924.0,
      "step": 49050
    },
    {
      "entropy": 5.876741075515747,
      "epoch": 4.286900288385913,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003284307273589627,
      "loss": 5.2953,
      "mean_token_accuracy": 0.20864846855401992,
      "num_tokens": 99387802.0,
      "step": 49055
    },
    {
      "entropy": 5.830845546722412,
      "epoch": 4.287337236738617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032840045711504597,
      "loss": 5.2298,
      "mean_token_accuracy": 0.20853020548820494,
      "num_tokens": 99398172.0,
      "step": 49060
    },
    {
      "entropy": 5.850990724563599,
      "epoch": 4.2877741850913225,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032837018584688045,
      "loss": 5.3286,
      "mean_token_accuracy": 0.20209128111600877,
      "num_tokens": 99408563.0,
      "step": 49065
    },
    {
      "entropy": 5.896893548965454,
      "epoch": 4.288211133444027,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003283399135550469,
      "loss": 5.33,
      "mean_token_accuracy": 0.20309910029172898,
      "num_tokens": 99418542.0,
      "step": 49070
    },
    {
      "entropy": 5.786329030990601,
      "epoch": 4.288648081796731,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003283096402401257,
      "loss": 5.332,
      "mean_token_accuracy": 0.2089870810508728,
      "num_tokens": 99428145.0,
      "step": 49075
    },
    {
      "entropy": 5.861512851715088,
      "epoch": 4.289085030149437,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00032827936590269774,
      "loss": 5.3044,
      "mean_token_accuracy": 0.2004219502210617,
      "num_tokens": 99439460.0,
      "step": 49080
    },
    {
      "entropy": 5.933067178726196,
      "epoch": 4.289521978502141,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003282490905433437,
      "loss": 5.35,
      "mean_token_accuracy": 0.19615016877651215,
      "num_tokens": 99448781.0,
      "step": 49085
    },
    {
      "entropy": 5.830363941192627,
      "epoch": 4.289958926854846,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032821881416264417,
      "loss": 5.2623,
      "mean_token_accuracy": 0.20522981286048889,
      "num_tokens": 99459945.0,
      "step": 49090
    },
    {
      "entropy": 5.849558591842651,
      "epoch": 4.29039587520755,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003281885367611798,
      "loss": 5.2959,
      "mean_token_accuracy": 0.20466530174016953,
      "num_tokens": 99469510.0,
      "step": 49095
    },
    {
      "entropy": 5.855089902877808,
      "epoch": 4.2908328235602555,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003281582583395315,
      "loss": 5.3675,
      "mean_token_accuracy": 0.2038872629404068,
      "num_tokens": 99480409.0,
      "step": 49100
    },
    {
      "entropy": 5.915462779998779,
      "epoch": 4.29126977191296,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032812797889827987,
      "loss": 5.3877,
      "mean_token_accuracy": 0.19499383121728897,
      "num_tokens": 99490044.0,
      "step": 49105
    },
    {
      "entropy": 5.962871503829956,
      "epoch": 4.2917067202656645,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032809769843800585,
      "loss": 5.351,
      "mean_token_accuracy": 0.19948346316814422,
      "num_tokens": 99500257.0,
      "step": 49110
    },
    {
      "entropy": 5.8807713985443115,
      "epoch": 4.292143668618369,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032806741695929015,
      "loss": 5.2993,
      "mean_token_accuracy": 0.2017137497663498,
      "num_tokens": 99509864.0,
      "step": 49115
    },
    {
      "entropy": 5.867462825775147,
      "epoch": 4.292580616971074,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032803713446271363,
      "loss": 5.3013,
      "mean_token_accuracy": 0.20546584874391555,
      "num_tokens": 99520277.0,
      "step": 49120
    },
    {
      "entropy": 5.965690612792969,
      "epoch": 4.293017565323779,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00032800685094885694,
      "loss": 5.4364,
      "mean_token_accuracy": 0.1863650470972061,
      "num_tokens": 99531890.0,
      "step": 49125
    },
    {
      "entropy": 5.832521915435791,
      "epoch": 4.293454513676483,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032797656641830116,
      "loss": 5.1364,
      "mean_token_accuracy": 0.2185782939195633,
      "num_tokens": 99541786.0,
      "step": 49130
    },
    {
      "entropy": 5.891370677947998,
      "epoch": 4.293891462029189,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003279462808716271,
      "loss": 5.2736,
      "mean_token_accuracy": 0.20599671751260756,
      "num_tokens": 99551665.0,
      "step": 49135
    },
    {
      "entropy": 5.834681510925293,
      "epoch": 4.294328410381893,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003279159943094156,
      "loss": 5.3137,
      "mean_token_accuracy": 0.196848002076149,
      "num_tokens": 99561350.0,
      "step": 49140
    },
    {
      "entropy": 5.867819452285767,
      "epoch": 4.2947653587345975,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003278857067322476,
      "loss": 5.3223,
      "mean_token_accuracy": 0.20852793306112288,
      "num_tokens": 99571473.0,
      "step": 49145
    },
    {
      "entropy": 5.962529516220092,
      "epoch": 4.295202307087302,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000327855418140704,
      "loss": 5.3295,
      "mean_token_accuracy": 0.20418438464403152,
      "num_tokens": 99580113.0,
      "step": 49150
    },
    {
      "entropy": 5.877592706680298,
      "epoch": 4.295639255440007,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003278251285353658,
      "loss": 5.3269,
      "mean_token_accuracy": 0.20254520922899247,
      "num_tokens": 99591660.0,
      "step": 49155
    },
    {
      "entropy": 5.833546590805054,
      "epoch": 4.296076203792712,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00032779483791681396,
      "loss": 5.2682,
      "mean_token_accuracy": 0.2053961217403412,
      "num_tokens": 99601470.0,
      "step": 49160
    },
    {
      "entropy": 5.878562641143799,
      "epoch": 4.296513152145416,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032776454628562945,
      "loss": 5.3448,
      "mean_token_accuracy": 0.20314813405275345,
      "num_tokens": 99612327.0,
      "step": 49165
    },
    {
      "entropy": 5.8568580627441404,
      "epoch": 4.296950100498121,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003277342536423933,
      "loss": 5.3061,
      "mean_token_accuracy": 0.20492361336946488,
      "num_tokens": 99621879.0,
      "step": 49170
    },
    {
      "entropy": 5.969712781906128,
      "epoch": 4.297387048850826,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003277039599876866,
      "loss": 5.3294,
      "mean_token_accuracy": 0.19414183497428894,
      "num_tokens": 99632052.0,
      "step": 49175
    },
    {
      "entropy": 6.013597202301026,
      "epoch": 4.297823997203531,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00032767366532209023,
      "loss": 5.4623,
      "mean_token_accuracy": 0.19446911215782164,
      "num_tokens": 99643128.0,
      "step": 49180
    },
    {
      "entropy": 5.825987195968628,
      "epoch": 4.298260945556235,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032764336964618553,
      "loss": 5.2723,
      "mean_token_accuracy": 0.2146165579557419,
      "num_tokens": 99654043.0,
      "step": 49185
    },
    {
      "entropy": 5.893337106704712,
      "epoch": 4.29869789390894,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032761307296055327,
      "loss": 5.3623,
      "mean_token_accuracy": 0.2081752300262451,
      "num_tokens": 99664889.0,
      "step": 49190
    },
    {
      "entropy": 5.944587135314942,
      "epoch": 4.299134842261645,
      "grad_norm": 1.0,
      "learning_rate": 0.00032758277526577474,
      "loss": 5.3945,
      "mean_token_accuracy": 0.1921470880508423,
      "num_tokens": 99675025.0,
      "step": 49195
    },
    {
      "entropy": 5.939509153366089,
      "epoch": 4.299571790614349,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032755247656243095,
      "loss": 5.4482,
      "mean_token_accuracy": 0.19362729787826538,
      "num_tokens": 99685947.0,
      "step": 49200
    },
    {
      "entropy": 5.9058084964752195,
      "epoch": 4.300008738967054,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003275221768511033,
      "loss": 5.3525,
      "mean_token_accuracy": 0.1973232537508011,
      "num_tokens": 99695787.0,
      "step": 49205
    },
    {
      "entropy": 5.869268274307251,
      "epoch": 4.300445687319759,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003274918761323726,
      "loss": 5.2737,
      "mean_token_accuracy": 0.20753171145915986,
      "num_tokens": 99705489.0,
      "step": 49210
    },
    {
      "entropy": 5.959701871871948,
      "epoch": 4.300882635672464,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032746157440682026,
      "loss": 5.3949,
      "mean_token_accuracy": 0.18911103159189224,
      "num_tokens": 99715580.0,
      "step": 49215
    },
    {
      "entropy": 5.88333945274353,
      "epoch": 4.301319584025168,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032743127167502746,
      "loss": 5.3599,
      "mean_token_accuracy": 0.19582410156726837,
      "num_tokens": 99725532.0,
      "step": 49220
    },
    {
      "entropy": 5.825571441650391,
      "epoch": 4.301756532377873,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032740096793757535,
      "loss": 5.2421,
      "mean_token_accuracy": 0.21244098246097565,
      "num_tokens": 99736673.0,
      "step": 49225
    },
    {
      "entropy": 5.92290678024292,
      "epoch": 4.302193480730578,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032737066319504527,
      "loss": 5.3256,
      "mean_token_accuracy": 0.20154813975095748,
      "num_tokens": 99747394.0,
      "step": 49230
    },
    {
      "entropy": 5.906637382507324,
      "epoch": 4.302630429083282,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003273403574480184,
      "loss": 5.3254,
      "mean_token_accuracy": 0.21006056070327758,
      "num_tokens": 99757671.0,
      "step": 49235
    },
    {
      "entropy": 5.861523246765136,
      "epoch": 4.303067377435987,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003273100506970761,
      "loss": 5.3232,
      "mean_token_accuracy": 0.2034192591905594,
      "num_tokens": 99767937.0,
      "step": 49240
    },
    {
      "entropy": 5.925798082351685,
      "epoch": 4.303504325788692,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032727974294279955,
      "loss": 5.389,
      "mean_token_accuracy": 0.20215173065662384,
      "num_tokens": 99778300.0,
      "step": 49245
    },
    {
      "entropy": 5.919865131378174,
      "epoch": 4.303941274141397,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003272494341857701,
      "loss": 5.2874,
      "mean_token_accuracy": 0.20366480946540833,
      "num_tokens": 99787740.0,
      "step": 49250
    },
    {
      "entropy": 5.872884273529053,
      "epoch": 4.304378222494101,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003272191244265693,
      "loss": 5.3626,
      "mean_token_accuracy": 0.1998731553554535,
      "num_tokens": 99797436.0,
      "step": 49255
    },
    {
      "entropy": 5.863197755813599,
      "epoch": 4.304815170846806,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032718881366577827,
      "loss": 5.3064,
      "mean_token_accuracy": 0.21064245253801345,
      "num_tokens": 99807293.0,
      "step": 49260
    },
    {
      "entropy": 5.842482423782348,
      "epoch": 4.305252119199511,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003271585019039783,
      "loss": 5.3377,
      "mean_token_accuracy": 0.2054900348186493,
      "num_tokens": 99816369.0,
      "step": 49265
    },
    {
      "entropy": 5.863571262359619,
      "epoch": 4.3056890675522155,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000327128189141751,
      "loss": 5.2758,
      "mean_token_accuracy": 0.20960303097963334,
      "num_tokens": 99825858.0,
      "step": 49270
    },
    {
      "entropy": 5.876490879058838,
      "epoch": 4.30612601590492,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003270978753796778,
      "loss": 5.4423,
      "mean_token_accuracy": 0.19128439724445342,
      "num_tokens": 99836664.0,
      "step": 49275
    },
    {
      "entropy": 5.924411725997925,
      "epoch": 4.306562964257624,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032706756061834006,
      "loss": 5.3884,
      "mean_token_accuracy": 0.199333755671978,
      "num_tokens": 99846497.0,
      "step": 49280
    },
    {
      "entropy": 5.960000610351562,
      "epoch": 4.30699991261033,
      "grad_norm": 1.125,
      "learning_rate": 0.0003270372448583191,
      "loss": 5.3962,
      "mean_token_accuracy": 0.2015078291296959,
      "num_tokens": 99856039.0,
      "step": 49285
    },
    {
      "entropy": 5.897904300689698,
      "epoch": 4.307436860963034,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032700692810019654,
      "loss": 5.3291,
      "mean_token_accuracy": 0.20176560878753663,
      "num_tokens": 99866541.0,
      "step": 49290
    },
    {
      "entropy": 5.8728940963745115,
      "epoch": 4.307873809315739,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00032697661034455396,
      "loss": 5.292,
      "mean_token_accuracy": 0.2024873286485672,
      "num_tokens": 99877550.0,
      "step": 49295
    },
    {
      "entropy": 5.819962835311889,
      "epoch": 4.308310757668443,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003269462915919726,
      "loss": 5.2333,
      "mean_token_accuracy": 0.2072040006518364,
      "num_tokens": 99887653.0,
      "step": 49300
    },
    {
      "entropy": 5.962124443054199,
      "epoch": 4.3087477060211485,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003269159718430343,
      "loss": 5.3189,
      "mean_token_accuracy": 0.20175447314977646,
      "num_tokens": 99896900.0,
      "step": 49305
    },
    {
      "entropy": 5.892336988449097,
      "epoch": 4.309184654373853,
      "grad_norm": 1.0,
      "learning_rate": 0.0003268856510983204,
      "loss": 5.3518,
      "mean_token_accuracy": 0.20465128719806672,
      "num_tokens": 99907019.0,
      "step": 49310
    },
    {
      "entropy": 5.845349359512329,
      "epoch": 4.3096216027265575,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003268553293584125,
      "loss": 5.324,
      "mean_token_accuracy": 0.20614369362592697,
      "num_tokens": 99917212.0,
      "step": 49315
    },
    {
      "entropy": 5.968334102630616,
      "epoch": 4.310058551079263,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00032682500662389223,
      "loss": 5.416,
      "mean_token_accuracy": 0.19297170341014863,
      "num_tokens": 99927239.0,
      "step": 49320
    },
    {
      "entropy": 5.855612897872925,
      "epoch": 4.310495499431967,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003267946828953411,
      "loss": 5.2896,
      "mean_token_accuracy": 0.2072415307164192,
      "num_tokens": 99937490.0,
      "step": 49325
    },
    {
      "entropy": 5.936912584304809,
      "epoch": 4.310932447784672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003267643581733409,
      "loss": 5.4207,
      "mean_token_accuracy": 0.19597044885158538,
      "num_tokens": 99947083.0,
      "step": 49330
    },
    {
      "entropy": 5.884358549118042,
      "epoch": 4.311369396137376,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003267340324584732,
      "loss": 5.3298,
      "mean_token_accuracy": 0.20311371833086014,
      "num_tokens": 99957842.0,
      "step": 49335
    },
    {
      "entropy": 5.883180856704712,
      "epoch": 4.311806344490082,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003267037057513196,
      "loss": 5.2968,
      "mean_token_accuracy": 0.2065809726715088,
      "num_tokens": 99968384.0,
      "step": 49340
    },
    {
      "entropy": 5.885250997543335,
      "epoch": 4.312243292842786,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003266733780524618,
      "loss": 5.323,
      "mean_token_accuracy": 0.20053360611200333,
      "num_tokens": 99978749.0,
      "step": 49345
    },
    {
      "entropy": 5.8294939517974855,
      "epoch": 4.3126802411954905,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003266430493624816,
      "loss": 5.3227,
      "mean_token_accuracy": 0.20349034965038298,
      "num_tokens": 99988335.0,
      "step": 49350
    },
    {
      "entropy": 5.836377334594727,
      "epoch": 4.313117189548195,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032661271968196067,
      "loss": 5.3233,
      "mean_token_accuracy": 0.20103525072336198,
      "num_tokens": 99999657.0,
      "step": 49355
    },
    {
      "entropy": 5.923674058914185,
      "epoch": 4.3135541379009,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032658238901148075,
      "loss": 5.2661,
      "mean_token_accuracy": 0.20659166425466538,
      "num_tokens": 100009604.0,
      "step": 49360
    },
    {
      "entropy": 5.951284980773925,
      "epoch": 4.313991086253605,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003265520573516235,
      "loss": 5.2844,
      "mean_token_accuracy": 0.20831457823514937,
      "num_tokens": 100019372.0,
      "step": 49365
    },
    {
      "entropy": 5.832951784133911,
      "epoch": 4.314428034606309,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003265217247029708,
      "loss": 5.3105,
      "mean_token_accuracy": 0.2064391106367111,
      "num_tokens": 100029097.0,
      "step": 49370
    },
    {
      "entropy": 5.849371147155762,
      "epoch": 4.314864982959014,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003264913910661044,
      "loss": 5.282,
      "mean_token_accuracy": 0.20513285100460052,
      "num_tokens": 100039506.0,
      "step": 49375
    },
    {
      "entropy": 5.866259527206421,
      "epoch": 4.315301931311719,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032646105644160617,
      "loss": 5.2259,
      "mean_token_accuracy": 0.2117846578359604,
      "num_tokens": 100049721.0,
      "step": 49380
    },
    {
      "entropy": 5.927819681167603,
      "epoch": 4.315738879664424,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003264307208300579,
      "loss": 5.4131,
      "mean_token_accuracy": 0.19301055371761322,
      "num_tokens": 100059329.0,
      "step": 49385
    },
    {
      "entropy": 5.846324920654297,
      "epoch": 4.316175828017128,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003264003842320415,
      "loss": 5.2845,
      "mean_token_accuracy": 0.20460135042667388,
      "num_tokens": 100069237.0,
      "step": 49390
    },
    {
      "entropy": 5.886916875839233,
      "epoch": 4.316612776369833,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032637004664813873,
      "loss": 5.4234,
      "mean_token_accuracy": 0.1907181367278099,
      "num_tokens": 100079475.0,
      "step": 49395
    },
    {
      "entropy": 5.890509986877442,
      "epoch": 4.317049724722538,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003263397080789316,
      "loss": 5.2919,
      "mean_token_accuracy": 0.20409605652093887,
      "num_tokens": 100089135.0,
      "step": 49400
    },
    {
      "entropy": 5.840731000900268,
      "epoch": 4.317486673075242,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032630936852500196,
      "loss": 5.2615,
      "mean_token_accuracy": 0.20653198212385177,
      "num_tokens": 100098465.0,
      "step": 49405
    },
    {
      "entropy": 5.861526393890381,
      "epoch": 4.317923621427947,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003262790279869317,
      "loss": 5.3175,
      "mean_token_accuracy": 0.20273620635271072,
      "num_tokens": 100110057.0,
      "step": 49410
    },
    {
      "entropy": 5.8954308986663815,
      "epoch": 4.318360569780652,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032624868646530285,
      "loss": 5.2696,
      "mean_token_accuracy": 0.2057241439819336,
      "num_tokens": 100119127.0,
      "step": 49415
    },
    {
      "entropy": 5.884216403961181,
      "epoch": 4.318797518133357,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003262183439606974,
      "loss": 5.3226,
      "mean_token_accuracy": 0.20065022110939026,
      "num_tokens": 100129528.0,
      "step": 49420
    },
    {
      "entropy": 5.882183933258057,
      "epoch": 4.319234466486061,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003261880004736972,
      "loss": 5.3149,
      "mean_token_accuracy": 0.20210431069135665,
      "num_tokens": 100139566.0,
      "step": 49425
    },
    {
      "entropy": 5.902682876586914,
      "epoch": 4.319671414838766,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032615765600488435,
      "loss": 5.3817,
      "mean_token_accuracy": 0.20254139900207518,
      "num_tokens": 100150173.0,
      "step": 49430
    },
    {
      "entropy": 5.8333110332489015,
      "epoch": 4.320108363191471,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003261273105548408,
      "loss": 5.246,
      "mean_token_accuracy": 0.2093818947672844,
      "num_tokens": 100160923.0,
      "step": 49435
    },
    {
      "entropy": 5.885039138793945,
      "epoch": 4.320545311544175,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003260969641241487,
      "loss": 5.312,
      "mean_token_accuracy": 0.2025030419230461,
      "num_tokens": 100170682.0,
      "step": 49440
    },
    {
      "entropy": 5.788444566726684,
      "epoch": 4.32098225989688,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032606661671339,
      "loss": 5.2286,
      "mean_token_accuracy": 0.20756722688674928,
      "num_tokens": 100180901.0,
      "step": 49445
    },
    {
      "entropy": 5.85057897567749,
      "epoch": 4.321419208249585,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032603626832314694,
      "loss": 5.3281,
      "mean_token_accuracy": 0.2027977854013443,
      "num_tokens": 100190842.0,
      "step": 49450
    },
    {
      "entropy": 5.882775449752808,
      "epoch": 4.32185615660229,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003260059189540014,
      "loss": 5.2361,
      "mean_token_accuracy": 0.20595088750123977,
      "num_tokens": 100200324.0,
      "step": 49455
    },
    {
      "entropy": 5.793231582641601,
      "epoch": 4.322293104954994,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003259755686065356,
      "loss": 5.1971,
      "mean_token_accuracy": 0.2176896810531616,
      "num_tokens": 100209977.0,
      "step": 49460
    },
    {
      "entropy": 5.812116241455078,
      "epoch": 4.322730053307699,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032594521728133177,
      "loss": 5.2689,
      "mean_token_accuracy": 0.20600373446941375,
      "num_tokens": 100219773.0,
      "step": 49465
    },
    {
      "entropy": 5.885090303421021,
      "epoch": 4.323167001660404,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000325914864978972,
      "loss": 5.298,
      "mean_token_accuracy": 0.20222324579954148,
      "num_tokens": 100229089.0,
      "step": 49470
    },
    {
      "entropy": 5.9421721458435055,
      "epoch": 4.3236039500131085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003258845117000384,
      "loss": 5.4408,
      "mean_token_accuracy": 0.1963436022400856,
      "num_tokens": 100239017.0,
      "step": 49475
    },
    {
      "entropy": 5.8355976104736325,
      "epoch": 4.324040898365813,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003258541574451131,
      "loss": 5.2784,
      "mean_token_accuracy": 0.20206902474164962,
      "num_tokens": 100250056.0,
      "step": 49480
    },
    {
      "entropy": 5.8573507308959964,
      "epoch": 4.324477846718517,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032582380221477853,
      "loss": 5.3019,
      "mean_token_accuracy": 0.20429997891187668,
      "num_tokens": 100260752.0,
      "step": 49485
    },
    {
      "entropy": 5.919102430343628,
      "epoch": 4.324914795071223,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032579344600961667,
      "loss": 5.3453,
      "mean_token_accuracy": 0.20334667563438416,
      "num_tokens": 100270379.0,
      "step": 49490
    },
    {
      "entropy": 5.893435716629028,
      "epoch": 4.325351743423927,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032576308883021013,
      "loss": 5.3223,
      "mean_token_accuracy": 0.20123157650232315,
      "num_tokens": 100280692.0,
      "step": 49495
    },
    {
      "entropy": 5.961850118637085,
      "epoch": 4.325788691776632,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032573273067714077,
      "loss": 5.3109,
      "mean_token_accuracy": 0.19536698311567308,
      "num_tokens": 100290726.0,
      "step": 49500
    },
    {
      "entropy": 5.892636299133301,
      "epoch": 4.326225640129337,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032570237155099107,
      "loss": 5.3593,
      "mean_token_accuracy": 0.20550315529108049,
      "num_tokens": 100301080.0,
      "step": 49505
    },
    {
      "entropy": 5.759251356124878,
      "epoch": 4.3266625884820415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003256720114523433,
      "loss": 5.2247,
      "mean_token_accuracy": 0.2250548854470253,
      "num_tokens": 100311412.0,
      "step": 49510
    },
    {
      "entropy": 5.815550422668457,
      "epoch": 4.327099536834746,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003256416503817799,
      "loss": 5.3034,
      "mean_token_accuracy": 0.2046065852046013,
      "num_tokens": 100321504.0,
      "step": 49515
    },
    {
      "entropy": 5.879633474349975,
      "epoch": 4.3275364851874505,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032561128833988307,
      "loss": 5.3234,
      "mean_token_accuracy": 0.20197924971580505,
      "num_tokens": 100332096.0,
      "step": 49520
    },
    {
      "entropy": 5.813748502731324,
      "epoch": 4.327973433540156,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032558092532723515,
      "loss": 5.2999,
      "mean_token_accuracy": 0.2012634441256523,
      "num_tokens": 100342544.0,
      "step": 49525
    },
    {
      "entropy": 5.881331968307495,
      "epoch": 4.32841038189286,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003255505613444186,
      "loss": 5.2795,
      "mean_token_accuracy": 0.1997422069311142,
      "num_tokens": 100351469.0,
      "step": 49530
    },
    {
      "entropy": 5.851439094543457,
      "epoch": 4.328847330245565,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003255201963920159,
      "loss": 5.3254,
      "mean_token_accuracy": 0.20356838703155516,
      "num_tokens": 100361378.0,
      "step": 49535
    },
    {
      "entropy": 5.8987898349761965,
      "epoch": 4.329284278598269,
      "grad_norm": 1.25,
      "learning_rate": 0.0003254898304706092,
      "loss": 5.3657,
      "mean_token_accuracy": 0.19564323723316193,
      "num_tokens": 100371851.0,
      "step": 49540
    },
    {
      "entropy": 5.841916942596436,
      "epoch": 4.329721226950975,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003254594635807812,
      "loss": 5.2881,
      "mean_token_accuracy": 0.20779860615730286,
      "num_tokens": 100382422.0,
      "step": 49545
    },
    {
      "entropy": 5.871819686889649,
      "epoch": 4.330158175303679,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00032542909572311424,
      "loss": 5.4054,
      "mean_token_accuracy": 0.20487880259752272,
      "num_tokens": 100393181.0,
      "step": 49550
    },
    {
      "entropy": 5.926227045059204,
      "epoch": 4.3305951236563835,
      "grad_norm": 0.984375,
      "learning_rate": 0.00032539872689819075,
      "loss": 5.3993,
      "mean_token_accuracy": 0.19632923305034639,
      "num_tokens": 100403464.0,
      "step": 49555
    },
    {
      "entropy": 5.949645757675171,
      "epoch": 4.331032072009089,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032536835710659335,
      "loss": 5.3682,
      "mean_token_accuracy": 0.19836936742067338,
      "num_tokens": 100413963.0,
      "step": 49560
    },
    {
      "entropy": 5.8804521560668945,
      "epoch": 4.331469020361793,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003253379863489044,
      "loss": 5.2277,
      "mean_token_accuracy": 0.21288405060768129,
      "num_tokens": 100423364.0,
      "step": 49565
    },
    {
      "entropy": 5.791508960723877,
      "epoch": 4.331905968714498,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032530761462570644,
      "loss": 5.2842,
      "mean_token_accuracy": 0.21091659963130951,
      "num_tokens": 100432444.0,
      "step": 49570
    },
    {
      "entropy": 5.8308838367462155,
      "epoch": 4.332342917067202,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032527724193758223,
      "loss": 5.3178,
      "mean_token_accuracy": 0.20449899286031722,
      "num_tokens": 100442840.0,
      "step": 49575
    },
    {
      "entropy": 5.959099197387696,
      "epoch": 4.332779865419908,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032524686828511407,
      "loss": 5.3196,
      "mean_token_accuracy": 0.19952136576175689,
      "num_tokens": 100452958.0,
      "step": 49580
    },
    {
      "entropy": 5.877829885482788,
      "epoch": 4.333216813772612,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032521649366888466,
      "loss": 5.2685,
      "mean_token_accuracy": 0.20787412971258162,
      "num_tokens": 100463532.0,
      "step": 49585
    },
    {
      "entropy": 5.891023588180542,
      "epoch": 4.333653762125317,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032518611808947655,
      "loss": 5.3448,
      "mean_token_accuracy": 0.2046072006225586,
      "num_tokens": 100474161.0,
      "step": 49590
    },
    {
      "entropy": 5.875240850448608,
      "epoch": 4.334090710478021,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032515574154747245,
      "loss": 5.422,
      "mean_token_accuracy": 0.19308695942163467,
      "num_tokens": 100484078.0,
      "step": 49595
    },
    {
      "entropy": 5.926823663711548,
      "epoch": 4.334527658830726,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032512536404345494,
      "loss": 5.3535,
      "mean_token_accuracy": 0.1964891329407692,
      "num_tokens": 100493722.0,
      "step": 49600
    },
    {
      "entropy": 5.89820408821106,
      "epoch": 4.334964607183431,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003250949855780067,
      "loss": 5.3088,
      "mean_token_accuracy": 0.21061260104179383,
      "num_tokens": 100504671.0,
      "step": 49605
    },
    {
      "entropy": 5.894103908538819,
      "epoch": 4.335401555536135,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032506460615171033,
      "loss": 5.2923,
      "mean_token_accuracy": 0.20551344603300095,
      "num_tokens": 100514452.0,
      "step": 49610
    },
    {
      "entropy": 5.793407678604126,
      "epoch": 4.335838503888841,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003250342257651487,
      "loss": 5.2244,
      "mean_token_accuracy": 0.21047049313783645,
      "num_tokens": 100524339.0,
      "step": 49615
    },
    {
      "entropy": 5.795385932922363,
      "epoch": 4.336275452241545,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003250038444189043,
      "loss": 5.2568,
      "mean_token_accuracy": 0.21153099089860916,
      "num_tokens": 100534724.0,
      "step": 49620
    },
    {
      "entropy": 5.860437202453613,
      "epoch": 4.33671240059425,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032497346211356,
      "loss": 5.3164,
      "mean_token_accuracy": 0.20400243401527404,
      "num_tokens": 100545309.0,
      "step": 49625
    },
    {
      "entropy": 5.843005657196045,
      "epoch": 4.337149348946954,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003249430788496984,
      "loss": 5.3262,
      "mean_token_accuracy": 0.2040720835328102,
      "num_tokens": 100555315.0,
      "step": 49630
    },
    {
      "entropy": 5.8440831184387205,
      "epoch": 4.3375862972996595,
      "grad_norm": 0.984375,
      "learning_rate": 0.00032491269462790257,
      "loss": 5.3289,
      "mean_token_accuracy": 0.1979516938328743,
      "num_tokens": 100565624.0,
      "step": 49635
    },
    {
      "entropy": 5.896436977386474,
      "epoch": 4.338023245652364,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032488230944875495,
      "loss": 5.3149,
      "mean_token_accuracy": 0.20385603606700897,
      "num_tokens": 100576430.0,
      "step": 49640
    },
    {
      "entropy": 5.867142581939698,
      "epoch": 4.338460194005068,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003248519233128386,
      "loss": 5.2617,
      "mean_token_accuracy": 0.2084368348121643,
      "num_tokens": 100586537.0,
      "step": 49645
    },
    {
      "entropy": 5.868064928054809,
      "epoch": 4.338897142357773,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003248215362207362,
      "loss": 5.2958,
      "mean_token_accuracy": 0.1997227117419243,
      "num_tokens": 100596694.0,
      "step": 49650
    },
    {
      "entropy": 5.964784479141235,
      "epoch": 4.339334090710478,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003247911481730306,
      "loss": 5.4191,
      "mean_token_accuracy": 0.19792273193597792,
      "num_tokens": 100606711.0,
      "step": 49655
    },
    {
      "entropy": 5.867602443695068,
      "epoch": 4.339771039063183,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032476075917030475,
      "loss": 5.2711,
      "mean_token_accuracy": 0.20254479348659515,
      "num_tokens": 100616778.0,
      "step": 49660
    },
    {
      "entropy": 5.857584476470947,
      "epoch": 4.340207987415887,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032473036921314136,
      "loss": 5.3174,
      "mean_token_accuracy": 0.2038287416100502,
      "num_tokens": 100627009.0,
      "step": 49665
    },
    {
      "entropy": 5.888661479949951,
      "epoch": 4.340644935768593,
      "grad_norm": 1.125,
      "learning_rate": 0.0003246999783021235,
      "loss": 5.3237,
      "mean_token_accuracy": 0.2051175132393837,
      "num_tokens": 100636735.0,
      "step": 49670
    },
    {
      "entropy": 5.88881664276123,
      "epoch": 4.341081884121297,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032466958643783404,
      "loss": 5.3076,
      "mean_token_accuracy": 0.20357435047626496,
      "num_tokens": 100646984.0,
      "step": 49675
    },
    {
      "entropy": 5.863287925720215,
      "epoch": 4.3415188324740015,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032463919362085583,
      "loss": 5.3484,
      "mean_token_accuracy": 0.20234831422567368,
      "num_tokens": 100656854.0,
      "step": 49680
    },
    {
      "entropy": 5.8934684753417965,
      "epoch": 4.341955780826706,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003246087998517719,
      "loss": 5.3366,
      "mean_token_accuracy": 0.19795399606227876,
      "num_tokens": 100666660.0,
      "step": 49685
    },
    {
      "entropy": 5.897421026229859,
      "epoch": 4.342392729179411,
      "grad_norm": 1.0,
      "learning_rate": 0.0003245784051311651,
      "loss": 5.3305,
      "mean_token_accuracy": 0.20321893692016602,
      "num_tokens": 100677451.0,
      "step": 49690
    },
    {
      "entropy": 5.881238746643066,
      "epoch": 4.342829677532116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032454800945961866,
      "loss": 5.3427,
      "mean_token_accuracy": 0.196255461871624,
      "num_tokens": 100687889.0,
      "step": 49695
    },
    {
      "entropy": 5.82961483001709,
      "epoch": 4.34326662588482,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003245176128377154,
      "loss": 5.278,
      "mean_token_accuracy": 0.2043263867497444,
      "num_tokens": 100697395.0,
      "step": 49700
    },
    {
      "entropy": 5.80074987411499,
      "epoch": 4.343703574237525,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003244872152660384,
      "loss": 5.223,
      "mean_token_accuracy": 0.20681294202804565,
      "num_tokens": 100707712.0,
      "step": 49705
    },
    {
      "entropy": 5.899450635910034,
      "epoch": 4.34414052259023,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032445681674517056,
      "loss": 5.327,
      "mean_token_accuracy": 0.202153742313385,
      "num_tokens": 100717979.0,
      "step": 49710
    },
    {
      "entropy": 5.803346443176269,
      "epoch": 4.344577470942935,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032442641727569517,
      "loss": 5.1987,
      "mean_token_accuracy": 0.21364924907684327,
      "num_tokens": 100727046.0,
      "step": 49715
    },
    {
      "entropy": 5.848292303085327,
      "epoch": 4.345014419295639,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003243960168581951,
      "loss": 5.2897,
      "mean_token_accuracy": 0.21259703636169433,
      "num_tokens": 100736645.0,
      "step": 49720
    },
    {
      "entropy": 5.788419103622436,
      "epoch": 4.345451367648344,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032436561549325366,
      "loss": 5.2405,
      "mean_token_accuracy": 0.21121986955404282,
      "num_tokens": 100746523.0,
      "step": 49725
    },
    {
      "entropy": 5.920471096038819,
      "epoch": 4.345888316001049,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003243352131814538,
      "loss": 5.3619,
      "mean_token_accuracy": 0.1986248552799225,
      "num_tokens": 100756122.0,
      "step": 49730
    },
    {
      "entropy": 5.840338945388794,
      "epoch": 4.346325264353753,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032430480992337867,
      "loss": 5.2936,
      "mean_token_accuracy": 0.2075578898191452,
      "num_tokens": 100766665.0,
      "step": 49735
    },
    {
      "entropy": 5.885763645172119,
      "epoch": 4.346762212706458,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003242744057196115,
      "loss": 5.2938,
      "mean_token_accuracy": 0.204337015748024,
      "num_tokens": 100775759.0,
      "step": 49740
    },
    {
      "entropy": 5.79798436164856,
      "epoch": 4.347199161059163,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00032424400057073536,
      "loss": 5.3124,
      "mean_token_accuracy": 0.20579316318035126,
      "num_tokens": 100785988.0,
      "step": 49745
    },
    {
      "entropy": 5.795202684402466,
      "epoch": 4.347636109411868,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003242135944773335,
      "loss": 5.2063,
      "mean_token_accuracy": 0.20818969011306762,
      "num_tokens": 100794352.0,
      "step": 49750
    },
    {
      "entropy": 5.793550109863281,
      "epoch": 4.348073057764572,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032418318743998917,
      "loss": 5.2249,
      "mean_token_accuracy": 0.21171856075525283,
      "num_tokens": 100804146.0,
      "step": 49755
    },
    {
      "entropy": 5.873235750198364,
      "epoch": 4.348510006117277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003241527794592855,
      "loss": 5.3074,
      "mean_token_accuracy": 0.20201647281646729,
      "num_tokens": 100815422.0,
      "step": 49760
    },
    {
      "entropy": 5.940159654617309,
      "epoch": 4.348946954469982,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032412237053580574,
      "loss": 5.4232,
      "mean_token_accuracy": 0.19510657787323,
      "num_tokens": 100826104.0,
      "step": 49765
    },
    {
      "entropy": 5.852405548095703,
      "epoch": 4.349383902822686,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003240919606701331,
      "loss": 5.2286,
      "mean_token_accuracy": 0.21485351920127868,
      "num_tokens": 100835175.0,
      "step": 49770
    },
    {
      "entropy": 5.858095741271972,
      "epoch": 4.349820851175391,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00032406154986285116,
      "loss": 5.301,
      "mean_token_accuracy": 0.21019960343837737,
      "num_tokens": 100845724.0,
      "step": 49775
    },
    {
      "entropy": 5.843391990661621,
      "epoch": 4.350257799528096,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032403113811454277,
      "loss": 5.3145,
      "mean_token_accuracy": 0.20842251777648926,
      "num_tokens": 100855902.0,
      "step": 49780
    },
    {
      "entropy": 5.844819355010986,
      "epoch": 4.350694747880801,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003240007254257915,
      "loss": 5.2851,
      "mean_token_accuracy": 0.2088502198457718,
      "num_tokens": 100866368.0,
      "step": 49785
    },
    {
      "entropy": 5.881498336791992,
      "epoch": 4.351131696233505,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032397031179718066,
      "loss": 5.2871,
      "mean_token_accuracy": 0.2067100927233696,
      "num_tokens": 100875631.0,
      "step": 49790
    },
    {
      "entropy": 5.836337995529175,
      "epoch": 4.35156864458621,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032393989722929355,
      "loss": 5.2719,
      "mean_token_accuracy": 0.21143371015787124,
      "num_tokens": 100885237.0,
      "step": 49795
    },
    {
      "entropy": 5.805927848815918,
      "epoch": 4.352005592938915,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003239094817227137,
      "loss": 5.331,
      "mean_token_accuracy": 0.2056270033121109,
      "num_tokens": 100895097.0,
      "step": 49800
    },
    {
      "entropy": 5.809764623641968,
      "epoch": 4.3524425412916194,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032387906527802426,
      "loss": 5.2421,
      "mean_token_accuracy": 0.20606577545404434,
      "num_tokens": 100904275.0,
      "step": 49805
    },
    {
      "entropy": 5.82324628829956,
      "epoch": 4.352879489644324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003238486478958087,
      "loss": 5.2315,
      "mean_token_accuracy": 0.20961406975984573,
      "num_tokens": 100913950.0,
      "step": 49810
    },
    {
      "entropy": 5.760082769393921,
      "epoch": 4.353316437997028,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003238182295766505,
      "loss": 5.2657,
      "mean_token_accuracy": 0.20854351669549942,
      "num_tokens": 100924446.0,
      "step": 49815
    },
    {
      "entropy": 5.850281524658203,
      "epoch": 4.353753386349734,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00032378781032113313,
      "loss": 5.2196,
      "mean_token_accuracy": 0.20909166634082793,
      "num_tokens": 100933717.0,
      "step": 49820
    },
    {
      "entropy": 5.864133071899414,
      "epoch": 4.354190334702438,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032375739012984004,
      "loss": 5.2579,
      "mean_token_accuracy": 0.20867629200220109,
      "num_tokens": 100943800.0,
      "step": 49825
    },
    {
      "entropy": 5.886387491226197,
      "epoch": 4.354627283055143,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032372696900335454,
      "loss": 5.3105,
      "mean_token_accuracy": 0.19692923873662949,
      "num_tokens": 100954733.0,
      "step": 49830
    },
    {
      "entropy": 5.856098222732544,
      "epoch": 4.355064231407848,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032369654694226025,
      "loss": 5.3653,
      "mean_token_accuracy": 0.1953582540154457,
      "num_tokens": 100964536.0,
      "step": 49835
    },
    {
      "entropy": 5.8725183486938475,
      "epoch": 4.3555011797605525,
      "grad_norm": 1.125,
      "learning_rate": 0.00032366612394714065,
      "loss": 5.3916,
      "mean_token_accuracy": 0.20001544803380966,
      "num_tokens": 100975650.0,
      "step": 49840
    },
    {
      "entropy": 5.9021909713745115,
      "epoch": 4.355938128113257,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032363570001857947,
      "loss": 5.3074,
      "mean_token_accuracy": 0.20171329826116563,
      "num_tokens": 100986439.0,
      "step": 49845
    },
    {
      "entropy": 5.850966501235962,
      "epoch": 4.3563750764659614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032360527515715994,
      "loss": 5.3334,
      "mean_token_accuracy": 0.19648855030536652,
      "num_tokens": 100996593.0,
      "step": 49850
    },
    {
      "entropy": 5.841660451889038,
      "epoch": 4.356812024818667,
      "grad_norm": 1.25,
      "learning_rate": 0.00032357484936346577,
      "loss": 5.3175,
      "mean_token_accuracy": 0.20641055703163147,
      "num_tokens": 101007614.0,
      "step": 49855
    },
    {
      "entropy": 5.9097185134887695,
      "epoch": 4.357248973171371,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032354442263808056,
      "loss": 5.3784,
      "mean_token_accuracy": 0.20197398066520691,
      "num_tokens": 101018234.0,
      "step": 49860
    },
    {
      "entropy": 5.84523572921753,
      "epoch": 4.357685921524076,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003235139949815878,
      "loss": 5.2766,
      "mean_token_accuracy": 0.21185355484485627,
      "num_tokens": 101028001.0,
      "step": 49865
    },
    {
      "entropy": 5.782727766036987,
      "epoch": 4.35812286987678,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003234835663945713,
      "loss": 5.2525,
      "mean_token_accuracy": 0.21068662852048875,
      "num_tokens": 101037857.0,
      "step": 49870
    },
    {
      "entropy": 5.922770261764526,
      "epoch": 4.358559818229486,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003234531368776145,
      "loss": 5.2766,
      "mean_token_accuracy": 0.2047962188720703,
      "num_tokens": 101047224.0,
      "step": 49875
    },
    {
      "entropy": 5.915990591049194,
      "epoch": 4.35899676658219,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032342270643130123,
      "loss": 5.2771,
      "mean_token_accuracy": 0.20556109696626662,
      "num_tokens": 101057124.0,
      "step": 49880
    },
    {
      "entropy": 5.864191293716431,
      "epoch": 4.3594337149348945,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032339227505621504,
      "loss": 5.3265,
      "mean_token_accuracy": 0.20050038546323776,
      "num_tokens": 101066712.0,
      "step": 49885
    },
    {
      "entropy": 5.86084098815918,
      "epoch": 4.3598706632876,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032336184275293974,
      "loss": 5.364,
      "mean_token_accuracy": 0.1966669052839279,
      "num_tokens": 101076787.0,
      "step": 49890
    },
    {
      "entropy": 5.905383443832397,
      "epoch": 4.360307611640304,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032333140952205884,
      "loss": 5.2683,
      "mean_token_accuracy": 0.20683312863111497,
      "num_tokens": 101087525.0,
      "step": 49895
    },
    {
      "entropy": 5.895298337936401,
      "epoch": 4.360744559993009,
      "grad_norm": 1.0,
      "learning_rate": 0.0003233009753641562,
      "loss": 5.3023,
      "mean_token_accuracy": 0.2040489211678505,
      "num_tokens": 101097114.0,
      "step": 49900
    },
    {
      "entropy": 5.786688804626465,
      "epoch": 4.361181508345713,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003232705402798156,
      "loss": 5.309,
      "mean_token_accuracy": 0.20751098692417144,
      "num_tokens": 101106732.0,
      "step": 49905
    },
    {
      "entropy": 5.868329572677612,
      "epoch": 4.361618456698419,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003232401042696207,
      "loss": 5.3208,
      "mean_token_accuracy": 0.20174745470285416,
      "num_tokens": 101116639.0,
      "step": 49910
    },
    {
      "entropy": 5.842535781860351,
      "epoch": 4.362055405051123,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032320966733415536,
      "loss": 5.3486,
      "mean_token_accuracy": 0.20591601282358168,
      "num_tokens": 101126716.0,
      "step": 49915
    },
    {
      "entropy": 5.871636915206909,
      "epoch": 4.362492353403828,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003231792294740033,
      "loss": 5.2441,
      "mean_token_accuracy": 0.2090708166360855,
      "num_tokens": 101137362.0,
      "step": 49920
    },
    {
      "entropy": 5.877208662033081,
      "epoch": 4.362929301756532,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032314879068974837,
      "loss": 5.3313,
      "mean_token_accuracy": 0.20137323588132858,
      "num_tokens": 101147437.0,
      "step": 49925
    },
    {
      "entropy": 5.914654684066773,
      "epoch": 4.363366250109237,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003231183509819744,
      "loss": 5.3868,
      "mean_token_accuracy": 0.20147329568862915,
      "num_tokens": 101157664.0,
      "step": 49930
    },
    {
      "entropy": 5.899207592010498,
      "epoch": 4.363803198461942,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003230879103512653,
      "loss": 5.3503,
      "mean_token_accuracy": 0.19809048026800155,
      "num_tokens": 101168996.0,
      "step": 49935
    },
    {
      "entropy": 5.8588286399841305,
      "epoch": 4.364240146814646,
      "grad_norm": 1.0,
      "learning_rate": 0.00032305746879820487,
      "loss": 5.3483,
      "mean_token_accuracy": 0.19640261232852935,
      "num_tokens": 101179089.0,
      "step": 49940
    },
    {
      "entropy": 5.921067380905152,
      "epoch": 4.364677095167351,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000323027026323377,
      "loss": 5.3642,
      "mean_token_accuracy": 0.20145345330238343,
      "num_tokens": 101189874.0,
      "step": 49945
    },
    {
      "entropy": 5.844922018051148,
      "epoch": 4.365114043520056,
      "grad_norm": 1.25,
      "learning_rate": 0.00032299658292736554,
      "loss": 5.2986,
      "mean_token_accuracy": 0.20120203346014023,
      "num_tokens": 101199715.0,
      "step": 49950
    },
    {
      "entropy": 5.85656213760376,
      "epoch": 4.365550991872761,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032296613861075454,
      "loss": 5.3358,
      "mean_token_accuracy": 0.19942925721406937,
      "num_tokens": 101209996.0,
      "step": 49955
    },
    {
      "entropy": 5.8996094226837155,
      "epoch": 4.365987940225465,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003229356933741278,
      "loss": 5.3392,
      "mean_token_accuracy": 0.20259799361228942,
      "num_tokens": 101220403.0,
      "step": 49960
    },
    {
      "entropy": 5.859089612960815,
      "epoch": 4.3664248885781705,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032290524721806943,
      "loss": 5.281,
      "mean_token_accuracy": 0.20646235048770906,
      "num_tokens": 101230490.0,
      "step": 49965
    },
    {
      "entropy": 5.875709915161133,
      "epoch": 4.366861836930875,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003228748001431633,
      "loss": 5.3082,
      "mean_token_accuracy": 0.2015449121594429,
      "num_tokens": 101240817.0,
      "step": 49970
    },
    {
      "entropy": 5.885822343826294,
      "epoch": 4.367298785283579,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003228443521499934,
      "loss": 5.3914,
      "mean_token_accuracy": 0.20158905237913133,
      "num_tokens": 101251893.0,
      "step": 49975
    },
    {
      "entropy": 5.926036739349366,
      "epoch": 4.367735733636284,
      "grad_norm": 1.125,
      "learning_rate": 0.00032281390323914366,
      "loss": 5.3144,
      "mean_token_accuracy": 0.20015927255153657,
      "num_tokens": 101262241.0,
      "step": 49980
    },
    {
      "entropy": 5.808478689193725,
      "epoch": 4.368172681988989,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003227834534111983,
      "loss": 5.3314,
      "mean_token_accuracy": 0.20356001555919648,
      "num_tokens": 101272285.0,
      "step": 49985
    },
    {
      "entropy": 5.894637393951416,
      "epoch": 4.368609630341694,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003227530026667413,
      "loss": 5.4134,
      "mean_token_accuracy": 0.1959490269422531,
      "num_tokens": 101283026.0,
      "step": 49990
    },
    {
      "entropy": 5.923023462295532,
      "epoch": 4.369046578694398,
      "grad_norm": 1.125,
      "learning_rate": 0.0003227225510063566,
      "loss": 5.3622,
      "mean_token_accuracy": 0.20670658648014067,
      "num_tokens": 101293376.0,
      "step": 49995
    },
    {
      "entropy": 5.9133992195129395,
      "epoch": 4.369483527047103,
      "grad_norm": 1.125,
      "learning_rate": 0.0003226920984306284,
      "loss": 5.3217,
      "mean_token_accuracy": 0.206332990527153,
      "num_tokens": 101305232.0,
      "step": 50000
    },
    {
      "entropy": 5.897661256790161,
      "epoch": 4.369920475399808,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032266164494014075,
      "loss": 5.3653,
      "mean_token_accuracy": 0.20663801878690718,
      "num_tokens": 101314942.0,
      "step": 50005
    },
    {
      "entropy": 5.879583787918091,
      "epoch": 4.3703574237525125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032263119053547784,
      "loss": 5.2717,
      "mean_token_accuracy": 0.20935308784246445,
      "num_tokens": 101323655.0,
      "step": 50010
    },
    {
      "entropy": 5.829888534545899,
      "epoch": 4.370794372105217,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003226007352172237,
      "loss": 5.1965,
      "mean_token_accuracy": 0.20816689729690552,
      "num_tokens": 101333594.0,
      "step": 50015
    },
    {
      "entropy": 5.837028360366821,
      "epoch": 4.371231320457922,
      "grad_norm": 1.0,
      "learning_rate": 0.0003225702789859625,
      "loss": 5.3072,
      "mean_token_accuracy": 0.20369311571121215,
      "num_tokens": 101344580.0,
      "step": 50020
    },
    {
      "entropy": 5.796280860900879,
      "epoch": 4.371668268810627,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032253982184227836,
      "loss": 5.2587,
      "mean_token_accuracy": 0.20865078419446945,
      "num_tokens": 101354574.0,
      "step": 50025
    },
    {
      "entropy": 5.844277191162109,
      "epoch": 4.372105217163331,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003225093637867556,
      "loss": 5.2572,
      "mean_token_accuracy": 0.20851811170578002,
      "num_tokens": 101365150.0,
      "step": 50030
    },
    {
      "entropy": 5.901932573318481,
      "epoch": 4.372542165516036,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003224789048199784,
      "loss": 5.3161,
      "mean_token_accuracy": 0.20590591728687285,
      "num_tokens": 101375726.0,
      "step": 50035
    },
    {
      "entropy": 5.912602233886719,
      "epoch": 4.372979113868741,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003224484449425308,
      "loss": 5.3809,
      "mean_token_accuracy": 0.19646709114313127,
      "num_tokens": 101385457.0,
      "step": 50040
    },
    {
      "entropy": 5.9006750106811525,
      "epoch": 4.3734160622214455,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003224179841549972,
      "loss": 5.361,
      "mean_token_accuracy": 0.19639085084199906,
      "num_tokens": 101395513.0,
      "step": 50045
    },
    {
      "entropy": 5.862729501724243,
      "epoch": 4.37385301057415,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032238752245796186,
      "loss": 5.2729,
      "mean_token_accuracy": 0.20392411798238755,
      "num_tokens": 101405886.0,
      "step": 50050
    },
    {
      "entropy": 5.8412374496459964,
      "epoch": 4.3742899589268545,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032235705985200904,
      "loss": 5.3248,
      "mean_token_accuracy": 0.20702305287122727,
      "num_tokens": 101416130.0,
      "step": 50055
    },
    {
      "entropy": 5.826675844192505,
      "epoch": 4.37472690727956,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003223265963377229,
      "loss": 5.2811,
      "mean_token_accuracy": 0.2154034420847893,
      "num_tokens": 101426817.0,
      "step": 50060
    },
    {
      "entropy": 5.904203605651856,
      "epoch": 4.375163855632264,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003222961319156879,
      "loss": 5.3259,
      "mean_token_accuracy": 0.2042616829276085,
      "num_tokens": 101436873.0,
      "step": 50065
    },
    {
      "entropy": 5.9485697746276855,
      "epoch": 4.375600803984969,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003222656665864884,
      "loss": 5.3457,
      "mean_token_accuracy": 0.20491712540388107,
      "num_tokens": 101447048.0,
      "step": 50070
    },
    {
      "entropy": 5.933070802688599,
      "epoch": 4.376037752337673,
      "grad_norm": 1.0,
      "learning_rate": 0.00032223520035070844,
      "loss": 5.3501,
      "mean_token_accuracy": 0.20048899352550506,
      "num_tokens": 101457536.0,
      "step": 50075
    },
    {
      "entropy": 5.86342978477478,
      "epoch": 4.376474700690379,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003222047332089328,
      "loss": 5.3146,
      "mean_token_accuracy": 0.20446366518735887,
      "num_tokens": 101467833.0,
      "step": 50080
    },
    {
      "entropy": 5.875989961624145,
      "epoch": 4.376911649043083,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032217426516174553,
      "loss": 5.3632,
      "mean_token_accuracy": 0.2024216130375862,
      "num_tokens": 101477960.0,
      "step": 50085
    },
    {
      "entropy": 5.929595470428467,
      "epoch": 4.3773485973957875,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032214379620973117,
      "loss": 5.3004,
      "mean_token_accuracy": 0.20665842890739441,
      "num_tokens": 101487192.0,
      "step": 50090
    },
    {
      "entropy": 5.8698381900787355,
      "epoch": 4.377785545748493,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032211332635347404,
      "loss": 5.2599,
      "mean_token_accuracy": 0.20919015109539033,
      "num_tokens": 101496835.0,
      "step": 50095
    },
    {
      "entropy": 5.815890979766846,
      "epoch": 4.378222494101197,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032208285559355863,
      "loss": 5.271,
      "mean_token_accuracy": 0.20388403683900833,
      "num_tokens": 101506584.0,
      "step": 50100
    },
    {
      "entropy": 5.902025032043457,
      "epoch": 4.378659442453902,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00032205238393056944,
      "loss": 5.3863,
      "mean_token_accuracy": 0.19725191295146943,
      "num_tokens": 101517428.0,
      "step": 50105
    },
    {
      "entropy": 5.8900251388549805,
      "epoch": 4.379096390806606,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032202191136509084,
      "loss": 5.3037,
      "mean_token_accuracy": 0.20488980263471604,
      "num_tokens": 101527594.0,
      "step": 50110
    },
    {
      "entropy": 5.897842693328857,
      "epoch": 4.379533339159312,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003219914378977073,
      "loss": 5.3237,
      "mean_token_accuracy": 0.2065684527158737,
      "num_tokens": 101538381.0,
      "step": 50115
    },
    {
      "entropy": 5.845895099639892,
      "epoch": 4.379970287512016,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032196096352900337,
      "loss": 5.2984,
      "mean_token_accuracy": 0.20941306948661803,
      "num_tokens": 101548110.0,
      "step": 50120
    },
    {
      "entropy": 5.913724565505982,
      "epoch": 4.380407235864721,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032193048825956353,
      "loss": 5.2645,
      "mean_token_accuracy": 0.2092040479183197,
      "num_tokens": 101556966.0,
      "step": 50125
    },
    {
      "entropy": 5.82501335144043,
      "epoch": 4.380844184217425,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032190001208997254,
      "loss": 5.2111,
      "mean_token_accuracy": 0.20923835635185242,
      "num_tokens": 101567310.0,
      "step": 50130
    },
    {
      "entropy": 5.8587963581085205,
      "epoch": 4.38128113257013,
      "grad_norm": 0.984375,
      "learning_rate": 0.00032186953502081456,
      "loss": 5.3352,
      "mean_token_accuracy": 0.1997484266757965,
      "num_tokens": 101577546.0,
      "step": 50135
    },
    {
      "entropy": 5.843367242813111,
      "epoch": 4.381718080922835,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032183905705267436,
      "loss": 5.3009,
      "mean_token_accuracy": 0.20396311283111573,
      "num_tokens": 101587856.0,
      "step": 50140
    },
    {
      "entropy": 5.872975778579712,
      "epoch": 4.382155029275539,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003218085781861365,
      "loss": 5.3374,
      "mean_token_accuracy": 0.19768889248371124,
      "num_tokens": 101597587.0,
      "step": 50145
    },
    {
      "entropy": 5.88378415107727,
      "epoch": 4.382591977628245,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003217780984217856,
      "loss": 5.3186,
      "mean_token_accuracy": 0.21481559574604034,
      "num_tokens": 101608239.0,
      "step": 50150
    },
    {
      "entropy": 5.854643440246582,
      "epoch": 4.383028925980949,
      "grad_norm": 1.0,
      "learning_rate": 0.0003217476177602062,
      "loss": 5.2741,
      "mean_token_accuracy": 0.2103445053100586,
      "num_tokens": 101618400.0,
      "step": 50155
    },
    {
      "entropy": 5.8245399475097654,
      "epoch": 4.383465874333654,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032171713620198313,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21237465888261794,
      "num_tokens": 101628434.0,
      "step": 50160
    },
    {
      "entropy": 5.8145304203033445,
      "epoch": 4.383902822686358,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003216866537477008,
      "loss": 5.2395,
      "mean_token_accuracy": 0.21035337448120117,
      "num_tokens": 101638209.0,
      "step": 50165
    },
    {
      "entropy": 5.946779489517212,
      "epoch": 4.3843397710390635,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000321656170397944,
      "loss": 5.4551,
      "mean_token_accuracy": 0.1939839765429497,
      "num_tokens": 101649162.0,
      "step": 50170
    },
    {
      "entropy": 5.941127634048462,
      "epoch": 4.384776719391768,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003216256861532974,
      "loss": 5.3198,
      "mean_token_accuracy": 0.20415695905685424,
      "num_tokens": 101658890.0,
      "step": 50175
    },
    {
      "entropy": 5.851134824752807,
      "epoch": 4.385213667744472,
      "grad_norm": 0.96875,
      "learning_rate": 0.00032159520101434567,
      "loss": 5.2543,
      "mean_token_accuracy": 0.21150397658348083,
      "num_tokens": 101670160.0,
      "step": 50180
    },
    {
      "entropy": 5.84287657737732,
      "epoch": 4.385650616097177,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003215647149816737,
      "loss": 5.2929,
      "mean_token_accuracy": 0.20340438783168793,
      "num_tokens": 101680025.0,
      "step": 50185
    },
    {
      "entropy": 5.92296953201294,
      "epoch": 4.386087564449882,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000321534228055866,
      "loss": 5.3655,
      "mean_token_accuracy": 0.19862531125545502,
      "num_tokens": 101690662.0,
      "step": 50190
    },
    {
      "entropy": 5.83783450126648,
      "epoch": 4.386524512802587,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032150374023750747,
      "loss": 5.2771,
      "mean_token_accuracy": 0.20253302305936813,
      "num_tokens": 101701605.0,
      "step": 50195
    },
    {
      "entropy": 5.859983825683594,
      "epoch": 4.386961461155291,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003214732515271828,
      "loss": 5.2906,
      "mean_token_accuracy": 0.20257220715284346,
      "num_tokens": 101711442.0,
      "step": 50200
    },
    {
      "entropy": 5.910368585586548,
      "epoch": 4.3873984095079965,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032144276192547683,
      "loss": 5.2862,
      "mean_token_accuracy": 0.21209139078855516,
      "num_tokens": 101721718.0,
      "step": 50205
    },
    {
      "entropy": 5.858285188674927,
      "epoch": 4.387835357860701,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032141227143297433,
      "loss": 5.3044,
      "mean_token_accuracy": 0.203427517414093,
      "num_tokens": 101731603.0,
      "step": 50210
    },
    {
      "entropy": 5.886444902420044,
      "epoch": 4.3882723062134055,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003213817800502601,
      "loss": 5.3108,
      "mean_token_accuracy": 0.20481548756361007,
      "num_tokens": 101742369.0,
      "step": 50215
    },
    {
      "entropy": 5.866548156738281,
      "epoch": 4.38870925456611,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003213512877779191,
      "loss": 5.3064,
      "mean_token_accuracy": 0.20048537999391555,
      "num_tokens": 101752893.0,
      "step": 50220
    },
    {
      "entropy": 5.815057611465454,
      "epoch": 4.389146202918815,
      "grad_norm": 1.0,
      "learning_rate": 0.0003213207946165361,
      "loss": 5.2525,
      "mean_token_accuracy": 0.2049351528286934,
      "num_tokens": 101763090.0,
      "step": 50225
    },
    {
      "entropy": 5.860318326950074,
      "epoch": 4.38958315127152,
      "grad_norm": 1.046875,
      "learning_rate": 0.000321290300566696,
      "loss": 5.2801,
      "mean_token_accuracy": 0.20306717306375505,
      "num_tokens": 101772052.0,
      "step": 50230
    },
    {
      "entropy": 5.852809190750122,
      "epoch": 4.390020099624224,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003212598056289836,
      "loss": 5.2442,
      "mean_token_accuracy": 0.21294985860586166,
      "num_tokens": 101782758.0,
      "step": 50235
    },
    {
      "entropy": 5.833488273620605,
      "epoch": 4.390457047976929,
      "grad_norm": 1.046875,
      "learning_rate": 0.000321229309803984,
      "loss": 5.2662,
      "mean_token_accuracy": 0.21364711672067643,
      "num_tokens": 101791681.0,
      "step": 50240
    },
    {
      "entropy": 5.875670003890991,
      "epoch": 4.390893996329634,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00032119881309228195,
      "loss": 5.3469,
      "mean_token_accuracy": 0.2038703128695488,
      "num_tokens": 101803446.0,
      "step": 50245
    },
    {
      "entropy": 5.878684377670288,
      "epoch": 4.3913309446823385,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032116831549446246,
      "loss": 5.3545,
      "mean_token_accuracy": 0.20191646814346315,
      "num_tokens": 101813907.0,
      "step": 50250
    },
    {
      "entropy": 5.776090097427368,
      "epoch": 4.391767893035043,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003211378170111105,
      "loss": 5.2581,
      "mean_token_accuracy": 0.20893080830574035,
      "num_tokens": 101823715.0,
      "step": 50255
    },
    {
      "entropy": 5.825240230560302,
      "epoch": 4.392204841387748,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032110731764281103,
      "loss": 5.322,
      "mean_token_accuracy": 0.2043122813105583,
      "num_tokens": 101834208.0,
      "step": 50260
    },
    {
      "entropy": 5.858873128890991,
      "epoch": 4.392641789740453,
      "grad_norm": 1.0,
      "learning_rate": 0.00032107681739014906,
      "loss": 5.2151,
      "mean_token_accuracy": 0.21682694554328918,
      "num_tokens": 101845354.0,
      "step": 50265
    },
    {
      "entropy": 5.930011129379272,
      "epoch": 4.393078738093157,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003210463162537096,
      "loss": 5.3462,
      "mean_token_accuracy": 0.20080042481422425,
      "num_tokens": 101855818.0,
      "step": 50270
    },
    {
      "entropy": 5.867638158798218,
      "epoch": 4.393515686445862,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032101581423407764,
      "loss": 5.332,
      "mean_token_accuracy": 0.20689690709114075,
      "num_tokens": 101866012.0,
      "step": 50275
    },
    {
      "entropy": 5.90093355178833,
      "epoch": 4.393952634798567,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032098531133183827,
      "loss": 5.2858,
      "mean_token_accuracy": 0.2045389950275421,
      "num_tokens": 101875834.0,
      "step": 50280
    },
    {
      "entropy": 5.898188924789428,
      "epoch": 4.394389583151272,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003209548075475765,
      "loss": 5.3238,
      "mean_token_accuracy": 0.20390056669712067,
      "num_tokens": 101885261.0,
      "step": 50285
    },
    {
      "entropy": 5.831753587722778,
      "epoch": 4.394826531503976,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032092430288187745,
      "loss": 5.2864,
      "mean_token_accuracy": 0.20680323839187623,
      "num_tokens": 101896290.0,
      "step": 50290
    },
    {
      "entropy": 5.913733196258545,
      "epoch": 4.3952634798566805,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032089379733532623,
      "loss": 5.3165,
      "mean_token_accuracy": 0.20724526196718215,
      "num_tokens": 101906830.0,
      "step": 50295
    },
    {
      "entropy": 5.890042018890381,
      "epoch": 4.395700428209386,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003208632909085079,
      "loss": 5.2867,
      "mean_token_accuracy": 0.20604747980833055,
      "num_tokens": 101917146.0,
      "step": 50300
    },
    {
      "entropy": 5.874347782135009,
      "epoch": 4.39613737656209,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003208327836020076,
      "loss": 5.3287,
      "mean_token_accuracy": 0.2027049258351326,
      "num_tokens": 101927326.0,
      "step": 50305
    },
    {
      "entropy": 5.875018119812012,
      "epoch": 4.396574324914795,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003208022754164106,
      "loss": 5.3981,
      "mean_token_accuracy": 0.1928459882736206,
      "num_tokens": 101938333.0,
      "step": 50310
    },
    {
      "entropy": 5.864415121078491,
      "epoch": 4.3970112732675,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003207717663523019,
      "loss": 5.2634,
      "mean_token_accuracy": 0.21246788203716277,
      "num_tokens": 101948847.0,
      "step": 50315
    },
    {
      "entropy": 5.895094156265259,
      "epoch": 4.397448221620205,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032074125641026676,
      "loss": 5.3571,
      "mean_token_accuracy": 0.1963951840996742,
      "num_tokens": 101959887.0,
      "step": 50320
    },
    {
      "entropy": 5.939965486526489,
      "epoch": 4.397885169972909,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003207107455908902,
      "loss": 5.2903,
      "mean_token_accuracy": 0.2053111881017685,
      "num_tokens": 101969336.0,
      "step": 50325
    },
    {
      "entropy": 5.862982988357544,
      "epoch": 4.398322118325614,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003206802338947578,
      "loss": 5.3655,
      "mean_token_accuracy": 0.20173436999320984,
      "num_tokens": 101979884.0,
      "step": 50330
    },
    {
      "entropy": 5.8953125953674315,
      "epoch": 4.398759066678319,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003206497213224544,
      "loss": 5.3655,
      "mean_token_accuracy": 0.2016034573316574,
      "num_tokens": 101989803.0,
      "step": 50335
    },
    {
      "entropy": 5.87347264289856,
      "epoch": 4.399196015031023,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032061920787456556,
      "loss": 5.3098,
      "mean_token_accuracy": 0.20395976006984712,
      "num_tokens": 101999519.0,
      "step": 50340
    },
    {
      "entropy": 5.874596357345581,
      "epoch": 4.399632963383728,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032058869355167634,
      "loss": 5.3005,
      "mean_token_accuracy": 0.20543740689754486,
      "num_tokens": 102009017.0,
      "step": 50345
    },
    {
      "entropy": 5.877021074295044,
      "epoch": 4.400069911736432,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003205581783543721,
      "loss": 5.3267,
      "mean_token_accuracy": 0.19878671020269395,
      "num_tokens": 102019357.0,
      "step": 50350
    },
    {
      "entropy": 5.876166772842407,
      "epoch": 4.400506860089138,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032052766228323815,
      "loss": 5.3015,
      "mean_token_accuracy": 0.20205287933349608,
      "num_tokens": 102029994.0,
      "step": 50355
    },
    {
      "entropy": 5.85344295501709,
      "epoch": 4.400943808441842,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032049714533885975,
      "loss": 5.2306,
      "mean_token_accuracy": 0.20877285301685333,
      "num_tokens": 102039314.0,
      "step": 50360
    },
    {
      "entropy": 5.886787605285645,
      "epoch": 4.401380756794547,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003204666275218223,
      "loss": 5.3492,
      "mean_token_accuracy": 0.19483829736709596,
      "num_tokens": 102049449.0,
      "step": 50365
    },
    {
      "entropy": 5.897764253616333,
      "epoch": 4.401817705147252,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003204361088327111,
      "loss": 5.365,
      "mean_token_accuracy": 0.2001473620533943,
      "num_tokens": 102059928.0,
      "step": 50370
    },
    {
      "entropy": 5.936738109588623,
      "epoch": 4.4022546534999565,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032040558927211147,
      "loss": 5.3318,
      "mean_token_accuracy": 0.204764661192894,
      "num_tokens": 102070065.0,
      "step": 50375
    },
    {
      "entropy": 5.942338466644287,
      "epoch": 4.402691601852661,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003203750688406089,
      "loss": 5.3389,
      "mean_token_accuracy": 0.19906821697950364,
      "num_tokens": 102080990.0,
      "step": 50380
    },
    {
      "entropy": 5.79874587059021,
      "epoch": 4.403128550205365,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032034454753878874,
      "loss": 5.2458,
      "mean_token_accuracy": 0.21493018716573714,
      "num_tokens": 102090237.0,
      "step": 50385
    },
    {
      "entropy": 5.824437141418457,
      "epoch": 4.403565498558071,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032031402536723634,
      "loss": 5.3218,
      "mean_token_accuracy": 0.20752741396427155,
      "num_tokens": 102100685.0,
      "step": 50390
    },
    {
      "entropy": 5.872645616531372,
      "epoch": 4.404002446910775,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003202835023265372,
      "loss": 5.3014,
      "mean_token_accuracy": 0.19973189681768416,
      "num_tokens": 102110218.0,
      "step": 50395
    },
    {
      "entropy": 5.939526033401489,
      "epoch": 4.40443939526348,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003202529784172768,
      "loss": 5.3859,
      "mean_token_accuracy": 0.19554729610681534,
      "num_tokens": 102121009.0,
      "step": 50400
    },
    {
      "entropy": 5.924735403060913,
      "epoch": 4.404876343616184,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032022245364004057,
      "loss": 5.2744,
      "mean_token_accuracy": 0.20450783520936966,
      "num_tokens": 102130840.0,
      "step": 50405
    },
    {
      "entropy": 5.848718976974487,
      "epoch": 4.4053132919688895,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032019192799541396,
      "loss": 5.3142,
      "mean_token_accuracy": 0.2075204148888588,
      "num_tokens": 102141289.0,
      "step": 50410
    },
    {
      "entropy": 5.797132968902588,
      "epoch": 4.405750240321594,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00032016140148398247,
      "loss": 5.263,
      "mean_token_accuracy": 0.21165309101343155,
      "num_tokens": 102150945.0,
      "step": 50415
    },
    {
      "entropy": 5.893374300003051,
      "epoch": 4.4061871886742985,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003201308741063317,
      "loss": 5.3731,
      "mean_token_accuracy": 0.19811987727880478,
      "num_tokens": 102161251.0,
      "step": 50420
    },
    {
      "entropy": 5.909700679779053,
      "epoch": 4.406624137027004,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000320100345863047,
      "loss": 5.3254,
      "mean_token_accuracy": 0.20100636631250382,
      "num_tokens": 102170848.0,
      "step": 50425
    },
    {
      "entropy": 5.925185585021973,
      "epoch": 4.407061085379708,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003200698167547141,
      "loss": 5.3562,
      "mean_token_accuracy": 0.2059213176369667,
      "num_tokens": 102181533.0,
      "step": 50430
    },
    {
      "entropy": 5.849519681930542,
      "epoch": 4.407498033732413,
      "grad_norm": 1.0,
      "learning_rate": 0.00032003928678191857,
      "loss": 5.2547,
      "mean_token_accuracy": 0.20621581971645356,
      "num_tokens": 102191868.0,
      "step": 50435
    },
    {
      "entropy": 5.837011528015137,
      "epoch": 4.407934982085117,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032000875594524584,
      "loss": 5.2772,
      "mean_token_accuracy": 0.2046761229634285,
      "num_tokens": 102201164.0,
      "step": 50440
    },
    {
      "entropy": 5.85961217880249,
      "epoch": 4.408371930437823,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031997822424528164,
      "loss": 5.3008,
      "mean_token_accuracy": 0.2109527349472046,
      "num_tokens": 102211309.0,
      "step": 50445
    },
    {
      "entropy": 5.945911073684693,
      "epoch": 4.408808878790527,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003199476916826115,
      "loss": 5.3826,
      "mean_token_accuracy": 0.19124091416597366,
      "num_tokens": 102221721.0,
      "step": 50450
    },
    {
      "entropy": 5.880290269851685,
      "epoch": 4.4092458271432315,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031991715825782104,
      "loss": 5.3359,
      "mean_token_accuracy": 0.19688940942287445,
      "num_tokens": 102231880.0,
      "step": 50455
    },
    {
      "entropy": 5.910099220275879,
      "epoch": 4.409682775495936,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031988662397149597,
      "loss": 5.313,
      "mean_token_accuracy": 0.20373724550008773,
      "num_tokens": 102243027.0,
      "step": 50460
    },
    {
      "entropy": 5.903230667114258,
      "epoch": 4.410119723848641,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003198560888242219,
      "loss": 5.274,
      "mean_token_accuracy": 0.2088819310069084,
      "num_tokens": 102253117.0,
      "step": 50465
    },
    {
      "entropy": 5.85102276802063,
      "epoch": 4.410556672201346,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003198255528165846,
      "loss": 5.2298,
      "mean_token_accuracy": 0.20714985132217406,
      "num_tokens": 102262483.0,
      "step": 50470
    },
    {
      "entropy": 5.7761835098266605,
      "epoch": 4.41099362055405,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003197950159491697,
      "loss": 5.1959,
      "mean_token_accuracy": 0.21096674650907515,
      "num_tokens": 102271213.0,
      "step": 50475
    },
    {
      "entropy": 5.895807456970215,
      "epoch": 4.411430568906756,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031976447822256295,
      "loss": 5.3732,
      "mean_token_accuracy": 0.1964925065636635,
      "num_tokens": 102280830.0,
      "step": 50480
    },
    {
      "entropy": 5.8930848121643065,
      "epoch": 4.41186751725946,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031973393963735,
      "loss": 5.3277,
      "mean_token_accuracy": 0.20629561245441436,
      "num_tokens": 102290869.0,
      "step": 50485
    },
    {
      "entropy": 5.815238809585571,
      "epoch": 4.412304465612165,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003197034001941166,
      "loss": 5.2795,
      "mean_token_accuracy": 0.20910938531160356,
      "num_tokens": 102301037.0,
      "step": 50490
    },
    {
      "entropy": 5.875391149520874,
      "epoch": 4.412741413964869,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031967285989344867,
      "loss": 5.3201,
      "mean_token_accuracy": 0.21016676872968673,
      "num_tokens": 102311485.0,
      "step": 50495
    },
    {
      "entropy": 5.911710548400879,
      "epoch": 4.413178362317574,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003196423187359319,
      "loss": 5.3101,
      "mean_token_accuracy": 0.2077197790145874,
      "num_tokens": 102320685.0,
      "step": 50500
    },
    {
      "entropy": 5.839983034133911,
      "epoch": 4.413615310670279,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031961177672215196,
      "loss": 5.2343,
      "mean_token_accuracy": 0.21864644587039947,
      "num_tokens": 102330946.0,
      "step": 50505
    },
    {
      "entropy": 5.91785078048706,
      "epoch": 4.414052259022983,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031958123385269485,
      "loss": 5.3245,
      "mean_token_accuracy": 0.2030795618891716,
      "num_tokens": 102340638.0,
      "step": 50510
    },
    {
      "entropy": 5.8934708595275875,
      "epoch": 4.414489207375688,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00031955069012814624,
      "loss": 5.3013,
      "mean_token_accuracy": 0.2085578426718712,
      "num_tokens": 102350691.0,
      "step": 50515
    },
    {
      "entropy": 5.923488569259644,
      "epoch": 4.414926155728393,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003195201455490921,
      "loss": 5.3609,
      "mean_token_accuracy": 0.19864840507507325,
      "num_tokens": 102361572.0,
      "step": 50520
    },
    {
      "entropy": 5.893928194046021,
      "epoch": 4.415363104081098,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031948960011611815,
      "loss": 5.3572,
      "mean_token_accuracy": 0.20068349689245224,
      "num_tokens": 102371599.0,
      "step": 50525
    },
    {
      "entropy": 5.829380464553833,
      "epoch": 4.415800052433802,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003194590538298105,
      "loss": 5.3107,
      "mean_token_accuracy": 0.20557239055633544,
      "num_tokens": 102381987.0,
      "step": 50530
    },
    {
      "entropy": 5.919538784027099,
      "epoch": 4.4162370007865075,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003194285066907548,
      "loss": 5.3514,
      "mean_token_accuracy": 0.1979216754436493,
      "num_tokens": 102392677.0,
      "step": 50535
    },
    {
      "entropy": 5.85387225151062,
      "epoch": 4.416673949139212,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003193979586995371,
      "loss": 5.2656,
      "mean_token_accuracy": 0.21486357152462005,
      "num_tokens": 102402553.0,
      "step": 50540
    },
    {
      "entropy": 5.84424352645874,
      "epoch": 4.417110897491916,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031936740985674325,
      "loss": 5.2499,
      "mean_token_accuracy": 0.20915690064430237,
      "num_tokens": 102412291.0,
      "step": 50545
    },
    {
      "entropy": 5.815459299087524,
      "epoch": 4.417547845844621,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003193368601629593,
      "loss": 5.2888,
      "mean_token_accuracy": 0.20805515795946122,
      "num_tokens": 102421749.0,
      "step": 50550
    },
    {
      "entropy": 5.815652179718017,
      "epoch": 4.417984794197326,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031930630961877105,
      "loss": 5.3213,
      "mean_token_accuracy": 0.20295144021511077,
      "num_tokens": 102432170.0,
      "step": 50555
    },
    {
      "entropy": 5.927330064773559,
      "epoch": 4.418421742550031,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003192757582247646,
      "loss": 5.3197,
      "mean_token_accuracy": 0.1992124781012535,
      "num_tokens": 102442454.0,
      "step": 50560
    },
    {
      "entropy": 5.833695983886718,
      "epoch": 4.418858690902735,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031924520598152595,
      "loss": 5.2414,
      "mean_token_accuracy": 0.21971308290958405,
      "num_tokens": 102454007.0,
      "step": 50565
    },
    {
      "entropy": 5.901796627044678,
      "epoch": 4.41929563925544,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031921465288964096,
      "loss": 5.3267,
      "mean_token_accuracy": 0.20461802035570145,
      "num_tokens": 102464601.0,
      "step": 50570
    },
    {
      "entropy": 5.883705043792725,
      "epoch": 4.419732587608145,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00031918409894969585,
      "loss": 5.3912,
      "mean_token_accuracy": 0.19896905720233918,
      "num_tokens": 102475552.0,
      "step": 50575
    },
    {
      "entropy": 5.918115615844727,
      "epoch": 4.4201695359608495,
      "grad_norm": 1.0,
      "learning_rate": 0.0003191535441622766,
      "loss": 5.3503,
      "mean_token_accuracy": 0.2053256541490555,
      "num_tokens": 102485934.0,
      "step": 50580
    },
    {
      "entropy": 5.9099020004272464,
      "epoch": 4.420606484313554,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003191229885279692,
      "loss": 5.3394,
      "mean_token_accuracy": 0.20199307650327683,
      "num_tokens": 102494987.0,
      "step": 50585
    },
    {
      "entropy": 5.886446952819824,
      "epoch": 4.421043432666258,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00031909243204735974,
      "loss": 5.2986,
      "mean_token_accuracy": 0.19748401939868926,
      "num_tokens": 102506556.0,
      "step": 50590
    },
    {
      "entropy": 5.921586561203003,
      "epoch": 4.421480381018964,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003190618747210343,
      "loss": 5.3178,
      "mean_token_accuracy": 0.20890088230371476,
      "num_tokens": 102516483.0,
      "step": 50595
    },
    {
      "entropy": 5.829165744781494,
      "epoch": 4.421917329371668,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003190313165495791,
      "loss": 5.2782,
      "mean_token_accuracy": 0.21105400919914247,
      "num_tokens": 102526502.0,
      "step": 50600
    },
    {
      "entropy": 5.919921922683716,
      "epoch": 4.422354277724373,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003190007575335801,
      "loss": 5.3152,
      "mean_token_accuracy": 0.20049750059843063,
      "num_tokens": 102536376.0,
      "step": 50605
    },
    {
      "entropy": 5.839821863174438,
      "epoch": 4.422791226077078,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003189701976736236,
      "loss": 5.2642,
      "mean_token_accuracy": 0.20881011784076692,
      "num_tokens": 102546221.0,
      "step": 50610
    },
    {
      "entropy": 5.838165235519409,
      "epoch": 4.4232281744297826,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00031893963697029554,
      "loss": 5.3124,
      "mean_token_accuracy": 0.21048667579889296,
      "num_tokens": 102556412.0,
      "step": 50615
    },
    {
      "entropy": 5.8766649723052975,
      "epoch": 4.423665122782487,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031890907542418234,
      "loss": 5.3115,
      "mean_token_accuracy": 0.20643449127674102,
      "num_tokens": 102566535.0,
      "step": 50620
    },
    {
      "entropy": 5.952707386016845,
      "epoch": 4.4241020711351915,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031887851303587,
      "loss": 5.4393,
      "mean_token_accuracy": 0.1909598618745804,
      "num_tokens": 102576902.0,
      "step": 50625
    },
    {
      "entropy": 5.9432250499725345,
      "epoch": 4.424539019487897,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003188479498059448,
      "loss": 5.3352,
      "mean_token_accuracy": 0.19689384251832961,
      "num_tokens": 102587604.0,
      "step": 50630
    },
    {
      "entropy": 5.8658106327056885,
      "epoch": 4.424975967840601,
      "grad_norm": 1.046875,
      "learning_rate": 0.000318817385734993,
      "loss": 5.2912,
      "mean_token_accuracy": 0.20287907868623734,
      "num_tokens": 102597978.0,
      "step": 50635
    },
    {
      "entropy": 5.926531791687012,
      "epoch": 4.425412916193306,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003187868208236007,
      "loss": 5.3491,
      "mean_token_accuracy": 0.20071059763431548,
      "num_tokens": 102608156.0,
      "step": 50640
    },
    {
      "entropy": 5.870338773727417,
      "epoch": 4.42584986454601,
      "grad_norm": 1.25,
      "learning_rate": 0.00031875625507235436,
      "loss": 5.3319,
      "mean_token_accuracy": 0.20762190371751785,
      "num_tokens": 102617726.0,
      "step": 50645
    },
    {
      "entropy": 5.863595771789551,
      "epoch": 4.426286812898716,
      "grad_norm": 1.125,
      "learning_rate": 0.00031872568848183996,
      "loss": 5.3243,
      "mean_token_accuracy": 0.2076472669839859,
      "num_tokens": 102628444.0,
      "step": 50650
    },
    {
      "entropy": 5.881782054901123,
      "epoch": 4.42672376125142,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031869512105264395,
      "loss": 5.2883,
      "mean_token_accuracy": 0.20793270021677018,
      "num_tokens": 102637751.0,
      "step": 50655
    },
    {
      "entropy": 5.886629962921143,
      "epoch": 4.4271607096041246,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003186645527853527,
      "loss": 5.3649,
      "mean_token_accuracy": 0.1985291436314583,
      "num_tokens": 102648276.0,
      "step": 50660
    },
    {
      "entropy": 5.836751794815063,
      "epoch": 4.42759765795683,
      "grad_norm": 0.9375,
      "learning_rate": 0.00031863398368055246,
      "loss": 5.2057,
      "mean_token_accuracy": 0.21322890371084213,
      "num_tokens": 102659113.0,
      "step": 50665
    },
    {
      "entropy": 5.828284788131714,
      "epoch": 4.428034606309534,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003186034137388296,
      "loss": 5.315,
      "mean_token_accuracy": 0.20133196115493773,
      "num_tokens": 102669502.0,
      "step": 50670
    },
    {
      "entropy": 5.842165327072143,
      "epoch": 4.428471554662239,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003185728429607703,
      "loss": 5.2235,
      "mean_token_accuracy": 0.2148209646344185,
      "num_tokens": 102679804.0,
      "step": 50675
    },
    {
      "entropy": 5.855023527145386,
      "epoch": 4.428908503014943,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00031854227134696107,
      "loss": 5.2302,
      "mean_token_accuracy": 0.2117108777165413,
      "num_tokens": 102689466.0,
      "step": 50680
    },
    {
      "entropy": 5.835519218444825,
      "epoch": 4.429345451367649,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031851169889798826,
      "loss": 5.2607,
      "mean_token_accuracy": 0.21183625757694244,
      "num_tokens": 102699957.0,
      "step": 50685
    },
    {
      "entropy": 5.833440351486206,
      "epoch": 4.429782399720353,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031848112561443825,
      "loss": 5.252,
      "mean_token_accuracy": 0.20774126648902894,
      "num_tokens": 102710575.0,
      "step": 50690
    },
    {
      "entropy": 5.854554319381714,
      "epoch": 4.430219348073058,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031845055149689754,
      "loss": 5.2729,
      "mean_token_accuracy": 0.20804637521505356,
      "num_tokens": 102720575.0,
      "step": 50695
    },
    {
      "entropy": 5.867639255523682,
      "epoch": 4.430656296425762,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031841997654595246,
      "loss": 5.3444,
      "mean_token_accuracy": 0.19832260012626649,
      "num_tokens": 102730202.0,
      "step": 50700
    },
    {
      "entropy": 5.899957084655762,
      "epoch": 4.431093244778467,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003183894007621894,
      "loss": 5.4318,
      "mean_token_accuracy": 0.19755795150995253,
      "num_tokens": 102741305.0,
      "step": 50705
    },
    {
      "entropy": 5.936587762832642,
      "epoch": 4.431530193131172,
      "grad_norm": 1.125,
      "learning_rate": 0.000318358824146195,
      "loss": 5.298,
      "mean_token_accuracy": 0.20223957002162934,
      "num_tokens": 102752453.0,
      "step": 50710
    },
    {
      "entropy": 5.871823740005493,
      "epoch": 4.431967141483876,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003183282466985555,
      "loss": 5.3241,
      "mean_token_accuracy": 0.20205827057361603,
      "num_tokens": 102763095.0,
      "step": 50715
    },
    {
      "entropy": 5.8472052097320555,
      "epoch": 4.432404089836581,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00031829766841985765,
      "loss": 5.3375,
      "mean_token_accuracy": 0.20384080857038497,
      "num_tokens": 102774239.0,
      "step": 50720
    },
    {
      "entropy": 5.833161211013794,
      "epoch": 4.432841038189286,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031826708931068776,
      "loss": 5.2359,
      "mean_token_accuracy": 0.2078963577747345,
      "num_tokens": 102783577.0,
      "step": 50725
    },
    {
      "entropy": 5.809207630157471,
      "epoch": 4.433277986541991,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031823650937163245,
      "loss": 5.1896,
      "mean_token_accuracy": 0.20745789110660554,
      "num_tokens": 102793647.0,
      "step": 50730
    },
    {
      "entropy": 5.810722398757934,
      "epoch": 4.433714934894695,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031820592860327826,
      "loss": 5.2015,
      "mean_token_accuracy": 0.2131941720843315,
      "num_tokens": 102804732.0,
      "step": 50735
    },
    {
      "entropy": 5.853714227676392,
      "epoch": 4.4341518832474005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031817534700621176,
      "loss": 5.2787,
      "mean_token_accuracy": 0.21171863675117492,
      "num_tokens": 102813783.0,
      "step": 50740
    },
    {
      "entropy": 5.768544387817383,
      "epoch": 4.434588831600105,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003181447645810194,
      "loss": 5.2291,
      "mean_token_accuracy": 0.21791806519031526,
      "num_tokens": 102824856.0,
      "step": 50745
    },
    {
      "entropy": 5.880929231643677,
      "epoch": 4.435025779952809,
      "grad_norm": 0.96875,
      "learning_rate": 0.00031811418132828776,
      "loss": 5.3491,
      "mean_token_accuracy": 0.19797232300043105,
      "num_tokens": 102836321.0,
      "step": 50750
    },
    {
      "entropy": 5.873777961730957,
      "epoch": 4.435462728305514,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031808359724860364,
      "loss": 5.3213,
      "mean_token_accuracy": 0.19629932940006256,
      "num_tokens": 102846446.0,
      "step": 50755
    },
    {
      "entropy": 5.879258155822754,
      "epoch": 4.435899676658219,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031805301234255363,
      "loss": 5.3263,
      "mean_token_accuracy": 0.20349138230085373,
      "num_tokens": 102856702.0,
      "step": 50760
    },
    {
      "entropy": 5.879237699508667,
      "epoch": 4.436336625010924,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003180224266107242,
      "loss": 5.2354,
      "mean_token_accuracy": 0.20939462631940842,
      "num_tokens": 102866271.0,
      "step": 50765
    },
    {
      "entropy": 5.868006086349487,
      "epoch": 4.436773573363628,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031799184005370205,
      "loss": 5.2941,
      "mean_token_accuracy": 0.20568432062864303,
      "num_tokens": 102876063.0,
      "step": 50770
    },
    {
      "entropy": 5.781429100036621,
      "epoch": 4.437210521716333,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003179612526720739,
      "loss": 5.1975,
      "mean_token_accuracy": 0.21499789953231813,
      "num_tokens": 102885237.0,
      "step": 50775
    },
    {
      "entropy": 5.85844464302063,
      "epoch": 4.437647470069038,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031793066446642645,
      "loss": 5.2854,
      "mean_token_accuracy": 0.21117707341909409,
      "num_tokens": 102895515.0,
      "step": 50780
    },
    {
      "entropy": 5.864838171005249,
      "epoch": 4.4380844184217425,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031790007543734634,
      "loss": 5.3357,
      "mean_token_accuracy": 0.20169627517461777,
      "num_tokens": 102905651.0,
      "step": 50785
    },
    {
      "entropy": 5.919083070755005,
      "epoch": 4.438521366774447,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003178694855854203,
      "loss": 5.3734,
      "mean_token_accuracy": 0.20326804220676423,
      "num_tokens": 102916195.0,
      "step": 50790
    },
    {
      "entropy": 5.865467739105225,
      "epoch": 4.438958315127152,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003178388949112351,
      "loss": 5.2252,
      "mean_token_accuracy": 0.21068420857191086,
      "num_tokens": 102925952.0,
      "step": 50795
    },
    {
      "entropy": 5.972745513916015,
      "epoch": 4.439395263479857,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003178083034153774,
      "loss": 5.4428,
      "mean_token_accuracy": 0.18932831585407256,
      "num_tokens": 102936313.0,
      "step": 50800
    },
    {
      "entropy": 5.871929454803467,
      "epoch": 4.439832211832561,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00031777771109843403,
      "loss": 5.3408,
      "mean_token_accuracy": 0.1989915981888771,
      "num_tokens": 102947031.0,
      "step": 50805
    },
    {
      "entropy": 5.949351835250854,
      "epoch": 4.440269160185266,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003177471179609918,
      "loss": 5.4033,
      "mean_token_accuracy": 0.2009573757648468,
      "num_tokens": 102958063.0,
      "step": 50810
    },
    {
      "entropy": 5.904688167572021,
      "epoch": 4.440706108537971,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031771652400363736,
      "loss": 5.3612,
      "mean_token_accuracy": 0.20153820365667344,
      "num_tokens": 102968234.0,
      "step": 50815
    },
    {
      "entropy": 5.907513236999511,
      "epoch": 4.441143056890676,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003176859292269576,
      "loss": 5.3256,
      "mean_token_accuracy": 0.20077465921640397,
      "num_tokens": 102977220.0,
      "step": 50820
    },
    {
      "entropy": 5.896244955062866,
      "epoch": 4.44158000524338,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031765533363153935,
      "loss": 5.3262,
      "mean_token_accuracy": 0.20217857211828233,
      "num_tokens": 102986190.0,
      "step": 50825
    },
    {
      "entropy": 5.765053129196167,
      "epoch": 4.4420169535960845,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031762473721796947,
      "loss": 5.2771,
      "mean_token_accuracy": 0.2132832035422325,
      "num_tokens": 102996299.0,
      "step": 50830
    },
    {
      "entropy": 5.89821949005127,
      "epoch": 4.44245390194879,
      "grad_norm": 1.0,
      "learning_rate": 0.0003175941399868349,
      "loss": 5.3175,
      "mean_token_accuracy": 0.20083008259534835,
      "num_tokens": 103005045.0,
      "step": 50835
    },
    {
      "entropy": 5.882098722457886,
      "epoch": 4.442890850301494,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031756354193872234,
      "loss": 5.3003,
      "mean_token_accuracy": 0.20810115039348603,
      "num_tokens": 103014299.0,
      "step": 50840
    },
    {
      "entropy": 5.922882413864135,
      "epoch": 4.443327798654199,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003175329430742187,
      "loss": 5.3099,
      "mean_token_accuracy": 0.20637218207120894,
      "num_tokens": 103023114.0,
      "step": 50845
    },
    {
      "entropy": 5.895956039428711,
      "epoch": 4.443764747006904,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031750234339391085,
      "loss": 5.3639,
      "mean_token_accuracy": 0.2004299744963646,
      "num_tokens": 103033405.0,
      "step": 50850
    },
    {
      "entropy": 5.797478055953979,
      "epoch": 4.444201695359609,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003174717428983859,
      "loss": 5.2168,
      "mean_token_accuracy": 0.2121412545442581,
      "num_tokens": 103042935.0,
      "step": 50855
    },
    {
      "entropy": 5.809089803695679,
      "epoch": 4.444638643712313,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031744114158823074,
      "loss": 5.2604,
      "mean_token_accuracy": 0.20997465550899505,
      "num_tokens": 103052104.0,
      "step": 50860
    },
    {
      "entropy": 5.810531568527222,
      "epoch": 4.445075592065018,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003174105394640321,
      "loss": 5.2529,
      "mean_token_accuracy": 0.21244822591543197,
      "num_tokens": 103061953.0,
      "step": 50865
    },
    {
      "entropy": 5.861509180068969,
      "epoch": 4.445512540417723,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003173799365263771,
      "loss": 5.2783,
      "mean_token_accuracy": 0.21729498505592346,
      "num_tokens": 103072168.0,
      "step": 50870
    },
    {
      "entropy": 5.857574987411499,
      "epoch": 4.445949488770427,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031734933277585275,
      "loss": 5.2744,
      "mean_token_accuracy": 0.20971888303756714,
      "num_tokens": 103083029.0,
      "step": 50875
    },
    {
      "entropy": 5.898807573318481,
      "epoch": 4.446386437123132,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031731872821304604,
      "loss": 5.4079,
      "mean_token_accuracy": 0.19788180887699128,
      "num_tokens": 103094631.0,
      "step": 50880
    },
    {
      "entropy": 5.885119867324829,
      "epoch": 4.446823385475836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031728812283854396,
      "loss": 5.3444,
      "mean_token_accuracy": 0.20049714148044587,
      "num_tokens": 103104906.0,
      "step": 50885
    },
    {
      "entropy": 5.861316299438476,
      "epoch": 4.447260333828542,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003172575166529335,
      "loss": 5.3128,
      "mean_token_accuracy": 0.21129896938800813,
      "num_tokens": 103115591.0,
      "step": 50890
    },
    {
      "entropy": 5.86256251335144,
      "epoch": 4.447697282181246,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031722690965680164,
      "loss": 5.2525,
      "mean_token_accuracy": 0.20610511302947998,
      "num_tokens": 103125808.0,
      "step": 50895
    },
    {
      "entropy": 5.818567895889283,
      "epoch": 4.448134230533951,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031719630185073555,
      "loss": 5.2984,
      "mean_token_accuracy": 0.21188496053218842,
      "num_tokens": 103136291.0,
      "step": 50900
    },
    {
      "entropy": 5.907498979568482,
      "epoch": 4.448571178886656,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003171656932353224,
      "loss": 5.3593,
      "mean_token_accuracy": 0.20101988017559053,
      "num_tokens": 103145949.0,
      "step": 50905
    },
    {
      "entropy": 5.86639633178711,
      "epoch": 4.4490081272393605,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031713508381114907,
      "loss": 5.2598,
      "mean_token_accuracy": 0.2101414456963539,
      "num_tokens": 103155375.0,
      "step": 50910
    },
    {
      "entropy": 5.787395095825195,
      "epoch": 4.449445075592065,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031710447357880277,
      "loss": 5.2136,
      "mean_token_accuracy": 0.2075529620051384,
      "num_tokens": 103164617.0,
      "step": 50915
    },
    {
      "entropy": 5.8531084060668945,
      "epoch": 4.449882023944769,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003170738625388706,
      "loss": 5.3224,
      "mean_token_accuracy": 0.199871489405632,
      "num_tokens": 103174688.0,
      "step": 50920
    },
    {
      "entropy": 5.928691625595093,
      "epoch": 4.450318972297475,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003170432506919397,
      "loss": 5.3814,
      "mean_token_accuracy": 0.19549475759267806,
      "num_tokens": 103185186.0,
      "step": 50925
    },
    {
      "entropy": 5.901504611968994,
      "epoch": 4.450755920650179,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003170126380385972,
      "loss": 5.3057,
      "mean_token_accuracy": 0.20249245911836625,
      "num_tokens": 103194818.0,
      "step": 50930
    },
    {
      "entropy": 5.820775890350342,
      "epoch": 4.451192869002884,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031698202457943034,
      "loss": 5.2602,
      "mean_token_accuracy": 0.20855837613344191,
      "num_tokens": 103205299.0,
      "step": 50935
    },
    {
      "entropy": 5.8773480415344235,
      "epoch": 4.451629817355588,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031695141031502615,
      "loss": 5.4097,
      "mean_token_accuracy": 0.19841071963310242,
      "num_tokens": 103216157.0,
      "step": 50940
    },
    {
      "entropy": 5.869400310516357,
      "epoch": 4.4520667657082935,
      "grad_norm": 1.09375,
      "learning_rate": 0.000316920795245972,
      "loss": 5.2669,
      "mean_token_accuracy": 0.2064067542552948,
      "num_tokens": 103227009.0,
      "step": 50945
    },
    {
      "entropy": 5.921848487854004,
      "epoch": 4.452503714060998,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031689017937285507,
      "loss": 5.2952,
      "mean_token_accuracy": 0.19992795139551162,
      "num_tokens": 103236848.0,
      "step": 50950
    },
    {
      "entropy": 5.875112724304199,
      "epoch": 4.4529406624137025,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031685956269626253,
      "loss": 5.3606,
      "mean_token_accuracy": 0.20338526964187623,
      "num_tokens": 103247664.0,
      "step": 50955
    },
    {
      "entropy": 5.8730775833129885,
      "epoch": 4.453377610766408,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003168289452167816,
      "loss": 5.3126,
      "mean_token_accuracy": 0.2031506285071373,
      "num_tokens": 103258605.0,
      "step": 50960
    },
    {
      "entropy": 5.883624505996704,
      "epoch": 4.453814559119112,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003167983269349996,
      "loss": 5.3051,
      "mean_token_accuracy": 0.20767066180706023,
      "num_tokens": 103267903.0,
      "step": 50965
    },
    {
      "entropy": 5.879829931259155,
      "epoch": 4.454251507471817,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003167677078515038,
      "loss": 5.2393,
      "mean_token_accuracy": 0.2163277119398117,
      "num_tokens": 103278189.0,
      "step": 50970
    },
    {
      "entropy": 5.832387018203735,
      "epoch": 4.454688455824521,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003167370879668815,
      "loss": 5.2621,
      "mean_token_accuracy": 0.20674961805343628,
      "num_tokens": 103289250.0,
      "step": 50975
    },
    {
      "entropy": 5.962663984298706,
      "epoch": 4.455125404177227,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031670646728172,
      "loss": 5.3465,
      "mean_token_accuracy": 0.1991713061928749,
      "num_tokens": 103298572.0,
      "step": 50980
    },
    {
      "entropy": 5.848229169845581,
      "epoch": 4.455562352529931,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003166758457966065,
      "loss": 5.2159,
      "mean_token_accuracy": 0.2146165519952774,
      "num_tokens": 103310560.0,
      "step": 50985
    },
    {
      "entropy": 5.8489946842193605,
      "epoch": 4.4559993008826355,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003166452235121285,
      "loss": 5.3232,
      "mean_token_accuracy": 0.20547462701797486,
      "num_tokens": 103319854.0,
      "step": 50990
    },
    {
      "entropy": 5.917729091644287,
      "epoch": 4.45643624923534,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031661460042887323,
      "loss": 5.3247,
      "mean_token_accuracy": 0.2030266508460045,
      "num_tokens": 103330286.0,
      "step": 50995
    },
    {
      "entropy": 5.92387433052063,
      "epoch": 4.456873197588045,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031658397654742826,
      "loss": 5.3625,
      "mean_token_accuracy": 0.20476399660110473,
      "num_tokens": 103340253.0,
      "step": 51000
    },
    {
      "epoch": 4.456873197588045,
      "eval_entropy": 5.615338788064479,
      "eval_loss": 5.382210731506348,
      "eval_mean_token_accuracy": 0.2087255459184623,
      "eval_num_tokens": 103340253.0,
      "eval_runtime": 27.7823,
      "eval_samples_per_second": 1332.285,
      "eval_steps_per_second": 166.545,
      "step": 51000
    },
    {
      "entropy": 5.877036905288696,
      "epoch": 4.45731014594075,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031655335186838075,
      "loss": 5.3236,
      "mean_token_accuracy": 0.2011795610189438,
      "num_tokens": 103349785.0,
      "step": 51005
    },
    {
      "entropy": 5.866184329986572,
      "epoch": 4.457747094293454,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031652272639231817,
      "loss": 5.3362,
      "mean_token_accuracy": 0.20196183323860167,
      "num_tokens": 103360378.0,
      "step": 51010
    },
    {
      "entropy": 6.020517778396607,
      "epoch": 4.45818404264616,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00031649210011982793,
      "loss": 5.3369,
      "mean_token_accuracy": 0.20402597039937972,
      "num_tokens": 103372101.0,
      "step": 51015
    },
    {
      "entropy": 5.945769596099853,
      "epoch": 4.458620990998864,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003164614730514975,
      "loss": 5.3207,
      "mean_token_accuracy": 0.20552937984466552,
      "num_tokens": 103381988.0,
      "step": 51020
    },
    {
      "entropy": 5.7956085205078125,
      "epoch": 4.459057939351569,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003164308451879143,
      "loss": 5.3475,
      "mean_token_accuracy": 0.20337550342082977,
      "num_tokens": 103392410.0,
      "step": 51025
    },
    {
      "entropy": 5.750772953033447,
      "epoch": 4.459494887704273,
      "grad_norm": 1.0,
      "learning_rate": 0.0003164002165296657,
      "loss": 5.1589,
      "mean_token_accuracy": 0.21754737794399262,
      "num_tokens": 103402502.0,
      "step": 51030
    },
    {
      "entropy": 5.872868537902832,
      "epoch": 4.459931836056978,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003163695870773393,
      "loss": 5.2953,
      "mean_token_accuracy": 0.20585173219442368,
      "num_tokens": 103412918.0,
      "step": 51035
    },
    {
      "entropy": 5.919824409484863,
      "epoch": 4.460368784409683,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031633895683152253,
      "loss": 5.3407,
      "mean_token_accuracy": 0.20133314579725264,
      "num_tokens": 103423160.0,
      "step": 51040
    },
    {
      "entropy": 5.894679021835327,
      "epoch": 4.460805732762387,
      "grad_norm": 1.046875,
      "learning_rate": 0.000316308325792803,
      "loss": 5.2774,
      "mean_token_accuracy": 0.206310498714447,
      "num_tokens": 103433617.0,
      "step": 51045
    },
    {
      "entropy": 5.895169830322265,
      "epoch": 4.461242681115092,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031627769396176805,
      "loss": 5.3559,
      "mean_token_accuracy": 0.20334461331367493,
      "num_tokens": 103444080.0,
      "step": 51050
    },
    {
      "entropy": 5.874560689926147,
      "epoch": 4.461679629467797,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031624706133900535,
      "loss": 5.2779,
      "mean_token_accuracy": 0.2093919262290001,
      "num_tokens": 103454562.0,
      "step": 51055
    },
    {
      "entropy": 5.909206819534302,
      "epoch": 4.462116577820502,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003162164279251023,
      "loss": 5.3659,
      "mean_token_accuracy": 0.20183541029691696,
      "num_tokens": 103465407.0,
      "step": 51060
    },
    {
      "entropy": 5.8697446346282955,
      "epoch": 4.462553526173206,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031618579372064663,
      "loss": 5.2703,
      "mean_token_accuracy": 0.21030523180961608,
      "num_tokens": 103475777.0,
      "step": 51065
    },
    {
      "entropy": 5.8373010635375975,
      "epoch": 4.4629904745259115,
      "grad_norm": 1.0,
      "learning_rate": 0.0003161551587262259,
      "loss": 5.2664,
      "mean_token_accuracy": 0.20628395825624465,
      "num_tokens": 103486380.0,
      "step": 51070
    },
    {
      "entropy": 5.86988205909729,
      "epoch": 4.463427422878616,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003161245229424276,
      "loss": 5.2699,
      "mean_token_accuracy": 0.2081194758415222,
      "num_tokens": 103495127.0,
      "step": 51075
    },
    {
      "entropy": 5.877475690841675,
      "epoch": 4.46386437123132,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003160938863698394,
      "loss": 5.3722,
      "mean_token_accuracy": 0.20313683450222014,
      "num_tokens": 103504781.0,
      "step": 51080
    },
    {
      "entropy": 5.790260457992554,
      "epoch": 4.464301319584025,
      "grad_norm": 1.09375,
      "learning_rate": 0.000316063249009049,
      "loss": 5.2817,
      "mean_token_accuracy": 0.20925038903951645,
      "num_tokens": 103514969.0,
      "step": 51085
    },
    {
      "entropy": 5.933901405334472,
      "epoch": 4.46473826793673,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003160326108606438,
      "loss": 5.3187,
      "mean_token_accuracy": 0.20030258744955062,
      "num_tokens": 103525097.0,
      "step": 51090
    },
    {
      "entropy": 5.909491109848022,
      "epoch": 4.465175216289435,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031600197192521186,
      "loss": 5.2949,
      "mean_token_accuracy": 0.2123803660273552,
      "num_tokens": 103535201.0,
      "step": 51095
    },
    {
      "entropy": 5.829965496063233,
      "epoch": 4.465612164642139,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003159713322033405,
      "loss": 5.2712,
      "mean_token_accuracy": 0.2048403650522232,
      "num_tokens": 103546257.0,
      "step": 51100
    },
    {
      "entropy": 5.87236213684082,
      "epoch": 4.466049112994844,
      "grad_norm": 1.0,
      "learning_rate": 0.0003159406916956174,
      "loss": 5.2808,
      "mean_token_accuracy": 0.2037589058279991,
      "num_tokens": 103556825.0,
      "step": 51105
    },
    {
      "entropy": 5.837741804122925,
      "epoch": 4.466486061347549,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003159100504026305,
      "loss": 5.2601,
      "mean_token_accuracy": 0.20723021179437637,
      "num_tokens": 103567819.0,
      "step": 51110
    },
    {
      "entropy": 5.884961032867432,
      "epoch": 4.4669230097002535,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003158794083249674,
      "loss": 5.3003,
      "mean_token_accuracy": 0.2046838343143463,
      "num_tokens": 103577629.0,
      "step": 51115
    },
    {
      "entropy": 5.836811828613281,
      "epoch": 4.467359958052958,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003158487654632158,
      "loss": 5.3167,
      "mean_token_accuracy": 0.20508363991975784,
      "num_tokens": 103587855.0,
      "step": 51120
    },
    {
      "entropy": 5.919727468490601,
      "epoch": 4.467796906405663,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003158181218179635,
      "loss": 5.3319,
      "mean_token_accuracy": 0.20124106705188752,
      "num_tokens": 103598022.0,
      "step": 51125
    },
    {
      "entropy": 5.842428684234619,
      "epoch": 4.468233854758368,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031578747738979826,
      "loss": 5.2659,
      "mean_token_accuracy": 0.21239344626665116,
      "num_tokens": 103608212.0,
      "step": 51130
    },
    {
      "entropy": 5.7919114112854,
      "epoch": 4.468670803111072,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003157568321793077,
      "loss": 5.2253,
      "mean_token_accuracy": 0.21402337402105331,
      "num_tokens": 103617394.0,
      "step": 51135
    },
    {
      "entropy": 5.788240718841553,
      "epoch": 4.469107751463777,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031572618618707987,
      "loss": 5.2508,
      "mean_token_accuracy": 0.2136405035853386,
      "num_tokens": 103627566.0,
      "step": 51140
    },
    {
      "entropy": 5.859026384353638,
      "epoch": 4.469544699816482,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031569553941370246,
      "loss": 5.1787,
      "mean_token_accuracy": 0.21145973354578018,
      "num_tokens": 103637066.0,
      "step": 51145
    },
    {
      "entropy": 5.876022005081177,
      "epoch": 4.4699816481691865,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031566489185976324,
      "loss": 5.3903,
      "mean_token_accuracy": 0.19737018942832946,
      "num_tokens": 103649944.0,
      "step": 51150
    },
    {
      "entropy": 5.906365966796875,
      "epoch": 4.470418596521891,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031563424352585,
      "loss": 5.2996,
      "mean_token_accuracy": 0.20522621870040894,
      "num_tokens": 103660703.0,
      "step": 51155
    },
    {
      "entropy": 5.860276317596435,
      "epoch": 4.4708555448745955,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031560359441255083,
      "loss": 5.2735,
      "mean_token_accuracy": 0.2097115322947502,
      "num_tokens": 103671062.0,
      "step": 51160
    },
    {
      "entropy": 5.834095048904419,
      "epoch": 4.471292493227301,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031557294452045335,
      "loss": 5.307,
      "mean_token_accuracy": 0.20818978548049927,
      "num_tokens": 103681470.0,
      "step": 51165
    },
    {
      "entropy": 5.878631114959717,
      "epoch": 4.471729441580005,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031554229385014553,
      "loss": 5.3333,
      "mean_token_accuracy": 0.19677036702632905,
      "num_tokens": 103691557.0,
      "step": 51170
    },
    {
      "entropy": 5.905198478698731,
      "epoch": 4.47216638993271,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003155116424022153,
      "loss": 5.3509,
      "mean_token_accuracy": 0.19202993661165238,
      "num_tokens": 103701152.0,
      "step": 51175
    },
    {
      "entropy": 5.837211084365845,
      "epoch": 4.472603338285415,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031548099017725056,
      "loss": 5.3223,
      "mean_token_accuracy": 0.20553489178419113,
      "num_tokens": 103711852.0,
      "step": 51180
    },
    {
      "entropy": 5.839199209213257,
      "epoch": 4.47304028663812,
      "grad_norm": 1.0,
      "learning_rate": 0.0003154503371758392,
      "loss": 5.339,
      "mean_token_accuracy": 0.2007738694548607,
      "num_tokens": 103722195.0,
      "step": 51185
    },
    {
      "entropy": 5.888858461380005,
      "epoch": 4.473477234990824,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003154196833985691,
      "loss": 5.3233,
      "mean_token_accuracy": 0.20601805597543715,
      "num_tokens": 103732575.0,
      "step": 51190
    },
    {
      "entropy": 5.844042015075684,
      "epoch": 4.4739141833435285,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031538902884602844,
      "loss": 5.2749,
      "mean_token_accuracy": 0.2105465665459633,
      "num_tokens": 103741837.0,
      "step": 51195
    },
    {
      "entropy": 5.916032600402832,
      "epoch": 4.474351131696234,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031535837351880493,
      "loss": 5.3697,
      "mean_token_accuracy": 0.20283145159482957,
      "num_tokens": 103753390.0,
      "step": 51200
    },
    {
      "entropy": 5.985875415802002,
      "epoch": 4.474788080048938,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003153277174174867,
      "loss": 5.394,
      "mean_token_accuracy": 0.1961669847369194,
      "num_tokens": 103762286.0,
      "step": 51205
    },
    {
      "entropy": 5.8720306873321535,
      "epoch": 4.475225028401643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003152970605426618,
      "loss": 5.3479,
      "mean_token_accuracy": 0.19736702740192413,
      "num_tokens": 103771962.0,
      "step": 51210
    },
    {
      "entropy": 5.870976877212525,
      "epoch": 4.475661976754347,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031526640289491803,
      "loss": 5.3126,
      "mean_token_accuracy": 0.2015109047293663,
      "num_tokens": 103782213.0,
      "step": 51215
    },
    {
      "entropy": 5.830065774917602,
      "epoch": 4.476098925107053,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003152357444748436,
      "loss": 5.3043,
      "mean_token_accuracy": 0.20190633535385133,
      "num_tokens": 103792848.0,
      "step": 51220
    },
    {
      "entropy": 5.872437858581543,
      "epoch": 4.476535873459757,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00031520508528302656,
      "loss": 5.2593,
      "mean_token_accuracy": 0.20942180156707763,
      "num_tokens": 103803162.0,
      "step": 51225
    },
    {
      "entropy": 5.780837202072144,
      "epoch": 4.476972821812462,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003151744253200548,
      "loss": 5.186,
      "mean_token_accuracy": 0.2161381646990776,
      "num_tokens": 103813335.0,
      "step": 51230
    },
    {
      "entropy": 5.800647878646851,
      "epoch": 4.477409770165167,
      "grad_norm": 0.96875,
      "learning_rate": 0.00031514376458651665,
      "loss": 5.2213,
      "mean_token_accuracy": 0.21145304441452026,
      "num_tokens": 103823135.0,
      "step": 51235
    },
    {
      "entropy": 5.878399133682251,
      "epoch": 4.477846718517871,
      "grad_norm": 1.0625,
      "learning_rate": 0.000315113103083,
      "loss": 5.3182,
      "mean_token_accuracy": 0.20147282779216766,
      "num_tokens": 103833178.0,
      "step": 51240
    },
    {
      "entropy": 5.839502429962158,
      "epoch": 4.478283666870576,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031508244081009297,
      "loss": 5.1865,
      "mean_token_accuracy": 0.21871925443410872,
      "num_tokens": 103843742.0,
      "step": 51245
    },
    {
      "entropy": 5.897921085357666,
      "epoch": 4.47872061522328,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031505177776838377,
      "loss": 5.3879,
      "mean_token_accuracy": 0.2017794817686081,
      "num_tokens": 103853287.0,
      "step": 51250
    },
    {
      "entropy": 5.881277322769165,
      "epoch": 4.479157563575986,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031502111395846047,
      "loss": 5.2876,
      "mean_token_accuracy": 0.2062872216105461,
      "num_tokens": 103863522.0,
      "step": 51255
    },
    {
      "entropy": 5.850173568725586,
      "epoch": 4.47959451192869,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031499044938091124,
      "loss": 5.3192,
      "mean_token_accuracy": 0.20703302472829818,
      "num_tokens": 103873749.0,
      "step": 51260
    },
    {
      "entropy": 5.784308910369873,
      "epoch": 4.480031460281395,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031495978403632427,
      "loss": 5.2417,
      "mean_token_accuracy": 0.2087349385023117,
      "num_tokens": 103884055.0,
      "step": 51265
    },
    {
      "entropy": 5.920846176147461,
      "epoch": 4.480468408634099,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003149291179252877,
      "loss": 5.3211,
      "mean_token_accuracy": 0.21123697012662887,
      "num_tokens": 103893242.0,
      "step": 51270
    },
    {
      "entropy": 5.9520632266998295,
      "epoch": 4.4809053569868045,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031489845104838964,
      "loss": 5.3396,
      "mean_token_accuracy": 0.20801038295030594,
      "num_tokens": 103902952.0,
      "step": 51275
    },
    {
      "entropy": 5.881765365600586,
      "epoch": 4.481342305339509,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003148677834062184,
      "loss": 5.2718,
      "mean_token_accuracy": 0.20942979007959367,
      "num_tokens": 103913659.0,
      "step": 51280
    },
    {
      "entropy": 5.862037611007691,
      "epoch": 4.481779253692213,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003148371149993623,
      "loss": 5.3138,
      "mean_token_accuracy": 0.20364563912153244,
      "num_tokens": 103923201.0,
      "step": 51285
    },
    {
      "entropy": 5.911166000366211,
      "epoch": 4.482216202044918,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00031480644582840934,
      "loss": 5.3366,
      "mean_token_accuracy": 0.19902945160865784,
      "num_tokens": 103933996.0,
      "step": 51290
    },
    {
      "entropy": 5.828790664672852,
      "epoch": 4.482653150397623,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003147757758939479,
      "loss": 5.2485,
      "mean_token_accuracy": 0.20767533779144287,
      "num_tokens": 103943404.0,
      "step": 51295
    },
    {
      "entropy": 5.853985691070557,
      "epoch": 4.483090098750328,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003147451051965663,
      "loss": 5.3123,
      "mean_token_accuracy": 0.20382374227046968,
      "num_tokens": 103952669.0,
      "step": 51300
    },
    {
      "entropy": 5.841844797134399,
      "epoch": 4.483527047103032,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031471443373685276,
      "loss": 5.2305,
      "mean_token_accuracy": 0.21460532546043395,
      "num_tokens": 103963138.0,
      "step": 51305
    },
    {
      "entropy": 5.884892082214355,
      "epoch": 4.4839639954557375,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003146837615153955,
      "loss": 5.2612,
      "mean_token_accuracy": 0.19900963604450225,
      "num_tokens": 103972268.0,
      "step": 51310
    },
    {
      "entropy": 5.891167306900025,
      "epoch": 4.484400943808442,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031465308853278296,
      "loss": 5.398,
      "mean_token_accuracy": 0.19711467772722244,
      "num_tokens": 103983536.0,
      "step": 51315
    },
    {
      "entropy": 5.872593593597412,
      "epoch": 4.4848378921611465,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031462241478960335,
      "loss": 5.3262,
      "mean_token_accuracy": 0.21203410923480986,
      "num_tokens": 103993402.0,
      "step": 51320
    },
    {
      "entropy": 5.834023523330688,
      "epoch": 4.485274840513851,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003145917402864451,
      "loss": 5.2727,
      "mean_token_accuracy": 0.20751055777072908,
      "num_tokens": 104003834.0,
      "step": 51325
    },
    {
      "entropy": 5.879668855667115,
      "epoch": 4.485711788866556,
      "grad_norm": 1.125,
      "learning_rate": 0.0003145610650238965,
      "loss": 5.3173,
      "mean_token_accuracy": 0.1992011100053787,
      "num_tokens": 104014608.0,
      "step": 51330
    },
    {
      "entropy": 5.905725765228271,
      "epoch": 4.486148737219261,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000314530389002546,
      "loss": 5.2795,
      "mean_token_accuracy": 0.20249331146478652,
      "num_tokens": 104024826.0,
      "step": 51335
    },
    {
      "entropy": 5.925774049758911,
      "epoch": 4.486585685571965,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031449971222298186,
      "loss": 5.388,
      "mean_token_accuracy": 0.19957863688468933,
      "num_tokens": 104035108.0,
      "step": 51340
    },
    {
      "entropy": 5.781083440780639,
      "epoch": 4.48702263392467,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031446903468579265,
      "loss": 5.2581,
      "mean_token_accuracy": 0.21266739666461945,
      "num_tokens": 104045402.0,
      "step": 51345
    },
    {
      "entropy": 5.836198997497559,
      "epoch": 4.487459582277375,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003144383563915666,
      "loss": 5.2957,
      "mean_token_accuracy": 0.20845185965299606,
      "num_tokens": 104055023.0,
      "step": 51350
    },
    {
      "entropy": 5.907045698165893,
      "epoch": 4.4878965306300795,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031440767734089214,
      "loss": 5.3707,
      "mean_token_accuracy": 0.20904904156923293,
      "num_tokens": 104065450.0,
      "step": 51355
    },
    {
      "entropy": 5.920812749862671,
      "epoch": 4.488333478982784,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003143769975343579,
      "loss": 5.3475,
      "mean_token_accuracy": 0.20074950456619262,
      "num_tokens": 104076601.0,
      "step": 51360
    },
    {
      "entropy": 5.856366777420044,
      "epoch": 4.4887704273354885,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003143463169725521,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20697934925556183,
      "num_tokens": 104086931.0,
      "step": 51365
    },
    {
      "entropy": 5.817307949066162,
      "epoch": 4.489207375688194,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031431563565606334,
      "loss": 5.2561,
      "mean_token_accuracy": 0.21464408040046692,
      "num_tokens": 104096930.0,
      "step": 51370
    },
    {
      "entropy": 5.808157014846802,
      "epoch": 4.489644324040898,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003142849535854802,
      "loss": 5.3016,
      "mean_token_accuracy": 0.20756500214338303,
      "num_tokens": 104107645.0,
      "step": 51375
    },
    {
      "entropy": 5.861813926696778,
      "epoch": 4.490081272393603,
      "grad_norm": 1.0,
      "learning_rate": 0.00031425427076139093,
      "loss": 5.2662,
      "mean_token_accuracy": 0.2099413678050041,
      "num_tokens": 104118217.0,
      "step": 51380
    },
    {
      "entropy": 5.864776659011841,
      "epoch": 4.490518220746308,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031422358718438426,
      "loss": 5.2753,
      "mean_token_accuracy": 0.21281817704439163,
      "num_tokens": 104127933.0,
      "step": 51385
    },
    {
      "entropy": 5.819220972061157,
      "epoch": 4.490955169099013,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003141929028550486,
      "loss": 5.2389,
      "mean_token_accuracy": 0.20774303674697875,
      "num_tokens": 104138025.0,
      "step": 51390
    },
    {
      "entropy": 5.819635677337646,
      "epoch": 4.491392117451717,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031416221777397257,
      "loss": 5.2572,
      "mean_token_accuracy": 0.21110234260559083,
      "num_tokens": 104147969.0,
      "step": 51395
    },
    {
      "entropy": 5.859591674804688,
      "epoch": 4.4918290658044215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003141315319417447,
      "loss": 5.3694,
      "mean_token_accuracy": 0.1973710373044014,
      "num_tokens": 104158245.0,
      "step": 51400
    },
    {
      "entropy": 5.867709016799926,
      "epoch": 4.492266014157127,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003141008453589535,
      "loss": 5.2938,
      "mean_token_accuracy": 0.20854728072881698,
      "num_tokens": 104168470.0,
      "step": 51405
    },
    {
      "entropy": 5.890962934494018,
      "epoch": 4.492702962509831,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003140701580261875,
      "loss": 5.3195,
      "mean_token_accuracy": 0.20129330903291703,
      "num_tokens": 104178198.0,
      "step": 51410
    },
    {
      "entropy": 5.886189889907837,
      "epoch": 4.493139910862536,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003140394699440355,
      "loss": 5.3432,
      "mean_token_accuracy": 0.20282064527273178,
      "num_tokens": 104188801.0,
      "step": 51415
    },
    {
      "entropy": 5.916756725311279,
      "epoch": 4.49357685921524,
      "grad_norm": 0.984375,
      "learning_rate": 0.000314008781113086,
      "loss": 5.3533,
      "mean_token_accuracy": 0.1989197239279747,
      "num_tokens": 104199871.0,
      "step": 51420
    },
    {
      "entropy": 5.832956695556641,
      "epoch": 4.494013807567946,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003139780915339277,
      "loss": 5.2611,
      "mean_token_accuracy": 0.20877244621515273,
      "num_tokens": 104209865.0,
      "step": 51425
    },
    {
      "entropy": 5.874456882476807,
      "epoch": 4.49445075592065,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003139474012071491,
      "loss": 5.2918,
      "mean_token_accuracy": 0.20658743679523467,
      "num_tokens": 104220375.0,
      "step": 51430
    },
    {
      "entropy": 5.828093957901001,
      "epoch": 4.494887704273355,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000313916710133339,
      "loss": 5.2524,
      "mean_token_accuracy": 0.21254068464040757,
      "num_tokens": 104230220.0,
      "step": 51435
    },
    {
      "entropy": 5.832824993133545,
      "epoch": 4.49532465262606,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003138860183130859,
      "loss": 5.298,
      "mean_token_accuracy": 0.20720189213752746,
      "num_tokens": 104241071.0,
      "step": 51440
    },
    {
      "entropy": 5.863577842712402,
      "epoch": 4.495761600978764,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003138553257469787,
      "loss": 5.2811,
      "mean_token_accuracy": 0.20395808517932892,
      "num_tokens": 104251450.0,
      "step": 51445
    },
    {
      "entropy": 5.819148111343384,
      "epoch": 4.496198549331469,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031382463243560593,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20586588233709335,
      "num_tokens": 104261898.0,
      "step": 51450
    },
    {
      "entropy": 5.837490081787109,
      "epoch": 4.496635497684173,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003137939383795564,
      "loss": 5.2468,
      "mean_token_accuracy": 0.2127468466758728,
      "num_tokens": 104270694.0,
      "step": 51455
    },
    {
      "entropy": 5.867347288131714,
      "epoch": 4.497072446036879,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031376324357941873,
      "loss": 5.3412,
      "mean_token_accuracy": 0.1985252544283867,
      "num_tokens": 104280755.0,
      "step": 51460
    },
    {
      "entropy": 5.8339035987854,
      "epoch": 4.497509394389583,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00031373254803578185,
      "loss": 5.3229,
      "mean_token_accuracy": 0.204010272026062,
      "num_tokens": 104290349.0,
      "step": 51465
    },
    {
      "entropy": 5.8299720764160154,
      "epoch": 4.497946342742288,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00031370185174923435,
      "loss": 5.2604,
      "mean_token_accuracy": 0.2110443115234375,
      "num_tokens": 104301343.0,
      "step": 51470
    },
    {
      "entropy": 5.843518590927124,
      "epoch": 4.498383291094992,
      "grad_norm": 1.125,
      "learning_rate": 0.000313671154720365,
      "loss": 5.3023,
      "mean_token_accuracy": 0.20943599790334702,
      "num_tokens": 104311678.0,
      "step": 51475
    },
    {
      "entropy": 5.890209722518921,
      "epoch": 4.4988202394476975,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003136404569497627,
      "loss": 5.3288,
      "mean_token_accuracy": 0.20546850115060805,
      "num_tokens": 104321936.0,
      "step": 51480
    },
    {
      "entropy": 5.9064202308654785,
      "epoch": 4.499257187800402,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003136097584380162,
      "loss": 5.3587,
      "mean_token_accuracy": 0.19888655841350555,
      "num_tokens": 104332281.0,
      "step": 51485
    },
    {
      "entropy": 5.894320964813232,
      "epoch": 4.499694136153106,
      "grad_norm": 1.125,
      "learning_rate": 0.0003135790591857143,
      "loss": 5.2749,
      "mean_token_accuracy": 0.20509438812732697,
      "num_tokens": 104341977.0,
      "step": 51490
    },
    {
      "entropy": 5.938681650161743,
      "epoch": 4.500131084505812,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031354835919344574,
      "loss": 5.4293,
      "mean_token_accuracy": 0.1930396795272827,
      "num_tokens": 104353707.0,
      "step": 51495
    },
    {
      "entropy": 5.941150236129761,
      "epoch": 4.500568032858516,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003135176584617996,
      "loss": 5.3746,
      "mean_token_accuracy": 0.1953096181154251,
      "num_tokens": 104363754.0,
      "step": 51500
    },
    {
      "entropy": 5.912969160079956,
      "epoch": 4.501004981211221,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003134869569913645,
      "loss": 5.3215,
      "mean_token_accuracy": 0.2049574911594391,
      "num_tokens": 104373229.0,
      "step": 51505
    },
    {
      "entropy": 5.8090308666229244,
      "epoch": 4.501441929563925,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031345625478272935,
      "loss": 5.2573,
      "mean_token_accuracy": 0.21530428677797317,
      "num_tokens": 104384142.0,
      "step": 51510
    },
    {
      "entropy": 5.825900506973267,
      "epoch": 4.5018788779166306,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003134255518364832,
      "loss": 5.3015,
      "mean_token_accuracy": 0.20848212093114854,
      "num_tokens": 104394649.0,
      "step": 51515
    },
    {
      "entropy": 5.859847974777222,
      "epoch": 4.502315826269335,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003133948481532148,
      "loss": 5.3003,
      "mean_token_accuracy": 0.20866943001747132,
      "num_tokens": 104405243.0,
      "step": 51520
    },
    {
      "entropy": 5.882281827926636,
      "epoch": 4.5027527746220395,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000313364143733513,
      "loss": 5.287,
      "mean_token_accuracy": 0.21180831789970397,
      "num_tokens": 104415628.0,
      "step": 51525
    },
    {
      "entropy": 5.815600824356079,
      "epoch": 4.503189722974744,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003133334385779669,
      "loss": 5.2402,
      "mean_token_accuracy": 0.21425874382257462,
      "num_tokens": 104425137.0,
      "step": 51530
    },
    {
      "entropy": 5.819349479675293,
      "epoch": 4.503626671327449,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031330273268716534,
      "loss": 5.194,
      "mean_token_accuracy": 0.22119349390268325,
      "num_tokens": 104435654.0,
      "step": 51535
    },
    {
      "entropy": 5.852815437316894,
      "epoch": 4.504063619680154,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031327202606169736,
      "loss": 5.3229,
      "mean_token_accuracy": 0.1986898198723793,
      "num_tokens": 104446050.0,
      "step": 51540
    },
    {
      "entropy": 5.844503355026245,
      "epoch": 4.504500568032858,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031324131870215176,
      "loss": 5.3012,
      "mean_token_accuracy": 0.2053817853331566,
      "num_tokens": 104456681.0,
      "step": 51545
    },
    {
      "entropy": 5.857231950759887,
      "epoch": 4.504937516385564,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003132106106091177,
      "loss": 5.328,
      "mean_token_accuracy": 0.20144789665937424,
      "num_tokens": 104467540.0,
      "step": 51550
    },
    {
      "entropy": 5.817353391647339,
      "epoch": 4.505374464738268,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003131799017831841,
      "loss": 5.2973,
      "mean_token_accuracy": 0.2010225385427475,
      "num_tokens": 104478072.0,
      "step": 51555
    },
    {
      "entropy": 5.865844345092773,
      "epoch": 4.5058114130909726,
      "grad_norm": 1.125,
      "learning_rate": 0.00031314919222493994,
      "loss": 5.2204,
      "mean_token_accuracy": 0.2186482384800911,
      "num_tokens": 104487741.0,
      "step": 51560
    },
    {
      "entropy": 5.868883800506592,
      "epoch": 4.506248361443677,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00031311848193497433,
      "loss": 5.352,
      "mean_token_accuracy": 0.20673674941062928,
      "num_tokens": 104497481.0,
      "step": 51565
    },
    {
      "entropy": 5.8026782989501955,
      "epoch": 4.506685309796382,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003130877709138762,
      "loss": 5.3094,
      "mean_token_accuracy": 0.20262749195098878,
      "num_tokens": 104507351.0,
      "step": 51570
    },
    {
      "entropy": 5.9208379745483395,
      "epoch": 4.507122258149087,
      "grad_norm": 1.0,
      "learning_rate": 0.0003130570591622348,
      "loss": 5.3541,
      "mean_token_accuracy": 0.20432856529951096,
      "num_tokens": 104517625.0,
      "step": 51575
    },
    {
      "entropy": 5.888882303237915,
      "epoch": 4.507559206501791,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003130263466806389,
      "loss": 5.3653,
      "mean_token_accuracy": 0.20462478399276735,
      "num_tokens": 104528467.0,
      "step": 51580
    },
    {
      "entropy": 5.872527837753296,
      "epoch": 4.507996154854496,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031299563346967786,
      "loss": 5.3355,
      "mean_token_accuracy": 0.20038727819919586,
      "num_tokens": 104538606.0,
      "step": 51585
    },
    {
      "entropy": 5.883844566345215,
      "epoch": 4.508433103207201,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031296491952994063,
      "loss": 5.3515,
      "mean_token_accuracy": 0.19890309125185013,
      "num_tokens": 104548963.0,
      "step": 51590
    },
    {
      "entropy": 5.877556276321411,
      "epoch": 4.508870051559906,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003129342048620163,
      "loss": 5.3164,
      "mean_token_accuracy": 0.20448651611804963,
      "num_tokens": 104558844.0,
      "step": 51595
    },
    {
      "entropy": 5.878137588500977,
      "epoch": 4.50930699991261,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031290348946649404,
      "loss": 5.3414,
      "mean_token_accuracy": 0.20172502398490905,
      "num_tokens": 104569334.0,
      "step": 51600
    },
    {
      "entropy": 5.900870323181152,
      "epoch": 4.509743948265315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003128727733439631,
      "loss": 5.3305,
      "mean_token_accuracy": 0.198254357278347,
      "num_tokens": 104579478.0,
      "step": 51605
    },
    {
      "entropy": 5.900650405883789,
      "epoch": 4.51018089661802,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003128420564950124,
      "loss": 5.3028,
      "mean_token_accuracy": 0.20116532295942308,
      "num_tokens": 104589963.0,
      "step": 51610
    },
    {
      "entropy": 5.877502584457398,
      "epoch": 4.510617844970724,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003128113389202313,
      "loss": 5.2559,
      "mean_token_accuracy": 0.20919694900512695,
      "num_tokens": 104599600.0,
      "step": 51615
    },
    {
      "entropy": 5.801351261138916,
      "epoch": 4.511054793323429,
      "grad_norm": 1.0,
      "learning_rate": 0.0003127806206202089,
      "loss": 5.2353,
      "mean_token_accuracy": 0.21518320143222808,
      "num_tokens": 104608771.0,
      "step": 51620
    },
    {
      "entropy": 5.790413093566895,
      "epoch": 4.511491741676134,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003127499015955344,
      "loss": 5.2825,
      "mean_token_accuracy": 0.20745428800582885,
      "num_tokens": 104619257.0,
      "step": 51625
    },
    {
      "entropy": 5.865118741989136,
      "epoch": 4.511928690028839,
      "grad_norm": 1.078125,
      "learning_rate": 0.000312719181846797,
      "loss": 5.2406,
      "mean_token_accuracy": 0.20997416526079177,
      "num_tokens": 104628493.0,
      "step": 51630
    },
    {
      "entropy": 5.838980054855346,
      "epoch": 4.512365638381543,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003126884613745859,
      "loss": 5.2454,
      "mean_token_accuracy": 0.2122738003730774,
      "num_tokens": 104637901.0,
      "step": 51635
    },
    {
      "entropy": 5.864096021652221,
      "epoch": 4.512802586734248,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031265774017949033,
      "loss": 5.2695,
      "mean_token_accuracy": 0.21117375642061234,
      "num_tokens": 104647104.0,
      "step": 51640
    },
    {
      "entropy": 5.8722536087036135,
      "epoch": 4.513239535086953,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003126270182620997,
      "loss": 5.2061,
      "mean_token_accuracy": 0.21211026459932328,
      "num_tokens": 104656125.0,
      "step": 51645
    },
    {
      "entropy": 5.820919179916382,
      "epoch": 4.513676483439657,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031259629562300297,
      "loss": 5.3228,
      "mean_token_accuracy": 0.21203346699476242,
      "num_tokens": 104666216.0,
      "step": 51650
    },
    {
      "entropy": 5.871844959259033,
      "epoch": 4.514113431792362,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003125655722627898,
      "loss": 5.3226,
      "mean_token_accuracy": 0.20340946316719055,
      "num_tokens": 104675239.0,
      "step": 51655
    },
    {
      "entropy": 5.844612121582031,
      "epoch": 4.514550380145067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031253484818204906,
      "loss": 5.2952,
      "mean_token_accuracy": 0.20282898396253585,
      "num_tokens": 104686324.0,
      "step": 51660
    },
    {
      "entropy": 5.888803339004516,
      "epoch": 4.514987328497772,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003125041233813704,
      "loss": 5.3551,
      "mean_token_accuracy": 0.20129266232252122,
      "num_tokens": 104696163.0,
      "step": 51665
    },
    {
      "entropy": 5.861810636520386,
      "epoch": 4.515424276850476,
      "grad_norm": 1.015625,
      "learning_rate": 0.000312473397861343,
      "loss": 5.2669,
      "mean_token_accuracy": 0.20584845542907715,
      "num_tokens": 104707126.0,
      "step": 51670
    },
    {
      "entropy": 5.853896427154541,
      "epoch": 4.515861225203181,
      "grad_norm": 0.984375,
      "learning_rate": 0.00031244267162255614,
      "loss": 5.289,
      "mean_token_accuracy": 0.2027246192097664,
      "num_tokens": 104718052.0,
      "step": 51675
    },
    {
      "entropy": 5.817909812927246,
      "epoch": 4.516298173555886,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031241194466559923,
      "loss": 5.2333,
      "mean_token_accuracy": 0.2164406254887581,
      "num_tokens": 104727956.0,
      "step": 51680
    },
    {
      "entropy": 5.877263021469116,
      "epoch": 4.5167351219085905,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003123812169910616,
      "loss": 5.3149,
      "mean_token_accuracy": 0.2030907765030861,
      "num_tokens": 104738258.0,
      "step": 51685
    },
    {
      "entropy": 5.846544647216797,
      "epoch": 4.517172070261295,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003123504885995327,
      "loss": 5.3029,
      "mean_token_accuracy": 0.20435622185468674,
      "num_tokens": 104748608.0,
      "step": 51690
    },
    {
      "entropy": 5.8374059200286865,
      "epoch": 4.517609018613999,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003123197594916018,
      "loss": 5.2836,
      "mean_token_accuracy": 0.21139603704214097,
      "num_tokens": 104758410.0,
      "step": 51695
    },
    {
      "entropy": 5.892313289642334,
      "epoch": 4.518045966966705,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003122890296678584,
      "loss": 5.3569,
      "mean_token_accuracy": 0.20183461755514145,
      "num_tokens": 104768657.0,
      "step": 51700
    },
    {
      "entropy": 5.9261139869689945,
      "epoch": 4.518482915319409,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031225829912889184,
      "loss": 5.3832,
      "mean_token_accuracy": 0.200539667904377,
      "num_tokens": 104779053.0,
      "step": 51705
    },
    {
      "entropy": 5.843643569946289,
      "epoch": 4.518919863672114,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031222756787529154,
      "loss": 5.2137,
      "mean_token_accuracy": 0.21019974648952483,
      "num_tokens": 104788652.0,
      "step": 51710
    },
    {
      "entropy": 5.862091588973999,
      "epoch": 4.519356812024819,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003121968359076471,
      "loss": 5.2862,
      "mean_token_accuracy": 0.2085829570889473,
      "num_tokens": 104798757.0,
      "step": 51715
    },
    {
      "entropy": 5.864850282669067,
      "epoch": 4.519793760377524,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031216610322654776,
      "loss": 5.3009,
      "mean_token_accuracy": 0.2063595548272133,
      "num_tokens": 104808174.0,
      "step": 51720
    },
    {
      "entropy": 5.858302402496338,
      "epoch": 4.520230708730228,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003121353698325831,
      "loss": 5.2638,
      "mean_token_accuracy": 0.2108748137950897,
      "num_tokens": 104817509.0,
      "step": 51725
    },
    {
      "entropy": 5.831945371627808,
      "epoch": 4.5206676570829325,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003121046357263426,
      "loss": 5.2093,
      "mean_token_accuracy": 0.2190042331814766,
      "num_tokens": 104827518.0,
      "step": 51730
    },
    {
      "entropy": 5.8631679058074955,
      "epoch": 4.521104605435638,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003120739009084158,
      "loss": 5.3183,
      "mean_token_accuracy": 0.2055252745747566,
      "num_tokens": 104837868.0,
      "step": 51735
    },
    {
      "entropy": 5.895939207077026,
      "epoch": 4.521541553788342,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031204316537939206,
      "loss": 5.3347,
      "mean_token_accuracy": 0.20103674829006196,
      "num_tokens": 104847849.0,
      "step": 51740
    },
    {
      "entropy": 5.844247579574585,
      "epoch": 4.521978502141047,
      "grad_norm": 1.09375,
      "learning_rate": 0.000312012429139861,
      "loss": 5.3086,
      "mean_token_accuracy": 0.20660857260227203,
      "num_tokens": 104857751.0,
      "step": 51745
    },
    {
      "entropy": 5.869261598587036,
      "epoch": 4.522415450493751,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003119816921904122,
      "loss": 5.2352,
      "mean_token_accuracy": 0.2081732302904129,
      "num_tokens": 104867092.0,
      "step": 51750
    },
    {
      "entropy": 5.9326332092285154,
      "epoch": 4.522852398846457,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003119509545316352,
      "loss": 5.3914,
      "mean_token_accuracy": 0.19762705713510514,
      "num_tokens": 104876791.0,
      "step": 51755
    },
    {
      "entropy": 5.879228210449218,
      "epoch": 4.523289347199161,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031192021616411946,
      "loss": 5.3121,
      "mean_token_accuracy": 0.2074132815003395,
      "num_tokens": 104886715.0,
      "step": 51760
    },
    {
      "entropy": 5.941941928863526,
      "epoch": 4.523726295551866,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031188947708845466,
      "loss": 5.3764,
      "mean_token_accuracy": 0.20055610239505767,
      "num_tokens": 104898204.0,
      "step": 51765
    },
    {
      "entropy": 5.8889320373535154,
      "epoch": 4.524163243904571,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003118587373052304,
      "loss": 5.2796,
      "mean_token_accuracy": 0.20800074487924575,
      "num_tokens": 104908875.0,
      "step": 51770
    },
    {
      "entropy": 5.790482139587402,
      "epoch": 4.524600192257275,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031182799681503615,
      "loss": 5.2393,
      "mean_token_accuracy": 0.21300617307424546,
      "num_tokens": 104918487.0,
      "step": 51775
    },
    {
      "entropy": 5.865699577331543,
      "epoch": 4.52503714060998,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003117972556184617,
      "loss": 5.3271,
      "mean_token_accuracy": 0.2043019324541092,
      "num_tokens": 104928899.0,
      "step": 51780
    },
    {
      "entropy": 5.89669189453125,
      "epoch": 4.525474088962684,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003117665137160966,
      "loss": 5.3237,
      "mean_token_accuracy": 0.20497480481863023,
      "num_tokens": 104939728.0,
      "step": 51785
    },
    {
      "entropy": 5.917776584625244,
      "epoch": 4.52591103731539,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003117357711085305,
      "loss": 5.3444,
      "mean_token_accuracy": 0.20241875797510148,
      "num_tokens": 104949181.0,
      "step": 51790
    },
    {
      "entropy": 5.867275619506836,
      "epoch": 4.526347985668094,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031170502779635296,
      "loss": 5.3766,
      "mean_token_accuracy": 0.19876141846179962,
      "num_tokens": 104959014.0,
      "step": 51795
    },
    {
      "entropy": 5.8995527744293215,
      "epoch": 4.526784934020799,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031167428378015396,
      "loss": 5.3165,
      "mean_token_accuracy": 0.20413813143968582,
      "num_tokens": 104970063.0,
      "step": 51800
    },
    {
      "entropy": 5.834655237197876,
      "epoch": 4.527221882373503,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031164353906052285,
      "loss": 5.2341,
      "mean_token_accuracy": 0.21043859273195267,
      "num_tokens": 104979452.0,
      "step": 51805
    },
    {
      "entropy": 5.854603815078735,
      "epoch": 4.5276588307262084,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003116127936380494,
      "loss": 5.278,
      "mean_token_accuracy": 0.21172107458114625,
      "num_tokens": 104989647.0,
      "step": 51810
    },
    {
      "entropy": 5.833832454681397,
      "epoch": 4.528095779078913,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031158204751332343,
      "loss": 5.2107,
      "mean_token_accuracy": 0.21595707386732102,
      "num_tokens": 105000148.0,
      "step": 51815
    },
    {
      "entropy": 5.950288009643555,
      "epoch": 4.528532727431617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031155130068693463,
      "loss": 5.3299,
      "mean_token_accuracy": 0.1973836526274681,
      "num_tokens": 105012057.0,
      "step": 51820
    },
    {
      "entropy": 5.861201810836792,
      "epoch": 4.528969675784323,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031152055315947273,
      "loss": 5.2744,
      "mean_token_accuracy": 0.20363034456968307,
      "num_tokens": 105022074.0,
      "step": 51825
    },
    {
      "entropy": 5.914483976364136,
      "epoch": 4.529406624137027,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003114898049315275,
      "loss": 5.3628,
      "mean_token_accuracy": 0.1995369851589203,
      "num_tokens": 105032589.0,
      "step": 51830
    },
    {
      "entropy": 5.849371814727784,
      "epoch": 4.529843572489732,
      "grad_norm": 0.953125,
      "learning_rate": 0.00031145905600368865,
      "loss": 5.3096,
      "mean_token_accuracy": 0.2102719411253929,
      "num_tokens": 105043592.0,
      "step": 51835
    },
    {
      "entropy": 5.810153198242188,
      "epoch": 4.530280520842436,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003114283063765461,
      "loss": 5.2324,
      "mean_token_accuracy": 0.21035866141319276,
      "num_tokens": 105053804.0,
      "step": 51840
    },
    {
      "entropy": 5.802827405929565,
      "epoch": 4.5307174691951415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003113975560506895,
      "loss": 5.3073,
      "mean_token_accuracy": 0.2036381646990776,
      "num_tokens": 105063341.0,
      "step": 51845
    },
    {
      "entropy": 5.871879577636719,
      "epoch": 4.531154417547846,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003113668050267086,
      "loss": 5.3547,
      "mean_token_accuracy": 0.20633741617202758,
      "num_tokens": 105075228.0,
      "step": 51850
    },
    {
      "entropy": 5.859414196014404,
      "epoch": 4.5315913659005504,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031133605330519343,
      "loss": 5.2082,
      "mean_token_accuracy": 0.22494516521692276,
      "num_tokens": 105085653.0,
      "step": 51855
    },
    {
      "entropy": 5.91516432762146,
      "epoch": 4.532028314253255,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003113053008867337,
      "loss": 5.3705,
      "mean_token_accuracy": 0.199093995988369,
      "num_tokens": 105095969.0,
      "step": 51860
    },
    {
      "entropy": 5.856011772155762,
      "epoch": 4.53246526260596,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003112745477719193,
      "loss": 5.2375,
      "mean_token_accuracy": 0.21070903092622756,
      "num_tokens": 105105260.0,
      "step": 51865
    },
    {
      "entropy": 5.868954133987427,
      "epoch": 4.532902210958665,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031124379396134003,
      "loss": 5.2646,
      "mean_token_accuracy": 0.20410395860671998,
      "num_tokens": 105114749.0,
      "step": 51870
    },
    {
      "entropy": 5.870032501220703,
      "epoch": 4.533339159311369,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031121303945558584,
      "loss": 5.3089,
      "mean_token_accuracy": 0.20614167153835297,
      "num_tokens": 105124520.0,
      "step": 51875
    },
    {
      "entropy": 5.921774816513062,
      "epoch": 4.533776107664075,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031118228425524653,
      "loss": 5.3442,
      "mean_token_accuracy": 0.20617344677448274,
      "num_tokens": 105133886.0,
      "step": 51880
    },
    {
      "entropy": 5.812576866149902,
      "epoch": 4.534213056016779,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031115152836091206,
      "loss": 5.1541,
      "mean_token_accuracy": 0.22370402961969377,
      "num_tokens": 105143781.0,
      "step": 51885
    },
    {
      "entropy": 5.85586519241333,
      "epoch": 4.5346500043694835,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003111207717731723,
      "loss": 5.3398,
      "mean_token_accuracy": 0.20764379501342772,
      "num_tokens": 105154383.0,
      "step": 51890
    },
    {
      "entropy": 5.865975761413575,
      "epoch": 4.535086952722188,
      "grad_norm": 1.125,
      "learning_rate": 0.00031109001449261726,
      "loss": 5.2541,
      "mean_token_accuracy": 0.2067123219370842,
      "num_tokens": 105163450.0,
      "step": 51895
    },
    {
      "entropy": 5.839752435684204,
      "epoch": 4.5355239010748924,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031105925651983687,
      "loss": 5.3776,
      "mean_token_accuracy": 0.19861715883016587,
      "num_tokens": 105174046.0,
      "step": 51900
    },
    {
      "entropy": 5.8590607166290285,
      "epoch": 4.535960849427598,
      "grad_norm": 1.015625,
      "learning_rate": 0.000311028497855421,
      "loss": 5.309,
      "mean_token_accuracy": 0.20049065053462983,
      "num_tokens": 105183527.0,
      "step": 51905
    },
    {
      "entropy": 5.863059139251709,
      "epoch": 4.536397797780302,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003109977384999596,
      "loss": 5.2608,
      "mean_token_accuracy": 0.2022978350520134,
      "num_tokens": 105194002.0,
      "step": 51910
    },
    {
      "entropy": 5.870242023468018,
      "epoch": 4.536834746133007,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003109669784540428,
      "loss": 5.3128,
      "mean_token_accuracy": 0.20052632540464402,
      "num_tokens": 105204722.0,
      "step": 51915
    },
    {
      "entropy": 5.85947618484497,
      "epoch": 4.537271694485712,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031093621771826046,
      "loss": 5.2411,
      "mean_token_accuracy": 0.21303628981113434,
      "num_tokens": 105214066.0,
      "step": 51920
    },
    {
      "entropy": 5.938489770889282,
      "epoch": 4.537708642838417,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031090545629320265,
      "loss": 5.3164,
      "mean_token_accuracy": 0.1996279239654541,
      "num_tokens": 105224943.0,
      "step": 51925
    },
    {
      "entropy": 5.931277942657471,
      "epoch": 4.538145591191121,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00031087469417945935,
      "loss": 5.3574,
      "mean_token_accuracy": 0.20054538547992706,
      "num_tokens": 105235102.0,
      "step": 51930
    },
    {
      "entropy": 5.815386486053467,
      "epoch": 4.538582539543826,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003108439313776207,
      "loss": 5.2138,
      "mean_token_accuracy": 0.20644993484020233,
      "num_tokens": 105245505.0,
      "step": 51935
    },
    {
      "entropy": 5.87437801361084,
      "epoch": 4.539019487896531,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031081316788827664,
      "loss": 5.3145,
      "mean_token_accuracy": 0.20791985988616943,
      "num_tokens": 105256386.0,
      "step": 51940
    },
    {
      "entropy": 5.899372291564942,
      "epoch": 4.539456436249235,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003107824037120172,
      "loss": 5.3163,
      "mean_token_accuracy": 0.2052408292889595,
      "num_tokens": 105266457.0,
      "step": 51945
    },
    {
      "entropy": 5.8395970344543455,
      "epoch": 4.53989338460194,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003107516388494325,
      "loss": 5.2745,
      "mean_token_accuracy": 0.20748749673366546,
      "num_tokens": 105276558.0,
      "step": 51950
    },
    {
      "entropy": 5.8289165019989015,
      "epoch": 4.540330332954644,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031072087330111264,
      "loss": 5.2886,
      "mean_token_accuracy": 0.2042071998119354,
      "num_tokens": 105285619.0,
      "step": 51955
    },
    {
      "entropy": 5.843512725830078,
      "epoch": 4.54076728130735,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003106901070676478,
      "loss": 5.1825,
      "mean_token_accuracy": 0.21247498989105223,
      "num_tokens": 105296269.0,
      "step": 51960
    },
    {
      "entropy": 5.854588222503662,
      "epoch": 4.541204229660054,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031065934014962794,
      "loss": 5.3014,
      "mean_token_accuracy": 0.20840517431497574,
      "num_tokens": 105305498.0,
      "step": 51965
    },
    {
      "entropy": 5.832265138626099,
      "epoch": 4.541641178012759,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00031062857254764327,
      "loss": 5.1979,
      "mean_token_accuracy": 0.21574923396110535,
      "num_tokens": 105316259.0,
      "step": 51970
    },
    {
      "entropy": 5.846461820602417,
      "epoch": 4.542078126365464,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031059780426228385,
      "loss": 5.2846,
      "mean_token_accuracy": 0.21201569586992264,
      "num_tokens": 105325802.0,
      "step": 51975
    },
    {
      "entropy": 5.871899127960205,
      "epoch": 4.542515074718168,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031056703529413994,
      "loss": 5.2396,
      "mean_token_accuracy": 0.21360841244459153,
      "num_tokens": 105335349.0,
      "step": 51980
    },
    {
      "entropy": 5.852225637435913,
      "epoch": 4.542952023070873,
      "grad_norm": 1.125,
      "learning_rate": 0.0003105362656438016,
      "loss": 5.3051,
      "mean_token_accuracy": 0.2009572431445122,
      "num_tokens": 105345706.0,
      "step": 51985
    },
    {
      "entropy": 5.814505958557129,
      "epoch": 4.543388971423578,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003105054953118591,
      "loss": 5.2204,
      "mean_token_accuracy": 0.2122189521789551,
      "num_tokens": 105356354.0,
      "step": 51990
    },
    {
      "entropy": 5.848420810699463,
      "epoch": 4.543825919776283,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031047472429890263,
      "loss": 5.264,
      "mean_token_accuracy": 0.20777972042560577,
      "num_tokens": 105367133.0,
      "step": 51995
    },
    {
      "entropy": 5.923508930206299,
      "epoch": 4.544262868128987,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003104439526055223,
      "loss": 5.3791,
      "mean_token_accuracy": 0.20043010264635086,
      "num_tokens": 105376480.0,
      "step": 52000
    },
    {
      "entropy": 5.881692886352539,
      "epoch": 4.544699816481692,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003104131802323084,
      "loss": 5.3141,
      "mean_token_accuracy": 0.20433265566825867,
      "num_tokens": 105386237.0,
      "step": 52005
    },
    {
      "entropy": 5.863932704925537,
      "epoch": 4.545136764834396,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031038240717985125,
      "loss": 5.2684,
      "mean_token_accuracy": 0.2048640191555023,
      "num_tokens": 105395687.0,
      "step": 52010
    },
    {
      "entropy": 5.910127639770508,
      "epoch": 4.5455737131871015,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003103516334487409,
      "loss": 5.2493,
      "mean_token_accuracy": 0.20771035254001619,
      "num_tokens": 105405465.0,
      "step": 52015
    },
    {
      "entropy": 5.953179121017456,
      "epoch": 4.546010661539806,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003103208590395676,
      "loss": 5.3002,
      "mean_token_accuracy": 0.21323944479227067,
      "num_tokens": 105415234.0,
      "step": 52020
    },
    {
      "entropy": 5.83829026222229,
      "epoch": 4.54644760989251,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003102900839529219,
      "loss": 5.2827,
      "mean_token_accuracy": 0.20890575349330903,
      "num_tokens": 105425773.0,
      "step": 52025
    },
    {
      "entropy": 5.865842628479004,
      "epoch": 4.546884558245216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003102593081893938,
      "loss": 5.3611,
      "mean_token_accuracy": 0.20109526067972183,
      "num_tokens": 105436247.0,
      "step": 52030
    },
    {
      "entropy": 5.863050174713135,
      "epoch": 4.54732150659792,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031022853174957375,
      "loss": 5.2447,
      "mean_token_accuracy": 0.2125392198562622,
      "num_tokens": 105446227.0,
      "step": 52035
    },
    {
      "entropy": 5.886109161376953,
      "epoch": 4.547758454950625,
      "grad_norm": 1.1875,
      "learning_rate": 0.000310197754634052,
      "loss": 5.3527,
      "mean_token_accuracy": 0.2059929132461548,
      "num_tokens": 105456368.0,
      "step": 52040
    },
    {
      "entropy": 5.855030488967896,
      "epoch": 4.548195403303329,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003101669768434188,
      "loss": 5.3415,
      "mean_token_accuracy": 0.20472541898489,
      "num_tokens": 105465530.0,
      "step": 52045
    },
    {
      "entropy": 5.921938133239746,
      "epoch": 4.5486323516560345,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003101361983782646,
      "loss": 5.3321,
      "mean_token_accuracy": 0.1955801084637642,
      "num_tokens": 105475723.0,
      "step": 52050
    },
    {
      "entropy": 5.94149079322815,
      "epoch": 4.549069300008739,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031010541923917964,
      "loss": 5.3081,
      "mean_token_accuracy": 0.2093450278043747,
      "num_tokens": 105484810.0,
      "step": 52055
    },
    {
      "entropy": 5.888721323013305,
      "epoch": 4.5495062483614435,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031007463942675446,
      "loss": 5.264,
      "mean_token_accuracy": 0.2076100617647171,
      "num_tokens": 105494748.0,
      "step": 52060
    },
    {
      "entropy": 5.851361656188965,
      "epoch": 4.549943196714148,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003100438589415793,
      "loss": 5.2874,
      "mean_token_accuracy": 0.20730283111333847,
      "num_tokens": 105505752.0,
      "step": 52065
    },
    {
      "entropy": 5.869140577316284,
      "epoch": 4.550380145066853,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003100130777842445,
      "loss": 5.2595,
      "mean_token_accuracy": 0.20644893497228622,
      "num_tokens": 105515842.0,
      "step": 52070
    },
    {
      "entropy": 5.823695135116577,
      "epoch": 4.550817093419558,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003099822959553405,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21255674064159394,
      "num_tokens": 105525786.0,
      "step": 52075
    },
    {
      "entropy": 5.852768993377685,
      "epoch": 4.551254041772262,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030995151345545777,
      "loss": 5.2594,
      "mean_token_accuracy": 0.209671650826931,
      "num_tokens": 105535034.0,
      "step": 52080
    },
    {
      "entropy": 5.7510936737060545,
      "epoch": 4.551690990124968,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003099207302851868,
      "loss": 5.3054,
      "mean_token_accuracy": 0.2066660612821579,
      "num_tokens": 105545313.0,
      "step": 52085
    },
    {
      "entropy": 5.940607786178589,
      "epoch": 4.552127938477672,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030988994644511774,
      "loss": 5.361,
      "mean_token_accuracy": 0.1998934969305992,
      "num_tokens": 105555034.0,
      "step": 52090
    },
    {
      "entropy": 5.925607776641845,
      "epoch": 4.5525648868303765,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030985916193584137,
      "loss": 5.262,
      "mean_token_accuracy": 0.2083752289414406,
      "num_tokens": 105564694.0,
      "step": 52095
    },
    {
      "entropy": 5.849529504776001,
      "epoch": 4.553001835183081,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030982837675794787,
      "loss": 5.3512,
      "mean_token_accuracy": 0.20089765638113022,
      "num_tokens": 105575052.0,
      "step": 52100
    },
    {
      "entropy": 5.796879053115845,
      "epoch": 4.553438783535786,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000309797590912028,
      "loss": 5.2457,
      "mean_token_accuracy": 0.2150793731212616,
      "num_tokens": 105585529.0,
      "step": 52105
    },
    {
      "entropy": 5.886821556091308,
      "epoch": 4.553875731888491,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003097668043986721,
      "loss": 5.3142,
      "mean_token_accuracy": 0.20440386831760407,
      "num_tokens": 105596464.0,
      "step": 52110
    },
    {
      "entropy": 5.950108814239502,
      "epoch": 4.554312680241195,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030973601721847054,
      "loss": 5.295,
      "mean_token_accuracy": 0.20049994587898254,
      "num_tokens": 105605493.0,
      "step": 52115
    },
    {
      "entropy": 5.825717258453369,
      "epoch": 4.5547496285939,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000309705229372014,
      "loss": 5.3023,
      "mean_token_accuracy": 0.20620983242988586,
      "num_tokens": 105616265.0,
      "step": 52120
    },
    {
      "entropy": 5.824825811386108,
      "epoch": 4.555186576946605,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030967444085989304,
      "loss": 5.3006,
      "mean_token_accuracy": 0.20459491163492202,
      "num_tokens": 105625889.0,
      "step": 52125
    },
    {
      "entropy": 5.8534752368927006,
      "epoch": 4.55562352529931,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030964365168269816,
      "loss": 5.3313,
      "mean_token_accuracy": 0.20614725351333618,
      "num_tokens": 105635821.0,
      "step": 52130
    },
    {
      "entropy": 5.8513476848602295,
      "epoch": 4.556060473652014,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00030961286184101984,
      "loss": 5.3133,
      "mean_token_accuracy": 0.2023557350039482,
      "num_tokens": 105646787.0,
      "step": 52135
    },
    {
      "entropy": 5.8859419345855715,
      "epoch": 4.556497422004719,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003095820713354487,
      "loss": 5.3228,
      "mean_token_accuracy": 0.20311077535152436,
      "num_tokens": 105658001.0,
      "step": 52140
    },
    {
      "entropy": 5.849445009231568,
      "epoch": 4.556934370357424,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003095512801665753,
      "loss": 5.3004,
      "mean_token_accuracy": 0.20330228209495543,
      "num_tokens": 105667543.0,
      "step": 52145
    },
    {
      "entropy": 5.88223967552185,
      "epoch": 4.557371318710128,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030952048833499026,
      "loss": 5.2866,
      "mean_token_accuracy": 0.20492685586214066,
      "num_tokens": 105677365.0,
      "step": 52150
    },
    {
      "entropy": 5.919702339172363,
      "epoch": 4.557808267062833,
      "grad_norm": 1.375,
      "learning_rate": 0.0003094896958412841,
      "loss": 5.3109,
      "mean_token_accuracy": 0.19942890703678132,
      "num_tokens": 105688241.0,
      "step": 52155
    },
    {
      "entropy": 5.838140249252319,
      "epoch": 4.558245215415538,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003094589026860476,
      "loss": 5.2294,
      "mean_token_accuracy": 0.21044907420873643,
      "num_tokens": 105697719.0,
      "step": 52160
    },
    {
      "entropy": 5.8484728813171385,
      "epoch": 4.558682163768243,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003094281088698712,
      "loss": 5.3218,
      "mean_token_accuracy": 0.19580818116664886,
      "num_tokens": 105708268.0,
      "step": 52165
    },
    {
      "entropy": 5.8918619632720945,
      "epoch": 4.559119112120947,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003093973143933456,
      "loss": 5.4006,
      "mean_token_accuracy": 0.19932521730661393,
      "num_tokens": 105718668.0,
      "step": 52170
    },
    {
      "entropy": 5.914939212799072,
      "epoch": 4.559556060473652,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030936651925706166,
      "loss": 5.372,
      "mean_token_accuracy": 0.20327403098344804,
      "num_tokens": 105730904.0,
      "step": 52175
    },
    {
      "entropy": 5.8216881275177,
      "epoch": 4.559993008826357,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003093357234616098,
      "loss": 5.2232,
      "mean_token_accuracy": 0.21071690022945405,
      "num_tokens": 105742146.0,
      "step": 52180
    },
    {
      "entropy": 5.886979818344116,
      "epoch": 4.560429957179061,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003093049270075807,
      "loss": 5.3242,
      "mean_token_accuracy": 0.2027843028306961,
      "num_tokens": 105753218.0,
      "step": 52185
    },
    {
      "entropy": 5.834444046020508,
      "epoch": 4.560866905531766,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003092741298955651,
      "loss": 5.2266,
      "mean_token_accuracy": 0.21193936914205552,
      "num_tokens": 105763133.0,
      "step": 52190
    },
    {
      "entropy": 5.867635774612427,
      "epoch": 4.561303853884471,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00030924333212615384,
      "loss": 5.3512,
      "mean_token_accuracy": 0.20383504629135132,
      "num_tokens": 105773810.0,
      "step": 52195
    },
    {
      "entropy": 5.88496241569519,
      "epoch": 4.561740802237176,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003092125336999375,
      "loss": 5.3032,
      "mean_token_accuracy": 0.21013808995485306,
      "num_tokens": 105783918.0,
      "step": 52200
    },
    {
      "entropy": 5.849149703979492,
      "epoch": 4.56217775058988,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003091817346175068,
      "loss": 5.2541,
      "mean_token_accuracy": 0.20839283913373946,
      "num_tokens": 105793799.0,
      "step": 52205
    },
    {
      "entropy": 5.840907573699951,
      "epoch": 4.562614698942585,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00030915093487945256,
      "loss": 5.2937,
      "mean_token_accuracy": 0.20509640276432037,
      "num_tokens": 105804311.0,
      "step": 52210
    },
    {
      "entropy": 5.7797770500183105,
      "epoch": 4.56305164729529,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003091201344863655,
      "loss": 5.2773,
      "mean_token_accuracy": 0.21166493594646454,
      "num_tokens": 105814132.0,
      "step": 52215
    },
    {
      "entropy": 5.845098972320557,
      "epoch": 4.5634885956479945,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030908933343883635,
      "loss": 5.2819,
      "mean_token_accuracy": 0.2070665881037712,
      "num_tokens": 105824693.0,
      "step": 52220
    },
    {
      "entropy": 5.9722754001617435,
      "epoch": 4.563925544000699,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030905853173745595,
      "loss": 5.4158,
      "mean_token_accuracy": 0.192659692466259,
      "num_tokens": 105835024.0,
      "step": 52225
    },
    {
      "entropy": 5.869674015045166,
      "epoch": 4.564362492353403,
      "grad_norm": 1.046875,
      "learning_rate": 0.000309027729382815,
      "loss": 5.266,
      "mean_token_accuracy": 0.20945727378129958,
      "num_tokens": 105844555.0,
      "step": 52230
    },
    {
      "entropy": 5.8707709312438965,
      "epoch": 4.564799440706109,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003089969263755045,
      "loss": 5.3299,
      "mean_token_accuracy": 0.20554242432117462,
      "num_tokens": 105854408.0,
      "step": 52235
    },
    {
      "entropy": 5.811615085601806,
      "epoch": 4.565236389058813,
      "grad_norm": 1.046875,
      "learning_rate": 0.000308966122716115,
      "loss": 5.2286,
      "mean_token_accuracy": 0.21754254251718522,
      "num_tokens": 105863491.0,
      "step": 52240
    },
    {
      "entropy": 5.862958765029907,
      "epoch": 4.565673337411518,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003089353184052376,
      "loss": 5.2404,
      "mean_token_accuracy": 0.2108889639377594,
      "num_tokens": 105873001.0,
      "step": 52245
    },
    {
      "entropy": 5.798494815826416,
      "epoch": 4.566110285764223,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030890451344346286,
      "loss": 5.2271,
      "mean_token_accuracy": 0.2159054085612297,
      "num_tokens": 105883200.0,
      "step": 52250
    },
    {
      "entropy": 5.867471694946289,
      "epoch": 4.5665472341169275,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030887370783138187,
      "loss": 5.3177,
      "mean_token_accuracy": 0.20390281826257706,
      "num_tokens": 105893558.0,
      "step": 52255
    },
    {
      "entropy": 5.869404220581055,
      "epoch": 4.566984182469632,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003088429015695854,
      "loss": 5.3057,
      "mean_token_accuracy": 0.20386411845684052,
      "num_tokens": 105903945.0,
      "step": 52260
    },
    {
      "entropy": 5.885552835464478,
      "epoch": 4.5674211308223365,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003088120946586643,
      "loss": 5.2857,
      "mean_token_accuracy": 0.20794419646263124,
      "num_tokens": 105915028.0,
      "step": 52265
    },
    {
      "entropy": 5.87179160118103,
      "epoch": 4.567858079175042,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030878128709920964,
      "loss": 5.3303,
      "mean_token_accuracy": 0.20191251933574678,
      "num_tokens": 105925881.0,
      "step": 52270
    },
    {
      "entropy": 5.910100078582763,
      "epoch": 4.568295027527746,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030875047889181205,
      "loss": 5.3725,
      "mean_token_accuracy": 0.20353536307811737,
      "num_tokens": 105936924.0,
      "step": 52275
    },
    {
      "entropy": 5.923104524612427,
      "epoch": 4.568731975880451,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003087196700370626,
      "loss": 5.3364,
      "mean_token_accuracy": 0.20621880739927292,
      "num_tokens": 105947507.0,
      "step": 52280
    },
    {
      "entropy": 5.858429956436157,
      "epoch": 4.569168924233155,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003086888605355523,
      "loss": 5.2614,
      "mean_token_accuracy": 0.20819020867347718,
      "num_tokens": 105957146.0,
      "step": 52285
    },
    {
      "entropy": 5.786646223068237,
      "epoch": 4.569605872585861,
      "grad_norm": 1.09375,
      "learning_rate": 0.000308658050387872,
      "loss": 5.2373,
      "mean_token_accuracy": 0.21288246661424637,
      "num_tokens": 105967339.0,
      "step": 52290
    },
    {
      "entropy": 5.8471965312957765,
      "epoch": 4.570042820938565,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030862723959461266,
      "loss": 5.263,
      "mean_token_accuracy": 0.2184256374835968,
      "num_tokens": 105977280.0,
      "step": 52295
    },
    {
      "entropy": 5.822481393814087,
      "epoch": 4.5704797692912695,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003085964281563652,
      "loss": 5.2836,
      "mean_token_accuracy": 0.21068477630615234,
      "num_tokens": 105986691.0,
      "step": 52300
    },
    {
      "entropy": 5.868315935134888,
      "epoch": 4.570916717643975,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030856561607372064,
      "loss": 5.285,
      "mean_token_accuracy": 0.20414730161428452,
      "num_tokens": 105997589.0,
      "step": 52305
    },
    {
      "entropy": 5.806434440612793,
      "epoch": 4.571353665996679,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003085348033472699,
      "loss": 5.2634,
      "mean_token_accuracy": 0.2027882605791092,
      "num_tokens": 106007537.0,
      "step": 52310
    },
    {
      "entropy": 5.88897852897644,
      "epoch": 4.571790614349384,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030850398997760414,
      "loss": 5.2053,
      "mean_token_accuracy": 0.20716944038867952,
      "num_tokens": 106017731.0,
      "step": 52315
    },
    {
      "entropy": 5.876692152023315,
      "epoch": 4.572227562702088,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003084731759653144,
      "loss": 5.3849,
      "mean_token_accuracy": 0.20154156386852265,
      "num_tokens": 106028670.0,
      "step": 52320
    },
    {
      "entropy": 5.866723155975341,
      "epoch": 4.572664511054794,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030844236131099146,
      "loss": 5.2786,
      "mean_token_accuracy": 0.20775924623012543,
      "num_tokens": 106038167.0,
      "step": 52325
    },
    {
      "entropy": 5.8802234649658205,
      "epoch": 4.573101459407498,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030841154601522655,
      "loss": 5.3322,
      "mean_token_accuracy": 0.20780277699232103,
      "num_tokens": 106048733.0,
      "step": 52330
    },
    {
      "entropy": 5.8636155128479,
      "epoch": 4.573538407760203,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030838073007861066,
      "loss": 5.1912,
      "mean_token_accuracy": 0.21549474000930785,
      "num_tokens": 106058643.0,
      "step": 52335
    },
    {
      "entropy": 5.865606641769409,
      "epoch": 4.573975356112907,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030834991350173484,
      "loss": 5.3584,
      "mean_token_accuracy": 0.2041687935590744,
      "num_tokens": 106068796.0,
      "step": 52340
    },
    {
      "entropy": 5.7986531257629395,
      "epoch": 4.574412304465612,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030831909628519026,
      "loss": 5.3015,
      "mean_token_accuracy": 0.21115064918994902,
      "num_tokens": 106078901.0,
      "step": 52345
    },
    {
      "entropy": 5.8864092350006105,
      "epoch": 4.574849252818317,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030828827842956796,
      "loss": 5.3276,
      "mean_token_accuracy": 0.20284238904714585,
      "num_tokens": 106088249.0,
      "step": 52350
    },
    {
      "entropy": 5.9189599514007565,
      "epoch": 4.575286201171021,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000308257459935459,
      "loss": 5.2794,
      "mean_token_accuracy": 0.20425979942083358,
      "num_tokens": 106098421.0,
      "step": 52355
    },
    {
      "entropy": 5.802982759475708,
      "epoch": 4.575723149523727,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003082266408034546,
      "loss": 5.2336,
      "mean_token_accuracy": 0.2064686805009842,
      "num_tokens": 106108988.0,
      "step": 52360
    },
    {
      "entropy": 5.878615665435791,
      "epoch": 4.576160097876431,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003081958210341458,
      "loss": 5.3312,
      "mean_token_accuracy": 0.20134294778108597,
      "num_tokens": 106119925.0,
      "step": 52365
    },
    {
      "entropy": 5.831479787826538,
      "epoch": 4.576597046229136,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030816500062812375,
      "loss": 5.2378,
      "mean_token_accuracy": 0.2111864760518074,
      "num_tokens": 106129575.0,
      "step": 52370
    },
    {
      "entropy": 5.8290973663330075,
      "epoch": 4.57703399458184,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030813417958597955,
      "loss": 5.3193,
      "mean_token_accuracy": 0.2042226031422615,
      "num_tokens": 106139338.0,
      "step": 52375
    },
    {
      "entropy": 5.884653615951538,
      "epoch": 4.5774709429345455,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030810335790830445,
      "loss": 5.2437,
      "mean_token_accuracy": 0.20775675475597383,
      "num_tokens": 106149558.0,
      "step": 52380
    },
    {
      "entropy": 5.889644765853882,
      "epoch": 4.57790789128725,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003080725355956897,
      "loss": 5.321,
      "mean_token_accuracy": 0.20387441664934158,
      "num_tokens": 106160018.0,
      "step": 52385
    },
    {
      "entropy": 5.830389308929443,
      "epoch": 4.578344839639954,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003080417126487263,
      "loss": 5.2993,
      "mean_token_accuracy": 0.20909770727157592,
      "num_tokens": 106171027.0,
      "step": 52390
    },
    {
      "entropy": 5.874608373641967,
      "epoch": 4.578781787992659,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003080108890680056,
      "loss": 5.2488,
      "mean_token_accuracy": 0.20967918932437896,
      "num_tokens": 106180625.0,
      "step": 52395
    },
    {
      "entropy": 5.689042854309082,
      "epoch": 4.579218736345364,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003079800648541186,
      "loss": 5.1737,
      "mean_token_accuracy": 0.22497160881757736,
      "num_tokens": 106190318.0,
      "step": 52400
    },
    {
      "entropy": 5.855074739456176,
      "epoch": 4.579655684698069,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003079492400076568,
      "loss": 5.287,
      "mean_token_accuracy": 0.20552212446928025,
      "num_tokens": 106200145.0,
      "step": 52405
    },
    {
      "entropy": 5.911000919342041,
      "epoch": 4.580092633050773,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030791841452921127,
      "loss": 5.2723,
      "mean_token_accuracy": 0.21083561331033707,
      "num_tokens": 106210861.0,
      "step": 52410
    },
    {
      "entropy": 5.906364917755127,
      "epoch": 4.5805295814034785,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030788758841937335,
      "loss": 5.3099,
      "mean_token_accuracy": 0.20231201350688935,
      "num_tokens": 106221810.0,
      "step": 52415
    },
    {
      "entropy": 5.918536710739136,
      "epoch": 4.580966529756183,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003078567616787342,
      "loss": 5.3347,
      "mean_token_accuracy": 0.20554220229387282,
      "num_tokens": 106232332.0,
      "step": 52420
    },
    {
      "entropy": 5.907801818847656,
      "epoch": 4.5814034781088875,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003078259343078852,
      "loss": 5.2913,
      "mean_token_accuracy": 0.20450949668884277,
      "num_tokens": 106243243.0,
      "step": 52425
    },
    {
      "entropy": 5.8178380012512205,
      "epoch": 4.581840426461592,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003077951063074175,
      "loss": 5.2228,
      "mean_token_accuracy": 0.2085361063480377,
      "num_tokens": 106252346.0,
      "step": 52430
    },
    {
      "entropy": 5.812678623199463,
      "epoch": 4.582277374814297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003077642776779225,
      "loss": 5.308,
      "mean_token_accuracy": 0.20617355555295944,
      "num_tokens": 106261849.0,
      "step": 52435
    },
    {
      "entropy": 5.801471424102783,
      "epoch": 4.582714323167002,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030773344841999154,
      "loss": 5.3513,
      "mean_token_accuracy": 0.20373540222644806,
      "num_tokens": 106272846.0,
      "step": 52440
    },
    {
      "entropy": 5.867638206481933,
      "epoch": 4.583151271519706,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030770261853421585,
      "loss": 5.2386,
      "mean_token_accuracy": 0.21235105097293855,
      "num_tokens": 106282625.0,
      "step": 52445
    },
    {
      "entropy": 5.837595701217651,
      "epoch": 4.583588219872411,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003076717880211868,
      "loss": 5.2337,
      "mean_token_accuracy": 0.21679312884807586,
      "num_tokens": 106292340.0,
      "step": 52450
    },
    {
      "entropy": 5.885555744171143,
      "epoch": 4.584025168225116,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030764095688149567,
      "loss": 5.3638,
      "mean_token_accuracy": 0.20004789233207704,
      "num_tokens": 106301795.0,
      "step": 52455
    },
    {
      "entropy": 5.9620349407196045,
      "epoch": 4.5844621165778205,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030761012511573406,
      "loss": 5.3352,
      "mean_token_accuracy": 0.20383714735507966,
      "num_tokens": 106313640.0,
      "step": 52460
    },
    {
      "entropy": 5.925408267974854,
      "epoch": 4.584899064930525,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030757929272449296,
      "loss": 5.2771,
      "mean_token_accuracy": 0.20357863306999208,
      "num_tokens": 106323627.0,
      "step": 52465
    },
    {
      "entropy": 5.871213054656982,
      "epoch": 4.58533601328323,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003075484597083641,
      "loss": 5.3695,
      "mean_token_accuracy": 0.20076726973056794,
      "num_tokens": 106334495.0,
      "step": 52470
    },
    {
      "entropy": 5.876040124893189,
      "epoch": 4.585772961635935,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030751762606793864,
      "loss": 5.3347,
      "mean_token_accuracy": 0.20598786771297456,
      "num_tokens": 106345209.0,
      "step": 52475
    },
    {
      "entropy": 5.927794599533081,
      "epoch": 4.586209909988639,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003074867918038081,
      "loss": 5.3357,
      "mean_token_accuracy": 0.19943747222423552,
      "num_tokens": 106355294.0,
      "step": 52480
    },
    {
      "entropy": 5.893305873870849,
      "epoch": 4.586646858341344,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003074559569165639,
      "loss": 5.3429,
      "mean_token_accuracy": 0.20479410141706467,
      "num_tokens": 106366340.0,
      "step": 52485
    },
    {
      "entropy": 5.860875415802002,
      "epoch": 4.587083806694049,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003074251214067974,
      "loss": 5.2641,
      "mean_token_accuracy": 0.2119127482175827,
      "num_tokens": 106376270.0,
      "step": 52490
    },
    {
      "entropy": 5.874554061889649,
      "epoch": 4.587520755046754,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003073942852751001,
      "loss": 5.3313,
      "mean_token_accuracy": 0.20272491872310638,
      "num_tokens": 106386231.0,
      "step": 52495
    },
    {
      "entropy": 5.84969162940979,
      "epoch": 4.587957703399458,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030736344852206346,
      "loss": 5.2293,
      "mean_token_accuracy": 0.20697942823171617,
      "num_tokens": 106396109.0,
      "step": 52500
    },
    {
      "entropy": 5.797194194793701,
      "epoch": 4.5883946517521625,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030733261114827885,
      "loss": 5.2734,
      "mean_token_accuracy": 0.20499747842550278,
      "num_tokens": 106406404.0,
      "step": 52505
    },
    {
      "entropy": 5.855047798156738,
      "epoch": 4.588831600104868,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030730177315433787,
      "loss": 5.2536,
      "mean_token_accuracy": 0.2134235069155693,
      "num_tokens": 106416762.0,
      "step": 52510
    },
    {
      "entropy": 5.891932439804077,
      "epoch": 4.589268548457572,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030727093454083193,
      "loss": 5.35,
      "mean_token_accuracy": 0.197241972386837,
      "num_tokens": 106427095.0,
      "step": 52515
    },
    {
      "entropy": 5.810720825195313,
      "epoch": 4.589705496810277,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030724009530835254,
      "loss": 5.2873,
      "mean_token_accuracy": 0.21761052757501603,
      "num_tokens": 106437576.0,
      "step": 52520
    },
    {
      "entropy": 5.742618465423584,
      "epoch": 4.590142445162982,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003072092554574912,
      "loss": 5.1774,
      "mean_token_accuracy": 0.21908022463321686,
      "num_tokens": 106448735.0,
      "step": 52525
    },
    {
      "entropy": 5.843540620803833,
      "epoch": 4.590579393515687,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030717841498883964,
      "loss": 5.315,
      "mean_token_accuracy": 0.20022810995578766,
      "num_tokens": 106459039.0,
      "step": 52530
    },
    {
      "entropy": 5.820000839233399,
      "epoch": 4.591016341868391,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000307147573902989,
      "loss": 5.2551,
      "mean_token_accuracy": 0.2056149035692215,
      "num_tokens": 106469257.0,
      "step": 52535
    },
    {
      "entropy": 5.8195164680480955,
      "epoch": 4.591453290221096,
      "grad_norm": 1.0,
      "learning_rate": 0.0003071167322005312,
      "loss": 5.2559,
      "mean_token_accuracy": 0.2106275051832199,
      "num_tokens": 106479564.0,
      "step": 52540
    },
    {
      "entropy": 5.916013574600219,
      "epoch": 4.591890238573801,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030708588988205755,
      "loss": 5.295,
      "mean_token_accuracy": 0.20608755201101303,
      "num_tokens": 106489606.0,
      "step": 52545
    },
    {
      "entropy": 5.882121658325195,
      "epoch": 4.592327186926505,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030705504694815986,
      "loss": 5.2386,
      "mean_token_accuracy": 0.21098972260951995,
      "num_tokens": 106500850.0,
      "step": 52550
    },
    {
      "entropy": 5.849615383148193,
      "epoch": 4.59276413527921,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030702420339942944,
      "loss": 5.3449,
      "mean_token_accuracy": 0.19743068814277648,
      "num_tokens": 106510563.0,
      "step": 52555
    },
    {
      "entropy": 5.840738725662232,
      "epoch": 4.593201083631914,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030699335923645805,
      "loss": 5.3295,
      "mean_token_accuracy": 0.20134874135255815,
      "num_tokens": 106520709.0,
      "step": 52560
    },
    {
      "entropy": 5.896492004394531,
      "epoch": 4.59363803198462,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030696251445983723,
      "loss": 5.3461,
      "mean_token_accuracy": 0.19882191866636276,
      "num_tokens": 106530946.0,
      "step": 52565
    },
    {
      "entropy": 5.896259403228759,
      "epoch": 4.594074980337324,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003069316690701587,
      "loss": 5.2772,
      "mean_token_accuracy": 0.20648026913404466,
      "num_tokens": 106540735.0,
      "step": 52570
    },
    {
      "entropy": 5.868773746490478,
      "epoch": 4.594511928690029,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000306900823068014,
      "loss": 5.3075,
      "mean_token_accuracy": 0.2089163228869438,
      "num_tokens": 106551128.0,
      "step": 52575
    },
    {
      "entropy": 5.876089429855346,
      "epoch": 4.594948877042734,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030686997645399484,
      "loss": 5.3563,
      "mean_token_accuracy": 0.2015736624598503,
      "num_tokens": 106562602.0,
      "step": 52580
    },
    {
      "entropy": 5.8807440280914305,
      "epoch": 4.5953858253954385,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00030683912922869274,
      "loss": 5.2728,
      "mean_token_accuracy": 0.2100341185927391,
      "num_tokens": 106573409.0,
      "step": 52585
    },
    {
      "entropy": 5.977831268310547,
      "epoch": 4.595822773748143,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003068082813926995,
      "loss": 5.3777,
      "mean_token_accuracy": 0.19678955972194673,
      "num_tokens": 106584342.0,
      "step": 52590
    },
    {
      "entropy": 5.969777536392212,
      "epoch": 4.596259722100847,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003067774329466068,
      "loss": 5.394,
      "mean_token_accuracy": 0.19629598408937454,
      "num_tokens": 106596154.0,
      "step": 52595
    },
    {
      "entropy": 5.800753736495972,
      "epoch": 4.596696670453552,
      "grad_norm": 1.125,
      "learning_rate": 0.00030674658389100623,
      "loss": 5.1799,
      "mean_token_accuracy": 0.21508463025093078,
      "num_tokens": 106605842.0,
      "step": 52600
    },
    {
      "entropy": 5.878994369506836,
      "epoch": 4.597133618806257,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030671573422648955,
      "loss": 5.3455,
      "mean_token_accuracy": 0.19965143203735353,
      "num_tokens": 106616319.0,
      "step": 52605
    },
    {
      "entropy": 5.935425662994385,
      "epoch": 4.597570567158962,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003066848839536485,
      "loss": 5.3504,
      "mean_token_accuracy": 0.20298164188861847,
      "num_tokens": 106627259.0,
      "step": 52610
    },
    {
      "entropy": 5.916572141647339,
      "epoch": 4.598007515511666,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003066540330730747,
      "loss": 5.313,
      "mean_token_accuracy": 0.20599785447120667,
      "num_tokens": 106637712.0,
      "step": 52615
    },
    {
      "entropy": 5.912780141830444,
      "epoch": 4.598444463864372,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030662318158536004,
      "loss": 5.286,
      "mean_token_accuracy": 0.22014850974082947,
      "num_tokens": 106648896.0,
      "step": 52620
    },
    {
      "entropy": 5.869231557846069,
      "epoch": 4.598881412217076,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030659232949109614,
      "loss": 5.2871,
      "mean_token_accuracy": 0.2105064570903778,
      "num_tokens": 106660016.0,
      "step": 52625
    },
    {
      "entropy": 5.818046998977661,
      "epoch": 4.5993183605697805,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003065614767908749,
      "loss": 5.278,
      "mean_token_accuracy": 0.2070907860994339,
      "num_tokens": 106669631.0,
      "step": 52630
    },
    {
      "entropy": 5.851933336257934,
      "epoch": 4.599755308922486,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030653062348528793,
      "loss": 5.3039,
      "mean_token_accuracy": 0.2088264510035515,
      "num_tokens": 106679218.0,
      "step": 52635
    },
    {
      "entropy": 5.890206050872803,
      "epoch": 4.60019225727519,
      "grad_norm": 1.046875,
      "learning_rate": 0.000306499769574927,
      "loss": 5.2375,
      "mean_token_accuracy": 0.2116873934864998,
      "num_tokens": 106688694.0,
      "step": 52640
    },
    {
      "entropy": 5.8670069694519045,
      "epoch": 4.600629205627895,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003064689150603841,
      "loss": 5.2757,
      "mean_token_accuracy": 0.20904888063669205,
      "num_tokens": 106698044.0,
      "step": 52645
    },
    {
      "entropy": 5.831884527206421,
      "epoch": 4.601066153980599,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003064380599422509,
      "loss": 5.2665,
      "mean_token_accuracy": 0.20762018412351607,
      "num_tokens": 106707973.0,
      "step": 52650
    },
    {
      "entropy": 5.799865293502807,
      "epoch": 4.601503102333304,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003064072042211192,
      "loss": 5.1702,
      "mean_token_accuracy": 0.21535350978374482,
      "num_tokens": 106717827.0,
      "step": 52655
    },
    {
      "entropy": 5.829851531982422,
      "epoch": 4.601940050686009,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003063763478975809,
      "loss": 5.2504,
      "mean_token_accuracy": 0.2196258932352066,
      "num_tokens": 106727168.0,
      "step": 52660
    },
    {
      "entropy": 5.905605792999268,
      "epoch": 4.602376999038714,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003063454909722278,
      "loss": 5.348,
      "mean_token_accuracy": 0.20262046456336974,
      "num_tokens": 106738686.0,
      "step": 52665
    },
    {
      "entropy": 5.906528806686401,
      "epoch": 4.602813947391418,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003063146334456517,
      "loss": 5.3673,
      "mean_token_accuracy": 0.20324983596801757,
      "num_tokens": 106748326.0,
      "step": 52670
    },
    {
      "entropy": 5.884072494506836,
      "epoch": 4.603250895744123,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003062837753184446,
      "loss": 5.2416,
      "mean_token_accuracy": 0.21008677035570145,
      "num_tokens": 106758255.0,
      "step": 52675
    },
    {
      "entropy": 5.832061529159546,
      "epoch": 4.603687844096828,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030625291659119825,
      "loss": 5.3086,
      "mean_token_accuracy": 0.20728712230920793,
      "num_tokens": 106767671.0,
      "step": 52680
    },
    {
      "entropy": 5.804366159439087,
      "epoch": 4.604124792449532,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003062220572645046,
      "loss": 5.2138,
      "mean_token_accuracy": 0.2150893360376358,
      "num_tokens": 106778217.0,
      "step": 52685
    },
    {
      "entropy": 5.841497611999512,
      "epoch": 4.604561740802237,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003061911973389555,
      "loss": 5.2319,
      "mean_token_accuracy": 0.2091813638806343,
      "num_tokens": 106788134.0,
      "step": 52690
    },
    {
      "entropy": 5.79999794960022,
      "epoch": 4.604998689154942,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000306160336815143,
      "loss": 5.208,
      "mean_token_accuracy": 0.20593417435884476,
      "num_tokens": 106797933.0,
      "step": 52695
    },
    {
      "entropy": 5.82149453163147,
      "epoch": 4.605435637507647,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00030612947569365876,
      "loss": 5.2909,
      "mean_token_accuracy": 0.20675983726978303,
      "num_tokens": 106808175.0,
      "step": 52700
    },
    {
      "entropy": 5.889187145233154,
      "epoch": 4.605872585860351,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000306098613975095,
      "loss": 5.3186,
      "mean_token_accuracy": 0.20509725511074067,
      "num_tokens": 106819197.0,
      "step": 52705
    },
    {
      "entropy": 5.84720516204834,
      "epoch": 4.606309534213056,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030606775166004343,
      "loss": 5.2253,
      "mean_token_accuracy": 0.2152615651488304,
      "num_tokens": 106830527.0,
      "step": 52710
    },
    {
      "entropy": 5.859371185302734,
      "epoch": 4.606746482565761,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003060368887490961,
      "loss": 5.3286,
      "mean_token_accuracy": 0.19465316236019134,
      "num_tokens": 106840208.0,
      "step": 52715
    },
    {
      "entropy": 5.792564153671265,
      "epoch": 4.607183430918465,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030600602524284503,
      "loss": 5.2853,
      "mean_token_accuracy": 0.20587335228919984,
      "num_tokens": 106850144.0,
      "step": 52720
    },
    {
      "entropy": 5.878941297531128,
      "epoch": 4.60762037927117,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030597516114188217,
      "loss": 5.2762,
      "mean_token_accuracy": 0.20580639392137529,
      "num_tokens": 106860585.0,
      "step": 52725
    },
    {
      "entropy": 5.892504787445068,
      "epoch": 4.608057327623875,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030594429644679945,
      "loss": 5.2455,
      "mean_token_accuracy": 0.20528485774993896,
      "num_tokens": 106871098.0,
      "step": 52730
    },
    {
      "entropy": 5.912678289413452,
      "epoch": 4.60849427597658,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030591343115818885,
      "loss": 5.3566,
      "mean_token_accuracy": 0.20087236911058426,
      "num_tokens": 106880921.0,
      "step": 52735
    },
    {
      "entropy": 5.859355068206787,
      "epoch": 4.608931224329284,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003058825652766426,
      "loss": 5.289,
      "mean_token_accuracy": 0.20776865333318711,
      "num_tokens": 106890490.0,
      "step": 52740
    },
    {
      "entropy": 5.875985193252563,
      "epoch": 4.609368172681989,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003058516988027524,
      "loss": 5.3176,
      "mean_token_accuracy": 0.2041855424642563,
      "num_tokens": 106900200.0,
      "step": 52745
    },
    {
      "entropy": 5.888804483413696,
      "epoch": 4.609805121034694,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030582083173711055,
      "loss": 5.354,
      "mean_token_accuracy": 0.2032389983534813,
      "num_tokens": 106910354.0,
      "step": 52750
    },
    {
      "entropy": 5.892531538009644,
      "epoch": 4.6102420693873984,
      "grad_norm": 1.078125,
      "learning_rate": 0.000305789964080309,
      "loss": 5.2237,
      "mean_token_accuracy": 0.2096411943435669,
      "num_tokens": 106920669.0,
      "step": 52755
    },
    {
      "entropy": 5.8839517593383786,
      "epoch": 4.610679017740103,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030575909583293974,
      "loss": 5.3027,
      "mean_token_accuracy": 0.20440252870321274,
      "num_tokens": 106930163.0,
      "step": 52760
    },
    {
      "entropy": 5.8694596767425535,
      "epoch": 4.611115966092807,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000305728226995595,
      "loss": 5.3055,
      "mean_token_accuracy": 0.20125974714756012,
      "num_tokens": 106939588.0,
      "step": 52765
    },
    {
      "entropy": 5.8596361637115475,
      "epoch": 4.611552914445513,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003056973575688667,
      "loss": 5.3411,
      "mean_token_accuracy": 0.19696817100048064,
      "num_tokens": 106950044.0,
      "step": 52770
    },
    {
      "entropy": 5.828532028198242,
      "epoch": 4.611989862798217,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030566648755334704,
      "loss": 5.2208,
      "mean_token_accuracy": 0.21610063910484315,
      "num_tokens": 106959970.0,
      "step": 52775
    },
    {
      "entropy": 5.85274863243103,
      "epoch": 4.612426811150922,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000305635616949628,
      "loss": 5.2042,
      "mean_token_accuracy": 0.21317313462495804,
      "num_tokens": 106970055.0,
      "step": 52780
    },
    {
      "entropy": 5.842352294921875,
      "epoch": 4.612863759503627,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003056047457583018,
      "loss": 5.2458,
      "mean_token_accuracy": 0.20905374884605407,
      "num_tokens": 106979598.0,
      "step": 52785
    },
    {
      "entropy": 5.83248987197876,
      "epoch": 4.6133007078563315,
      "grad_norm": 1.125,
      "learning_rate": 0.0003055738739799606,
      "loss": 5.2744,
      "mean_token_accuracy": 0.21373941302299498,
      "num_tokens": 106990662.0,
      "step": 52790
    },
    {
      "entropy": 5.8514772891998295,
      "epoch": 4.613737656209036,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030554300161519645,
      "loss": 5.2948,
      "mean_token_accuracy": 0.21311238557100295,
      "num_tokens": 107000942.0,
      "step": 52795
    },
    {
      "entropy": 5.923272705078125,
      "epoch": 4.6141746045617404,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00030551212866460146,
      "loss": 5.3684,
      "mean_token_accuracy": 0.19815043807029725,
      "num_tokens": 107010928.0,
      "step": 52800
    },
    {
      "entropy": 5.804409313201904,
      "epoch": 4.614611552914446,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003054812551287679,
      "loss": 5.2545,
      "mean_token_accuracy": 0.21034546494483947,
      "num_tokens": 107020563.0,
      "step": 52805
    },
    {
      "entropy": 5.87577805519104,
      "epoch": 4.61504850126715,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030545038100828793,
      "loss": 5.301,
      "mean_token_accuracy": 0.21124560832977296,
      "num_tokens": 107032260.0,
      "step": 52810
    },
    {
      "entropy": 5.8824687004089355,
      "epoch": 4.615485449619855,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030541950630375367,
      "loss": 5.3184,
      "mean_token_accuracy": 0.19940262138843537,
      "num_tokens": 107041866.0,
      "step": 52815
    },
    {
      "entropy": 5.888841724395752,
      "epoch": 4.615922397972559,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003053886310157573,
      "loss": 5.3438,
      "mean_token_accuracy": 0.20049359947443007,
      "num_tokens": 107052621.0,
      "step": 52820
    },
    {
      "entropy": 5.903606653213501,
      "epoch": 4.616359346325265,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003053577551448912,
      "loss": 5.2829,
      "mean_token_accuracy": 0.2064453184604645,
      "num_tokens": 107061634.0,
      "step": 52825
    },
    {
      "entropy": 5.833686780929566,
      "epoch": 4.616796294677969,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003053268786917473,
      "loss": 5.3026,
      "mean_token_accuracy": 0.20806241482496263,
      "num_tokens": 107071683.0,
      "step": 52830
    },
    {
      "entropy": 5.785531139373779,
      "epoch": 4.6172332430306735,
      "grad_norm": 1.015625,
      "learning_rate": 0.000305296001656918,
      "loss": 5.2222,
      "mean_token_accuracy": 0.2130257472395897,
      "num_tokens": 107081654.0,
      "step": 52835
    },
    {
      "entropy": 5.8354325771331785,
      "epoch": 4.617670191383379,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030526512404099556,
      "loss": 5.2273,
      "mean_token_accuracy": 0.21562103778123856,
      "num_tokens": 107091853.0,
      "step": 52840
    },
    {
      "entropy": 5.873558139801025,
      "epoch": 4.618107139736083,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030523424584457217,
      "loss": 5.2313,
      "mean_token_accuracy": 0.21163812428712844,
      "num_tokens": 107101485.0,
      "step": 52845
    },
    {
      "entropy": 5.844868183135986,
      "epoch": 4.618544088088788,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003052033670682401,
      "loss": 5.3772,
      "mean_token_accuracy": 0.20091265588998794,
      "num_tokens": 107112051.0,
      "step": 52850
    },
    {
      "entropy": 5.903227376937866,
      "epoch": 4.618981036441492,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003051724877125916,
      "loss": 5.3495,
      "mean_token_accuracy": 0.1980556294322014,
      "num_tokens": 107121954.0,
      "step": 52855
    },
    {
      "entropy": 5.851165294647217,
      "epoch": 4.619417984794198,
      "grad_norm": 1.0625,
      "learning_rate": 0.000305141607778219,
      "loss": 5.2145,
      "mean_token_accuracy": 0.2158074527978897,
      "num_tokens": 107132521.0,
      "step": 52860
    },
    {
      "entropy": 5.871708011627197,
      "epoch": 4.619854933146902,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003051107272657146,
      "loss": 5.235,
      "mean_token_accuracy": 0.20895729809999466,
      "num_tokens": 107142795.0,
      "step": 52865
    },
    {
      "entropy": 5.841083908081055,
      "epoch": 4.620291881499607,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003050798461756706,
      "loss": 5.3326,
      "mean_token_accuracy": 0.20224103927612305,
      "num_tokens": 107153660.0,
      "step": 52870
    },
    {
      "entropy": 5.893749475479126,
      "epoch": 4.620728829852311,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00030504896450867934,
      "loss": 5.3265,
      "mean_token_accuracy": 0.20593298524618148,
      "num_tokens": 107163840.0,
      "step": 52875
    },
    {
      "entropy": 5.884877347946167,
      "epoch": 4.621165778205016,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030501808226533327,
      "loss": 5.2975,
      "mean_token_accuracy": 0.20697800070047379,
      "num_tokens": 107173471.0,
      "step": 52880
    },
    {
      "entropy": 5.889902257919312,
      "epoch": 4.621602726557721,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003049871994462247,
      "loss": 5.2306,
      "mean_token_accuracy": 0.21214406341314315,
      "num_tokens": 107183077.0,
      "step": 52885
    },
    {
      "entropy": 5.828957271575928,
      "epoch": 4.622039674910425,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030495631605194584,
      "loss": 5.2255,
      "mean_token_accuracy": 0.21700474470853806,
      "num_tokens": 107193330.0,
      "step": 52890
    },
    {
      "entropy": 5.816842842102051,
      "epoch": 4.622476623263131,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003049254320830892,
      "loss": 5.276,
      "mean_token_accuracy": 0.21100682467222215,
      "num_tokens": 107203528.0,
      "step": 52895
    },
    {
      "entropy": 5.856206655502319,
      "epoch": 4.622913571615835,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003048945475402471,
      "loss": 5.3073,
      "mean_token_accuracy": 0.19548185765743256,
      "num_tokens": 107213195.0,
      "step": 52900
    },
    {
      "entropy": 5.8157860279083256,
      "epoch": 4.62335051996854,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003048636624240119,
      "loss": 5.2114,
      "mean_token_accuracy": 0.21736407577991484,
      "num_tokens": 107222178.0,
      "step": 52905
    },
    {
      "entropy": 5.887387466430664,
      "epoch": 4.623787468321244,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030483277673497603,
      "loss": 5.3066,
      "mean_token_accuracy": 0.20761168748140335,
      "num_tokens": 107232982.0,
      "step": 52910
    },
    {
      "entropy": 5.836064577102661,
      "epoch": 4.6242244166739495,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030480189047373186,
      "loss": 5.2017,
      "mean_token_accuracy": 0.2100437745451927,
      "num_tokens": 107241851.0,
      "step": 52915
    },
    {
      "entropy": 5.854775094985962,
      "epoch": 4.624661365026654,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003047710036408718,
      "loss": 5.2124,
      "mean_token_accuracy": 0.20583388060331345,
      "num_tokens": 107251170.0,
      "step": 52920
    },
    {
      "entropy": 5.927915859222412,
      "epoch": 4.625098313379358,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030474011623698826,
      "loss": 5.3239,
      "mean_token_accuracy": 0.2084981769323349,
      "num_tokens": 107260944.0,
      "step": 52925
    },
    {
      "entropy": 5.895732164382935,
      "epoch": 4.625535261732063,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003047092282626738,
      "loss": 5.3184,
      "mean_token_accuracy": 0.2037600964307785,
      "num_tokens": 107271181.0,
      "step": 52930
    },
    {
      "entropy": 5.828516149520874,
      "epoch": 4.625972210084768,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003046783397185207,
      "loss": 5.3135,
      "mean_token_accuracy": 0.20968573689460754,
      "num_tokens": 107280809.0,
      "step": 52935
    },
    {
      "entropy": 5.816805839538574,
      "epoch": 4.626409158437473,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003046474506051214,
      "loss": 5.1913,
      "mean_token_accuracy": 0.21158453375101088,
      "num_tokens": 107290653.0,
      "step": 52940
    },
    {
      "entropy": 5.901458311080932,
      "epoch": 4.626846106790177,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003046165609230686,
      "loss": 5.3113,
      "mean_token_accuracy": 0.20427082031965255,
      "num_tokens": 107300518.0,
      "step": 52945
    },
    {
      "entropy": 5.7963282585144045,
      "epoch": 4.6272830551428825,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003045856706729547,
      "loss": 5.1823,
      "mean_token_accuracy": 0.21761911362409592,
      "num_tokens": 107310514.0,
      "step": 52950
    },
    {
      "entropy": 5.825556087493896,
      "epoch": 4.627720003495587,
      "grad_norm": 1.109375,
      "learning_rate": 0.000304554779855372,
      "loss": 5.3189,
      "mean_token_accuracy": 0.2099808305501938,
      "num_tokens": 107320383.0,
      "step": 52955
    },
    {
      "entropy": 5.876955795288086,
      "epoch": 4.6281569518482915,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030452388847091317,
      "loss": 5.3079,
      "mean_token_accuracy": 0.20981258153915405,
      "num_tokens": 107331228.0,
      "step": 52960
    },
    {
      "entropy": 5.8475090026855465,
      "epoch": 4.628593900200996,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003044929965201707,
      "loss": 5.2606,
      "mean_token_accuracy": 0.20787498354911804,
      "num_tokens": 107340543.0,
      "step": 52965
    },
    {
      "entropy": 5.789859056472778,
      "epoch": 4.629030848553701,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003044621040037371,
      "loss": 5.2903,
      "mean_token_accuracy": 0.21583101749420167,
      "num_tokens": 107351657.0,
      "step": 52970
    },
    {
      "entropy": 5.94443531036377,
      "epoch": 4.629467796906406,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030443121092220486,
      "loss": 5.3957,
      "mean_token_accuracy": 0.19547804743051528,
      "num_tokens": 107361630.0,
      "step": 52975
    },
    {
      "entropy": 5.92670955657959,
      "epoch": 4.62990474525911,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003044003172761667,
      "loss": 5.2978,
      "mean_token_accuracy": 0.20259809941053392,
      "num_tokens": 107371488.0,
      "step": 52980
    },
    {
      "entropy": 5.81478910446167,
      "epoch": 4.630341693611815,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030436942306621495,
      "loss": 5.2581,
      "mean_token_accuracy": 0.20595831125974656,
      "num_tokens": 107381906.0,
      "step": 52985
    },
    {
      "entropy": 5.863779497146607,
      "epoch": 4.63077864196452,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003043385282929423,
      "loss": 5.3256,
      "mean_token_accuracy": 0.20642978250980376,
      "num_tokens": 107392488.0,
      "step": 52990
    },
    {
      "entropy": 5.868684387207031,
      "epoch": 4.6312155903172245,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003043076329569412,
      "loss": 5.208,
      "mean_token_accuracy": 0.21297542452812196,
      "num_tokens": 107402479.0,
      "step": 52995
    },
    {
      "entropy": 5.878337621688843,
      "epoch": 4.631652538669929,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030427673705880453,
      "loss": 5.4229,
      "mean_token_accuracy": 0.19912266880273818,
      "num_tokens": 107412870.0,
      "step": 53000
    },
    {
      "entropy": 5.910529136657715,
      "epoch": 4.632089487022634,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003042458405991246,
      "loss": 5.2589,
      "mean_token_accuracy": 0.2104918211698532,
      "num_tokens": 107422923.0,
      "step": 53005
    },
    {
      "entropy": 5.820309162139893,
      "epoch": 4.632526435375339,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003042149435784942,
      "loss": 5.2613,
      "mean_token_accuracy": 0.2059025287628174,
      "num_tokens": 107432233.0,
      "step": 53010
    },
    {
      "entropy": 5.809059381484985,
      "epoch": 4.632963383728043,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030418404599750576,
      "loss": 5.2504,
      "mean_token_accuracy": 0.21145391315221787,
      "num_tokens": 107442838.0,
      "step": 53015
    },
    {
      "entropy": 5.845526885986328,
      "epoch": 4.633400332080748,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00030415314785675206,
      "loss": 5.2921,
      "mean_token_accuracy": 0.2062000721693039,
      "num_tokens": 107453542.0,
      "step": 53020
    },
    {
      "entropy": 5.847106504440307,
      "epoch": 4.633837280433453,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003041222491568258,
      "loss": 5.2667,
      "mean_token_accuracy": 0.21496604681015014,
      "num_tokens": 107463139.0,
      "step": 53025
    },
    {
      "entropy": 5.815287494659424,
      "epoch": 4.634274228786158,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030409134989831945,
      "loss": 5.26,
      "mean_token_accuracy": 0.20599113702774047,
      "num_tokens": 107474213.0,
      "step": 53030
    },
    {
      "entropy": 5.901761627197265,
      "epoch": 4.634711177138862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030406045008182574,
      "loss": 5.3466,
      "mean_token_accuracy": 0.20299067795276643,
      "num_tokens": 107485317.0,
      "step": 53035
    },
    {
      "entropy": 5.950935506820679,
      "epoch": 4.6351481254915665,
      "grad_norm": 1.125,
      "learning_rate": 0.00030402954970793747,
      "loss": 5.3136,
      "mean_token_accuracy": 0.20581701695919036,
      "num_tokens": 107494460.0,
      "step": 53040
    },
    {
      "entropy": 5.849438714981079,
      "epoch": 4.635585073844272,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003039986487772471,
      "loss": 5.2603,
      "mean_token_accuracy": 0.20625395625829696,
      "num_tokens": 107504042.0,
      "step": 53045
    },
    {
      "entropy": 5.845937871932984,
      "epoch": 4.636022022196976,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003039677472903476,
      "loss": 5.2721,
      "mean_token_accuracy": 0.20742824524641038,
      "num_tokens": 107513642.0,
      "step": 53050
    },
    {
      "entropy": 5.888844633102417,
      "epoch": 4.636458970549681,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030393684524783135,
      "loss": 5.2964,
      "mean_token_accuracy": 0.20011549592018127,
      "num_tokens": 107523060.0,
      "step": 53055
    },
    {
      "entropy": 5.822275066375733,
      "epoch": 4.636895918902386,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030390594265029136,
      "loss": 5.264,
      "mean_token_accuracy": 0.21745940446853637,
      "num_tokens": 107533238.0,
      "step": 53060
    },
    {
      "entropy": 5.689874696731567,
      "epoch": 4.637332867255091,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00030387503949832025,
      "loss": 5.1591,
      "mean_token_accuracy": 0.2173513650894165,
      "num_tokens": 107543076.0,
      "step": 53065
    },
    {
      "entropy": 5.83575177192688,
      "epoch": 4.637769815607795,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030384413579251075,
      "loss": 5.3094,
      "mean_token_accuracy": 0.20973700135946274,
      "num_tokens": 107554288.0,
      "step": 53070
    },
    {
      "entropy": 5.857014465332031,
      "epoch": 4.6382067639605,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003038132315334556,
      "loss": 5.2088,
      "mean_token_accuracy": 0.2080821380019188,
      "num_tokens": 107564346.0,
      "step": 53075
    },
    {
      "entropy": 5.840096282958984,
      "epoch": 4.638643712313205,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003037823267217475,
      "loss": 5.213,
      "mean_token_accuracy": 0.21004650890827178,
      "num_tokens": 107573941.0,
      "step": 53080
    },
    {
      "entropy": 5.78811616897583,
      "epoch": 4.639080660665909,
      "grad_norm": 1.125,
      "learning_rate": 0.0003037514213579793,
      "loss": 5.2896,
      "mean_token_accuracy": 0.20334740430116655,
      "num_tokens": 107583812.0,
      "step": 53085
    },
    {
      "entropy": 5.857134056091309,
      "epoch": 4.639517609018614,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003037205154427439,
      "loss": 5.2597,
      "mean_token_accuracy": 0.21155748069286345,
      "num_tokens": 107593377.0,
      "step": 53090
    },
    {
      "entropy": 5.809724760055542,
      "epoch": 4.639954557371318,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030368960897663393,
      "loss": 5.3051,
      "mean_token_accuracy": 0.22053600698709488,
      "num_tokens": 107604765.0,
      "step": 53095
    },
    {
      "entropy": 5.902327060699463,
      "epoch": 4.640391505724024,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030365870196024216,
      "loss": 5.3235,
      "mean_token_accuracy": 0.20422317683696747,
      "num_tokens": 107615976.0,
      "step": 53100
    },
    {
      "entropy": 5.881492042541504,
      "epoch": 4.640828454076728,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003036277943941615,
      "loss": 5.2267,
      "mean_token_accuracy": 0.21471490412950517,
      "num_tokens": 107625098.0,
      "step": 53105
    },
    {
      "entropy": 5.80077805519104,
      "epoch": 4.641265402429433,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030359688627898466,
      "loss": 5.2608,
      "mean_token_accuracy": 0.20750023424625397,
      "num_tokens": 107634488.0,
      "step": 53110
    },
    {
      "entropy": 5.842446613311767,
      "epoch": 4.641702350782138,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003035659776153047,
      "loss": 5.2763,
      "mean_token_accuracy": 0.20732713490724564,
      "num_tokens": 107644030.0,
      "step": 53115
    },
    {
      "entropy": 5.858380699157715,
      "epoch": 4.6421392991348425,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003035350684037143,
      "loss": 5.2054,
      "mean_token_accuracy": 0.21362187713384628,
      "num_tokens": 107653790.0,
      "step": 53120
    },
    {
      "entropy": 5.846860885620117,
      "epoch": 4.642576247487547,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003035041586448063,
      "loss": 5.2008,
      "mean_token_accuracy": 0.20443208366632462,
      "num_tokens": 107664133.0,
      "step": 53125
    },
    {
      "entropy": 5.828344345092773,
      "epoch": 4.643013195840251,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030347324833917357,
      "loss": 5.2868,
      "mean_token_accuracy": 0.20591903179883958,
      "num_tokens": 107673259.0,
      "step": 53130
    },
    {
      "entropy": 5.935131549835205,
      "epoch": 4.643450144192957,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000303442337487409,
      "loss": 5.3178,
      "mean_token_accuracy": 0.20248263329267502,
      "num_tokens": 107683250.0,
      "step": 53135
    },
    {
      "entropy": 5.842086887359619,
      "epoch": 4.643887092545661,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003034114260901056,
      "loss": 5.2007,
      "mean_token_accuracy": 0.2142081767320633,
      "num_tokens": 107692166.0,
      "step": 53140
    },
    {
      "entropy": 5.801650285720825,
      "epoch": 4.644324040898366,
      "grad_norm": 1.125,
      "learning_rate": 0.0003033805141478561,
      "loss": 5.2811,
      "mean_token_accuracy": 0.21194937974214553,
      "num_tokens": 107701906.0,
      "step": 53145
    },
    {
      "entropy": 5.839269495010376,
      "epoch": 4.64476098925107,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030334960166125344,
      "loss": 5.3052,
      "mean_token_accuracy": 0.20916569232940674,
      "num_tokens": 107712139.0,
      "step": 53150
    },
    {
      "entropy": 5.948177862167358,
      "epoch": 4.6451979376037755,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030331868863089055,
      "loss": 5.2868,
      "mean_token_accuracy": 0.20285044759511947,
      "num_tokens": 107721836.0,
      "step": 53155
    },
    {
      "entropy": 5.88347020149231,
      "epoch": 4.64563488595648,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030328777505736037,
      "loss": 5.2897,
      "mean_token_accuracy": 0.20625058561563492,
      "num_tokens": 107731682.0,
      "step": 53160
    },
    {
      "entropy": 5.842969512939453,
      "epoch": 4.6460718343091845,
      "grad_norm": 1.125,
      "learning_rate": 0.0003032568609412559,
      "loss": 5.2455,
      "mean_token_accuracy": 0.2161852791905403,
      "num_tokens": 107741870.0,
      "step": 53165
    },
    {
      "entropy": 5.84500641822815,
      "epoch": 4.64650878266189,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030322594628317005,
      "loss": 5.2816,
      "mean_token_accuracy": 0.2116066962480545,
      "num_tokens": 107752507.0,
      "step": 53170
    },
    {
      "entropy": 5.875972032546997,
      "epoch": 4.646945731014594,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003031950310836957,
      "loss": 5.2797,
      "mean_token_accuracy": 0.21060299426317214,
      "num_tokens": 107762842.0,
      "step": 53175
    },
    {
      "entropy": 5.856182098388672,
      "epoch": 4.647382679367299,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003031641153434259,
      "loss": 5.2691,
      "mean_token_accuracy": 0.20477842837572097,
      "num_tokens": 107771796.0,
      "step": 53180
    },
    {
      "entropy": 5.827564287185669,
      "epoch": 4.647819627720003,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003031331990629536,
      "loss": 5.3218,
      "mean_token_accuracy": 0.20841550827026367,
      "num_tokens": 107781264.0,
      "step": 53185
    },
    {
      "entropy": 5.845240831375122,
      "epoch": 4.648256576072709,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003031022822428719,
      "loss": 5.2826,
      "mean_token_accuracy": 0.2094491720199585,
      "num_tokens": 107791405.0,
      "step": 53190
    },
    {
      "entropy": 5.945262861251831,
      "epoch": 4.648693524425413,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030307136488377356,
      "loss": 5.3539,
      "mean_token_accuracy": 0.19890140742063522,
      "num_tokens": 107802168.0,
      "step": 53195
    },
    {
      "entropy": 5.910614728927612,
      "epoch": 4.6491304727781175,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003030404469862518,
      "loss": 5.3383,
      "mean_token_accuracy": 0.20124973356723785,
      "num_tokens": 107811900.0,
      "step": 53200
    },
    {
      "entropy": 5.856690359115601,
      "epoch": 4.649567421130822,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003030095285508995,
      "loss": 5.314,
      "mean_token_accuracy": 0.20257431119680405,
      "num_tokens": 107821906.0,
      "step": 53205
    },
    {
      "entropy": 5.8716590881347654,
      "epoch": 4.650004369483527,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030297860957830984,
      "loss": 5.2834,
      "mean_token_accuracy": 0.20088582932949067,
      "num_tokens": 107831268.0,
      "step": 53210
    },
    {
      "entropy": 5.9239274024963375,
      "epoch": 4.650441317836232,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030294769006907585,
      "loss": 5.3704,
      "mean_token_accuracy": 0.20287147015333176,
      "num_tokens": 107841284.0,
      "step": 53215
    },
    {
      "entropy": 5.905216264724731,
      "epoch": 4.650878266188936,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003029167700237904,
      "loss": 5.3048,
      "mean_token_accuracy": 0.2083453431725502,
      "num_tokens": 107853119.0,
      "step": 53220
    },
    {
      "entropy": 5.890584516525268,
      "epoch": 4.651315214541642,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030288584944304674,
      "loss": 5.2647,
      "mean_token_accuracy": 0.2118932858109474,
      "num_tokens": 107864162.0,
      "step": 53225
    },
    {
      "entropy": 5.856568431854248,
      "epoch": 4.651752162894346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003028549283274378,
      "loss": 5.245,
      "mean_token_accuracy": 0.21333284825086593,
      "num_tokens": 107874921.0,
      "step": 53230
    },
    {
      "entropy": 5.852432727813721,
      "epoch": 4.652189111247051,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003028240066775568,
      "loss": 5.3297,
      "mean_token_accuracy": 0.2001931831240654,
      "num_tokens": 107885841.0,
      "step": 53235
    },
    {
      "entropy": 5.878617286682129,
      "epoch": 4.652626059599755,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003027930844939968,
      "loss": 5.2954,
      "mean_token_accuracy": 0.20386195480823516,
      "num_tokens": 107896651.0,
      "step": 53240
    },
    {
      "entropy": 5.885091972351074,
      "epoch": 4.6530630079524595,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00030276216177735083,
      "loss": 5.3243,
      "mean_token_accuracy": 0.20358554720878602,
      "num_tokens": 107907540.0,
      "step": 53245
    },
    {
      "entropy": 5.900114440917969,
      "epoch": 4.653499956305165,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030273123852821203,
      "loss": 5.2314,
      "mean_token_accuracy": 0.20673041492700578,
      "num_tokens": 107916894.0,
      "step": 53250
    },
    {
      "entropy": 5.85543007850647,
      "epoch": 4.653936904657869,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030270031474717357,
      "loss": 5.3545,
      "mean_token_accuracy": 0.19938847422599792,
      "num_tokens": 107927160.0,
      "step": 53255
    },
    {
      "entropy": 5.872903203964233,
      "epoch": 4.654373853010574,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003026693904348286,
      "loss": 5.347,
      "mean_token_accuracy": 0.19890491515398026,
      "num_tokens": 107938511.0,
      "step": 53260
    },
    {
      "entropy": 5.866177701950074,
      "epoch": 4.654810801363279,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003026384655917702,
      "loss": 5.2652,
      "mean_token_accuracy": 0.20873110890388488,
      "num_tokens": 107948942.0,
      "step": 53265
    },
    {
      "entropy": 5.909026145935059,
      "epoch": 4.655247749715984,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030260754021859147,
      "loss": 5.337,
      "mean_token_accuracy": 0.20426613539457322,
      "num_tokens": 107959768.0,
      "step": 53270
    },
    {
      "entropy": 5.912074375152588,
      "epoch": 4.655684698068688,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003025766143158858,
      "loss": 5.2922,
      "mean_token_accuracy": 0.20736121237277985,
      "num_tokens": 107970956.0,
      "step": 53275
    },
    {
      "entropy": 5.903384780883789,
      "epoch": 4.6561216464213935,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003025456878842461,
      "loss": 5.3101,
      "mean_token_accuracy": 0.211184960603714,
      "num_tokens": 107980786.0,
      "step": 53280
    },
    {
      "entropy": 5.880833625793457,
      "epoch": 4.656558594774098,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030251476092426575,
      "loss": 5.3675,
      "mean_token_accuracy": 0.20007702112197875,
      "num_tokens": 107990578.0,
      "step": 53285
    },
    {
      "entropy": 5.896550559997559,
      "epoch": 4.656995543126802,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003024838334365378,
      "loss": 5.3509,
      "mean_token_accuracy": 0.2034713387489319,
      "num_tokens": 108000417.0,
      "step": 53290
    },
    {
      "entropy": 5.821872282028198,
      "epoch": 4.657432491479507,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030245290542165555,
      "loss": 5.2894,
      "mean_token_accuracy": 0.2067921668291092,
      "num_tokens": 108010145.0,
      "step": 53295
    },
    {
      "entropy": 5.837266302108764,
      "epoch": 4.657869439832211,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003024219768802122,
      "loss": 5.2783,
      "mean_token_accuracy": 0.20652322620153427,
      "num_tokens": 108020510.0,
      "step": 53300
    },
    {
      "entropy": 5.881914472579956,
      "epoch": 4.658306388184917,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003023910478128009,
      "loss": 5.3233,
      "mean_token_accuracy": 0.20659493654966354,
      "num_tokens": 108029763.0,
      "step": 53305
    },
    {
      "entropy": 5.961287355422973,
      "epoch": 4.658743336537621,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030236011822001505,
      "loss": 5.3881,
      "mean_token_accuracy": 0.19644173830747605,
      "num_tokens": 108039944.0,
      "step": 53310
    },
    {
      "entropy": 5.919487857818604,
      "epoch": 4.659180284890326,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003023291881024477,
      "loss": 5.3435,
      "mean_token_accuracy": 0.2045857235789299,
      "num_tokens": 108051030.0,
      "step": 53315
    },
    {
      "entropy": 5.83052339553833,
      "epoch": 4.659617233243031,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030229825746069225,
      "loss": 5.2245,
      "mean_token_accuracy": 0.21871947795152663,
      "num_tokens": 108061537.0,
      "step": 53320
    },
    {
      "entropy": 5.838040590286255,
      "epoch": 4.6600541815957355,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003022673262953419,
      "loss": 5.2415,
      "mean_token_accuracy": 0.2132135435938835,
      "num_tokens": 108072142.0,
      "step": 53325
    },
    {
      "entropy": 5.8518165111541744,
      "epoch": 4.66049112994844,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030223639460699,
      "loss": 5.3164,
      "mean_token_accuracy": 0.1997606188058853,
      "num_tokens": 108083201.0,
      "step": 53330
    },
    {
      "entropy": 5.840442657470703,
      "epoch": 4.660928078301144,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003022054623962297,
      "loss": 5.269,
      "mean_token_accuracy": 0.20848754197359085,
      "num_tokens": 108093175.0,
      "step": 53335
    },
    {
      "entropy": 5.85776801109314,
      "epoch": 4.66136502665385,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030217452966365446,
      "loss": 5.2403,
      "mean_token_accuracy": 0.21162635684013367,
      "num_tokens": 108103918.0,
      "step": 53340
    },
    {
      "entropy": 5.832322883605957,
      "epoch": 4.661801975006554,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003021435964098574,
      "loss": 5.2918,
      "mean_token_accuracy": 0.2103957414627075,
      "num_tokens": 108115060.0,
      "step": 53345
    },
    {
      "entropy": 5.816421175003052,
      "epoch": 4.662238923359259,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030211266263543204,
      "loss": 5.2095,
      "mean_token_accuracy": 0.2159522444009781,
      "num_tokens": 108124262.0,
      "step": 53350
    },
    {
      "entropy": 5.822730636596679,
      "epoch": 4.662675871711963,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030208172834097157,
      "loss": 5.2437,
      "mean_token_accuracy": 0.2130796954035759,
      "num_tokens": 108132852.0,
      "step": 53355
    },
    {
      "entropy": 5.793469619750977,
      "epoch": 4.6631128200646685,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030205079352706935,
      "loss": 5.2419,
      "mean_token_accuracy": 0.21184832900762557,
      "num_tokens": 108142533.0,
      "step": 53360
    },
    {
      "entropy": 5.819828128814697,
      "epoch": 4.663549768417373,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003020198581943188,
      "loss": 5.2482,
      "mean_token_accuracy": 0.20965854227542877,
      "num_tokens": 108151781.0,
      "step": 53365
    },
    {
      "entropy": 5.9249590873718265,
      "epoch": 4.6639867167700775,
      "grad_norm": 1.0,
      "learning_rate": 0.0003019889223433131,
      "loss": 5.3531,
      "mean_token_accuracy": 0.19750808030366898,
      "num_tokens": 108162005.0,
      "step": 53370
    },
    {
      "entropy": 5.896938037872315,
      "epoch": 4.664423665122783,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030195798597464573,
      "loss": 5.3667,
      "mean_token_accuracy": 0.20242626816034318,
      "num_tokens": 108172009.0,
      "step": 53375
    },
    {
      "entropy": 5.8971813201904295,
      "epoch": 4.664860613475487,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003019270490889101,
      "loss": 5.3268,
      "mean_token_accuracy": 0.21058740168809892,
      "num_tokens": 108181584.0,
      "step": 53380
    },
    {
      "entropy": 5.9085102558135985,
      "epoch": 4.665297561828192,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003018961116866996,
      "loss": 5.3111,
      "mean_token_accuracy": 0.20292955189943312,
      "num_tokens": 108190846.0,
      "step": 53385
    },
    {
      "entropy": 5.881638431549073,
      "epoch": 4.665734510180896,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030186517376860753,
      "loss": 5.2399,
      "mean_token_accuracy": 0.20916635692119598,
      "num_tokens": 108199650.0,
      "step": 53390
    },
    {
      "entropy": 5.905157899856567,
      "epoch": 4.666171458533602,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003018342353352274,
      "loss": 5.3851,
      "mean_token_accuracy": 0.2031741186976433,
      "num_tokens": 108210294.0,
      "step": 53395
    },
    {
      "entropy": 5.929882955551148,
      "epoch": 4.666608406886306,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030180329638715246,
      "loss": 5.3385,
      "mean_token_accuracy": 0.20260522663593292,
      "num_tokens": 108218981.0,
      "step": 53400
    },
    {
      "entropy": 5.858361911773682,
      "epoch": 4.6670453552390105,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003017723569249764,
      "loss": 5.2657,
      "mean_token_accuracy": 0.21043899655342102,
      "num_tokens": 108229613.0,
      "step": 53405
    },
    {
      "entropy": 5.816173362731933,
      "epoch": 4.667482303591715,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030174141694929243,
      "loss": 5.2868,
      "mean_token_accuracy": 0.21066283881664277,
      "num_tokens": 108239269.0,
      "step": 53410
    },
    {
      "entropy": 5.829517412185669,
      "epoch": 4.66791925194442,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030171047646069404,
      "loss": 5.2783,
      "mean_token_accuracy": 0.20313953906297683,
      "num_tokens": 108248942.0,
      "step": 53415
    },
    {
      "entropy": 5.9046307563781735,
      "epoch": 4.668356200297125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003016795354597748,
      "loss": 5.3042,
      "mean_token_accuracy": 0.20059469640254973,
      "num_tokens": 108260479.0,
      "step": 53420
    },
    {
      "entropy": 5.835871410369873,
      "epoch": 4.668793148649829,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003016485939471281,
      "loss": 5.291,
      "mean_token_accuracy": 0.20746456384658812,
      "num_tokens": 108269530.0,
      "step": 53425
    },
    {
      "entropy": 5.86724009513855,
      "epoch": 4.669230097002535,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030161765192334735,
      "loss": 5.2953,
      "mean_token_accuracy": 0.21752146631479263,
      "num_tokens": 108280650.0,
      "step": 53430
    },
    {
      "entropy": 5.900255250930786,
      "epoch": 4.669667045355239,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030158670938902605,
      "loss": 5.2504,
      "mean_token_accuracy": 0.207812337577343,
      "num_tokens": 108290266.0,
      "step": 53435
    },
    {
      "entropy": 5.773008632659912,
      "epoch": 4.670103993707944,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003015557663447578,
      "loss": 5.2215,
      "mean_token_accuracy": 0.21886873990297318,
      "num_tokens": 108300911.0,
      "step": 53440
    },
    {
      "entropy": 5.782088565826416,
      "epoch": 4.670540942060648,
      "grad_norm": 1.046875,
      "learning_rate": 0.000301524822791136,
      "loss": 5.1829,
      "mean_token_accuracy": 0.21683214008808135,
      "num_tokens": 108311000.0,
      "step": 53445
    },
    {
      "entropy": 5.855199861526489,
      "epoch": 4.670977890413353,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003014938787287542,
      "loss": 5.2734,
      "mean_token_accuracy": 0.20853521525859833,
      "num_tokens": 108321817.0,
      "step": 53450
    },
    {
      "entropy": 5.9526465892791744,
      "epoch": 4.671414838766058,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030146293415820593,
      "loss": 5.4008,
      "mean_token_accuracy": 0.19617196768522263,
      "num_tokens": 108332917.0,
      "step": 53455
    },
    {
      "entropy": 5.87481164932251,
      "epoch": 4.671851787118762,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003014319890800847,
      "loss": 5.2116,
      "mean_token_accuracy": 0.20943575948476792,
      "num_tokens": 108344013.0,
      "step": 53460
    },
    {
      "entropy": 5.9126183032989506,
      "epoch": 4.672288735471467,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003014010434949841,
      "loss": 5.4037,
      "mean_token_accuracy": 0.19809839129447937,
      "num_tokens": 108353849.0,
      "step": 53465
    },
    {
      "entropy": 5.936564636230469,
      "epoch": 4.672725683824172,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030137009740349767,
      "loss": 5.3445,
      "mean_token_accuracy": 0.20283572226762772,
      "num_tokens": 108363437.0,
      "step": 53470
    },
    {
      "entropy": 5.82892107963562,
      "epoch": 4.673162632176877,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030133915080621894,
      "loss": 5.293,
      "mean_token_accuracy": 0.2102851465344429,
      "num_tokens": 108372477.0,
      "step": 53475
    },
    {
      "entropy": 5.880655765533447,
      "epoch": 4.673599580529581,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030130820370374145,
      "loss": 5.3056,
      "mean_token_accuracy": 0.20615922659635544,
      "num_tokens": 108382693.0,
      "step": 53480
    },
    {
      "entropy": 5.8763525009155275,
      "epoch": 4.6740365288822865,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003012772560966588,
      "loss": 5.2008,
      "mean_token_accuracy": 0.21859612464904785,
      "num_tokens": 108392044.0,
      "step": 53485
    },
    {
      "entropy": 5.805232381820678,
      "epoch": 4.674473477234991,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00030124630798556475,
      "loss": 5.2006,
      "mean_token_accuracy": 0.2120540514588356,
      "num_tokens": 108401541.0,
      "step": 53490
    },
    {
      "entropy": 5.782459115982055,
      "epoch": 4.674910425587695,
      "grad_norm": 1.0,
      "learning_rate": 0.0003012153593710527,
      "loss": 5.2045,
      "mean_token_accuracy": 0.2129358932375908,
      "num_tokens": 108412337.0,
      "step": 53495
    },
    {
      "entropy": 5.871257162094116,
      "epoch": 4.6753473739404,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003011844102537163,
      "loss": 5.298,
      "mean_token_accuracy": 0.20486894994974136,
      "num_tokens": 108421965.0,
      "step": 53500
    },
    {
      "entropy": 5.774171638488769,
      "epoch": 4.675784322293105,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003011534606341492,
      "loss": 5.2045,
      "mean_token_accuracy": 0.2238602429628372,
      "num_tokens": 108432380.0,
      "step": 53505
    },
    {
      "entropy": 5.794019746780395,
      "epoch": 4.67622127064581,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000301122510512945,
      "loss": 5.1895,
      "mean_token_accuracy": 0.2165067732334137,
      "num_tokens": 108441893.0,
      "step": 53510
    },
    {
      "entropy": 5.801884031295776,
      "epoch": 4.676658218998514,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003010915598906973,
      "loss": 5.2135,
      "mean_token_accuracy": 0.218349190056324,
      "num_tokens": 108452188.0,
      "step": 53515
    },
    {
      "entropy": 5.847303104400635,
      "epoch": 4.677095167351219,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030106060876799995,
      "loss": 5.2514,
      "mean_token_accuracy": 0.21708798110485078,
      "num_tokens": 108462071.0,
      "step": 53520
    },
    {
      "entropy": 5.772377061843872,
      "epoch": 4.677532115703924,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003010296571454464,
      "loss": 5.1838,
      "mean_token_accuracy": 0.2211538940668106,
      "num_tokens": 108470988.0,
      "step": 53525
    },
    {
      "entropy": 5.874396324157715,
      "epoch": 4.6779690640566285,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030099870502363037,
      "loss": 5.2952,
      "mean_token_accuracy": 0.20744970738887786,
      "num_tokens": 108480534.0,
      "step": 53530
    },
    {
      "entropy": 5.903978157043457,
      "epoch": 4.678406012409333,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003009677524031456,
      "loss": 5.3396,
      "mean_token_accuracy": 0.2075077176094055,
      "num_tokens": 108491311.0,
      "step": 53535
    },
    {
      "entropy": 5.947551012039185,
      "epoch": 4.678842960762038,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003009367992845857,
      "loss": 5.3009,
      "mean_token_accuracy": 0.21492071002721785,
      "num_tokens": 108501407.0,
      "step": 53540
    },
    {
      "entropy": 5.88347020149231,
      "epoch": 4.679279909114743,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003009058456685444,
      "loss": 5.3612,
      "mean_token_accuracy": 0.19442747831344603,
      "num_tokens": 108511149.0,
      "step": 53545
    },
    {
      "entropy": 5.85161190032959,
      "epoch": 4.679716857467447,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003008748915556154,
      "loss": 5.3019,
      "mean_token_accuracy": 0.20619666427373887,
      "num_tokens": 108521224.0,
      "step": 53550
    },
    {
      "entropy": 5.893539333343506,
      "epoch": 4.680153805820152,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003008439369463924,
      "loss": 5.265,
      "mean_token_accuracy": 0.20555381774902343,
      "num_tokens": 108530860.0,
      "step": 53555
    },
    {
      "entropy": 5.836408424377441,
      "epoch": 4.680590754172857,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003008129818414692,
      "loss": 5.334,
      "mean_token_accuracy": 0.20480705350637435,
      "num_tokens": 108540460.0,
      "step": 53560
    },
    {
      "entropy": 5.916872930526734,
      "epoch": 4.6810277025255616,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030078202624143945,
      "loss": 5.296,
      "mean_token_accuracy": 0.20409023761749268,
      "num_tokens": 108550588.0,
      "step": 53565
    },
    {
      "entropy": 5.9089573383331295,
      "epoch": 4.681464650878266,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030075107014689695,
      "loss": 5.3228,
      "mean_token_accuracy": 0.20486582666635514,
      "num_tokens": 108561107.0,
      "step": 53570
    },
    {
      "entropy": 5.891446352005005,
      "epoch": 4.6819015992309705,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003007201135584354,
      "loss": 5.3143,
      "mean_token_accuracy": 0.20348425954580307,
      "num_tokens": 108571834.0,
      "step": 53575
    },
    {
      "entropy": 5.908519840240478,
      "epoch": 4.682338547583676,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003006891564766486,
      "loss": 5.3357,
      "mean_token_accuracy": 0.20368096232414246,
      "num_tokens": 108581264.0,
      "step": 53580
    },
    {
      "entropy": 5.953167057037353,
      "epoch": 4.68277549593638,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030065819890213037,
      "loss": 5.3907,
      "mean_token_accuracy": 0.19763894379138947,
      "num_tokens": 108591302.0,
      "step": 53585
    },
    {
      "entropy": 5.865368032455445,
      "epoch": 4.683212444289085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030062724083547437,
      "loss": 5.2779,
      "mean_token_accuracy": 0.21098823845386505,
      "num_tokens": 108601834.0,
      "step": 53590
    },
    {
      "entropy": 5.897441959381103,
      "epoch": 4.68364939264179,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003005962822772745,
      "loss": 5.3628,
      "mean_token_accuracy": 0.20100731104612352,
      "num_tokens": 108612455.0,
      "step": 53595
    },
    {
      "entropy": 5.892339372634888,
      "epoch": 4.684086340994495,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003005653232281245,
      "loss": 5.275,
      "mean_token_accuracy": 0.20812000036239625,
      "num_tokens": 108622999.0,
      "step": 53600
    },
    {
      "entropy": 5.854294490814209,
      "epoch": 4.684523289347199,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003005343636886182,
      "loss": 5.2853,
      "mean_token_accuracy": 0.21017930805683135,
      "num_tokens": 108633710.0,
      "step": 53605
    },
    {
      "entropy": 5.809257125854492,
      "epoch": 4.6849602376999036,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003005034036593495,
      "loss": 5.2132,
      "mean_token_accuracy": 0.2025788977742195,
      "num_tokens": 108644128.0,
      "step": 53610
    },
    {
      "entropy": 5.876645755767822,
      "epoch": 4.685397186052609,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003004724431409121,
      "loss": 5.2426,
      "mean_token_accuracy": 0.20753360837697982,
      "num_tokens": 108654259.0,
      "step": 53615
    },
    {
      "entropy": 5.879355955123901,
      "epoch": 4.685834134405313,
      "grad_norm": 1.125,
      "learning_rate": 0.0003004414821338999,
      "loss": 5.266,
      "mean_token_accuracy": 0.21129074245691298,
      "num_tokens": 108664835.0,
      "step": 53620
    },
    {
      "entropy": 5.784032344818115,
      "epoch": 4.686271082758018,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003004105206389067,
      "loss": 5.2607,
      "mean_token_accuracy": 0.21074849963188172,
      "num_tokens": 108675083.0,
      "step": 53625
    },
    {
      "entropy": 5.855156946182251,
      "epoch": 4.686708031110722,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00030037955865652645,
      "loss": 5.2765,
      "mean_token_accuracy": 0.20779765546321868,
      "num_tokens": 108685295.0,
      "step": 53630
    },
    {
      "entropy": 5.837475347518921,
      "epoch": 4.687144979463428,
      "grad_norm": 1.078125,
      "learning_rate": 0.000300348596187353,
      "loss": 5.2517,
      "mean_token_accuracy": 0.21530783474445342,
      "num_tokens": 108694957.0,
      "step": 53635
    },
    {
      "entropy": 5.9067059516906735,
      "epoch": 4.687581927816132,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030031763323198007,
      "loss": 5.3503,
      "mean_token_accuracy": 0.20904072672128676,
      "num_tokens": 108705410.0,
      "step": 53640
    },
    {
      "entropy": 5.905647850036621,
      "epoch": 4.688018876168837,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003002866697910018,
      "loss": 5.2683,
      "mean_token_accuracy": 0.21299065500497819,
      "num_tokens": 108715189.0,
      "step": 53645
    },
    {
      "entropy": 5.8940431594848635,
      "epoch": 4.688455824521542,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003002557058650118,
      "loss": 5.3506,
      "mean_token_accuracy": 0.20504713505506517,
      "num_tokens": 108725225.0,
      "step": 53650
    },
    {
      "entropy": 5.914198493957519,
      "epoch": 4.688892772874246,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003002247414546043,
      "loss": 5.3463,
      "mean_token_accuracy": 0.19901479333639144,
      "num_tokens": 108735933.0,
      "step": 53655
    },
    {
      "entropy": 5.932635068893433,
      "epoch": 4.689329721226951,
      "grad_norm": 1.125,
      "learning_rate": 0.00030019377656037295,
      "loss": 5.3061,
      "mean_token_accuracy": 0.2046133831143379,
      "num_tokens": 108745336.0,
      "step": 53660
    },
    {
      "entropy": 5.877285528182983,
      "epoch": 4.689766669579655,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003001628111829118,
      "loss": 5.3153,
      "mean_token_accuracy": 0.19752372056245804,
      "num_tokens": 108756583.0,
      "step": 53665
    },
    {
      "entropy": 5.877013063430786,
      "epoch": 4.690203617932361,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030013184532281465,
      "loss": 5.3186,
      "mean_token_accuracy": 0.20341752767562865,
      "num_tokens": 108766606.0,
      "step": 53670
    },
    {
      "entropy": 5.756172561645508,
      "epoch": 4.690640566285065,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00030010087898067566,
      "loss": 5.2163,
      "mean_token_accuracy": 0.22077204138040543,
      "num_tokens": 108776803.0,
      "step": 53675
    },
    {
      "entropy": 5.840844964981079,
      "epoch": 4.69107751463777,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00030006991215708855,
      "loss": 5.2932,
      "mean_token_accuracy": 0.21061870008707045,
      "num_tokens": 108787033.0,
      "step": 53680
    },
    {
      "entropy": 5.818985462188721,
      "epoch": 4.691514462990474,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030003894485264754,
      "loss": 5.2023,
      "mean_token_accuracy": 0.21773965507745743,
      "num_tokens": 108797593.0,
      "step": 53685
    },
    {
      "entropy": 5.848167371749878,
      "epoch": 4.6919514113431795,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030000797706794633,
      "loss": 5.2685,
      "mean_token_accuracy": 0.20884234607219695,
      "num_tokens": 108807065.0,
      "step": 53690
    },
    {
      "entropy": 5.862428426742554,
      "epoch": 4.692388359695884,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029997700880357906,
      "loss": 5.2368,
      "mean_token_accuracy": 0.21342889368534088,
      "num_tokens": 108816689.0,
      "step": 53695
    },
    {
      "entropy": 5.860279130935669,
      "epoch": 4.692825308048588,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029994604006013966,
      "loss": 5.2622,
      "mean_token_accuracy": 0.21397175192832946,
      "num_tokens": 108826630.0,
      "step": 53700
    },
    {
      "entropy": 5.806341075897217,
      "epoch": 4.693262256401294,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002999150708382221,
      "loss": 5.1897,
      "mean_token_accuracy": 0.22223153263330458,
      "num_tokens": 108836268.0,
      "step": 53705
    },
    {
      "entropy": 5.855404663085937,
      "epoch": 4.693699204753998,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002998841011384205,
      "loss": 5.2485,
      "mean_token_accuracy": 0.2099264606833458,
      "num_tokens": 108845865.0,
      "step": 53710
    },
    {
      "entropy": 5.895630168914795,
      "epoch": 4.694136153106703,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002998531309613288,
      "loss": 5.2721,
      "mean_token_accuracy": 0.207812038064003,
      "num_tokens": 108855628.0,
      "step": 53715
    },
    {
      "entropy": 5.884434080123901,
      "epoch": 4.694573101459407,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029982216030754094,
      "loss": 5.3123,
      "mean_token_accuracy": 0.20897051692008972,
      "num_tokens": 108866323.0,
      "step": 53720
    },
    {
      "entropy": 5.97613000869751,
      "epoch": 4.695010049812113,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002997911891776511,
      "loss": 5.3897,
      "mean_token_accuracy": 0.19672513902187347,
      "num_tokens": 108876783.0,
      "step": 53725
    },
    {
      "entropy": 5.8414712905883786,
      "epoch": 4.695446998164817,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029976021757225317,
      "loss": 5.2529,
      "mean_token_accuracy": 0.21540887653827667,
      "num_tokens": 108887171.0,
      "step": 53730
    },
    {
      "entropy": 5.853727960586548,
      "epoch": 4.6958839465175215,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002997292454919415,
      "loss": 5.2099,
      "mean_token_accuracy": 0.20889489203691483,
      "num_tokens": 108897441.0,
      "step": 53735
    },
    {
      "entropy": 5.868468713760376,
      "epoch": 4.696320894870226,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029969827293730973,
      "loss": 5.239,
      "mean_token_accuracy": 0.20892244726419448,
      "num_tokens": 108906608.0,
      "step": 53740
    },
    {
      "entropy": 5.875193929672241,
      "epoch": 4.696757843222931,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00029966729990895226,
      "loss": 5.312,
      "mean_token_accuracy": 0.2104366034269333,
      "num_tokens": 108917038.0,
      "step": 53745
    },
    {
      "entropy": 5.934405851364136,
      "epoch": 4.697194791575636,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000299636326407463,
      "loss": 5.3536,
      "mean_token_accuracy": 0.19594601541757584,
      "num_tokens": 108926682.0,
      "step": 53750
    },
    {
      "entropy": 5.882221031188965,
      "epoch": 4.69763173992834,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029960535243343616,
      "loss": 5.2837,
      "mean_token_accuracy": 0.2159831702709198,
      "num_tokens": 108936936.0,
      "step": 53755
    },
    {
      "entropy": 5.927259922027588,
      "epoch": 4.698068688281046,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002995743779874657,
      "loss": 5.3128,
      "mean_token_accuracy": 0.209363155066967,
      "num_tokens": 108947635.0,
      "step": 53760
    },
    {
      "entropy": 5.848855066299438,
      "epoch": 4.69850563663375,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002995434030701457,
      "loss": 5.2347,
      "mean_token_accuracy": 0.21671396344900132,
      "num_tokens": 108957085.0,
      "step": 53765
    },
    {
      "entropy": 5.809519815444946,
      "epoch": 4.698942584986455,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002995124276820705,
      "loss": 5.3115,
      "mean_token_accuracy": 0.20156436860561372,
      "num_tokens": 108966639.0,
      "step": 53770
    },
    {
      "entropy": 5.878196859359742,
      "epoch": 4.699379533339159,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029948145182383414,
      "loss": 5.2385,
      "mean_token_accuracy": 0.215505151450634,
      "num_tokens": 108977300.0,
      "step": 53775
    },
    {
      "entropy": 5.866387367248535,
      "epoch": 4.699816481691864,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002994504754960306,
      "loss": 5.3277,
      "mean_token_accuracy": 0.2071644738316536,
      "num_tokens": 108987281.0,
      "step": 53780
    },
    {
      "entropy": 5.89580225944519,
      "epoch": 4.700253430044569,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029941949869925424,
      "loss": 5.2922,
      "mean_token_accuracy": 0.20477039664983748,
      "num_tokens": 108996676.0,
      "step": 53785
    },
    {
      "entropy": 5.8763079166412355,
      "epoch": 4.700690378397273,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029938852143409896,
      "loss": 5.3065,
      "mean_token_accuracy": 0.2116333618760109,
      "num_tokens": 109006633.0,
      "step": 53790
    },
    {
      "entropy": 5.868525505065918,
      "epoch": 4.701127326749978,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029935754370115906,
      "loss": 5.2233,
      "mean_token_accuracy": 0.21168725043535233,
      "num_tokens": 109016876.0,
      "step": 53795
    },
    {
      "entropy": 5.85492959022522,
      "epoch": 4.701564275102683,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029932656550102886,
      "loss": 5.2389,
      "mean_token_accuracy": 0.20729394257068634,
      "num_tokens": 109027107.0,
      "step": 53800
    },
    {
      "entropy": 5.954433155059815,
      "epoch": 4.702001223455388,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002992955868343023,
      "loss": 5.3345,
      "mean_token_accuracy": 0.19551678746938705,
      "num_tokens": 109037809.0,
      "step": 53805
    },
    {
      "entropy": 5.870394325256347,
      "epoch": 4.702438171808092,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002992646077015737,
      "loss": 5.3182,
      "mean_token_accuracy": 0.2000947564840317,
      "num_tokens": 109048913.0,
      "step": 53810
    },
    {
      "entropy": 5.858957958221436,
      "epoch": 4.7028751201607975,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029923362810343714,
      "loss": 5.2445,
      "mean_token_accuracy": 0.20862781405448913,
      "num_tokens": 109059482.0,
      "step": 53815
    },
    {
      "entropy": 5.8850343227386475,
      "epoch": 4.703312068513502,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029920264804048687,
      "loss": 5.3455,
      "mean_token_accuracy": 0.20314404368400574,
      "num_tokens": 109069236.0,
      "step": 53820
    },
    {
      "entropy": 5.879355716705322,
      "epoch": 4.703749016866206,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002991716675133172,
      "loss": 5.3136,
      "mean_token_accuracy": 0.20278285294771195,
      "num_tokens": 109079215.0,
      "step": 53825
    },
    {
      "entropy": 5.8655767917633055,
      "epoch": 4.704185965218911,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029914068652252225,
      "loss": 5.2776,
      "mean_token_accuracy": 0.20657743364572526,
      "num_tokens": 109088695.0,
      "step": 53830
    },
    {
      "entropy": 5.900309467315674,
      "epoch": 4.704622913571616,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029910970506869634,
      "loss": 5.3459,
      "mean_token_accuracy": 0.2019328162074089,
      "num_tokens": 109099203.0,
      "step": 53835
    },
    {
      "entropy": 5.985786628723145,
      "epoch": 4.705059861924321,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029907872315243354,
      "loss": 5.4386,
      "mean_token_accuracy": 0.19239737540483476,
      "num_tokens": 109110917.0,
      "step": 53840
    },
    {
      "entropy": 5.897764205932617,
      "epoch": 4.705496810277025,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029904774077432826,
      "loss": 5.3003,
      "mean_token_accuracy": 0.20467474907636643,
      "num_tokens": 109121277.0,
      "step": 53845
    },
    {
      "entropy": 5.894669771194458,
      "epoch": 4.70593375862973,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002990167579349748,
      "loss": 5.2409,
      "mean_token_accuracy": 0.20915245711803437,
      "num_tokens": 109130485.0,
      "step": 53850
    },
    {
      "entropy": 5.853766345977784,
      "epoch": 4.706370706982435,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002989857746349672,
      "loss": 5.2311,
      "mean_token_accuracy": 0.21299024522304535,
      "num_tokens": 109139872.0,
      "step": 53855
    },
    {
      "entropy": 5.903519821166992,
      "epoch": 4.7068076553351395,
      "grad_norm": 1.0,
      "learning_rate": 0.00029895479087489995,
      "loss": 5.3153,
      "mean_token_accuracy": 0.20413741916418077,
      "num_tokens": 109150004.0,
      "step": 53860
    },
    {
      "entropy": 5.846348714828491,
      "epoch": 4.707244603687844,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029892380665536727,
      "loss": 5.3187,
      "mean_token_accuracy": 0.20399902611970902,
      "num_tokens": 109160078.0,
      "step": 53865
    },
    {
      "entropy": 5.859825468063354,
      "epoch": 4.707681552040549,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002988928219769634,
      "loss": 5.2551,
      "mean_token_accuracy": 0.2074936032295227,
      "num_tokens": 109168829.0,
      "step": 53870
    },
    {
      "entropy": 5.764869546890258,
      "epoch": 4.708118500393254,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029886183684028276,
      "loss": 5.1939,
      "mean_token_accuracy": 0.22241052091121674,
      "num_tokens": 109178909.0,
      "step": 53875
    },
    {
      "entropy": 5.926632022857666,
      "epoch": 4.708555448745958,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002988308512459196,
      "loss": 5.3303,
      "mean_token_accuracy": 0.20549999922513962,
      "num_tokens": 109188156.0,
      "step": 53880
    },
    {
      "entropy": 5.9495666980743405,
      "epoch": 4.708992397098663,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029879986519446814,
      "loss": 5.3692,
      "mean_token_accuracy": 0.20267427414655687,
      "num_tokens": 109198777.0,
      "step": 53885
    },
    {
      "entropy": 5.9063169956207275,
      "epoch": 4.709429345451367,
      "grad_norm": 1.125,
      "learning_rate": 0.00029876887868652283,
      "loss": 5.2843,
      "mean_token_accuracy": 0.20968568623065947,
      "num_tokens": 109207942.0,
      "step": 53890
    },
    {
      "entropy": 5.840945911407471,
      "epoch": 4.7098662938040725,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029873789172267794,
      "loss": 5.2449,
      "mean_token_accuracy": 0.21101270020008087,
      "num_tokens": 109217636.0,
      "step": 53895
    },
    {
      "entropy": 5.861267042160034,
      "epoch": 4.710303242156777,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002987069043035279,
      "loss": 5.2944,
      "mean_token_accuracy": 0.2076340362429619,
      "num_tokens": 109228379.0,
      "step": 53900
    },
    {
      "entropy": 5.807565450668335,
      "epoch": 4.7107401905094815,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029867591642966694,
      "loss": 5.2636,
      "mean_token_accuracy": 0.2173785001039505,
      "num_tokens": 109237750.0,
      "step": 53905
    },
    {
      "entropy": 5.917830181121826,
      "epoch": 4.711177138862187,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029864492810168945,
      "loss": 5.3175,
      "mean_token_accuracy": 0.20665715932846068,
      "num_tokens": 109248411.0,
      "step": 53910
    },
    {
      "entropy": 5.8761146068573,
      "epoch": 4.711614087214891,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029861393932018997,
      "loss": 5.2929,
      "mean_token_accuracy": 0.20844463855028153,
      "num_tokens": 109257495.0,
      "step": 53915
    },
    {
      "entropy": 5.815168857574463,
      "epoch": 4.712051035567596,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002985829500857628,
      "loss": 5.2732,
      "mean_token_accuracy": 0.21342180371284486,
      "num_tokens": 109267003.0,
      "step": 53920
    },
    {
      "entropy": 5.803250694274903,
      "epoch": 4.712487983920301,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029855196039900217,
      "loss": 5.23,
      "mean_token_accuracy": 0.21651377230882646,
      "num_tokens": 109276336.0,
      "step": 53925
    },
    {
      "entropy": 5.9202272415161135,
      "epoch": 4.712924932273006,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002985209702605027,
      "loss": 5.3177,
      "mean_token_accuracy": 0.20627596378326415,
      "num_tokens": 109286639.0,
      "step": 53930
    },
    {
      "entropy": 5.906519603729248,
      "epoch": 4.71336188062571,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002984899796708586,
      "loss": 5.2298,
      "mean_token_accuracy": 0.21293804347515105,
      "num_tokens": 109296994.0,
      "step": 53935
    },
    {
      "entropy": 5.898672008514405,
      "epoch": 4.7137988289784145,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002984589886306644,
      "loss": 5.3206,
      "mean_token_accuracy": 0.20472271293401717,
      "num_tokens": 109306729.0,
      "step": 53940
    },
    {
      "entropy": 5.835019731521607,
      "epoch": 4.714235777331119,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002984279971405145,
      "loss": 5.37,
      "mean_token_accuracy": 0.20269441306591035,
      "num_tokens": 109318664.0,
      "step": 53945
    },
    {
      "entropy": 5.9055084705352785,
      "epoch": 4.714672725683824,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002983970052010033,
      "loss": 5.2921,
      "mean_token_accuracy": 0.20747271329164504,
      "num_tokens": 109327950.0,
      "step": 53950
    },
    {
      "entropy": 5.891663694381714,
      "epoch": 4.715109674036529,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029836601281272543,
      "loss": 5.2662,
      "mean_token_accuracy": 0.20751525163650514,
      "num_tokens": 109338046.0,
      "step": 53955
    },
    {
      "entropy": 5.908524751663208,
      "epoch": 4.715546622389233,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000298335019976275,
      "loss": 5.2644,
      "mean_token_accuracy": 0.21091226637363433,
      "num_tokens": 109348738.0,
      "step": 53960
    },
    {
      "entropy": 5.920520877838134,
      "epoch": 4.715983570741939,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002983040266922469,
      "loss": 5.3238,
      "mean_token_accuracy": 0.20655420124530793,
      "num_tokens": 109358729.0,
      "step": 53965
    },
    {
      "entropy": 5.883314609527588,
      "epoch": 4.716420519094643,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002982730329612352,
      "loss": 5.3125,
      "mean_token_accuracy": 0.20403342247009276,
      "num_tokens": 109368912.0,
      "step": 53970
    },
    {
      "entropy": 5.8737428188323975,
      "epoch": 4.716857467447348,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029824203878383454,
      "loss": 5.2711,
      "mean_token_accuracy": 0.207236947119236,
      "num_tokens": 109378703.0,
      "step": 53975
    },
    {
      "entropy": 5.9087118148803714,
      "epoch": 4.717294415800053,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002982110441606394,
      "loss": 5.3341,
      "mean_token_accuracy": 0.20691514015197754,
      "num_tokens": 109388839.0,
      "step": 53980
    },
    {
      "entropy": 5.830571174621582,
      "epoch": 4.717731364152757,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029818004909224426,
      "loss": 5.1777,
      "mean_token_accuracy": 0.21933842301368714,
      "num_tokens": 109399661.0,
      "step": 53985
    },
    {
      "entropy": 5.750387859344483,
      "epoch": 4.718168312505462,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029814905357924374,
      "loss": 5.1306,
      "mean_token_accuracy": 0.2244100645184517,
      "num_tokens": 109409755.0,
      "step": 53990
    },
    {
      "entropy": 5.8431761264801025,
      "epoch": 4.718605260858166,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002981180576222321,
      "loss": 5.2522,
      "mean_token_accuracy": 0.2076107978820801,
      "num_tokens": 109421119.0,
      "step": 53995
    },
    {
      "entropy": 5.864750719070434,
      "epoch": 4.719042209210871,
      "grad_norm": 1.0,
      "learning_rate": 0.00029808706122180404,
      "loss": 5.2508,
      "mean_token_accuracy": 0.21262926012277603,
      "num_tokens": 109431136.0,
      "step": 54000
    },
    {
      "epoch": 4.719042209210871,
      "eval_entropy": 5.585795371867885,
      "eval_loss": 5.352627754211426,
      "eval_mean_token_accuracy": 0.21234075893348667,
      "eval_num_tokens": 109431136.0,
      "eval_runtime": 27.9293,
      "eval_samples_per_second": 1325.276,
      "eval_steps_per_second": 165.668,
      "step": 54000
    },
    {
      "entropy": 5.867594242095947,
      "epoch": 4.719479157563576,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002980560643785541,
      "loss": 5.2623,
      "mean_token_accuracy": 0.20485040694475173,
      "num_tokens": 109441898.0,
      "step": 54005
    },
    {
      "entropy": 5.940377426147461,
      "epoch": 4.719916105916281,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002980250670930767,
      "loss": 5.3119,
      "mean_token_accuracy": 0.20392631143331527,
      "num_tokens": 109451067.0,
      "step": 54010
    },
    {
      "entropy": 5.814860248565674,
      "epoch": 4.720353054268985,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002979940693659666,
      "loss": 5.1967,
      "mean_token_accuracy": 0.22202742099761963,
      "num_tokens": 109461131.0,
      "step": 54015
    },
    {
      "entropy": 5.828780889511108,
      "epoch": 4.7207900026216905,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029796307119781803,
      "loss": 5.3066,
      "mean_token_accuracy": 0.20557685047388077,
      "num_tokens": 109471495.0,
      "step": 54020
    },
    {
      "entropy": 5.789646005630493,
      "epoch": 4.721226950974395,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029793207258922575,
      "loss": 5.2451,
      "mean_token_accuracy": 0.21183938533067703,
      "num_tokens": 109481571.0,
      "step": 54025
    },
    {
      "entropy": 5.859884786605835,
      "epoch": 4.721663899327099,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029790107354078435,
      "loss": 5.196,
      "mean_token_accuracy": 0.21150265634059906,
      "num_tokens": 109491233.0,
      "step": 54030
    },
    {
      "entropy": 5.8468084812164305,
      "epoch": 4.722100847679804,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029787007405308825,
      "loss": 5.2634,
      "mean_token_accuracy": 0.20718976557254792,
      "num_tokens": 109501722.0,
      "step": 54035
    },
    {
      "entropy": 5.875467348098755,
      "epoch": 4.722537796032509,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029783907412673236,
      "loss": 5.2418,
      "mean_token_accuracy": 0.21403346359729766,
      "num_tokens": 109510423.0,
      "step": 54040
    },
    {
      "entropy": 5.8596446990966795,
      "epoch": 4.722974744385214,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002978080737623108,
      "loss": 5.3064,
      "mean_token_accuracy": 0.2031472235918045,
      "num_tokens": 109520746.0,
      "step": 54045
    },
    {
      "entropy": 5.8513593673706055,
      "epoch": 4.723411692737918,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002977770729604186,
      "loss": 5.2991,
      "mean_token_accuracy": 0.20680481493473052,
      "num_tokens": 109530677.0,
      "step": 54050
    },
    {
      "entropy": 5.866330575942993,
      "epoch": 4.723848641090623,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002977460717216502,
      "loss": 5.2659,
      "mean_token_accuracy": 0.21031666696071624,
      "num_tokens": 109539923.0,
      "step": 54055
    },
    {
      "entropy": 5.858955430984497,
      "epoch": 4.724285589443328,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002977150700466002,
      "loss": 5.2087,
      "mean_token_accuracy": 0.21721208095550537,
      "num_tokens": 109548925.0,
      "step": 54060
    },
    {
      "entropy": 5.840574312210083,
      "epoch": 4.7247225377960325,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029768406793586325,
      "loss": 5.2217,
      "mean_token_accuracy": 0.20832999646663666,
      "num_tokens": 109558842.0,
      "step": 54065
    },
    {
      "entropy": 5.826891136169434,
      "epoch": 4.725159486148737,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029765306539003393,
      "loss": 5.1895,
      "mean_token_accuracy": 0.2152777150273323,
      "num_tokens": 109567757.0,
      "step": 54070
    },
    {
      "entropy": 5.8934544086456295,
      "epoch": 4.725596434501442,
      "grad_norm": 1.078125,
      "learning_rate": 0.000297622062409707,
      "loss": 5.3105,
      "mean_token_accuracy": 0.20383282601833344,
      "num_tokens": 109577635.0,
      "step": 54075
    },
    {
      "entropy": 5.853840684890747,
      "epoch": 4.726033382854147,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002975910589954771,
      "loss": 5.2509,
      "mean_token_accuracy": 0.2064347580075264,
      "num_tokens": 109586800.0,
      "step": 54080
    },
    {
      "entropy": 5.845836925506592,
      "epoch": 4.726470331206851,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029756005514793883,
      "loss": 5.2081,
      "mean_token_accuracy": 0.21233228892087935,
      "num_tokens": 109596681.0,
      "step": 54085
    },
    {
      "entropy": 5.844128274917603,
      "epoch": 4.726907279559556,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002975290508676869,
      "loss": 5.3325,
      "mean_token_accuracy": 0.2068369060754776,
      "num_tokens": 109606829.0,
      "step": 54090
    },
    {
      "entropy": 5.882937812805176,
      "epoch": 4.727344227912261,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029749804615531594,
      "loss": 5.3199,
      "mean_token_accuracy": 0.2020780086517334,
      "num_tokens": 109616521.0,
      "step": 54095
    },
    {
      "entropy": 5.943563747406006,
      "epoch": 4.7277811762649655,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029746704101142065,
      "loss": 5.2933,
      "mean_token_accuracy": 0.1995037764310837,
      "num_tokens": 109627184.0,
      "step": 54100
    },
    {
      "entropy": 5.871136903762817,
      "epoch": 4.72821812461767,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002974360354365957,
      "loss": 5.1716,
      "mean_token_accuracy": 0.22120787799358368,
      "num_tokens": 109637680.0,
      "step": 54105
    },
    {
      "entropy": 5.841470193862915,
      "epoch": 4.7286550729703745,
      "grad_norm": 0.953125,
      "learning_rate": 0.00029740502943143596,
      "loss": 5.2638,
      "mean_token_accuracy": 0.2043451651930809,
      "num_tokens": 109647521.0,
      "step": 54110
    },
    {
      "entropy": 5.79041166305542,
      "epoch": 4.72909202132308,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029737402299653587,
      "loss": 5.2416,
      "mean_token_accuracy": 0.21475017666816712,
      "num_tokens": 109656889.0,
      "step": 54115
    },
    {
      "entropy": 5.820567607879639,
      "epoch": 4.729528969675784,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029734301613249035,
      "loss": 5.1504,
      "mean_token_accuracy": 0.21729123890399932,
      "num_tokens": 109666681.0,
      "step": 54120
    },
    {
      "entropy": 5.93259596824646,
      "epoch": 4.729965918028489,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029731200883989414,
      "loss": 5.3813,
      "mean_token_accuracy": 0.1988944098353386,
      "num_tokens": 109677657.0,
      "step": 54125
    },
    {
      "entropy": 5.916926908493042,
      "epoch": 4.730402866381194,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00029728100111934186,
      "loss": 5.2377,
      "mean_token_accuracy": 0.2079031765460968,
      "num_tokens": 109688031.0,
      "step": 54130
    },
    {
      "entropy": 5.873408222198487,
      "epoch": 4.730839814733899,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002972499929714283,
      "loss": 5.3339,
      "mean_token_accuracy": 0.20666728615760804,
      "num_tokens": 109699652.0,
      "step": 54135
    },
    {
      "entropy": 5.809854125976562,
      "epoch": 4.731276763086603,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002972189843967482,
      "loss": 5.1633,
      "mean_token_accuracy": 0.21339148581027984,
      "num_tokens": 109709674.0,
      "step": 54140
    },
    {
      "entropy": 5.757018375396728,
      "epoch": 4.7317137114393075,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002971879753958963,
      "loss": 5.1742,
      "mean_token_accuracy": 0.21600909829139708,
      "num_tokens": 109720374.0,
      "step": 54145
    },
    {
      "entropy": 5.847087430953979,
      "epoch": 4.732150659792013,
      "grad_norm": 1.125,
      "learning_rate": 0.0002971569659694675,
      "loss": 5.2832,
      "mean_token_accuracy": 0.20446348339319229,
      "num_tokens": 109730613.0,
      "step": 54150
    },
    {
      "entropy": 5.8624317169189455,
      "epoch": 4.732587608144717,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002971259561180565,
      "loss": 5.2535,
      "mean_token_accuracy": 0.20924047380685806,
      "num_tokens": 109741593.0,
      "step": 54155
    },
    {
      "entropy": 5.933083248138428,
      "epoch": 4.733024556497422,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000297094945842258,
      "loss": 5.4177,
      "mean_token_accuracy": 0.19074669629335403,
      "num_tokens": 109752178.0,
      "step": 54160
    },
    {
      "entropy": 5.975879335403443,
      "epoch": 4.733461504850126,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002970639351426669,
      "loss": 5.3506,
      "mean_token_accuracy": 0.1987948849797249,
      "num_tokens": 109762833.0,
      "step": 54165
    },
    {
      "entropy": 5.881414937973022,
      "epoch": 4.733898453202832,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00029703292401987784,
      "loss": 5.2891,
      "mean_token_accuracy": 0.214098060131073,
      "num_tokens": 109773449.0,
      "step": 54170
    },
    {
      "entropy": 5.852913188934326,
      "epoch": 4.734335401555536,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029700191247448586,
      "loss": 5.2993,
      "mean_token_accuracy": 0.20235639363527297,
      "num_tokens": 109783637.0,
      "step": 54175
    },
    {
      "entropy": 5.878699588775635,
      "epoch": 4.734772349908241,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029697090050708575,
      "loss": 5.2502,
      "mean_token_accuracy": 0.2111206278204918,
      "num_tokens": 109793443.0,
      "step": 54180
    },
    {
      "entropy": 5.892764663696289,
      "epoch": 4.735209298260946,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002969398881182721,
      "loss": 5.3047,
      "mean_token_accuracy": 0.2021851047873497,
      "num_tokens": 109804253.0,
      "step": 54185
    },
    {
      "entropy": 5.86941933631897,
      "epoch": 4.73564624661365,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029690887530864,
      "loss": 5.2432,
      "mean_token_accuracy": 0.20905410200357438,
      "num_tokens": 109814827.0,
      "step": 54190
    },
    {
      "entropy": 5.855991172790527,
      "epoch": 4.736083194966355,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029687786207878414,
      "loss": 5.26,
      "mean_token_accuracy": 0.2105087623000145,
      "num_tokens": 109824388.0,
      "step": 54195
    },
    {
      "entropy": 5.926966667175293,
      "epoch": 4.736520143319059,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002968468484292995,
      "loss": 5.3832,
      "mean_token_accuracy": 0.20147550851106644,
      "num_tokens": 109833967.0,
      "step": 54200
    },
    {
      "entropy": 5.777616786956787,
      "epoch": 4.736957091671765,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029681583436078074,
      "loss": 5.1412,
      "mean_token_accuracy": 0.23194549083709717,
      "num_tokens": 109842522.0,
      "step": 54205
    },
    {
      "entropy": 5.835075569152832,
      "epoch": 4.737394040024469,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029678481987382294,
      "loss": 5.2349,
      "mean_token_accuracy": 0.20810457020998002,
      "num_tokens": 109853015.0,
      "step": 54210
    },
    {
      "entropy": 5.918501949310302,
      "epoch": 4.737830988377174,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002967538049690208,
      "loss": 5.2831,
      "mean_token_accuracy": 0.20216262042522432,
      "num_tokens": 109862801.0,
      "step": 54215
    },
    {
      "entropy": 5.865246152877807,
      "epoch": 4.738267936729878,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029672278964696936,
      "loss": 5.2454,
      "mean_token_accuracy": 0.21179793328046798,
      "num_tokens": 109871888.0,
      "step": 54220
    },
    {
      "entropy": 5.793857288360596,
      "epoch": 4.7387048850825835,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002966917739082634,
      "loss": 5.256,
      "mean_token_accuracy": 0.20709623098373414,
      "num_tokens": 109881933.0,
      "step": 54225
    },
    {
      "entropy": 5.843588781356812,
      "epoch": 4.739141833435288,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00029666075775349783,
      "loss": 5.2513,
      "mean_token_accuracy": 0.21464868187904357,
      "num_tokens": 109892210.0,
      "step": 54230
    },
    {
      "entropy": 5.881244754791259,
      "epoch": 4.739578781787992,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029662974118326756,
      "loss": 5.2892,
      "mean_token_accuracy": 0.20882684290409087,
      "num_tokens": 109902507.0,
      "step": 54235
    },
    {
      "entropy": 5.869935846328735,
      "epoch": 4.740015730140698,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002965987241981676,
      "loss": 5.3037,
      "mean_token_accuracy": 0.2053201675415039,
      "num_tokens": 109912052.0,
      "step": 54240
    },
    {
      "entropy": 5.888818979263306,
      "epoch": 4.740452678493402,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002965677067987927,
      "loss": 5.2179,
      "mean_token_accuracy": 0.21190074235200881,
      "num_tokens": 109922591.0,
      "step": 54245
    },
    {
      "entropy": 5.904103708267212,
      "epoch": 4.740889626846107,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000296536688985738,
      "loss": 5.3009,
      "mean_token_accuracy": 0.2054843410849571,
      "num_tokens": 109932015.0,
      "step": 54250
    },
    {
      "entropy": 5.860475397109985,
      "epoch": 4.741326575198811,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002965056707595982,
      "loss": 5.2887,
      "mean_token_accuracy": 0.21008395552635192,
      "num_tokens": 109943184.0,
      "step": 54255
    },
    {
      "entropy": 5.828697967529297,
      "epoch": 4.7417635235515165,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002964746521209684,
      "loss": 5.2706,
      "mean_token_accuracy": 0.21413087397813796,
      "num_tokens": 109952013.0,
      "step": 54260
    },
    {
      "entropy": 5.849738931655883,
      "epoch": 4.742200471904221,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029644363307044353,
      "loss": 5.1923,
      "mean_token_accuracy": 0.2134503409266472,
      "num_tokens": 109962163.0,
      "step": 54265
    },
    {
      "entropy": 5.837800407409668,
      "epoch": 4.7426374202569255,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002964126136086185,
      "loss": 5.331,
      "mean_token_accuracy": 0.21469750702381135,
      "num_tokens": 109971479.0,
      "step": 54270
    },
    {
      "entropy": 5.849053812026978,
      "epoch": 4.74307436860963,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002963815937360884,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21530047804117203,
      "num_tokens": 109981407.0,
      "step": 54275
    },
    {
      "entropy": 5.895948839187622,
      "epoch": 4.743511316962335,
      "grad_norm": 1.0,
      "learning_rate": 0.0002963505734534481,
      "loss": 5.3522,
      "mean_token_accuracy": 0.20401151478290558,
      "num_tokens": 109991404.0,
      "step": 54280
    },
    {
      "entropy": 5.845859575271606,
      "epoch": 4.74394826531504,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029631955276129265,
      "loss": 5.2348,
      "mean_token_accuracy": 0.21364212930202484,
      "num_tokens": 110001483.0,
      "step": 54285
    },
    {
      "entropy": 5.924649620056153,
      "epoch": 4.744385213667744,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029628853166021696,
      "loss": 5.3216,
      "mean_token_accuracy": 0.19797641336917876,
      "num_tokens": 110011460.0,
      "step": 54290
    },
    {
      "entropy": 5.825711584091186,
      "epoch": 4.74482216202045,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002962575101508161,
      "loss": 5.2081,
      "mean_token_accuracy": 0.21088789850473405,
      "num_tokens": 110021199.0,
      "step": 54295
    },
    {
      "entropy": 5.892941570281982,
      "epoch": 4.745259110373154,
      "grad_norm": 1.03125,
      "learning_rate": 0.000296226488233685,
      "loss": 5.3316,
      "mean_token_accuracy": 0.2027154266834259,
      "num_tokens": 110031127.0,
      "step": 54300
    },
    {
      "entropy": 5.885359287261963,
      "epoch": 4.7456960587258585,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002961954659094187,
      "loss": 5.2668,
      "mean_token_accuracy": 0.2126220256090164,
      "num_tokens": 110041541.0,
      "step": 54305
    },
    {
      "entropy": 5.899575424194336,
      "epoch": 4.746133007078563,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002961644431786124,
      "loss": 5.2752,
      "mean_token_accuracy": 0.2081721156835556,
      "num_tokens": 110050882.0,
      "step": 54310
    },
    {
      "entropy": 5.878344535827637,
      "epoch": 4.746569955431268,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002961334200418609,
      "loss": 5.3351,
      "mean_token_accuracy": 0.2014761745929718,
      "num_tokens": 110062510.0,
      "step": 54315
    },
    {
      "entropy": 5.894688081741333,
      "epoch": 4.747006903783973,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002961023964997594,
      "loss": 5.2788,
      "mean_token_accuracy": 0.21046175360679625,
      "num_tokens": 110072625.0,
      "step": 54320
    },
    {
      "entropy": 5.870384454727173,
      "epoch": 4.747443852136677,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029607137255290277,
      "loss": 5.2371,
      "mean_token_accuracy": 0.21590915322303772,
      "num_tokens": 110082525.0,
      "step": 54325
    },
    {
      "entropy": 5.85167932510376,
      "epoch": 4.747880800489382,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029604034820188623,
      "loss": 5.291,
      "mean_token_accuracy": 0.20984893143177033,
      "num_tokens": 110092137.0,
      "step": 54330
    },
    {
      "entropy": 5.869189071655273,
      "epoch": 4.748317748842087,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002960093234473048,
      "loss": 5.3342,
      "mean_token_accuracy": 0.20245331525802612,
      "num_tokens": 110103677.0,
      "step": 54335
    },
    {
      "entropy": 5.861517477035522,
      "epoch": 4.748754697194792,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029597829828975354,
      "loss": 5.2404,
      "mean_token_accuracy": 0.2110649511218071,
      "num_tokens": 110113131.0,
      "step": 54340
    },
    {
      "entropy": 5.910169363021851,
      "epoch": 4.749191645547496,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002959472727298275,
      "loss": 5.3257,
      "mean_token_accuracy": 0.20888693928718566,
      "num_tokens": 110124777.0,
      "step": 54345
    },
    {
      "entropy": 5.9198229789733885,
      "epoch": 4.749628593900201,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002959162467681218,
      "loss": 5.2568,
      "mean_token_accuracy": 0.20923299193382264,
      "num_tokens": 110134234.0,
      "step": 54350
    },
    {
      "entropy": 5.844200611114502,
      "epoch": 4.750065542252906,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002958852204052315,
      "loss": 5.3008,
      "mean_token_accuracy": 0.20031464993953704,
      "num_tokens": 110144135.0,
      "step": 54355
    },
    {
      "entropy": 5.796900653839112,
      "epoch": 4.75050249060561,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029585419364175174,
      "loss": 5.2014,
      "mean_token_accuracy": 0.21044834405183793,
      "num_tokens": 110154009.0,
      "step": 54360
    },
    {
      "entropy": 5.954784965515136,
      "epoch": 4.750939438958315,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029582316647827774,
      "loss": 5.3441,
      "mean_token_accuracy": 0.20350825190544128,
      "num_tokens": 110163628.0,
      "step": 54365
    },
    {
      "entropy": 5.911629295349121,
      "epoch": 4.75137638731102,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029579213891540437,
      "loss": 5.2671,
      "mean_token_accuracy": 0.20774114429950713,
      "num_tokens": 110174321.0,
      "step": 54370
    },
    {
      "entropy": 5.813344955444336,
      "epoch": 4.751813335663725,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029576111095372685,
      "loss": 5.2018,
      "mean_token_accuracy": 0.21726952642202377,
      "num_tokens": 110184674.0,
      "step": 54375
    },
    {
      "entropy": 5.729461669921875,
      "epoch": 4.752250284016429,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029573008259384037,
      "loss": 5.1796,
      "mean_token_accuracy": 0.2192919060587883,
      "num_tokens": 110195090.0,
      "step": 54380
    },
    {
      "entropy": 5.853987169265747,
      "epoch": 4.752687232369134,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002956990538363401,
      "loss": 5.2653,
      "mean_token_accuracy": 0.20558559596538545,
      "num_tokens": 110206237.0,
      "step": 54385
    },
    {
      "entropy": 5.874744939804077,
      "epoch": 4.753124180721839,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002956680246818211,
      "loss": 5.3077,
      "mean_token_accuracy": 0.21303244829177856,
      "num_tokens": 110216434.0,
      "step": 54390
    },
    {
      "entropy": 5.8653663158416744,
      "epoch": 4.753561129074543,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002956369951308786,
      "loss": 5.319,
      "mean_token_accuracy": 0.20293537080287932,
      "num_tokens": 110226847.0,
      "step": 54395
    },
    {
      "entropy": 5.766601800918579,
      "epoch": 4.753998077427248,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029560596518410766,
      "loss": 5.1223,
      "mean_token_accuracy": 0.226563960313797,
      "num_tokens": 110238113.0,
      "step": 54400
    },
    {
      "entropy": 5.878798532485962,
      "epoch": 4.754435025779953,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002955749348421035,
      "loss": 5.2794,
      "mean_token_accuracy": 0.210342375934124,
      "num_tokens": 110248338.0,
      "step": 54405
    },
    {
      "entropy": 5.796452379226684,
      "epoch": 4.754871974132658,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029554390410546134,
      "loss": 5.2288,
      "mean_token_accuracy": 0.2156543418765068,
      "num_tokens": 110259193.0,
      "step": 54410
    },
    {
      "entropy": 5.8989180564880375,
      "epoch": 4.755308922485362,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002955128729747764,
      "loss": 5.2763,
      "mean_token_accuracy": 0.2035280078649521,
      "num_tokens": 110269929.0,
      "step": 54415
    },
    {
      "entropy": 5.771977758407592,
      "epoch": 4.755745870838067,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002954818414506437,
      "loss": 5.1437,
      "mean_token_accuracy": 0.21440295577049256,
      "num_tokens": 110279591.0,
      "step": 54420
    },
    {
      "entropy": 5.710013008117675,
      "epoch": 4.756182819190772,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002954508095336586,
      "loss": 5.1695,
      "mean_token_accuracy": 0.22099486738443375,
      "num_tokens": 110289781.0,
      "step": 54425
    },
    {
      "entropy": 5.898360061645508,
      "epoch": 4.7566197675434765,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002954197772244162,
      "loss": 5.2899,
      "mean_token_accuracy": 0.20719857513904572,
      "num_tokens": 110299581.0,
      "step": 54430
    },
    {
      "entropy": 5.9238368511199955,
      "epoch": 4.757056715896181,
      "grad_norm": 1.0,
      "learning_rate": 0.00029538874452351185,
      "loss": 5.3029,
      "mean_token_accuracy": 0.20126113444566726,
      "num_tokens": 110309343.0,
      "step": 54435
    },
    {
      "entropy": 5.869533061981201,
      "epoch": 4.757493664248885,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029535771143154074,
      "loss": 5.316,
      "mean_token_accuracy": 0.2098039135336876,
      "num_tokens": 110320346.0,
      "step": 54440
    },
    {
      "entropy": 5.836766529083252,
      "epoch": 4.757930612601591,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029532667794909795,
      "loss": 5.2898,
      "mean_token_accuracy": 0.21031882464885712,
      "num_tokens": 110330775.0,
      "step": 54445
    },
    {
      "entropy": 5.88843092918396,
      "epoch": 4.758367560954295,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029529564407677883,
      "loss": 5.2795,
      "mean_token_accuracy": 0.2135935381054878,
      "num_tokens": 110340372.0,
      "step": 54450
    },
    {
      "entropy": 5.8190971374511715,
      "epoch": 4.758804509307,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002952646098151786,
      "loss": 5.2343,
      "mean_token_accuracy": 0.21279652118682862,
      "num_tokens": 110349988.0,
      "step": 54455
    },
    {
      "entropy": 5.874308443069458,
      "epoch": 4.759241457659705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029523357516489265,
      "loss": 5.2828,
      "mean_token_accuracy": 0.20506299734115602,
      "num_tokens": 110360072.0,
      "step": 54460
    },
    {
      "entropy": 5.852837324142456,
      "epoch": 4.7596784060124095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002952025401265161,
      "loss": 5.2511,
      "mean_token_accuracy": 0.21577712893486023,
      "num_tokens": 110370289.0,
      "step": 54465
    },
    {
      "entropy": 5.89428677558899,
      "epoch": 4.760115354365114,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002951715047006441,
      "loss": 5.2973,
      "mean_token_accuracy": 0.20886960476636887,
      "num_tokens": 110380777.0,
      "step": 54470
    },
    {
      "entropy": 5.889464330673218,
      "epoch": 4.7605523027178185,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002951404688878722,
      "loss": 5.28,
      "mean_token_accuracy": 0.2065955713391304,
      "num_tokens": 110390800.0,
      "step": 54475
    },
    {
      "entropy": 5.856869888305664,
      "epoch": 4.760989251070524,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002951094326887955,
      "loss": 5.3635,
      "mean_token_accuracy": 0.19477435052394867,
      "num_tokens": 110400582.0,
      "step": 54480
    },
    {
      "entropy": 5.899149084091187,
      "epoch": 4.761426199423228,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029507839610400945,
      "loss": 5.2719,
      "mean_token_accuracy": 0.20857807099819184,
      "num_tokens": 110410341.0,
      "step": 54485
    },
    {
      "entropy": 5.869025087356567,
      "epoch": 4.761863147775933,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002950473591341091,
      "loss": 5.2726,
      "mean_token_accuracy": 0.21030950397253037,
      "num_tokens": 110420804.0,
      "step": 54490
    },
    {
      "entropy": 5.849012804031372,
      "epoch": 4.762300096128637,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00029501632177969,
      "loss": 5.1896,
      "mean_token_accuracy": 0.21861943006515502,
      "num_tokens": 110431485.0,
      "step": 54495
    },
    {
      "entropy": 5.8670401096344,
      "epoch": 4.762737044481343,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002949852840413473,
      "loss": 5.2541,
      "mean_token_accuracy": 0.20898856818675995,
      "num_tokens": 110442368.0,
      "step": 54500
    },
    {
      "entropy": 5.724662446975708,
      "epoch": 4.763173992834047,
      "grad_norm": 1.125,
      "learning_rate": 0.00029495424591967644,
      "loss": 5.1612,
      "mean_token_accuracy": 0.2223414659500122,
      "num_tokens": 110453087.0,
      "step": 54505
    },
    {
      "entropy": 5.8886146068573,
      "epoch": 4.7636109411867515,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002949232074152726,
      "loss": 5.331,
      "mean_token_accuracy": 0.20543254166841507,
      "num_tokens": 110463086.0,
      "step": 54510
    },
    {
      "entropy": 5.894073390960694,
      "epoch": 4.764047889539457,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002948921685287313,
      "loss": 5.2912,
      "mean_token_accuracy": 0.20283401310443877,
      "num_tokens": 110472918.0,
      "step": 54515
    },
    {
      "entropy": 5.928156423568725,
      "epoch": 4.764484837892161,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029486112926064775,
      "loss": 5.2865,
      "mean_token_accuracy": 0.2005913510918617,
      "num_tokens": 110482966.0,
      "step": 54520
    },
    {
      "entropy": 5.886723566055298,
      "epoch": 4.764921786244866,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002948300896116173,
      "loss": 5.3036,
      "mean_token_accuracy": 0.20899414271116257,
      "num_tokens": 110493343.0,
      "step": 54525
    },
    {
      "entropy": 5.881261920928955,
      "epoch": 4.76535873459757,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002947990495822354,
      "loss": 5.1988,
      "mean_token_accuracy": 0.21422846168279647,
      "num_tokens": 110503978.0,
      "step": 54530
    },
    {
      "entropy": 5.841882848739624,
      "epoch": 4.765795682950275,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029476800917309733,
      "loss": 5.2323,
      "mean_token_accuracy": 0.21228781938552857,
      "num_tokens": 110513689.0,
      "step": 54535
    },
    {
      "entropy": 5.847158527374267,
      "epoch": 4.76623263130298,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029473696838479843,
      "loss": 5.2482,
      "mean_token_accuracy": 0.21567221432924272,
      "num_tokens": 110523002.0,
      "step": 54540
    },
    {
      "entropy": 5.864076137542725,
      "epoch": 4.766669579655685,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029470592721793414,
      "loss": 5.3485,
      "mean_token_accuracy": 0.20407938957214355,
      "num_tokens": 110533834.0,
      "step": 54545
    },
    {
      "entropy": 5.824592256546021,
      "epoch": 4.767106528008389,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029467488567309983,
      "loss": 5.1814,
      "mean_token_accuracy": 0.2252017989754677,
      "num_tokens": 110544331.0,
      "step": 54550
    },
    {
      "entropy": 5.916950607299805,
      "epoch": 4.767543476361094,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000294643843750891,
      "loss": 5.3132,
      "mean_token_accuracy": 0.203237147629261,
      "num_tokens": 110554356.0,
      "step": 54555
    },
    {
      "entropy": 5.87756462097168,
      "epoch": 4.767980424713799,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002946128014519029,
      "loss": 5.3857,
      "mean_token_accuracy": 0.20323435068130494,
      "num_tokens": 110564794.0,
      "step": 54560
    },
    {
      "entropy": 5.908736705780029,
      "epoch": 4.768417373066503,
      "grad_norm": 1.109375,
      "learning_rate": 0.000294581758776731,
      "loss": 5.308,
      "mean_token_accuracy": 0.20222229659557342,
      "num_tokens": 110574523.0,
      "step": 54565
    },
    {
      "entropy": 5.838738107681275,
      "epoch": 4.768854321419209,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002945507157259707,
      "loss": 5.2472,
      "mean_token_accuracy": 0.20931720435619355,
      "num_tokens": 110585051.0,
      "step": 54570
    },
    {
      "entropy": 5.875579786300659,
      "epoch": 4.769291269771913,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002945196723002174,
      "loss": 5.2437,
      "mean_token_accuracy": 0.21658874452114105,
      "num_tokens": 110595987.0,
      "step": 54575
    },
    {
      "entropy": 5.949211168289184,
      "epoch": 4.769728218124618,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029448862850006664,
      "loss": 5.3653,
      "mean_token_accuracy": 0.19977940171957015,
      "num_tokens": 110605986.0,
      "step": 54580
    },
    {
      "entropy": 5.916871166229248,
      "epoch": 4.770165166477322,
      "grad_norm": 1.125,
      "learning_rate": 0.00029445758432611373,
      "loss": 5.3387,
      "mean_token_accuracy": 0.20281503051519395,
      "num_tokens": 110616243.0,
      "step": 54585
    },
    {
      "entropy": 5.871824598312378,
      "epoch": 4.770602114830027,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002944265397789541,
      "loss": 5.302,
      "mean_token_accuracy": 0.205795493721962,
      "num_tokens": 110625752.0,
      "step": 54590
    },
    {
      "entropy": 5.9025559425354,
      "epoch": 4.771039063182732,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029439549485918334,
      "loss": 5.3324,
      "mean_token_accuracy": 0.21126218438148497,
      "num_tokens": 110635607.0,
      "step": 54595
    },
    {
      "entropy": 5.922963190078735,
      "epoch": 4.771476011535436,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029436444956739674,
      "loss": 5.2967,
      "mean_token_accuracy": 0.20960184633731843,
      "num_tokens": 110646376.0,
      "step": 54600
    },
    {
      "entropy": 5.873734712600708,
      "epoch": 4.771912959888141,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029433340390418996,
      "loss": 5.3194,
      "mean_token_accuracy": 0.2001795619726181,
      "num_tokens": 110656697.0,
      "step": 54605
    },
    {
      "entropy": 5.835140895843506,
      "epoch": 4.772349908240846,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002943023578701583,
      "loss": 5.3239,
      "mean_token_accuracy": 0.21243527233600618,
      "num_tokens": 110665897.0,
      "step": 54610
    },
    {
      "entropy": 5.807705211639404,
      "epoch": 4.772786856593551,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0002942713114658973,
      "loss": 5.2347,
      "mean_token_accuracy": 0.21856972128152846,
      "num_tokens": 110676282.0,
      "step": 54615
    },
    {
      "entropy": 5.950540828704834,
      "epoch": 4.773223804946255,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029424026469200245,
      "loss": 5.3599,
      "mean_token_accuracy": 0.20466499924659728,
      "num_tokens": 110686097.0,
      "step": 54620
    },
    {
      "entropy": 5.890060234069824,
      "epoch": 4.773660753298961,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029420921754906923,
      "loss": 5.2519,
      "mean_token_accuracy": 0.20643078088760375,
      "num_tokens": 110696378.0,
      "step": 54625
    },
    {
      "entropy": 5.910047197341919,
      "epoch": 4.774097701651665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002941781700376932,
      "loss": 5.3212,
      "mean_token_accuracy": 0.20640910118818284,
      "num_tokens": 110707144.0,
      "step": 54630
    },
    {
      "entropy": 5.834984302520752,
      "epoch": 4.7745346500043695,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029414712215846973,
      "loss": 5.2294,
      "mean_token_accuracy": 0.21562298387289047,
      "num_tokens": 110715993.0,
      "step": 54635
    },
    {
      "entropy": 5.918476581573486,
      "epoch": 4.774971598357074,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029411607391199445,
      "loss": 5.3119,
      "mean_token_accuracy": 0.1998613655567169,
      "num_tokens": 110726627.0,
      "step": 54640
    },
    {
      "entropy": 5.937845993041992,
      "epoch": 4.775408546709778,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029408502529886286,
      "loss": 5.3597,
      "mean_token_accuracy": 0.20488191097974778,
      "num_tokens": 110736958.0,
      "step": 54645
    },
    {
      "entropy": 5.858523893356323,
      "epoch": 4.775845495062484,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002940539763196706,
      "loss": 5.2078,
      "mean_token_accuracy": 0.21628126949071885,
      "num_tokens": 110747057.0,
      "step": 54650
    },
    {
      "entropy": 5.869193172454834,
      "epoch": 4.776282443415188,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029402292697501293,
      "loss": 5.3242,
      "mean_token_accuracy": 0.20465673953294755,
      "num_tokens": 110756430.0,
      "step": 54655
    },
    {
      "entropy": 5.896829509735108,
      "epoch": 4.776719391767893,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029399187726548564,
      "loss": 5.2493,
      "mean_token_accuracy": 0.20835819095373154,
      "num_tokens": 110766566.0,
      "step": 54660
    },
    {
      "entropy": 5.8153746128082275,
      "epoch": 4.777156340120598,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002939608271916841,
      "loss": 5.2258,
      "mean_token_accuracy": 0.21748212575912476,
      "num_tokens": 110775934.0,
      "step": 54665
    },
    {
      "entropy": 5.909921312332154,
      "epoch": 4.777593288473303,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002939297767542041,
      "loss": 5.2894,
      "mean_token_accuracy": 0.20615816563367845,
      "num_tokens": 110787608.0,
      "step": 54670
    },
    {
      "entropy": 5.885834741592407,
      "epoch": 4.778030236826007,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000293898725953641,
      "loss": 5.2545,
      "mean_token_accuracy": 0.21437402367591857,
      "num_tokens": 110797265.0,
      "step": 54675
    },
    {
      "entropy": 5.786981534957886,
      "epoch": 4.7784671851787115,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002938676747905904,
      "loss": 5.1812,
      "mean_token_accuracy": 0.21650377810001373,
      "num_tokens": 110807059.0,
      "step": 54680
    },
    {
      "entropy": 5.8346254348754885,
      "epoch": 4.778904133531417,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000293836623265648,
      "loss": 5.2058,
      "mean_token_accuracy": 0.2146538034081459,
      "num_tokens": 110816056.0,
      "step": 54685
    },
    {
      "entropy": 5.921615076065064,
      "epoch": 4.779341081884121,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029380557137940924,
      "loss": 5.3132,
      "mean_token_accuracy": 0.20457857549190522,
      "num_tokens": 110825601.0,
      "step": 54690
    },
    {
      "entropy": 5.869565725326538,
      "epoch": 4.779778030236826,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029377451913246977,
      "loss": 5.2028,
      "mean_token_accuracy": 0.2113985985517502,
      "num_tokens": 110835435.0,
      "step": 54695
    },
    {
      "entropy": 5.853565788269043,
      "epoch": 4.78021497858953,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002937434665254252,
      "loss": 5.2591,
      "mean_token_accuracy": 0.20895344614982606,
      "num_tokens": 110845607.0,
      "step": 54700
    },
    {
      "entropy": 5.848150730133057,
      "epoch": 4.780651926942236,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002937124135588711,
      "loss": 5.2675,
      "mean_token_accuracy": 0.2095123216509819,
      "num_tokens": 110856089.0,
      "step": 54705
    },
    {
      "entropy": 5.926081371307373,
      "epoch": 4.78108887529494,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000293681360233403,
      "loss": 5.3464,
      "mean_token_accuracy": 0.20298898965120316,
      "num_tokens": 110865491.0,
      "step": 54710
    },
    {
      "entropy": 5.8899911403656,
      "epoch": 4.781525823647645,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002936503065496167,
      "loss": 5.3219,
      "mean_token_accuracy": 0.20502635538578035,
      "num_tokens": 110875579.0,
      "step": 54715
    },
    {
      "entropy": 5.8632889747619625,
      "epoch": 4.78196277200035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029361925250810786,
      "loss": 5.269,
      "mean_token_accuracy": 0.20812940150499343,
      "num_tokens": 110886452.0,
      "step": 54720
    },
    {
      "entropy": 5.946552801132202,
      "epoch": 4.782399720353054,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002935881981094719,
      "loss": 5.3566,
      "mean_token_accuracy": 0.19995759874582292,
      "num_tokens": 110897854.0,
      "step": 54725
    },
    {
      "entropy": 5.895529937744141,
      "epoch": 4.782836668705759,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00029355714335430456,
      "loss": 5.2402,
      "mean_token_accuracy": 0.20866269916296004,
      "num_tokens": 110907942.0,
      "step": 54730
    },
    {
      "entropy": 5.86646466255188,
      "epoch": 4.783273617058463,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002935260882432015,
      "loss": 5.3157,
      "mean_token_accuracy": 0.21305381804704665,
      "num_tokens": 110917381.0,
      "step": 54735
    },
    {
      "entropy": 5.896022939682007,
      "epoch": 4.783710565411169,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002934950327767583,
      "loss": 5.2985,
      "mean_token_accuracy": 0.2067648231983185,
      "num_tokens": 110927343.0,
      "step": 54740
    },
    {
      "entropy": 5.899972248077392,
      "epoch": 4.784147513763873,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002934639769555708,
      "loss": 5.2872,
      "mean_token_accuracy": 0.2054971292614937,
      "num_tokens": 110937475.0,
      "step": 54745
    },
    {
      "entropy": 5.8599207401275635,
      "epoch": 4.784584462116578,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002934329207802345,
      "loss": 5.22,
      "mean_token_accuracy": 0.21906166523694992,
      "num_tokens": 110947411.0,
      "step": 54750
    },
    {
      "entropy": 5.815550947189331,
      "epoch": 4.785021410469282,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029340186425134507,
      "loss": 5.2544,
      "mean_token_accuracy": 0.2111246481537819,
      "num_tokens": 110957625.0,
      "step": 54755
    },
    {
      "entropy": 5.878488445281983,
      "epoch": 4.7854583588219874,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002933708073694983,
      "loss": 5.3374,
      "mean_token_accuracy": 0.20308366268873215,
      "num_tokens": 110967848.0,
      "step": 54760
    },
    {
      "entropy": 5.9170619487762455,
      "epoch": 4.785895307174692,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002933397501352898,
      "loss": 5.288,
      "mean_token_accuracy": 0.2057332441210747,
      "num_tokens": 110978109.0,
      "step": 54765
    },
    {
      "entropy": 5.923997402191162,
      "epoch": 4.786332255527396,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002933086925493153,
      "loss": 5.2425,
      "mean_token_accuracy": 0.21334272921085357,
      "num_tokens": 110987803.0,
      "step": 54770
    },
    {
      "entropy": 5.8190357208251955,
      "epoch": 4.786769203880102,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002932776346121704,
      "loss": 5.2345,
      "mean_token_accuracy": 0.21021364331245423,
      "num_tokens": 110997568.0,
      "step": 54775
    },
    {
      "entropy": 5.875457334518432,
      "epoch": 4.787206152232806,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000293246576324451,
      "loss": 5.2791,
      "mean_token_accuracy": 0.21105333119630815,
      "num_tokens": 111006922.0,
      "step": 54780
    },
    {
      "entropy": 5.899103450775146,
      "epoch": 4.787643100585511,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029321551768675256,
      "loss": 5.2829,
      "mean_token_accuracy": 0.20313888639211655,
      "num_tokens": 111017455.0,
      "step": 54785
    },
    {
      "entropy": 5.878425741195679,
      "epoch": 4.788080048938215,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002931844586996711,
      "loss": 5.3139,
      "mean_token_accuracy": 0.2081313133239746,
      "num_tokens": 111027668.0,
      "step": 54790
    },
    {
      "entropy": 5.888793087005615,
      "epoch": 4.7885169972909205,
      "grad_norm": 1.0,
      "learning_rate": 0.00029315339936380216,
      "loss": 5.2975,
      "mean_token_accuracy": 0.2120113790035248,
      "num_tokens": 111038199.0,
      "step": 54795
    },
    {
      "entropy": 5.947935485839844,
      "epoch": 4.788953945643625,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002931223396797415,
      "loss": 5.4127,
      "mean_token_accuracy": 0.19976502209901809,
      "num_tokens": 111047721.0,
      "step": 54800
    },
    {
      "entropy": 5.9091253757476805,
      "epoch": 4.7893908939963294,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002930912796480848,
      "loss": 5.2402,
      "mean_token_accuracy": 0.217880979180336,
      "num_tokens": 111057771.0,
      "step": 54805
    },
    {
      "entropy": 5.914538860321045,
      "epoch": 4.789827842349034,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029306021926942796,
      "loss": 5.2679,
      "mean_token_accuracy": 0.21377795040607453,
      "num_tokens": 111066994.0,
      "step": 54810
    },
    {
      "entropy": 5.833503675460816,
      "epoch": 4.790264790701739,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002930291585443667,
      "loss": 5.2291,
      "mean_token_accuracy": 0.2099042922258377,
      "num_tokens": 111076280.0,
      "step": 54815
    },
    {
      "entropy": 5.815505456924439,
      "epoch": 4.790701739054444,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002929980974734966,
      "loss": 5.306,
      "mean_token_accuracy": 0.20501339733600615,
      "num_tokens": 111087034.0,
      "step": 54820
    },
    {
      "entropy": 5.822227191925049,
      "epoch": 4.791138687407148,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002929670360574136,
      "loss": 5.1504,
      "mean_token_accuracy": 0.2132887750864029,
      "num_tokens": 111097896.0,
      "step": 54825
    },
    {
      "entropy": 5.864399528503418,
      "epoch": 4.791575635759854,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029293597429671343,
      "loss": 5.2993,
      "mean_token_accuracy": 0.2036902204155922,
      "num_tokens": 111108464.0,
      "step": 54830
    },
    {
      "entropy": 5.873274326324463,
      "epoch": 4.792012584112558,
      "grad_norm": 1.125,
      "learning_rate": 0.0002929049121919919,
      "loss": 5.3171,
      "mean_token_accuracy": 0.21150413751602173,
      "num_tokens": 111118670.0,
      "step": 54835
    },
    {
      "entropy": 5.913125801086426,
      "epoch": 4.7924495324652625,
      "grad_norm": 0.984375,
      "learning_rate": 0.00029287384974384476,
      "loss": 5.2909,
      "mean_token_accuracy": 0.20452796518802643,
      "num_tokens": 111129451.0,
      "step": 54840
    },
    {
      "entropy": 5.916761684417724,
      "epoch": 4.792886480817967,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002928427869528678,
      "loss": 5.3405,
      "mean_token_accuracy": 0.19907034188508987,
      "num_tokens": 111139852.0,
      "step": 54845
    },
    {
      "entropy": 5.949840688705445,
      "epoch": 4.793323429170672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029281172381965685,
      "loss": 5.3191,
      "mean_token_accuracy": 0.20229703187942505,
      "num_tokens": 111149408.0,
      "step": 54850
    },
    {
      "entropy": 5.894603872299195,
      "epoch": 4.793760377523377,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002927806603448077,
      "loss": 5.2817,
      "mean_token_accuracy": 0.20562144964933396,
      "num_tokens": 111159182.0,
      "step": 54855
    },
    {
      "entropy": 5.814043092727661,
      "epoch": 4.794197325876081,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029274959652891624,
      "loss": 5.2539,
      "mean_token_accuracy": 0.21017750203609467,
      "num_tokens": 111168747.0,
      "step": 54860
    },
    {
      "entropy": 5.9175080299377445,
      "epoch": 4.794634274228786,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002927185323725782,
      "loss": 5.3395,
      "mean_token_accuracy": 0.20174660682678222,
      "num_tokens": 111178850.0,
      "step": 54865
    },
    {
      "entropy": 5.88908839225769,
      "epoch": 4.795071222581491,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029268746787638935,
      "loss": 5.2574,
      "mean_token_accuracy": 0.20933501273393632,
      "num_tokens": 111188173.0,
      "step": 54870
    },
    {
      "entropy": 5.876942205429077,
      "epoch": 4.795508170934196,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002926564030409456,
      "loss": 5.2902,
      "mean_token_accuracy": 0.19858522415161134,
      "num_tokens": 111198564.0,
      "step": 54875
    },
    {
      "entropy": 5.844913816452026,
      "epoch": 4.7959451192869,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00029262533786684276,
      "loss": 5.2323,
      "mean_token_accuracy": 0.20960234999656677,
      "num_tokens": 111208546.0,
      "step": 54880
    },
    {
      "entropy": 5.841584348678589,
      "epoch": 4.796382067639605,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002925942723546768,
      "loss": 5.1988,
      "mean_token_accuracy": 0.2109857752919197,
      "num_tokens": 111218903.0,
      "step": 54885
    },
    {
      "entropy": 5.834330987930298,
      "epoch": 4.79681901599231,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002925632065050434,
      "loss": 5.2412,
      "mean_token_accuracy": 0.20531760901212692,
      "num_tokens": 111229420.0,
      "step": 54890
    },
    {
      "entropy": 5.901593923568726,
      "epoch": 4.797255964345014,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00029253214031853855,
      "loss": 5.2595,
      "mean_token_accuracy": 0.21059868186712266,
      "num_tokens": 111238764.0,
      "step": 54895
    },
    {
      "entropy": 5.905568981170655,
      "epoch": 4.797692912697719,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002925010737957581,
      "loss": 5.3668,
      "mean_token_accuracy": 0.20195332020521164,
      "num_tokens": 111249082.0,
      "step": 54900
    },
    {
      "entropy": 5.907672548294068,
      "epoch": 4.798129861050424,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002924700069372978,
      "loss": 5.3743,
      "mean_token_accuracy": 0.20226949602365493,
      "num_tokens": 111259141.0,
      "step": 54905
    },
    {
      "entropy": 5.904357099533081,
      "epoch": 4.798566809403129,
      "grad_norm": 1.125,
      "learning_rate": 0.0002924389397437537,
      "loss": 5.2535,
      "mean_token_accuracy": 0.21509561985731124,
      "num_tokens": 111269356.0,
      "step": 54910
    },
    {
      "entropy": 5.904075241088867,
      "epoch": 4.799003757755833,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029240787221572153,
      "loss": 5.189,
      "mean_token_accuracy": 0.2144182652235031,
      "num_tokens": 111278550.0,
      "step": 54915
    },
    {
      "entropy": 5.853848934173584,
      "epoch": 4.799440706108538,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002923768043537972,
      "loss": 5.2845,
      "mean_token_accuracy": 0.20663340538740158,
      "num_tokens": 111289231.0,
      "step": 54920
    },
    {
      "entropy": 5.87928729057312,
      "epoch": 4.799877654461243,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029234573615857676,
      "loss": 5.3257,
      "mean_token_accuracy": 0.20140114724636077,
      "num_tokens": 111298401.0,
      "step": 54925
    },
    {
      "entropy": 5.8831220149993895,
      "epoch": 4.800314602813947,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029231466763065595,
      "loss": 5.2931,
      "mean_token_accuracy": 0.21084158569574357,
      "num_tokens": 111309752.0,
      "step": 54930
    },
    {
      "entropy": 5.880131483078003,
      "epoch": 4.800751551166652,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002922835987706308,
      "loss": 5.2614,
      "mean_token_accuracy": 0.20736350119113922,
      "num_tokens": 111319852.0,
      "step": 54935
    },
    {
      "entropy": 5.834808731079102,
      "epoch": 4.801188499519357,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002922525295790971,
      "loss": 5.2182,
      "mean_token_accuracy": 0.21616055369377135,
      "num_tokens": 111329465.0,
      "step": 54940
    },
    {
      "entropy": 5.875842714309693,
      "epoch": 4.801625447872062,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002922214600566508,
      "loss": 5.3018,
      "mean_token_accuracy": 0.20215645730495452,
      "num_tokens": 111338949.0,
      "step": 54945
    },
    {
      "entropy": 5.942536020278931,
      "epoch": 4.802062396224766,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029219039020388803,
      "loss": 5.3608,
      "mean_token_accuracy": 0.2023554190993309,
      "num_tokens": 111349451.0,
      "step": 54950
    },
    {
      "entropy": 5.832452297210693,
      "epoch": 4.802499344577471,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002921593200214045,
      "loss": 5.2074,
      "mean_token_accuracy": 0.21552193015813828,
      "num_tokens": 111359349.0,
      "step": 54955
    },
    {
      "entropy": 5.877748155593872,
      "epoch": 4.802936292930176,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002921282495097962,
      "loss": 5.324,
      "mean_token_accuracy": 0.20693793445825576,
      "num_tokens": 111369618.0,
      "step": 54960
    },
    {
      "entropy": 5.869707155227661,
      "epoch": 4.8033732412828805,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029209717866965904,
      "loss": 5.2617,
      "mean_token_accuracy": 0.21428151726722716,
      "num_tokens": 111379395.0,
      "step": 54965
    },
    {
      "entropy": 5.9124139785766605,
      "epoch": 4.803810189635585,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002920661075015891,
      "loss": 5.3709,
      "mean_token_accuracy": 0.20548734217882156,
      "num_tokens": 111389722.0,
      "step": 54970
    },
    {
      "entropy": 5.872347164154053,
      "epoch": 4.804247137988289,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002920350360061823,
      "loss": 5.2888,
      "mean_token_accuracy": 0.20584156960248948,
      "num_tokens": 111400470.0,
      "step": 54975
    },
    {
      "entropy": 5.885905408859253,
      "epoch": 4.804684086340995,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00029200396418403463,
      "loss": 5.2728,
      "mean_token_accuracy": 0.2059742659330368,
      "num_tokens": 111410296.0,
      "step": 54980
    },
    {
      "entropy": 5.8593768119812015,
      "epoch": 4.805121034693699,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002919728920357419,
      "loss": 5.2287,
      "mean_token_accuracy": 0.21886928528547286,
      "num_tokens": 111421790.0,
      "step": 54985
    },
    {
      "entropy": 5.869637155532837,
      "epoch": 4.805557983046404,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029194181956190027,
      "loss": 5.3587,
      "mean_token_accuracy": 0.20584065467119217,
      "num_tokens": 111431764.0,
      "step": 54990
    },
    {
      "entropy": 5.874060153961182,
      "epoch": 4.805994931399109,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002919107467631056,
      "loss": 5.2928,
      "mean_token_accuracy": 0.209069362282753,
      "num_tokens": 111441334.0,
      "step": 54995
    },
    {
      "entropy": 5.895236015319824,
      "epoch": 4.8064318797518135,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000291879673639954,
      "loss": 5.3435,
      "mean_token_accuracy": 0.2068013310432434,
      "num_tokens": 111451123.0,
      "step": 55000
    },
    {
      "entropy": 5.870291614532471,
      "epoch": 4.806868828104518,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029184860019304136,
      "loss": 5.2419,
      "mean_token_accuracy": 0.21898613423109053,
      "num_tokens": 111461157.0,
      "step": 55005
    },
    {
      "entropy": 5.883331155776977,
      "epoch": 4.8073057764572225,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002918175264229638,
      "loss": 5.3008,
      "mean_token_accuracy": 0.20544177293777466,
      "num_tokens": 111471164.0,
      "step": 55010
    },
    {
      "entropy": 5.9291339874267575,
      "epoch": 4.807742724809928,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0002917864523303172,
      "loss": 5.4153,
      "mean_token_accuracy": 0.19793327897787094,
      "num_tokens": 111481870.0,
      "step": 55015
    },
    {
      "entropy": 5.871652460098266,
      "epoch": 4.808179673162632,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029175537791569775,
      "loss": 5.2884,
      "mean_token_accuracy": 0.21620973348617553,
      "num_tokens": 111491321.0,
      "step": 55020
    },
    {
      "entropy": 5.881215333938599,
      "epoch": 4.808616621515337,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029172430317970133,
      "loss": 5.2705,
      "mean_token_accuracy": 0.2052223041653633,
      "num_tokens": 111500518.0,
      "step": 55025
    },
    {
      "entropy": 5.870839357376099,
      "epoch": 4.809053569868041,
      "grad_norm": 1.109375,
      "learning_rate": 0.000291693228122924,
      "loss": 5.2284,
      "mean_token_accuracy": 0.21155571341514587,
      "num_tokens": 111510452.0,
      "step": 55030
    },
    {
      "entropy": 5.798665237426758,
      "epoch": 4.809490518220747,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029166215274596186,
      "loss": 5.1435,
      "mean_token_accuracy": 0.21780920773744583,
      "num_tokens": 111520930.0,
      "step": 55035
    },
    {
      "entropy": 5.88686089515686,
      "epoch": 4.809927466573451,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029163107704941084,
      "loss": 5.3092,
      "mean_token_accuracy": 0.20108796358108522,
      "num_tokens": 111531704.0,
      "step": 55040
    },
    {
      "entropy": 5.900704288482666,
      "epoch": 4.8103644149261555,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029160000103386704,
      "loss": 5.2573,
      "mean_token_accuracy": 0.2065522089600563,
      "num_tokens": 111542015.0,
      "step": 55045
    },
    {
      "entropy": 5.805512237548828,
      "epoch": 4.810801363278861,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029156892469992657,
      "loss": 5.224,
      "mean_token_accuracy": 0.21934160590171814,
      "num_tokens": 111551198.0,
      "step": 55050
    },
    {
      "entropy": 5.85687141418457,
      "epoch": 4.811238311631565,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002915378480481854,
      "loss": 5.231,
      "mean_token_accuracy": 0.21013998091220856,
      "num_tokens": 111561872.0,
      "step": 55055
    },
    {
      "entropy": 5.940856218338013,
      "epoch": 4.81167525998427,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002915067710792396,
      "loss": 5.3609,
      "mean_token_accuracy": 0.19844009429216386,
      "num_tokens": 111572686.0,
      "step": 55060
    },
    {
      "entropy": 5.877993106842041,
      "epoch": 4.812112208336974,
      "grad_norm": 1.125,
      "learning_rate": 0.0002914756937936854,
      "loss": 5.2821,
      "mean_token_accuracy": 0.20672052055597306,
      "num_tokens": 111582656.0,
      "step": 55065
    },
    {
      "entropy": 5.902146339416504,
      "epoch": 4.81254915668968,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002914446161921187,
      "loss": 5.313,
      "mean_token_accuracy": 0.20470982044935226,
      "num_tokens": 111594182.0,
      "step": 55070
    },
    {
      "entropy": 5.892113256454468,
      "epoch": 4.812986105042384,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002914135382751356,
      "loss": 5.3179,
      "mean_token_accuracy": 0.20376531779766083,
      "num_tokens": 111605174.0,
      "step": 55075
    },
    {
      "entropy": 5.940579128265381,
      "epoch": 4.813423053395089,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002913824600433323,
      "loss": 5.3214,
      "mean_token_accuracy": 0.20911879688501359,
      "num_tokens": 111617489.0,
      "step": 55080
    },
    {
      "entropy": 5.9578718662261965,
      "epoch": 4.813860001747793,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029135138149730485,
      "loss": 5.3786,
      "mean_token_accuracy": 0.19748386591672898,
      "num_tokens": 111627250.0,
      "step": 55085
    },
    {
      "entropy": 5.90429916381836,
      "epoch": 4.814296950100498,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029132030263764923,
      "loss": 5.2808,
      "mean_token_accuracy": 0.20491136312484742,
      "num_tokens": 111637726.0,
      "step": 55090
    },
    {
      "entropy": 5.873023509979248,
      "epoch": 4.814733898453203,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002912892234649617,
      "loss": 5.2164,
      "mean_token_accuracy": 0.2061282992362976,
      "num_tokens": 111647630.0,
      "step": 55095
    },
    {
      "entropy": 5.844771766662598,
      "epoch": 4.815170846805907,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002912581439798384,
      "loss": 5.2743,
      "mean_token_accuracy": 0.20737397223711013,
      "num_tokens": 111657428.0,
      "step": 55100
    },
    {
      "entropy": 5.959333086013794,
      "epoch": 4.815607795158613,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002912270641828753,
      "loss": 5.3624,
      "mean_token_accuracy": 0.203133662045002,
      "num_tokens": 111668614.0,
      "step": 55105
    },
    {
      "entropy": 5.828033828735352,
      "epoch": 4.816044743511317,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029119598407466866,
      "loss": 5.1896,
      "mean_token_accuracy": 0.21756235659122466,
      "num_tokens": 111677506.0,
      "step": 55110
    },
    {
      "entropy": 5.849728393554687,
      "epoch": 4.816481691864022,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002911649036558145,
      "loss": 5.2488,
      "mean_token_accuracy": 0.21182367503643035,
      "num_tokens": 111686316.0,
      "step": 55115
    },
    {
      "entropy": 5.856533002853394,
      "epoch": 4.816918640216726,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002911338229269091,
      "loss": 5.2622,
      "mean_token_accuracy": 0.21457097679376602,
      "num_tokens": 111696788.0,
      "step": 55120
    },
    {
      "entropy": 5.878426170349121,
      "epoch": 4.8173555885694315,
      "grad_norm": 1.0,
      "learning_rate": 0.00029110274188854843,
      "loss": 5.2103,
      "mean_token_accuracy": 0.2157828390598297,
      "num_tokens": 111706868.0,
      "step": 55125
    },
    {
      "entropy": 5.905268669128418,
      "epoch": 4.817792536922136,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002910716605413288,
      "loss": 5.3199,
      "mean_token_accuracy": 0.20501870959997176,
      "num_tokens": 111716401.0,
      "step": 55130
    },
    {
      "entropy": 5.834462404251099,
      "epoch": 4.81822948527484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002910405788858462,
      "loss": 5.242,
      "mean_token_accuracy": 0.21797147542238235,
      "num_tokens": 111725797.0,
      "step": 55135
    },
    {
      "entropy": 5.767148208618164,
      "epoch": 4.818666433627545,
      "grad_norm": 1.09375,
      "learning_rate": 0.000291009496922697,
      "loss": 5.1944,
      "mean_token_accuracy": 0.21483552008867263,
      "num_tokens": 111735906.0,
      "step": 55140
    },
    {
      "entropy": 5.767161464691162,
      "epoch": 4.81910338198025,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002909784146524773,
      "loss": 5.1949,
      "mean_token_accuracy": 0.21783833503723143,
      "num_tokens": 111745368.0,
      "step": 55145
    },
    {
      "entropy": 5.8548276901245115,
      "epoch": 4.819540330332955,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00029094733207578316,
      "loss": 5.3158,
      "mean_token_accuracy": 0.2073303073644638,
      "num_tokens": 111755196.0,
      "step": 55150
    },
    {
      "entropy": 5.882082414627075,
      "epoch": 4.819977278685659,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002909162491932108,
      "loss": 5.2841,
      "mean_token_accuracy": 0.20820630639791488,
      "num_tokens": 111765617.0,
      "step": 55155
    },
    {
      "entropy": 5.888442325592041,
      "epoch": 4.8204142270383645,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029088516600535644,
      "loss": 5.3151,
      "mean_token_accuracy": 0.21001758724451064,
      "num_tokens": 111776667.0,
      "step": 55160
    },
    {
      "entropy": 5.875148868560791,
      "epoch": 4.820851175391069,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002908540825128163,
      "loss": 5.2454,
      "mean_token_accuracy": 0.21291887760162354,
      "num_tokens": 111785558.0,
      "step": 55165
    },
    {
      "entropy": 5.903106355667115,
      "epoch": 4.8212881237437735,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002908229987161866,
      "loss": 5.3093,
      "mean_token_accuracy": 0.20741340070962905,
      "num_tokens": 111795406.0,
      "step": 55170
    },
    {
      "entropy": 5.9190246105194095,
      "epoch": 4.821725072096478,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029079191461606343,
      "loss": 5.3488,
      "mean_token_accuracy": 0.2003259316086769,
      "num_tokens": 111806426.0,
      "step": 55175
    },
    {
      "entropy": 5.880667304992675,
      "epoch": 4.822162020449183,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000290760830213043,
      "loss": 5.2975,
      "mean_token_accuracy": 0.20913487523794175,
      "num_tokens": 111816140.0,
      "step": 55180
    },
    {
      "entropy": 5.919016551971436,
      "epoch": 4.822598968801888,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002907297455077217,
      "loss": 5.2726,
      "mean_token_accuracy": 0.2129566803574562,
      "num_tokens": 111826349.0,
      "step": 55185
    },
    {
      "entropy": 5.808802461624145,
      "epoch": 4.823035917154592,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002906986605006956,
      "loss": 5.1851,
      "mean_token_accuracy": 0.2164750173687935,
      "num_tokens": 111837164.0,
      "step": 55190
    },
    {
      "entropy": 5.802021026611328,
      "epoch": 4.823472865507297,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029066757519256105,
      "loss": 5.1964,
      "mean_token_accuracy": 0.21165265887975693,
      "num_tokens": 111847268.0,
      "step": 55195
    },
    {
      "entropy": 5.886240720748901,
      "epoch": 4.823909813860002,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029063648958391403,
      "loss": 5.2744,
      "mean_token_accuracy": 0.2043464347720146,
      "num_tokens": 111856154.0,
      "step": 55200
    },
    {
      "entropy": 5.799288368225097,
      "epoch": 4.8243467622127065,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002906054036753511,
      "loss": 5.244,
      "mean_token_accuracy": 0.19886420667171478,
      "num_tokens": 111865903.0,
      "step": 55205
    },
    {
      "entropy": 5.887459802627563,
      "epoch": 4.824783710565411,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029057431746746826,
      "loss": 5.2666,
      "mean_token_accuracy": 0.20236852765083313,
      "num_tokens": 111875873.0,
      "step": 55210
    },
    {
      "entropy": 5.865136051177979,
      "epoch": 4.825220658918116,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029054323096086186,
      "loss": 5.2679,
      "mean_token_accuracy": 0.2065800577402115,
      "num_tokens": 111886874.0,
      "step": 55215
    },
    {
      "entropy": 5.8690488815307615,
      "epoch": 4.825657607270821,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002905121441561282,
      "loss": 5.2892,
      "mean_token_accuracy": 0.21093617528676986,
      "num_tokens": 111896889.0,
      "step": 55220
    },
    {
      "entropy": 5.867450094223022,
      "epoch": 4.826094555623525,
      "grad_norm": 1.125,
      "learning_rate": 0.00029048105705386346,
      "loss": 5.2443,
      "mean_token_accuracy": 0.20267465561628342,
      "num_tokens": 111906155.0,
      "step": 55225
    },
    {
      "entropy": 5.860227108001709,
      "epoch": 4.82653150397623,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029044996965466393,
      "loss": 5.2573,
      "mean_token_accuracy": 0.21327962428331376,
      "num_tokens": 111915244.0,
      "step": 55230
    },
    {
      "entropy": 5.84359769821167,
      "epoch": 4.826968452328934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029041888195912587,
      "loss": 5.3139,
      "mean_token_accuracy": 0.2106443852186203,
      "num_tokens": 111925576.0,
      "step": 55235
    },
    {
      "entropy": 5.879868745803833,
      "epoch": 4.82740540068164,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0002903877939678456,
      "loss": 5.283,
      "mean_token_accuracy": 0.20918617993593216,
      "num_tokens": 111937015.0,
      "step": 55240
    },
    {
      "entropy": 5.885421228408814,
      "epoch": 4.827842349034344,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029035670568141936,
      "loss": 5.3146,
      "mean_token_accuracy": 0.20686940252780914,
      "num_tokens": 111946730.0,
      "step": 55245
    },
    {
      "entropy": 5.897366905212403,
      "epoch": 4.8282792973870485,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002903256171004434,
      "loss": 5.2369,
      "mean_token_accuracy": 0.21303732544183732,
      "num_tokens": 111956026.0,
      "step": 55250
    },
    {
      "entropy": 5.8556060791015625,
      "epoch": 4.828716245739754,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002902945282255142,
      "loss": 5.2993,
      "mean_token_accuracy": 0.21106134951114655,
      "num_tokens": 111966453.0,
      "step": 55255
    },
    {
      "entropy": 5.861486911773682,
      "epoch": 4.829153194092458,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002902634390572279,
      "loss": 5.2356,
      "mean_token_accuracy": 0.20740047991275787,
      "num_tokens": 111977318.0,
      "step": 55260
    },
    {
      "entropy": 5.8886782169342045,
      "epoch": 4.829590142445163,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0002902323495961808,
      "loss": 5.3252,
      "mean_token_accuracy": 0.20337578803300857,
      "num_tokens": 111987448.0,
      "step": 55265
    },
    {
      "entropy": 5.9322474002838135,
      "epoch": 4.830027090797868,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029020125984296927,
      "loss": 5.3208,
      "mean_token_accuracy": 0.2006827786564827,
      "num_tokens": 111997222.0,
      "step": 55270
    },
    {
      "entropy": 5.944171714782715,
      "epoch": 4.830464039150573,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0002901701697981896,
      "loss": 5.2868,
      "mean_token_accuracy": 0.209459887444973,
      "num_tokens": 112007537.0,
      "step": 55275
    },
    {
      "entropy": 5.83535737991333,
      "epoch": 4.830900987503277,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002901390794624381,
      "loss": 5.2336,
      "mean_token_accuracy": 0.20735537260770798,
      "num_tokens": 112017511.0,
      "step": 55280
    },
    {
      "entropy": 5.806063842773438,
      "epoch": 4.831337935855982,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002901079888363112,
      "loss": 5.1953,
      "mean_token_accuracy": 0.21436322927474977,
      "num_tokens": 112027139.0,
      "step": 55285
    },
    {
      "entropy": 5.8414891242980955,
      "epoch": 4.831774884208686,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002900768979204051,
      "loss": 5.2674,
      "mean_token_accuracy": 0.21500944197177888,
      "num_tokens": 112037643.0,
      "step": 55290
    },
    {
      "entropy": 5.850915145874024,
      "epoch": 4.832211832561391,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029004580671531624,
      "loss": 5.2621,
      "mean_token_accuracy": 0.21409803181886672,
      "num_tokens": 112048763.0,
      "step": 55295
    },
    {
      "entropy": 5.891600656509399,
      "epoch": 4.832648780914096,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002900147152216408,
      "loss": 5.3058,
      "mean_token_accuracy": 0.20355388522148132,
      "num_tokens": 112058883.0,
      "step": 55300
    },
    {
      "entropy": 5.849521636962891,
      "epoch": 4.8330857292668,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028998362343997543,
      "loss": 5.2655,
      "mean_token_accuracy": 0.2059768706560135,
      "num_tokens": 112069038.0,
      "step": 55305
    },
    {
      "entropy": 5.8833253383636475,
      "epoch": 4.833522677619506,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002899525313709162,
      "loss": 5.3069,
      "mean_token_accuracy": 0.20507784336805343,
      "num_tokens": 112078599.0,
      "step": 55310
    },
    {
      "entropy": 5.9492600440979,
      "epoch": 4.83395962597221,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028992143901505955,
      "loss": 5.2883,
      "mean_token_accuracy": 0.2053718239068985,
      "num_tokens": 112088671.0,
      "step": 55315
    },
    {
      "entropy": 5.906651258468628,
      "epoch": 4.834396574324915,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028989034637300206,
      "loss": 5.2929,
      "mean_token_accuracy": 0.20827051401138305,
      "num_tokens": 112098428.0,
      "step": 55320
    },
    {
      "entropy": 5.8451635360717775,
      "epoch": 4.834833522677619,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028985925344533975,
      "loss": 5.251,
      "mean_token_accuracy": 0.21501570641994477,
      "num_tokens": 112109271.0,
      "step": 55325
    },
    {
      "entropy": 5.8306769847869875,
      "epoch": 4.8352704710303245,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002898281602326692,
      "loss": 5.2102,
      "mean_token_accuracy": 0.21473384499549866,
      "num_tokens": 112118755.0,
      "step": 55330
    },
    {
      "entropy": 5.911168766021729,
      "epoch": 4.835707419383029,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002897970667355868,
      "loss": 5.3529,
      "mean_token_accuracy": 0.20371102541685104,
      "num_tokens": 112129628.0,
      "step": 55335
    },
    {
      "entropy": 5.898406839370727,
      "epoch": 4.836144367735733,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002897659729546889,
      "loss": 5.3102,
      "mean_token_accuracy": 0.20583580136299134,
      "num_tokens": 112140188.0,
      "step": 55340
    },
    {
      "entropy": 5.798867416381836,
      "epoch": 4.836581316088438,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002897348788905719,
      "loss": 5.2363,
      "mean_token_accuracy": 0.21885496824979783,
      "num_tokens": 112149807.0,
      "step": 55345
    },
    {
      "entropy": 5.837567758560181,
      "epoch": 4.837018264441143,
      "grad_norm": 1.125,
      "learning_rate": 0.00028970378454383223,
      "loss": 5.2744,
      "mean_token_accuracy": 0.21210371404886247,
      "num_tokens": 112160531.0,
      "step": 55350
    },
    {
      "entropy": 5.8688530921936035,
      "epoch": 4.837455212793848,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028967268991506625,
      "loss": 5.2703,
      "mean_token_accuracy": 0.2128053143620491,
      "num_tokens": 112171173.0,
      "step": 55355
    },
    {
      "entropy": 5.902270650863647,
      "epoch": 4.837892161146552,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028964159500487034,
      "loss": 5.3102,
      "mean_token_accuracy": 0.20407804399728774,
      "num_tokens": 112181345.0,
      "step": 55360
    },
    {
      "entropy": 5.825448083877563,
      "epoch": 4.8383291094992575,
      "grad_norm": 1.015625,
      "learning_rate": 0.000289610499813841,
      "loss": 5.2236,
      "mean_token_accuracy": 0.21367825716733932,
      "num_tokens": 112190798.0,
      "step": 55365
    },
    {
      "entropy": 5.861500358581543,
      "epoch": 4.838766057851962,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002895794043425746,
      "loss": 5.2367,
      "mean_token_accuracy": 0.2120460346341133,
      "num_tokens": 112199836.0,
      "step": 55370
    },
    {
      "entropy": 5.830715274810791,
      "epoch": 4.8392030062046665,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00028954830859166767,
      "loss": 5.2038,
      "mean_token_accuracy": 0.2163579374551773,
      "num_tokens": 112209885.0,
      "step": 55375
    },
    {
      "entropy": 5.895305252075195,
      "epoch": 4.839639954557371,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028951721256171653,
      "loss": 5.2734,
      "mean_token_accuracy": 0.20377109497785567,
      "num_tokens": 112219775.0,
      "step": 55380
    },
    {
      "entropy": 5.8472637176513675,
      "epoch": 4.840076902910076,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028948611625331756,
      "loss": 5.2709,
      "mean_token_accuracy": 0.21497959345579148,
      "num_tokens": 112230354.0,
      "step": 55385
    },
    {
      "entropy": 5.884412860870361,
      "epoch": 4.840513851262781,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002894550196670674,
      "loss": 5.2899,
      "mean_token_accuracy": 0.20500167906284333,
      "num_tokens": 112240821.0,
      "step": 55390
    },
    {
      "entropy": 5.9183002471923825,
      "epoch": 4.840950799615485,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028942392280356227,
      "loss": 5.2791,
      "mean_token_accuracy": 0.20533842742443084,
      "num_tokens": 112250561.0,
      "step": 55395
    },
    {
      "entropy": 5.861183023452758,
      "epoch": 4.84138774796819,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028939282566339884,
      "loss": 5.2154,
      "mean_token_accuracy": 0.2188319206237793,
      "num_tokens": 112261829.0,
      "step": 55400
    },
    {
      "entropy": 5.866769504547119,
      "epoch": 4.841824696320895,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002893617282471734,
      "loss": 5.2697,
      "mean_token_accuracy": 0.2062443196773529,
      "num_tokens": 112272277.0,
      "step": 55405
    },
    {
      "entropy": 5.883980751037598,
      "epoch": 4.8422616446735995,
      "grad_norm": 1.0,
      "learning_rate": 0.0002893306305554825,
      "loss": 5.2593,
      "mean_token_accuracy": 0.21428372859954833,
      "num_tokens": 112283629.0,
      "step": 55410
    },
    {
      "entropy": 5.880274105072021,
      "epoch": 4.842698593026304,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028929953258892264,
      "loss": 5.2443,
      "mean_token_accuracy": 0.2133172407746315,
      "num_tokens": 112293278.0,
      "step": 55415
    },
    {
      "entropy": 5.8834082126617435,
      "epoch": 4.843135541379009,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028926843434809023,
      "loss": 5.2053,
      "mean_token_accuracy": 0.21185715943574907,
      "num_tokens": 112303293.0,
      "step": 55420
    },
    {
      "entropy": 5.797517538070679,
      "epoch": 4.843572489731714,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002892373358335819,
      "loss": 5.2086,
      "mean_token_accuracy": 0.21600128412246705,
      "num_tokens": 112313659.0,
      "step": 55425
    },
    {
      "entropy": 5.9375334739685055,
      "epoch": 4.844009438084418,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028920623704599386,
      "loss": 5.369,
      "mean_token_accuracy": 0.20393929183483123,
      "num_tokens": 112324923.0,
      "step": 55430
    },
    {
      "entropy": 5.874921464920044,
      "epoch": 4.844446386437123,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002891751379859228,
      "loss": 5.2443,
      "mean_token_accuracy": 0.2113190174102783,
      "num_tokens": 112334803.0,
      "step": 55435
    },
    {
      "entropy": 5.783056831359863,
      "epoch": 4.844883334789828,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002891440386539651,
      "loss": 5.2405,
      "mean_token_accuracy": 0.21791887730360032,
      "num_tokens": 112344909.0,
      "step": 55440
    },
    {
      "entropy": 5.915611267089844,
      "epoch": 4.845320283142533,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028911293905071735,
      "loss": 5.232,
      "mean_token_accuracy": 0.21088257282972336,
      "num_tokens": 112355191.0,
      "step": 55445
    },
    {
      "entropy": 5.964656496047974,
      "epoch": 4.845757231495237,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028908183917677603,
      "loss": 5.2806,
      "mean_token_accuracy": 0.2125750795006752,
      "num_tokens": 112365055.0,
      "step": 55450
    },
    {
      "entropy": 5.911442422866822,
      "epoch": 4.8461941798479415,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002890507390327377,
      "loss": 5.3334,
      "mean_token_accuracy": 0.20106270164251328,
      "num_tokens": 112375775.0,
      "step": 55455
    },
    {
      "entropy": 5.890569972991943,
      "epoch": 4.846631128200647,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028901963861919876,
      "loss": 5.2967,
      "mean_token_accuracy": 0.20648290812969208,
      "num_tokens": 112385734.0,
      "step": 55460
    },
    {
      "entropy": 5.840050220489502,
      "epoch": 4.847068076553351,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002889885379367559,
      "loss": 5.1868,
      "mean_token_accuracy": 0.2111409440636635,
      "num_tokens": 112396212.0,
      "step": 55465
    },
    {
      "entropy": 5.86109676361084,
      "epoch": 4.847505024906056,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002889574369860055,
      "loss": 5.3134,
      "mean_token_accuracy": 0.20373644828796386,
      "num_tokens": 112405982.0,
      "step": 55470
    },
    {
      "entropy": 5.925565338134765,
      "epoch": 4.847941973258761,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002889263357675442,
      "loss": 5.2991,
      "mean_token_accuracy": 0.2071339577436447,
      "num_tokens": 112415976.0,
      "step": 55475
    },
    {
      "entropy": 5.902926301956176,
      "epoch": 4.848378921611466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002888952342819684,
      "loss": 5.2819,
      "mean_token_accuracy": 0.20524488389492035,
      "num_tokens": 112426284.0,
      "step": 55480
    },
    {
      "entropy": 5.841900873184204,
      "epoch": 4.84881586996417,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028886413252987483,
      "loss": 5.1668,
      "mean_token_accuracy": 0.2092384874820709,
      "num_tokens": 112436467.0,
      "step": 55485
    },
    {
      "entropy": 5.842377138137818,
      "epoch": 4.849252818316875,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028883303051185975,
      "loss": 5.2403,
      "mean_token_accuracy": 0.2082191750407219,
      "num_tokens": 112446086.0,
      "step": 55490
    },
    {
      "entropy": 5.85979151725769,
      "epoch": 4.84968976666958,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028880192822852,
      "loss": 5.28,
      "mean_token_accuracy": 0.2095283403992653,
      "num_tokens": 112456202.0,
      "step": 55495
    },
    {
      "entropy": 5.872143650054932,
      "epoch": 4.850126715022284,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002887708256804521,
      "loss": 5.2723,
      "mean_token_accuracy": 0.20677807480096816,
      "num_tokens": 112465836.0,
      "step": 55500
    },
    {
      "entropy": 5.85922384262085,
      "epoch": 4.850563663374989,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002887397228682525,
      "loss": 5.2942,
      "mean_token_accuracy": 0.20710992068052292,
      "num_tokens": 112475966.0,
      "step": 55505
    },
    {
      "entropy": 5.981885623931885,
      "epoch": 4.851000611727693,
      "grad_norm": 1.125,
      "learning_rate": 0.0002887086197925178,
      "loss": 5.4184,
      "mean_token_accuracy": 0.19179779142141343,
      "num_tokens": 112485232.0,
      "step": 55510
    },
    {
      "entropy": 5.850045013427734,
      "epoch": 4.851437560080399,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002886775164538446,
      "loss": 5.2502,
      "mean_token_accuracy": 0.20540927946567536,
      "num_tokens": 112495954.0,
      "step": 55515
    },
    {
      "entropy": 5.8222880363464355,
      "epoch": 4.851874508433103,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028864641285282944,
      "loss": 5.2223,
      "mean_token_accuracy": 0.214474219083786,
      "num_tokens": 112505492.0,
      "step": 55520
    },
    {
      "entropy": 5.8251501560211185,
      "epoch": 4.852311456785808,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000288615308990069,
      "loss": 5.256,
      "mean_token_accuracy": 0.21418195068836213,
      "num_tokens": 112515216.0,
      "step": 55525
    },
    {
      "entropy": 5.876076507568359,
      "epoch": 4.852748405138513,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002885842048661598,
      "loss": 5.1948,
      "mean_token_accuracy": 0.21582616865634918,
      "num_tokens": 112524797.0,
      "step": 55530
    },
    {
      "entropy": 5.882627344131469,
      "epoch": 4.8531853534912175,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002885531004816983,
      "loss": 5.306,
      "mean_token_accuracy": 0.20937279462814332,
      "num_tokens": 112535057.0,
      "step": 55535
    },
    {
      "entropy": 5.915195369720459,
      "epoch": 4.853622301843922,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002885219958372813,
      "loss": 5.2987,
      "mean_token_accuracy": 0.21249077916145326,
      "num_tokens": 112545926.0,
      "step": 55540
    },
    {
      "entropy": 5.865259504318237,
      "epoch": 4.854059250196626,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002884908909335054,
      "loss": 5.2613,
      "mean_token_accuracy": 0.20950970351696013,
      "num_tokens": 112556367.0,
      "step": 55545
    },
    {
      "entropy": 5.929406976699829,
      "epoch": 4.854496198549332,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002884597857709671,
      "loss": 5.3336,
      "mean_token_accuracy": 0.2120959386229515,
      "num_tokens": 112566862.0,
      "step": 55550
    },
    {
      "entropy": 5.882200479507446,
      "epoch": 4.854933146902036,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000288428680350263,
      "loss": 5.3028,
      "mean_token_accuracy": 0.20537632405757905,
      "num_tokens": 112577306.0,
      "step": 55555
    },
    {
      "entropy": 5.917066192626953,
      "epoch": 4.855370095254741,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028839757467198987,
      "loss": 5.2914,
      "mean_token_accuracy": 0.21145587861537934,
      "num_tokens": 112587865.0,
      "step": 55560
    },
    {
      "entropy": 5.892334127426148,
      "epoch": 4.855807043607445,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002883664687367442,
      "loss": 5.2884,
      "mean_token_accuracy": 0.20885468721389772,
      "num_tokens": 112597103.0,
      "step": 55565
    },
    {
      "entropy": 5.867527151107788,
      "epoch": 4.8562439919601506,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028833536254512264,
      "loss": 5.2259,
      "mean_token_accuracy": 0.21312750726938248,
      "num_tokens": 112606971.0,
      "step": 55570
    },
    {
      "entropy": 5.929556560516358,
      "epoch": 4.856680940312855,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002883042560977219,
      "loss": 5.3774,
      "mean_token_accuracy": 0.19978595674037933,
      "num_tokens": 112617588.0,
      "step": 55575
    },
    {
      "entropy": 5.929070997238159,
      "epoch": 4.8571178886655595,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028827314939513843,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21869638413190842,
      "num_tokens": 112627854.0,
      "step": 55580
    },
    {
      "entropy": 5.885299301147461,
      "epoch": 4.857554837018265,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028824204243796907,
      "loss": 5.3081,
      "mean_token_accuracy": 0.20784508883953096,
      "num_tokens": 112637969.0,
      "step": 55585
    },
    {
      "entropy": 5.890111589431763,
      "epoch": 4.857991785370969,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028821093522681044,
      "loss": 5.2888,
      "mean_token_accuracy": 0.2058829739689827,
      "num_tokens": 112648482.0,
      "step": 55590
    },
    {
      "entropy": 5.882221031188965,
      "epoch": 4.858428733723674,
      "grad_norm": 1.125,
      "learning_rate": 0.0002881798277622591,
      "loss": 5.2796,
      "mean_token_accuracy": 0.21162212193012236,
      "num_tokens": 112658256.0,
      "step": 55595
    },
    {
      "entropy": 5.829293727874756,
      "epoch": 4.858865682076378,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002881487200449118,
      "loss": 5.1741,
      "mean_token_accuracy": 0.22234107553958893,
      "num_tokens": 112668301.0,
      "step": 55600
    },
    {
      "entropy": 5.898016548156738,
      "epoch": 4.859302630429084,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002881176120753651,
      "loss": 5.2537,
      "mean_token_accuracy": 0.21123354583978654,
      "num_tokens": 112679215.0,
      "step": 55605
    },
    {
      "entropy": 5.96496262550354,
      "epoch": 4.859739578781788,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028808650385421573,
      "loss": 5.3518,
      "mean_token_accuracy": 0.19665527045726777,
      "num_tokens": 112689304.0,
      "step": 55610
    },
    {
      "entropy": 5.856265544891357,
      "epoch": 4.8601765271344926,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028805539538206043,
      "loss": 5.2013,
      "mean_token_accuracy": 0.21050291508436203,
      "num_tokens": 112698726.0,
      "step": 55615
    },
    {
      "entropy": 5.883113813400269,
      "epoch": 4.860613475487197,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028802428665949584,
      "loss": 5.2862,
      "mean_token_accuracy": 0.2032075896859169,
      "num_tokens": 112708500.0,
      "step": 55620
    },
    {
      "entropy": 5.844793367385864,
      "epoch": 4.861050423839902,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028799317768711844,
      "loss": 5.3595,
      "mean_token_accuracy": 0.1994740828871727,
      "num_tokens": 112719195.0,
      "step": 55625
    },
    {
      "entropy": 5.88963565826416,
      "epoch": 4.861487372192607,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002879620684655252,
      "loss": 5.303,
      "mean_token_accuracy": 0.20484840720891953,
      "num_tokens": 112728610.0,
      "step": 55630
    },
    {
      "entropy": 5.8642274856567385,
      "epoch": 4.861924320545311,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002879309589953126,
      "loss": 5.2597,
      "mean_token_accuracy": 0.20465251356363295,
      "num_tokens": 112738818.0,
      "step": 55635
    },
    {
      "entropy": 5.8765928745269775,
      "epoch": 4.862361268898017,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002878998492770776,
      "loss": 5.3117,
      "mean_token_accuracy": 0.20391668677330016,
      "num_tokens": 112748965.0,
      "step": 55640
    },
    {
      "entropy": 5.9306073665618895,
      "epoch": 4.862798217250721,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028786873931141655,
      "loss": 5.3176,
      "mean_token_accuracy": 0.20539166629314423,
      "num_tokens": 112758638.0,
      "step": 55645
    },
    {
      "entropy": 5.860184049606323,
      "epoch": 4.863235165603426,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002878376290989264,
      "loss": 5.2667,
      "mean_token_accuracy": 0.21190934330224992,
      "num_tokens": 112768891.0,
      "step": 55650
    },
    {
      "entropy": 5.909431028366089,
      "epoch": 4.86367211395613,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028780651864020385,
      "loss": 5.2558,
      "mean_token_accuracy": 0.20373493283987046,
      "num_tokens": 112779938.0,
      "step": 55655
    },
    {
      "entropy": 5.893035173416138,
      "epoch": 4.864109062308835,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002877754079358455,
      "loss": 5.3455,
      "mean_token_accuracy": 0.20286061614751816,
      "num_tokens": 112791601.0,
      "step": 55660
    },
    {
      "entropy": 5.853157567977905,
      "epoch": 4.86454601066154,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002877442969864482,
      "loss": 5.2258,
      "mean_token_accuracy": 0.21361281126737594,
      "num_tokens": 112802502.0,
      "step": 55665
    },
    {
      "entropy": 5.885803127288819,
      "epoch": 4.864982959014244,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028771318579260855,
      "loss": 5.2268,
      "mean_token_accuracy": 0.2141124650835991,
      "num_tokens": 112812336.0,
      "step": 55670
    },
    {
      "entropy": 5.827940034866333,
      "epoch": 4.865419907366949,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002876820743549234,
      "loss": 5.2515,
      "mean_token_accuracy": 0.21219036728143692,
      "num_tokens": 112822572.0,
      "step": 55675
    },
    {
      "entropy": 5.885257911682129,
      "epoch": 4.865856855719654,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002876509626739893,
      "loss": 5.3249,
      "mean_token_accuracy": 0.2005326732993126,
      "num_tokens": 112832523.0,
      "step": 55680
    },
    {
      "entropy": 5.929747247695923,
      "epoch": 4.866293804072359,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002876198507504032,
      "loss": 5.2919,
      "mean_token_accuracy": 0.21209944486618043,
      "num_tokens": 112842511.0,
      "step": 55685
    },
    {
      "entropy": 5.938538265228272,
      "epoch": 4.866730752425063,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028758873858476174,
      "loss": 5.2364,
      "mean_token_accuracy": 0.21694501787424086,
      "num_tokens": 112852207.0,
      "step": 55690
    },
    {
      "entropy": 5.972259664535523,
      "epoch": 4.8671677007777685,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028755762617766167,
      "loss": 5.315,
      "mean_token_accuracy": 0.19635798484086991,
      "num_tokens": 112862574.0,
      "step": 55695
    },
    {
      "entropy": 5.8592160701751705,
      "epoch": 4.867604649130473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002875265135296997,
      "loss": 5.2537,
      "mean_token_accuracy": 0.21194346398115158,
      "num_tokens": 112872356.0,
      "step": 55700
    },
    {
      "entropy": 5.852695369720459,
      "epoch": 4.868041597483177,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028749540064147266,
      "loss": 5.2512,
      "mean_token_accuracy": 0.21366229504346848,
      "num_tokens": 112882958.0,
      "step": 55705
    },
    {
      "entropy": 5.911819839477539,
      "epoch": 4.868478545835882,
      "grad_norm": 1.0,
      "learning_rate": 0.0002874642875135773,
      "loss": 5.243,
      "mean_token_accuracy": 0.2104844018816948,
      "num_tokens": 112892061.0,
      "step": 55710
    },
    {
      "entropy": 5.865098714828491,
      "epoch": 4.868915494188587,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002874331741466104,
      "loss": 5.2753,
      "mean_token_accuracy": 0.21022414714097976,
      "num_tokens": 112901940.0,
      "step": 55715
    },
    {
      "entropy": 5.941733646392822,
      "epoch": 4.869352442541292,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002874020605411686,
      "loss": 5.308,
      "mean_token_accuracy": 0.20455810129642488,
      "num_tokens": 112911451.0,
      "step": 55720
    },
    {
      "entropy": 5.8852132797241214,
      "epoch": 4.869789390893996,
      "grad_norm": 1.0,
      "learning_rate": 0.0002873709466978489,
      "loss": 5.2199,
      "mean_token_accuracy": 0.21084349006414413,
      "num_tokens": 112921183.0,
      "step": 55725
    },
    {
      "entropy": 5.890749168395996,
      "epoch": 4.870226339246701,
      "grad_norm": 1.0625,
      "learning_rate": 0.000287339832617248,
      "loss": 5.3218,
      "mean_token_accuracy": 0.20838457047939302,
      "num_tokens": 112931658.0,
      "step": 55730
    },
    {
      "entropy": 5.884297800064087,
      "epoch": 4.870663287599406,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002873087182999626,
      "loss": 5.2084,
      "mean_token_accuracy": 0.21317316144704818,
      "num_tokens": 112942398.0,
      "step": 55735
    },
    {
      "entropy": 5.962948942184449,
      "epoch": 4.8711002359521105,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028727760374658946,
      "loss": 5.4438,
      "mean_token_accuracy": 0.19194798767566681,
      "num_tokens": 112953012.0,
      "step": 55740
    },
    {
      "entropy": 5.8881042957305905,
      "epoch": 4.871537184304815,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002872464889577255,
      "loss": 5.2831,
      "mean_token_accuracy": 0.20537519305944443,
      "num_tokens": 112962740.0,
      "step": 55745
    },
    {
      "entropy": 5.939539384841919,
      "epoch": 4.87197413265752,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002872153739339674,
      "loss": 5.3258,
      "mean_token_accuracy": 0.2016407370567322,
      "num_tokens": 112973424.0,
      "step": 55750
    },
    {
      "entropy": 5.8444170475006105,
      "epoch": 4.872411081010225,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028718425867591213,
      "loss": 5.1571,
      "mean_token_accuracy": 0.20989980697631835,
      "num_tokens": 112982582.0,
      "step": 55755
    },
    {
      "entropy": 5.82037034034729,
      "epoch": 4.872848029362929,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002871531431841563,
      "loss": 5.2208,
      "mean_token_accuracy": 0.2132863759994507,
      "num_tokens": 112992131.0,
      "step": 55760
    },
    {
      "entropy": 5.840242004394531,
      "epoch": 4.873284977715634,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002871220274592969,
      "loss": 5.2558,
      "mean_token_accuracy": 0.21027023196220399,
      "num_tokens": 113002176.0,
      "step": 55765
    },
    {
      "entropy": 5.8938216209411625,
      "epoch": 4.873721926068339,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002870909115019306,
      "loss": 5.2362,
      "mean_token_accuracy": 0.21060003191232682,
      "num_tokens": 113011463.0,
      "step": 55770
    },
    {
      "entropy": 5.939411926269531,
      "epoch": 4.874158874421044,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002870597953126543,
      "loss": 5.3058,
      "mean_token_accuracy": 0.20444785803556442,
      "num_tokens": 113021012.0,
      "step": 55775
    },
    {
      "entropy": 5.938031435012817,
      "epoch": 4.874595822773748,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028702867889206484,
      "loss": 5.3283,
      "mean_token_accuracy": 0.20035931169986726,
      "num_tokens": 113030623.0,
      "step": 55780
    },
    {
      "entropy": 5.918456888198852,
      "epoch": 4.8750327711264525,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028699756224075893,
      "loss": 5.2784,
      "mean_token_accuracy": 0.20729545801877974,
      "num_tokens": 113040059.0,
      "step": 55785
    },
    {
      "entropy": 5.845688962936402,
      "epoch": 4.875469719479158,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002869664453593336,
      "loss": 5.2329,
      "mean_token_accuracy": 0.20801586359739305,
      "num_tokens": 113049102.0,
      "step": 55790
    },
    {
      "entropy": 5.838420724868774,
      "epoch": 4.875906667831862,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002869353282483855,
      "loss": 5.1932,
      "mean_token_accuracy": 0.21695750206708908,
      "num_tokens": 113059155.0,
      "step": 55795
    },
    {
      "entropy": 5.899631977081299,
      "epoch": 4.876343616184567,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028690421090851156,
      "loss": 5.2896,
      "mean_token_accuracy": 0.21505147516727446,
      "num_tokens": 113069716.0,
      "step": 55800
    },
    {
      "entropy": 5.862986469268799,
      "epoch": 4.876780564537272,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002868730933403087,
      "loss": 5.3052,
      "mean_token_accuracy": 0.20469502210617066,
      "num_tokens": 113078826.0,
      "step": 55805
    },
    {
      "entropy": 5.818486976623535,
      "epoch": 4.877217512889977,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00028684197554437355,
      "loss": 5.2314,
      "mean_token_accuracy": 0.21021707504987716,
      "num_tokens": 113088626.0,
      "step": 55810
    },
    {
      "entropy": 5.940396165847778,
      "epoch": 4.877654461242681,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002868108575213032,
      "loss": 5.3766,
      "mean_token_accuracy": 0.1978781044483185,
      "num_tokens": 113097917.0,
      "step": 55815
    },
    {
      "entropy": 5.901240015029908,
      "epoch": 4.878091409595386,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028677973927169445,
      "loss": 5.277,
      "mean_token_accuracy": 0.2084656149148941,
      "num_tokens": 113109522.0,
      "step": 55820
    },
    {
      "entropy": 5.9409716606140135,
      "epoch": 4.878528357948091,
      "grad_norm": 1.078125,
      "learning_rate": 0.000286748620796144,
      "loss": 5.3198,
      "mean_token_accuracy": 0.20528215914964676,
      "num_tokens": 113119063.0,
      "step": 55825
    },
    {
      "entropy": 5.855813026428223,
      "epoch": 4.878965306300795,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002867175020952489,
      "loss": 5.2638,
      "mean_token_accuracy": 0.2098736807703972,
      "num_tokens": 113130013.0,
      "step": 55830
    },
    {
      "entropy": 5.875828790664673,
      "epoch": 4.8794022546535,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028668638316960604,
      "loss": 5.3033,
      "mean_token_accuracy": 0.2081551432609558,
      "num_tokens": 113139900.0,
      "step": 55835
    },
    {
      "entropy": 5.884455823898316,
      "epoch": 4.879839203006204,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002866552640198121,
      "loss": 5.3507,
      "mean_token_accuracy": 0.19671900421380997,
      "num_tokens": 113149378.0,
      "step": 55840
    },
    {
      "entropy": 5.914575290679932,
      "epoch": 4.88027615135891,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002866241446464642,
      "loss": 5.3442,
      "mean_token_accuracy": 0.20133416205644608,
      "num_tokens": 113159637.0,
      "step": 55845
    },
    {
      "entropy": 5.923979043960571,
      "epoch": 4.880713099711614,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028659302505015913,
      "loss": 5.2505,
      "mean_token_accuracy": 0.20492694824934005,
      "num_tokens": 113169663.0,
      "step": 55850
    },
    {
      "entropy": 5.862129163742066,
      "epoch": 4.881150048064319,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028656190523149367,
      "loss": 5.27,
      "mean_token_accuracy": 0.21107799261808396,
      "num_tokens": 113179492.0,
      "step": 55855
    },
    {
      "entropy": 5.825142478942871,
      "epoch": 4.881586996417024,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028653078519106485,
      "loss": 5.2218,
      "mean_token_accuracy": 0.21565364599227904,
      "num_tokens": 113190626.0,
      "step": 55860
    },
    {
      "entropy": 5.9002213954925535,
      "epoch": 4.8820239447697285,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002864996649294695,
      "loss": 5.2312,
      "mean_token_accuracy": 0.2164963960647583,
      "num_tokens": 113200783.0,
      "step": 55865
    },
    {
      "entropy": 5.895705938339233,
      "epoch": 4.882460893122433,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0002864685444473046,
      "loss": 5.3225,
      "mean_token_accuracy": 0.20595137476921083,
      "num_tokens": 113211221.0,
      "step": 55870
    },
    {
      "entropy": 5.7729167461395265,
      "epoch": 4.882897841475137,
      "grad_norm": 1.015625,
      "learning_rate": 0.000286437423745167,
      "loss": 5.1486,
      "mean_token_accuracy": 0.22285314053297042,
      "num_tokens": 113221170.0,
      "step": 55875
    },
    {
      "entropy": 5.761284112930298,
      "epoch": 4.883334789827842,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002864063028236536,
      "loss": 5.2027,
      "mean_token_accuracy": 0.22666311115026475,
      "num_tokens": 113231748.0,
      "step": 55880
    },
    {
      "entropy": 5.849656963348389,
      "epoch": 4.883771738180547,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028637518168336134,
      "loss": 5.2388,
      "mean_token_accuracy": 0.22016551792621614,
      "num_tokens": 113241801.0,
      "step": 55885
    },
    {
      "entropy": 5.926892995834351,
      "epoch": 4.884208686533252,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002863440603248871,
      "loss": 5.3353,
      "mean_token_accuracy": 0.20679707378149031,
      "num_tokens": 113251181.0,
      "step": 55890
    },
    {
      "entropy": 5.872051954269409,
      "epoch": 4.884645634885956,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002863129387488279,
      "loss": 5.2199,
      "mean_token_accuracy": 0.21332307457923888,
      "num_tokens": 113260249.0,
      "step": 55895
    },
    {
      "entropy": 5.913862800598144,
      "epoch": 4.8850825832386615,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002862818169557805,
      "loss": 5.2156,
      "mean_token_accuracy": 0.2067580282688141,
      "num_tokens": 113270481.0,
      "step": 55900
    },
    {
      "entropy": 5.7230006694793705,
      "epoch": 4.885519531591366,
      "grad_norm": 1.0,
      "learning_rate": 0.00028625069494634206,
      "loss": 5.2415,
      "mean_token_accuracy": 0.21440920233726501,
      "num_tokens": 113280990.0,
      "step": 55905
    },
    {
      "entropy": 5.924447298049927,
      "epoch": 4.8859564799440705,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028621957272110935,
      "loss": 5.3289,
      "mean_token_accuracy": 0.19922223091125488,
      "num_tokens": 113291454.0,
      "step": 55910
    },
    {
      "entropy": 5.884297657012939,
      "epoch": 4.886393428296776,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002861884502806794,
      "loss": 5.2317,
      "mean_token_accuracy": 0.203391432762146,
      "num_tokens": 113300825.0,
      "step": 55915
    },
    {
      "entropy": 5.9219895362854,
      "epoch": 4.88683037664948,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028615732762564904,
      "loss": 5.2589,
      "mean_token_accuracy": 0.21630858033895492,
      "num_tokens": 113311236.0,
      "step": 55920
    },
    {
      "entropy": 5.948619794845581,
      "epoch": 4.887267325002185,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028612620475661535,
      "loss": 5.3625,
      "mean_token_accuracy": 0.20067594349384307,
      "num_tokens": 113321072.0,
      "step": 55925
    },
    {
      "entropy": 5.927085113525391,
      "epoch": 4.887704273354889,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028609508167417517,
      "loss": 5.3136,
      "mean_token_accuracy": 0.2078840270638466,
      "num_tokens": 113331330.0,
      "step": 55930
    },
    {
      "entropy": 5.869311428070068,
      "epoch": 4.888141221707594,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002860639583789255,
      "loss": 5.3817,
      "mean_token_accuracy": 0.20112004727125168,
      "num_tokens": 113342290.0,
      "step": 55935
    },
    {
      "entropy": 5.8683031558990475,
      "epoch": 4.888578170060299,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002860328348714634,
      "loss": 5.2272,
      "mean_token_accuracy": 0.2113775849342346,
      "num_tokens": 113351903.0,
      "step": 55940
    },
    {
      "entropy": 5.87104058265686,
      "epoch": 4.8890151184130035,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002860017111523857,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2139774590730667,
      "num_tokens": 113361658.0,
      "step": 55945
    },
    {
      "entropy": 5.880601835250855,
      "epoch": 4.889452066765708,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002859705872222894,
      "loss": 5.2849,
      "mean_token_accuracy": 0.20671185553073884,
      "num_tokens": 113372148.0,
      "step": 55950
    },
    {
      "entropy": 5.908346271514892,
      "epoch": 4.889889015118413,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028593946308177147,
      "loss": 5.3692,
      "mean_token_accuracy": 0.20445099771022796,
      "num_tokens": 113383775.0,
      "step": 55955
    },
    {
      "entropy": 5.903768873214721,
      "epoch": 4.890325963471118,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028590833873142895,
      "loss": 5.2498,
      "mean_token_accuracy": 0.2154083102941513,
      "num_tokens": 113393854.0,
      "step": 55960
    },
    {
      "entropy": 5.926236772537232,
      "epoch": 4.890762911823822,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028587721417185877,
      "loss": 5.2655,
      "mean_token_accuracy": 0.20804270952939988,
      "num_tokens": 113404562.0,
      "step": 55965
    },
    {
      "entropy": 5.875795459747314,
      "epoch": 4.891199860176527,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002858460894036579,
      "loss": 5.2767,
      "mean_token_accuracy": 0.20589140057563782,
      "num_tokens": 113414082.0,
      "step": 55970
    },
    {
      "entropy": 5.886067581176758,
      "epoch": 4.891636808529232,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002858149644274234,
      "loss": 5.2847,
      "mean_token_accuracy": 0.19927649945020676,
      "num_tokens": 113424990.0,
      "step": 55975
    },
    {
      "entropy": 5.940613079071045,
      "epoch": 4.892073756881937,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002857838392437522,
      "loss": 5.3383,
      "mean_token_accuracy": 0.20703790485858917,
      "num_tokens": 113435850.0,
      "step": 55980
    },
    {
      "entropy": 5.925675344467163,
      "epoch": 4.892510705234641,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028575271385324123,
      "loss": 5.3004,
      "mean_token_accuracy": 0.20805398523807525,
      "num_tokens": 113445987.0,
      "step": 55985
    },
    {
      "entropy": 5.931181955337524,
      "epoch": 4.8929476535873455,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002857215882564877,
      "loss": 5.3567,
      "mean_token_accuracy": 0.19895374923944473,
      "num_tokens": 113454879.0,
      "step": 55990
    },
    {
      "entropy": 5.856374454498291,
      "epoch": 4.893384601940051,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002856904624540884,
      "loss": 5.2892,
      "mean_token_accuracy": 0.20379792302846908,
      "num_tokens": 113464840.0,
      "step": 55995
    },
    {
      "entropy": 5.893730497360229,
      "epoch": 4.893821550292755,
      "grad_norm": 1.0,
      "learning_rate": 0.00028565933644664036,
      "loss": 5.3436,
      "mean_token_accuracy": 0.20493225008249283,
      "num_tokens": 113474279.0,
      "step": 56000
    },
    {
      "entropy": 5.876547145843506,
      "epoch": 4.89425849864546,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002856282102347408,
      "loss": 5.2445,
      "mean_token_accuracy": 0.21470918506383896,
      "num_tokens": 113484238.0,
      "step": 56005
    },
    {
      "entropy": 5.919389533996582,
      "epoch": 4.894695446998165,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028559708381898654,
      "loss": 5.259,
      "mean_token_accuracy": 0.20821300148963928,
      "num_tokens": 113494961.0,
      "step": 56010
    },
    {
      "entropy": 5.91197829246521,
      "epoch": 4.89513239535087,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002855659571999747,
      "loss": 5.3875,
      "mean_token_accuracy": 0.209483040869236,
      "num_tokens": 113505893.0,
      "step": 56015
    },
    {
      "entropy": 5.875579881668091,
      "epoch": 4.895569343703574,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028553483037830225,
      "loss": 5.2557,
      "mean_token_accuracy": 0.20887023508548735,
      "num_tokens": 113515919.0,
      "step": 56020
    },
    {
      "entropy": 5.825322484970092,
      "epoch": 4.896006292056279,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002855037033545662,
      "loss": 5.1761,
      "mean_token_accuracy": 0.21960469484329223,
      "num_tokens": 113525748.0,
      "step": 56025
    },
    {
      "entropy": 5.91252555847168,
      "epoch": 4.896443240408984,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002854725761293636,
      "loss": 5.3444,
      "mean_token_accuracy": 0.20211513340473175,
      "num_tokens": 113536803.0,
      "step": 56030
    },
    {
      "entropy": 5.857369232177734,
      "epoch": 4.896880188761688,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028544144870329153,
      "loss": 5.2024,
      "mean_token_accuracy": 0.2154713898897171,
      "num_tokens": 113547099.0,
      "step": 56035
    },
    {
      "entropy": 5.87781662940979,
      "epoch": 4.897317137114393,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002854103210769471,
      "loss": 5.3101,
      "mean_token_accuracy": 0.20592548996210097,
      "num_tokens": 113557334.0,
      "step": 56040
    },
    {
      "entropy": 5.865287637710571,
      "epoch": 4.897754085467097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028537919325092714,
      "loss": 5.2913,
      "mean_token_accuracy": 0.2068054646253586,
      "num_tokens": 113566112.0,
      "step": 56045
    },
    {
      "entropy": 5.894245862960815,
      "epoch": 4.898191033819803,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028534806522582885,
      "loss": 5.3236,
      "mean_token_accuracy": 0.20500457286834717,
      "num_tokens": 113576069.0,
      "step": 56050
    },
    {
      "entropy": 5.894721460342407,
      "epoch": 4.898627982172507,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028531693700224926,
      "loss": 5.2956,
      "mean_token_accuracy": 0.20858899801969527,
      "num_tokens": 113585935.0,
      "step": 56055
    },
    {
      "entropy": 5.873931837081909,
      "epoch": 4.899064930525212,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028528580858078537,
      "loss": 5.2275,
      "mean_token_accuracy": 0.21442634612321854,
      "num_tokens": 113597413.0,
      "step": 56060
    },
    {
      "entropy": 5.8839937210083,
      "epoch": 4.899501878877917,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028525467996203425,
      "loss": 5.2767,
      "mean_token_accuracy": 0.21001888066530228,
      "num_tokens": 113607446.0,
      "step": 56065
    },
    {
      "entropy": 5.858818197250367,
      "epoch": 4.8999388272306215,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000285223551146593,
      "loss": 5.2055,
      "mean_token_accuracy": 0.21761797666549682,
      "num_tokens": 113616847.0,
      "step": 56070
    },
    {
      "entropy": 5.909807538986206,
      "epoch": 4.900375775583326,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002851924221350588,
      "loss": 5.3228,
      "mean_token_accuracy": 0.20104700624942778,
      "num_tokens": 113626809.0,
      "step": 56075
    },
    {
      "entropy": 5.880495405197143,
      "epoch": 4.90081272393603,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002851612929280285,
      "loss": 5.2038,
      "mean_token_accuracy": 0.21392233967781066,
      "num_tokens": 113636695.0,
      "step": 56080
    },
    {
      "entropy": 5.897242450714112,
      "epoch": 4.901249672288736,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002851301635260993,
      "loss": 5.2738,
      "mean_token_accuracy": 0.2149894893169403,
      "num_tokens": 113646522.0,
      "step": 56085
    },
    {
      "entropy": 5.8108766078948975,
      "epoch": 4.90168662064144,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002850990339298683,
      "loss": 5.1836,
      "mean_token_accuracy": 0.21947732716798782,
      "num_tokens": 113656844.0,
      "step": 56090
    },
    {
      "entropy": 5.858398723602295,
      "epoch": 4.902123568994145,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002850679041399325,
      "loss": 5.2662,
      "mean_token_accuracy": 0.21138952821493148,
      "num_tokens": 113667466.0,
      "step": 56095
    },
    {
      "entropy": 5.874529647827148,
      "epoch": 4.902560517346849,
      "grad_norm": 1.046875,
      "learning_rate": 0.000285036774156889,
      "loss": 5.2862,
      "mean_token_accuracy": 0.20686406791210174,
      "num_tokens": 113677614.0,
      "step": 56100
    },
    {
      "entropy": 5.934945583343506,
      "epoch": 4.9029974656995545,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028500564398133496,
      "loss": 5.2973,
      "mean_token_accuracy": 0.20919312983751298,
      "num_tokens": 113687289.0,
      "step": 56105
    },
    {
      "entropy": 5.889858865737915,
      "epoch": 4.903434414052259,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028497451361386743,
      "loss": 5.3129,
      "mean_token_accuracy": 0.20683043897151948,
      "num_tokens": 113696695.0,
      "step": 56110
    },
    {
      "entropy": 5.842043399810791,
      "epoch": 4.9038713624049635,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028494338305508343,
      "loss": 5.2398,
      "mean_token_accuracy": 0.20924404710531236,
      "num_tokens": 113706952.0,
      "step": 56115
    },
    {
      "entropy": 5.8511310577392575,
      "epoch": 4.904308310757669,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028491225230558014,
      "loss": 5.3251,
      "mean_token_accuracy": 0.21151094287633895,
      "num_tokens": 113717774.0,
      "step": 56120
    },
    {
      "entropy": 5.882019758224487,
      "epoch": 4.904745259110373,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028488112136595473,
      "loss": 5.2403,
      "mean_token_accuracy": 0.21099464297294618,
      "num_tokens": 113727479.0,
      "step": 56125
    },
    {
      "entropy": 5.8651604652404785,
      "epoch": 4.905182207463078,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00028484999023680417,
      "loss": 5.2222,
      "mean_token_accuracy": 0.2150694027543068,
      "num_tokens": 113738212.0,
      "step": 56130
    },
    {
      "entropy": 5.921840381622315,
      "epoch": 4.905619155815782,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028481885891872574,
      "loss": 5.3443,
      "mean_token_accuracy": 0.20687018930912018,
      "num_tokens": 113748054.0,
      "step": 56135
    },
    {
      "entropy": 5.868416881561279,
      "epoch": 4.906056104168488,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028478772741231636,
      "loss": 5.2235,
      "mean_token_accuracy": 0.21263317316770552,
      "num_tokens": 113757756.0,
      "step": 56140
    },
    {
      "entropy": 5.835989761352539,
      "epoch": 4.906493052521192,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028475659571817325,
      "loss": 5.2246,
      "mean_token_accuracy": 0.21575887203216554,
      "num_tokens": 113767812.0,
      "step": 56145
    },
    {
      "entropy": 5.824907493591309,
      "epoch": 4.9069300008738965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028472546383689353,
      "loss": 5.2639,
      "mean_token_accuracy": 0.2036179259419441,
      "num_tokens": 113779473.0,
      "step": 56150
    },
    {
      "entropy": 5.8406189441680905,
      "epoch": 4.907366949226601,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028469433176907434,
      "loss": 5.2357,
      "mean_token_accuracy": 0.21148528009653092,
      "num_tokens": 113788545.0,
      "step": 56155
    },
    {
      "entropy": 5.833606910705567,
      "epoch": 4.907803897579306,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002846631995153128,
      "loss": 5.2674,
      "mean_token_accuracy": 0.20812720209360122,
      "num_tokens": 113798713.0,
      "step": 56160
    },
    {
      "entropy": 5.921920442581177,
      "epoch": 4.908240845932011,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028463206707620597,
      "loss": 5.3086,
      "mean_token_accuracy": 0.20812164545059203,
      "num_tokens": 113809387.0,
      "step": 56165
    },
    {
      "entropy": 5.940256786346436,
      "epoch": 4.908677794284715,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002846009344523511,
      "loss": 5.2899,
      "mean_token_accuracy": 0.20786315351724624,
      "num_tokens": 113819867.0,
      "step": 56170
    },
    {
      "entropy": 5.881246852874756,
      "epoch": 4.909114742637421,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002845698016443452,
      "loss": 5.2868,
      "mean_token_accuracy": 0.20551809221506118,
      "num_tokens": 113830723.0,
      "step": 56175
    },
    {
      "entropy": 5.87975697517395,
      "epoch": 4.909551690990125,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028453866865278557,
      "loss": 5.253,
      "mean_token_accuracy": 0.2094110831618309,
      "num_tokens": 113839905.0,
      "step": 56180
    },
    {
      "entropy": 5.844990205764771,
      "epoch": 4.90998863934283,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028450753547826923,
      "loss": 5.2391,
      "mean_token_accuracy": 0.2111005961894989,
      "num_tokens": 113848851.0,
      "step": 56185
    },
    {
      "entropy": 5.843260860443115,
      "epoch": 4.910425587695534,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028447640212139334,
      "loss": 5.3047,
      "mean_token_accuracy": 0.21033355146646499,
      "num_tokens": 113859157.0,
      "step": 56190
    },
    {
      "entropy": 5.836468744277954,
      "epoch": 4.910862536048239,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028444526858275514,
      "loss": 5.2108,
      "mean_token_accuracy": 0.21477921605110167,
      "num_tokens": 113869083.0,
      "step": 56195
    },
    {
      "entropy": 5.927158164978027,
      "epoch": 4.911299484400944,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00028441413486295173,
      "loss": 5.3527,
      "mean_token_accuracy": 0.20822176337242126,
      "num_tokens": 113879665.0,
      "step": 56200
    },
    {
      "entropy": 5.932024574279785,
      "epoch": 4.911736432753648,
      "grad_norm": 1.125,
      "learning_rate": 0.00028438300096258036,
      "loss": 5.3406,
      "mean_token_accuracy": 0.20177732706069945,
      "num_tokens": 113889271.0,
      "step": 56205
    },
    {
      "entropy": 5.92923789024353,
      "epoch": 4.912173381106353,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028435186688223805,
      "loss": 5.325,
      "mean_token_accuracy": 0.20543894171714783,
      "num_tokens": 113900504.0,
      "step": 56210
    },
    {
      "entropy": 5.888232898712158,
      "epoch": 4.912610329459058,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000284320732622522,
      "loss": 5.325,
      "mean_token_accuracy": 0.20206792056560516,
      "num_tokens": 113910892.0,
      "step": 56215
    },
    {
      "entropy": 5.908745765686035,
      "epoch": 4.913047277811763,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00028428959818402945,
      "loss": 5.3221,
      "mean_token_accuracy": 0.20897235423326493,
      "num_tokens": 113921722.0,
      "step": 56220
    },
    {
      "entropy": 5.916287612915039,
      "epoch": 4.913484226164467,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028425846356735753,
      "loss": 5.2636,
      "mean_token_accuracy": 0.2047579050064087,
      "num_tokens": 113933183.0,
      "step": 56225
    },
    {
      "entropy": 5.861678409576416,
      "epoch": 4.9139211745171725,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002842273287731035,
      "loss": 5.2581,
      "mean_token_accuracy": 0.22279146313667297,
      "num_tokens": 113943608.0,
      "step": 56230
    },
    {
      "entropy": 5.8389284133911135,
      "epoch": 4.914358122869877,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028419619380186433,
      "loss": 5.2446,
      "mean_token_accuracy": 0.20983339995145797,
      "num_tokens": 113953065.0,
      "step": 56235
    },
    {
      "entropy": 5.647596836090088,
      "epoch": 4.914795071222581,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002841650586542374,
      "loss": 5.0454,
      "mean_token_accuracy": 0.23703864514827727,
      "num_tokens": 113964188.0,
      "step": 56240
    },
    {
      "entropy": 5.819708013534546,
      "epoch": 4.915232019575286,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002841339233308198,
      "loss": 5.2586,
      "mean_token_accuracy": 0.21105241030454636,
      "num_tokens": 113973857.0,
      "step": 56245
    },
    {
      "entropy": 5.960319137573242,
      "epoch": 4.915668967927991,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028410278783220887,
      "loss": 5.3162,
      "mean_token_accuracy": 0.20048186480998992,
      "num_tokens": 113984158.0,
      "step": 56250
    },
    {
      "entropy": 5.921494007110596,
      "epoch": 4.916105916280696,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028407165215900166,
      "loss": 5.2832,
      "mean_token_accuracy": 0.21042756885290145,
      "num_tokens": 113994875.0,
      "step": 56255
    },
    {
      "entropy": 5.898594236373901,
      "epoch": 4.9165428646334,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002840405163117954,
      "loss": 5.3254,
      "mean_token_accuracy": 0.2079416811466217,
      "num_tokens": 114004947.0,
      "step": 56260
    },
    {
      "entropy": 5.902395343780517,
      "epoch": 4.916979812986105,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028400938029118726,
      "loss": 5.302,
      "mean_token_accuracy": 0.2105281800031662,
      "num_tokens": 114015592.0,
      "step": 56265
    },
    {
      "entropy": 5.840075159072876,
      "epoch": 4.91741676133881,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002839782440977745,
      "loss": 5.1995,
      "mean_token_accuracy": 0.21179545372724534,
      "num_tokens": 114025538.0,
      "step": 56270
    },
    {
      "entropy": 5.867462253570556,
      "epoch": 4.9178537096915145,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002839471077321544,
      "loss": 5.2412,
      "mean_token_accuracy": 0.21128964573144912,
      "num_tokens": 114035196.0,
      "step": 56275
    },
    {
      "entropy": 5.832903623580933,
      "epoch": 4.918290658044219,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000283915971194924,
      "loss": 5.2362,
      "mean_token_accuracy": 0.2062760353088379,
      "num_tokens": 114045896.0,
      "step": 56280
    },
    {
      "entropy": 5.852113819122314,
      "epoch": 4.918727606396924,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028388483448668064,
      "loss": 5.3053,
      "mean_token_accuracy": 0.2069786384701729,
      "num_tokens": 114055854.0,
      "step": 56285
    },
    {
      "entropy": 5.894686841964722,
      "epoch": 4.919164554749629,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002838536976080214,
      "loss": 5.2691,
      "mean_token_accuracy": 0.20992080122232437,
      "num_tokens": 114066349.0,
      "step": 56290
    },
    {
      "entropy": 5.878214740753174,
      "epoch": 4.919601503102333,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002838225605595438,
      "loss": 5.3221,
      "mean_token_accuracy": 0.2039284735918045,
      "num_tokens": 114077374.0,
      "step": 56295
    },
    {
      "entropy": 5.840265130996704,
      "epoch": 4.920038451455038,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002837914233418447,
      "loss": 5.219,
      "mean_token_accuracy": 0.21462631076574326,
      "num_tokens": 114088276.0,
      "step": 56300
    },
    {
      "entropy": 5.838958263397217,
      "epoch": 4.920475399807743,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028376028595552156,
      "loss": 5.314,
      "mean_token_accuracy": 0.20951756685972214,
      "num_tokens": 114097983.0,
      "step": 56305
    },
    {
      "entropy": 5.824288845062256,
      "epoch": 4.9209123481604475,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002837291484011715,
      "loss": 5.2151,
      "mean_token_accuracy": 0.21711032837629318,
      "num_tokens": 114107958.0,
      "step": 56310
    },
    {
      "entropy": 5.898913812637329,
      "epoch": 4.921349296513152,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002836980106793918,
      "loss": 5.2545,
      "mean_token_accuracy": 0.20683859139680863,
      "num_tokens": 114118426.0,
      "step": 56315
    },
    {
      "entropy": 5.85609393119812,
      "epoch": 4.9217862448658565,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028366687279077977,
      "loss": 5.2134,
      "mean_token_accuracy": 0.22169376462697982,
      "num_tokens": 114128014.0,
      "step": 56320
    },
    {
      "entropy": 5.878192281723022,
      "epoch": 4.922223193218562,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028363573473593254,
      "loss": 5.2447,
      "mean_token_accuracy": 0.21167627424001695,
      "num_tokens": 114137827.0,
      "step": 56325
    },
    {
      "entropy": 5.867742443084717,
      "epoch": 4.922660141571266,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028360459651544735,
      "loss": 5.3371,
      "mean_token_accuracy": 0.1991913542151451,
      "num_tokens": 114148171.0,
      "step": 56330
    },
    {
      "entropy": 5.903535079956055,
      "epoch": 4.923097089923971,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002835734581299216,
      "loss": 5.2666,
      "mean_token_accuracy": 0.20804971158504487,
      "num_tokens": 114157923.0,
      "step": 56335
    },
    {
      "entropy": 5.902818489074707,
      "epoch": 4.923534038276676,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002835423195799523,
      "loss": 5.207,
      "mean_token_accuracy": 0.21390482634305955,
      "num_tokens": 114167471.0,
      "step": 56340
    },
    {
      "entropy": 5.945140218734741,
      "epoch": 4.923970986629381,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002835111808661369,
      "loss": 5.3387,
      "mean_token_accuracy": 0.2102211207151413,
      "num_tokens": 114177337.0,
      "step": 56345
    },
    {
      "entropy": 5.8214599609375,
      "epoch": 4.924407934982085,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028348004198907255,
      "loss": 5.2382,
      "mean_token_accuracy": 0.2123619168996811,
      "num_tokens": 114187019.0,
      "step": 56350
    },
    {
      "entropy": 5.773484373092652,
      "epoch": 4.9248448833347895,
      "grad_norm": 1.125,
      "learning_rate": 0.00028344890294935667,
      "loss": 5.2102,
      "mean_token_accuracy": 0.21351995915174485,
      "num_tokens": 114196553.0,
      "step": 56355
    },
    {
      "entropy": 5.933808469772339,
      "epoch": 4.925281831687495,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002834177637475863,
      "loss": 5.3156,
      "mean_token_accuracy": 0.20305173844099045,
      "num_tokens": 114207096.0,
      "step": 56360
    },
    {
      "entropy": 5.876393270492554,
      "epoch": 4.925718780040199,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002833866243843587,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21305223405361176,
      "num_tokens": 114218294.0,
      "step": 56365
    },
    {
      "entropy": 5.9192750453948975,
      "epoch": 4.926155728392904,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028335548486027145,
      "loss": 5.2417,
      "mean_token_accuracy": 0.20813332498073578,
      "num_tokens": 114228520.0,
      "step": 56370
    },
    {
      "entropy": 5.8464244365692135,
      "epoch": 4.926592676745608,
      "grad_norm": 1.0,
      "learning_rate": 0.0002833243451759216,
      "loss": 5.2719,
      "mean_token_accuracy": 0.21247351616621019,
      "num_tokens": 114238491.0,
      "step": 56375
    },
    {
      "entropy": 5.919299364089966,
      "epoch": 4.927029625098314,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028329320533190637,
      "loss": 5.2811,
      "mean_token_accuracy": 0.20889726132154465,
      "num_tokens": 114248040.0,
      "step": 56380
    },
    {
      "entropy": 5.8519124507904055,
      "epoch": 4.927466573451018,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028326206532882313,
      "loss": 5.2431,
      "mean_token_accuracy": 0.21499415040016173,
      "num_tokens": 114258834.0,
      "step": 56385
    },
    {
      "entropy": 5.866873645782471,
      "epoch": 4.927903521803723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028323092516726917,
      "loss": 5.2679,
      "mean_token_accuracy": 0.2128313109278679,
      "num_tokens": 114268822.0,
      "step": 56390
    },
    {
      "entropy": 5.847755241394043,
      "epoch": 4.928340470156428,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002831997848478418,
      "loss": 5.2766,
      "mean_token_accuracy": 0.21025744676589966,
      "num_tokens": 114278844.0,
      "step": 56395
    },
    {
      "entropy": 5.830688428878784,
      "epoch": 4.928777418509132,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028316864437113814,
      "loss": 5.156,
      "mean_token_accuracy": 0.22103468626737593,
      "num_tokens": 114287361.0,
      "step": 56400
    },
    {
      "entropy": 5.8817144393920895,
      "epoch": 4.929214366861837,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028313750373775565,
      "loss": 5.2775,
      "mean_token_accuracy": 0.21438289433717728,
      "num_tokens": 114297971.0,
      "step": 56405
    },
    {
      "entropy": 5.9138130187988285,
      "epoch": 4.929651315214541,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028310636294829163,
      "loss": 5.2725,
      "mean_token_accuracy": 0.20475396066904067,
      "num_tokens": 114308477.0,
      "step": 56410
    },
    {
      "entropy": 5.844355630874634,
      "epoch": 4.930088263567247,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002830752220033433,
      "loss": 5.2304,
      "mean_token_accuracy": 0.21291066855192184,
      "num_tokens": 114317453.0,
      "step": 56415
    },
    {
      "entropy": 5.898775100708008,
      "epoch": 4.930525211919951,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028304408090350794,
      "loss": 5.3098,
      "mean_token_accuracy": 0.199411016702652,
      "num_tokens": 114326320.0,
      "step": 56420
    },
    {
      "entropy": 5.85786280632019,
      "epoch": 4.930962160272656,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028301293964938285,
      "loss": 5.2621,
      "mean_token_accuracy": 0.2098367154598236,
      "num_tokens": 114336552.0,
      "step": 56425
    },
    {
      "entropy": 5.914081668853759,
      "epoch": 4.93139910862536,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002829817982415654,
      "loss": 5.2619,
      "mean_token_accuracy": 0.20721474289894104,
      "num_tokens": 114346711.0,
      "step": 56430
    },
    {
      "entropy": 5.940443754196167,
      "epoch": 4.9318360569780655,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000282950656680653,
      "loss": 5.3301,
      "mean_token_accuracy": 0.21051183342933655,
      "num_tokens": 114356000.0,
      "step": 56435
    },
    {
      "entropy": 5.863522720336914,
      "epoch": 4.93227300533077,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028291951496724275,
      "loss": 5.274,
      "mean_token_accuracy": 0.2189063236117363,
      "num_tokens": 114366727.0,
      "step": 56440
    },
    {
      "entropy": 5.9327093124389645,
      "epoch": 4.932709953683474,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028288837310193213,
      "loss": 5.2592,
      "mean_token_accuracy": 0.2077247679233551,
      "num_tokens": 114376880.0,
      "step": 56445
    },
    {
      "entropy": 5.945688629150391,
      "epoch": 4.93314690203618,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028285723108531827,
      "loss": 5.3131,
      "mean_token_accuracy": 0.19741540104150773,
      "num_tokens": 114386823.0,
      "step": 56450
    },
    {
      "entropy": 5.869300413131714,
      "epoch": 4.933583850388884,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002828260889179986,
      "loss": 5.3015,
      "mean_token_accuracy": 0.20652665942907333,
      "num_tokens": 114396335.0,
      "step": 56455
    },
    {
      "entropy": 5.873976707458496,
      "epoch": 4.934020798741589,
      "grad_norm": 1.125,
      "learning_rate": 0.00028279494660057056,
      "loss": 5.2574,
      "mean_token_accuracy": 0.2076423868536949,
      "num_tokens": 114407196.0,
      "step": 56460
    },
    {
      "entropy": 5.8970733165740965,
      "epoch": 4.934457747094293,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002827638041336312,
      "loss": 5.2817,
      "mean_token_accuracy": 0.20223770588636397,
      "num_tokens": 114417226.0,
      "step": 56465
    },
    {
      "entropy": 5.896813774108887,
      "epoch": 4.9348946954469985,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002827326615177781,
      "loss": 5.2568,
      "mean_token_accuracy": 0.2141839399933815,
      "num_tokens": 114427197.0,
      "step": 56470
    },
    {
      "entropy": 5.883169555664063,
      "epoch": 4.935331643799703,
      "grad_norm": 1.125,
      "learning_rate": 0.0002827015187536085,
      "loss": 5.294,
      "mean_token_accuracy": 0.20705011934041978,
      "num_tokens": 114438509.0,
      "step": 56475
    },
    {
      "entropy": 5.909549331665039,
      "epoch": 4.9357685921524075,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028267037584171973,
      "loss": 5.3002,
      "mean_token_accuracy": 0.20891426205635072,
      "num_tokens": 114448660.0,
      "step": 56480
    },
    {
      "entropy": 5.893820905685425,
      "epoch": 4.936205540505112,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002826392327827091,
      "loss": 5.2531,
      "mean_token_accuracy": 0.21201108247041703,
      "num_tokens": 114458869.0,
      "step": 56485
    },
    {
      "entropy": 5.8461174964904785,
      "epoch": 4.936642488857817,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002826080895771741,
      "loss": 5.2098,
      "mean_token_accuracy": 0.217110738158226,
      "num_tokens": 114468572.0,
      "step": 56490
    },
    {
      "entropy": 5.879475927352905,
      "epoch": 4.937079437210522,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002825769462257119,
      "loss": 5.3277,
      "mean_token_accuracy": 0.2036124810576439,
      "num_tokens": 114478692.0,
      "step": 56495
    },
    {
      "entropy": 5.968720245361328,
      "epoch": 4.937516385563226,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00028254580272891983,
      "loss": 5.3745,
      "mean_token_accuracy": 0.1969718113541603,
      "num_tokens": 114490578.0,
      "step": 56500
    },
    {
      "entropy": 5.990403509140014,
      "epoch": 4.937953333915932,
      "grad_norm": 1.0,
      "learning_rate": 0.0002825146590873953,
      "loss": 5.2956,
      "mean_token_accuracy": 0.20566201508045195,
      "num_tokens": 114500979.0,
      "step": 56505
    },
    {
      "entropy": 5.904782438278199,
      "epoch": 4.938390282268636,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002824835153017358,
      "loss": 5.2897,
      "mean_token_accuracy": 0.21204884648323058,
      "num_tokens": 114510437.0,
      "step": 56510
    },
    {
      "entropy": 5.820358562469482,
      "epoch": 4.9388272306213405,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028245237137253845,
      "loss": 5.2636,
      "mean_token_accuracy": 0.21246273815631866,
      "num_tokens": 114521458.0,
      "step": 56515
    },
    {
      "entropy": 5.780618715286255,
      "epoch": 4.939264178974045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002824212273004007,
      "loss": 5.1214,
      "mean_token_accuracy": 0.22280751913785934,
      "num_tokens": 114531134.0,
      "step": 56520
    },
    {
      "entropy": 5.849256944656372,
      "epoch": 4.9397011273267495,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028239008308591996,
      "loss": 5.2442,
      "mean_token_accuracy": 0.21322206556797027,
      "num_tokens": 114541671.0,
      "step": 56525
    },
    {
      "entropy": 5.841625118255616,
      "epoch": 4.940138075679455,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002823589387296936,
      "loss": 5.2433,
      "mean_token_accuracy": 0.20764423459768294,
      "num_tokens": 114551454.0,
      "step": 56530
    },
    {
      "entropy": 5.88473105430603,
      "epoch": 4.940575024032159,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028232779423231893,
      "loss": 5.2548,
      "mean_token_accuracy": 0.21112620532512666,
      "num_tokens": 114561168.0,
      "step": 56535
    },
    {
      "entropy": 5.83260703086853,
      "epoch": 4.941011972384864,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002822966495943932,
      "loss": 5.2222,
      "mean_token_accuracy": 0.2160140246152878,
      "num_tokens": 114570164.0,
      "step": 56540
    },
    {
      "entropy": 5.748507738113403,
      "epoch": 4.941448920737569,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000282265504816514,
      "loss": 5.1259,
      "mean_token_accuracy": 0.22420865893363953,
      "num_tokens": 114582608.0,
      "step": 56545
    },
    {
      "entropy": 5.910902738571167,
      "epoch": 4.941885869090274,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028223435989927864,
      "loss": 5.3129,
      "mean_token_accuracy": 0.20196628719568252,
      "num_tokens": 114594768.0,
      "step": 56550
    },
    {
      "entropy": 5.975846195220948,
      "epoch": 4.942322817442978,
      "grad_norm": 1.125,
      "learning_rate": 0.0002822032148432844,
      "loss": 5.2718,
      "mean_token_accuracy": 0.20387360751628875,
      "num_tokens": 114605023.0,
      "step": 56555
    },
    {
      "entropy": 5.917460203170776,
      "epoch": 4.942759765795683,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028217206964912877,
      "loss": 5.2737,
      "mean_token_accuracy": 0.20920604318380356,
      "num_tokens": 114615794.0,
      "step": 56560
    },
    {
      "entropy": 5.832402181625366,
      "epoch": 4.943196714148388,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028214092431740913,
      "loss": 5.2377,
      "mean_token_accuracy": 0.2102249950170517,
      "num_tokens": 114625389.0,
      "step": 56565
    },
    {
      "entropy": 5.8560810565948485,
      "epoch": 4.943633662501092,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002821097788487227,
      "loss": 5.2991,
      "mean_token_accuracy": 0.20456133484840394,
      "num_tokens": 114635594.0,
      "step": 56570
    },
    {
      "entropy": 5.909289121627808,
      "epoch": 4.944070610853797,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028207863324366704,
      "loss": 5.2369,
      "mean_token_accuracy": 0.2150420993566513,
      "num_tokens": 114645730.0,
      "step": 56575
    },
    {
      "entropy": 5.91946759223938,
      "epoch": 4.944507559206501,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028204748750283953,
      "loss": 5.2289,
      "mean_token_accuracy": 0.2057092934846878,
      "num_tokens": 114655048.0,
      "step": 56580
    },
    {
      "entropy": 5.847956323623658,
      "epoch": 4.944944507559207,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028201634162683756,
      "loss": 5.2952,
      "mean_token_accuracy": 0.20775219202041625,
      "num_tokens": 114665965.0,
      "step": 56585
    },
    {
      "entropy": 5.802372455596924,
      "epoch": 4.945381455911911,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028198519561625833,
      "loss": 5.1785,
      "mean_token_accuracy": 0.21679890155792236,
      "num_tokens": 114676395.0,
      "step": 56590
    },
    {
      "entropy": 5.935219478607178,
      "epoch": 4.945818404264616,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002819540494716995,
      "loss": 5.2823,
      "mean_token_accuracy": 0.19855900704860688,
      "num_tokens": 114687302.0,
      "step": 56595
    },
    {
      "entropy": 5.839901256561279,
      "epoch": 4.946255352617321,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028192290319375827,
      "loss": 5.1977,
      "mean_token_accuracy": 0.2190491661429405,
      "num_tokens": 114697252.0,
      "step": 56600
    },
    {
      "entropy": 5.838676595687867,
      "epoch": 4.946692300970025,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028189175678303217,
      "loss": 5.2339,
      "mean_token_accuracy": 0.21693686544895172,
      "num_tokens": 114706997.0,
      "step": 56605
    },
    {
      "entropy": 5.836890411376953,
      "epoch": 4.94712924932273,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028186061024011857,
      "loss": 5.2689,
      "mean_token_accuracy": 0.20665887445211412,
      "num_tokens": 114716452.0,
      "step": 56610
    },
    {
      "entropy": 5.877643013000489,
      "epoch": 4.947566197675435,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002818294635656149,
      "loss": 5.2946,
      "mean_token_accuracy": 0.2078608378767967,
      "num_tokens": 114726763.0,
      "step": 56615
    },
    {
      "entropy": 5.975151491165161,
      "epoch": 4.94800314602814,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028179831676011845,
      "loss": 5.3358,
      "mean_token_accuracy": 0.2074110820889473,
      "num_tokens": 114736903.0,
      "step": 56620
    },
    {
      "entropy": 5.886218547821045,
      "epoch": 4.948440094380844,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002817671698242268,
      "loss": 5.3181,
      "mean_token_accuracy": 0.20494204461574556,
      "num_tokens": 114746125.0,
      "step": 56625
    },
    {
      "entropy": 5.838334083557129,
      "epoch": 4.948877042733549,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028173602275853734,
      "loss": 5.2352,
      "mean_token_accuracy": 0.21462323814630507,
      "num_tokens": 114755576.0,
      "step": 56630
    },
    {
      "entropy": 5.845485401153565,
      "epoch": 4.949313991086253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002817048755636474,
      "loss": 5.2273,
      "mean_token_accuracy": 0.21430494487285615,
      "num_tokens": 114765982.0,
      "step": 56635
    },
    {
      "entropy": 5.867333507537841,
      "epoch": 4.9497509394389585,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028167372824015435,
      "loss": 5.2615,
      "mean_token_accuracy": 0.20701224058866502,
      "num_tokens": 114776434.0,
      "step": 56640
    },
    {
      "entropy": 5.891823387145996,
      "epoch": 4.950187887791663,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028164258078865575,
      "loss": 5.2928,
      "mean_token_accuracy": 0.20598137825727464,
      "num_tokens": 114786613.0,
      "step": 56645
    },
    {
      "entropy": 5.86541781425476,
      "epoch": 4.950624836144367,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028161143320974897,
      "loss": 5.2455,
      "mean_token_accuracy": 0.21150336861610414,
      "num_tokens": 114796276.0,
      "step": 56650
    },
    {
      "entropy": 5.910002946853638,
      "epoch": 4.951061784497073,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002815802855040314,
      "loss": 5.2517,
      "mean_token_accuracy": 0.21561769992113114,
      "num_tokens": 114807535.0,
      "step": 56655
    },
    {
      "entropy": 5.8624529361724855,
      "epoch": 4.951498732849777,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002815491376721006,
      "loss": 5.1696,
      "mean_token_accuracy": 0.21441082060337066,
      "num_tokens": 114817004.0,
      "step": 56660
    },
    {
      "entropy": 5.922556161880493,
      "epoch": 4.951935681202482,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028151798971455376,
      "loss": 5.3807,
      "mean_token_accuracy": 0.19875044077634813,
      "num_tokens": 114826980.0,
      "step": 56665
    },
    {
      "entropy": 5.824126958847046,
      "epoch": 4.952372629555186,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028148684163198856,
      "loss": 5.2462,
      "mean_token_accuracy": 0.21116557568311692,
      "num_tokens": 114837406.0,
      "step": 56670
    },
    {
      "entropy": 5.892506313323975,
      "epoch": 4.952809577907892,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002814556934250024,
      "loss": 5.2371,
      "mean_token_accuracy": 0.21795143336057662,
      "num_tokens": 114847405.0,
      "step": 56675
    },
    {
      "entropy": 5.900516986846924,
      "epoch": 4.953246526260596,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028142454509419256,
      "loss": 5.2746,
      "mean_token_accuracy": 0.21082980036735535,
      "num_tokens": 114857317.0,
      "step": 56680
    },
    {
      "entropy": 5.891704559326172,
      "epoch": 4.9536834746133005,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002813933966401566,
      "loss": 5.3036,
      "mean_token_accuracy": 0.201605860888958,
      "num_tokens": 114866579.0,
      "step": 56685
    },
    {
      "entropy": 5.850552320480347,
      "epoch": 4.954120422966005,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002813622480634919,
      "loss": 5.3103,
      "mean_token_accuracy": 0.2026976838707924,
      "num_tokens": 114875638.0,
      "step": 56690
    },
    {
      "entropy": 5.87047758102417,
      "epoch": 4.95455737131871,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000281331099364796,
      "loss": 5.263,
      "mean_token_accuracy": 0.21272734552621841,
      "num_tokens": 114885496.0,
      "step": 56695
    },
    {
      "entropy": 5.9094805240631105,
      "epoch": 4.954994319671415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028129995054466635,
      "loss": 5.2201,
      "mean_token_accuracy": 0.21282554864883424,
      "num_tokens": 114894733.0,
      "step": 56700
    },
    {
      "entropy": 5.931108522415161,
      "epoch": 4.955431268024119,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002812688016037003,
      "loss": 5.2851,
      "mean_token_accuracy": 0.2051680266857147,
      "num_tokens": 114904991.0,
      "step": 56705
    },
    {
      "entropy": 5.857975339889526,
      "epoch": 4.955868216376825,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002812376525424953,
      "loss": 5.2851,
      "mean_token_accuracy": 0.2131008341908455,
      "num_tokens": 114915296.0,
      "step": 56710
    },
    {
      "entropy": 5.904486179351807,
      "epoch": 4.956305164729529,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002812065033616489,
      "loss": 5.2846,
      "mean_token_accuracy": 0.20681697279214858,
      "num_tokens": 114924614.0,
      "step": 56715
    },
    {
      "entropy": 5.8177046298980715,
      "epoch": 4.956742113082234,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028117535406175847,
      "loss": 5.2175,
      "mean_token_accuracy": 0.2153631940484047,
      "num_tokens": 114934252.0,
      "step": 56720
    },
    {
      "entropy": 5.886100339889526,
      "epoch": 4.957179061434938,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002811442046434216,
      "loss": 5.2783,
      "mean_token_accuracy": 0.20758417397737502,
      "num_tokens": 114944163.0,
      "step": 56725
    },
    {
      "entropy": 5.8623589992523195,
      "epoch": 4.957616009787643,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002811130551072356,
      "loss": 5.1713,
      "mean_token_accuracy": 0.21257021129131318,
      "num_tokens": 114953844.0,
      "step": 56730
    },
    {
      "entropy": 5.86158561706543,
      "epoch": 4.958052958140348,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000281081905453798,
      "loss": 5.1816,
      "mean_token_accuracy": 0.2204958602786064,
      "num_tokens": 114964395.0,
      "step": 56735
    },
    {
      "entropy": 5.914341020584106,
      "epoch": 4.958489906493052,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0002810507556837063,
      "loss": 5.3764,
      "mean_token_accuracy": 0.20391901731491088,
      "num_tokens": 114975479.0,
      "step": 56740
    },
    {
      "entropy": 5.84303994178772,
      "epoch": 4.958926854845757,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002810196057975579,
      "loss": 5.2296,
      "mean_token_accuracy": 0.21520532369613649,
      "num_tokens": 114984580.0,
      "step": 56745
    },
    {
      "entropy": 5.903133296966553,
      "epoch": 4.959363803198462,
      "grad_norm": 1.125,
      "learning_rate": 0.0002809884557959503,
      "loss": 5.1892,
      "mean_token_accuracy": 0.21794378757476807,
      "num_tokens": 114994623.0,
      "step": 56750
    },
    {
      "entropy": 5.847520637512207,
      "epoch": 4.959800751551167,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028095730567948105,
      "loss": 5.2283,
      "mean_token_accuracy": 0.21393343806266785,
      "num_tokens": 115003593.0,
      "step": 56755
    },
    {
      "entropy": 5.8733104228973385,
      "epoch": 4.960237699903871,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002809261554487475,
      "loss": 5.3018,
      "mean_token_accuracy": 0.20550901293754578,
      "num_tokens": 115013781.0,
      "step": 56760
    },
    {
      "entropy": 5.91970419883728,
      "epoch": 4.9606746482565764,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028089500510434726,
      "loss": 5.3489,
      "mean_token_accuracy": 0.20693806260824205,
      "num_tokens": 115023392.0,
      "step": 56765
    },
    {
      "entropy": 5.879947662353516,
      "epoch": 4.961111596609281,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028086385464687764,
      "loss": 5.3036,
      "mean_token_accuracy": 0.20208583921194076,
      "num_tokens": 115033774.0,
      "step": 56770
    },
    {
      "entropy": 5.930194139480591,
      "epoch": 4.961548544961985,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002808327040769362,
      "loss": 5.277,
      "mean_token_accuracy": 0.20570868104696274,
      "num_tokens": 115044536.0,
      "step": 56775
    },
    {
      "entropy": 5.898545646667481,
      "epoch": 4.96198549331469,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028080155339512043,
      "loss": 5.1998,
      "mean_token_accuracy": 0.2109905168414116,
      "num_tokens": 115055611.0,
      "step": 56780
    },
    {
      "entropy": 5.793706369400025,
      "epoch": 4.962422441667395,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002807704026020279,
      "loss": 5.2092,
      "mean_token_accuracy": 0.22088915705680848,
      "num_tokens": 115065082.0,
      "step": 56785
    },
    {
      "entropy": 5.853241014480591,
      "epoch": 4.9628593900201,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028073925169825605,
      "loss": 5.2758,
      "mean_token_accuracy": 0.20642005354166032,
      "num_tokens": 115074680.0,
      "step": 56790
    },
    {
      "entropy": 5.90534234046936,
      "epoch": 4.963296338372804,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002807081006844023,
      "loss": 5.3016,
      "mean_token_accuracy": 0.2083236590027809,
      "num_tokens": 115083966.0,
      "step": 56795
    },
    {
      "entropy": 5.955673408508301,
      "epoch": 4.963733286725509,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028067694956106416,
      "loss": 5.279,
      "mean_token_accuracy": 0.20654804557561873,
      "num_tokens": 115093962.0,
      "step": 56800
    },
    {
      "entropy": 5.864563131332398,
      "epoch": 4.964170235078214,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002806457983288392,
      "loss": 5.2306,
      "mean_token_accuracy": 0.2147394150495529,
      "num_tokens": 115104401.0,
      "step": 56805
    },
    {
      "entropy": 5.85514268875122,
      "epoch": 4.9646071834309184,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002806146469883248,
      "loss": 5.2692,
      "mean_token_accuracy": 0.2008014589548111,
      "num_tokens": 115114705.0,
      "step": 56810
    },
    {
      "entropy": 5.914950132369995,
      "epoch": 4.965044131783623,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002805834955401186,
      "loss": 5.3065,
      "mean_token_accuracy": 0.20801512748003007,
      "num_tokens": 115124499.0,
      "step": 56815
    },
    {
      "entropy": 5.861837673187256,
      "epoch": 4.965481080136328,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000280552343984818,
      "loss": 5.2374,
      "mean_token_accuracy": 0.2126064583659172,
      "num_tokens": 115134285.0,
      "step": 56820
    },
    {
      "entropy": 5.888640975952148,
      "epoch": 4.965918028489033,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002805211923230205,
      "loss": 5.3148,
      "mean_token_accuracy": 0.20498149394989013,
      "num_tokens": 115144984.0,
      "step": 56825
    },
    {
      "entropy": 5.914325857162476,
      "epoch": 4.966354976841737,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028049004055532374,
      "loss": 5.3041,
      "mean_token_accuracy": 0.21081950515508652,
      "num_tokens": 115154375.0,
      "step": 56830
    },
    {
      "entropy": 5.847874784469605,
      "epoch": 4.966791925194442,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028045888868232497,
      "loss": 5.221,
      "mean_token_accuracy": 0.21751132011413574,
      "num_tokens": 115164433.0,
      "step": 56835
    },
    {
      "entropy": 5.8698677062988285,
      "epoch": 4.967228873547147,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028042773670462203,
      "loss": 5.2508,
      "mean_token_accuracy": 0.21140339970588684,
      "num_tokens": 115174235.0,
      "step": 56840
    },
    {
      "entropy": 5.822379016876221,
      "epoch": 4.9676658218998515,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028039658462281216,
      "loss": 5.2174,
      "mean_token_accuracy": 0.2253788009285927,
      "num_tokens": 115184149.0,
      "step": 56845
    },
    {
      "entropy": 5.804949951171875,
      "epoch": 4.968102770252556,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028036543243749296,
      "loss": 5.2032,
      "mean_token_accuracy": 0.21007646024227142,
      "num_tokens": 115195093.0,
      "step": 56850
    },
    {
      "entropy": 5.8390364170074465,
      "epoch": 4.9685397186052604,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000280334280149262,
      "loss": 5.2874,
      "mean_token_accuracy": 0.21202534884214402,
      "num_tokens": 115205972.0,
      "step": 56855
    },
    {
      "entropy": 5.904137516021729,
      "epoch": 4.968976666957966,
      "grad_norm": 1.125,
      "learning_rate": 0.00028030312775871667,
      "loss": 5.2738,
      "mean_token_accuracy": 0.2089862510561943,
      "num_tokens": 115216260.0,
      "step": 56860
    },
    {
      "entropy": 5.853262710571289,
      "epoch": 4.96941361531067,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028027197526645463,
      "loss": 5.2071,
      "mean_token_accuracy": 0.21369143277406694,
      "num_tokens": 115225845.0,
      "step": 56865
    },
    {
      "entropy": 5.9266985893249515,
      "epoch": 4.969850563663375,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002802408226730734,
      "loss": 5.3498,
      "mean_token_accuracy": 0.19510104954242707,
      "num_tokens": 115235951.0,
      "step": 56870
    },
    {
      "entropy": 5.873304414749145,
      "epoch": 4.97028751201608,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028020966997917035,
      "loss": 5.2666,
      "mean_token_accuracy": 0.2130429044365883,
      "num_tokens": 115246382.0,
      "step": 56875
    },
    {
      "entropy": 5.841899394989014,
      "epoch": 4.970724460368785,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002801785171853431,
      "loss": 5.1933,
      "mean_token_accuracy": 0.21930330097675324,
      "num_tokens": 115256566.0,
      "step": 56880
    },
    {
      "entropy": 5.841558742523193,
      "epoch": 4.971161408721489,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028014736429218926,
      "loss": 5.2099,
      "mean_token_accuracy": 0.21835994571447373,
      "num_tokens": 115266506.0,
      "step": 56885
    },
    {
      "entropy": 5.872181749343872,
      "epoch": 4.9715983570741935,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028011621130030626,
      "loss": 5.2729,
      "mean_token_accuracy": 0.20600930154323577,
      "num_tokens": 115276671.0,
      "step": 56890
    },
    {
      "entropy": 5.853465557098389,
      "epoch": 4.972035305426899,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002800850582102916,
      "loss": 5.2827,
      "mean_token_accuracy": 0.20804230719804764,
      "num_tokens": 115285911.0,
      "step": 56895
    },
    {
      "entropy": 5.7562402248382565,
      "epoch": 4.972472253779603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028005390502274283,
      "loss": 5.1563,
      "mean_token_accuracy": 0.22809349298477172,
      "num_tokens": 115294931.0,
      "step": 56900
    },
    {
      "entropy": 5.864391565322876,
      "epoch": 4.972909202132308,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002800227517382576,
      "loss": 5.3101,
      "mean_token_accuracy": 0.2026732087135315,
      "num_tokens": 115304568.0,
      "step": 56905
    },
    {
      "entropy": 5.864074563980102,
      "epoch": 4.973346150485012,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002799915983574333,
      "loss": 5.2683,
      "mean_token_accuracy": 0.21391597092151643,
      "num_tokens": 115315067.0,
      "step": 56910
    },
    {
      "entropy": 5.87061505317688,
      "epoch": 4.973783098837718,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002799604448808676,
      "loss": 5.2389,
      "mean_token_accuracy": 0.21416021883487701,
      "num_tokens": 115324963.0,
      "step": 56915
    },
    {
      "entropy": 5.856711435317993,
      "epoch": 4.974220047190422,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002799292913091579,
      "loss": 5.31,
      "mean_token_accuracy": 0.20867951065301896,
      "num_tokens": 115335671.0,
      "step": 56920
    },
    {
      "entropy": 5.858908843994141,
      "epoch": 4.974656995543127,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002798981376429018,
      "loss": 5.2218,
      "mean_token_accuracy": 0.21351610273122787,
      "num_tokens": 115345020.0,
      "step": 56925
    },
    {
      "entropy": 5.876742649078369,
      "epoch": 4.975093943895832,
      "grad_norm": 1.375,
      "learning_rate": 0.0002798669838826968,
      "loss": 5.2567,
      "mean_token_accuracy": 0.21000246703624725,
      "num_tokens": 115354584.0,
      "step": 56930
    },
    {
      "entropy": 5.928784561157227,
      "epoch": 4.975530892248536,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002798358300291407,
      "loss": 5.3095,
      "mean_token_accuracy": 0.20568659603595735,
      "num_tokens": 115365451.0,
      "step": 56935
    },
    {
      "entropy": 5.9414434909820555,
      "epoch": 4.975967840601241,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027980467608283065,
      "loss": 5.3071,
      "mean_token_accuracy": 0.2076416626572609,
      "num_tokens": 115375073.0,
      "step": 56940
    },
    {
      "entropy": 5.864425182342529,
      "epoch": 4.976404788953945,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027977352204436453,
      "loss": 5.2431,
      "mean_token_accuracy": 0.20711090862751008,
      "num_tokens": 115385564.0,
      "step": 56945
    },
    {
      "entropy": 5.893436050415039,
      "epoch": 4.976841737306651,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002797423679143396,
      "loss": 5.3405,
      "mean_token_accuracy": 0.20415582060813903,
      "num_tokens": 115397118.0,
      "step": 56950
    },
    {
      "entropy": 5.860381650924682,
      "epoch": 4.977278685659355,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002797112136933537,
      "loss": 5.2304,
      "mean_token_accuracy": 0.21426864713430405,
      "num_tokens": 115407681.0,
      "step": 56955
    },
    {
      "entropy": 5.8589186668396,
      "epoch": 4.97771563401206,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002796800593820041,
      "loss": 5.1951,
      "mean_token_accuracy": 0.20989157855510712,
      "num_tokens": 115417449.0,
      "step": 56960
    },
    {
      "entropy": 5.864449548721313,
      "epoch": 4.978152582364764,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027964890498088875,
      "loss": 5.1722,
      "mean_token_accuracy": 0.21426600366830825,
      "num_tokens": 115426932.0,
      "step": 56965
    },
    {
      "entropy": 5.957550001144409,
      "epoch": 4.9785895307174695,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002796177504906048,
      "loss": 5.4083,
      "mean_token_accuracy": 0.19875203520059587,
      "num_tokens": 115437381.0,
      "step": 56970
    },
    {
      "entropy": 5.86564269065857,
      "epoch": 4.979026479070174,
      "grad_norm": 1.046875,
      "learning_rate": 0.00027958659591175,
      "loss": 5.3727,
      "mean_token_accuracy": 0.20553605854511262,
      "num_tokens": 115448452.0,
      "step": 56975
    },
    {
      "entropy": 5.90580382347107,
      "epoch": 4.979463427422878,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00027955544124492196,
      "loss": 5.2879,
      "mean_token_accuracy": 0.19809358268976213,
      "num_tokens": 115458466.0,
      "step": 56980
    },
    {
      "entropy": 5.887865877151489,
      "epoch": 4.979900375775584,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002795242864907181,
      "loss": 5.3099,
      "mean_token_accuracy": 0.20254219621419906,
      "num_tokens": 115468330.0,
      "step": 56985
    },
    {
      "entropy": 5.88407883644104,
      "epoch": 4.980337324128288,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00027949313164973605,
      "loss": 5.2323,
      "mean_token_accuracy": 0.21418397575616838,
      "num_tokens": 115478385.0,
      "step": 56990
    },
    {
      "entropy": 5.921640205383301,
      "epoch": 4.980774272480993,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002794619767225733,
      "loss": 5.374,
      "mean_token_accuracy": 0.1965426906943321,
      "num_tokens": 115488907.0,
      "step": 56995
    },
    {
      "entropy": 5.832983112335205,
      "epoch": 4.981211220833697,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002794308217098276,
      "loss": 5.1961,
      "mean_token_accuracy": 0.21596582233905792,
      "num_tokens": 115498399.0,
      "step": 57000
    },
    {
      "epoch": 4.981211220833697,
      "eval_entropy": 5.593895349282025,
      "eval_loss": 5.330423355102539,
      "eval_mean_token_accuracy": 0.2153679423939146,
      "eval_num_tokens": 115498399.0,
      "eval_runtime": 27.8711,
      "eval_samples_per_second": 1328.043,
      "eval_steps_per_second": 166.014,
      "step": 57000
    },
    {
      "entropy": 5.8339958667755125,
      "epoch": 4.9816481691864025,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027939966661209647,
      "loss": 5.1468,
      "mean_token_accuracy": 0.21723688691854476,
      "num_tokens": 115508145.0,
      "step": 57005
    },
    {
      "entropy": 5.848921775817871,
      "epoch": 4.982085117539107,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027936851142997736,
      "loss": 5.2729,
      "mean_token_accuracy": 0.20736686885356903,
      "num_tokens": 115517816.0,
      "step": 57010
    },
    {
      "entropy": 5.817902231216431,
      "epoch": 4.9825220658918115,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002793373561640679,
      "loss": 5.232,
      "mean_token_accuracy": 0.22021067142486572,
      "num_tokens": 115527673.0,
      "step": 57015
    },
    {
      "entropy": 5.850135278701782,
      "epoch": 4.982959014244516,
      "grad_norm": 1.125,
      "learning_rate": 0.00027930620081496573,
      "loss": 5.169,
      "mean_token_accuracy": 0.2111182540655136,
      "num_tokens": 115537507.0,
      "step": 57020
    },
    {
      "entropy": 5.864220428466797,
      "epoch": 4.983395962597221,
      "grad_norm": 1.125,
      "learning_rate": 0.0002792750453832683,
      "loss": 5.238,
      "mean_token_accuracy": 0.21328955590724946,
      "num_tokens": 115547913.0,
      "step": 57025
    },
    {
      "entropy": 5.827035808563233,
      "epoch": 4.983832910949926,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002792438898695734,
      "loss": 5.2199,
      "mean_token_accuracy": 0.22412694841623307,
      "num_tokens": 115558074.0,
      "step": 57030
    },
    {
      "entropy": 5.831992530822754,
      "epoch": 4.98426985930263,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027921273427447833,
      "loss": 5.2361,
      "mean_token_accuracy": 0.2133249372243881,
      "num_tokens": 115568161.0,
      "step": 57035
    },
    {
      "entropy": 5.937843084335327,
      "epoch": 4.984706807655336,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027918157859858087,
      "loss": 5.3501,
      "mean_token_accuracy": 0.20367964059114457,
      "num_tokens": 115578334.0,
      "step": 57040
    },
    {
      "entropy": 5.885875844955445,
      "epoch": 4.98514375600804,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027915042284247846,
      "loss": 5.286,
      "mean_token_accuracy": 0.211648653447628,
      "num_tokens": 115588251.0,
      "step": 57045
    },
    {
      "entropy": 5.755224752426147,
      "epoch": 4.9855807043607445,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002791192670067689,
      "loss": 5.221,
      "mean_token_accuracy": 0.2130904823541641,
      "num_tokens": 115598748.0,
      "step": 57050
    },
    {
      "entropy": 5.863641691207886,
      "epoch": 4.986017652713449,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002790881110920495,
      "loss": 5.2727,
      "mean_token_accuracy": 0.20822433531284332,
      "num_tokens": 115608154.0,
      "step": 57055
    },
    {
      "entropy": 5.916394662857056,
      "epoch": 4.986454601066154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027905695509891805,
      "loss": 5.3006,
      "mean_token_accuracy": 0.21145616024732589,
      "num_tokens": 115617409.0,
      "step": 57060
    },
    {
      "entropy": 5.854255485534668,
      "epoch": 4.986891549418859,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002790257990279721,
      "loss": 5.2001,
      "mean_token_accuracy": 0.21629361808300018,
      "num_tokens": 115627994.0,
      "step": 57065
    },
    {
      "entropy": 5.838038206100464,
      "epoch": 4.987328497771563,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002789946428798092,
      "loss": 5.2246,
      "mean_token_accuracy": 0.2155321106314659,
      "num_tokens": 115638314.0,
      "step": 57070
    },
    {
      "entropy": 5.902239084243774,
      "epoch": 4.987765446124268,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027896348665502695,
      "loss": 5.2959,
      "mean_token_accuracy": 0.2053772896528244,
      "num_tokens": 115648411.0,
      "step": 57075
    },
    {
      "entropy": 5.8544477939605715,
      "epoch": 4.988202394476973,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000278932330354223,
      "loss": 5.2416,
      "mean_token_accuracy": 0.21594252437353134,
      "num_tokens": 115658850.0,
      "step": 57080
    },
    {
      "entropy": 5.836382865905762,
      "epoch": 4.988639342829678,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002789011739779948,
      "loss": 5.2565,
      "mean_token_accuracy": 0.20675360411405563,
      "num_tokens": 115668413.0,
      "step": 57085
    },
    {
      "entropy": 5.8764525890350345,
      "epoch": 4.989076291182382,
      "grad_norm": 1.015625,
      "learning_rate": 0.00027887001752694004,
      "loss": 5.2409,
      "mean_token_accuracy": 0.2102430760860443,
      "num_tokens": 115677872.0,
      "step": 57090
    },
    {
      "entropy": 5.835366725921631,
      "epoch": 4.989513239535087,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027883886100165637,
      "loss": 5.2347,
      "mean_token_accuracy": 0.215859255194664,
      "num_tokens": 115689589.0,
      "step": 57095
    },
    {
      "entropy": 5.927762222290039,
      "epoch": 4.989950187887792,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002788077044027413,
      "loss": 5.3364,
      "mean_token_accuracy": 0.2121294468641281,
      "num_tokens": 115699951.0,
      "step": 57100
    },
    {
      "entropy": 5.897933673858643,
      "epoch": 4.990387136240496,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027877654773079253,
      "loss": 5.2851,
      "mean_token_accuracy": 0.20758446753025056,
      "num_tokens": 115710296.0,
      "step": 57105
    },
    {
      "entropy": 5.845287799835205,
      "epoch": 4.990824084593201,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002787453909864075,
      "loss": 5.2287,
      "mean_token_accuracy": 0.20966563373804092,
      "num_tokens": 115719627.0,
      "step": 57110
    },
    {
      "entropy": 5.751111268997192,
      "epoch": 4.991261032945906,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000278714234170184,
      "loss": 5.0948,
      "mean_token_accuracy": 0.22704849094152452,
      "num_tokens": 115729839.0,
      "step": 57115
    },
    {
      "entropy": 5.841661214828491,
      "epoch": 4.991697981298611,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027868307728271945,
      "loss": 5.2224,
      "mean_token_accuracy": 0.21429407000541686,
      "num_tokens": 115739664.0,
      "step": 57120
    },
    {
      "entropy": 5.9356879711151125,
      "epoch": 4.992134929651315,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002786519203246117,
      "loss": 5.3214,
      "mean_token_accuracy": 0.19930350333452224,
      "num_tokens": 115749927.0,
      "step": 57125
    },
    {
      "entropy": 5.861351585388183,
      "epoch": 4.99257187800402,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027862076329645803,
      "loss": 5.2427,
      "mean_token_accuracy": 0.21110230237245559,
      "num_tokens": 115759804.0,
      "step": 57130
    },
    {
      "entropy": 5.879577589035034,
      "epoch": 4.993008826356725,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027858960619885623,
      "loss": 5.24,
      "mean_token_accuracy": 0.2062641277909279,
      "num_tokens": 115770482.0,
      "step": 57135
    },
    {
      "entropy": 5.85199384689331,
      "epoch": 4.993445774709429,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000278558449032404,
      "loss": 5.2648,
      "mean_token_accuracy": 0.21007518768310546,
      "num_tokens": 115780691.0,
      "step": 57140
    },
    {
      "entropy": 5.897588586807251,
      "epoch": 4.993882723062134,
      "grad_norm": 1.125,
      "learning_rate": 0.0002785272917976989,
      "loss": 5.3105,
      "mean_token_accuracy": 0.21201534271240235,
      "num_tokens": 115790416.0,
      "step": 57145
    },
    {
      "entropy": 5.973320388793946,
      "epoch": 4.994319671414839,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027849613449533835,
      "loss": 5.331,
      "mean_token_accuracy": 0.20661496818065644,
      "num_tokens": 115800327.0,
      "step": 57150
    },
    {
      "entropy": 5.946268510818482,
      "epoch": 4.994756619767544,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027846497712592024,
      "loss": 5.2823,
      "mean_token_accuracy": 0.206516270339489,
      "num_tokens": 115809472.0,
      "step": 57155
    },
    {
      "entropy": 5.877649784088135,
      "epoch": 4.995193568120248,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002784338196900419,
      "loss": 5.3363,
      "mean_token_accuracy": 0.20253763049840928,
      "num_tokens": 115820113.0,
      "step": 57160
    },
    {
      "entropy": 5.8816712379455565,
      "epoch": 4.995630516472953,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027840266218830123,
      "loss": 5.2656,
      "mean_token_accuracy": 0.21045792549848558,
      "num_tokens": 115830647.0,
      "step": 57165
    },
    {
      "entropy": 5.994711828231812,
      "epoch": 4.996067464825657,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027837150462129567,
      "loss": 5.3928,
      "mean_token_accuracy": 0.19574316143989562,
      "num_tokens": 115840918.0,
      "step": 57170
    },
    {
      "entropy": 5.923894023895263,
      "epoch": 4.9965044131783625,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00027834034698962293,
      "loss": 5.3071,
      "mean_token_accuracy": 0.20805004090070725,
      "num_tokens": 115851323.0,
      "step": 57175
    },
    {
      "entropy": 5.879899168014527,
      "epoch": 4.996941361531067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002783091892938805,
      "loss": 5.309,
      "mean_token_accuracy": 0.21127871721982955,
      "num_tokens": 115862182.0,
      "step": 57180
    },
    {
      "entropy": 5.874031019210816,
      "epoch": 4.997378309883771,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002782780315346661,
      "loss": 5.2198,
      "mean_token_accuracy": 0.21486542522907257,
      "num_tokens": 115872437.0,
      "step": 57185
    },
    {
      "entropy": 5.93183536529541,
      "epoch": 4.997815258236477,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027824687371257744,
      "loss": 5.3551,
      "mean_token_accuracy": 0.20161037743091584,
      "num_tokens": 115882943.0,
      "step": 57190
    },
    {
      "entropy": 5.838115501403808,
      "epoch": 4.998252206589181,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002782157158282119,
      "loss": 5.1576,
      "mean_token_accuracy": 0.2182507961988449,
      "num_tokens": 115892539.0,
      "step": 57195
    },
    {
      "entropy": 5.8771164894104,
      "epoch": 4.998689154941886,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027818455788216733,
      "loss": 5.3295,
      "mean_token_accuracy": 0.20728321075439454,
      "num_tokens": 115902273.0,
      "step": 57200
    },
    {
      "entropy": 5.901480102539063,
      "epoch": 4.999126103294591,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002781533998750413,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21384452283382416,
      "num_tokens": 115912286.0,
      "step": 57205
    },
    {
      "entropy": 5.828431177139282,
      "epoch": 4.9995630516472955,
      "grad_norm": 1.125,
      "learning_rate": 0.0002781222418074313,
      "loss": 5.1884,
      "mean_token_accuracy": 0.2262457564473152,
      "num_tokens": 115922476.0,
      "step": 57210
    },
    {
      "entropy": 5.846165943145752,
      "epoch": 5.0,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027809108367993516,
      "loss": 5.2104,
      "mean_token_accuracy": 0.21287480145692825,
      "num_tokens": 115933010.0,
      "step": 57215
    },
    {
      "entropy": 5.861330986022949,
      "epoch": 5.0004369483527045,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002780599254931504,
      "loss": 5.2214,
      "mean_token_accuracy": 0.20994778126478195,
      "num_tokens": 115942333.0,
      "step": 57220
    },
    {
      "entropy": 5.934155797958374,
      "epoch": 5.00087389670541,
      "grad_norm": 1.25,
      "learning_rate": 0.0002780287672476746,
      "loss": 5.232,
      "mean_token_accuracy": 0.2161656454205513,
      "num_tokens": 115951821.0,
      "step": 57225
    },
    {
      "entropy": 5.891773891448975,
      "epoch": 5.001310845058114,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002779976089441055,
      "loss": 5.2051,
      "mean_token_accuracy": 0.2194888949394226,
      "num_tokens": 115961830.0,
      "step": 57230
    },
    {
      "entropy": 5.853494882583618,
      "epoch": 5.001747793410819,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002779664505830407,
      "loss": 5.2328,
      "mean_token_accuracy": 0.2155277669429779,
      "num_tokens": 115971584.0,
      "step": 57235
    },
    {
      "entropy": 5.877412748336792,
      "epoch": 5.002184741763523,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002779352921650779,
      "loss": 5.2403,
      "mean_token_accuracy": 0.21528198570013046,
      "num_tokens": 115981949.0,
      "step": 57240
    },
    {
      "entropy": 5.898882055282593,
      "epoch": 5.002621690116229,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00027790413369081454,
      "loss": 5.2045,
      "mean_token_accuracy": 0.21395424008369446,
      "num_tokens": 115992301.0,
      "step": 57245
    },
    {
      "entropy": 5.849624729156494,
      "epoch": 5.003058638468933,
      "grad_norm": 1.125,
      "learning_rate": 0.0002778729751608484,
      "loss": 5.1918,
      "mean_token_accuracy": 0.21927930265665055,
      "num_tokens": 116002205.0,
      "step": 57250
    },
    {
      "entropy": 5.876811456680298,
      "epoch": 5.0034955868216375,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002778418165757771,
      "loss": 5.259,
      "mean_token_accuracy": 0.21375482976436616,
      "num_tokens": 116011876.0,
      "step": 57255
    },
    {
      "entropy": 5.8821855068206785,
      "epoch": 5.003932535174342,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002778106579361983,
      "loss": 5.186,
      "mean_token_accuracy": 0.2202381372451782,
      "num_tokens": 116021813.0,
      "step": 57260
    },
    {
      "entropy": 5.855607318878174,
      "epoch": 5.004369483527047,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002777794992427096,
      "loss": 5.1685,
      "mean_token_accuracy": 0.21779284179210662,
      "num_tokens": 116031480.0,
      "step": 57265
    },
    {
      "entropy": 5.742363452911377,
      "epoch": 5.004806431879752,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027774834049590857,
      "loss": 5.0787,
      "mean_token_accuracy": 0.228927443921566,
      "num_tokens": 116041476.0,
      "step": 57270
    },
    {
      "entropy": 5.850942134857178,
      "epoch": 5.005243380232456,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000277717181696393,
      "loss": 5.2155,
      "mean_token_accuracy": 0.2162558302283287,
      "num_tokens": 116052750.0,
      "step": 57275
    },
    {
      "entropy": 5.827357006072998,
      "epoch": 5.005680328585162,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002776860228447604,
      "loss": 5.1211,
      "mean_token_accuracy": 0.22354120314121245,
      "num_tokens": 116062827.0,
      "step": 57280
    },
    {
      "entropy": 5.818589067459106,
      "epoch": 5.006117276937866,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002776548639416085,
      "loss": 5.2498,
      "mean_token_accuracy": 0.22036105990409852,
      "num_tokens": 116073623.0,
      "step": 57285
    },
    {
      "entropy": 5.8477527618408205,
      "epoch": 5.006554225290571,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027762370498753505,
      "loss": 5.1674,
      "mean_token_accuracy": 0.21221791505813598,
      "num_tokens": 116084143.0,
      "step": 57290
    },
    {
      "entropy": 5.828296041488647,
      "epoch": 5.006991173643275,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027759254598313745,
      "loss": 5.184,
      "mean_token_accuracy": 0.216710065305233,
      "num_tokens": 116094147.0,
      "step": 57295
    },
    {
      "entropy": 5.895185279846191,
      "epoch": 5.00742812199598,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027756138692901347,
      "loss": 5.2714,
      "mean_token_accuracy": 0.2069227233529091,
      "num_tokens": 116104438.0,
      "step": 57300
    },
    {
      "entropy": 5.906028366088867,
      "epoch": 5.007865070348685,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002775302278257606,
      "loss": 5.2041,
      "mean_token_accuracy": 0.21709794998168946,
      "num_tokens": 116115036.0,
      "step": 57305
    },
    {
      "entropy": 5.793821144104004,
      "epoch": 5.008302018701389,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00027749906867397683,
      "loss": 5.1611,
      "mean_token_accuracy": 0.22164538502693176,
      "num_tokens": 116125632.0,
      "step": 57310
    },
    {
      "entropy": 5.847276115417481,
      "epoch": 5.008738967054094,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002774679094742595,
      "loss": 5.2858,
      "mean_token_accuracy": 0.21164631396532058,
      "num_tokens": 116136289.0,
      "step": 57315
    },
    {
      "entropy": 5.851640462875366,
      "epoch": 5.009175915406799,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027743675022720647,
      "loss": 5.2271,
      "mean_token_accuracy": 0.2169113725423813,
      "num_tokens": 116146479.0,
      "step": 57320
    },
    {
      "entropy": 5.918398332595825,
      "epoch": 5.009612863759504,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002774055909334152,
      "loss": 5.2221,
      "mean_token_accuracy": 0.21294637024402618,
      "num_tokens": 116156168.0,
      "step": 57325
    },
    {
      "entropy": 5.9114399433135985,
      "epoch": 5.010049812112208,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027737443159348354,
      "loss": 5.2823,
      "mean_token_accuracy": 0.20495997071266175,
      "num_tokens": 116165881.0,
      "step": 57330
    },
    {
      "entropy": 5.880865001678467,
      "epoch": 5.0104867604649135,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027734327220800905,
      "loss": 5.2174,
      "mean_token_accuracy": 0.21251231878995896,
      "num_tokens": 116176002.0,
      "step": 57335
    },
    {
      "entropy": 5.92426872253418,
      "epoch": 5.010923708817618,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002773121127775893,
      "loss": 5.2342,
      "mean_token_accuracy": 0.20947461873292922,
      "num_tokens": 116186561.0,
      "step": 57340
    },
    {
      "entropy": 5.8816112041473385,
      "epoch": 5.011360657170322,
      "grad_norm": 1.09375,
      "learning_rate": 0.000277280953302822,
      "loss": 5.2935,
      "mean_token_accuracy": 0.21550005227327346,
      "num_tokens": 116197817.0,
      "step": 57345
    },
    {
      "entropy": 5.816081905364991,
      "epoch": 5.011797605523027,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027724979378430486,
      "loss": 5.2156,
      "mean_token_accuracy": 0.2193106397986412,
      "num_tokens": 116208369.0,
      "step": 57350
    },
    {
      "entropy": 5.936599254608154,
      "epoch": 5.012234553875732,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002772186342226355,
      "loss": 5.2896,
      "mean_token_accuracy": 0.20888464152812958,
      "num_tokens": 116218745.0,
      "step": 57355
    },
    {
      "entropy": 5.912152194976807,
      "epoch": 5.012671502228437,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027718747461841155,
      "loss": 5.2439,
      "mean_token_accuracy": 0.21392017602920532,
      "num_tokens": 116228985.0,
      "step": 57360
    },
    {
      "entropy": 5.90400185585022,
      "epoch": 5.013108450581141,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00027715631497223067,
      "loss": 5.2916,
      "mean_token_accuracy": 0.20596380531787872,
      "num_tokens": 116239358.0,
      "step": 57365
    },
    {
      "entropy": 5.8644514083862305,
      "epoch": 5.013545398933846,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027712515528469057,
      "loss": 5.185,
      "mean_token_accuracy": 0.21378031522035598,
      "num_tokens": 116249768.0,
      "step": 57370
    },
    {
      "entropy": 5.846675205230713,
      "epoch": 5.013982347286551,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027709399555638886,
      "loss": 5.2238,
      "mean_token_accuracy": 0.21811042129993438,
      "num_tokens": 116260290.0,
      "step": 57375
    },
    {
      "entropy": 5.852652359008789,
      "epoch": 5.0144192956392555,
      "grad_norm": 1.046875,
      "learning_rate": 0.00027706283578792326,
      "loss": 5.1337,
      "mean_token_accuracy": 0.21898607760667801,
      "num_tokens": 116270638.0,
      "step": 57380
    },
    {
      "entropy": 5.875527334213257,
      "epoch": 5.01485624399196,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002770316759798913,
      "loss": 5.1962,
      "mean_token_accuracy": 0.2189212515950203,
      "num_tokens": 116280868.0,
      "step": 57385
    },
    {
      "entropy": 5.841114950180054,
      "epoch": 5.015293192344664,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002770005161328908,
      "loss": 5.1286,
      "mean_token_accuracy": 0.22563961893320084,
      "num_tokens": 116290379.0,
      "step": 57390
    },
    {
      "entropy": 5.8604518413543705,
      "epoch": 5.01573014069737,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027696935624751934,
      "loss": 5.2148,
      "mean_token_accuracy": 0.21877849102020264,
      "num_tokens": 116301058.0,
      "step": 57395
    },
    {
      "entropy": 5.8251348495483395,
      "epoch": 5.016167089050074,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002769381963243746,
      "loss": 5.1916,
      "mean_token_accuracy": 0.21262338608503342,
      "num_tokens": 116309756.0,
      "step": 57400
    },
    {
      "entropy": 5.885043334960938,
      "epoch": 5.016604037402779,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002769070363640543,
      "loss": 5.228,
      "mean_token_accuracy": 0.22090415954589843,
      "num_tokens": 116319533.0,
      "step": 57405
    },
    {
      "entropy": 5.867009496688842,
      "epoch": 5.017040985755484,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002768758763671559,
      "loss": 5.201,
      "mean_token_accuracy": 0.21657511591911316,
      "num_tokens": 116329805.0,
      "step": 57410
    },
    {
      "entropy": 5.888688373565674,
      "epoch": 5.0174779341081885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002768447163342773,
      "loss": 5.192,
      "mean_token_accuracy": 0.22266086637973787,
      "num_tokens": 116340921.0,
      "step": 57415
    },
    {
      "entropy": 5.894060707092285,
      "epoch": 5.017914882460893,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027681355626601605,
      "loss": 5.2572,
      "mean_token_accuracy": 0.2141995757818222,
      "num_tokens": 116351808.0,
      "step": 57420
    },
    {
      "entropy": 5.853298139572144,
      "epoch": 5.0183518308135975,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027678239616296983,
      "loss": 5.222,
      "mean_token_accuracy": 0.2154070869088173,
      "num_tokens": 116362213.0,
      "step": 57425
    },
    {
      "entropy": 5.94111123085022,
      "epoch": 5.018788779166303,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002767512360257364,
      "loss": 5.2458,
      "mean_token_accuracy": 0.21290275156497956,
      "num_tokens": 116373843.0,
      "step": 57430
    },
    {
      "entropy": 5.93305082321167,
      "epoch": 5.019225727519007,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027672007585491327,
      "loss": 5.3763,
      "mean_token_accuracy": 0.2002154529094696,
      "num_tokens": 116384200.0,
      "step": 57435
    },
    {
      "entropy": 5.795771789550781,
      "epoch": 5.019662675871712,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002766889156510982,
      "loss": 5.1708,
      "mean_token_accuracy": 0.2260590136051178,
      "num_tokens": 116394969.0,
      "step": 57440
    },
    {
      "entropy": 5.8576075553894045,
      "epoch": 5.020099624224416,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002766577554148888,
      "loss": 5.1614,
      "mean_token_accuracy": 0.2160823494195938,
      "num_tokens": 116404611.0,
      "step": 57445
    },
    {
      "entropy": 5.900273418426513,
      "epoch": 5.020536572577122,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027662659514688287,
      "loss": 5.2435,
      "mean_token_accuracy": 0.2098928987979889,
      "num_tokens": 116414614.0,
      "step": 57450
    },
    {
      "entropy": 5.87584285736084,
      "epoch": 5.020973520929826,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027659543484767797,
      "loss": 5.2032,
      "mean_token_accuracy": 0.21906380206346512,
      "num_tokens": 116424178.0,
      "step": 57455
    },
    {
      "entropy": 5.936579608917237,
      "epoch": 5.0214104692825305,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027656427451787174,
      "loss": 5.2799,
      "mean_token_accuracy": 0.20881435871124268,
      "num_tokens": 116434376.0,
      "step": 57460
    },
    {
      "entropy": 5.869890403747559,
      "epoch": 5.021847417635236,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027653311415806194,
      "loss": 5.1429,
      "mean_token_accuracy": 0.22018850296735765,
      "num_tokens": 116443424.0,
      "step": 57465
    },
    {
      "entropy": 5.842494916915894,
      "epoch": 5.02228436598794,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002765019537688462,
      "loss": 5.2136,
      "mean_token_accuracy": 0.21324381828308106,
      "num_tokens": 116452515.0,
      "step": 57470
    },
    {
      "entropy": 5.831165027618408,
      "epoch": 5.022721314340645,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002764707933508222,
      "loss": 5.228,
      "mean_token_accuracy": 0.21359838992357255,
      "num_tokens": 116463699.0,
      "step": 57475
    },
    {
      "entropy": 5.871450710296631,
      "epoch": 5.023158262693349,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027643963290458764,
      "loss": 5.2083,
      "mean_token_accuracy": 0.20636752396821975,
      "num_tokens": 116474650.0,
      "step": 57480
    },
    {
      "entropy": 5.87013611793518,
      "epoch": 5.023595211046055,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027640847243074005,
      "loss": 5.2541,
      "mean_token_accuracy": 0.20676225423812866,
      "num_tokens": 116484703.0,
      "step": 57485
    },
    {
      "entropy": 5.8528882503509525,
      "epoch": 5.024032159398759,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002763773119298773,
      "loss": 5.1521,
      "mean_token_accuracy": 0.2185731440782547,
      "num_tokens": 116495247.0,
      "step": 57490
    },
    {
      "entropy": 5.947031259536743,
      "epoch": 5.024469107751464,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002763461514025971,
      "loss": 5.2345,
      "mean_token_accuracy": 0.21469509601593018,
      "num_tokens": 116505100.0,
      "step": 57495
    },
    {
      "entropy": 5.859530639648438,
      "epoch": 5.024906056104168,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002763149908494969,
      "loss": 5.2149,
      "mean_token_accuracy": 0.21126665323972701,
      "num_tokens": 116515573.0,
      "step": 57500
    },
    {
      "entropy": 5.817746782302857,
      "epoch": 5.025343004456873,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027628383027117446,
      "loss": 5.1873,
      "mean_token_accuracy": 0.21747245341539384,
      "num_tokens": 116525234.0,
      "step": 57505
    },
    {
      "entropy": 5.896178388595581,
      "epoch": 5.025779952809578,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002762526696682275,
      "loss": 5.2583,
      "mean_token_accuracy": 0.2055425226688385,
      "num_tokens": 116535759.0,
      "step": 57510
    },
    {
      "entropy": 5.898248910903931,
      "epoch": 5.026216901162282,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027622150904125364,
      "loss": 5.2355,
      "mean_token_accuracy": 0.21095171719789504,
      "num_tokens": 116546533.0,
      "step": 57515
    },
    {
      "entropy": 5.8248778820037845,
      "epoch": 5.026653849514988,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002761903483908507,
      "loss": 5.2092,
      "mean_token_accuracy": 0.2157468780875206,
      "num_tokens": 116557999.0,
      "step": 57520
    },
    {
      "entropy": 5.855276346206665,
      "epoch": 5.027090797867692,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002761591877176163,
      "loss": 5.1833,
      "mean_token_accuracy": 0.22343987673521043,
      "num_tokens": 116567975.0,
      "step": 57525
    },
    {
      "entropy": 5.918731594085694,
      "epoch": 5.027527746220397,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027612802702214797,
      "loss": 5.3013,
      "mean_token_accuracy": 0.21021223515272142,
      "num_tokens": 116578384.0,
      "step": 57530
    },
    {
      "entropy": 5.909542465209961,
      "epoch": 5.027964694573101,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002760968663050434,
      "loss": 5.2688,
      "mean_token_accuracy": 0.20523979961872102,
      "num_tokens": 116588881.0,
      "step": 57535
    },
    {
      "entropy": 5.9253737449646,
      "epoch": 5.0284016429258065,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002760657055669005,
      "loss": 5.262,
      "mean_token_accuracy": 0.20755659490823747,
      "num_tokens": 116599114.0,
      "step": 57540
    },
    {
      "entropy": 5.842081069946289,
      "epoch": 5.028838591278511,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002760345448083168,
      "loss": 5.2584,
      "mean_token_accuracy": 0.21654251366853713,
      "num_tokens": 116609124.0,
      "step": 57545
    },
    {
      "entropy": 5.876411294937133,
      "epoch": 5.029275539631215,
      "grad_norm": 1.046875,
      "learning_rate": 0.00027600338402988995,
      "loss": 5.2992,
      "mean_token_accuracy": 0.20865776538848876,
      "num_tokens": 116619437.0,
      "step": 57550
    },
    {
      "entropy": 5.919462251663208,
      "epoch": 5.02971248798392,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027597222323221763,
      "loss": 5.2635,
      "mean_token_accuracy": 0.2114575371146202,
      "num_tokens": 116629648.0,
      "step": 57555
    },
    {
      "entropy": 5.866595983505249,
      "epoch": 5.030149436336625,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002759410624158976,
      "loss": 5.1821,
      "mean_token_accuracy": 0.21973170191049576,
      "num_tokens": 116639335.0,
      "step": 57560
    },
    {
      "entropy": 5.863292789459228,
      "epoch": 5.03058638468933,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002759099015815275,
      "loss": 5.1896,
      "mean_token_accuracy": 0.21401297450065612,
      "num_tokens": 116649345.0,
      "step": 57565
    },
    {
      "entropy": 5.872456026077271,
      "epoch": 5.031023333042034,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002758787407297051,
      "loss": 5.2236,
      "mean_token_accuracy": 0.21067134588956832,
      "num_tokens": 116659285.0,
      "step": 57570
    },
    {
      "entropy": 5.823410320281982,
      "epoch": 5.0314602813947396,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027584757986102786,
      "loss": 5.1503,
      "mean_token_accuracy": 0.222203029692173,
      "num_tokens": 116669753.0,
      "step": 57575
    },
    {
      "entropy": 5.774070167541504,
      "epoch": 5.031897229747444,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002758164189760936,
      "loss": 5.1962,
      "mean_token_accuracy": 0.2248554840683937,
      "num_tokens": 116679853.0,
      "step": 57580
    },
    {
      "entropy": 5.843558073043823,
      "epoch": 5.0323341781001485,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002757852580755001,
      "loss": 5.2272,
      "mean_token_accuracy": 0.21061434745788574,
      "num_tokens": 116689659.0,
      "step": 57585
    },
    {
      "entropy": 5.827411413192749,
      "epoch": 5.032771126452853,
      "grad_norm": 1.125,
      "learning_rate": 0.00027575409715984486,
      "loss": 5.1508,
      "mean_token_accuracy": 0.21939564347267151,
      "num_tokens": 116699627.0,
      "step": 57590
    },
    {
      "entropy": 5.889414310455322,
      "epoch": 5.033208074805558,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027572293622972574,
      "loss": 5.2887,
      "mean_token_accuracy": 0.21048369109630585,
      "num_tokens": 116711112.0,
      "step": 57595
    },
    {
      "entropy": 5.91390061378479,
      "epoch": 5.033645023158263,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002756917752857401,
      "loss": 5.1597,
      "mean_token_accuracy": 0.22295346558094026,
      "num_tokens": 116720375.0,
      "step": 57600
    },
    {
      "entropy": 5.820985603332519,
      "epoch": 5.034081971510967,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027566061432848603,
      "loss": 5.1482,
      "mean_token_accuracy": 0.2218545362353325,
      "num_tokens": 116730016.0,
      "step": 57605
    },
    {
      "entropy": 5.8815079689025875,
      "epoch": 5.034518919863672,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000275629453358561,
      "loss": 5.2401,
      "mean_token_accuracy": 0.21234291791915894,
      "num_tokens": 116741364.0,
      "step": 57610
    },
    {
      "entropy": 5.8466661930084225,
      "epoch": 5.034955868216377,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002755982923765627,
      "loss": 5.1551,
      "mean_token_accuracy": 0.22410043179988862,
      "num_tokens": 116750434.0,
      "step": 57615
    },
    {
      "entropy": 5.844673299789429,
      "epoch": 5.0353928165690816,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027556713138308887,
      "loss": 5.267,
      "mean_token_accuracy": 0.21148400753736496,
      "num_tokens": 116760020.0,
      "step": 57620
    },
    {
      "entropy": 5.908726692199707,
      "epoch": 5.035829764921786,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002755359703787372,
      "loss": 5.3011,
      "mean_token_accuracy": 0.21239988058805465,
      "num_tokens": 116770970.0,
      "step": 57625
    },
    {
      "entropy": 5.934309625625611,
      "epoch": 5.036266713274491,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002755048093641053,
      "loss": 5.2304,
      "mean_token_accuracy": 0.21589442640542983,
      "num_tokens": 116781489.0,
      "step": 57630
    },
    {
      "entropy": 5.863969326019287,
      "epoch": 5.036703661627196,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027547364833979085,
      "loss": 5.2748,
      "mean_token_accuracy": 0.21003906279802323,
      "num_tokens": 116791910.0,
      "step": 57635
    },
    {
      "entropy": 5.878126764297486,
      "epoch": 5.0371406099799,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027544248730639157,
      "loss": 5.2604,
      "mean_token_accuracy": 0.20608228594064712,
      "num_tokens": 116802580.0,
      "step": 57640
    },
    {
      "entropy": 5.859083271026611,
      "epoch": 5.037577558332605,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002754113262645052,
      "loss": 5.1561,
      "mean_token_accuracy": 0.2175288036465645,
      "num_tokens": 116811925.0,
      "step": 57645
    },
    {
      "entropy": 5.915993118286133,
      "epoch": 5.03801450668531,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027538016521472936,
      "loss": 5.1857,
      "mean_token_accuracy": 0.21169117242097854,
      "num_tokens": 116820425.0,
      "step": 57650
    },
    {
      "entropy": 5.810034227371216,
      "epoch": 5.038451455038015,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002753490041576617,
      "loss": 5.1193,
      "mean_token_accuracy": 0.22444578856229783,
      "num_tokens": 116830789.0,
      "step": 57655
    },
    {
      "entropy": 5.814937257766724,
      "epoch": 5.038888403390719,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002753178430939,
      "loss": 5.2315,
      "mean_token_accuracy": 0.2073592558503151,
      "num_tokens": 116840608.0,
      "step": 57660
    },
    {
      "entropy": 5.90686182975769,
      "epoch": 5.0393253517434236,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00027528668202404197,
      "loss": 5.2458,
      "mean_token_accuracy": 0.21525943726301194,
      "num_tokens": 116850464.0,
      "step": 57665
    },
    {
      "entropy": 5.929163408279419,
      "epoch": 5.039762300096129,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002752555209486851,
      "loss": 5.3108,
      "mean_token_accuracy": 0.21227057427167892,
      "num_tokens": 116859902.0,
      "step": 57670
    },
    {
      "entropy": 5.848665952682495,
      "epoch": 5.040199248448833,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002752243598684273,
      "loss": 5.2019,
      "mean_token_accuracy": 0.21922681331634522,
      "num_tokens": 116870346.0,
      "step": 57675
    },
    {
      "entropy": 5.827651643753052,
      "epoch": 5.040636196801538,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002751931987838662,
      "loss": 5.1978,
      "mean_token_accuracy": 0.2224365770816803,
      "num_tokens": 116881529.0,
      "step": 57680
    },
    {
      "entropy": 5.835590553283692,
      "epoch": 5.041073145154243,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002751620376955994,
      "loss": 5.1845,
      "mean_token_accuracy": 0.22405117601156235,
      "num_tokens": 116891517.0,
      "step": 57685
    },
    {
      "entropy": 5.829622888565064,
      "epoch": 5.041510093506948,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002751308766042246,
      "loss": 5.1921,
      "mean_token_accuracy": 0.21830118149518968,
      "num_tokens": 116901959.0,
      "step": 57690
    },
    {
      "entropy": 5.844024515151977,
      "epoch": 5.041947041859652,
      "grad_norm": 1.046875,
      "learning_rate": 0.00027509971551033946,
      "loss": 5.2577,
      "mean_token_accuracy": 0.209248948097229,
      "num_tokens": 116912517.0,
      "step": 57695
    },
    {
      "entropy": 5.909501743316651,
      "epoch": 5.042383990212357,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002750685544145418,
      "loss": 5.2973,
      "mean_token_accuracy": 0.20580539405345916,
      "num_tokens": 116922282.0,
      "step": 57700
    },
    {
      "entropy": 5.902874565124511,
      "epoch": 5.042820938565062,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002750373933174292,
      "loss": 5.2174,
      "mean_token_accuracy": 0.20903718918561937,
      "num_tokens": 116933280.0,
      "step": 57705
    },
    {
      "entropy": 5.9135175228118895,
      "epoch": 5.043257886917766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027500623221959946,
      "loss": 5.28,
      "mean_token_accuracy": 0.20973046869039536,
      "num_tokens": 116943951.0,
      "step": 57710
    },
    {
      "entropy": 5.881491565704346,
      "epoch": 5.043694835270471,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027497507112165013,
      "loss": 5.207,
      "mean_token_accuracy": 0.22025787830352783,
      "num_tokens": 116954707.0,
      "step": 57715
    },
    {
      "entropy": 5.887699556350708,
      "epoch": 5.044131783623175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027494391002417894,
      "loss": 5.1918,
      "mean_token_accuracy": 0.2115066796541214,
      "num_tokens": 116964643.0,
      "step": 57720
    },
    {
      "entropy": 5.821962642669678,
      "epoch": 5.044568731975881,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002749127489277836,
      "loss": 5.1875,
      "mean_token_accuracy": 0.21501814126968383,
      "num_tokens": 116975122.0,
      "step": 57725
    },
    {
      "entropy": 5.8392219066619875,
      "epoch": 5.045005680328585,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002748815878330618,
      "loss": 5.201,
      "mean_token_accuracy": 0.2214598074555397,
      "num_tokens": 116985927.0,
      "step": 57730
    },
    {
      "entropy": 5.891806697845459,
      "epoch": 5.04544262868129,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027485042674061115,
      "loss": 5.2553,
      "mean_token_accuracy": 0.20978247076272966,
      "num_tokens": 116996401.0,
      "step": 57735
    },
    {
      "entropy": 5.916150999069214,
      "epoch": 5.045879577033994,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002748192656510295,
      "loss": 5.2822,
      "mean_token_accuracy": 0.2068115770816803,
      "num_tokens": 117007201.0,
      "step": 57740
    },
    {
      "entropy": 5.9607501983642575,
      "epoch": 5.0463165253866995,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027478810456491433,
      "loss": 5.3302,
      "mean_token_accuracy": 0.2069412276148796,
      "num_tokens": 117016960.0,
      "step": 57745
    },
    {
      "entropy": 5.843442964553833,
      "epoch": 5.046753473739404,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00027475694348286337,
      "loss": 5.1554,
      "mean_token_accuracy": 0.2227377861738205,
      "num_tokens": 117026160.0,
      "step": 57750
    },
    {
      "entropy": 5.845069694519043,
      "epoch": 5.047190422092108,
      "grad_norm": 1.125,
      "learning_rate": 0.0002747257824054744,
      "loss": 5.194,
      "mean_token_accuracy": 0.22335070371627808,
      "num_tokens": 117036779.0,
      "step": 57755
    },
    {
      "entropy": 5.8943624019622805,
      "epoch": 5.047627370444814,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002746946213333451,
      "loss": 5.2737,
      "mean_token_accuracy": 0.21342905163764953,
      "num_tokens": 117047114.0,
      "step": 57760
    },
    {
      "entropy": 5.8341296195983885,
      "epoch": 5.048064318797518,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002746634602670732,
      "loss": 5.185,
      "mean_token_accuracy": 0.21636570543050765,
      "num_tokens": 117057114.0,
      "step": 57765
    },
    {
      "entropy": 5.848310279846191,
      "epoch": 5.048501267150223,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027463229920725616,
      "loss": 5.1634,
      "mean_token_accuracy": 0.21551994532346724,
      "num_tokens": 117067162.0,
      "step": 57770
    },
    {
      "entropy": 5.894377136230469,
      "epoch": 5.048938215502927,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002746011381544919,
      "loss": 5.1778,
      "mean_token_accuracy": 0.21550687551498413,
      "num_tokens": 117076231.0,
      "step": 57775
    },
    {
      "entropy": 5.831473636627197,
      "epoch": 5.049375163855633,
      "grad_norm": 1.21875,
      "learning_rate": 0.000274569977109378,
      "loss": 5.1526,
      "mean_token_accuracy": 0.22218507826328276,
      "num_tokens": 117085266.0,
      "step": 57780
    },
    {
      "entropy": 5.793817520141602,
      "epoch": 5.049812112208337,
      "grad_norm": 1.125,
      "learning_rate": 0.00027453881607251225,
      "loss": 5.1442,
      "mean_token_accuracy": 0.22050172388553618,
      "num_tokens": 117095461.0,
      "step": 57785
    },
    {
      "entropy": 5.85306191444397,
      "epoch": 5.0502490605610415,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002745076550444922,
      "loss": 5.2783,
      "mean_token_accuracy": 0.21288507282733918,
      "num_tokens": 117106539.0,
      "step": 57790
    },
    {
      "entropy": 5.850473165512085,
      "epoch": 5.050686008913746,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027447649402591545,
      "loss": 5.1842,
      "mean_token_accuracy": 0.21202616542577743,
      "num_tokens": 117117825.0,
      "step": 57795
    },
    {
      "entropy": 5.860927104949951,
      "epoch": 5.051122957266451,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00027444533301738,
      "loss": 5.1594,
      "mean_token_accuracy": 0.22158513963222504,
      "num_tokens": 117128336.0,
      "step": 57800
    },
    {
      "entropy": 5.840927410125732,
      "epoch": 5.051559905619156,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027441417201948324,
      "loss": 5.1517,
      "mean_token_accuracy": 0.2194037452340126,
      "num_tokens": 117138496.0,
      "step": 57805
    },
    {
      "entropy": 5.7886643409729,
      "epoch": 5.05199685397186,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027438301103282304,
      "loss": 5.1376,
      "mean_token_accuracy": 0.22852829992771148,
      "num_tokens": 117147398.0,
      "step": 57810
    },
    {
      "entropy": 5.779140329360962,
      "epoch": 5.052433802324566,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027435185005799696,
      "loss": 5.1235,
      "mean_token_accuracy": 0.21912945806980133,
      "num_tokens": 117157341.0,
      "step": 57815
    },
    {
      "entropy": 5.860627698898315,
      "epoch": 5.05287075067727,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002743206890956028,
      "loss": 5.2533,
      "mean_token_accuracy": 0.20807507932186126,
      "num_tokens": 117168093.0,
      "step": 57820
    },
    {
      "entropy": 5.829985284805298,
      "epoch": 5.053307699029975,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027428952814623816,
      "loss": 5.1357,
      "mean_token_accuracy": 0.22938753068447112,
      "num_tokens": 117179172.0,
      "step": 57825
    },
    {
      "entropy": 5.877688932418823,
      "epoch": 5.053744647382679,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027425836721050066,
      "loss": 5.2308,
      "mean_token_accuracy": 0.21281818747520448,
      "num_tokens": 117188801.0,
      "step": 57830
    },
    {
      "entropy": 5.861796712875366,
      "epoch": 5.054181595735384,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002742272062889881,
      "loss": 5.2106,
      "mean_token_accuracy": 0.21315712332725525,
      "num_tokens": 117200303.0,
      "step": 57835
    },
    {
      "entropy": 5.9048257827758786,
      "epoch": 5.054618544088089,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027419604538229817,
      "loss": 5.2385,
      "mean_token_accuracy": 0.21337031424045563,
      "num_tokens": 117210647.0,
      "step": 57840
    },
    {
      "entropy": 5.8280096530914305,
      "epoch": 5.055055492440793,
      "grad_norm": 1.125,
      "learning_rate": 0.0002741648844910285,
      "loss": 5.2251,
      "mean_token_accuracy": 0.21738872528076172,
      "num_tokens": 117220059.0,
      "step": 57845
    },
    {
      "entropy": 5.812351179122925,
      "epoch": 5.055492440793498,
      "grad_norm": 1.125,
      "learning_rate": 0.0002741337236157768,
      "loss": 5.2232,
      "mean_token_accuracy": 0.2149779811501503,
      "num_tokens": 117229809.0,
      "step": 57850
    },
    {
      "entropy": 5.868808031082153,
      "epoch": 5.055929389146203,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027410256275714066,
      "loss": 5.2252,
      "mean_token_accuracy": 0.21212174892425537,
      "num_tokens": 117239947.0,
      "step": 57855
    },
    {
      "entropy": 5.951613664627075,
      "epoch": 5.056366337498908,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002740714019157179,
      "loss": 5.3114,
      "mean_token_accuracy": 0.20218253433704375,
      "num_tokens": 117249785.0,
      "step": 57860
    },
    {
      "entropy": 5.843499565124512,
      "epoch": 5.056803285851612,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002740402410921061,
      "loss": 5.2159,
      "mean_token_accuracy": 0.21316303163766862,
      "num_tokens": 117260828.0,
      "step": 57865
    },
    {
      "entropy": 5.854537439346314,
      "epoch": 5.0572402342043175,
      "grad_norm": 1.046875,
      "learning_rate": 0.000274009080286903,
      "loss": 5.2277,
      "mean_token_accuracy": 0.2150226354598999,
      "num_tokens": 117271188.0,
      "step": 57870
    },
    {
      "entropy": 5.882730531692505,
      "epoch": 5.057677182557022,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002739779195007063,
      "loss": 5.2829,
      "mean_token_accuracy": 0.20889674574136735,
      "num_tokens": 117281682.0,
      "step": 57875
    },
    {
      "entropy": 5.850439977645874,
      "epoch": 5.058114130909726,
      "grad_norm": 1.125,
      "learning_rate": 0.00027394675873411363,
      "loss": 5.1187,
      "mean_token_accuracy": 0.2285830244421959,
      "num_tokens": 117292032.0,
      "step": 57880
    },
    {
      "entropy": 5.8730096340179445,
      "epoch": 5.058551079262431,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002739155979877226,
      "loss": 5.1887,
      "mean_token_accuracy": 0.21835064142942429,
      "num_tokens": 117302104.0,
      "step": 57885
    },
    {
      "entropy": 5.813538932800293,
      "epoch": 5.058988027615136,
      "grad_norm": 1.03125,
      "learning_rate": 0.000273884437262131,
      "loss": 5.1616,
      "mean_token_accuracy": 0.22085177153348923,
      "num_tokens": 117312181.0,
      "step": 57890
    },
    {
      "entropy": 5.8431884765625,
      "epoch": 5.059424975967841,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002738532765579365,
      "loss": 5.2325,
      "mean_token_accuracy": 0.221943698823452,
      "num_tokens": 117322812.0,
      "step": 57895
    },
    {
      "entropy": 5.911433410644531,
      "epoch": 5.059861924320545,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002738221158757368,
      "loss": 5.2141,
      "mean_token_accuracy": 0.211869677901268,
      "num_tokens": 117333154.0,
      "step": 57900
    },
    {
      "entropy": 5.947068071365356,
      "epoch": 5.06029887267325,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002737909552161295,
      "loss": 5.3027,
      "mean_token_accuracy": 0.20496292263269425,
      "num_tokens": 117342803.0,
      "step": 57905
    },
    {
      "entropy": 5.9245771884918215,
      "epoch": 5.060735821025955,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002737597945797123,
      "loss": 5.2858,
      "mean_token_accuracy": 0.21329572349786757,
      "num_tokens": 117352256.0,
      "step": 57910
    },
    {
      "entropy": 5.859976434707642,
      "epoch": 5.0611727693786595,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027372863396708287,
      "loss": 5.2303,
      "mean_token_accuracy": 0.21881221234798431,
      "num_tokens": 117362637.0,
      "step": 57915
    },
    {
      "entropy": 5.822133111953735,
      "epoch": 5.061609717731364,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027369747337883894,
      "loss": 5.2072,
      "mean_token_accuracy": 0.21628188490867614,
      "num_tokens": 117374163.0,
      "step": 57920
    },
    {
      "entropy": 5.85673975944519,
      "epoch": 5.062046666084069,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002736663128155781,
      "loss": 5.1992,
      "mean_token_accuracy": 0.2194063737988472,
      "num_tokens": 117383201.0,
      "step": 57925
    },
    {
      "entropy": 5.866451215744019,
      "epoch": 5.062483614436774,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002736351522778982,
      "loss": 5.2141,
      "mean_token_accuracy": 0.21692034006118774,
      "num_tokens": 117393091.0,
      "step": 57930
    },
    {
      "entropy": 5.890122556686402,
      "epoch": 5.062920562789478,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002736039917663967,
      "loss": 5.2134,
      "mean_token_accuracy": 0.21736280620098114,
      "num_tokens": 117403210.0,
      "step": 57935
    },
    {
      "entropy": 5.89981746673584,
      "epoch": 5.063357511142183,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027357283128167135,
      "loss": 5.2999,
      "mean_token_accuracy": 0.20793347358703612,
      "num_tokens": 117414190.0,
      "step": 57940
    },
    {
      "entropy": 5.870461082458496,
      "epoch": 5.063794459494888,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002735416708243199,
      "loss": 5.2116,
      "mean_token_accuracy": 0.2192936807870865,
      "num_tokens": 117423961.0,
      "step": 57945
    },
    {
      "entropy": 5.85518970489502,
      "epoch": 5.0642314078475925,
      "grad_norm": 1.0,
      "learning_rate": 0.00027351051039493994,
      "loss": 5.2433,
      "mean_token_accuracy": 0.2155206620693207,
      "num_tokens": 117434748.0,
      "step": 57950
    },
    {
      "entropy": 5.891202020645141,
      "epoch": 5.064668356200297,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002734793499941292,
      "loss": 5.2882,
      "mean_token_accuracy": 0.21299016177654267,
      "num_tokens": 117445335.0,
      "step": 57955
    },
    {
      "entropy": 5.881016778945923,
      "epoch": 5.0651053045530015,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002734481896224853,
      "loss": 5.2037,
      "mean_token_accuracy": 0.22072263658046723,
      "num_tokens": 117455964.0,
      "step": 57960
    },
    {
      "entropy": 5.904594659805298,
      "epoch": 5.065542252905707,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002734170292806059,
      "loss": 5.3154,
      "mean_token_accuracy": 0.20654846578836442,
      "num_tokens": 117466641.0,
      "step": 57965
    },
    {
      "entropy": 5.914243173599243,
      "epoch": 5.065979201258411,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002733858689690888,
      "loss": 5.2708,
      "mean_token_accuracy": 0.2093452036380768,
      "num_tokens": 117477522.0,
      "step": 57970
    },
    {
      "entropy": 5.890363025665283,
      "epoch": 5.066416149611116,
      "grad_norm": 1.125,
      "learning_rate": 0.00027335470868853154,
      "loss": 5.2331,
      "mean_token_accuracy": 0.21809052526950837,
      "num_tokens": 117486894.0,
      "step": 57975
    },
    {
      "entropy": 5.810613346099854,
      "epoch": 5.066853097963821,
      "grad_norm": 1.015625,
      "learning_rate": 0.00027332354843953176,
      "loss": 5.1133,
      "mean_token_accuracy": 0.22838521897792816,
      "num_tokens": 117497159.0,
      "step": 57980
    },
    {
      "entropy": 5.798126888275147,
      "epoch": 5.067290046316526,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027329238822268725,
      "loss": 5.0705,
      "mean_token_accuracy": 0.2321518287062645,
      "num_tokens": 117507412.0,
      "step": 57985
    },
    {
      "entropy": 5.85184268951416,
      "epoch": 5.06772699466923,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002732612280385956,
      "loss": 5.1874,
      "mean_token_accuracy": 0.2153654009103775,
      "num_tokens": 117517242.0,
      "step": 57990
    },
    {
      "entropy": 5.837071418762207,
      "epoch": 5.0681639430219345,
      "grad_norm": 1.125,
      "learning_rate": 0.0002732300678878546,
      "loss": 5.2092,
      "mean_token_accuracy": 0.21582173705101013,
      "num_tokens": 117527469.0,
      "step": 57995
    },
    {
      "entropy": 5.90761251449585,
      "epoch": 5.06860089137464,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027319890777106177,
      "loss": 5.2387,
      "mean_token_accuracy": 0.21284492462873458,
      "num_tokens": 117537918.0,
      "step": 58000
    },
    {
      "entropy": 5.842866325378418,
      "epoch": 5.069037839727344,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027316774768881483,
      "loss": 5.1739,
      "mean_token_accuracy": 0.216383758187294,
      "num_tokens": 117548339.0,
      "step": 58005
    },
    {
      "entropy": 5.786742639541626,
      "epoch": 5.069474788080049,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002731365876417114,
      "loss": 5.2088,
      "mean_token_accuracy": 0.2118971139192581,
      "num_tokens": 117558306.0,
      "step": 58010
    },
    {
      "entropy": 5.8334794521331785,
      "epoch": 5.069911736432753,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027310542763034926,
      "loss": 5.2448,
      "mean_token_accuracy": 0.2085582435131073,
      "num_tokens": 117569408.0,
      "step": 58015
    },
    {
      "entropy": 5.846066427230835,
      "epoch": 5.070348684785459,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000273074267655326,
      "loss": 5.1701,
      "mean_token_accuracy": 0.21802989691495894,
      "num_tokens": 117579744.0,
      "step": 58020
    },
    {
      "entropy": 5.911066007614136,
      "epoch": 5.070785633138163,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002730431077172393,
      "loss": 5.2058,
      "mean_token_accuracy": 0.21519635170698165,
      "num_tokens": 117589246.0,
      "step": 58025
    },
    {
      "entropy": 5.818168354034424,
      "epoch": 5.071222581490868,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002730119478166868,
      "loss": 5.1931,
      "mean_token_accuracy": 0.20913739651441574,
      "num_tokens": 117600045.0,
      "step": 58030
    },
    {
      "entropy": 5.892168617248535,
      "epoch": 5.071659529843572,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027298078795426616,
      "loss": 5.2408,
      "mean_token_accuracy": 0.20943233370780945,
      "num_tokens": 117610019.0,
      "step": 58035
    },
    {
      "entropy": 5.884922504425049,
      "epoch": 5.072096478196277,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002729496281305751,
      "loss": 5.1554,
      "mean_token_accuracy": 0.22930606752634047,
      "num_tokens": 117620642.0,
      "step": 58040
    },
    {
      "entropy": 5.875205326080322,
      "epoch": 5.072533426548982,
      "grad_norm": 1.046875,
      "learning_rate": 0.00027291846834621124,
      "loss": 5.1589,
      "mean_token_accuracy": 0.21381037533283234,
      "num_tokens": 117630154.0,
      "step": 58045
    },
    {
      "entropy": 5.86173677444458,
      "epoch": 5.072970374901686,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002728873086017723,
      "loss": 5.2305,
      "mean_token_accuracy": 0.21336643993854523,
      "num_tokens": 117640028.0,
      "step": 58050
    },
    {
      "entropy": 5.918142604827881,
      "epoch": 5.073407323254392,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00027285614889785583,
      "loss": 5.1714,
      "mean_token_accuracy": 0.217954058945179,
      "num_tokens": 117651166.0,
      "step": 58055
    },
    {
      "entropy": 5.9112616062164305,
      "epoch": 5.073844271607096,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027282498923505955,
      "loss": 5.2779,
      "mean_token_accuracy": 0.20873309373855592,
      "num_tokens": 117660589.0,
      "step": 58060
    },
    {
      "entropy": 5.8393584251403805,
      "epoch": 5.074281219959801,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002727938296139811,
      "loss": 5.1836,
      "mean_token_accuracy": 0.2130659431219101,
      "num_tokens": 117671805.0,
      "step": 58065
    },
    {
      "entropy": 5.900846385955811,
      "epoch": 5.074718168312505,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002727626700352182,
      "loss": 5.3042,
      "mean_token_accuracy": 0.20662341713905336,
      "num_tokens": 117682363.0,
      "step": 58070
    },
    {
      "entropy": 5.867915391921997,
      "epoch": 5.0751551166652105,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027273151049936843,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21229285895824432,
      "num_tokens": 117692855.0,
      "step": 58075
    },
    {
      "entropy": 5.847911167144775,
      "epoch": 5.075592065017915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002727003510070295,
      "loss": 5.2258,
      "mean_token_accuracy": 0.215971140563488,
      "num_tokens": 117702631.0,
      "step": 58080
    },
    {
      "entropy": 5.861561393737793,
      "epoch": 5.076029013370619,
      "grad_norm": 1.234375,
      "learning_rate": 0.000272669191558799,
      "loss": 5.2622,
      "mean_token_accuracy": 0.21690926253795623,
      "num_tokens": 117712682.0,
      "step": 58085
    },
    {
      "entropy": 5.822045040130615,
      "epoch": 5.076465961723324,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027263803215527466,
      "loss": 5.2255,
      "mean_token_accuracy": 0.21099784970283508,
      "num_tokens": 117722529.0,
      "step": 58090
    },
    {
      "entropy": 5.940535402297973,
      "epoch": 5.076902910076029,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002726068727970541,
      "loss": 5.3475,
      "mean_token_accuracy": 0.20030897557735444,
      "num_tokens": 117732994.0,
      "step": 58095
    },
    {
      "entropy": 5.953585052490235,
      "epoch": 5.077339858428734,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000272575713484735,
      "loss": 5.2856,
      "mean_token_accuracy": 0.21281123012304307,
      "num_tokens": 117743224.0,
      "step": 58100
    },
    {
      "entropy": 5.940330219268799,
      "epoch": 5.077776806781438,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002725445542189149,
      "loss": 5.2895,
      "mean_token_accuracy": 0.20591940283775328,
      "num_tokens": 117753775.0,
      "step": 58105
    },
    {
      "entropy": 5.923979711532593,
      "epoch": 5.0782137551341435,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002725133950001915,
      "loss": 5.2646,
      "mean_token_accuracy": 0.2133520320057869,
      "num_tokens": 117764039.0,
      "step": 58110
    },
    {
      "entropy": 5.905444097518921,
      "epoch": 5.078650703486848,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002724822358291627,
      "loss": 5.2273,
      "mean_token_accuracy": 0.21153703182935715,
      "num_tokens": 117774704.0,
      "step": 58115
    },
    {
      "entropy": 5.8797986030578615,
      "epoch": 5.0790876518395525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002724510767064257,
      "loss": 5.2822,
      "mean_token_accuracy": 0.21046726405620575,
      "num_tokens": 117784352.0,
      "step": 58120
    },
    {
      "entropy": 5.840762662887573,
      "epoch": 5.079524600192257,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002724199176325785,
      "loss": 5.2233,
      "mean_token_accuracy": 0.21289851665496826,
      "num_tokens": 117794469.0,
      "step": 58125
    },
    {
      "entropy": 5.898886060714721,
      "epoch": 5.079961548544962,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027238875860821865,
      "loss": 5.2174,
      "mean_token_accuracy": 0.21146588325500487,
      "num_tokens": 117803547.0,
      "step": 58130
    },
    {
      "entropy": 5.96879472732544,
      "epoch": 5.080398496897667,
      "grad_norm": 1.125,
      "learning_rate": 0.00027235759963394377,
      "loss": 5.3347,
      "mean_token_accuracy": 0.20647358894348145,
      "num_tokens": 117813978.0,
      "step": 58135
    },
    {
      "entropy": 5.9032612323760985,
      "epoch": 5.080835445250371,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002723264407103515,
      "loss": 5.1807,
      "mean_token_accuracy": 0.2169306233525276,
      "num_tokens": 117823609.0,
      "step": 58140
    },
    {
      "entropy": 5.830365467071533,
      "epoch": 5.081272393603076,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002722952818380395,
      "loss": 5.2202,
      "mean_token_accuracy": 0.2154015764594078,
      "num_tokens": 117834920.0,
      "step": 58145
    },
    {
      "entropy": 5.818089723587036,
      "epoch": 5.081709341955781,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002722641230176054,
      "loss": 5.1776,
      "mean_token_accuracy": 0.21941329538822174,
      "num_tokens": 117844994.0,
      "step": 58150
    },
    {
      "entropy": 5.823651742935181,
      "epoch": 5.0821462903084855,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002722329642496469,
      "loss": 5.1376,
      "mean_token_accuracy": 0.2228752166032791,
      "num_tokens": 117854300.0,
      "step": 58155
    },
    {
      "entropy": 5.840351581573486,
      "epoch": 5.08258323866119,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002722018055347615,
      "loss": 5.1342,
      "mean_token_accuracy": 0.22936027348041535,
      "num_tokens": 117864162.0,
      "step": 58160
    },
    {
      "entropy": 5.866607522964477,
      "epoch": 5.083020187013895,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027217064687354704,
      "loss": 5.1892,
      "mean_token_accuracy": 0.2202967196702957,
      "num_tokens": 117874325.0,
      "step": 58165
    },
    {
      "entropy": 5.862828016281128,
      "epoch": 5.0834571353666,
      "grad_norm": 1.15625,
      "learning_rate": 0.000272139488266601,
      "loss": 5.2084,
      "mean_token_accuracy": 0.21764974445104598,
      "num_tokens": 117884143.0,
      "step": 58170
    },
    {
      "entropy": 5.9233503341674805,
      "epoch": 5.083894083719304,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002721083297145211,
      "loss": 5.2726,
      "mean_token_accuracy": 0.2066458910703659,
      "num_tokens": 117895007.0,
      "step": 58175
    },
    {
      "entropy": 5.846984004974365,
      "epoch": 5.084331032072009,
      "grad_norm": 1.015625,
      "learning_rate": 0.00027207717121790494,
      "loss": 5.206,
      "mean_token_accuracy": 0.21764918714761733,
      "num_tokens": 117905902.0,
      "step": 58180
    },
    {
      "entropy": 5.9099733352661135,
      "epoch": 5.084767980424714,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002720460127773502,
      "loss": 5.2472,
      "mean_token_accuracy": 0.20722784996032714,
      "num_tokens": 117916714.0,
      "step": 58185
    },
    {
      "entropy": 5.894626045227051,
      "epoch": 5.085204928777419,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002720148543934544,
      "loss": 5.2451,
      "mean_token_accuracy": 0.20837949961423874,
      "num_tokens": 117926305.0,
      "step": 58190
    },
    {
      "entropy": 5.815827894210815,
      "epoch": 5.085641877130123,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002719836960668153,
      "loss": 5.1328,
      "mean_token_accuracy": 0.22459251135587693,
      "num_tokens": 117935350.0,
      "step": 58195
    },
    {
      "entropy": 5.842511510848999,
      "epoch": 5.0860788254828275,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027195253779803045,
      "loss": 5.2565,
      "mean_token_accuracy": 0.219296233355999,
      "num_tokens": 117944965.0,
      "step": 58200
    },
    {
      "entropy": 5.8063701629638675,
      "epoch": 5.086515773835533,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027192137958769765,
      "loss": 5.1419,
      "mean_token_accuracy": 0.22167333513498305,
      "num_tokens": 117954199.0,
      "step": 58205
    },
    {
      "entropy": 5.882345056533813,
      "epoch": 5.086952722188237,
      "grad_norm": 1.125,
      "learning_rate": 0.0002718902214364143,
      "loss": 5.2434,
      "mean_token_accuracy": 0.21776071190834045,
      "num_tokens": 117964579.0,
      "step": 58210
    },
    {
      "entropy": 5.887089824676513,
      "epoch": 5.087389670540942,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027185906334477815,
      "loss": 5.1688,
      "mean_token_accuracy": 0.21379120647907257,
      "num_tokens": 117973664.0,
      "step": 58215
    },
    {
      "entropy": 5.8671492576599125,
      "epoch": 5.087826618893647,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002718279053133868,
      "loss": 5.2111,
      "mean_token_accuracy": 0.21439493745565413,
      "num_tokens": 117983253.0,
      "step": 58220
    },
    {
      "entropy": 5.805482244491577,
      "epoch": 5.088263567246352,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002717967473428379,
      "loss": 5.1428,
      "mean_token_accuracy": 0.22936399579048156,
      "num_tokens": 117992418.0,
      "step": 58225
    },
    {
      "entropy": 5.874980354309082,
      "epoch": 5.088700515599056,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002717655894337291,
      "loss": 5.2554,
      "mean_token_accuracy": 0.22110625356435776,
      "num_tokens": 118001882.0,
      "step": 58230
    },
    {
      "entropy": 5.901996326446533,
      "epoch": 5.089137463951761,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027173443158665794,
      "loss": 5.2736,
      "mean_token_accuracy": 0.2182972714304924,
      "num_tokens": 118013322.0,
      "step": 58235
    },
    {
      "entropy": 5.899669456481933,
      "epoch": 5.089574412304466,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002717032738022221,
      "loss": 5.1888,
      "mean_token_accuracy": 0.21341657191514968,
      "num_tokens": 118023452.0,
      "step": 58240
    },
    {
      "entropy": 5.870774555206299,
      "epoch": 5.09001136065717,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027167211608101924,
      "loss": 5.2031,
      "mean_token_accuracy": 0.2250499829649925,
      "num_tokens": 118033401.0,
      "step": 58245
    },
    {
      "entropy": 5.776428747177124,
      "epoch": 5.090448309009875,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002716409584236469,
      "loss": 5.1432,
      "mean_token_accuracy": 0.2233434110879898,
      "num_tokens": 118044001.0,
      "step": 58250
    },
    {
      "entropy": 5.831004476547241,
      "epoch": 5.090885257362579,
      "grad_norm": 1.015625,
      "learning_rate": 0.00027160980083070277,
      "loss": 5.1763,
      "mean_token_accuracy": 0.2141701579093933,
      "num_tokens": 118053730.0,
      "step": 58255
    },
    {
      "entropy": 5.838170862197876,
      "epoch": 5.091322205715285,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002715786433027844,
      "loss": 5.1529,
      "mean_token_accuracy": 0.21522029042243956,
      "num_tokens": 118063527.0,
      "step": 58260
    },
    {
      "entropy": 5.736724042892456,
      "epoch": 5.091759154067989,
      "grad_norm": 1.125,
      "learning_rate": 0.0002715474858404894,
      "loss": 5.1194,
      "mean_token_accuracy": 0.2265746310353279,
      "num_tokens": 118074411.0,
      "step": 58265
    },
    {
      "entropy": 5.779613447189331,
      "epoch": 5.092196102420694,
      "grad_norm": 0.984375,
      "learning_rate": 0.00027151632844441546,
      "loss": 5.1854,
      "mean_token_accuracy": 0.22463594526052474,
      "num_tokens": 118084570.0,
      "step": 58270
    },
    {
      "entropy": 5.893780899047852,
      "epoch": 5.092633050773399,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027148517111516026,
      "loss": 5.1508,
      "mean_token_accuracy": 0.2187260016798973,
      "num_tokens": 118094695.0,
      "step": 58275
    },
    {
      "entropy": 5.893911647796631,
      "epoch": 5.0930699991261035,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002714540138533212,
      "loss": 5.279,
      "mean_token_accuracy": 0.21112193614244462,
      "num_tokens": 118105400.0,
      "step": 58280
    },
    {
      "entropy": 5.81316728591919,
      "epoch": 5.093506947478808,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027142285665949606,
      "loss": 5.1916,
      "mean_token_accuracy": 0.21741385459899903,
      "num_tokens": 118116507.0,
      "step": 58285
    },
    {
      "entropy": 5.860620546340942,
      "epoch": 5.093943895831512,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002713916995342824,
      "loss": 5.1838,
      "mean_token_accuracy": 0.22165905833244323,
      "num_tokens": 118126485.0,
      "step": 58290
    },
    {
      "entropy": 5.893037939071656,
      "epoch": 5.094380844184218,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002713605424782778,
      "loss": 5.2406,
      "mean_token_accuracy": 0.2153599202632904,
      "num_tokens": 118136730.0,
      "step": 58295
    },
    {
      "entropy": 5.850124168395996,
      "epoch": 5.094817792536922,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027132938549208005,
      "loss": 5.1268,
      "mean_token_accuracy": 0.22202152758836746,
      "num_tokens": 118147379.0,
      "step": 58300
    },
    {
      "entropy": 5.834736633300781,
      "epoch": 5.095254740889627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027129822857628643,
      "loss": 5.2634,
      "mean_token_accuracy": 0.20880017429590225,
      "num_tokens": 118157815.0,
      "step": 58305
    },
    {
      "entropy": 5.787024116516113,
      "epoch": 5.095691689242331,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027126707173149477,
      "loss": 5.1128,
      "mean_token_accuracy": 0.22041702419519424,
      "num_tokens": 118168150.0,
      "step": 58310
    },
    {
      "entropy": 5.817406558990479,
      "epoch": 5.0961286375950365,
      "grad_norm": 1.0,
      "learning_rate": 0.0002712359149583027,
      "loss": 5.1455,
      "mean_token_accuracy": 0.2253435045480728,
      "num_tokens": 118179299.0,
      "step": 58315
    },
    {
      "entropy": 5.87364616394043,
      "epoch": 5.096565585947741,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002712047582573077,
      "loss": 5.2357,
      "mean_token_accuracy": 0.2080863207578659,
      "num_tokens": 118189455.0,
      "step": 58320
    },
    {
      "entropy": 5.773622941970825,
      "epoch": 5.0970025343004455,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002711736016291074,
      "loss": 5.1364,
      "mean_token_accuracy": 0.2246266260743141,
      "num_tokens": 118198916.0,
      "step": 58325
    },
    {
      "entropy": 5.928527164459228,
      "epoch": 5.097439482653151,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027114244507429945,
      "loss": 5.3039,
      "mean_token_accuracy": 0.20333373546600342,
      "num_tokens": 118209122.0,
      "step": 58330
    },
    {
      "entropy": 5.851162052154541,
      "epoch": 5.097876431005855,
      "grad_norm": 1.125,
      "learning_rate": 0.00027111128859348146,
      "loss": 5.1964,
      "mean_token_accuracy": 0.21787362098693847,
      "num_tokens": 118219197.0,
      "step": 58335
    },
    {
      "entropy": 5.917755508422852,
      "epoch": 5.09831337935856,
      "grad_norm": 1.09375,
      "learning_rate": 0.000271080132187251,
      "loss": 5.2645,
      "mean_token_accuracy": 0.2167102262377739,
      "num_tokens": 118229979.0,
      "step": 58340
    },
    {
      "entropy": 5.8591207504272464,
      "epoch": 5.098750327711264,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002710489758562056,
      "loss": 5.2236,
      "mean_token_accuracy": 0.21736396700143815,
      "num_tokens": 118238611.0,
      "step": 58345
    },
    {
      "entropy": 5.829409551620484,
      "epoch": 5.09918727606397,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027101781960094297,
      "loss": 5.1935,
      "mean_token_accuracy": 0.2245839238166809,
      "num_tokens": 118247979.0,
      "step": 58350
    },
    {
      "entropy": 5.9013739109039305,
      "epoch": 5.099624224416674,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002709866634220606,
      "loss": 5.2291,
      "mean_token_accuracy": 0.219404174387455,
      "num_tokens": 118258807.0,
      "step": 58355
    },
    {
      "entropy": 5.834315633773803,
      "epoch": 5.1000611727693785,
      "grad_norm": 1.125,
      "learning_rate": 0.0002709555073201561,
      "loss": 5.218,
      "mean_token_accuracy": 0.2191125750541687,
      "num_tokens": 118269538.0,
      "step": 58360
    },
    {
      "entropy": 5.8624104976654055,
      "epoch": 5.100498121122083,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00027092435129582714,
      "loss": 5.1404,
      "mean_token_accuracy": 0.2189393609762192,
      "num_tokens": 118280011.0,
      "step": 58365
    },
    {
      "entropy": 5.81089038848877,
      "epoch": 5.100935069474788,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002708931953496712,
      "loss": 5.1963,
      "mean_token_accuracy": 0.22192478328943252,
      "num_tokens": 118290477.0,
      "step": 58370
    },
    {
      "entropy": 5.765202617645263,
      "epoch": 5.101372017827493,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027086203948228606,
      "loss": 5.0863,
      "mean_token_accuracy": 0.2243970513343811,
      "num_tokens": 118300219.0,
      "step": 58375
    },
    {
      "entropy": 5.934050226211548,
      "epoch": 5.101808966180197,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002708308836942691,
      "loss": 5.3184,
      "mean_token_accuracy": 0.2052357405424118,
      "num_tokens": 118310711.0,
      "step": 58380
    },
    {
      "entropy": 5.890861129760742,
      "epoch": 5.102245914532903,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002707997279862179,
      "loss": 5.1969,
      "mean_token_accuracy": 0.20821432024240494,
      "num_tokens": 118320793.0,
      "step": 58385
    },
    {
      "entropy": 5.828047895431519,
      "epoch": 5.102682862885607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002707685723587302,
      "loss": 5.1999,
      "mean_token_accuracy": 0.21507903933525085,
      "num_tokens": 118330651.0,
      "step": 58390
    },
    {
      "entropy": 5.878360891342163,
      "epoch": 5.103119811238312,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002707374168124035,
      "loss": 5.1894,
      "mean_token_accuracy": 0.21870106607675552,
      "num_tokens": 118340097.0,
      "step": 58395
    },
    {
      "entropy": 5.816684913635254,
      "epoch": 5.103556759591016,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027070626134783536,
      "loss": 5.2343,
      "mean_token_accuracy": 0.21230284571647645,
      "num_tokens": 118350508.0,
      "step": 58400
    },
    {
      "entropy": 5.83207950592041,
      "epoch": 5.103993707943721,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002706751059656233,
      "loss": 5.194,
      "mean_token_accuracy": 0.2107810854911804,
      "num_tokens": 118360634.0,
      "step": 58405
    },
    {
      "entropy": 5.899861860275268,
      "epoch": 5.104430656296426,
      "grad_norm": 1.0,
      "learning_rate": 0.0002706439506663651,
      "loss": 5.2516,
      "mean_token_accuracy": 0.20898328721523285,
      "num_tokens": 118370836.0,
      "step": 58410
    },
    {
      "entropy": 5.983456993103028,
      "epoch": 5.10486760464913,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027061279545065813,
      "loss": 5.3465,
      "mean_token_accuracy": 0.19916608035564423,
      "num_tokens": 118381887.0,
      "step": 58415
    },
    {
      "entropy": 5.823404121398926,
      "epoch": 5.105304553001835,
      "grad_norm": 1.125,
      "learning_rate": 0.00027058164031910014,
      "loss": 5.1321,
      "mean_token_accuracy": 0.22268588244915008,
      "num_tokens": 118391609.0,
      "step": 58420
    },
    {
      "entropy": 5.880451250076294,
      "epoch": 5.10574150135454,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027055048527228854,
      "loss": 5.2578,
      "mean_token_accuracy": 0.21181749552488327,
      "num_tokens": 118401424.0,
      "step": 58425
    },
    {
      "entropy": 5.858693981170655,
      "epoch": 5.106178449707245,
      "grad_norm": 1.25,
      "learning_rate": 0.0002705193303108209,
      "loss": 5.2091,
      "mean_token_accuracy": 0.2205878585577011,
      "num_tokens": 118411195.0,
      "step": 58430
    },
    {
      "entropy": 5.840194797515869,
      "epoch": 5.106615398059949,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002704881754352949,
      "loss": 5.2142,
      "mean_token_accuracy": 0.21645492762327195,
      "num_tokens": 118421064.0,
      "step": 58435
    },
    {
      "entropy": 5.78735842704773,
      "epoch": 5.107052346412654,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027045702064630806,
      "loss": 5.146,
      "mean_token_accuracy": 0.22766676545143127,
      "num_tokens": 118431126.0,
      "step": 58440
    },
    {
      "entropy": 5.83124008178711,
      "epoch": 5.107489294765359,
      "grad_norm": 1.21875,
      "learning_rate": 0.000270425865944458,
      "loss": 5.1396,
      "mean_token_accuracy": 0.22337112128734588,
      "num_tokens": 118441007.0,
      "step": 58445
    },
    {
      "entropy": 5.863634824752808,
      "epoch": 5.107926243118063,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027039471133034215,
      "loss": 5.1707,
      "mean_token_accuracy": 0.2197271317243576,
      "num_tokens": 118449807.0,
      "step": 58450
    },
    {
      "entropy": 5.827907276153565,
      "epoch": 5.108363191470768,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002703635568045582,
      "loss": 5.219,
      "mean_token_accuracy": 0.22035948932170868,
      "num_tokens": 118459803.0,
      "step": 58455
    },
    {
      "entropy": 5.801023006439209,
      "epoch": 5.108800139823473,
      "grad_norm": 1.125,
      "learning_rate": 0.0002703324023677036,
      "loss": 5.136,
      "mean_token_accuracy": 0.2220948100090027,
      "num_tokens": 118469630.0,
      "step": 58460
    },
    {
      "entropy": 5.850785589218139,
      "epoch": 5.109237088176178,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027030124802037596,
      "loss": 5.2381,
      "mean_token_accuracy": 0.2150669887661934,
      "num_tokens": 118480549.0,
      "step": 58465
    },
    {
      "entropy": 5.897498273849488,
      "epoch": 5.109674036528882,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002702700937631729,
      "loss": 5.2906,
      "mean_token_accuracy": 0.2096800372004509,
      "num_tokens": 118491192.0,
      "step": 58470
    },
    {
      "entropy": 5.916086292266845,
      "epoch": 5.110110984881587,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002702389395966919,
      "loss": 5.2993,
      "mean_token_accuracy": 0.2070713922381401,
      "num_tokens": 118501208.0,
      "step": 58475
    },
    {
      "entropy": 5.965551900863647,
      "epoch": 5.110547933234292,
      "grad_norm": 1.125,
      "learning_rate": 0.0002702077855215306,
      "loss": 5.3148,
      "mean_token_accuracy": 0.20570918023586274,
      "num_tokens": 118511736.0,
      "step": 58480
    },
    {
      "entropy": 5.907826471328735,
      "epoch": 5.1109848815869965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002701766315382864,
      "loss": 5.2563,
      "mean_token_accuracy": 0.21690727025270462,
      "num_tokens": 118522160.0,
      "step": 58485
    },
    {
      "entropy": 5.8804925918579105,
      "epoch": 5.111421829939701,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000270145477647557,
      "loss": 5.1989,
      "mean_token_accuracy": 0.21372421234846115,
      "num_tokens": 118531825.0,
      "step": 58490
    },
    {
      "entropy": 5.8607666015625,
      "epoch": 5.111858778292405,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002701143238499398,
      "loss": 5.2488,
      "mean_token_accuracy": 0.212752665579319,
      "num_tokens": 118541084.0,
      "step": 58495
    },
    {
      "entropy": 5.861398983001709,
      "epoch": 5.112295726645111,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002700831701460325,
      "loss": 5.202,
      "mean_token_accuracy": 0.21571240872144698,
      "num_tokens": 118550524.0,
      "step": 58500
    },
    {
      "entropy": 5.875755071640015,
      "epoch": 5.112732674997815,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002700520165364325,
      "loss": 5.2288,
      "mean_token_accuracy": 0.21352440863847733,
      "num_tokens": 118560250.0,
      "step": 58505
    },
    {
      "entropy": 5.87928318977356,
      "epoch": 5.11316962335052,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027002086302173756,
      "loss": 5.2414,
      "mean_token_accuracy": 0.21100766211748123,
      "num_tokens": 118570574.0,
      "step": 58510
    },
    {
      "entropy": 5.813499593734742,
      "epoch": 5.113606571703225,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026998970960254496,
      "loss": 5.2201,
      "mean_token_accuracy": 0.21239992082118989,
      "num_tokens": 118582408.0,
      "step": 58515
    },
    {
      "entropy": 5.770941066741943,
      "epoch": 5.1140435200559295,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002699585562794524,
      "loss": 5.1646,
      "mean_token_accuracy": 0.22097823023796082,
      "num_tokens": 118592277.0,
      "step": 58520
    },
    {
      "entropy": 5.82296142578125,
      "epoch": 5.114480468408634,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002699274030530573,
      "loss": 5.1561,
      "mean_token_accuracy": 0.22250492125749588,
      "num_tokens": 118601943.0,
      "step": 58525
    },
    {
      "entropy": 5.884820365905762,
      "epoch": 5.1149174167613385,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026989624992395734,
      "loss": 5.1825,
      "mean_token_accuracy": 0.21345241963863373,
      "num_tokens": 118611450.0,
      "step": 58530
    },
    {
      "entropy": 5.872974252700805,
      "epoch": 5.115354365114044,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026986509689274995,
      "loss": 5.2907,
      "mean_token_accuracy": 0.20435562878847122,
      "num_tokens": 118621660.0,
      "step": 58535
    },
    {
      "entropy": 5.869198322296143,
      "epoch": 5.115791313466748,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002698339439600327,
      "loss": 5.254,
      "mean_token_accuracy": 0.21285576075315477,
      "num_tokens": 118631455.0,
      "step": 58540
    },
    {
      "entropy": 5.821994400024414,
      "epoch": 5.116228261819453,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002698027911264031,
      "loss": 5.1408,
      "mean_token_accuracy": 0.216756209731102,
      "num_tokens": 118641208.0,
      "step": 58545
    },
    {
      "entropy": 5.862035655975342,
      "epoch": 5.116665210172157,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002697716383924587,
      "loss": 5.1686,
      "mean_token_accuracy": 0.2158595696091652,
      "num_tokens": 118651802.0,
      "step": 58550
    },
    {
      "entropy": 5.838028049468994,
      "epoch": 5.117102158524863,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026974048575879704,
      "loss": 5.1659,
      "mean_token_accuracy": 0.21846652925014495,
      "num_tokens": 118661623.0,
      "step": 58555
    },
    {
      "entropy": 5.82995924949646,
      "epoch": 5.117539106877567,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002697093332260156,
      "loss": 5.2339,
      "mean_token_accuracy": 0.20868349969387054,
      "num_tokens": 118671714.0,
      "step": 58560
    },
    {
      "entropy": 5.846002721786499,
      "epoch": 5.1179760552302715,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002696781807947119,
      "loss": 5.1431,
      "mean_token_accuracy": 0.2180471181869507,
      "num_tokens": 118681990.0,
      "step": 58565
    },
    {
      "entropy": 5.839434480667114,
      "epoch": 5.118413003582977,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002696470284654835,
      "loss": 5.1492,
      "mean_token_accuracy": 0.2133224844932556,
      "num_tokens": 118691465.0,
      "step": 58570
    },
    {
      "entropy": 5.753275299072266,
      "epoch": 5.118849951935681,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002696158762389279,
      "loss": 5.1374,
      "mean_token_accuracy": 0.21762757003307343,
      "num_tokens": 118700975.0,
      "step": 58575
    },
    {
      "entropy": 5.864735698699951,
      "epoch": 5.119286900288386,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002695847241156427,
      "loss": 5.2998,
      "mean_token_accuracy": 0.20922466814517976,
      "num_tokens": 118711249.0,
      "step": 58580
    },
    {
      "entropy": 5.943218851089478,
      "epoch": 5.11972384864109,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026955357209622517,
      "loss": 5.2511,
      "mean_token_accuracy": 0.2105296805500984,
      "num_tokens": 118721182.0,
      "step": 58585
    },
    {
      "entropy": 5.851770877838135,
      "epoch": 5.120160796993796,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026952242018127303,
      "loss": 5.145,
      "mean_token_accuracy": 0.22069993019104003,
      "num_tokens": 118730346.0,
      "step": 58590
    },
    {
      "entropy": 5.836177921295166,
      "epoch": 5.1205977453465,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002694912683713838,
      "loss": 5.2682,
      "mean_token_accuracy": 0.21285433769226075,
      "num_tokens": 118740622.0,
      "step": 58595
    },
    {
      "entropy": 5.8458915710449215,
      "epoch": 5.121034693699205,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026946011666715484,
      "loss": 5.2133,
      "mean_token_accuracy": 0.21062613576650618,
      "num_tokens": 118752900.0,
      "step": 58600
    },
    {
      "entropy": 5.942108011245727,
      "epoch": 5.121471642051909,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026942896506918385,
      "loss": 5.2423,
      "mean_token_accuracy": 0.21201675087213517,
      "num_tokens": 118762430.0,
      "step": 58605
    },
    {
      "entropy": 5.855388212203979,
      "epoch": 5.121908590404614,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026939781357806814,
      "loss": 5.2346,
      "mean_token_accuracy": 0.21420695185661315,
      "num_tokens": 118772670.0,
      "step": 58610
    },
    {
      "entropy": 5.860794878005981,
      "epoch": 5.122345538757319,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026936666219440537,
      "loss": 5.1899,
      "mean_token_accuracy": 0.21101368963718414,
      "num_tokens": 118783785.0,
      "step": 58615
    },
    {
      "entropy": 5.856268501281738,
      "epoch": 5.122782487110023,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026933551091879295,
      "loss": 5.1849,
      "mean_token_accuracy": 0.2191990688443184,
      "num_tokens": 118793283.0,
      "step": 58620
    },
    {
      "entropy": 5.909734773635864,
      "epoch": 5.123219435462729,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002693043597518283,
      "loss": 5.2423,
      "mean_token_accuracy": 0.21079994440078736,
      "num_tokens": 118803249.0,
      "step": 58625
    },
    {
      "entropy": 5.84733738899231,
      "epoch": 5.123656383815433,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002692732086941092,
      "loss": 5.2015,
      "mean_token_accuracy": 0.21832391023635864,
      "num_tokens": 118814194.0,
      "step": 58630
    },
    {
      "entropy": 5.753260231018066,
      "epoch": 5.124093332168138,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002692420577462328,
      "loss": 5.096,
      "mean_token_accuracy": 0.22769162952899932,
      "num_tokens": 118824203.0,
      "step": 58635
    },
    {
      "entropy": 5.8398760795593265,
      "epoch": 5.124530280520842,
      "grad_norm": 1.03125,
      "learning_rate": 0.00026921090690879685,
      "loss": 5.2385,
      "mean_token_accuracy": 0.21429131478071212,
      "num_tokens": 118834071.0,
      "step": 58640
    },
    {
      "entropy": 5.8740404605865475,
      "epoch": 5.1249672288735475,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026917975618239863,
      "loss": 5.1988,
      "mean_token_accuracy": 0.2199619010090828,
      "num_tokens": 118844256.0,
      "step": 58645
    },
    {
      "entropy": 5.877405595779419,
      "epoch": 5.125404177226252,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002691486055676359,
      "loss": 5.2043,
      "mean_token_accuracy": 0.216410356760025,
      "num_tokens": 118854025.0,
      "step": 58650
    },
    {
      "entropy": 5.915388488769532,
      "epoch": 5.125841125578956,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026911745506510586,
      "loss": 5.2826,
      "mean_token_accuracy": 0.21251437216997146,
      "num_tokens": 118864313.0,
      "step": 58655
    },
    {
      "entropy": 5.838672208786011,
      "epoch": 5.126278073931661,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026908630467540614,
      "loss": 5.1451,
      "mean_token_accuracy": 0.21902606785297393,
      "num_tokens": 118873726.0,
      "step": 58660
    },
    {
      "entropy": 5.821819067001343,
      "epoch": 5.126715022284366,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002690551543991342,
      "loss": 5.1599,
      "mean_token_accuracy": 0.22249167710542678,
      "num_tokens": 118883995.0,
      "step": 58665
    },
    {
      "entropy": 5.808149147033691,
      "epoch": 5.127151970637071,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002690240042368875,
      "loss": 5.2185,
      "mean_token_accuracy": 0.21991421580314635,
      "num_tokens": 118893717.0,
      "step": 58670
    },
    {
      "entropy": 5.841803503036499,
      "epoch": 5.127588918989775,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002689928541892635,
      "loss": 5.2136,
      "mean_token_accuracy": 0.21481496393680571,
      "num_tokens": 118902888.0,
      "step": 58675
    },
    {
      "entropy": 5.985000801086426,
      "epoch": 5.12802586734248,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026896170425685977,
      "loss": 5.3164,
      "mean_token_accuracy": 0.2043130725622177,
      "num_tokens": 118913201.0,
      "step": 58680
    },
    {
      "entropy": 5.859775400161743,
      "epoch": 5.128462815695185,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002689305544402737,
      "loss": 5.1893,
      "mean_token_accuracy": 0.21534129679203035,
      "num_tokens": 118922256.0,
      "step": 58685
    },
    {
      "entropy": 5.902875089645386,
      "epoch": 5.1288997640478895,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002688994047401028,
      "loss": 5.2173,
      "mean_token_accuracy": 0.20947923511266708,
      "num_tokens": 118932083.0,
      "step": 58690
    },
    {
      "entropy": 5.797876691818237,
      "epoch": 5.129336712400594,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002688682551569444,
      "loss": 5.1945,
      "mean_token_accuracy": 0.2172691747546196,
      "num_tokens": 118941499.0,
      "step": 58695
    },
    {
      "entropy": 5.841184043884278,
      "epoch": 5.129773660753299,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026883710569139624,
      "loss": 5.1505,
      "mean_token_accuracy": 0.22691810578107835,
      "num_tokens": 118951375.0,
      "step": 58700
    },
    {
      "entropy": 5.8297278881073,
      "epoch": 5.130210609106004,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026880595634405554,
      "loss": 5.2439,
      "mean_token_accuracy": 0.21488988697528838,
      "num_tokens": 118962616.0,
      "step": 58705
    },
    {
      "entropy": 5.877161645889283,
      "epoch": 5.130647557458708,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002687748071155198,
      "loss": 5.2985,
      "mean_token_accuracy": 0.20231076776981355,
      "num_tokens": 118973209.0,
      "step": 58710
    },
    {
      "entropy": 5.89851336479187,
      "epoch": 5.131084505811413,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026874365800638657,
      "loss": 5.2681,
      "mean_token_accuracy": 0.2094871237874031,
      "num_tokens": 118983808.0,
      "step": 58715
    },
    {
      "entropy": 5.936266326904297,
      "epoch": 5.131521454164118,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026871250901725326,
      "loss": 5.2393,
      "mean_token_accuracy": 0.2048746958374977,
      "num_tokens": 118993557.0,
      "step": 58720
    },
    {
      "entropy": 5.879603910446167,
      "epoch": 5.131958402516823,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002686813601487174,
      "loss": 5.2391,
      "mean_token_accuracy": 0.2148989737033844,
      "num_tokens": 119004078.0,
      "step": 58725
    },
    {
      "entropy": 5.8783581256866455,
      "epoch": 5.132395350869527,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026865021140137617,
      "loss": 5.2632,
      "mean_token_accuracy": 0.21347761005163193,
      "num_tokens": 119014822.0,
      "step": 58730
    },
    {
      "entropy": 5.866932344436646,
      "epoch": 5.1328322992222315,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002686190627758274,
      "loss": 5.2592,
      "mean_token_accuracy": 0.20917898714542388,
      "num_tokens": 119025593.0,
      "step": 58735
    },
    {
      "entropy": 5.847176885604858,
      "epoch": 5.133269247574937,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002685879142726682,
      "loss": 5.239,
      "mean_token_accuracy": 0.2159674271941185,
      "num_tokens": 119035656.0,
      "step": 58740
    },
    {
      "entropy": 5.958669805526734,
      "epoch": 5.133706195927641,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002685567658924962,
      "loss": 5.3224,
      "mean_token_accuracy": 0.20827839225530625,
      "num_tokens": 119045897.0,
      "step": 58745
    },
    {
      "entropy": 5.9316894054412845,
      "epoch": 5.134143144280346,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002685256176359089,
      "loss": 5.255,
      "mean_token_accuracy": 0.20970599204301835,
      "num_tokens": 119057387.0,
      "step": 58750
    },
    {
      "entropy": 5.839749813079834,
      "epoch": 5.134580092633051,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002684944695035036,
      "loss": 5.197,
      "mean_token_accuracy": 0.21978786289691926,
      "num_tokens": 119067506.0,
      "step": 58755
    },
    {
      "entropy": 5.822545957565308,
      "epoch": 5.135017040985756,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026846332149587775,
      "loss": 5.189,
      "mean_token_accuracy": 0.21329203695058824,
      "num_tokens": 119078336.0,
      "step": 58760
    },
    {
      "entropy": 5.87808837890625,
      "epoch": 5.13545398933846,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002684321736136289,
      "loss": 5.2312,
      "mean_token_accuracy": 0.21470196545124054,
      "num_tokens": 119088062.0,
      "step": 58765
    },
    {
      "entropy": 5.900221300125122,
      "epoch": 5.135890937691165,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026840102585735427,
      "loss": 5.2675,
      "mean_token_accuracy": 0.21369886994361878,
      "num_tokens": 119097668.0,
      "step": 58770
    },
    {
      "entropy": 5.8718091487884525,
      "epoch": 5.13632788604387,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026836987822765154,
      "loss": 5.1976,
      "mean_token_accuracy": 0.2122106671333313,
      "num_tokens": 119109133.0,
      "step": 58775
    },
    {
      "entropy": 5.82197642326355,
      "epoch": 5.136764834396574,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000268338730725118,
      "loss": 5.154,
      "mean_token_accuracy": 0.21848034411668776,
      "num_tokens": 119119149.0,
      "step": 58780
    },
    {
      "entropy": 5.903691387176513,
      "epoch": 5.137201782749279,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026830758335035106,
      "loss": 5.2215,
      "mean_token_accuracy": 0.21151596754789354,
      "num_tokens": 119129004.0,
      "step": 58785
    },
    {
      "entropy": 5.836928224563598,
      "epoch": 5.137638731101983,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026827643610394816,
      "loss": 5.1837,
      "mean_token_accuracy": 0.2169938176870346,
      "num_tokens": 119139894.0,
      "step": 58790
    },
    {
      "entropy": 5.807089853286743,
      "epoch": 5.138075679454689,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002682452889865067,
      "loss": 5.1535,
      "mean_token_accuracy": 0.22396302223205566,
      "num_tokens": 119149224.0,
      "step": 58795
    },
    {
      "entropy": 5.857544565200806,
      "epoch": 5.138512627807393,
      "grad_norm": 1.125,
      "learning_rate": 0.00026821414199862425,
      "loss": 5.1882,
      "mean_token_accuracy": 0.2136734679341316,
      "num_tokens": 119159040.0,
      "step": 58800
    },
    {
      "entropy": 5.937079954147339,
      "epoch": 5.138949576160098,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026818299514089807,
      "loss": 5.2498,
      "mean_token_accuracy": 0.21158597767353057,
      "num_tokens": 119168287.0,
      "step": 58805
    },
    {
      "entropy": 5.9113270282745365,
      "epoch": 5.139386524512803,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002681518484139255,
      "loss": 5.2687,
      "mean_token_accuracy": 0.2142147272825241,
      "num_tokens": 119179077.0,
      "step": 58810
    },
    {
      "entropy": 5.922895240783691,
      "epoch": 5.1398234728655074,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002681207018183042,
      "loss": 5.3229,
      "mean_token_accuracy": 0.20568869411945342,
      "num_tokens": 119189580.0,
      "step": 58815
    },
    {
      "entropy": 5.931754016876221,
      "epoch": 5.140260421218212,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002680895553546314,
      "loss": 5.256,
      "mean_token_accuracy": 0.2117808058857918,
      "num_tokens": 119199532.0,
      "step": 58820
    },
    {
      "entropy": 5.863183784484863,
      "epoch": 5.140697369570916,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002680584090235046,
      "loss": 5.2099,
      "mean_token_accuracy": 0.21792035847902297,
      "num_tokens": 119211358.0,
      "step": 58825
    },
    {
      "entropy": 5.866349411010742,
      "epoch": 5.141134317923622,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002680272628255211,
      "loss": 5.1412,
      "mean_token_accuracy": 0.21305144131183623,
      "num_tokens": 119221841.0,
      "step": 58830
    },
    {
      "entropy": 5.847177648544312,
      "epoch": 5.141571266276326,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002679961167612784,
      "loss": 5.2142,
      "mean_token_accuracy": 0.21379073709249496,
      "num_tokens": 119230619.0,
      "step": 58835
    },
    {
      "entropy": 5.89723801612854,
      "epoch": 5.142008214629031,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002679649708313738,
      "loss": 5.1782,
      "mean_token_accuracy": 0.21397195905447006,
      "num_tokens": 119240516.0,
      "step": 58840
    },
    {
      "entropy": 5.910413074493408,
      "epoch": 5.142445162981735,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002679338250364048,
      "loss": 5.2724,
      "mean_token_accuracy": 0.2083301231265068,
      "num_tokens": 119250438.0,
      "step": 58845
    },
    {
      "entropy": 5.923629570007324,
      "epoch": 5.1428821113344405,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002679026793769687,
      "loss": 5.2532,
      "mean_token_accuracy": 0.21602341681718826,
      "num_tokens": 119260528.0,
      "step": 58850
    },
    {
      "entropy": 5.78213415145874,
      "epoch": 5.143319059687145,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002678715338536629,
      "loss": 5.155,
      "mean_token_accuracy": 0.2213892176747322,
      "num_tokens": 119270018.0,
      "step": 58855
    },
    {
      "entropy": 5.813081932067871,
      "epoch": 5.1437560080398494,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002678403884670849,
      "loss": 5.214,
      "mean_token_accuracy": 0.21984408497810365,
      "num_tokens": 119280901.0,
      "step": 58860
    },
    {
      "entropy": 5.9042740821838375,
      "epoch": 5.144192956392555,
      "grad_norm": 1.25,
      "learning_rate": 0.000267809243217832,
      "loss": 5.2813,
      "mean_token_accuracy": 0.20587062388658522,
      "num_tokens": 119290554.0,
      "step": 58865
    },
    {
      "entropy": 5.8498553276062015,
      "epoch": 5.144629904745259,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002677780981065014,
      "loss": 5.1957,
      "mean_token_accuracy": 0.21775908023118973,
      "num_tokens": 119300488.0,
      "step": 58870
    },
    {
      "entropy": 5.85945839881897,
      "epoch": 5.145066853097964,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026774695313369086,
      "loss": 5.254,
      "mean_token_accuracy": 0.21083336919546128,
      "num_tokens": 119311237.0,
      "step": 58875
    },
    {
      "entropy": 5.871705436706543,
      "epoch": 5.145503801450668,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002677158082999975,
      "loss": 5.2354,
      "mean_token_accuracy": 0.20274345725774764,
      "num_tokens": 119322503.0,
      "step": 58880
    },
    {
      "entropy": 5.906850862503052,
      "epoch": 5.145940749803374,
      "grad_norm": 1.125,
      "learning_rate": 0.0002676846636060188,
      "loss": 5.2445,
      "mean_token_accuracy": 0.21321820616722106,
      "num_tokens": 119332705.0,
      "step": 58885
    },
    {
      "entropy": 5.94787859916687,
      "epoch": 5.146377698156078,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000267653519052352,
      "loss": 5.2829,
      "mean_token_accuracy": 0.21331517547369003,
      "num_tokens": 119342602.0,
      "step": 58890
    },
    {
      "entropy": 5.890360021591187,
      "epoch": 5.1468146465087825,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002676223746395946,
      "loss": 5.2306,
      "mean_token_accuracy": 0.21527994871139527,
      "num_tokens": 119351918.0,
      "step": 58895
    },
    {
      "entropy": 5.692033195495606,
      "epoch": 5.147251594861487,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002675912303683439,
      "loss": 5.0641,
      "mean_token_accuracy": 0.23541464060544967,
      "num_tokens": 119360947.0,
      "step": 58900
    },
    {
      "entropy": 5.852478265762329,
      "epoch": 5.147688543214192,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002675600862391972,
      "loss": 5.2001,
      "mean_token_accuracy": 0.20881661027669907,
      "num_tokens": 119370176.0,
      "step": 58905
    },
    {
      "entropy": 5.913375949859619,
      "epoch": 5.148125491566897,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026752894225275217,
      "loss": 5.2458,
      "mean_token_accuracy": 0.21227305382490158,
      "num_tokens": 119380177.0,
      "step": 58910
    },
    {
      "entropy": 5.8933909893035885,
      "epoch": 5.148562439919601,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026749779840960574,
      "loss": 5.1967,
      "mean_token_accuracy": 0.2168618306517601,
      "num_tokens": 119389898.0,
      "step": 58915
    },
    {
      "entropy": 5.820146799087524,
      "epoch": 5.148999388272307,
      "grad_norm": 1.125,
      "learning_rate": 0.0002674666547103556,
      "loss": 5.2003,
      "mean_token_accuracy": 0.2198764219880104,
      "num_tokens": 119400221.0,
      "step": 58920
    },
    {
      "entropy": 5.8074195861816404,
      "epoch": 5.149436336625011,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026743551115559886,
      "loss": 5.192,
      "mean_token_accuracy": 0.21899962276220322,
      "num_tokens": 119410122.0,
      "step": 58925
    },
    {
      "entropy": 5.87906584739685,
      "epoch": 5.149873284977716,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000267404367745933,
      "loss": 5.2442,
      "mean_token_accuracy": 0.217616768181324,
      "num_tokens": 119419892.0,
      "step": 58930
    },
    {
      "entropy": 5.916546392440796,
      "epoch": 5.15031023333042,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026737322448195543,
      "loss": 5.197,
      "mean_token_accuracy": 0.21798429638147354,
      "num_tokens": 119429463.0,
      "step": 58935
    },
    {
      "entropy": 5.900277805328369,
      "epoch": 5.150747181683125,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00026734208136426337,
      "loss": 5.2469,
      "mean_token_accuracy": 0.213668192923069,
      "num_tokens": 119440588.0,
      "step": 58940
    },
    {
      "entropy": 5.91395320892334,
      "epoch": 5.15118413003583,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002673109383934542,
      "loss": 5.2707,
      "mean_token_accuracy": 0.2041370913386345,
      "num_tokens": 119450257.0,
      "step": 58945
    },
    {
      "entropy": 5.923146295547485,
      "epoch": 5.151621078388534,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026727979557012523,
      "loss": 5.2046,
      "mean_token_accuracy": 0.21243277788162232,
      "num_tokens": 119459257.0,
      "step": 58950
    },
    {
      "entropy": 5.854554080963135,
      "epoch": 5.152058026741239,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026724865289487384,
      "loss": 5.2264,
      "mean_token_accuracy": 0.22035559564828872,
      "num_tokens": 119469754.0,
      "step": 58955
    },
    {
      "entropy": 5.774784564971924,
      "epoch": 5.152494975093944,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026721751036829736,
      "loss": 5.1503,
      "mean_token_accuracy": 0.2265697181224823,
      "num_tokens": 119480154.0,
      "step": 58960
    },
    {
      "entropy": 5.771295785903931,
      "epoch": 5.152931923446649,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026718636799099313,
      "loss": 5.2053,
      "mean_token_accuracy": 0.22311335951089858,
      "num_tokens": 119490932.0,
      "step": 58965
    },
    {
      "entropy": 5.836139488220215,
      "epoch": 5.153368871799353,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002671552257635584,
      "loss": 5.1681,
      "mean_token_accuracy": 0.22894561141729355,
      "num_tokens": 119501675.0,
      "step": 58970
    },
    {
      "entropy": 5.958424377441406,
      "epoch": 5.1538058201520585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002671240836865906,
      "loss": 5.2606,
      "mean_token_accuracy": 0.21454501152038574,
      "num_tokens": 119512273.0,
      "step": 58975
    },
    {
      "entropy": 5.8231902599334715,
      "epoch": 5.154242768504763,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026709294176068693,
      "loss": 5.1689,
      "mean_token_accuracy": 0.21580301225185394,
      "num_tokens": 119521909.0,
      "step": 58980
    },
    {
      "entropy": 5.811031007766724,
      "epoch": 5.154679716857467,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026706179998644484,
      "loss": 5.2096,
      "mean_token_accuracy": 0.22020357102155685,
      "num_tokens": 119533032.0,
      "step": 58985
    },
    {
      "entropy": 5.947742462158203,
      "epoch": 5.155116665210172,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002670306583644616,
      "loss": 5.3176,
      "mean_token_accuracy": 0.19874042868614197,
      "num_tokens": 119544000.0,
      "step": 58990
    },
    {
      "entropy": 5.872696304321289,
      "epoch": 5.155553613562877,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026699951689533445,
      "loss": 5.1942,
      "mean_token_accuracy": 0.21967371851205825,
      "num_tokens": 119554540.0,
      "step": 58995
    },
    {
      "entropy": 5.879238414764404,
      "epoch": 5.155990561915582,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002669683755796608,
      "loss": 5.2479,
      "mean_token_accuracy": 0.2080775111913681,
      "num_tokens": 119564847.0,
      "step": 59000
    },
    {
      "entropy": 5.889899826049804,
      "epoch": 5.156427510268286,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026693723441803796,
      "loss": 5.2323,
      "mean_token_accuracy": 0.20534978210926055,
      "num_tokens": 119575029.0,
      "step": 59005
    },
    {
      "entropy": 5.910179185867309,
      "epoch": 5.156864458620991,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026690609341106314,
      "loss": 5.2813,
      "mean_token_accuracy": 0.20571995824575423,
      "num_tokens": 119586394.0,
      "step": 59010
    },
    {
      "entropy": 5.944313955307007,
      "epoch": 5.157301406973696,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002668749525593337,
      "loss": 5.2727,
      "mean_token_accuracy": 0.20757764130830764,
      "num_tokens": 119596890.0,
      "step": 59015
    },
    {
      "entropy": 5.827403831481933,
      "epoch": 5.1577383553264005,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026684381186344686,
      "loss": 5.1604,
      "mean_token_accuracy": 0.21952022463083268,
      "num_tokens": 119606745.0,
      "step": 59020
    },
    {
      "entropy": 5.859563159942627,
      "epoch": 5.158175303679105,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026681267132400003,
      "loss": 5.2605,
      "mean_token_accuracy": 0.21358362287282945,
      "num_tokens": 119615757.0,
      "step": 59025
    },
    {
      "entropy": 5.912339067459106,
      "epoch": 5.15861225203181,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002667815309415905,
      "loss": 5.258,
      "mean_token_accuracy": 0.2057099461555481,
      "num_tokens": 119626433.0,
      "step": 59030
    },
    {
      "entropy": 5.903936195373535,
      "epoch": 5.159049200384515,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00026675039071681546,
      "loss": 5.2081,
      "mean_token_accuracy": 0.21682633012533187,
      "num_tokens": 119636554.0,
      "step": 59035
    },
    {
      "entropy": 5.828812789916992,
      "epoch": 5.159486148737219,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002667192506502723,
      "loss": 5.1877,
      "mean_token_accuracy": 0.22071186751127242,
      "num_tokens": 119645636.0,
      "step": 59040
    },
    {
      "entropy": 5.851348638534546,
      "epoch": 5.159923097089924,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026668811074255814,
      "loss": 5.2825,
      "mean_token_accuracy": 0.20983781218528746,
      "num_tokens": 119656828.0,
      "step": 59045
    },
    {
      "entropy": 5.847495460510254,
      "epoch": 5.160360045442629,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026665697099427044,
      "loss": 5.1375,
      "mean_token_accuracy": 0.2176382288336754,
      "num_tokens": 119666438.0,
      "step": 59050
    },
    {
      "entropy": 5.825823211669922,
      "epoch": 5.1607969937953335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002666258314060065,
      "loss": 5.2203,
      "mean_token_accuracy": 0.21942706704139708,
      "num_tokens": 119677080.0,
      "step": 59055
    },
    {
      "entropy": 5.842266273498535,
      "epoch": 5.161233942148038,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002665946919783634,
      "loss": 5.1799,
      "mean_token_accuracy": 0.21928858608007432,
      "num_tokens": 119687401.0,
      "step": 59060
    },
    {
      "entropy": 5.866951131820679,
      "epoch": 5.1616708905007425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002665635527119385,
      "loss": 5.2071,
      "mean_token_accuracy": 0.21299756616353988,
      "num_tokens": 119697314.0,
      "step": 59065
    },
    {
      "entropy": 5.944546508789062,
      "epoch": 5.162107838853448,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002665324136073291,
      "loss": 5.2693,
      "mean_token_accuracy": 0.2113379120826721,
      "num_tokens": 119707609.0,
      "step": 59070
    },
    {
      "entropy": 5.791769313812256,
      "epoch": 5.162544787206152,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002665012746651325,
      "loss": 5.1359,
      "mean_token_accuracy": 0.2210620313882828,
      "num_tokens": 119717304.0,
      "step": 59075
    },
    {
      "entropy": 5.838575220108032,
      "epoch": 5.162981735558857,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002664701358859459,
      "loss": 5.1996,
      "mean_token_accuracy": 0.21377744227647782,
      "num_tokens": 119727708.0,
      "step": 59080
    },
    {
      "entropy": 5.8408232688903805,
      "epoch": 5.163418683911561,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026643899727036647,
      "loss": 5.2158,
      "mean_token_accuracy": 0.21636127978563308,
      "num_tokens": 119739248.0,
      "step": 59085
    },
    {
      "entropy": 5.8804656028747555,
      "epoch": 5.163855632264267,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002664078588189916,
      "loss": 5.1927,
      "mean_token_accuracy": 0.2120744287967682,
      "num_tokens": 119748157.0,
      "step": 59090
    },
    {
      "entropy": 5.806981420516967,
      "epoch": 5.164292580616971,
      "grad_norm": 1.25,
      "learning_rate": 0.00026637672053241853,
      "loss": 5.1562,
      "mean_token_accuracy": 0.21835724711418153,
      "num_tokens": 119757125.0,
      "step": 59095
    },
    {
      "entropy": 5.883220291137695,
      "epoch": 5.1647295289696755,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002663455824112444,
      "loss": 5.2598,
      "mean_token_accuracy": 0.20813976526260375,
      "num_tokens": 119767276.0,
      "step": 59100
    },
    {
      "entropy": 5.907597970962525,
      "epoch": 5.165166477322381,
      "grad_norm": 1.0,
      "learning_rate": 0.0002663144444560666,
      "loss": 5.276,
      "mean_token_accuracy": 0.20927760004997253,
      "num_tokens": 119777718.0,
      "step": 59105
    },
    {
      "entropy": 5.930367183685303,
      "epoch": 5.165603425675085,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002662833066674823,
      "loss": 5.2473,
      "mean_token_accuracy": 0.20660935640335082,
      "num_tokens": 119787342.0,
      "step": 59110
    },
    {
      "entropy": 5.8371976852417,
      "epoch": 5.16604037402779,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026625216904608873,
      "loss": 5.1275,
      "mean_token_accuracy": 0.22354452610015868,
      "num_tokens": 119796585.0,
      "step": 59115
    },
    {
      "entropy": 5.875998020172119,
      "epoch": 5.166477322380494,
      "grad_norm": 1.125,
      "learning_rate": 0.00026622103159248316,
      "loss": 5.2219,
      "mean_token_accuracy": 0.21157582402229308,
      "num_tokens": 119806486.0,
      "step": 59120
    },
    {
      "entropy": 5.86383843421936,
      "epoch": 5.1669142707332,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002661898943072628,
      "loss": 5.2474,
      "mean_token_accuracy": 0.21484647691249847,
      "num_tokens": 119817075.0,
      "step": 59125
    },
    {
      "entropy": 5.817796277999878,
      "epoch": 5.167351219085904,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026615875719102495,
      "loss": 5.1622,
      "mean_token_accuracy": 0.21690565794706346,
      "num_tokens": 119826397.0,
      "step": 59130
    },
    {
      "entropy": 5.909416246414184,
      "epoch": 5.167788167438609,
      "grad_norm": 1.125,
      "learning_rate": 0.00026612762024436667,
      "loss": 5.2419,
      "mean_token_accuracy": 0.2135734364390373,
      "num_tokens": 119837106.0,
      "step": 59135
    },
    {
      "entropy": 5.928016948699951,
      "epoch": 5.168225115791313,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002660964834678853,
      "loss": 5.1947,
      "mean_token_accuracy": 0.21689920574426652,
      "num_tokens": 119847942.0,
      "step": 59140
    },
    {
      "entropy": 5.908745765686035,
      "epoch": 5.168662064144018,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026606534686217804,
      "loss": 5.2702,
      "mean_token_accuracy": 0.207001394033432,
      "num_tokens": 119857731.0,
      "step": 59145
    },
    {
      "entropy": 5.801943063735962,
      "epoch": 5.169099012496723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002660342104278422,
      "loss": 5.1521,
      "mean_token_accuracy": 0.223113115131855,
      "num_tokens": 119867418.0,
      "step": 59150
    },
    {
      "entropy": 5.701635694503784,
      "epoch": 5.169535960849427,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002660030741654748,
      "loss": 5.0363,
      "mean_token_accuracy": 0.2411941647529602,
      "num_tokens": 119876166.0,
      "step": 59155
    },
    {
      "entropy": 5.8147563457489015,
      "epoch": 5.169972909202133,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002659719380756732,
      "loss": 5.1461,
      "mean_token_accuracy": 0.22072310149669647,
      "num_tokens": 119885519.0,
      "step": 59160
    },
    {
      "entropy": 5.844015073776245,
      "epoch": 5.170409857554837,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026594080215903445,
      "loss": 5.2478,
      "mean_token_accuracy": 0.21307166963815688,
      "num_tokens": 119895947.0,
      "step": 59165
    },
    {
      "entropy": 5.790354871749878,
      "epoch": 5.170846805907542,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002659096664161559,
      "loss": 5.1507,
      "mean_token_accuracy": 0.22279942780733109,
      "num_tokens": 119905521.0,
      "step": 59170
    },
    {
      "entropy": 5.855911159515381,
      "epoch": 5.171283754260246,
      "grad_norm": 1.125,
      "learning_rate": 0.00026587853084763483,
      "loss": 5.2052,
      "mean_token_accuracy": 0.2132493034005165,
      "num_tokens": 119916282.0,
      "step": 59175
    },
    {
      "entropy": 5.887851858139038,
      "epoch": 5.1717207026129515,
      "grad_norm": 1.125,
      "learning_rate": 0.00026584739545406825,
      "loss": 5.2532,
      "mean_token_accuracy": 0.20758175998926162,
      "num_tokens": 119925757.0,
      "step": 59180
    },
    {
      "entropy": 5.90660982131958,
      "epoch": 5.172157650965656,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002658162602360534,
      "loss": 5.2473,
      "mean_token_accuracy": 0.21508698612451554,
      "num_tokens": 119935974.0,
      "step": 59185
    },
    {
      "entropy": 5.9010203838348385,
      "epoch": 5.17259459931836,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002657851251941875,
      "loss": 5.2047,
      "mean_token_accuracy": 0.21469840705394744,
      "num_tokens": 119947215.0,
      "step": 59190
    },
    {
      "entropy": 5.907526063919067,
      "epoch": 5.173031547671065,
      "grad_norm": 1.25,
      "learning_rate": 0.0002657539903290677,
      "loss": 5.211,
      "mean_token_accuracy": 0.2178231507539749,
      "num_tokens": 119957276.0,
      "step": 59195
    },
    {
      "entropy": 5.878737878799439,
      "epoch": 5.17346849602377,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002657228556412913,
      "loss": 5.2462,
      "mean_token_accuracy": 0.20780488699674607,
      "num_tokens": 119966636.0,
      "step": 59200
    },
    {
      "entropy": 5.841320657730103,
      "epoch": 5.173905444376475,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002656917211314553,
      "loss": 5.1977,
      "mean_token_accuracy": 0.21214244812726973,
      "num_tokens": 119975860.0,
      "step": 59205
    },
    {
      "entropy": 5.873690986633301,
      "epoch": 5.174342392729179,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000265660586800157,
      "loss": 5.2161,
      "mean_token_accuracy": 0.2094064712524414,
      "num_tokens": 119985167.0,
      "step": 59210
    },
    {
      "entropy": 5.9225709438323975,
      "epoch": 5.1747793410818845,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002656294526479936,
      "loss": 5.2298,
      "mean_token_accuracy": 0.2161889433860779,
      "num_tokens": 119995384.0,
      "step": 59215
    },
    {
      "entropy": 5.780989313125611,
      "epoch": 5.175216289434589,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002655983186755621,
      "loss": 5.0992,
      "mean_token_accuracy": 0.2289832800626755,
      "num_tokens": 120005081.0,
      "step": 59220
    },
    {
      "entropy": 5.8928121566772464,
      "epoch": 5.1756532377872935,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002655671848834598,
      "loss": 5.1971,
      "mean_token_accuracy": 0.21354618072509765,
      "num_tokens": 120015407.0,
      "step": 59225
    },
    {
      "entropy": 5.829603147506714,
      "epoch": 5.176090186139998,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026553605127228386,
      "loss": 5.1938,
      "mean_token_accuracy": 0.21446175277233123,
      "num_tokens": 120025894.0,
      "step": 59230
    },
    {
      "entropy": 5.839618682861328,
      "epoch": 5.176527134492703,
      "grad_norm": 1.375,
      "learning_rate": 0.0002655049178426315,
      "loss": 5.139,
      "mean_token_accuracy": 0.2147107571363449,
      "num_tokens": 120035267.0,
      "step": 59235
    },
    {
      "entropy": 5.876053428649902,
      "epoch": 5.176964082845408,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002654737845950997,
      "loss": 5.2662,
      "mean_token_accuracy": 0.21188654601573945,
      "num_tokens": 120045489.0,
      "step": 59240
    },
    {
      "entropy": 5.956955194473267,
      "epoch": 5.177401031198112,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002654426515302857,
      "loss": 5.2622,
      "mean_token_accuracy": 0.21149833798408507,
      "num_tokens": 120054416.0,
      "step": 59245
    },
    {
      "entropy": 5.877848148345947,
      "epoch": 5.177837979550817,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00026541151864878664,
      "loss": 5.2213,
      "mean_token_accuracy": 0.21284887045621873,
      "num_tokens": 120064656.0,
      "step": 59250
    },
    {
      "entropy": 5.837874031066894,
      "epoch": 5.178274927903522,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002653803859511997,
      "loss": 5.1863,
      "mean_token_accuracy": 0.21651487201452255,
      "num_tokens": 120075254.0,
      "step": 59255
    },
    {
      "entropy": 5.876355123519898,
      "epoch": 5.1787118762562265,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000265349253438122,
      "loss": 5.2291,
      "mean_token_accuracy": 0.216213358938694,
      "num_tokens": 120085162.0,
      "step": 59260
    },
    {
      "entropy": 5.853385162353516,
      "epoch": 5.179148824608931,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026531812111015067,
      "loss": 5.2015,
      "mean_token_accuracy": 0.21141074299812318,
      "num_tokens": 120094390.0,
      "step": 59265
    },
    {
      "entropy": 5.749490118026733,
      "epoch": 5.179585772961636,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002652869889678829,
      "loss": 5.1151,
      "mean_token_accuracy": 0.2256644368171692,
      "num_tokens": 120104869.0,
      "step": 59270
    },
    {
      "entropy": 5.846583700180053,
      "epoch": 5.180022721314341,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002652558570119157,
      "loss": 5.1621,
      "mean_token_accuracy": 0.2228758454322815,
      "num_tokens": 120115254.0,
      "step": 59275
    },
    {
      "entropy": 5.799851608276367,
      "epoch": 5.180459669667045,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002652247252428463,
      "loss": 5.1879,
      "mean_token_accuracy": 0.21921376436948775,
      "num_tokens": 120125740.0,
      "step": 59280
    },
    {
      "entropy": 5.818235921859741,
      "epoch": 5.18089661801975,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026519359366127183,
      "loss": 5.2288,
      "mean_token_accuracy": 0.21737515032291413,
      "num_tokens": 120135811.0,
      "step": 59285
    },
    {
      "entropy": 5.780989122390747,
      "epoch": 5.181333566372455,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002651624622677893,
      "loss": 5.0844,
      "mean_token_accuracy": 0.23559398502111434,
      "num_tokens": 120146145.0,
      "step": 59290
    },
    {
      "entropy": 5.8541570663452145,
      "epoch": 5.18177051472516,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002651313310629959,
      "loss": 5.1818,
      "mean_token_accuracy": 0.21854296326637268,
      "num_tokens": 120156330.0,
      "step": 59295
    },
    {
      "entropy": 5.855679941177368,
      "epoch": 5.182207463077864,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002651002000474888,
      "loss": 5.2457,
      "mean_token_accuracy": 0.21393772959709167,
      "num_tokens": 120165809.0,
      "step": 59300
    },
    {
      "entropy": 5.864467000961303,
      "epoch": 5.1826444114305685,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026506906922186494,
      "loss": 5.2425,
      "mean_token_accuracy": 0.21312965601682662,
      "num_tokens": 120176118.0,
      "step": 59305
    },
    {
      "entropy": 5.813090419769287,
      "epoch": 5.183081359783274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002650379385867217,
      "loss": 5.1646,
      "mean_token_accuracy": 0.21714054495096208,
      "num_tokens": 120185589.0,
      "step": 59310
    },
    {
      "entropy": 5.8204833984375,
      "epoch": 5.183518308135978,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002650068081426559,
      "loss": 5.1432,
      "mean_token_accuracy": 0.2231563851237297,
      "num_tokens": 120195849.0,
      "step": 59315
    },
    {
      "entropy": 5.82243161201477,
      "epoch": 5.183955256488683,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026497567789026484,
      "loss": 5.2079,
      "mean_token_accuracy": 0.21389820873737336,
      "num_tokens": 120206305.0,
      "step": 59320
    },
    {
      "entropy": 5.869015741348266,
      "epoch": 5.184392204841387,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002649445478301454,
      "loss": 5.1698,
      "mean_token_accuracy": 0.22272848337888718,
      "num_tokens": 120216142.0,
      "step": 59325
    },
    {
      "entropy": 5.944114637374878,
      "epoch": 5.184829153194093,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002649134179628949,
      "loss": 5.2666,
      "mean_token_accuracy": 0.2068385198712349,
      "num_tokens": 120226561.0,
      "step": 59330
    },
    {
      "entropy": 5.815418815612793,
      "epoch": 5.185266101546797,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002648822882891104,
      "loss": 5.1344,
      "mean_token_accuracy": 0.2262500301003456,
      "num_tokens": 120236614.0,
      "step": 59335
    },
    {
      "entropy": 5.899397134780884,
      "epoch": 5.185703049899502,
      "grad_norm": 1.125,
      "learning_rate": 0.00026485115880938874,
      "loss": 5.3298,
      "mean_token_accuracy": 0.20210694819688796,
      "num_tokens": 120246519.0,
      "step": 59340
    },
    {
      "entropy": 5.857924461364746,
      "epoch": 5.186139998252207,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026482002952432724,
      "loss": 5.1966,
      "mean_token_accuracy": 0.223619681596756,
      "num_tokens": 120256249.0,
      "step": 59345
    },
    {
      "entropy": 5.898159074783325,
      "epoch": 5.186576946604911,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026478890043452296,
      "loss": 5.2054,
      "mean_token_accuracy": 0.22234148383140565,
      "num_tokens": 120267175.0,
      "step": 59350
    },
    {
      "entropy": 5.918980646133423,
      "epoch": 5.187013894957616,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026475777154057287,
      "loss": 5.2058,
      "mean_token_accuracy": 0.21659871488809584,
      "num_tokens": 120277312.0,
      "step": 59355
    },
    {
      "entropy": 5.883548450469971,
      "epoch": 5.18745084331032,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00026472664284307414,
      "loss": 5.2508,
      "mean_token_accuracy": 0.210866117477417,
      "num_tokens": 120287163.0,
      "step": 59360
    },
    {
      "entropy": 5.868800497055053,
      "epoch": 5.187887791663026,
      "grad_norm": 1.046875,
      "learning_rate": 0.00026469551434262373,
      "loss": 5.2608,
      "mean_token_accuracy": 0.21082033812999726,
      "num_tokens": 120297695.0,
      "step": 59365
    },
    {
      "entropy": 5.857542991638184,
      "epoch": 5.18832474001573,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002646643860398187,
      "loss": 5.2267,
      "mean_token_accuracy": 0.21306970864534377,
      "num_tokens": 120309104.0,
      "step": 59370
    },
    {
      "entropy": 5.83421630859375,
      "epoch": 5.188761688368435,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0002646332579352563,
      "loss": 5.181,
      "mean_token_accuracy": 0.226076577603817,
      "num_tokens": 120320123.0,
      "step": 59375
    },
    {
      "entropy": 5.81157865524292,
      "epoch": 5.189198636721139,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002646021300295334,
      "loss": 5.1644,
      "mean_token_accuracy": 0.22309521734714508,
      "num_tokens": 120330948.0,
      "step": 59380
    },
    {
      "entropy": 5.846994686126709,
      "epoch": 5.1896355850738445,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000264571002323247,
      "loss": 5.2165,
      "mean_token_accuracy": 0.21483008563518524,
      "num_tokens": 120340668.0,
      "step": 59385
    },
    {
      "entropy": 5.86342191696167,
      "epoch": 5.190072533426549,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002645398748169944,
      "loss": 5.1627,
      "mean_token_accuracy": 0.21828193515539168,
      "num_tokens": 120351089.0,
      "step": 59390
    },
    {
      "entropy": 5.812619256973266,
      "epoch": 5.190509481779253,
      "grad_norm": 1.796875,
      "learning_rate": 0.00026450874751137235,
      "loss": 5.1973,
      "mean_token_accuracy": 0.2190316155552864,
      "num_tokens": 120362014.0,
      "step": 59395
    },
    {
      "entropy": 5.969311046600342,
      "epoch": 5.190946430131959,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000264477620406978,
      "loss": 5.3652,
      "mean_token_accuracy": 0.2062755510210991,
      "num_tokens": 120372264.0,
      "step": 59400
    },
    {
      "entropy": 5.912919807434082,
      "epoch": 5.191383378484663,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002644464935044085,
      "loss": 5.2089,
      "mean_token_accuracy": 0.21328670382499695,
      "num_tokens": 120382387.0,
      "step": 59405
    },
    {
      "entropy": 5.9017982006073,
      "epoch": 5.191820326837368,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002644153668042608,
      "loss": 5.2439,
      "mean_token_accuracy": 0.21621603071689605,
      "num_tokens": 120392951.0,
      "step": 59410
    },
    {
      "entropy": 5.858920431137085,
      "epoch": 5.192257275190072,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002643842403071319,
      "loss": 5.2258,
      "mean_token_accuracy": 0.20987599343061447,
      "num_tokens": 120401856.0,
      "step": 59415
    },
    {
      "entropy": 5.890224838256836,
      "epoch": 5.1926942235427775,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026435311401361874,
      "loss": 5.2674,
      "mean_token_accuracy": 0.2085893452167511,
      "num_tokens": 120411090.0,
      "step": 59420
    },
    {
      "entropy": 5.868609142303467,
      "epoch": 5.193131171895482,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002643219879243185,
      "loss": 5.1295,
      "mean_token_accuracy": 0.21915851831436156,
      "num_tokens": 120421298.0,
      "step": 59425
    },
    {
      "entropy": 5.8265749454498295,
      "epoch": 5.1935681202481865,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026429086203982806,
      "loss": 5.2001,
      "mean_token_accuracy": 0.21621939837932586,
      "num_tokens": 120432113.0,
      "step": 59430
    },
    {
      "entropy": 5.908547210693359,
      "epoch": 5.194005068600891,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026425973636074466,
      "loss": 5.2994,
      "mean_token_accuracy": 0.20968687087297438,
      "num_tokens": 120442353.0,
      "step": 59435
    },
    {
      "entropy": 5.871396017074585,
      "epoch": 5.194442016953596,
      "grad_norm": 1.09375,
      "learning_rate": 0.000264228610887665,
      "loss": 5.1913,
      "mean_token_accuracy": 0.2151532545685768,
      "num_tokens": 120452161.0,
      "step": 59440
    },
    {
      "entropy": 5.836061477661133,
      "epoch": 5.194878965306301,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026419748562118625,
      "loss": 5.129,
      "mean_token_accuracy": 0.22038486152887343,
      "num_tokens": 120462025.0,
      "step": 59445
    },
    {
      "entropy": 5.843501234054566,
      "epoch": 5.195315913659005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026416636056190534,
      "loss": 5.2392,
      "mean_token_accuracy": 0.20980894118547438,
      "num_tokens": 120472251.0,
      "step": 59450
    },
    {
      "entropy": 5.875318193435669,
      "epoch": 5.195752862011711,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002641352357104194,
      "loss": 5.2049,
      "mean_token_accuracy": 0.21145355999469756,
      "num_tokens": 120482070.0,
      "step": 59455
    },
    {
      "entropy": 5.899602365493775,
      "epoch": 5.196189810364415,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00026410411106732534,
      "loss": 5.1898,
      "mean_token_accuracy": 0.22190581560134887,
      "num_tokens": 120492399.0,
      "step": 59460
    },
    {
      "entropy": 5.8420877933502195,
      "epoch": 5.1966267587171195,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002640729866332201,
      "loss": 5.1997,
      "mean_token_accuracy": 0.2162471055984497,
      "num_tokens": 120503557.0,
      "step": 59465
    },
    {
      "entropy": 5.906124925613403,
      "epoch": 5.197063707069824,
      "grad_norm": 1.03125,
      "learning_rate": 0.00026404186240870073,
      "loss": 5.3214,
      "mean_token_accuracy": 0.20574678033590316,
      "num_tokens": 120513247.0,
      "step": 59470
    },
    {
      "entropy": 5.974802398681641,
      "epoch": 5.197500655422529,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002640107383943641,
      "loss": 5.3491,
      "mean_token_accuracy": 0.20699489563703538,
      "num_tokens": 120524479.0,
      "step": 59475
    },
    {
      "entropy": 5.837312459945679,
      "epoch": 5.197937603775234,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026397961459080735,
      "loss": 5.1979,
      "mean_token_accuracy": 0.21709142476320267,
      "num_tokens": 120533868.0,
      "step": 59480
    },
    {
      "entropy": 5.92918438911438,
      "epoch": 5.198374552127938,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026394849099862735,
      "loss": 5.2004,
      "mean_token_accuracy": 0.217412531375885,
      "num_tokens": 120543714.0,
      "step": 59485
    },
    {
      "entropy": 5.848719644546509,
      "epoch": 5.198811500480643,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026391736761842107,
      "loss": 5.1988,
      "mean_token_accuracy": 0.21706725060939788,
      "num_tokens": 120553033.0,
      "step": 59490
    },
    {
      "entropy": 5.861321973800659,
      "epoch": 5.199248448833348,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002638862444507855,
      "loss": 5.1491,
      "mean_token_accuracy": 0.2181893065571785,
      "num_tokens": 120562551.0,
      "step": 59495
    },
    {
      "entropy": 5.887213134765625,
      "epoch": 5.199685397186053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002638551214963175,
      "loss": 5.3135,
      "mean_token_accuracy": 0.20836560428142548,
      "num_tokens": 120572956.0,
      "step": 59500
    },
    {
      "entropy": 5.898671674728393,
      "epoch": 5.200122345538757,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026382399875561423,
      "loss": 5.2039,
      "mean_token_accuracy": 0.21435188949108125,
      "num_tokens": 120582917.0,
      "step": 59505
    },
    {
      "entropy": 5.897777032852173,
      "epoch": 5.200559293891462,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026379287622927246,
      "loss": 5.199,
      "mean_token_accuracy": 0.2156781792640686,
      "num_tokens": 120593252.0,
      "step": 59510
    },
    {
      "entropy": 5.919966554641723,
      "epoch": 5.200996242244167,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002637617539178892,
      "loss": 5.2864,
      "mean_token_accuracy": 0.21626256853342057,
      "num_tokens": 120603511.0,
      "step": 59515
    },
    {
      "entropy": 5.829275417327881,
      "epoch": 5.201433190596871,
      "grad_norm": 1.03125,
      "learning_rate": 0.00026373063182206136,
      "loss": 5.15,
      "mean_token_accuracy": 0.22844166606664656,
      "num_tokens": 120614504.0,
      "step": 59520
    },
    {
      "entropy": 5.792450904846191,
      "epoch": 5.201870138949576,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026369950994238595,
      "loss": 5.0744,
      "mean_token_accuracy": 0.22639795243740082,
      "num_tokens": 120623660.0,
      "step": 59525
    },
    {
      "entropy": 5.858392333984375,
      "epoch": 5.202307087302281,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026366838827945985,
      "loss": 5.2323,
      "mean_token_accuracy": 0.2134037896990776,
      "num_tokens": 120634787.0,
      "step": 59530
    },
    {
      "entropy": 5.904288625717163,
      "epoch": 5.202744035654986,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026363726683387997,
      "loss": 5.2033,
      "mean_token_accuracy": 0.21934842467308044,
      "num_tokens": 120644456.0,
      "step": 59535
    },
    {
      "entropy": 5.940155792236328,
      "epoch": 5.20318098400769,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002636061456062433,
      "loss": 5.2333,
      "mean_token_accuracy": 0.21226486265659333,
      "num_tokens": 120653713.0,
      "step": 59540
    },
    {
      "entropy": 5.827235460281372,
      "epoch": 5.203617932360395,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002635750245971468,
      "loss": 5.1862,
      "mean_token_accuracy": 0.21476999223232268,
      "num_tokens": 120662979.0,
      "step": 59545
    },
    {
      "entropy": 5.802068471908569,
      "epoch": 5.2040548807131,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026354390380718717,
      "loss": 5.2144,
      "mean_token_accuracy": 0.21009233146905898,
      "num_tokens": 120672963.0,
      "step": 59550
    },
    {
      "entropy": 5.875054407119751,
      "epoch": 5.204491829065804,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026351278323696154,
      "loss": 5.2343,
      "mean_token_accuracy": 0.21222781389951706,
      "num_tokens": 120682010.0,
      "step": 59555
    },
    {
      "entropy": 5.8595099449157715,
      "epoch": 5.204928777418509,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002634816628870668,
      "loss": 5.1739,
      "mean_token_accuracy": 0.22173442989587783,
      "num_tokens": 120692146.0,
      "step": 59560
    },
    {
      "entropy": 5.86117467880249,
      "epoch": 5.205365725771214,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026345054275809973,
      "loss": 5.183,
      "mean_token_accuracy": 0.22119509279727936,
      "num_tokens": 120701956.0,
      "step": 59565
    },
    {
      "entropy": 5.8042584419250485,
      "epoch": 5.205802674123919,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026341942285065735,
      "loss": 5.1127,
      "mean_token_accuracy": 0.2146507531404495,
      "num_tokens": 120711667.0,
      "step": 59570
    },
    {
      "entropy": 5.8373575687408445,
      "epoch": 5.206239622476623,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026338830316533646,
      "loss": 5.1635,
      "mean_token_accuracy": 0.21944501698017121,
      "num_tokens": 120721765.0,
      "step": 59575
    },
    {
      "entropy": 5.863487386703492,
      "epoch": 5.206676570829328,
      "grad_norm": 1.140625,
      "learning_rate": 0.000263357183702734,
      "loss": 5.2201,
      "mean_token_accuracy": 0.21468270123004912,
      "num_tokens": 120731933.0,
      "step": 59580
    },
    {
      "entropy": 5.89162392616272,
      "epoch": 5.207113519182033,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002633260644634469,
      "loss": 5.1906,
      "mean_token_accuracy": 0.2211417943239212,
      "num_tokens": 120743303.0,
      "step": 59585
    },
    {
      "entropy": 5.8304356098175045,
      "epoch": 5.2075504675347375,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026329494544807195,
      "loss": 5.2035,
      "mean_token_accuracy": 0.2091735601425171,
      "num_tokens": 120754264.0,
      "step": 59590
    },
    {
      "entropy": 5.858281898498535,
      "epoch": 5.207987415887442,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002632638266572061,
      "loss": 5.2284,
      "mean_token_accuracy": 0.2162725418806076,
      "num_tokens": 120763973.0,
      "step": 59595
    },
    {
      "entropy": 5.9268763065338135,
      "epoch": 5.208424364240146,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002632327080914462,
      "loss": 5.2539,
      "mean_token_accuracy": 0.20668623894453048,
      "num_tokens": 120773577.0,
      "step": 59600
    },
    {
      "entropy": 5.82003493309021,
      "epoch": 5.208861312592852,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002632015897513891,
      "loss": 5.1415,
      "mean_token_accuracy": 0.2265907347202301,
      "num_tokens": 120783130.0,
      "step": 59605
    },
    {
      "entropy": 5.864716529846191,
      "epoch": 5.209298260945556,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026317047163763174,
      "loss": 5.2629,
      "mean_token_accuracy": 0.21019758135080338,
      "num_tokens": 120792783.0,
      "step": 59610
    },
    {
      "entropy": 5.833669185638428,
      "epoch": 5.209735209298261,
      "grad_norm": 1.046875,
      "learning_rate": 0.000263139353750771,
      "loss": 5.1401,
      "mean_token_accuracy": 0.21307841837406158,
      "num_tokens": 120802739.0,
      "step": 59615
    },
    {
      "entropy": 5.807094955444336,
      "epoch": 5.210172157650966,
      "grad_norm": 1.03125,
      "learning_rate": 0.00026310823609140355,
      "loss": 5.1413,
      "mean_token_accuracy": 0.2200099840760231,
      "num_tokens": 120812557.0,
      "step": 59620
    },
    {
      "entropy": 5.876937770843506,
      "epoch": 5.2106091060036706,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002630771186601264,
      "loss": 5.1786,
      "mean_token_accuracy": 0.21439557522535324,
      "num_tokens": 120821496.0,
      "step": 59625
    },
    {
      "entropy": 5.828439807891845,
      "epoch": 5.211046054356375,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026304600145753636,
      "loss": 5.2318,
      "mean_token_accuracy": 0.21151545345783235,
      "num_tokens": 120831878.0,
      "step": 59630
    },
    {
      "entropy": 5.8557085514068605,
      "epoch": 5.2114830027090795,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026301488448423027,
      "loss": 5.1956,
      "mean_token_accuracy": 0.21431923657655716,
      "num_tokens": 120841521.0,
      "step": 59635
    },
    {
      "entropy": 5.877801704406738,
      "epoch": 5.211919951061785,
      "grad_norm": 1.203125,
      "learning_rate": 0.000262983767740805,
      "loss": 5.2175,
      "mean_token_accuracy": 0.21632504910230638,
      "num_tokens": 120851632.0,
      "step": 59640
    },
    {
      "entropy": 5.895924377441406,
      "epoch": 5.212356899414489,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002629526512278574,
      "loss": 5.2977,
      "mean_token_accuracy": 0.20647418946027757,
      "num_tokens": 120862657.0,
      "step": 59645
    },
    {
      "entropy": 5.88104395866394,
      "epoch": 5.212793847767194,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002629215349459842,
      "loss": 5.1629,
      "mean_token_accuracy": 0.22061215490102767,
      "num_tokens": 120874346.0,
      "step": 59650
    },
    {
      "entropy": 5.815547609329224,
      "epoch": 5.213230796119898,
      "grad_norm": 1.0,
      "learning_rate": 0.00026289041889578226,
      "loss": 5.1193,
      "mean_token_accuracy": 0.2317645013332367,
      "num_tokens": 120884507.0,
      "step": 59655
    },
    {
      "entropy": 5.814928770065308,
      "epoch": 5.213667744472604,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026285930307784846,
      "loss": 5.1836,
      "mean_token_accuracy": 0.2194572702050209,
      "num_tokens": 120893937.0,
      "step": 59660
    },
    {
      "entropy": 5.867707538604736,
      "epoch": 5.214104692825308,
      "grad_norm": 1.125,
      "learning_rate": 0.00026282818749277966,
      "loss": 5.1969,
      "mean_token_accuracy": 0.21402524560689926,
      "num_tokens": 120903301.0,
      "step": 59665
    },
    {
      "entropy": 5.81575345993042,
      "epoch": 5.2145416411780126,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002627970721411725,
      "loss": 5.1615,
      "mean_token_accuracy": 0.2180611252784729,
      "num_tokens": 120914191.0,
      "step": 59670
    },
    {
      "entropy": 5.856165599822998,
      "epoch": 5.214978589530718,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026276595702362396,
      "loss": 5.1968,
      "mean_token_accuracy": 0.21825837939977646,
      "num_tokens": 120924227.0,
      "step": 59675
    },
    {
      "entropy": 5.858949041366577,
      "epoch": 5.215415537883422,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026273484214073067,
      "loss": 5.2388,
      "mean_token_accuracy": 0.21113680452108383,
      "num_tokens": 120934528.0,
      "step": 59680
    },
    {
      "entropy": 5.913558721542358,
      "epoch": 5.215852486236127,
      "grad_norm": 1.125,
      "learning_rate": 0.0002627037274930896,
      "loss": 5.2269,
      "mean_token_accuracy": 0.2112354025244713,
      "num_tokens": 120943815.0,
      "step": 59685
    },
    {
      "entropy": 5.900870990753174,
      "epoch": 5.216289434588831,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002626726130812975,
      "loss": 5.2524,
      "mean_token_accuracy": 0.21527790576219558,
      "num_tokens": 120954021.0,
      "step": 59690
    },
    {
      "entropy": 5.863891983032227,
      "epoch": 5.216726382941537,
      "grad_norm": 1.125,
      "learning_rate": 0.0002626414989059512,
      "loss": 5.2533,
      "mean_token_accuracy": 0.2147976875305176,
      "num_tokens": 120963018.0,
      "step": 59695
    },
    {
      "entropy": 5.910703420639038,
      "epoch": 5.217163331294241,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002626103849676473,
      "loss": 5.2603,
      "mean_token_accuracy": 0.21211109459400176,
      "num_tokens": 120972777.0,
      "step": 59700
    },
    {
      "entropy": 5.77928147315979,
      "epoch": 5.217600279646946,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026257927126698276,
      "loss": 5.0596,
      "mean_token_accuracy": 0.23184471130371093,
      "num_tokens": 120982449.0,
      "step": 59705
    },
    {
      "entropy": 5.891949033737182,
      "epoch": 5.21803722799965,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026254815780455434,
      "loss": 5.1945,
      "mean_token_accuracy": 0.21335447430610657,
      "num_tokens": 120991706.0,
      "step": 59710
    },
    {
      "entropy": 5.872481012344361,
      "epoch": 5.218474176352355,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00026251704458095875,
      "loss": 5.2349,
      "mean_token_accuracy": 0.2126376748085022,
      "num_tokens": 121001857.0,
      "step": 59715
    },
    {
      "entropy": 5.839682674407959,
      "epoch": 5.21891112470506,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002624859315967928,
      "loss": 5.1823,
      "mean_token_accuracy": 0.2189657896757126,
      "num_tokens": 121010957.0,
      "step": 59720
    },
    {
      "entropy": 5.878854036331177,
      "epoch": 5.219348073057764,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026245481885265324,
      "loss": 5.2606,
      "mean_token_accuracy": 0.2052150249481201,
      "num_tokens": 121021229.0,
      "step": 59725
    },
    {
      "entropy": 5.8604163646698,
      "epoch": 5.21978502141047,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026242370634913675,
      "loss": 5.1755,
      "mean_token_accuracy": 0.2118426114320755,
      "num_tokens": 121031999.0,
      "step": 59730
    },
    {
      "entropy": 5.8892356872558596,
      "epoch": 5.220221969763174,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026239259408684023,
      "loss": 5.2337,
      "mean_token_accuracy": 0.2204602912068367,
      "num_tokens": 121042220.0,
      "step": 59735
    },
    {
      "entropy": 5.826741743087768,
      "epoch": 5.220658918115879,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002623614820663604,
      "loss": 5.2137,
      "mean_token_accuracy": 0.20912555009126663,
      "num_tokens": 121053347.0,
      "step": 59740
    },
    {
      "entropy": 5.900650835037231,
      "epoch": 5.221095866468583,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026233037028829386,
      "loss": 5.179,
      "mean_token_accuracy": 0.21821261644363404,
      "num_tokens": 121063364.0,
      "step": 59745
    },
    {
      "entropy": 5.8945635795593265,
      "epoch": 5.2215328148212885,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002622992587532375,
      "loss": 5.2509,
      "mean_token_accuracy": 0.21345202773809432,
      "num_tokens": 121074530.0,
      "step": 59750
    },
    {
      "entropy": 5.872925519943237,
      "epoch": 5.221969763173993,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002622681474617881,
      "loss": 5.2185,
      "mean_token_accuracy": 0.21497586816549302,
      "num_tokens": 121084852.0,
      "step": 59755
    },
    {
      "entropy": 5.74947338104248,
      "epoch": 5.222406711526697,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026223703641454224,
      "loss": 5.1233,
      "mean_token_accuracy": 0.2290520414710045,
      "num_tokens": 121094108.0,
      "step": 59760
    },
    {
      "entropy": 5.83327350616455,
      "epoch": 5.222843659879402,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026220592561209674,
      "loss": 5.2115,
      "mean_token_accuracy": 0.2160020425915718,
      "num_tokens": 121103411.0,
      "step": 59765
    },
    {
      "entropy": 5.8650720596313475,
      "epoch": 5.223280608232107,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002621748150550482,
      "loss": 5.2103,
      "mean_token_accuracy": 0.21564893722534179,
      "num_tokens": 121112739.0,
      "step": 59770
    },
    {
      "entropy": 5.935981750488281,
      "epoch": 5.223717556584812,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026214370474399347,
      "loss": 5.2236,
      "mean_token_accuracy": 0.21107665747404097,
      "num_tokens": 121122460.0,
      "step": 59775
    },
    {
      "entropy": 5.8167712688446045,
      "epoch": 5.224154504937516,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002621125946795292,
      "loss": 5.1892,
      "mean_token_accuracy": 0.21832638829946518,
      "num_tokens": 121131733.0,
      "step": 59780
    },
    {
      "entropy": 5.797769927978516,
      "epoch": 5.224591453290221,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002620814848622522,
      "loss": 5.136,
      "mean_token_accuracy": 0.22034626454114914,
      "num_tokens": 121141241.0,
      "step": 59785
    },
    {
      "entropy": 5.886603832244873,
      "epoch": 5.225028401642926,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026205037529275905,
      "loss": 5.2265,
      "mean_token_accuracy": 0.21348435133695604,
      "num_tokens": 121151665.0,
      "step": 59790
    },
    {
      "entropy": 5.898061275482178,
      "epoch": 5.2254653499956305,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026201926597164645,
      "loss": 5.2701,
      "mean_token_accuracy": 0.20723362416028976,
      "num_tokens": 121161823.0,
      "step": 59795
    },
    {
      "entropy": 5.8310445785522464,
      "epoch": 5.225902298348335,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002619881568995111,
      "loss": 5.1653,
      "mean_token_accuracy": 0.22152884304523468,
      "num_tokens": 121170848.0,
      "step": 59800
    },
    {
      "entropy": 5.865543794631958,
      "epoch": 5.22633924670104,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002619570480769498,
      "loss": 5.1936,
      "mean_token_accuracy": 0.22015177458524704,
      "num_tokens": 121181158.0,
      "step": 59805
    },
    {
      "entropy": 5.8546429634094235,
      "epoch": 5.226776195053745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026192593950455915,
      "loss": 5.2,
      "mean_token_accuracy": 0.21351416707038878,
      "num_tokens": 121191362.0,
      "step": 59810
    },
    {
      "entropy": 5.886496591567993,
      "epoch": 5.227213143406449,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026189483118293583,
      "loss": 5.2874,
      "mean_token_accuracy": 0.21233968138694764,
      "num_tokens": 121201381.0,
      "step": 59815
    },
    {
      "entropy": 5.882571792602539,
      "epoch": 5.227650091759154,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026186372311267644,
      "loss": 5.1844,
      "mean_token_accuracy": 0.22107996940612792,
      "num_tokens": 121211633.0,
      "step": 59820
    },
    {
      "entropy": 5.836688613891601,
      "epoch": 5.228087040111859,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002618326152943778,
      "loss": 5.1633,
      "mean_token_accuracy": 0.21716644763946533,
      "num_tokens": 121221354.0,
      "step": 59825
    },
    {
      "entropy": 5.828885746002197,
      "epoch": 5.228523988464564,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002618015077286364,
      "loss": 5.2043,
      "mean_token_accuracy": 0.2165147989988327,
      "num_tokens": 121231424.0,
      "step": 59830
    },
    {
      "entropy": 5.853335952758789,
      "epoch": 5.228960936817268,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026177040041604915,
      "loss": 5.1397,
      "mean_token_accuracy": 0.22017797976732253,
      "num_tokens": 121240765.0,
      "step": 59835
    },
    {
      "entropy": 5.873388862609863,
      "epoch": 5.2293978851699725,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002617392933572125,
      "loss": 5.2315,
      "mean_token_accuracy": 0.21456877291202545,
      "num_tokens": 121249550.0,
      "step": 59840
    },
    {
      "entropy": 5.899400043487549,
      "epoch": 5.229834833522678,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002617081865527231,
      "loss": 5.2002,
      "mean_token_accuracy": 0.2168462574481964,
      "num_tokens": 121258212.0,
      "step": 59845
    },
    {
      "entropy": 5.791489219665527,
      "epoch": 5.230271781875382,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002616770800031776,
      "loss": 5.1646,
      "mean_token_accuracy": 0.22458794265985488,
      "num_tokens": 121267865.0,
      "step": 59850
    },
    {
      "entropy": 5.8536584854125975,
      "epoch": 5.230708730228087,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026164597370917283,
      "loss": 5.1868,
      "mean_token_accuracy": 0.2242321714758873,
      "num_tokens": 121277833.0,
      "step": 59855
    },
    {
      "entropy": 5.916000556945801,
      "epoch": 5.231145678580792,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002616148676713052,
      "loss": 5.2882,
      "mean_token_accuracy": 0.20751732885837554,
      "num_tokens": 121288646.0,
      "step": 59860
    },
    {
      "entropy": 5.872376298904419,
      "epoch": 5.231582626933497,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002615837618901714,
      "loss": 5.1799,
      "mean_token_accuracy": 0.21650264412164688,
      "num_tokens": 121298392.0,
      "step": 59865
    },
    {
      "entropy": 5.887739801406861,
      "epoch": 5.232019575286201,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002615526563663681,
      "loss": 5.27,
      "mean_token_accuracy": 0.21559444963932037,
      "num_tokens": 121308222.0,
      "step": 59870
    },
    {
      "entropy": 5.930021667480469,
      "epoch": 5.232456523638906,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026152155110049187,
      "loss": 5.3076,
      "mean_token_accuracy": 0.21431175321340562,
      "num_tokens": 121318699.0,
      "step": 59875
    },
    {
      "entropy": 5.90586404800415,
      "epoch": 5.232893471991611,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002614904460931394,
      "loss": 5.2424,
      "mean_token_accuracy": 0.21388984024524688,
      "num_tokens": 121329256.0,
      "step": 59880
    },
    {
      "entropy": 5.8526959896087645,
      "epoch": 5.233330420344315,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002614593413449072,
      "loss": 5.2199,
      "mean_token_accuracy": 0.2163341909646988,
      "num_tokens": 121340654.0,
      "step": 59885
    },
    {
      "entropy": 5.817443990707398,
      "epoch": 5.23376736869702,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000261428236856392,
      "loss": 5.1526,
      "mean_token_accuracy": 0.21750302165746688,
      "num_tokens": 121350171.0,
      "step": 59890
    },
    {
      "entropy": 5.924517488479614,
      "epoch": 5.234204317049724,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002613971326281903,
      "loss": 5.3792,
      "mean_token_accuracy": 0.19948514699935913,
      "num_tokens": 121360650.0,
      "step": 59895
    },
    {
      "entropy": 5.936907625198364,
      "epoch": 5.23464126540243,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002613660286608987,
      "loss": 5.238,
      "mean_token_accuracy": 0.21010339707136155,
      "num_tokens": 121370800.0,
      "step": 59900
    },
    {
      "entropy": 5.871545600891113,
      "epoch": 5.235078213755134,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026133492495511386,
      "loss": 5.2347,
      "mean_token_accuracy": 0.21444839835166932,
      "num_tokens": 121380857.0,
      "step": 59905
    },
    {
      "entropy": 5.913450431823731,
      "epoch": 5.235515162107839,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002613038215114323,
      "loss": 5.2973,
      "mean_token_accuracy": 0.20016632825136185,
      "num_tokens": 121391656.0,
      "step": 59910
    },
    {
      "entropy": 5.917642879486084,
      "epoch": 5.235952110460544,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026127271833045057,
      "loss": 5.2384,
      "mean_token_accuracy": 0.21589977592229842,
      "num_tokens": 121401840.0,
      "step": 59915
    },
    {
      "entropy": 5.863652753829956,
      "epoch": 5.2363890588132485,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002612416154127654,
      "loss": 5.1881,
      "mean_token_accuracy": 0.2227205291390419,
      "num_tokens": 121411435.0,
      "step": 59920
    },
    {
      "entropy": 5.856557607650757,
      "epoch": 5.236826007165953,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002612105127589732,
      "loss": 5.1858,
      "mean_token_accuracy": 0.22204433083534242,
      "num_tokens": 121420952.0,
      "step": 59925
    },
    {
      "entropy": 5.846410894393921,
      "epoch": 5.237262955518657,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002611794103696706,
      "loss": 5.223,
      "mean_token_accuracy": 0.21636005938053132,
      "num_tokens": 121430855.0,
      "step": 59930
    },
    {
      "entropy": 5.8894425392150875,
      "epoch": 5.237699903871363,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002611483082454541,
      "loss": 5.2011,
      "mean_token_accuracy": 0.21901308447122575,
      "num_tokens": 121439876.0,
      "step": 59935
    },
    {
      "entropy": 5.8911243915557865,
      "epoch": 5.238136852224067,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002611172063869203,
      "loss": 5.2582,
      "mean_token_accuracy": 0.21807927042245864,
      "num_tokens": 121450295.0,
      "step": 59940
    },
    {
      "entropy": 5.901226615905761,
      "epoch": 5.238573800576772,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002610861047946659,
      "loss": 5.2502,
      "mean_token_accuracy": 0.21144324243068696,
      "num_tokens": 121460269.0,
      "step": 59945
    },
    {
      "entropy": 5.874485445022583,
      "epoch": 5.239010748929476,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002610550034692872,
      "loss": 5.2291,
      "mean_token_accuracy": 0.2158267855644226,
      "num_tokens": 121470384.0,
      "step": 59950
    },
    {
      "entropy": 5.799777412414551,
      "epoch": 5.2394476972821815,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002610239024113808,
      "loss": 5.1478,
      "mean_token_accuracy": 0.22807978093624115,
      "num_tokens": 121479622.0,
      "step": 59955
    },
    {
      "entropy": 5.794666862487793,
      "epoch": 5.239884645634886,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002609928016215433,
      "loss": 5.1882,
      "mean_token_accuracy": 0.21545412987470627,
      "num_tokens": 121489253.0,
      "step": 59960
    },
    {
      "entropy": 5.89273509979248,
      "epoch": 5.2403215939875905,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002609617011003712,
      "loss": 5.2431,
      "mean_token_accuracy": 0.20826376527547835,
      "num_tokens": 121498954.0,
      "step": 59965
    },
    {
      "entropy": 5.924697637557983,
      "epoch": 5.240758542340295,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002609306008484611,
      "loss": 5.2223,
      "mean_token_accuracy": 0.21894704550504684,
      "num_tokens": 121509055.0,
      "step": 59970
    },
    {
      "entropy": 5.895440101623535,
      "epoch": 5.241195490693,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002608995008664094,
      "loss": 5.2121,
      "mean_token_accuracy": 0.2235142856836319,
      "num_tokens": 121519940.0,
      "step": 59975
    },
    {
      "entropy": 5.8982984066009525,
      "epoch": 5.241632439045705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026086840115481264,
      "loss": 5.2178,
      "mean_token_accuracy": 0.2131478816270828,
      "num_tokens": 121530576.0,
      "step": 59980
    },
    {
      "entropy": 5.8004625797271725,
      "epoch": 5.242069387398409,
      "grad_norm": 1.640625,
      "learning_rate": 0.0002608373017142674,
      "loss": 5.0896,
      "mean_token_accuracy": 0.22968045771121978,
      "num_tokens": 121540934.0,
      "step": 59985
    },
    {
      "entropy": 5.870791339874268,
      "epoch": 5.242506335751115,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00026080620254537005,
      "loss": 5.2371,
      "mean_token_accuracy": 0.2174873545765877,
      "num_tokens": 121551222.0,
      "step": 59990
    },
    {
      "entropy": 5.807149124145508,
      "epoch": 5.242943284103819,
      "grad_norm": 1.125,
      "learning_rate": 0.00026077510364871725,
      "loss": 5.1538,
      "mean_token_accuracy": 0.22264270782470702,
      "num_tokens": 121560900.0,
      "step": 59995
    },
    {
      "entropy": 5.894606256484986,
      "epoch": 5.2433802324565235,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026074400502490544,
      "loss": 5.2708,
      "mean_token_accuracy": 0.21274588704109193,
      "num_tokens": 121570859.0,
      "step": 60000
    },
    {
      "epoch": 5.2433802324565235,
      "eval_entropy": 5.634189322722867,
      "eval_loss": 5.312753677368164,
      "eval_mean_token_accuracy": 0.21801283864273427,
      "eval_num_tokens": 121570859.0,
      "eval_runtime": 27.8935,
      "eval_samples_per_second": 1326.977,
      "eval_steps_per_second": 165.881,
      "step": 60000
    },
    {
      "entropy": 5.885384559631348,
      "epoch": 5.243817180809228,
      "grad_norm": 1.0,
      "learning_rate": 0.000260712906674531,
      "loss": 5.22,
      "mean_token_accuracy": 0.21718786358833314,
      "num_tokens": 121581386.0,
      "step": 60005
    },
    {
      "entropy": 5.8596776008605955,
      "epoch": 5.244254129161933,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026068180859819054,
      "loss": 5.1976,
      "mean_token_accuracy": 0.21996743232011795,
      "num_tokens": 121590900.0,
      "step": 60010
    },
    {
      "entropy": 5.911012411117554,
      "epoch": 5.244691077514638,
      "grad_norm": 1.125,
      "learning_rate": 0.00026065071079648045,
      "loss": 5.3055,
      "mean_token_accuracy": 0.20941568911075592,
      "num_tokens": 121600358.0,
      "step": 60015
    },
    {
      "entropy": 5.906136512756348,
      "epoch": 5.245128025867342,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026061961326999736,
      "loss": 5.3031,
      "mean_token_accuracy": 0.20146828293800353,
      "num_tokens": 121611373.0,
      "step": 60020
    },
    {
      "entropy": 5.889150381088257,
      "epoch": 5.245564974220047,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002605885160193375,
      "loss": 5.1868,
      "mean_token_accuracy": 0.21717663556337358,
      "num_tokens": 121621632.0,
      "step": 60025
    },
    {
      "entropy": 5.887444353103637,
      "epoch": 5.246001922572752,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002605574190450975,
      "loss": 5.2142,
      "mean_token_accuracy": 0.2119697079062462,
      "num_tokens": 121631501.0,
      "step": 60030
    },
    {
      "entropy": 5.863863897323609,
      "epoch": 5.246438870925457,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002605263223478738,
      "loss": 5.2059,
      "mean_token_accuracy": 0.2185255229473114,
      "num_tokens": 121641628.0,
      "step": 60035
    },
    {
      "entropy": 5.8797064304351805,
      "epoch": 5.246875819278161,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002604952259282628,
      "loss": 5.197,
      "mean_token_accuracy": 0.21827581226825715,
      "num_tokens": 121651582.0,
      "step": 60040
    },
    {
      "entropy": 5.893228769302368,
      "epoch": 5.247312767630866,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026046412978686096,
      "loss": 5.1996,
      "mean_token_accuracy": 0.2249930754303932,
      "num_tokens": 121662947.0,
      "step": 60045
    },
    {
      "entropy": 5.870632886886597,
      "epoch": 5.247749715983571,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026043303392426473,
      "loss": 5.1842,
      "mean_token_accuracy": 0.22083420157432557,
      "num_tokens": 121673634.0,
      "step": 60050
    },
    {
      "entropy": 5.867474794387817,
      "epoch": 5.248186664336275,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002604019383410705,
      "loss": 5.2428,
      "mean_token_accuracy": 0.21600246280431748,
      "num_tokens": 121683982.0,
      "step": 60055
    },
    {
      "entropy": 5.90163369178772,
      "epoch": 5.24862361268898,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002603708430378748,
      "loss": 5.2828,
      "mean_token_accuracy": 0.21219853013753892,
      "num_tokens": 121694319.0,
      "step": 60060
    },
    {
      "entropy": 5.897940015792846,
      "epoch": 5.249060561041685,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000260339748015274,
      "loss": 5.1917,
      "mean_token_accuracy": 0.22031523585319518,
      "num_tokens": 121704560.0,
      "step": 60065
    },
    {
      "entropy": 5.850846147537231,
      "epoch": 5.24949750939439,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002603086532738645,
      "loss": 5.1179,
      "mean_token_accuracy": 0.22137697041034698,
      "num_tokens": 121714403.0,
      "step": 60070
    },
    {
      "entropy": 5.7762586116790775,
      "epoch": 5.249934457747094,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002602775588142427,
      "loss": 5.0979,
      "mean_token_accuracy": 0.2215091735124588,
      "num_tokens": 121724104.0,
      "step": 60075
    },
    {
      "entropy": 5.83536810874939,
      "epoch": 5.250371406099799,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002602464646370051,
      "loss": 5.2554,
      "mean_token_accuracy": 0.21206511408090592,
      "num_tokens": 121733956.0,
      "step": 60080
    },
    {
      "entropy": 5.8713164806365965,
      "epoch": 5.250808354452504,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002602153707427479,
      "loss": 5.2319,
      "mean_token_accuracy": 0.20734913349151612,
      "num_tokens": 121743135.0,
      "step": 60085
    },
    {
      "entropy": 5.908465003967285,
      "epoch": 5.251245302805208,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026018427713206784,
      "loss": 5.2337,
      "mean_token_accuracy": 0.21997734904289246,
      "num_tokens": 121753118.0,
      "step": 60090
    },
    {
      "entropy": 5.838523483276367,
      "epoch": 5.251682251157913,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026015318380556094,
      "loss": 5.1527,
      "mean_token_accuracy": 0.22286375910043715,
      "num_tokens": 121763505.0,
      "step": 60095
    },
    {
      "entropy": 5.827185535430909,
      "epoch": 5.252119199510618,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002601220907638238,
      "loss": 5.2266,
      "mean_token_accuracy": 0.20966774523258208,
      "num_tokens": 121773492.0,
      "step": 60100
    },
    {
      "entropy": 5.866214847564697,
      "epoch": 5.252556147863323,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026009099800745276,
      "loss": 5.1672,
      "mean_token_accuracy": 0.22165706306695937,
      "num_tokens": 121782674.0,
      "step": 60105
    },
    {
      "entropy": 5.827052021026612,
      "epoch": 5.252993096216027,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002600599055370442,
      "loss": 5.2071,
      "mean_token_accuracy": 0.21737801432609558,
      "num_tokens": 121793288.0,
      "step": 60110
    },
    {
      "entropy": 5.878500461578369,
      "epoch": 5.253430044568732,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002600288133531945,
      "loss": 5.1949,
      "mean_token_accuracy": 0.22066044956445693,
      "num_tokens": 121804653.0,
      "step": 60115
    },
    {
      "entropy": 5.895565366744995,
      "epoch": 5.253866992921437,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025999772145650005,
      "loss": 5.175,
      "mean_token_accuracy": 0.21330095082521439,
      "num_tokens": 121815586.0,
      "step": 60120
    },
    {
      "entropy": 5.90427074432373,
      "epoch": 5.2543039412741415,
      "grad_norm": 1.171875,
      "learning_rate": 0.000259966629847557,
      "loss": 5.2279,
      "mean_token_accuracy": 0.21497530490159988,
      "num_tokens": 121825345.0,
      "step": 60125
    },
    {
      "entropy": 5.8160826683044435,
      "epoch": 5.254740889626846,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000259935538526962,
      "loss": 5.2634,
      "mean_token_accuracy": 0.21659717261791228,
      "num_tokens": 121835502.0,
      "step": 60130
    },
    {
      "entropy": 5.79560694694519,
      "epoch": 5.25517783797955,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025990444749531124,
      "loss": 5.1513,
      "mean_token_accuracy": 0.22261205762624742,
      "num_tokens": 121845318.0,
      "step": 60135
    },
    {
      "entropy": 5.815024518966675,
      "epoch": 5.255614786332256,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002598733567532011,
      "loss": 5.1317,
      "mean_token_accuracy": 0.22404042035341262,
      "num_tokens": 121855465.0,
      "step": 60140
    },
    {
      "entropy": 5.957713794708252,
      "epoch": 5.25605173468496,
      "grad_norm": 1.03125,
      "learning_rate": 0.00025984226630122785,
      "loss": 5.3151,
      "mean_token_accuracy": 0.2082614153623581,
      "num_tokens": 121866095.0,
      "step": 60145
    },
    {
      "entropy": 5.906035423278809,
      "epoch": 5.256488683037665,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00025981117613998794,
      "loss": 5.2112,
      "mean_token_accuracy": 0.21807369440793992,
      "num_tokens": 121876342.0,
      "step": 60150
    },
    {
      "entropy": 5.870071744918823,
      "epoch": 5.25692563139037,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025978008627007754,
      "loss": 5.1659,
      "mean_token_accuracy": 0.21952869296073912,
      "num_tokens": 121887258.0,
      "step": 60155
    },
    {
      "entropy": 5.873815870285034,
      "epoch": 5.2573625797430745,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025974899669209314,
      "loss": 5.225,
      "mean_token_accuracy": 0.21013241857290268,
      "num_tokens": 121897725.0,
      "step": 60160
    },
    {
      "entropy": 5.862902402877808,
      "epoch": 5.257799528095779,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000259717907406631,
      "loss": 5.2284,
      "mean_token_accuracy": 0.21652440428733827,
      "num_tokens": 121907911.0,
      "step": 60165
    },
    {
      "entropy": 5.826011896133423,
      "epoch": 5.2582364764484835,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002596868184142873,
      "loss": 5.1779,
      "mean_token_accuracy": 0.21111556887626648,
      "num_tokens": 121919137.0,
      "step": 60170
    },
    {
      "entropy": 5.829453420639038,
      "epoch": 5.258673424801189,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025965572971565847,
      "loss": 5.2684,
      "mean_token_accuracy": 0.21481546014547348,
      "num_tokens": 121930586.0,
      "step": 60175
    },
    {
      "entropy": 5.876383829116821,
      "epoch": 5.259110373153893,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025962464131134086,
      "loss": 5.1971,
      "mean_token_accuracy": 0.21773429363965988,
      "num_tokens": 121940344.0,
      "step": 60180
    },
    {
      "entropy": 5.838567543029785,
      "epoch": 5.259547321506598,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025959355320193063,
      "loss": 5.1098,
      "mean_token_accuracy": 0.22547689080238342,
      "num_tokens": 121949828.0,
      "step": 60185
    },
    {
      "entropy": 5.850296545028686,
      "epoch": 5.259984269859302,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025956246538802414,
      "loss": 5.2529,
      "mean_token_accuracy": 0.2106336086988449,
      "num_tokens": 121958723.0,
      "step": 60190
    },
    {
      "entropy": 5.795763826370239,
      "epoch": 5.260421218212008,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002595313778702176,
      "loss": 5.0965,
      "mean_token_accuracy": 0.22563048750162124,
      "num_tokens": 121970144.0,
      "step": 60195
    },
    {
      "entropy": 5.793441915512085,
      "epoch": 5.260858166564712,
      "grad_norm": 1.03125,
      "learning_rate": 0.00025950029064910745,
      "loss": 5.1497,
      "mean_token_accuracy": 0.22209599912166594,
      "num_tokens": 121979992.0,
      "step": 60200
    },
    {
      "entropy": 5.929102993011474,
      "epoch": 5.2612951149174165,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025946920372528975,
      "loss": 5.2481,
      "mean_token_accuracy": 0.20984494537115098,
      "num_tokens": 121990191.0,
      "step": 60205
    },
    {
      "entropy": 5.888847923278808,
      "epoch": 5.261732063270122,
      "grad_norm": 1.046875,
      "learning_rate": 0.00025943811709936093,
      "loss": 5.1581,
      "mean_token_accuracy": 0.22158268690109253,
      "num_tokens": 122000792.0,
      "step": 60210
    },
    {
      "entropy": 5.845481491088867,
      "epoch": 5.262169011622826,
      "grad_norm": 1.125,
      "learning_rate": 0.00025940703077191715,
      "loss": 5.2288,
      "mean_token_accuracy": 0.2192952439188957,
      "num_tokens": 122012021.0,
      "step": 60215
    },
    {
      "entropy": 5.895156002044677,
      "epoch": 5.262605959975531,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002593759447435547,
      "loss": 5.1954,
      "mean_token_accuracy": 0.21357356607913972,
      "num_tokens": 122021425.0,
      "step": 60220
    },
    {
      "entropy": 6.016922807693481,
      "epoch": 5.263042908328235,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00025934485901486976,
      "loss": 5.3694,
      "mean_token_accuracy": 0.20517150312662125,
      "num_tokens": 122031061.0,
      "step": 60225
    },
    {
      "entropy": 5.920328998565674,
      "epoch": 5.263479856680941,
      "grad_norm": 1.078125,
      "learning_rate": 0.00025931377358645857,
      "loss": 5.2692,
      "mean_token_accuracy": 0.20999946892261506,
      "num_tokens": 122040279.0,
      "step": 60230
    },
    {
      "entropy": 5.855244112014771,
      "epoch": 5.263916805033645,
      "grad_norm": 1.015625,
      "learning_rate": 0.00025928268845891754,
      "loss": 5.2146,
      "mean_token_accuracy": 0.21685483008623124,
      "num_tokens": 122050241.0,
      "step": 60235
    },
    {
      "entropy": 5.861480331420898,
      "epoch": 5.26435375338635,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00025925160363284277,
      "loss": 5.1774,
      "mean_token_accuracy": 0.21639186441898345,
      "num_tokens": 122060765.0,
      "step": 60240
    },
    {
      "entropy": 5.862980461120605,
      "epoch": 5.264790701739054,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002592205191088304,
      "loss": 5.1509,
      "mean_token_accuracy": 0.2253304362297058,
      "num_tokens": 122070271.0,
      "step": 60245
    },
    {
      "entropy": 5.877711009979248,
      "epoch": 5.265227650091759,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00025918943488747694,
      "loss": 5.2118,
      "mean_token_accuracy": 0.21632240563631058,
      "num_tokens": 122080401.0,
      "step": 60250
    },
    {
      "entropy": 5.825073289871216,
      "epoch": 5.265664598444464,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025915835096937824,
      "loss": 5.1972,
      "mean_token_accuracy": 0.2228411078453064,
      "num_tokens": 122090731.0,
      "step": 60255
    },
    {
      "entropy": 5.895104646682739,
      "epoch": 5.266101546797168,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002591272673551307,
      "loss": 5.2707,
      "mean_token_accuracy": 0.21444420665502548,
      "num_tokens": 122100838.0,
      "step": 60260
    },
    {
      "entropy": 5.861852312088013,
      "epoch": 5.266538495149874,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002590961840453305,
      "loss": 5.1859,
      "mean_token_accuracy": 0.21606209725141526,
      "num_tokens": 122110063.0,
      "step": 60265
    },
    {
      "entropy": 5.87294454574585,
      "epoch": 5.266975443502578,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025906510104057386,
      "loss": 5.2139,
      "mean_token_accuracy": 0.2190433219075203,
      "num_tokens": 122119900.0,
      "step": 60270
    },
    {
      "entropy": 5.874405336380005,
      "epoch": 5.267412391855283,
      "grad_norm": 1.046875,
      "learning_rate": 0.00025903401834145693,
      "loss": 5.153,
      "mean_token_accuracy": 0.2137095957994461,
      "num_tokens": 122129424.0,
      "step": 60275
    },
    {
      "entropy": 5.934931516647339,
      "epoch": 5.267849340207987,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002590029359485759,
      "loss": 5.3548,
      "mean_token_accuracy": 0.2084837943315506,
      "num_tokens": 122140729.0,
      "step": 60280
    },
    {
      "entropy": 5.836497020721436,
      "epoch": 5.2682862885606925,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00025897185386252684,
      "loss": 5.1485,
      "mean_token_accuracy": 0.21889135390520095,
      "num_tokens": 122151872.0,
      "step": 60285
    },
    {
      "entropy": 5.89937424659729,
      "epoch": 5.268723236913397,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025894077208390607,
      "loss": 5.2575,
      "mean_token_accuracy": 0.20793858915567398,
      "num_tokens": 122161501.0,
      "step": 60290
    },
    {
      "entropy": 5.843084812164307,
      "epoch": 5.269160185266101,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025890969061330974,
      "loss": 5.1129,
      "mean_token_accuracy": 0.22194865494966506,
      "num_tokens": 122171039.0,
      "step": 60295
    },
    {
      "entropy": 5.824951314926148,
      "epoch": 5.269597133618806,
      "grad_norm": 1.15625,
      "learning_rate": 0.000258878609451334,
      "loss": 5.2167,
      "mean_token_accuracy": 0.21862400472164153,
      "num_tokens": 122180933.0,
      "step": 60300
    },
    {
      "entropy": 5.864669132232666,
      "epoch": 5.270034081971511,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002588475285985749,
      "loss": 5.2161,
      "mean_token_accuracy": 0.21286333948373795,
      "num_tokens": 122190925.0,
      "step": 60305
    },
    {
      "entropy": 5.825633859634399,
      "epoch": 5.270471030324216,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002588164480556287,
      "loss": 5.1457,
      "mean_token_accuracy": 0.21826748698949813,
      "num_tokens": 122200386.0,
      "step": 60310
    },
    {
      "entropy": 5.809813785552978,
      "epoch": 5.27090797867692,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002587853678230915,
      "loss": 5.1434,
      "mean_token_accuracy": 0.22054464966058732,
      "num_tokens": 122210812.0,
      "step": 60315
    },
    {
      "entropy": 5.872324275970459,
      "epoch": 5.2713449270296255,
      "grad_norm": 1.0625,
      "learning_rate": 0.00025875428790155945,
      "loss": 5.1915,
      "mean_token_accuracy": 0.2114668920636177,
      "num_tokens": 122220993.0,
      "step": 60320
    },
    {
      "entropy": 5.918923568725586,
      "epoch": 5.27178187538233,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002587232082916288,
      "loss": 5.244,
      "mean_token_accuracy": 0.21513839960098266,
      "num_tokens": 122231371.0,
      "step": 60325
    },
    {
      "entropy": 5.893101930618286,
      "epoch": 5.2722188237350345,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002586921289938953,
      "loss": 5.2057,
      "mean_token_accuracy": 0.21310580372810364,
      "num_tokens": 122241273.0,
      "step": 60330
    },
    {
      "entropy": 5.879921913146973,
      "epoch": 5.272655772087739,
      "grad_norm": 1.015625,
      "learning_rate": 0.00025866105000895545,
      "loss": 5.1926,
      "mean_token_accuracy": 0.21301777809858322,
      "num_tokens": 122251709.0,
      "step": 60335
    },
    {
      "entropy": 5.829068851470947,
      "epoch": 5.273092720440444,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025862997133740517,
      "loss": 5.1729,
      "mean_token_accuracy": 0.21455208957195282,
      "num_tokens": 122262083.0,
      "step": 60340
    },
    {
      "entropy": 5.930835103988647,
      "epoch": 5.273529668793149,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00025859889297984075,
      "loss": 5.2585,
      "mean_token_accuracy": 0.2138911798596382,
      "num_tokens": 122271429.0,
      "step": 60345
    },
    {
      "entropy": 5.879729557037353,
      "epoch": 5.273966617145853,
      "grad_norm": 1.078125,
      "learning_rate": 0.000258567814936858,
      "loss": 5.2512,
      "mean_token_accuracy": 0.21445049345493317,
      "num_tokens": 122282136.0,
      "step": 60350
    },
    {
      "entropy": 5.8395774364471436,
      "epoch": 5.274403565498558,
      "grad_norm": 1.0625,
      "learning_rate": 0.00025853673720905326,
      "loss": 5.1634,
      "mean_token_accuracy": 0.21638136953115464,
      "num_tokens": 122291178.0,
      "step": 60355
    },
    {
      "entropy": 5.863810777664185,
      "epoch": 5.274840513851263,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025850565979702243,
      "loss": 5.2325,
      "mean_token_accuracy": 0.21474552601575853,
      "num_tokens": 122301300.0,
      "step": 60360
    },
    {
      "entropy": 5.855230665206909,
      "epoch": 5.2752774622039675,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00025847458270136174,
      "loss": 5.2333,
      "mean_token_accuracy": 0.21659792810678483,
      "num_tokens": 122312520.0,
      "step": 60365
    },
    {
      "entropy": 5.812800931930542,
      "epoch": 5.275714410556672,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002584435059226672,
      "loss": 5.1042,
      "mean_token_accuracy": 0.2294999450445175,
      "num_tokens": 122322070.0,
      "step": 60370
    },
    {
      "entropy": 5.879456233978272,
      "epoch": 5.276151358909377,
      "grad_norm": 1.171875,
      "learning_rate": 0.000258412429461535,
      "loss": 5.1738,
      "mean_token_accuracy": 0.21753329783678055,
      "num_tokens": 122332329.0,
      "step": 60375
    },
    {
      "entropy": 5.817227840423584,
      "epoch": 5.276588307262082,
      "grad_norm": 1.109375,
      "learning_rate": 0.000258381353318561,
      "loss": 5.1219,
      "mean_token_accuracy": 0.21954327076673508,
      "num_tokens": 122342107.0,
      "step": 60380
    },
    {
      "entropy": 5.83520622253418,
      "epoch": 5.277025255614786,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002583502774943413,
      "loss": 5.2757,
      "mean_token_accuracy": 0.21710471510887147,
      "num_tokens": 122352694.0,
      "step": 60385
    },
    {
      "entropy": 5.865703105926514,
      "epoch": 5.277462203967491,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025831920198947204,
      "loss": 5.1673,
      "mean_token_accuracy": 0.21967174410820006,
      "num_tokens": 122362599.0,
      "step": 60390
    },
    {
      "entropy": 5.8800302028656,
      "epoch": 5.277899152320196,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025828812680454935,
      "loss": 5.1283,
      "mean_token_accuracy": 0.22134342640638352,
      "num_tokens": 122371952.0,
      "step": 60395
    },
    {
      "entropy": 5.86614031791687,
      "epoch": 5.278336100672901,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025825705194016903,
      "loss": 5.2003,
      "mean_token_accuracy": 0.21858495622873306,
      "num_tokens": 122382105.0,
      "step": 60400
    },
    {
      "entropy": 5.899086904525757,
      "epoch": 5.278773049025605,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025822597739692723,
      "loss": 5.2551,
      "mean_token_accuracy": 0.20745190531015395,
      "num_tokens": 122391005.0,
      "step": 60405
    },
    {
      "entropy": 5.849132823944092,
      "epoch": 5.2792099973783095,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002581949031754199,
      "loss": 5.1778,
      "mean_token_accuracy": 0.21599004715681075,
      "num_tokens": 122401987.0,
      "step": 60410
    },
    {
      "entropy": 5.8209610939025875,
      "epoch": 5.279646945731015,
      "grad_norm": 1.0625,
      "learning_rate": 0.00025816382927624327,
      "loss": 5.2012,
      "mean_token_accuracy": 0.2242193639278412,
      "num_tokens": 122413134.0,
      "step": 60415
    },
    {
      "entropy": 5.858914804458618,
      "epoch": 5.280083894083719,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025813275569999325,
      "loss": 5.1803,
      "mean_token_accuracy": 0.22258348464965821,
      "num_tokens": 122423242.0,
      "step": 60420
    },
    {
      "entropy": 5.8232283115386965,
      "epoch": 5.280520842436424,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002581016824472657,
      "loss": 5.1825,
      "mean_token_accuracy": 0.22445365488529206,
      "num_tokens": 122432855.0,
      "step": 60425
    },
    {
      "entropy": 5.88244595527649,
      "epoch": 5.280957790789129,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002580706095186567,
      "loss": 5.2383,
      "mean_token_accuracy": 0.2188464105129242,
      "num_tokens": 122443452.0,
      "step": 60430
    },
    {
      "entropy": 5.835411024093628,
      "epoch": 5.281394739141834,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002580395369147623,
      "loss": 5.2314,
      "mean_token_accuracy": 0.21807456910610198,
      "num_tokens": 122453453.0,
      "step": 60435
    },
    {
      "entropy": 5.849581384658814,
      "epoch": 5.281831687494538,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025800846463617855,
      "loss": 5.1678,
      "mean_token_accuracy": 0.22218048870563506,
      "num_tokens": 122463417.0,
      "step": 60440
    },
    {
      "entropy": 5.914114046096802,
      "epoch": 5.282268635847243,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025797739268350133,
      "loss": 5.2344,
      "mean_token_accuracy": 0.2118669033050537,
      "num_tokens": 122472673.0,
      "step": 60445
    },
    {
      "entropy": 5.923945045471191,
      "epoch": 5.282705584199948,
      "grad_norm": 1.125,
      "learning_rate": 0.0002579463210573266,
      "loss": 5.2233,
      "mean_token_accuracy": 0.21714343279600143,
      "num_tokens": 122483172.0,
      "step": 60450
    },
    {
      "entropy": 5.909732866287231,
      "epoch": 5.283142532552652,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025791524975825043,
      "loss": 5.3517,
      "mean_token_accuracy": 0.2103021040558815,
      "num_tokens": 122494053.0,
      "step": 60455
    },
    {
      "entropy": 5.868380117416382,
      "epoch": 5.283579480905357,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025788417878686866,
      "loss": 5.2345,
      "mean_token_accuracy": 0.21363915950059892,
      "num_tokens": 122504926.0,
      "step": 60460
    },
    {
      "entropy": 5.8714484691619875,
      "epoch": 5.284016429258061,
      "grad_norm": 1.0625,
      "learning_rate": 0.00025785310814377724,
      "loss": 5.2792,
      "mean_token_accuracy": 0.21097590029239655,
      "num_tokens": 122515069.0,
      "step": 60465
    },
    {
      "entropy": 5.834456062316894,
      "epoch": 5.284453377610767,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002578220378295724,
      "loss": 5.192,
      "mean_token_accuracy": 0.2180214300751686,
      "num_tokens": 122524251.0,
      "step": 60470
    },
    {
      "entropy": 5.852326726913452,
      "epoch": 5.284890325963471,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002577909678448497,
      "loss": 5.2362,
      "mean_token_accuracy": 0.21656683087348938,
      "num_tokens": 122534451.0,
      "step": 60475
    },
    {
      "entropy": 5.868758964538574,
      "epoch": 5.285327274316176,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025775989819020535,
      "loss": 5.1583,
      "mean_token_accuracy": 0.22664730101823807,
      "num_tokens": 122544179.0,
      "step": 60480
    },
    {
      "entropy": 5.8687645435333256,
      "epoch": 5.28576422266888,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002577288288662351,
      "loss": 5.1919,
      "mean_token_accuracy": 0.2130729839205742,
      "num_tokens": 122552831.0,
      "step": 60485
    },
    {
      "entropy": 5.801574659347534,
      "epoch": 5.2862011710215855,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000257697759873535,
      "loss": 5.1907,
      "mean_token_accuracy": 0.21813541948795317,
      "num_tokens": 122561676.0,
      "step": 60490
    },
    {
      "entropy": 5.77241358757019,
      "epoch": 5.28663811937429,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000257666691212701,
      "loss": 5.1622,
      "mean_token_accuracy": 0.21662204414606095,
      "num_tokens": 122572177.0,
      "step": 60495
    },
    {
      "entropy": 5.88501091003418,
      "epoch": 5.287075067726994,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002576356228843289,
      "loss": 5.2084,
      "mean_token_accuracy": 0.20549858957529069,
      "num_tokens": 122582089.0,
      "step": 60500
    },
    {
      "entropy": 5.8590563297271725,
      "epoch": 5.2875120160797,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002576045548890146,
      "loss": 5.1443,
      "mean_token_accuracy": 0.22499787360429763,
      "num_tokens": 122592365.0,
      "step": 60505
    },
    {
      "entropy": 5.912256860733033,
      "epoch": 5.287948964432404,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002575734872273541,
      "loss": 5.2547,
      "mean_token_accuracy": 0.2099597692489624,
      "num_tokens": 122602909.0,
      "step": 60510
    },
    {
      "entropy": 5.879155206680298,
      "epoch": 5.288385912785109,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00025754241989994325,
      "loss": 5.2301,
      "mean_token_accuracy": 0.21656302362680435,
      "num_tokens": 122613331.0,
      "step": 60515
    },
    {
      "entropy": 5.863555908203125,
      "epoch": 5.288822861137813,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025751135290737794,
      "loss": 5.152,
      "mean_token_accuracy": 0.22037249356508254,
      "num_tokens": 122623240.0,
      "step": 60520
    },
    {
      "entropy": 5.835490083694458,
      "epoch": 5.2892598094905185,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000257480286250254,
      "loss": 5.2004,
      "mean_token_accuracy": 0.2202698439359665,
      "num_tokens": 122633122.0,
      "step": 60525
    },
    {
      "entropy": 5.8141396045684814,
      "epoch": 5.289696757843223,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00025744921992916736,
      "loss": 5.2224,
      "mean_token_accuracy": 0.21693599671125413,
      "num_tokens": 122643360.0,
      "step": 60530
    },
    {
      "entropy": 5.87205057144165,
      "epoch": 5.2901337061959275,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002574181539447139,
      "loss": 5.1681,
      "mean_token_accuracy": 0.21734708547592163,
      "num_tokens": 122653919.0,
      "step": 60535
    },
    {
      "entropy": 5.876134443283081,
      "epoch": 5.290570654548632,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025738708829748947,
      "loss": 5.186,
      "mean_token_accuracy": 0.2238428235054016,
      "num_tokens": 122662772.0,
      "step": 60540
    },
    {
      "entropy": 5.878548812866211,
      "epoch": 5.291007602901337,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00025735602298808993,
      "loss": 5.1934,
      "mean_token_accuracy": 0.22071337550878525,
      "num_tokens": 122673841.0,
      "step": 60545
    },
    {
      "entropy": 5.894378042221069,
      "epoch": 5.291444551254042,
      "grad_norm": 1.125,
      "learning_rate": 0.000257324958017111,
      "loss": 5.2878,
      "mean_token_accuracy": 0.2112657830119133,
      "num_tokens": 122683951.0,
      "step": 60550
    },
    {
      "entropy": 5.873309659957886,
      "epoch": 5.291881499606746,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00025729389338514875,
      "loss": 5.1872,
      "mean_token_accuracy": 0.21717616617679597,
      "num_tokens": 122693907.0,
      "step": 60555
    },
    {
      "entropy": 5.842406129837036,
      "epoch": 5.292318447959451,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002572628290927989,
      "loss": 5.1119,
      "mean_token_accuracy": 0.22946038693189622,
      "num_tokens": 122704060.0,
      "step": 60560
    },
    {
      "entropy": 5.8483811378479,
      "epoch": 5.292755396312156,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002572317651406573,
      "loss": 5.2471,
      "mean_token_accuracy": 0.20557643473148346,
      "num_tokens": 122715754.0,
      "step": 60565
    },
    {
      "entropy": 5.850751066207886,
      "epoch": 5.2931923446648605,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002572007015293197,
      "loss": 5.2196,
      "mean_token_accuracy": 0.21568377763032914,
      "num_tokens": 122726031.0,
      "step": 60570
    },
    {
      "entropy": 5.9398236751556395,
      "epoch": 5.293629293017565,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00025716963825938194,
      "loss": 5.2564,
      "mean_token_accuracy": 0.21251377761363982,
      "num_tokens": 122736730.0,
      "step": 60575
    },
    {
      "entropy": 5.84017014503479,
      "epoch": 5.29406624137027,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002571385753314399,
      "loss": 5.2449,
      "mean_token_accuracy": 0.2092633828520775,
      "num_tokens": 122746399.0,
      "step": 60580
    },
    {
      "entropy": 5.854525470733643,
      "epoch": 5.294503189722975,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002571075127460893,
      "loss": 5.1521,
      "mean_token_accuracy": 0.21465857028961183,
      "num_tokens": 122756612.0,
      "step": 60585
    },
    {
      "entropy": 5.871263694763184,
      "epoch": 5.294940138075679,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025707645050392596,
      "loss": 5.2292,
      "mean_token_accuracy": 0.21401250660419463,
      "num_tokens": 122766789.0,
      "step": 60590
    },
    {
      "entropy": 5.90920877456665,
      "epoch": 5.295377086428384,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025704538860554576,
      "loss": 5.2849,
      "mean_token_accuracy": 0.20419915318489074,
      "num_tokens": 122776912.0,
      "step": 60595
    },
    {
      "entropy": 5.866057014465332,
      "epoch": 5.295814034781089,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025701432705154445,
      "loss": 5.1662,
      "mean_token_accuracy": 0.21447132229804994,
      "num_tokens": 122786509.0,
      "step": 60600
    },
    {
      "entropy": 5.904999876022339,
      "epoch": 5.296250983133794,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002569832658425177,
      "loss": 5.2517,
      "mean_token_accuracy": 0.21132341027259827,
      "num_tokens": 122797092.0,
      "step": 60605
    },
    {
      "entropy": 5.859048509597779,
      "epoch": 5.296687931486498,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002569522049790614,
      "loss": 5.1634,
      "mean_token_accuracy": 0.2215537518262863,
      "num_tokens": 122807708.0,
      "step": 60610
    },
    {
      "entropy": 5.862647819519043,
      "epoch": 5.2971248798392025,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002569211444617713,
      "loss": 5.2091,
      "mean_token_accuracy": 0.2215864434838295,
      "num_tokens": 122817963.0,
      "step": 60615
    },
    {
      "entropy": 5.880216455459594,
      "epoch": 5.297561828191908,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025689008429124305,
      "loss": 5.255,
      "mean_token_accuracy": 0.209730327129364,
      "num_tokens": 122826882.0,
      "step": 60620
    },
    {
      "entropy": 5.869483709335327,
      "epoch": 5.297998776544612,
      "grad_norm": 1.0625,
      "learning_rate": 0.00025685902446807244,
      "loss": 5.1339,
      "mean_token_accuracy": 0.2244601234793663,
      "num_tokens": 122836910.0,
      "step": 60625
    },
    {
      "entropy": 5.852020692825318,
      "epoch": 5.298435724897317,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025682796499285537,
      "loss": 5.2017,
      "mean_token_accuracy": 0.2219776302576065,
      "num_tokens": 122846490.0,
      "step": 60630
    },
    {
      "entropy": 5.85709137916565,
      "epoch": 5.298872673250022,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025679690586618735,
      "loss": 5.2189,
      "mean_token_accuracy": 0.21801847517490386,
      "num_tokens": 122856568.0,
      "step": 60635
    },
    {
      "entropy": 5.904139804840088,
      "epoch": 5.299309621602727,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002567658470886643,
      "loss": 5.3013,
      "mean_token_accuracy": 0.20360370576381684,
      "num_tokens": 122866164.0,
      "step": 60640
    },
    {
      "entropy": 5.861127614974976,
      "epoch": 5.299746569955431,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025673478866088177,
      "loss": 5.1875,
      "mean_token_accuracy": 0.22219510674476622,
      "num_tokens": 122875673.0,
      "step": 60645
    },
    {
      "entropy": 5.869850587844849,
      "epoch": 5.300183518308136,
      "grad_norm": 1.0625,
      "learning_rate": 0.00025670373058343565,
      "loss": 5.2096,
      "mean_token_accuracy": 0.21625005155801774,
      "num_tokens": 122886745.0,
      "step": 60650
    },
    {
      "entropy": 5.917657136917114,
      "epoch": 5.300620466660841,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002566726728569215,
      "loss": 5.2506,
      "mean_token_accuracy": 0.20992915481328964,
      "num_tokens": 122897500.0,
      "step": 60655
    },
    {
      "entropy": 5.897189569473267,
      "epoch": 5.301057415013545,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002566416154819351,
      "loss": 5.2463,
      "mean_token_accuracy": 0.21100812256336213,
      "num_tokens": 122907256.0,
      "step": 60660
    },
    {
      "entropy": 5.843786573410034,
      "epoch": 5.30149436336625,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002566105584590721,
      "loss": 5.1755,
      "mean_token_accuracy": 0.2194210782647133,
      "num_tokens": 122916866.0,
      "step": 60665
    },
    {
      "entropy": 5.882147312164307,
      "epoch": 5.301931311718954,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002565795017889283,
      "loss": 5.2435,
      "mean_token_accuracy": 0.2102186992764473,
      "num_tokens": 122928115.0,
      "step": 60670
    },
    {
      "entropy": 5.846619129180908,
      "epoch": 5.30236826007166,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025654844547209926,
      "loss": 5.1636,
      "mean_token_accuracy": 0.2109671115875244,
      "num_tokens": 122937957.0,
      "step": 60675
    },
    {
      "entropy": 5.86161527633667,
      "epoch": 5.302805208424364,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025651738950918077,
      "loss": 5.1311,
      "mean_token_accuracy": 0.22744261771440505,
      "num_tokens": 122947393.0,
      "step": 60680
    },
    {
      "entropy": 5.867906379699707,
      "epoch": 5.303242156777069,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025648633390076836,
      "loss": 5.1884,
      "mean_token_accuracy": 0.214944751560688,
      "num_tokens": 122956669.0,
      "step": 60685
    },
    {
      "entropy": 5.837695550918579,
      "epoch": 5.303679105129774,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00025645527864745775,
      "loss": 5.1836,
      "mean_token_accuracy": 0.218546587228775,
      "num_tokens": 122966482.0,
      "step": 60690
    },
    {
      "entropy": 5.835694217681885,
      "epoch": 5.3041160534824785,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002564242237498447,
      "loss": 5.2288,
      "mean_token_accuracy": 0.2200414150953293,
      "num_tokens": 122976177.0,
      "step": 60695
    },
    {
      "entropy": 5.828300046920776,
      "epoch": 5.304553001835183,
      "grad_norm": 1.0625,
      "learning_rate": 0.00025639316920852474,
      "loss": 5.1306,
      "mean_token_accuracy": 0.22064952552318573,
      "num_tokens": 122985569.0,
      "step": 60700
    },
    {
      "entropy": 5.806789684295654,
      "epoch": 5.304989950187887,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002563621150240935,
      "loss": 5.1176,
      "mean_token_accuracy": 0.22097227722406387,
      "num_tokens": 122995023.0,
      "step": 60705
    },
    {
      "entropy": 5.789667129516602,
      "epoch": 5.305426898540593,
      "grad_norm": 1.125,
      "learning_rate": 0.0002563310611971467,
      "loss": 5.2007,
      "mean_token_accuracy": 0.21697116941213607,
      "num_tokens": 123004756.0,
      "step": 60710
    },
    {
      "entropy": 5.860283374786377,
      "epoch": 5.305863846893297,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025630000772827987,
      "loss": 5.1977,
      "mean_token_accuracy": 0.2195188283920288,
      "num_tokens": 123014476.0,
      "step": 60715
    },
    {
      "entropy": 5.884975957870483,
      "epoch": 5.306300795246002,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025626895461808877,
      "loss": 5.1936,
      "mean_token_accuracy": 0.21882154196500778,
      "num_tokens": 123024750.0,
      "step": 60720
    },
    {
      "entropy": 5.934785461425781,
      "epoch": 5.306737743598706,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025623790186716887,
      "loss": 5.2691,
      "mean_token_accuracy": 0.21273654699325562,
      "num_tokens": 123035583.0,
      "step": 60725
    },
    {
      "entropy": 5.878716039657593,
      "epoch": 5.307174691951412,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002562068494761159,
      "loss": 5.2737,
      "mean_token_accuracy": 0.20665253698825836,
      "num_tokens": 123045664.0,
      "step": 60730
    },
    {
      "entropy": 5.822194576263428,
      "epoch": 5.307611640304116,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002561757974455254,
      "loss": 5.1708,
      "mean_token_accuracy": 0.2170628562569618,
      "num_tokens": 123055539.0,
      "step": 60735
    },
    {
      "entropy": 5.854221630096435,
      "epoch": 5.3080485886568205,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002561447457759929,
      "loss": 5.1916,
      "mean_token_accuracy": 0.21111034452915192,
      "num_tokens": 123065295.0,
      "step": 60740
    },
    {
      "entropy": 5.9285359382629395,
      "epoch": 5.308485537009526,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002561136944681141,
      "loss": 5.2656,
      "mean_token_accuracy": 0.20916201472282409,
      "num_tokens": 123075836.0,
      "step": 60745
    },
    {
      "entropy": 5.923833274841309,
      "epoch": 5.30892248536223,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002560826435224846,
      "loss": 5.2401,
      "mean_token_accuracy": 0.21316773295402527,
      "num_tokens": 123086277.0,
      "step": 60750
    },
    {
      "entropy": 5.877942371368408,
      "epoch": 5.309359433714935,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002560515929396998,
      "loss": 5.1792,
      "mean_token_accuracy": 0.22613172382116317,
      "num_tokens": 123095954.0,
      "step": 60755
    },
    {
      "entropy": 5.901020097732544,
      "epoch": 5.309796382067639,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025602054272035545,
      "loss": 5.2657,
      "mean_token_accuracy": 0.21213342249393463,
      "num_tokens": 123105128.0,
      "step": 60760
    },
    {
      "entropy": 5.782530832290649,
      "epoch": 5.310233330420345,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000255989492865047,
      "loss": 5.1328,
      "mean_token_accuracy": 0.21912401914596558,
      "num_tokens": 123114233.0,
      "step": 60765
    },
    {
      "entropy": 5.823461055755615,
      "epoch": 5.310670278773049,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025595844337437,
      "loss": 5.2257,
      "mean_token_accuracy": 0.21463125199079514,
      "num_tokens": 123124302.0,
      "step": 60770
    },
    {
      "entropy": 5.875054502487183,
      "epoch": 5.311107227125754,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002559273942489202,
      "loss": 5.2323,
      "mean_token_accuracy": 0.2090788319706917,
      "num_tokens": 123134647.0,
      "step": 60775
    },
    {
      "entropy": 5.9542930126190186,
      "epoch": 5.311544175478458,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002558963454892928,
      "loss": 5.2351,
      "mean_token_accuracy": 0.21143606752157212,
      "num_tokens": 123143880.0,
      "step": 60780
    },
    {
      "entropy": 5.896210241317749,
      "epoch": 5.311981123831163,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002558652970960836,
      "loss": 5.2208,
      "mean_token_accuracy": 0.21611683964729309,
      "num_tokens": 123153856.0,
      "step": 60785
    },
    {
      "entropy": 5.939654397964477,
      "epoch": 5.312418072183868,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025583424906988795,
      "loss": 5.315,
      "mean_token_accuracy": 0.20966374427080153,
      "num_tokens": 123164153.0,
      "step": 60790
    },
    {
      "entropy": 5.797943592071533,
      "epoch": 5.312855020536572,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002558032014113015,
      "loss": 5.1208,
      "mean_token_accuracy": 0.22567618638277054,
      "num_tokens": 123173953.0,
      "step": 60795
    },
    {
      "entropy": 5.818247222900391,
      "epoch": 5.313291968889278,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002557721541209197,
      "loss": 5.2045,
      "mean_token_accuracy": 0.22029811441898345,
      "num_tokens": 123185427.0,
      "step": 60800
    },
    {
      "entropy": 5.693183088302613,
      "epoch": 5.313728917241982,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025574110719933803,
      "loss": 5.0504,
      "mean_token_accuracy": 0.23972921222448348,
      "num_tokens": 123196038.0,
      "step": 60805
    },
    {
      "entropy": 5.888056707382202,
      "epoch": 5.314165865594687,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025571006064715204,
      "loss": 5.2608,
      "mean_token_accuracy": 0.21401135623455048,
      "num_tokens": 123206032.0,
      "step": 60810
    },
    {
      "entropy": 5.977963972091675,
      "epoch": 5.314602813947391,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025567901446495715,
      "loss": 5.2325,
      "mean_token_accuracy": 0.21187733262777328,
      "num_tokens": 123214918.0,
      "step": 60815
    },
    {
      "entropy": 5.905537033081055,
      "epoch": 5.3150397623000964,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000255647968653349,
      "loss": 5.232,
      "mean_token_accuracy": 0.21391105502843857,
      "num_tokens": 123225317.0,
      "step": 60820
    },
    {
      "entropy": 5.860505771636963,
      "epoch": 5.315476710652801,
      "grad_norm": 1.125,
      "learning_rate": 0.00025561692321292284,
      "loss": 5.2541,
      "mean_token_accuracy": 0.2066854178905487,
      "num_tokens": 123235989.0,
      "step": 60825
    },
    {
      "entropy": 5.842749214172363,
      "epoch": 5.315913659005505,
      "grad_norm": 1.078125,
      "learning_rate": 0.00025558587814427427,
      "loss": 5.1524,
      "mean_token_accuracy": 0.22029493600130082,
      "num_tokens": 123247071.0,
      "step": 60830
    },
    {
      "entropy": 5.8542724609375,
      "epoch": 5.31635060735821,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002555548334479987,
      "loss": 5.159,
      "mean_token_accuracy": 0.21926596164703369,
      "num_tokens": 123257565.0,
      "step": 60835
    },
    {
      "entropy": 5.786494970321655,
      "epoch": 5.316787555710915,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002555237891246917,
      "loss": 5.1189,
      "mean_token_accuracy": 0.222915418446064,
      "num_tokens": 123268949.0,
      "step": 60840
    },
    {
      "entropy": 5.840277528762817,
      "epoch": 5.31722450406362,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002554927451749486,
      "loss": 5.1813,
      "mean_token_accuracy": 0.21547521650791168,
      "num_tokens": 123279743.0,
      "step": 60845
    },
    {
      "entropy": 5.910766124725342,
      "epoch": 5.317661452416324,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002554617015993649,
      "loss": 5.2013,
      "mean_token_accuracy": 0.21113799065351485,
      "num_tokens": 123289256.0,
      "step": 60850
    },
    {
      "entropy": 5.817729902267456,
      "epoch": 5.3180984007690295,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002554306583985359,
      "loss": 5.1393,
      "mean_token_accuracy": 0.2253887891769409,
      "num_tokens": 123299834.0,
      "step": 60855
    },
    {
      "entropy": 5.857950496673584,
      "epoch": 5.318535349121734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025539961557305717,
      "loss": 5.2514,
      "mean_token_accuracy": 0.21063498854637147,
      "num_tokens": 123310290.0,
      "step": 60860
    },
    {
      "entropy": 5.881827068328858,
      "epoch": 5.3189722974744384,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025536857312352413,
      "loss": 5.1597,
      "mean_token_accuracy": 0.21349766999483108,
      "num_tokens": 123320286.0,
      "step": 60865
    },
    {
      "entropy": 5.879464101791382,
      "epoch": 5.319409245827143,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002553375310505321,
      "loss": 5.2417,
      "mean_token_accuracy": 0.21327976137399673,
      "num_tokens": 123331035.0,
      "step": 60870
    },
    {
      "entropy": 5.873933458328247,
      "epoch": 5.319846194179848,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002553064893546766,
      "loss": 5.2218,
      "mean_token_accuracy": 0.21234396547079087,
      "num_tokens": 123341090.0,
      "step": 60875
    },
    {
      "entropy": 5.888038587570191,
      "epoch": 5.320283142532553,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002552754480365529,
      "loss": 5.2273,
      "mean_token_accuracy": 0.21380477100610734,
      "num_tokens": 123351411.0,
      "step": 60880
    },
    {
      "entropy": 5.942503786087036,
      "epoch": 5.320720090885257,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002552444070967564,
      "loss": 5.2403,
      "mean_token_accuracy": 0.20870566964149476,
      "num_tokens": 123361473.0,
      "step": 60885
    },
    {
      "entropy": 5.889030694961548,
      "epoch": 5.321157039237962,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025521336653588255,
      "loss": 5.2313,
      "mean_token_accuracy": 0.21792213320732118,
      "num_tokens": 123371331.0,
      "step": 60890
    },
    {
      "entropy": 5.915173053741455,
      "epoch": 5.321593987590667,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002551823263545267,
      "loss": 5.2928,
      "mean_token_accuracy": 0.2040253534913063,
      "num_tokens": 123382057.0,
      "step": 60895
    },
    {
      "entropy": 5.899051570892334,
      "epoch": 5.3220309359433715,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025515128655328426,
      "loss": 5.2149,
      "mean_token_accuracy": 0.21755172163248063,
      "num_tokens": 123391580.0,
      "step": 60900
    },
    {
      "entropy": 5.84075174331665,
      "epoch": 5.322467884296076,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002551202471327505,
      "loss": 5.1975,
      "mean_token_accuracy": 0.22139672487974166,
      "num_tokens": 123401773.0,
      "step": 60905
    },
    {
      "entropy": 5.892189645767212,
      "epoch": 5.322904832648781,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025508920809352083,
      "loss": 5.292,
      "mean_token_accuracy": 0.19900889992713927,
      "num_tokens": 123411476.0,
      "step": 60910
    },
    {
      "entropy": 5.838903141021729,
      "epoch": 5.323341781001486,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002550581694361905,
      "loss": 5.1473,
      "mean_token_accuracy": 0.2184552639722824,
      "num_tokens": 123420924.0,
      "step": 60915
    },
    {
      "entropy": 5.885335159301758,
      "epoch": 5.32377872935419,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025502713116135506,
      "loss": 5.2682,
      "mean_token_accuracy": 0.21244069933891296,
      "num_tokens": 123430556.0,
      "step": 60920
    },
    {
      "entropy": 5.9266613006591795,
      "epoch": 5.324215677706895,
      "grad_norm": 1.078125,
      "learning_rate": 0.00025499609326960966,
      "loss": 5.3118,
      "mean_token_accuracy": 0.21322576105594634,
      "num_tokens": 123440589.0,
      "step": 60925
    },
    {
      "entropy": 5.893042135238647,
      "epoch": 5.3246526260596,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002549650557615497,
      "loss": 5.2043,
      "mean_token_accuracy": 0.21941966116428374,
      "num_tokens": 123451047.0,
      "step": 60930
    },
    {
      "entropy": 5.9017815589904785,
      "epoch": 5.325089574412305,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025493401863777037,
      "loss": 5.2221,
      "mean_token_accuracy": 0.21106606870889663,
      "num_tokens": 123462011.0,
      "step": 60935
    },
    {
      "entropy": 5.91640625,
      "epoch": 5.325526522765009,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002549029818988671,
      "loss": 5.2887,
      "mean_token_accuracy": 0.2112216904759407,
      "num_tokens": 123471691.0,
      "step": 60940
    },
    {
      "entropy": 5.88974347114563,
      "epoch": 5.3259634711177135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002548719455454352,
      "loss": 5.1654,
      "mean_token_accuracy": 0.21987870335578918,
      "num_tokens": 123482321.0,
      "step": 60945
    },
    {
      "entropy": 5.881533575057984,
      "epoch": 5.326400419470419,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002548409095780699,
      "loss": 5.2085,
      "mean_token_accuracy": 0.21626691222190858,
      "num_tokens": 123491810.0,
      "step": 60950
    },
    {
      "entropy": 5.850592184066772,
      "epoch": 5.326837367823123,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025480987399736666,
      "loss": 5.2582,
      "mean_token_accuracy": 0.2147575631737709,
      "num_tokens": 123502519.0,
      "step": 60955
    },
    {
      "entropy": 5.813942956924438,
      "epoch": 5.327274316175828,
      "grad_norm": 1.125,
      "learning_rate": 0.0002547788388039205,
      "loss": 5.1549,
      "mean_token_accuracy": 0.21928419023752213,
      "num_tokens": 123513690.0,
      "step": 60960
    },
    {
      "entropy": 5.863082456588745,
      "epoch": 5.327711264528533,
      "grad_norm": 1.125,
      "learning_rate": 0.00025474780399832684,
      "loss": 5.207,
      "mean_token_accuracy": 0.21722501814365386,
      "num_tokens": 123522703.0,
      "step": 60965
    },
    {
      "entropy": 5.896283197402954,
      "epoch": 5.328148212881238,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025471676958118084,
      "loss": 5.1975,
      "mean_token_accuracy": 0.21434501856565474,
      "num_tokens": 123532286.0,
      "step": 60970
    },
    {
      "entropy": 5.878910207748413,
      "epoch": 5.328585161233942,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002546857355530779,
      "loss": 5.255,
      "mean_token_accuracy": 0.21474431157112123,
      "num_tokens": 123541772.0,
      "step": 60975
    },
    {
      "entropy": 5.811774015426636,
      "epoch": 5.329022109586647,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00025465470191461323,
      "loss": 5.1407,
      "mean_token_accuracy": 0.22771652191877365,
      "num_tokens": 123551979.0,
      "step": 60980
    },
    {
      "entropy": 5.860636806488037,
      "epoch": 5.329459057939352,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025462366866638197,
      "loss": 5.3006,
      "mean_token_accuracy": 0.21153673380613328,
      "num_tokens": 123562275.0,
      "step": 60985
    },
    {
      "entropy": 5.860629415512085,
      "epoch": 5.329896006292056,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002545926358089794,
      "loss": 5.1968,
      "mean_token_accuracy": 0.21663876473903657,
      "num_tokens": 123571919.0,
      "step": 60990
    },
    {
      "entropy": 5.925621080398559,
      "epoch": 5.330332954644761,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002545616033430008,
      "loss": 5.2321,
      "mean_token_accuracy": 0.21534604877233504,
      "num_tokens": 123582543.0,
      "step": 60995
    },
    {
      "entropy": 5.853824329376221,
      "epoch": 5.330769902997465,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002545305712690414,
      "loss": 5.2139,
      "mean_token_accuracy": 0.2212715893983841,
      "num_tokens": 123592987.0,
      "step": 61000
    },
    {
      "entropy": 5.855568981170654,
      "epoch": 5.331206851350171,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002544995395876964,
      "loss": 5.1785,
      "mean_token_accuracy": 0.21695187240839003,
      "num_tokens": 123602466.0,
      "step": 61005
    },
    {
      "entropy": 5.886905097961426,
      "epoch": 5.331643799702875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025446850829956085,
      "loss": 5.1882,
      "mean_token_accuracy": 0.21552692949771882,
      "num_tokens": 123612895.0,
      "step": 61010
    },
    {
      "entropy": 5.867119979858399,
      "epoch": 5.33208074805558,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025443747740523014,
      "loss": 5.2664,
      "mean_token_accuracy": 0.2140299618244171,
      "num_tokens": 123624405.0,
      "step": 61015
    },
    {
      "entropy": 5.836073160171509,
      "epoch": 5.332517696408285,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002544064469052993,
      "loss": 5.1785,
      "mean_token_accuracy": 0.22240012735128403,
      "num_tokens": 123633982.0,
      "step": 61020
    },
    {
      "entropy": 5.850583124160766,
      "epoch": 5.3329546447609895,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002543754168003637,
      "loss": 5.2011,
      "mean_token_accuracy": 0.22229291051626204,
      "num_tokens": 123644516.0,
      "step": 61025
    },
    {
      "entropy": 5.819261837005615,
      "epoch": 5.333391593113694,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002543443870910184,
      "loss": 5.2012,
      "mean_token_accuracy": 0.22347129732370377,
      "num_tokens": 123655634.0,
      "step": 61030
    },
    {
      "entropy": 5.880191421508789,
      "epoch": 5.333828541466398,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002543133577778586,
      "loss": 5.1594,
      "mean_token_accuracy": 0.22426874786615372,
      "num_tokens": 123665560.0,
      "step": 61035
    },
    {
      "entropy": 5.806635856628418,
      "epoch": 5.334265489819104,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002542823288614794,
      "loss": 5.1724,
      "mean_token_accuracy": 0.22596853524446486,
      "num_tokens": 123675136.0,
      "step": 61040
    },
    {
      "entropy": 5.8526890754699705,
      "epoch": 5.334702438171808,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025425130034247603,
      "loss": 5.182,
      "mean_token_accuracy": 0.21674935817718505,
      "num_tokens": 123684730.0,
      "step": 61045
    },
    {
      "entropy": 5.883206415176391,
      "epoch": 5.335139386524513,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002542202722214435,
      "loss": 5.2394,
      "mean_token_accuracy": 0.21695251166820526,
      "num_tokens": 123694636.0,
      "step": 61050
    },
    {
      "entropy": 5.891243648529053,
      "epoch": 5.335576334877217,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002541892444989771,
      "loss": 5.2227,
      "mean_token_accuracy": 0.2071736767888069,
      "num_tokens": 123704352.0,
      "step": 61055
    },
    {
      "entropy": 5.902839851379395,
      "epoch": 5.3360132832299225,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002541582171756719,
      "loss": 5.2313,
      "mean_token_accuracy": 0.21297977715730668,
      "num_tokens": 123714425.0,
      "step": 61060
    },
    {
      "entropy": 5.871789836883545,
      "epoch": 5.336450231582627,
      "grad_norm": 1.078125,
      "learning_rate": 0.000254127190252123,
      "loss": 5.1681,
      "mean_token_accuracy": 0.2178148150444031,
      "num_tokens": 123724626.0,
      "step": 61065
    },
    {
      "entropy": 5.850432682037353,
      "epoch": 5.3368871799353315,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025409616372892553,
      "loss": 5.2121,
      "mean_token_accuracy": 0.2139788493514061,
      "num_tokens": 123734738.0,
      "step": 61070
    },
    {
      "entropy": 5.824569892883301,
      "epoch": 5.337324128288037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002540651376066745,
      "loss": 5.1603,
      "mean_token_accuracy": 0.22604036778211595,
      "num_tokens": 123745154.0,
      "step": 61075
    },
    {
      "entropy": 5.82666425704956,
      "epoch": 5.337761076640741,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002540341118859653,
      "loss": 5.148,
      "mean_token_accuracy": 0.2181192934513092,
      "num_tokens": 123755603.0,
      "step": 61080
    },
    {
      "entropy": 5.886415815353393,
      "epoch": 5.338198024993446,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025400308656739267,
      "loss": 5.1795,
      "mean_token_accuracy": 0.2214914232492447,
      "num_tokens": 123765023.0,
      "step": 61085
    },
    {
      "entropy": 5.870741701126098,
      "epoch": 5.33863497334615,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002539720616515519,
      "loss": 5.2094,
      "mean_token_accuracy": 0.2210501566529274,
      "num_tokens": 123775033.0,
      "step": 61090
    },
    {
      "entropy": 5.86631441116333,
      "epoch": 5.339071921698856,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000253941037139038,
      "loss": 5.3082,
      "mean_token_accuracy": 0.21061419397592546,
      "num_tokens": 123786105.0,
      "step": 61095
    },
    {
      "entropy": 5.848397970199585,
      "epoch": 5.33950887005156,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025391001303044607,
      "loss": 5.2206,
      "mean_token_accuracy": 0.2173908993601799,
      "num_tokens": 123795639.0,
      "step": 61100
    },
    {
      "entropy": 5.89074673652649,
      "epoch": 5.3399458184042645,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002538789893263711,
      "loss": 5.2698,
      "mean_token_accuracy": 0.21302756071090698,
      "num_tokens": 123805728.0,
      "step": 61105
    },
    {
      "entropy": 5.853855180740356,
      "epoch": 5.340382766756969,
      "grad_norm": 1.03125,
      "learning_rate": 0.00025384796602740816,
      "loss": 5.1883,
      "mean_token_accuracy": 0.2194962501525879,
      "num_tokens": 123816239.0,
      "step": 61110
    },
    {
      "entropy": 5.894834804534912,
      "epoch": 5.340819715109674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00025381694313415234,
      "loss": 5.2053,
      "mean_token_accuracy": 0.22072957903146745,
      "num_tokens": 123826375.0,
      "step": 61115
    },
    {
      "entropy": 5.931826782226563,
      "epoch": 5.341256663462379,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025378592064719864,
      "loss": 5.2402,
      "mean_token_accuracy": 0.210785810649395,
      "num_tokens": 123836565.0,
      "step": 61120
    },
    {
      "entropy": 5.857781791687012,
      "epoch": 5.341693611815083,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025375489856714216,
      "loss": 5.1851,
      "mean_token_accuracy": 0.2185870587825775,
      "num_tokens": 123848031.0,
      "step": 61125
    },
    {
      "entropy": 5.890960359573365,
      "epoch": 5.342130560167788,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025372387689457775,
      "loss": 5.2567,
      "mean_token_accuracy": 0.21225516051054,
      "num_tokens": 123858467.0,
      "step": 61130
    },
    {
      "entropy": 5.876202726364136,
      "epoch": 5.342567508520493,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002536928556301006,
      "loss": 5.2744,
      "mean_token_accuracy": 0.20789990723133087,
      "num_tokens": 123868874.0,
      "step": 61135
    },
    {
      "entropy": 5.772492551803589,
      "epoch": 5.343004456873198,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025366183477430564,
      "loss": 5.1182,
      "mean_token_accuracy": 0.23607994318008424,
      "num_tokens": 123880781.0,
      "step": 61140
    },
    {
      "entropy": 5.947469282150268,
      "epoch": 5.343441405225902,
      "grad_norm": 1.0625,
      "learning_rate": 0.00025363081432778784,
      "loss": 5.2647,
      "mean_token_accuracy": 0.21060526967048646,
      "num_tokens": 123891636.0,
      "step": 61145
    },
    {
      "entropy": 5.909274053573609,
      "epoch": 5.343878353578607,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025359979429114227,
      "loss": 5.2068,
      "mean_token_accuracy": 0.21453756392002105,
      "num_tokens": 123902224.0,
      "step": 61150
    },
    {
      "entropy": 5.837630033493042,
      "epoch": 5.344315301931312,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002535687746649638,
      "loss": 5.1466,
      "mean_token_accuracy": 0.2230236902832985,
      "num_tokens": 123911322.0,
      "step": 61155
    },
    {
      "entropy": 5.801965856552124,
      "epoch": 5.344752250284016,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002535377554498475,
      "loss": 5.1498,
      "mean_token_accuracy": 0.21448559314012527,
      "num_tokens": 123921338.0,
      "step": 61160
    },
    {
      "entropy": 5.853262853622437,
      "epoch": 5.345189198636721,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002535067366463883,
      "loss": 5.1951,
      "mean_token_accuracy": 0.2153836876153946,
      "num_tokens": 123930873.0,
      "step": 61165
    },
    {
      "entropy": 5.861759901046753,
      "epoch": 5.345626146989426,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002534757182551811,
      "loss": 5.1907,
      "mean_token_accuracy": 0.2203558251261711,
      "num_tokens": 123940275.0,
      "step": 61170
    },
    {
      "entropy": 5.819519424438477,
      "epoch": 5.346063095342131,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025344470027682104,
      "loss": 5.1545,
      "mean_token_accuracy": 0.2219121888279915,
      "num_tokens": 123950407.0,
      "step": 61175
    },
    {
      "entropy": 5.8479491710662845,
      "epoch": 5.346500043694835,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002534136827119028,
      "loss": 5.2094,
      "mean_token_accuracy": 0.2157697334885597,
      "num_tokens": 123960742.0,
      "step": 61180
    },
    {
      "entropy": 5.850207185745239,
      "epoch": 5.34693699204754,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002533826655610214,
      "loss": 5.1731,
      "mean_token_accuracy": 0.22133970111608506,
      "num_tokens": 123970423.0,
      "step": 61185
    },
    {
      "entropy": 5.864758586883545,
      "epoch": 5.347373940400245,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002533516488247719,
      "loss": 5.2351,
      "mean_token_accuracy": 0.20844423174858093,
      "num_tokens": 123981059.0,
      "step": 61190
    },
    {
      "entropy": 5.838087129592895,
      "epoch": 5.347810888752949,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002533206325037491,
      "loss": 5.1893,
      "mean_token_accuracy": 0.21985120922327042,
      "num_tokens": 123991201.0,
      "step": 61195
    },
    {
      "entropy": 5.881005001068115,
      "epoch": 5.348247837105654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025328961659854796,
      "loss": 5.239,
      "mean_token_accuracy": 0.2083499938249588,
      "num_tokens": 124001454.0,
      "step": 61200
    },
    {
      "entropy": 5.823829698562622,
      "epoch": 5.348684785458359,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002532586011097632,
      "loss": 5.1447,
      "mean_token_accuracy": 0.22244221568107606,
      "num_tokens": 124011846.0,
      "step": 61205
    },
    {
      "entropy": 5.85544056892395,
      "epoch": 5.349121733811064,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025322758603799,
      "loss": 5.1878,
      "mean_token_accuracy": 0.21580157727003096,
      "num_tokens": 124022623.0,
      "step": 61210
    },
    {
      "entropy": 5.8796610832214355,
      "epoch": 5.349558682163768,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002531965713838231,
      "loss": 5.2422,
      "mean_token_accuracy": 0.20790733993053437,
      "num_tokens": 124033234.0,
      "step": 61215
    },
    {
      "entropy": 5.879020929336548,
      "epoch": 5.349995630516473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025316555714785733,
      "loss": 5.2173,
      "mean_token_accuracy": 0.21309770345687867,
      "num_tokens": 124042053.0,
      "step": 61220
    },
    {
      "entropy": 5.8831134796142575,
      "epoch": 5.350432578869178,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025313454333068764,
      "loss": 5.2316,
      "mean_token_accuracy": 0.2221636727452278,
      "num_tokens": 124052373.0,
      "step": 61225
    },
    {
      "entropy": 5.91700758934021,
      "epoch": 5.3508695272218825,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00025310352993290893,
      "loss": 5.1943,
      "mean_token_accuracy": 0.21688240319490432,
      "num_tokens": 124062442.0,
      "step": 61230
    },
    {
      "entropy": 5.870856094360351,
      "epoch": 5.351306475574587,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002530725169551159,
      "loss": 5.264,
      "mean_token_accuracy": 0.21291148960590361,
      "num_tokens": 124073625.0,
      "step": 61235
    },
    {
      "entropy": 5.857592821121216,
      "epoch": 5.351743423927291,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025304150439790345,
      "loss": 5.2398,
      "mean_token_accuracy": 0.21618484556674958,
      "num_tokens": 124083867.0,
      "step": 61240
    },
    {
      "entropy": 5.870092964172363,
      "epoch": 5.352180372279997,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025301049226186654,
      "loss": 5.15,
      "mean_token_accuracy": 0.21761249005794525,
      "num_tokens": 124093479.0,
      "step": 61245
    },
    {
      "entropy": 5.822082042694092,
      "epoch": 5.352617320632701,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002529794805475999,
      "loss": 5.235,
      "mean_token_accuracy": 0.21368245631456376,
      "num_tokens": 124105038.0,
      "step": 61250
    },
    {
      "entropy": 5.899104642868042,
      "epoch": 5.353054268985406,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002529484692556984,
      "loss": 5.1992,
      "mean_token_accuracy": 0.21809294819831848,
      "num_tokens": 124114739.0,
      "step": 61255
    },
    {
      "entropy": 5.933018636703491,
      "epoch": 5.35349121733811,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025291745838675674,
      "loss": 5.308,
      "mean_token_accuracy": 0.21069658249616624,
      "num_tokens": 124124283.0,
      "step": 61260
    },
    {
      "entropy": 5.922083759307862,
      "epoch": 5.3539281656908155,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002528864479413698,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20924428552389146,
      "num_tokens": 124134547.0,
      "step": 61265
    },
    {
      "entropy": 5.78387131690979,
      "epoch": 5.35436511404352,
      "grad_norm": 1.125,
      "learning_rate": 0.0002528554379201324,
      "loss": 5.1406,
      "mean_token_accuracy": 0.2240740567445755,
      "num_tokens": 124145349.0,
      "step": 61270
    },
    {
      "entropy": 5.968286180496216,
      "epoch": 5.3548020623962245,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002528244283236393,
      "loss": 5.354,
      "mean_token_accuracy": 0.20209286361932755,
      "num_tokens": 124156632.0,
      "step": 61275
    },
    {
      "entropy": 5.8479047298431395,
      "epoch": 5.35523901074893,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002527934191524853,
      "loss": 5.1108,
      "mean_token_accuracy": 0.22611351311206818,
      "num_tokens": 124166707.0,
      "step": 61280
    },
    {
      "entropy": 5.794693088531494,
      "epoch": 5.355675959101634,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002527624104072651,
      "loss": 5.1235,
      "mean_token_accuracy": 0.22755440920591355,
      "num_tokens": 124177516.0,
      "step": 61285
    },
    {
      "entropy": 5.846703004837036,
      "epoch": 5.356112907454339,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002527314020885736,
      "loss": 5.2316,
      "mean_token_accuracy": 0.21614260375499725,
      "num_tokens": 124186979.0,
      "step": 61290
    },
    {
      "entropy": 5.87879638671875,
      "epoch": 5.356549855807043,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00025270039419700535,
      "loss": 5.1925,
      "mean_token_accuracy": 0.21231168359518052,
      "num_tokens": 124197169.0,
      "step": 61295
    },
    {
      "entropy": 5.802665710449219,
      "epoch": 5.356986804159749,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00025266938673315534,
      "loss": 5.1576,
      "mean_token_accuracy": 0.21808529645204544,
      "num_tokens": 124206308.0,
      "step": 61300
    },
    {
      "entropy": 5.786713743209839,
      "epoch": 5.357423752512453,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025263837969761814,
      "loss": 5.1322,
      "mean_token_accuracy": 0.2284994602203369,
      "num_tokens": 124217286.0,
      "step": 61305
    },
    {
      "entropy": 5.837676239013672,
      "epoch": 5.3578607008651575,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00025260737309098857,
      "loss": 5.2427,
      "mean_token_accuracy": 0.21950914412736894,
      "num_tokens": 124227485.0,
      "step": 61310
    },
    {
      "entropy": 5.915628242492676,
      "epoch": 5.358297649217862,
      "grad_norm": 1.046875,
      "learning_rate": 0.00025257636691386134,
      "loss": 5.2726,
      "mean_token_accuracy": 0.210996775329113,
      "num_tokens": 124237091.0,
      "step": 61315
    },
    {
      "entropy": 5.909593820571899,
      "epoch": 5.358734597570567,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002525453611668311,
      "loss": 5.1853,
      "mean_token_accuracy": 0.21876852512359618,
      "num_tokens": 124247369.0,
      "step": 61320
    },
    {
      "entropy": 5.913056993484497,
      "epoch": 5.359171545923272,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002525143558504926,
      "loss": 5.2608,
      "mean_token_accuracy": 0.21294804960489272,
      "num_tokens": 124255877.0,
      "step": 61325
    },
    {
      "entropy": 5.864094877243042,
      "epoch": 5.359608494275976,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025248335096544054,
      "loss": 5.2191,
      "mean_token_accuracy": 0.21591641157865524,
      "num_tokens": 124266409.0,
      "step": 61330
    },
    {
      "entropy": 5.891663074493408,
      "epoch": 5.360045442628682,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002524523465122697,
      "loss": 5.2265,
      "mean_token_accuracy": 0.21813777536153794,
      "num_tokens": 124276881.0,
      "step": 61335
    },
    {
      "entropy": 5.854880619049072,
      "epoch": 5.360482390981386,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002524213424915745,
      "loss": 5.2239,
      "mean_token_accuracy": 0.21696258336305618,
      "num_tokens": 124287979.0,
      "step": 61340
    },
    {
      "entropy": 5.86840329170227,
      "epoch": 5.360919339334091,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025239033890394984,
      "loss": 5.2132,
      "mean_token_accuracy": 0.217117078602314,
      "num_tokens": 124297551.0,
      "step": 61345
    },
    {
      "entropy": 5.858966302871704,
      "epoch": 5.361356287686795,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002523593357499904,
      "loss": 5.2387,
      "mean_token_accuracy": 0.21045951098203658,
      "num_tokens": 124308150.0,
      "step": 61350
    },
    {
      "entropy": 5.833465719223023,
      "epoch": 5.3617932360395,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025232833303029075,
      "loss": 5.2217,
      "mean_token_accuracy": 0.2077915072441101,
      "num_tokens": 124319397.0,
      "step": 61355
    },
    {
      "entropy": 5.892832851409912,
      "epoch": 5.362230184392205,
      "grad_norm": 1.03125,
      "learning_rate": 0.00025229733074544553,
      "loss": 5.1872,
      "mean_token_accuracy": 0.2159242182970047,
      "num_tokens": 124330318.0,
      "step": 61360
    },
    {
      "entropy": 5.883864498138427,
      "epoch": 5.362667132744909,
      "grad_norm": 1.046875,
      "learning_rate": 0.00025226632889604935,
      "loss": 5.2243,
      "mean_token_accuracy": 0.21798815578222275,
      "num_tokens": 124340200.0,
      "step": 61365
    },
    {
      "entropy": 5.849551773071289,
      "epoch": 5.363104081097614,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025223532748269695,
      "loss": 5.1777,
      "mean_token_accuracy": 0.22098178416490555,
      "num_tokens": 124349767.0,
      "step": 61370
    },
    {
      "entropy": 5.871389579772949,
      "epoch": 5.363541029450319,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00025220432650598285,
      "loss": 5.2268,
      "mean_token_accuracy": 0.2110409840941429,
      "num_tokens": 124360213.0,
      "step": 61375
    },
    {
      "entropy": 5.860149145126343,
      "epoch": 5.363977977803024,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025217332596650174,
      "loss": 5.1365,
      "mean_token_accuracy": 0.21279842853546144,
      "num_tokens": 124371432.0,
      "step": 61380
    },
    {
      "entropy": 5.845922327041626,
      "epoch": 5.364414926155728,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025214232586484825,
      "loss": 5.2046,
      "mean_token_accuracy": 0.21088181138038636,
      "num_tokens": 124381742.0,
      "step": 61385
    },
    {
      "entropy": 5.793655824661255,
      "epoch": 5.3648518745084335,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025211132620161684,
      "loss": 5.0677,
      "mean_token_accuracy": 0.23254308104515076,
      "num_tokens": 124391474.0,
      "step": 61390
    },
    {
      "entropy": 5.790002870559692,
      "epoch": 5.365288822861138,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00025208032697740225,
      "loss": 5.1329,
      "mean_token_accuracy": 0.22647491842508316,
      "num_tokens": 124402052.0,
      "step": 61395
    },
    {
      "entropy": 5.841784334182739,
      "epoch": 5.365725771213842,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025204932819279896,
      "loss": 5.2144,
      "mean_token_accuracy": 0.21816882342100144,
      "num_tokens": 124411636.0,
      "step": 61400
    },
    {
      "entropy": 5.860776138305664,
      "epoch": 5.366162719566547,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002520183298484017,
      "loss": 5.1878,
      "mean_token_accuracy": 0.21735899299383163,
      "num_tokens": 124421831.0,
      "step": 61405
    },
    {
      "entropy": 5.84753270149231,
      "epoch": 5.366599667919252,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002519873319448047,
      "loss": 5.169,
      "mean_token_accuracy": 0.21970075368881226,
      "num_tokens": 124431149.0,
      "step": 61410
    },
    {
      "entropy": 5.880486965179443,
      "epoch": 5.367036616271957,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002519563344826029,
      "loss": 5.1834,
      "mean_token_accuracy": 0.21580072045326232,
      "num_tokens": 124440993.0,
      "step": 61415
    },
    {
      "entropy": 5.8834394931793215,
      "epoch": 5.367473564624661,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002519253374623906,
      "loss": 5.2267,
      "mean_token_accuracy": 0.21231116950511933,
      "num_tokens": 124451337.0,
      "step": 61420
    },
    {
      "entropy": 5.807259178161621,
      "epoch": 5.367910512977366,
      "grad_norm": 1.0,
      "learning_rate": 0.00025189434088476243,
      "loss": 5.1255,
      "mean_token_accuracy": 0.22672712057828903,
      "num_tokens": 124461191.0,
      "step": 61425
    },
    {
      "entropy": 5.907026672363282,
      "epoch": 5.368347461330071,
      "grad_norm": 1.203125,
      "learning_rate": 0.000251863344750313,
      "loss": 5.1924,
      "mean_token_accuracy": 0.2136141374707222,
      "num_tokens": 124469239.0,
      "step": 61430
    },
    {
      "entropy": 5.757833433151245,
      "epoch": 5.3687844096827755,
      "grad_norm": 1.125,
      "learning_rate": 0.0002518323490596366,
      "loss": 5.1318,
      "mean_token_accuracy": 0.2292174682021141,
      "num_tokens": 124478717.0,
      "step": 61435
    },
    {
      "entropy": 5.809430456161499,
      "epoch": 5.36922135803548,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000251801353813328,
      "loss": 5.133,
      "mean_token_accuracy": 0.22027506977319716,
      "num_tokens": 124489270.0,
      "step": 61440
    },
    {
      "entropy": 5.917975187301636,
      "epoch": 5.369658306388185,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025177035901198145,
      "loss": 5.2623,
      "mean_token_accuracy": 0.210924956202507,
      "num_tokens": 124499233.0,
      "step": 61445
    },
    {
      "entropy": 5.853517532348633,
      "epoch": 5.37009525474089,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025173936465619175,
      "loss": 5.2469,
      "mean_token_accuracy": 0.2133162349462509,
      "num_tokens": 124509881.0,
      "step": 61450
    },
    {
      "entropy": 5.904454278945923,
      "epoch": 5.370532203093594,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00025170837074655315,
      "loss": 5.2272,
      "mean_token_accuracy": 0.20818551182746886,
      "num_tokens": 124519466.0,
      "step": 61455
    },
    {
      "entropy": 5.802977657318115,
      "epoch": 5.370969151446299,
      "grad_norm": 1.125,
      "learning_rate": 0.0002516773772836602,
      "loss": 5.077,
      "mean_token_accuracy": 0.2274857297539711,
      "num_tokens": 124528399.0,
      "step": 61460
    },
    {
      "entropy": 5.885010147094727,
      "epoch": 5.371406099799004,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025164638426810736,
      "loss": 5.2802,
      "mean_token_accuracy": 0.21252359598875045,
      "num_tokens": 124539807.0,
      "step": 61465
    },
    {
      "entropy": 5.871362638473511,
      "epoch": 5.3718430481517085,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00025161539170048906,
      "loss": 5.2427,
      "mean_token_accuracy": 0.2142006665468216,
      "num_tokens": 124550681.0,
      "step": 61470
    },
    {
      "entropy": 5.91728138923645,
      "epoch": 5.372279996504413,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002515843995813999,
      "loss": 5.2479,
      "mean_token_accuracy": 0.21524734944105148,
      "num_tokens": 124560623.0,
      "step": 61475
    },
    {
      "entropy": 5.873931741714477,
      "epoch": 5.3727169448571175,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025155340791143424,
      "loss": 5.2326,
      "mean_token_accuracy": 0.21358968764543534,
      "num_tokens": 124571558.0,
      "step": 61480
    },
    {
      "entropy": 5.874268960952759,
      "epoch": 5.373153893209823,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002515224166911864,
      "loss": 5.2303,
      "mean_token_accuracy": 0.21392686516046525,
      "num_tokens": 124582180.0,
      "step": 61485
    },
    {
      "entropy": 5.872440195083618,
      "epoch": 5.373590841562527,
      "grad_norm": 1.140625,
      "learning_rate": 0.000251491425921251,
      "loss": 5.189,
      "mean_token_accuracy": 0.2135837584733963,
      "num_tokens": 124592453.0,
      "step": 61490
    },
    {
      "entropy": 5.84668345451355,
      "epoch": 5.374027789915232,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002514604356022223,
      "loss": 5.1859,
      "mean_token_accuracy": 0.22116338908672334,
      "num_tokens": 124601872.0,
      "step": 61495
    },
    {
      "entropy": 5.897156381607056,
      "epoch": 5.374464738267937,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002514294457346948,
      "loss": 5.2574,
      "mean_token_accuracy": 0.21151820421218873,
      "num_tokens": 124612835.0,
      "step": 61500
    },
    {
      "entropy": 5.894557666778565,
      "epoch": 5.374901686620642,
      "grad_norm": 1.03125,
      "learning_rate": 0.00025139845631926293,
      "loss": 5.1521,
      "mean_token_accuracy": 0.22028412222862243,
      "num_tokens": 124623158.0,
      "step": 61505
    },
    {
      "entropy": 5.905017518997193,
      "epoch": 5.375338634973346,
      "grad_norm": 1.171875,
      "learning_rate": 0.000251367467356521,
      "loss": 5.2155,
      "mean_token_accuracy": 0.21163926273584366,
      "num_tokens": 124632700.0,
      "step": 61510
    },
    {
      "entropy": 5.858822202682495,
      "epoch": 5.3757755833260505,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002513364788470634,
      "loss": 5.2097,
      "mean_token_accuracy": 0.21782552748918532,
      "num_tokens": 124643586.0,
      "step": 61515
    },
    {
      "entropy": 5.855399513244629,
      "epoch": 5.376212531678756,
      "grad_norm": 1.125,
      "learning_rate": 0.00025130549079148455,
      "loss": 5.2464,
      "mean_token_accuracy": 0.2119087189435959,
      "num_tokens": 124653513.0,
      "step": 61520
    },
    {
      "entropy": 5.869025135040284,
      "epoch": 5.37664948003146,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002512745031903788,
      "loss": 5.218,
      "mean_token_accuracy": 0.21985130310058593,
      "num_tokens": 124663725.0,
      "step": 61525
    },
    {
      "entropy": 5.914868879318237,
      "epoch": 5.377086428384165,
      "grad_norm": 1.078125,
      "learning_rate": 0.00025124351604434057,
      "loss": 5.2146,
      "mean_token_accuracy": 0.22075712233781813,
      "num_tokens": 124673803.0,
      "step": 61530
    },
    {
      "entropy": 5.831185245513916,
      "epoch": 5.377523376736869,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002512125293539641,
      "loss": 5.1086,
      "mean_token_accuracy": 0.2200693041086197,
      "num_tokens": 124683628.0,
      "step": 61535
    },
    {
      "entropy": 5.819927406311035,
      "epoch": 5.377960325089575,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002511815431198438,
      "loss": 5.1732,
      "mean_token_accuracy": 0.22212169617414473,
      "num_tokens": 124694214.0,
      "step": 61540
    },
    {
      "entropy": 5.835945892333984,
      "epoch": 5.378397273442279,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025115055734257395,
      "loss": 5.1369,
      "mean_token_accuracy": 0.21820179224014283,
      "num_tokens": 124703559.0,
      "step": 61545
    },
    {
      "entropy": 5.916050958633423,
      "epoch": 5.378834221794984,
      "grad_norm": 1.109375,
      "learning_rate": 0.000251119572022749,
      "loss": 5.2638,
      "mean_token_accuracy": 0.21557379961013795,
      "num_tokens": 124715611.0,
      "step": 61550
    },
    {
      "entropy": 5.883503437042236,
      "epoch": 5.379271170147689,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00025108858716096303,
      "loss": 5.2086,
      "mean_token_accuracy": 0.21683592200279236,
      "num_tokens": 124726144.0,
      "step": 61555
    },
    {
      "entropy": 5.8511800289154055,
      "epoch": 5.379708118500393,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002510576027578106,
      "loss": 5.1864,
      "mean_token_accuracy": 0.22357126027345658,
      "num_tokens": 124735684.0,
      "step": 61560
    },
    {
      "entropy": 5.806182479858398,
      "epoch": 5.380145066853098,
      "grad_norm": 1.125,
      "learning_rate": 0.0002510266188138858,
      "loss": 5.2558,
      "mean_token_accuracy": 0.2136545166373253,
      "num_tokens": 124745914.0,
      "step": 61565
    },
    {
      "entropy": 5.854673194885254,
      "epoch": 5.380582015205802,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002509956353297831,
      "loss": 5.1115,
      "mean_token_accuracy": 0.22456012219190596,
      "num_tokens": 124756766.0,
      "step": 61570
    },
    {
      "entropy": 5.773026466369629,
      "epoch": 5.381018963558508,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002509646523060967,
      "loss": 5.0716,
      "mean_token_accuracy": 0.22738643288612365,
      "num_tokens": 124767095.0,
      "step": 61575
    },
    {
      "entropy": 5.843450403213501,
      "epoch": 5.381455911911212,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002509336697434208,
      "loss": 5.2079,
      "mean_token_accuracy": 0.21683936864137648,
      "num_tokens": 124777154.0,
      "step": 61580
    },
    {
      "entropy": 5.881764984130859,
      "epoch": 5.381892860263917,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025090268764234975,
      "loss": 5.2146,
      "mean_token_accuracy": 0.21887979358434678,
      "num_tokens": 124786228.0,
      "step": 61585
    },
    {
      "entropy": 5.817010068893433,
      "epoch": 5.382329808616621,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025087170600347774,
      "loss": 5.1686,
      "mean_token_accuracy": 0.21632589697837828,
      "num_tokens": 124796337.0,
      "step": 61590
    },
    {
      "entropy": 5.907052993774414,
      "epoch": 5.3827667569693265,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025084072482739906,
      "loss": 5.2738,
      "mean_token_accuracy": 0.20904128700494767,
      "num_tokens": 124806418.0,
      "step": 61595
    },
    {
      "entropy": 5.826500320434571,
      "epoch": 5.383203705322031,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000250809744114708,
      "loss": 5.1419,
      "mean_token_accuracy": 0.22677528560161592,
      "num_tokens": 124815961.0,
      "step": 61600
    },
    {
      "entropy": 5.819893312454224,
      "epoch": 5.383640653674735,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025077876386599867,
      "loss": 5.1751,
      "mean_token_accuracy": 0.22429176419973373,
      "num_tokens": 124825345.0,
      "step": 61605
    },
    {
      "entropy": 5.843751764297485,
      "epoch": 5.384077602027441,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025074778408186536,
      "loss": 5.16,
      "mean_token_accuracy": 0.216802941262722,
      "num_tokens": 124835688.0,
      "step": 61610
    },
    {
      "entropy": 5.896868276596069,
      "epoch": 5.384514550380145,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025071680476290226,
      "loss": 5.2548,
      "mean_token_accuracy": 0.20818147361278533,
      "num_tokens": 124845579.0,
      "step": 61615
    },
    {
      "entropy": 5.876398706436158,
      "epoch": 5.38495149873285,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002506858259097036,
      "loss": 5.1851,
      "mean_token_accuracy": 0.22002973556518554,
      "num_tokens": 124855381.0,
      "step": 61620
    },
    {
      "entropy": 5.822246360778808,
      "epoch": 5.385388447085554,
      "grad_norm": 1.125,
      "learning_rate": 0.00025065484752286347,
      "loss": 5.2151,
      "mean_token_accuracy": 0.21564441174268723,
      "num_tokens": 124865186.0,
      "step": 61625
    },
    {
      "entropy": 5.84189076423645,
      "epoch": 5.3858253954382596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025062386960297616,
      "loss": 5.2304,
      "mean_token_accuracy": 0.21220458298921585,
      "num_tokens": 124875423.0,
      "step": 61630
    },
    {
      "entropy": 5.836026000976562,
      "epoch": 5.386262343790964,
      "grad_norm": 1.125,
      "learning_rate": 0.0002505928921506358,
      "loss": 5.1667,
      "mean_token_accuracy": 0.22025548964738845,
      "num_tokens": 124885423.0,
      "step": 61635
    },
    {
      "entropy": 6.0069293022155765,
      "epoch": 5.3866992921436685,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025056191516643656,
      "loss": 5.3692,
      "mean_token_accuracy": 0.20387116372585296,
      "num_tokens": 124895903.0,
      "step": 61640
    },
    {
      "entropy": 5.855905294418335,
      "epoch": 5.387136240496373,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002505309386509726,
      "loss": 5.1903,
      "mean_token_accuracy": 0.21556751132011415,
      "num_tokens": 124905943.0,
      "step": 61645
    },
    {
      "entropy": 5.781299352645874,
      "epoch": 5.387573188849078,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025049996260483796,
      "loss": 4.9844,
      "mean_token_accuracy": 0.23413843661546707,
      "num_tokens": 124916475.0,
      "step": 61650
    },
    {
      "entropy": 5.812895679473877,
      "epoch": 5.388010137201783,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025046898702862696,
      "loss": 5.137,
      "mean_token_accuracy": 0.22369788438081742,
      "num_tokens": 124926077.0,
      "step": 61655
    },
    {
      "entropy": 5.754429626464844,
      "epoch": 5.388447085554487,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025043801192293363,
      "loss": 5.123,
      "mean_token_accuracy": 0.22251260727643968,
      "num_tokens": 124935264.0,
      "step": 61660
    },
    {
      "entropy": 5.8387425422668455,
      "epoch": 5.388884033907193,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002504070372883521,
      "loss": 5.191,
      "mean_token_accuracy": 0.22663494497537612,
      "num_tokens": 124945503.0,
      "step": 61665
    },
    {
      "entropy": 5.847274494171143,
      "epoch": 5.389320982259897,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025037606312547654,
      "loss": 5.1478,
      "mean_token_accuracy": 0.22353509217500686,
      "num_tokens": 124955353.0,
      "step": 61670
    },
    {
      "entropy": 5.903543424606323,
      "epoch": 5.3897579306126016,
      "grad_norm": 1.125,
      "learning_rate": 0.0002503450894349009,
      "loss": 5.2198,
      "mean_token_accuracy": 0.2114457055926323,
      "num_tokens": 124966001.0,
      "step": 61675
    },
    {
      "entropy": 5.872127199172974,
      "epoch": 5.390194878965306,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025031411621721946,
      "loss": 5.1985,
      "mean_token_accuracy": 0.21630174070596694,
      "num_tokens": 124975516.0,
      "step": 61680
    },
    {
      "entropy": 5.831107091903687,
      "epoch": 5.390631827318011,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002502831434730262,
      "loss": 5.1627,
      "mean_token_accuracy": 0.22235678881406784,
      "num_tokens": 124984624.0,
      "step": 61685
    },
    {
      "entropy": 5.816422319412231,
      "epoch": 5.391068775670716,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002502521712029151,
      "loss": 5.1493,
      "mean_token_accuracy": 0.22757192999124526,
      "num_tokens": 124994856.0,
      "step": 61690
    },
    {
      "entropy": 5.842279767990112,
      "epoch": 5.39150572402342,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002502211994074804,
      "loss": 5.1803,
      "mean_token_accuracy": 0.21842234581708908,
      "num_tokens": 125004154.0,
      "step": 61695
    },
    {
      "entropy": 5.931442975997925,
      "epoch": 5.391942672376125,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002501902280873161,
      "loss": 5.2596,
      "mean_token_accuracy": 0.2146630421280861,
      "num_tokens": 125013287.0,
      "step": 61700
    },
    {
      "entropy": 5.890445899963379,
      "epoch": 5.39237962072883,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025015925724301613,
      "loss": 5.2643,
      "mean_token_accuracy": 0.2071490168571472,
      "num_tokens": 125023445.0,
      "step": 61705
    },
    {
      "entropy": 5.921852111816406,
      "epoch": 5.392816569081535,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002501282868751747,
      "loss": 5.2501,
      "mean_token_accuracy": 0.20831114649772645,
      "num_tokens": 125034443.0,
      "step": 61710
    },
    {
      "entropy": 5.887500190734864,
      "epoch": 5.393253517434239,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00025009731698438575,
      "loss": 5.2053,
      "mean_token_accuracy": 0.21866389065980912,
      "num_tokens": 125044110.0,
      "step": 61715
    },
    {
      "entropy": 5.844639301300049,
      "epoch": 5.393690465786944,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002500663475712433,
      "loss": 5.16,
      "mean_token_accuracy": 0.21466989070177078,
      "num_tokens": 125054017.0,
      "step": 61720
    },
    {
      "entropy": 5.869560813903808,
      "epoch": 5.394127414139649,
      "grad_norm": 1.125,
      "learning_rate": 0.00025003537863634137,
      "loss": 5.2278,
      "mean_token_accuracy": 0.21033093631267546,
      "num_tokens": 125064217.0,
      "step": 61725
    },
    {
      "entropy": 5.870755004882812,
      "epoch": 5.394564362492353,
      "grad_norm": 1.125,
      "learning_rate": 0.00025000441018027393,
      "loss": 5.2862,
      "mean_token_accuracy": 0.21177329123020172,
      "num_tokens": 125074768.0,
      "step": 61730
    },
    {
      "entropy": 5.8358229637146,
      "epoch": 5.395001310845058,
      "grad_norm": 1.03125,
      "learning_rate": 0.00024997344220363505,
      "loss": 5.1488,
      "mean_token_accuracy": 0.22251201868057252,
      "num_tokens": 125085343.0,
      "step": 61735
    },
    {
      "entropy": 5.776456212997436,
      "epoch": 5.395438259197763,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024994247470701866,
      "loss": 5.1742,
      "mean_token_accuracy": 0.22690171450376512,
      "num_tokens": 125095385.0,
      "step": 61740
    },
    {
      "entropy": 5.876190900802612,
      "epoch": 5.395875207550468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002499115076910187,
      "loss": 5.1555,
      "mean_token_accuracy": 0.2164514183998108,
      "num_tokens": 125105100.0,
      "step": 61745
    },
    {
      "entropy": 5.87057671546936,
      "epoch": 5.396312155903172,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024988054115622916,
      "loss": 5.227,
      "mean_token_accuracy": 0.2125218242406845,
      "num_tokens": 125115003.0,
      "step": 61750
    },
    {
      "entropy": 5.802651357650757,
      "epoch": 5.396749104255877,
      "grad_norm": 1.15625,
      "learning_rate": 0.000249849575103244,
      "loss": 5.1893,
      "mean_token_accuracy": 0.2232845738530159,
      "num_tokens": 125124926.0,
      "step": 61755
    },
    {
      "entropy": 5.812474155426026,
      "epoch": 5.397186052608582,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024981860953265727,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21393606662750245,
      "num_tokens": 125133731.0,
      "step": 61760
    },
    {
      "entropy": 5.896534442901611,
      "epoch": 5.397623000961286,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024978764444506276,
      "loss": 5.2125,
      "mean_token_accuracy": 0.2127764567732811,
      "num_tokens": 125142522.0,
      "step": 61765
    },
    {
      "entropy": 5.881294250488281,
      "epoch": 5.398059949313991,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002497566798410544,
      "loss": 5.1069,
      "mean_token_accuracy": 0.22477137297391891,
      "num_tokens": 125152816.0,
      "step": 61770
    },
    {
      "entropy": 5.91558837890625,
      "epoch": 5.398496897666695,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002497257157212261,
      "loss": 5.2654,
      "mean_token_accuracy": 0.21078816056251526,
      "num_tokens": 125162005.0,
      "step": 61775
    },
    {
      "entropy": 5.875582933425903,
      "epoch": 5.398933846019401,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002496947520861719,
      "loss": 5.2414,
      "mean_token_accuracy": 0.21323656439781188,
      "num_tokens": 125172856.0,
      "step": 61780
    },
    {
      "entropy": 5.827754259109497,
      "epoch": 5.399370794372105,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024966378893648554,
      "loss": 5.0754,
      "mean_token_accuracy": 0.2268151506781578,
      "num_tokens": 125182028.0,
      "step": 61785
    },
    {
      "entropy": 5.929358768463135,
      "epoch": 5.39980774272481,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024963282627276105,
      "loss": 5.2691,
      "mean_token_accuracy": 0.2093011200428009,
      "num_tokens": 125191257.0,
      "step": 61790
    },
    {
      "entropy": 5.899605798721313,
      "epoch": 5.400244691077515,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024960186409559224,
      "loss": 5.209,
      "mean_token_accuracy": 0.21606461703777313,
      "num_tokens": 125201448.0,
      "step": 61795
    },
    {
      "entropy": 5.808444833755493,
      "epoch": 5.4006816394302195,
      "grad_norm": 1.046875,
      "learning_rate": 0.00024957090240557297,
      "loss": 5.1788,
      "mean_token_accuracy": 0.21662469506263732,
      "num_tokens": 125212199.0,
      "step": 61800
    },
    {
      "entropy": 5.845820522308349,
      "epoch": 5.401118587782924,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002495399412032971,
      "loss": 5.251,
      "mean_token_accuracy": 0.21060725301504135,
      "num_tokens": 125222201.0,
      "step": 61805
    },
    {
      "entropy": 5.892104244232177,
      "epoch": 5.401555536135628,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024950898048935855,
      "loss": 5.1873,
      "mean_token_accuracy": 0.2143678292632103,
      "num_tokens": 125232329.0,
      "step": 61810
    },
    {
      "entropy": 5.846663141250611,
      "epoch": 5.401992484488334,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002494780202643511,
      "loss": 5.2,
      "mean_token_accuracy": 0.22627917528152466,
      "num_tokens": 125242920.0,
      "step": 61815
    },
    {
      "entropy": 5.884505653381348,
      "epoch": 5.402429432841038,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024944706052886856,
      "loss": 5.2581,
      "mean_token_accuracy": 0.21247004866600036,
      "num_tokens": 125252014.0,
      "step": 61820
    },
    {
      "entropy": 5.842872285842896,
      "epoch": 5.402866381193743,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002494161012835048,
      "loss": 5.1232,
      "mean_token_accuracy": 0.2270761325955391,
      "num_tokens": 125261807.0,
      "step": 61825
    },
    {
      "entropy": 5.855425691604614,
      "epoch": 5.403303329546447,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002493851425288536,
      "loss": 5.1369,
      "mean_token_accuracy": 0.21618829369544984,
      "num_tokens": 125272586.0,
      "step": 61830
    },
    {
      "entropy": 5.922100639343261,
      "epoch": 5.403740277899153,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002493541842655088,
      "loss": 5.2544,
      "mean_token_accuracy": 0.20852405577898026,
      "num_tokens": 125282738.0,
      "step": 61835
    },
    {
      "entropy": 5.813332986831665,
      "epoch": 5.404177226251857,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002493232264940642,
      "loss": 5.1308,
      "mean_token_accuracy": 0.2228452295064926,
      "num_tokens": 125291449.0,
      "step": 61840
    },
    {
      "entropy": 5.837984037399292,
      "epoch": 5.4046141746045615,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024929226921511355,
      "loss": 5.2155,
      "mean_token_accuracy": 0.2188211649656296,
      "num_tokens": 125301373.0,
      "step": 61845
    },
    {
      "entropy": 5.859469604492188,
      "epoch": 5.405051122957267,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002492613124292506,
      "loss": 5.2259,
      "mean_token_accuracy": 0.2175227776169777,
      "num_tokens": 125311475.0,
      "step": 61850
    },
    {
      "entropy": 5.874275970458984,
      "epoch": 5.405488071309971,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024923035613706935,
      "loss": 5.2523,
      "mean_token_accuracy": 0.20675774067640304,
      "num_tokens": 125321981.0,
      "step": 61855
    },
    {
      "entropy": 5.854508638381958,
      "epoch": 5.405925019662676,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002491994003391632,
      "loss": 5.2125,
      "mean_token_accuracy": 0.22316638082265855,
      "num_tokens": 125332491.0,
      "step": 61860
    },
    {
      "entropy": 5.905590438842774,
      "epoch": 5.40636196801538,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00024916844503612614,
      "loss": 5.2294,
      "mean_token_accuracy": 0.2168899729847908,
      "num_tokens": 125342153.0,
      "step": 61865
    },
    {
      "entropy": 5.826717901229858,
      "epoch": 5.406798916368086,
      "grad_norm": 1.125,
      "learning_rate": 0.0002491374902285518,
      "loss": 5.2204,
      "mean_token_accuracy": 0.21249188780784606,
      "num_tokens": 125352830.0,
      "step": 61870
    },
    {
      "entropy": 5.845785140991211,
      "epoch": 5.40723586472079,
      "grad_norm": 1.125,
      "learning_rate": 0.00024910653591703397,
      "loss": 5.2279,
      "mean_token_accuracy": 0.21772327572107314,
      "num_tokens": 125363782.0,
      "step": 61875
    },
    {
      "entropy": 5.860337018966675,
      "epoch": 5.407672813073495,
      "grad_norm": 1.25,
      "learning_rate": 0.0002490755821021664,
      "loss": 5.1941,
      "mean_token_accuracy": 0.20967482924461364,
      "num_tokens": 125373403.0,
      "step": 61880
    },
    {
      "entropy": 5.899634742736817,
      "epoch": 5.408109761426199,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002490446287845427,
      "loss": 5.3021,
      "mean_token_accuracy": 0.20293980985879898,
      "num_tokens": 125383647.0,
      "step": 61885
    },
    {
      "entropy": 5.902106475830078,
      "epoch": 5.408546709778904,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002490136759647566,
      "loss": 5.2587,
      "mean_token_accuracy": 0.21502847224473953,
      "num_tokens": 125393840.0,
      "step": 61890
    },
    {
      "entropy": 5.907250022888183,
      "epoch": 5.408983658131609,
      "grad_norm": 1.125,
      "learning_rate": 0.00024898272364340184,
      "loss": 5.3221,
      "mean_token_accuracy": 0.21405290812253952,
      "num_tokens": 125404550.0,
      "step": 61895
    },
    {
      "entropy": 5.9040735244750975,
      "epoch": 5.409420606484313,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00024895177182107203,
      "loss": 5.2492,
      "mean_token_accuracy": 0.21917207092046737,
      "num_tokens": 125414557.0,
      "step": 61900
    },
    {
      "entropy": 5.859119319915772,
      "epoch": 5.409857554837018,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024892082049836095,
      "loss": 5.1699,
      "mean_token_accuracy": 0.22321218699216844,
      "num_tokens": 125424961.0,
      "step": 61905
    },
    {
      "entropy": 5.852442932128906,
      "epoch": 5.410294503189723,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002488898696758622,
      "loss": 5.2289,
      "mean_token_accuracy": 0.217837093770504,
      "num_tokens": 125434271.0,
      "step": 61910
    },
    {
      "entropy": 5.872655439376831,
      "epoch": 5.410731451542428,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024885891935416937,
      "loss": 5.2311,
      "mean_token_accuracy": 0.2158433198928833,
      "num_tokens": 125444732.0,
      "step": 61915
    },
    {
      "entropy": 5.872144651412964,
      "epoch": 5.411168399895132,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024882796953387616,
      "loss": 5.1647,
      "mean_token_accuracy": 0.21561498939990997,
      "num_tokens": 125454276.0,
      "step": 61920
    },
    {
      "entropy": 5.855586576461792,
      "epoch": 5.4116053482478375,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024879702021557627,
      "loss": 5.3132,
      "mean_token_accuracy": 0.21136273294687272,
      "num_tokens": 125464417.0,
      "step": 61925
    },
    {
      "entropy": 5.830375528335571,
      "epoch": 5.412042296600542,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002487660713998632,
      "loss": 5.1769,
      "mean_token_accuracy": 0.22169426530599595,
      "num_tokens": 125474174.0,
      "step": 61930
    },
    {
      "entropy": 5.8350990295410154,
      "epoch": 5.412479244953246,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024873512308733073,
      "loss": 5.1149,
      "mean_token_accuracy": 0.22543274164199828,
      "num_tokens": 125483557.0,
      "step": 61935
    },
    {
      "entropy": 5.849109792709351,
      "epoch": 5.412916193305951,
      "grad_norm": 1.046875,
      "learning_rate": 0.00024870417527857224,
      "loss": 5.158,
      "mean_token_accuracy": 0.21990798860788346,
      "num_tokens": 125494250.0,
      "step": 61940
    },
    {
      "entropy": 5.825496673583984,
      "epoch": 5.413353141658656,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002486732279741815,
      "loss": 5.1651,
      "mean_token_accuracy": 0.21892015039920806,
      "num_tokens": 125503151.0,
      "step": 61945
    },
    {
      "entropy": 5.874508476257324,
      "epoch": 5.413790090011361,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024864228117475203,
      "loss": 5.2767,
      "mean_token_accuracy": 0.20792326480150222,
      "num_tokens": 125512187.0,
      "step": 61950
    },
    {
      "entropy": 5.830628871917725,
      "epoch": 5.414227038364065,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024861133488087746,
      "loss": 5.1543,
      "mean_token_accuracy": 0.2228178545832634,
      "num_tokens": 125522416.0,
      "step": 61955
    },
    {
      "entropy": 5.954314184188843,
      "epoch": 5.41466398671677,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002485803890931513,
      "loss": 5.3285,
      "mean_token_accuracy": 0.1972593918442726,
      "num_tokens": 125532530.0,
      "step": 61960
    },
    {
      "entropy": 5.7966361999511715,
      "epoch": 5.415100935069475,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002485494438121671,
      "loss": 5.0819,
      "mean_token_accuracy": 0.23311345130205155,
      "num_tokens": 125542282.0,
      "step": 61965
    },
    {
      "entropy": 5.86147255897522,
      "epoch": 5.4155378834221795,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00024851849903851844,
      "loss": 5.1822,
      "mean_token_accuracy": 0.21809669882059096,
      "num_tokens": 125552404.0,
      "step": 61970
    },
    {
      "entropy": 5.877524137496948,
      "epoch": 5.415974831774884,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002484875547727988,
      "loss": 5.2619,
      "mean_token_accuracy": 0.20829402804374694,
      "num_tokens": 125562285.0,
      "step": 61975
    },
    {
      "entropy": 5.830034923553467,
      "epoch": 5.416411780127589,
      "grad_norm": 1.03125,
      "learning_rate": 0.00024845661101560186,
      "loss": 5.2272,
      "mean_token_accuracy": 0.2138385519385338,
      "num_tokens": 125572795.0,
      "step": 61980
    },
    {
      "entropy": 5.90495982170105,
      "epoch": 5.416848728480294,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002484256677675209,
      "loss": 5.2245,
      "mean_token_accuracy": 0.2134304866194725,
      "num_tokens": 125583592.0,
      "step": 61985
    },
    {
      "entropy": 5.8657886505126955,
      "epoch": 5.417285676832998,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002483947250291496,
      "loss": 5.2302,
      "mean_token_accuracy": 0.2118639439344406,
      "num_tokens": 125594018.0,
      "step": 61990
    },
    {
      "entropy": 5.820235824584961,
      "epoch": 5.417722625185703,
      "grad_norm": 1.125,
      "learning_rate": 0.0002483637828010815,
      "loss": 5.1978,
      "mean_token_accuracy": 0.21960858404636383,
      "num_tokens": 125604946.0,
      "step": 61995
    },
    {
      "entropy": 5.8642502784729,
      "epoch": 5.418159573538408,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024833284108390986,
      "loss": 5.21,
      "mean_token_accuracy": 0.21630256474018097,
      "num_tokens": 125614243.0,
      "step": 62000
    },
    {
      "entropy": 5.902847051620483,
      "epoch": 5.4185965218911125,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00024830189987822834,
      "loss": 5.2443,
      "mean_token_accuracy": 0.21059580296278,
      "num_tokens": 125625251.0,
      "step": 62005
    },
    {
      "entropy": 5.853356838226318,
      "epoch": 5.419033470243817,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002482709591846304,
      "loss": 5.1908,
      "mean_token_accuracy": 0.22566444277763367,
      "num_tokens": 125635312.0,
      "step": 62010
    },
    {
      "entropy": 5.908271074295044,
      "epoch": 5.4194704185965215,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002482400190037095,
      "loss": 5.3415,
      "mean_token_accuracy": 0.2001085177063942,
      "num_tokens": 125645618.0,
      "step": 62015
    },
    {
      "entropy": 5.875373458862304,
      "epoch": 5.419907366949227,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024820907933605904,
      "loss": 5.2079,
      "mean_token_accuracy": 0.22218888998031616,
      "num_tokens": 125656445.0,
      "step": 62020
    },
    {
      "entropy": 5.831534719467163,
      "epoch": 5.420344315301931,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024817814018227247,
      "loss": 5.1319,
      "mean_token_accuracy": 0.23081128299236298,
      "num_tokens": 125666282.0,
      "step": 62025
    },
    {
      "entropy": 5.87288908958435,
      "epoch": 5.420781263654636,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002481472015429432,
      "loss": 5.207,
      "mean_token_accuracy": 0.21932636350393295,
      "num_tokens": 125676196.0,
      "step": 62030
    },
    {
      "entropy": 5.846082401275635,
      "epoch": 5.421218212007341,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024811626341866464,
      "loss": 5.2592,
      "mean_token_accuracy": 0.20895850360393525,
      "num_tokens": 125686601.0,
      "step": 62035
    },
    {
      "entropy": 5.926618146896362,
      "epoch": 5.421655160360046,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024808532581003037,
      "loss": 5.2203,
      "mean_token_accuracy": 0.21051522046327592,
      "num_tokens": 125696787.0,
      "step": 62040
    },
    {
      "entropy": 5.876022481918335,
      "epoch": 5.42209210871275,
      "grad_norm": 1.125,
      "learning_rate": 0.0002480543887176336,
      "loss": 5.1988,
      "mean_token_accuracy": 0.21372314244508744,
      "num_tokens": 125707858.0,
      "step": 62045
    },
    {
      "entropy": 5.854219198226929,
      "epoch": 5.4225290570654545,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024802345214206765,
      "loss": 5.1854,
      "mean_token_accuracy": 0.22183957695960999,
      "num_tokens": 125718172.0,
      "step": 62050
    },
    {
      "entropy": 5.8953554153442385,
      "epoch": 5.42296600541816,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002479925160839262,
      "loss": 5.1994,
      "mean_token_accuracy": 0.22023999691009521,
      "num_tokens": 125729060.0,
      "step": 62055
    },
    {
      "entropy": 5.860545873641968,
      "epoch": 5.423402953770864,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002479615805438023,
      "loss": 5.1746,
      "mean_token_accuracy": 0.22217163890600206,
      "num_tokens": 125738819.0,
      "step": 62060
    },
    {
      "entropy": 5.884372663497925,
      "epoch": 5.423839902123569,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024793064552228953,
      "loss": 5.1817,
      "mean_token_accuracy": 0.2169395700097084,
      "num_tokens": 125748128.0,
      "step": 62065
    },
    {
      "entropy": 5.8359888076782225,
      "epoch": 5.424276850476273,
      "grad_norm": 1.125,
      "learning_rate": 0.0002478997110199812,
      "loss": 5.1432,
      "mean_token_accuracy": 0.22000993937253951,
      "num_tokens": 125758622.0,
      "step": 62070
    },
    {
      "entropy": 5.850030755996704,
      "epoch": 5.424713798828979,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024786877703747055,
      "loss": 5.1756,
      "mean_token_accuracy": 0.22064959704875947,
      "num_tokens": 125767806.0,
      "step": 62075
    },
    {
      "entropy": 5.938510131835938,
      "epoch": 5.425150747181683,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000247837843575351,
      "loss": 5.2947,
      "mean_token_accuracy": 0.21098721027374268,
      "num_tokens": 125777373.0,
      "step": 62080
    },
    {
      "entropy": 5.88432674407959,
      "epoch": 5.425587695534388,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002478069106342158,
      "loss": 5.2325,
      "mean_token_accuracy": 0.21626638621091843,
      "num_tokens": 125787701.0,
      "step": 62085
    },
    {
      "entropy": 5.849546480178833,
      "epoch": 5.426024643887093,
      "grad_norm": 1.25,
      "learning_rate": 0.0002477759782146584,
      "loss": 5.1678,
      "mean_token_accuracy": 0.22390058338642121,
      "num_tokens": 125797063.0,
      "step": 62090
    },
    {
      "entropy": 5.870355415344238,
      "epoch": 5.426461592239797,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002477450463172719,
      "loss": 5.2281,
      "mean_token_accuracy": 0.217142254114151,
      "num_tokens": 125806922.0,
      "step": 62095
    },
    {
      "entropy": 5.863258409500122,
      "epoch": 5.426898540592502,
      "grad_norm": 1.046875,
      "learning_rate": 0.00024771411494264975,
      "loss": 5.2683,
      "mean_token_accuracy": 0.20099122673273087,
      "num_tokens": 125817167.0,
      "step": 62100
    },
    {
      "entropy": 5.857768774032593,
      "epoch": 5.427335488945206,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002476831840913851,
      "loss": 5.1488,
      "mean_token_accuracy": 0.22312836050987245,
      "num_tokens": 125827082.0,
      "step": 62105
    },
    {
      "entropy": 5.872993469238281,
      "epoch": 5.427772437297912,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00024765225376407134,
      "loss": 5.2502,
      "mean_token_accuracy": 0.2157857432961464,
      "num_tokens": 125837727.0,
      "step": 62110
    },
    {
      "entropy": 5.8792346000671385,
      "epoch": 5.428209385650616,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00024762132396130175,
      "loss": 5.2598,
      "mean_token_accuracy": 0.2138613671064377,
      "num_tokens": 125847900.0,
      "step": 62115
    },
    {
      "entropy": 5.789610958099365,
      "epoch": 5.428646334003321,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00024759039468366943,
      "loss": 5.1349,
      "mean_token_accuracy": 0.22313386499881743,
      "num_tokens": 125857816.0,
      "step": 62120
    },
    {
      "entropy": 5.84381685256958,
      "epoch": 5.429083282356025,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002475594659317678,
      "loss": 5.2352,
      "mean_token_accuracy": 0.21959846317768097,
      "num_tokens": 125868734.0,
      "step": 62125
    },
    {
      "entropy": 5.8962174415588375,
      "epoch": 5.4295202307087305,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002475285377061898,
      "loss": 5.2483,
      "mean_token_accuracy": 0.20706566721200942,
      "num_tokens": 125878987.0,
      "step": 62130
    },
    {
      "entropy": 5.847641134262085,
      "epoch": 5.429957179061435,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024749761000752903,
      "loss": 5.2176,
      "mean_token_accuracy": 0.21811597645282746,
      "num_tokens": 125888389.0,
      "step": 62135
    },
    {
      "entropy": 5.812290191650391,
      "epoch": 5.430394127414139,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024746668283637843,
      "loss": 5.1333,
      "mean_token_accuracy": 0.22547849267721176,
      "num_tokens": 125897951.0,
      "step": 62140
    },
    {
      "entropy": 5.866216468811035,
      "epoch": 5.430831075766845,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002474357561933313,
      "loss": 5.2243,
      "mean_token_accuracy": 0.2164355233311653,
      "num_tokens": 125908535.0,
      "step": 62145
    },
    {
      "entropy": 5.8238043785095215,
      "epoch": 5.431268024119549,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002474048300789808,
      "loss": 5.1215,
      "mean_token_accuracy": 0.22643649578094482,
      "num_tokens": 125918415.0,
      "step": 62150
    },
    {
      "entropy": 5.8276325225830075,
      "epoch": 5.431704972472254,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002473739044939201,
      "loss": 5.1521,
      "mean_token_accuracy": 0.22017334550619125,
      "num_tokens": 125928679.0,
      "step": 62155
    },
    {
      "entropy": 5.8324949741363525,
      "epoch": 5.432141920824958,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024734297943874246,
      "loss": 5.1954,
      "mean_token_accuracy": 0.21605506390333176,
      "num_tokens": 125938649.0,
      "step": 62160
    },
    {
      "entropy": 5.877627229690551,
      "epoch": 5.4325788691776635,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002473120549140409,
      "loss": 5.2108,
      "mean_token_accuracy": 0.21960764974355698,
      "num_tokens": 125949188.0,
      "step": 62165
    },
    {
      "entropy": 5.854521799087524,
      "epoch": 5.433015817530368,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002472811309204086,
      "loss": 5.1454,
      "mean_token_accuracy": 0.22885323017835618,
      "num_tokens": 125958672.0,
      "step": 62170
    },
    {
      "entropy": 5.859076595306396,
      "epoch": 5.4334527658830725,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002472502074584388,
      "loss": 5.155,
      "mean_token_accuracy": 0.22021180093288423,
      "num_tokens": 125968874.0,
      "step": 62175
    },
    {
      "entropy": 5.889335203170776,
      "epoch": 5.433889714235777,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002472192845287245,
      "loss": 5.2588,
      "mean_token_accuracy": 0.2056918904185295,
      "num_tokens": 125979607.0,
      "step": 62180
    },
    {
      "entropy": 5.8843334197998045,
      "epoch": 5.434326662588482,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024718836213185897,
      "loss": 5.3085,
      "mean_token_accuracy": 0.20537298768758774,
      "num_tokens": 125990100.0,
      "step": 62185
    },
    {
      "entropy": 5.87847933769226,
      "epoch": 5.434763610941187,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024715744026843517,
      "loss": 5.1657,
      "mean_token_accuracy": 0.20875435322523117,
      "num_tokens": 126000022.0,
      "step": 62190
    },
    {
      "entropy": 5.81868987083435,
      "epoch": 5.435200559293891,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024712651893904627,
      "loss": 5.1933,
      "mean_token_accuracy": 0.22288817316293716,
      "num_tokens": 126009857.0,
      "step": 62195
    },
    {
      "entropy": 5.737622451782227,
      "epoch": 5.435637507646597,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002470955981442853,
      "loss": 5.043,
      "mean_token_accuracy": 0.2351227268576622,
      "num_tokens": 126020267.0,
      "step": 62200
    },
    {
      "entropy": 5.848813152313232,
      "epoch": 5.436074455999301,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002470646778847455,
      "loss": 5.2256,
      "mean_token_accuracy": 0.21273237466812134,
      "num_tokens": 126030735.0,
      "step": 62205
    },
    {
      "entropy": 5.8266764163970945,
      "epoch": 5.4365114043520055,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002470337581610197,
      "loss": 5.1543,
      "mean_token_accuracy": 0.21751252114772796,
      "num_tokens": 126041054.0,
      "step": 62210
    },
    {
      "entropy": 5.818909072875977,
      "epoch": 5.43694835270471,
      "grad_norm": 1.140625,
      "learning_rate": 0.00024700283897370113,
      "loss": 5.1363,
      "mean_token_accuracy": 0.22933305501937867,
      "num_tokens": 126051258.0,
      "step": 62215
    },
    {
      "entropy": 5.833260107040405,
      "epoch": 5.437385301057415,
      "grad_norm": 1.125,
      "learning_rate": 0.00024697192032338273,
      "loss": 5.1466,
      "mean_token_accuracy": 0.22648336440324784,
      "num_tokens": 126060485.0,
      "step": 62220
    },
    {
      "entropy": 5.860839033126831,
      "epoch": 5.43782224941012,
      "grad_norm": 1.25,
      "learning_rate": 0.0002469410022106576,
      "loss": 5.1617,
      "mean_token_accuracy": 0.2246729999780655,
      "num_tokens": 126071080.0,
      "step": 62225
    },
    {
      "entropy": 5.921841669082641,
      "epoch": 5.438259197762824,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024691008463611873,
      "loss": 5.3255,
      "mean_token_accuracy": 0.20430052429437637,
      "num_tokens": 126081057.0,
      "step": 62230
    },
    {
      "entropy": 5.798677730560303,
      "epoch": 5.438696146115529,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024687916760035924,
      "loss": 5.1061,
      "mean_token_accuracy": 0.2230978325009346,
      "num_tokens": 126091312.0,
      "step": 62235
    },
    {
      "entropy": 5.817716598510742,
      "epoch": 5.439133094468234,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024684825110397196,
      "loss": 5.1194,
      "mean_token_accuracy": 0.22501124143600465,
      "num_tokens": 126100864.0,
      "step": 62240
    },
    {
      "entropy": 5.850128602981568,
      "epoch": 5.439570042820939,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024681733514755,
      "loss": 5.2002,
      "mean_token_accuracy": 0.21466539055109024,
      "num_tokens": 126110662.0,
      "step": 62245
    },
    {
      "entropy": 5.82912802696228,
      "epoch": 5.440006991173643,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024678641973168634,
      "loss": 5.1706,
      "mean_token_accuracy": 0.21928295940160752,
      "num_tokens": 126120409.0,
      "step": 62250
    },
    {
      "entropy": 5.820542192459106,
      "epoch": 5.440443939526348,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002467555048569739,
      "loss": 5.1704,
      "mean_token_accuracy": 0.21467126160860062,
      "num_tokens": 126130541.0,
      "step": 62255
    },
    {
      "entropy": 5.888935518264771,
      "epoch": 5.440880887879053,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024672459052400576,
      "loss": 5.2344,
      "mean_token_accuracy": 0.21613105088472367,
      "num_tokens": 126141015.0,
      "step": 62260
    },
    {
      "entropy": 5.851725006103516,
      "epoch": 5.441317836231757,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002466936767333747,
      "loss": 5.2157,
      "mean_token_accuracy": 0.21376904994249343,
      "num_tokens": 126150376.0,
      "step": 62265
    },
    {
      "entropy": 5.934937810897827,
      "epoch": 5.441754784584462,
      "grad_norm": 1.125,
      "learning_rate": 0.0002466627634856738,
      "loss": 5.3102,
      "mean_token_accuracy": 0.2048228055238724,
      "num_tokens": 126160980.0,
      "step": 62270
    },
    {
      "entropy": 5.892531824111939,
      "epoch": 5.442191732937167,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002466318507814959,
      "loss": 5.2111,
      "mean_token_accuracy": 0.21627283245325088,
      "num_tokens": 126170181.0,
      "step": 62275
    },
    {
      "entropy": 5.896350574493408,
      "epoch": 5.442628681289872,
      "grad_norm": 1.1875,
      "learning_rate": 0.000246600938621434,
      "loss": 5.2756,
      "mean_token_accuracy": 0.20884174406528472,
      "num_tokens": 126179836.0,
      "step": 62280
    },
    {
      "entropy": 5.907616329193115,
      "epoch": 5.443065629642576,
      "grad_norm": 1.03125,
      "learning_rate": 0.000246570027006081,
      "loss": 5.239,
      "mean_token_accuracy": 0.21395530551671982,
      "num_tokens": 126190021.0,
      "step": 62285
    },
    {
      "entropy": 5.8078882694244385,
      "epoch": 5.443502577995281,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024653911593602976,
      "loss": 5.1246,
      "mean_token_accuracy": 0.22679390907287597,
      "num_tokens": 126200642.0,
      "step": 62290
    },
    {
      "entropy": 5.775629711151123,
      "epoch": 5.443939526347986,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024650820541187316,
      "loss": 5.1048,
      "mean_token_accuracy": 0.23264601826667786,
      "num_tokens": 126210711.0,
      "step": 62295
    },
    {
      "entropy": 5.80217227935791,
      "epoch": 5.44437647470069,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002464772954342041,
      "loss": 5.134,
      "mean_token_accuracy": 0.223495751619339,
      "num_tokens": 126220739.0,
      "step": 62300
    },
    {
      "entropy": 5.9553240776062015,
      "epoch": 5.444813423053395,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002464463860036155,
      "loss": 5.2915,
      "mean_token_accuracy": 0.20948550403118132,
      "num_tokens": 126231498.0,
      "step": 62305
    },
    {
      "entropy": 5.874520587921142,
      "epoch": 5.4452503714061,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024641547712070014,
      "loss": 5.2394,
      "mean_token_accuracy": 0.21230707615613936,
      "num_tokens": 126241172.0,
      "step": 62310
    },
    {
      "entropy": 5.91136155128479,
      "epoch": 5.445687319758805,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024638456878605096,
      "loss": 5.3102,
      "mean_token_accuracy": 0.20578216761350632,
      "num_tokens": 126250941.0,
      "step": 62315
    },
    {
      "entropy": 5.836944055557251,
      "epoch": 5.446124268111509,
      "grad_norm": 1.015625,
      "learning_rate": 0.00024635366100026076,
      "loss": 5.1456,
      "mean_token_accuracy": 0.21939098536968232,
      "num_tokens": 126261071.0,
      "step": 62320
    },
    {
      "entropy": 5.789814615249634,
      "epoch": 5.446561216464214,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00024632275376392225,
      "loss": 5.2078,
      "mean_token_accuracy": 0.2177569419145584,
      "num_tokens": 126271068.0,
      "step": 62325
    },
    {
      "entropy": 5.868464994430542,
      "epoch": 5.446998164816919,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024629184707762846,
      "loss": 5.2358,
      "mean_token_accuracy": 0.21695550680160522,
      "num_tokens": 126280218.0,
      "step": 62330
    },
    {
      "entropy": 5.854587745666504,
      "epoch": 5.4474351131696235,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00024626094094197206,
      "loss": 5.0725,
      "mean_token_accuracy": 0.22595454901456832,
      "num_tokens": 126289813.0,
      "step": 62335
    },
    {
      "entropy": 5.7975996971130375,
      "epoch": 5.447872061522328,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002462300353575458,
      "loss": 5.1423,
      "mean_token_accuracy": 0.22674878239631652,
      "num_tokens": 126300125.0,
      "step": 62340
    },
    {
      "entropy": 5.753090572357178,
      "epoch": 5.448309009875032,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002461991303249426,
      "loss": 5.1739,
      "mean_token_accuracy": 0.21930561810731888,
      "num_tokens": 126310537.0,
      "step": 62345
    },
    {
      "entropy": 5.8507452487945555,
      "epoch": 5.448745958227738,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002461682258447551,
      "loss": 5.2626,
      "mean_token_accuracy": 0.21196883916854858,
      "num_tokens": 126320644.0,
      "step": 62350
    },
    {
      "entropy": 5.891303634643554,
      "epoch": 5.449182906580442,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024613732191757615,
      "loss": 5.2078,
      "mean_token_accuracy": 0.2170141875743866,
      "num_tokens": 126331486.0,
      "step": 62355
    },
    {
      "entropy": 5.961665248870849,
      "epoch": 5.449619854933147,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002461064185439985,
      "loss": 5.2646,
      "mean_token_accuracy": 0.21417986750602722,
      "num_tokens": 126341684.0,
      "step": 62360
    },
    {
      "entropy": 5.881649208068848,
      "epoch": 5.450056803285852,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024607551572461493,
      "loss": 5.2365,
      "mean_token_accuracy": 0.2178796723484993,
      "num_tokens": 126352264.0,
      "step": 62365
    },
    {
      "entropy": 5.879782342910767,
      "epoch": 5.4504937516385565,
      "grad_norm": 1.0625,
      "learning_rate": 0.00024604461346001814,
      "loss": 5.281,
      "mean_token_accuracy": 0.2132712185382843,
      "num_tokens": 126363869.0,
      "step": 62370
    },
    {
      "entropy": 5.890753602981567,
      "epoch": 5.450930699991261,
      "grad_norm": 1.125,
      "learning_rate": 0.00024601371175080077,
      "loss": 5.2047,
      "mean_token_accuracy": 0.22076687812805176,
      "num_tokens": 126374763.0,
      "step": 62375
    },
    {
      "entropy": 5.872761964797974,
      "epoch": 5.4513676483439655,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00024598281059755566,
      "loss": 5.254,
      "mean_token_accuracy": 0.2142028659582138,
      "num_tokens": 126384179.0,
      "step": 62380
    },
    {
      "entropy": 5.889883089065552,
      "epoch": 5.451804596696671,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024595191000087544,
      "loss": 5.2234,
      "mean_token_accuracy": 0.2174105778336525,
      "num_tokens": 126394350.0,
      "step": 62385
    },
    {
      "entropy": 5.834632158279419,
      "epoch": 5.452241545049375,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002459210099613527,
      "loss": 5.1643,
      "mean_token_accuracy": 0.21854951977729797,
      "num_tokens": 126404818.0,
      "step": 62390
    },
    {
      "entropy": 5.86187252998352,
      "epoch": 5.45267849340208,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024589011047958036,
      "loss": 5.2335,
      "mean_token_accuracy": 0.21743471175432205,
      "num_tokens": 126415561.0,
      "step": 62395
    },
    {
      "entropy": 5.83949875831604,
      "epoch": 5.453115441754784,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002458592115561509,
      "loss": 5.1085,
      "mean_token_accuracy": 0.22290666550397872,
      "num_tokens": 126425620.0,
      "step": 62400
    },
    {
      "entropy": 5.866055822372436,
      "epoch": 5.45355239010749,
      "grad_norm": 1.125,
      "learning_rate": 0.00024582831319165703,
      "loss": 5.1522,
      "mean_token_accuracy": 0.2159618020057678,
      "num_tokens": 126434870.0,
      "step": 62405
    },
    {
      "entropy": 5.80110011100769,
      "epoch": 5.453989338460194,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002457974153866914,
      "loss": 5.198,
      "mean_token_accuracy": 0.2178236111998558,
      "num_tokens": 126444508.0,
      "step": 62410
    },
    {
      "entropy": 5.891092443466187,
      "epoch": 5.4544262868128985,
      "grad_norm": 1.125,
      "learning_rate": 0.0002457665181418467,
      "loss": 5.2042,
      "mean_token_accuracy": 0.21467438340187073,
      "num_tokens": 126455040.0,
      "step": 62415
    },
    {
      "entropy": 5.920819425582886,
      "epoch": 5.454863235165604,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002457356214577155,
      "loss": 5.2929,
      "mean_token_accuracy": 0.21212466210126876,
      "num_tokens": 126465890.0,
      "step": 62420
    },
    {
      "entropy": 5.819927978515625,
      "epoch": 5.455300183518308,
      "grad_norm": 1.0,
      "learning_rate": 0.00024570472533489037,
      "loss": 5.1408,
      "mean_token_accuracy": 0.22441952377557756,
      "num_tokens": 126476177.0,
      "step": 62425
    },
    {
      "entropy": 5.853008460998535,
      "epoch": 5.455737131871013,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002456738297739639,
      "loss": 5.2565,
      "mean_token_accuracy": 0.21059761941432953,
      "num_tokens": 126486816.0,
      "step": 62430
    },
    {
      "entropy": 5.880437850952148,
      "epoch": 5.456174080223717,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002456429347755288,
      "loss": 5.2107,
      "mean_token_accuracy": 0.2132252499461174,
      "num_tokens": 126497017.0,
      "step": 62435
    },
    {
      "entropy": 5.894294261932373,
      "epoch": 5.456611028576423,
      "grad_norm": 1.140625,
      "learning_rate": 0.00024561204034017763,
      "loss": 5.1985,
      "mean_token_accuracy": 0.20995407402515412,
      "num_tokens": 126506432.0,
      "step": 62440
    },
    {
      "entropy": 5.891908311843872,
      "epoch": 5.457047976929127,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002455811464685029,
      "loss": 5.2508,
      "mean_token_accuracy": 0.21251547783613206,
      "num_tokens": 126516047.0,
      "step": 62445
    },
    {
      "entropy": 5.835444164276123,
      "epoch": 5.457484925281832,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024555025316109715,
      "loss": 5.1312,
      "mean_token_accuracy": 0.22774304300546647,
      "num_tokens": 126525733.0,
      "step": 62450
    },
    {
      "entropy": 5.814400720596313,
      "epoch": 5.457921873634536,
      "grad_norm": 1.109375,
      "learning_rate": 0.000245519360418553,
      "loss": 5.1905,
      "mean_token_accuracy": 0.21437804102897645,
      "num_tokens": 126534967.0,
      "step": 62455
    },
    {
      "entropy": 5.779571008682251,
      "epoch": 5.458358821987241,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000245488468241463,
      "loss": 5.1084,
      "mean_token_accuracy": 0.22545215487480164,
      "num_tokens": 126544800.0,
      "step": 62460
    },
    {
      "entropy": 5.824568176269532,
      "epoch": 5.458795770339946,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002454575766304196,
      "loss": 5.2204,
      "mean_token_accuracy": 0.2144869789481163,
      "num_tokens": 126554856.0,
      "step": 62465
    },
    {
      "entropy": 5.933547496795654,
      "epoch": 5.45923271869265,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002454266855860154,
      "loss": 5.2753,
      "mean_token_accuracy": 0.2084350124001503,
      "num_tokens": 126565398.0,
      "step": 62470
    },
    {
      "entropy": 5.827747440338134,
      "epoch": 5.459669667045355,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002453957951088428,
      "loss": 5.0969,
      "mean_token_accuracy": 0.2210448056459427,
      "num_tokens": 126575427.0,
      "step": 62475
    },
    {
      "entropy": 5.799513339996338,
      "epoch": 5.46010661539806,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002453649051994944,
      "loss": 5.2434,
      "mean_token_accuracy": 0.2148277521133423,
      "num_tokens": 126586197.0,
      "step": 62480
    },
    {
      "entropy": 5.8879475593566895,
      "epoch": 5.460543563750765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024533401585856267,
      "loss": 5.2159,
      "mean_token_accuracy": 0.2194475769996643,
      "num_tokens": 126595707.0,
      "step": 62485
    },
    {
      "entropy": 5.843797492980957,
      "epoch": 5.460980512103469,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024530312708664003,
      "loss": 5.2255,
      "mean_token_accuracy": 0.2158366024494171,
      "num_tokens": 126604594.0,
      "step": 62490
    },
    {
      "entropy": 5.846048974990845,
      "epoch": 5.4614174604561745,
      "grad_norm": 1.125,
      "learning_rate": 0.00024527223888431906,
      "loss": 5.1474,
      "mean_token_accuracy": 0.222189062833786,
      "num_tokens": 126614490.0,
      "step": 62495
    },
    {
      "entropy": 5.877003383636475,
      "epoch": 5.461854408808879,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000245241351252192,
      "loss": 5.2025,
      "mean_token_accuracy": 0.21959995478391647,
      "num_tokens": 126624227.0,
      "step": 62500
    },
    {
      "entropy": 5.916490602493286,
      "epoch": 5.462291357161583,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002452104641908515,
      "loss": 5.1966,
      "mean_token_accuracy": 0.22050365507602693,
      "num_tokens": 126633698.0,
      "step": 62505
    },
    {
      "entropy": 5.802379989624024,
      "epoch": 5.462728305514288,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024517957770088983,
      "loss": 5.1924,
      "mean_token_accuracy": 0.21567383855581285,
      "num_tokens": 126643703.0,
      "step": 62510
    },
    {
      "entropy": 5.853786993026733,
      "epoch": 5.463165253866993,
      "grad_norm": 1.015625,
      "learning_rate": 0.00024514869178289956,
      "loss": 5.1955,
      "mean_token_accuracy": 0.21426486372947692,
      "num_tokens": 126653609.0,
      "step": 62515
    },
    {
      "entropy": 5.8523077964782715,
      "epoch": 5.463602202219698,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000245117806437473,
      "loss": 5.1664,
      "mean_token_accuracy": 0.22594016492366792,
      "num_tokens": 126663881.0,
      "step": 62520
    },
    {
      "entropy": 5.8432700634002686,
      "epoch": 5.464039150572402,
      "grad_norm": 1.03125,
      "learning_rate": 0.00024508692166520254,
      "loss": 5.1526,
      "mean_token_accuracy": 0.22137845009565355,
      "num_tokens": 126673771.0,
      "step": 62525
    },
    {
      "entropy": 5.838052606582641,
      "epoch": 5.464476098925107,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002450560374666806,
      "loss": 5.1319,
      "mean_token_accuracy": 0.22441843897104263,
      "num_tokens": 126682703.0,
      "step": 62530
    },
    {
      "entropy": 5.807549619674683,
      "epoch": 5.464913047277812,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024502515384249956,
      "loss": 5.214,
      "mean_token_accuracy": 0.22037265449762344,
      "num_tokens": 126692089.0,
      "step": 62535
    },
    {
      "entropy": 5.819486951828003,
      "epoch": 5.4653499956305165,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024499427079325187,
      "loss": 5.1337,
      "mean_token_accuracy": 0.22570179849863053,
      "num_tokens": 126701993.0,
      "step": 62540
    },
    {
      "entropy": 5.907558250427246,
      "epoch": 5.465786943983221,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002449633883195297,
      "loss": 5.2032,
      "mean_token_accuracy": 0.2149224102497101,
      "num_tokens": 126713431.0,
      "step": 62545
    },
    {
      "entropy": 5.838865566253662,
      "epoch": 5.466223892335925,
      "grad_norm": 1.125,
      "learning_rate": 0.0002449325064219255,
      "loss": 5.1809,
      "mean_token_accuracy": 0.22068215906620026,
      "num_tokens": 126723480.0,
      "step": 62550
    },
    {
      "entropy": 5.849808883666992,
      "epoch": 5.466660840688631,
      "grad_norm": 1.015625,
      "learning_rate": 0.00024490162510103154,
      "loss": 5.1479,
      "mean_token_accuracy": 0.21048859357833863,
      "num_tokens": 126733637.0,
      "step": 62555
    },
    {
      "entropy": 5.871515846252441,
      "epoch": 5.467097789041335,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024487074435744017,
      "loss": 5.2459,
      "mean_token_accuracy": 0.2171633407473564,
      "num_tokens": 126744226.0,
      "step": 62560
    },
    {
      "entropy": 5.847854232788086,
      "epoch": 5.46753473739404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024483986419174383,
      "loss": 5.161,
      "mean_token_accuracy": 0.21805773824453353,
      "num_tokens": 126753772.0,
      "step": 62565
    },
    {
      "entropy": 5.85986442565918,
      "epoch": 5.467971685746745,
      "grad_norm": 1.046875,
      "learning_rate": 0.00024480898460453463,
      "loss": 5.2108,
      "mean_token_accuracy": 0.2139774888753891,
      "num_tokens": 126764053.0,
      "step": 62570
    },
    {
      "entropy": 5.9412524700164795,
      "epoch": 5.4684086340994496,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024477810559640485,
      "loss": 5.332,
      "mean_token_accuracy": 0.20985397845506668,
      "num_tokens": 126775218.0,
      "step": 62575
    },
    {
      "entropy": 5.89195237159729,
      "epoch": 5.468845582452154,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002447472271679469,
      "loss": 5.2555,
      "mean_token_accuracy": 0.21184714436531066,
      "num_tokens": 126785449.0,
      "step": 62580
    },
    {
      "entropy": 5.9141875267028805,
      "epoch": 5.4692825308048585,
      "grad_norm": 1.25,
      "learning_rate": 0.000244716349319753,
      "loss": 5.2271,
      "mean_token_accuracy": 0.21590752750635148,
      "num_tokens": 126795618.0,
      "step": 62585
    },
    {
      "entropy": 5.926526308059692,
      "epoch": 5.469719479157564,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002446854720524154,
      "loss": 5.2025,
      "mean_token_accuracy": 0.2202798992395401,
      "num_tokens": 126806352.0,
      "step": 62590
    },
    {
      "entropy": 5.832966375350952,
      "epoch": 5.470156427510268,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002446545953665263,
      "loss": 5.2106,
      "mean_token_accuracy": 0.21713675558567047,
      "num_tokens": 126817765.0,
      "step": 62595
    },
    {
      "entropy": 5.893620300292969,
      "epoch": 5.470593375862973,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000244623719262678,
      "loss": 5.3218,
      "mean_token_accuracy": 0.2082711473107338,
      "num_tokens": 126827804.0,
      "step": 62600
    },
    {
      "entropy": 5.903477573394776,
      "epoch": 5.471030324215677,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002445928437414626,
      "loss": 5.1911,
      "mean_token_accuracy": 0.22291302680969238,
      "num_tokens": 126837677.0,
      "step": 62605
    },
    {
      "entropy": 5.897715854644775,
      "epoch": 5.471467272568383,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002445619688034725,
      "loss": 5.2723,
      "mean_token_accuracy": 0.21545772701501847,
      "num_tokens": 126847358.0,
      "step": 62610
    },
    {
      "entropy": 5.78155255317688,
      "epoch": 5.471904220921087,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002445310944492997,
      "loss": 5.1847,
      "mean_token_accuracy": 0.21809262484312059,
      "num_tokens": 126857638.0,
      "step": 62615
    },
    {
      "entropy": 5.898239326477051,
      "epoch": 5.4723411692737916,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024450022067953653,
      "loss": 5.2129,
      "mean_token_accuracy": 0.21163169145584107,
      "num_tokens": 126868342.0,
      "step": 62620
    },
    {
      "entropy": 5.8786077976226805,
      "epoch": 5.472778117626497,
      "grad_norm": 1.015625,
      "learning_rate": 0.00024446934749477507,
      "loss": 5.2145,
      "mean_token_accuracy": 0.21890401244163513,
      "num_tokens": 126878267.0,
      "step": 62625
    },
    {
      "entropy": 5.818154001235962,
      "epoch": 5.473215065979201,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002444384748956075,
      "loss": 5.2132,
      "mean_token_accuracy": 0.2186027780175209,
      "num_tokens": 126888200.0,
      "step": 62630
    },
    {
      "entropy": 5.79249210357666,
      "epoch": 5.473652014331906,
      "grad_norm": 1.0625,
      "learning_rate": 0.00024440760288262607,
      "loss": 5.1155,
      "mean_token_accuracy": 0.22783997356891633,
      "num_tokens": 126898515.0,
      "step": 62635
    },
    {
      "entropy": 5.846873378753662,
      "epoch": 5.47408896268461,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002443767314564228,
      "loss": 5.2317,
      "mean_token_accuracy": 0.22103234082460405,
      "num_tokens": 126908211.0,
      "step": 62640
    },
    {
      "entropy": 5.877137899398804,
      "epoch": 5.474525911037316,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002443458606175899,
      "loss": 5.2043,
      "mean_token_accuracy": 0.2177169293165207,
      "num_tokens": 126917734.0,
      "step": 62645
    },
    {
      "entropy": 5.894374895095825,
      "epoch": 5.47496285939002,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024431499036671944,
      "loss": 5.2157,
      "mean_token_accuracy": 0.20891683995723725,
      "num_tokens": 126927827.0,
      "step": 62650
    },
    {
      "entropy": 5.842324686050415,
      "epoch": 5.475399807742725,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002442841207044035,
      "loss": 5.1961,
      "mean_token_accuracy": 0.2171446844935417,
      "num_tokens": 126937844.0,
      "step": 62655
    },
    {
      "entropy": 5.858688926696777,
      "epoch": 5.475836756095429,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002442532516312343,
      "loss": 5.1332,
      "mean_token_accuracy": 0.21868069767951964,
      "num_tokens": 126947066.0,
      "step": 62660
    },
    {
      "entropy": 5.805534553527832,
      "epoch": 5.476273704448134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024422238314780384,
      "loss": 5.2161,
      "mean_token_accuracy": 0.21843433529138565,
      "num_tokens": 126957809.0,
      "step": 62665
    },
    {
      "entropy": 5.878379487991333,
      "epoch": 5.476710652800839,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024419151525470413,
      "loss": 5.2501,
      "mean_token_accuracy": 0.21748052388429642,
      "num_tokens": 126967242.0,
      "step": 62670
    },
    {
      "entropy": 5.858044767379761,
      "epoch": 5.477147601153543,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002441606479525273,
      "loss": 5.1465,
      "mean_token_accuracy": 0.21996057033538818,
      "num_tokens": 126976956.0,
      "step": 62675
    },
    {
      "entropy": 5.736375522613526,
      "epoch": 5.477584549506249,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002441297812418654,
      "loss": 5.0756,
      "mean_token_accuracy": 0.23142136335372926,
      "num_tokens": 126986369.0,
      "step": 62680
    },
    {
      "entropy": 5.86336350440979,
      "epoch": 5.478021497858953,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024409891512331056,
      "loss": 5.2262,
      "mean_token_accuracy": 0.21207311898469924,
      "num_tokens": 126995383.0,
      "step": 62685
    },
    {
      "entropy": 5.8120777130126955,
      "epoch": 5.478458446211658,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024406804959745465,
      "loss": 5.1233,
      "mean_token_accuracy": 0.22734883278608323,
      "num_tokens": 127005441.0,
      "step": 62690
    },
    {
      "entropy": 5.883257961273193,
      "epoch": 5.478895394564362,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024403718466488978,
      "loss": 5.2047,
      "mean_token_accuracy": 0.21730054169893265,
      "num_tokens": 127015882.0,
      "step": 62695
    },
    {
      "entropy": 5.842234897613525,
      "epoch": 5.4793323429170675,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002440063203262079,
      "loss": 5.1965,
      "mean_token_accuracy": 0.22086513489484788,
      "num_tokens": 127025854.0,
      "step": 62700
    },
    {
      "entropy": 5.864479827880859,
      "epoch": 5.479769291269772,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024397545658200106,
      "loss": 5.2624,
      "mean_token_accuracy": 0.21162248402833939,
      "num_tokens": 127036411.0,
      "step": 62705
    },
    {
      "entropy": 5.856620168685913,
      "epoch": 5.480206239622476,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00024394459343286113,
      "loss": 5.2519,
      "mean_token_accuracy": 0.2159654378890991,
      "num_tokens": 127047097.0,
      "step": 62710
    },
    {
      "entropy": 5.861390733718872,
      "epoch": 5.480643187975181,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024391373087938025,
      "loss": 5.1598,
      "mean_token_accuracy": 0.2192693293094635,
      "num_tokens": 127056666.0,
      "step": 62715
    },
    {
      "entropy": 5.815508890151977,
      "epoch": 5.481080136327886,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024388286892215033,
      "loss": 5.1994,
      "mean_token_accuracy": 0.21302491128444673,
      "num_tokens": 127066553.0,
      "step": 62720
    },
    {
      "entropy": 5.828025674819946,
      "epoch": 5.481517084680591,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00024385200756176328,
      "loss": 5.1648,
      "mean_token_accuracy": 0.21609130799770354,
      "num_tokens": 127076242.0,
      "step": 62725
    },
    {
      "entropy": 5.848207139968872,
      "epoch": 5.481954033033295,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024382114679881106,
      "loss": 5.206,
      "mean_token_accuracy": 0.2150135025382042,
      "num_tokens": 127085907.0,
      "step": 62730
    },
    {
      "entropy": 5.8358582019805905,
      "epoch": 5.482390981386001,
      "grad_norm": 1.125,
      "learning_rate": 0.00024379028663388554,
      "loss": 5.1476,
      "mean_token_accuracy": 0.22170584350824357,
      "num_tokens": 127095811.0,
      "step": 62735
    },
    {
      "entropy": 5.7355842113494875,
      "epoch": 5.482827929738705,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002437594270675787,
      "loss": 5.1215,
      "mean_token_accuracy": 0.23554034233093263,
      "num_tokens": 127105418.0,
      "step": 62740
    },
    {
      "entropy": 5.763470411300659,
      "epoch": 5.4832648780914095,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024372856810048245,
      "loss": 5.0482,
      "mean_token_accuracy": 0.23000542521476747,
      "num_tokens": 127114809.0,
      "step": 62745
    },
    {
      "entropy": 5.829774999618531,
      "epoch": 5.483701826444114,
      "grad_norm": 1.0625,
      "learning_rate": 0.00024369770973318856,
      "loss": 5.2209,
      "mean_token_accuracy": 0.21207588911056519,
      "num_tokens": 127124173.0,
      "step": 62750
    },
    {
      "entropy": 5.843751287460327,
      "epoch": 5.484138774796819,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00024366685196628907,
      "loss": 5.1053,
      "mean_token_accuracy": 0.23221080452203752,
      "num_tokens": 127134846.0,
      "step": 62755
    },
    {
      "entropy": 5.899185276031494,
      "epoch": 5.484575723149524,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024363599480037575,
      "loss": 5.2203,
      "mean_token_accuracy": 0.219889335334301,
      "num_tokens": 127144742.0,
      "step": 62760
    },
    {
      "entropy": 5.869937515258789,
      "epoch": 5.485012671502228,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024360513823604046,
      "loss": 5.1297,
      "mean_token_accuracy": 0.22180704176425933,
      "num_tokens": 127154781.0,
      "step": 62765
    },
    {
      "entropy": 5.880482769012451,
      "epoch": 5.485449619854933,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024357428227387513,
      "loss": 5.2882,
      "mean_token_accuracy": 0.20705097019672394,
      "num_tokens": 127164594.0,
      "step": 62770
    },
    {
      "entropy": 5.851327037811279,
      "epoch": 5.485886568207638,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002435434269144715,
      "loss": 5.2081,
      "mean_token_accuracy": 0.2164146348834038,
      "num_tokens": 127174596.0,
      "step": 62775
    },
    {
      "entropy": 5.843204307556152,
      "epoch": 5.486323516560343,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024351257215842143,
      "loss": 5.1965,
      "mean_token_accuracy": 0.21819870173931122,
      "num_tokens": 127184774.0,
      "step": 62780
    },
    {
      "entropy": 5.876123285293579,
      "epoch": 5.486760464913047,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024348171800631668,
      "loss": 5.1607,
      "mean_token_accuracy": 0.21901041269302368,
      "num_tokens": 127193933.0,
      "step": 62785
    },
    {
      "entropy": 5.8310130596160885,
      "epoch": 5.487197413265752,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002434508644587492,
      "loss": 5.1571,
      "mean_token_accuracy": 0.2195672556757927,
      "num_tokens": 127203607.0,
      "step": 62790
    },
    {
      "entropy": 5.844692182540894,
      "epoch": 5.487634361618457,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002434200115163106,
      "loss": 5.2308,
      "mean_token_accuracy": 0.2085227131843567,
      "num_tokens": 127212694.0,
      "step": 62795
    },
    {
      "entropy": 5.879489707946777,
      "epoch": 5.488071309971161,
      "grad_norm": 1.125,
      "learning_rate": 0.00024338915917959276,
      "loss": 5.2376,
      "mean_token_accuracy": 0.21241752207279205,
      "num_tokens": 127222725.0,
      "step": 62800
    },
    {
      "entropy": 5.918185424804688,
      "epoch": 5.488508258323866,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024335830744918736,
      "loss": 5.2253,
      "mean_token_accuracy": 0.21568699181079865,
      "num_tokens": 127233092.0,
      "step": 62805
    },
    {
      "entropy": 5.88067307472229,
      "epoch": 5.488945206676571,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024332745632568625,
      "loss": 5.138,
      "mean_token_accuracy": 0.2222173810005188,
      "num_tokens": 127242339.0,
      "step": 62810
    },
    {
      "entropy": 5.836260271072388,
      "epoch": 5.489382155029276,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00024329660580968104,
      "loss": 5.164,
      "mean_token_accuracy": 0.2237920194864273,
      "num_tokens": 127254033.0,
      "step": 62815
    },
    {
      "entropy": 5.8586958885192875,
      "epoch": 5.48981910338198,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002432657559017637,
      "loss": 5.1735,
      "mean_token_accuracy": 0.21237639486789703,
      "num_tokens": 127263420.0,
      "step": 62820
    },
    {
      "entropy": 5.78330717086792,
      "epoch": 5.490256051734685,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002432349066025257,
      "loss": 5.1511,
      "mean_token_accuracy": 0.22862517237663268,
      "num_tokens": 127273368.0,
      "step": 62825
    },
    {
      "entropy": 5.837042284011841,
      "epoch": 5.49069300008739,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002432040579125588,
      "loss": 5.2278,
      "mean_token_accuracy": 0.21263562589883805,
      "num_tokens": 127283656.0,
      "step": 62830
    },
    {
      "entropy": 5.8630129337310795,
      "epoch": 5.491129948440094,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024317320983245478,
      "loss": 5.259,
      "mean_token_accuracy": 0.20743469893932343,
      "num_tokens": 127293188.0,
      "step": 62835
    },
    {
      "entropy": 5.900797176361084,
      "epoch": 5.491566896792799,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024314236236280525,
      "loss": 5.189,
      "mean_token_accuracy": 0.22161209732294082,
      "num_tokens": 127303113.0,
      "step": 62840
    },
    {
      "entropy": 5.886916160583496,
      "epoch": 5.492003845145504,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024311151550420192,
      "loss": 5.1567,
      "mean_token_accuracy": 0.21906621307134627,
      "num_tokens": 127312653.0,
      "step": 62845
    },
    {
      "entropy": 5.879986667633057,
      "epoch": 5.492440793498209,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024308066925723642,
      "loss": 5.2312,
      "mean_token_accuracy": 0.2088206797838211,
      "num_tokens": 127322749.0,
      "step": 62850
    },
    {
      "entropy": 5.893346738815308,
      "epoch": 5.492877741850913,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002430498236225004,
      "loss": 5.2608,
      "mean_token_accuracy": 0.20864401757717133,
      "num_tokens": 127333832.0,
      "step": 62855
    },
    {
      "entropy": 5.847002935409546,
      "epoch": 5.493314690203618,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024301897860058548,
      "loss": 5.1433,
      "mean_token_accuracy": 0.2270314708352089,
      "num_tokens": 127343535.0,
      "step": 62860
    },
    {
      "entropy": 5.87496919631958,
      "epoch": 5.493751638556323,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024298813419208333,
      "loss": 5.2266,
      "mean_token_accuracy": 0.21104514896869658,
      "num_tokens": 127352990.0,
      "step": 62865
    },
    {
      "entropy": 5.880280876159668,
      "epoch": 5.4941885869090274,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024295729039758562,
      "loss": 5.2702,
      "mean_token_accuracy": 0.2141884371638298,
      "num_tokens": 127362900.0,
      "step": 62870
    },
    {
      "entropy": 5.894758319854736,
      "epoch": 5.494625535261732,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00024292644721768375,
      "loss": 5.2047,
      "mean_token_accuracy": 0.2079462707042694,
      "num_tokens": 127372910.0,
      "step": 62875
    },
    {
      "entropy": 5.807213211059571,
      "epoch": 5.495062483614436,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002428956046529695,
      "loss": 5.0918,
      "mean_token_accuracy": 0.2249908849596977,
      "num_tokens": 127382968.0,
      "step": 62880
    },
    {
      "entropy": 5.848834753036499,
      "epoch": 5.495499431967142,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024286476270403434,
      "loss": 5.2255,
      "mean_token_accuracy": 0.21522408723831177,
      "num_tokens": 127393151.0,
      "step": 62885
    },
    {
      "entropy": 5.898263835906983,
      "epoch": 5.495936380319846,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024283392137146992,
      "loss": 5.2215,
      "mean_token_accuracy": 0.21051024198532103,
      "num_tokens": 127403016.0,
      "step": 62890
    },
    {
      "entropy": 5.8125440120697025,
      "epoch": 5.496373328672551,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002428030806558677,
      "loss": 5.1423,
      "mean_token_accuracy": 0.22514323890209198,
      "num_tokens": 127412934.0,
      "step": 62895
    },
    {
      "entropy": 5.805845642089844,
      "epoch": 5.496810277025256,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024277224055781926,
      "loss": 5.1671,
      "mean_token_accuracy": 0.22417303621768953,
      "num_tokens": 127422283.0,
      "step": 62900
    },
    {
      "entropy": 5.79053316116333,
      "epoch": 5.4972472253779605,
      "grad_norm": 1.046875,
      "learning_rate": 0.00024274140107791614,
      "loss": 5.1507,
      "mean_token_accuracy": 0.22013373970985411,
      "num_tokens": 127431739.0,
      "step": 62905
    },
    {
      "entropy": 5.805102157592773,
      "epoch": 5.497684173730665,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024271056221674987,
      "loss": 5.1687,
      "mean_token_accuracy": 0.22257886230945587,
      "num_tokens": 127441637.0,
      "step": 62910
    },
    {
      "entropy": 5.851957130432129,
      "epoch": 5.4981211220833694,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024267972397491196,
      "loss": 5.1777,
      "mean_token_accuracy": 0.21699855774641036,
      "num_tokens": 127452523.0,
      "step": 62915
    },
    {
      "entropy": 5.942371654510498,
      "epoch": 5.498558070436075,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024264888635299386,
      "loss": 5.2259,
      "mean_token_accuracy": 0.22026608139276505,
      "num_tokens": 127463235.0,
      "step": 62920
    },
    {
      "entropy": 5.869737148284912,
      "epoch": 5.498995018788779,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024261804935158705,
      "loss": 5.1452,
      "mean_token_accuracy": 0.22524067014455795,
      "num_tokens": 127472663.0,
      "step": 62925
    },
    {
      "entropy": 5.8772436618804935,
      "epoch": 5.499431967141484,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000242587212971283,
      "loss": 5.271,
      "mean_token_accuracy": 0.2130527451634407,
      "num_tokens": 127483368.0,
      "step": 62930
    },
    {
      "entropy": 5.849616622924804,
      "epoch": 5.499868915494188,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002425563772126732,
      "loss": 5.2085,
      "mean_token_accuracy": 0.21426059305667877,
      "num_tokens": 127493075.0,
      "step": 62935
    },
    {
      "entropy": 5.788430547714233,
      "epoch": 5.500305863846894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00024252554207634914,
      "loss": 5.1701,
      "mean_token_accuracy": 0.21993221938610077,
      "num_tokens": 127503035.0,
      "step": 62940
    },
    {
      "entropy": 5.899667358398437,
      "epoch": 5.500742812199598,
      "grad_norm": 1.25,
      "learning_rate": 0.00024249470756290215,
      "loss": 5.2031,
      "mean_token_accuracy": 0.2137356922030449,
      "num_tokens": 127513104.0,
      "step": 62945
    },
    {
      "entropy": 5.826998853683472,
      "epoch": 5.5011797605523025,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024246387367292372,
      "loss": 5.1264,
      "mean_token_accuracy": 0.23051348179578782,
      "num_tokens": 127522983.0,
      "step": 62950
    },
    {
      "entropy": 5.831675577163696,
      "epoch": 5.501616708905008,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002424330404070052,
      "loss": 5.1446,
      "mean_token_accuracy": 0.21995947360992432,
      "num_tokens": 127533151.0,
      "step": 62955
    },
    {
      "entropy": 5.792768764495849,
      "epoch": 5.502053657257712,
      "grad_norm": 1.125,
      "learning_rate": 0.00024240220776573806,
      "loss": 5.1285,
      "mean_token_accuracy": 0.2204357534646988,
      "num_tokens": 127542778.0,
      "step": 62960
    },
    {
      "entropy": 5.787067699432373,
      "epoch": 5.502490605610417,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002423713757497137,
      "loss": 5.209,
      "mean_token_accuracy": 0.22153698056936263,
      "num_tokens": 127552599.0,
      "step": 62965
    },
    {
      "entropy": 5.818133068084717,
      "epoch": 5.502927553963121,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002423405443595234,
      "loss": 5.1628,
      "mean_token_accuracy": 0.22275475561618804,
      "num_tokens": 127563402.0,
      "step": 62970
    },
    {
      "entropy": 5.952702283859253,
      "epoch": 5.503364502315827,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024230971359575853,
      "loss": 5.2746,
      "mean_token_accuracy": 0.2109834685921669,
      "num_tokens": 127574456.0,
      "step": 62975
    },
    {
      "entropy": 5.859111499786377,
      "epoch": 5.503801450668531,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024227888345901055,
      "loss": 5.1843,
      "mean_token_accuracy": 0.21484122127294542,
      "num_tokens": 127584678.0,
      "step": 62980
    },
    {
      "entropy": 5.8567475318908695,
      "epoch": 5.504238399021236,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002422480539498707,
      "loss": 5.1897,
      "mean_token_accuracy": 0.2238506406545639,
      "num_tokens": 127594019.0,
      "step": 62985
    },
    {
      "entropy": 5.826978874206543,
      "epoch": 5.50467534737394,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002422172250689304,
      "loss": 5.2586,
      "mean_token_accuracy": 0.21775880157947541,
      "num_tokens": 127604794.0,
      "step": 62990
    },
    {
      "entropy": 5.865322256088257,
      "epoch": 5.505112295726645,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024218639681678085,
      "loss": 5.2032,
      "mean_token_accuracy": 0.21398858428001405,
      "num_tokens": 127614566.0,
      "step": 62995
    },
    {
      "entropy": 5.824462127685547,
      "epoch": 5.50554924407935,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024215556919401343,
      "loss": 5.1762,
      "mean_token_accuracy": 0.219043830037117,
      "num_tokens": 127624923.0,
      "step": 63000
    },
    {
      "epoch": 5.50554924407935,
      "eval_entropy": 5.579384114022437,
      "eval_loss": 5.300252437591553,
      "eval_mean_token_accuracy": 0.21989593003528896,
      "eval_num_tokens": 127624923.0,
      "eval_runtime": 27.8195,
      "eval_samples_per_second": 1330.504,
      "eval_steps_per_second": 166.322,
      "step": 63000
    },
    {
      "entropy": 5.771523571014404,
      "epoch": 5.505986192432054,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024212474220121932,
      "loss": 5.1625,
      "mean_token_accuracy": 0.22690026760101317,
      "num_tokens": 127634592.0,
      "step": 63005
    },
    {
      "entropy": 5.801569080352783,
      "epoch": 5.50642314078476,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024209391583898993,
      "loss": 5.2184,
      "mean_token_accuracy": 0.21757798492908478,
      "num_tokens": 127645476.0,
      "step": 63010
    },
    {
      "entropy": 5.862243223190307,
      "epoch": 5.506860089137464,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024206309010791656,
      "loss": 5.1755,
      "mean_token_accuracy": 0.22206952571868896,
      "num_tokens": 127654900.0,
      "step": 63015
    },
    {
      "entropy": 5.8246084690094,
      "epoch": 5.507297037490169,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024203226500859026,
      "loss": 5.0962,
      "mean_token_accuracy": 0.22352208495140075,
      "num_tokens": 127664311.0,
      "step": 63020
    },
    {
      "entropy": 5.830549049377441,
      "epoch": 5.507733985842873,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024200144054160245,
      "loss": 5.1967,
      "mean_token_accuracy": 0.21837690174579621,
      "num_tokens": 127674619.0,
      "step": 63025
    },
    {
      "entropy": 5.8446990013122555,
      "epoch": 5.5081709341955785,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002419706167075443,
      "loss": 5.1186,
      "mean_token_accuracy": 0.21572850048542022,
      "num_tokens": 127683468.0,
      "step": 63030
    },
    {
      "entropy": 5.876030540466308,
      "epoch": 5.508607882548283,
      "grad_norm": 1.015625,
      "learning_rate": 0.000241939793507007,
      "loss": 5.2066,
      "mean_token_accuracy": 0.21581943482160568,
      "num_tokens": 127694027.0,
      "step": 63035
    },
    {
      "entropy": 5.8801360607147215,
      "epoch": 5.509044830900987,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024190897094058183,
      "loss": 5.2298,
      "mean_token_accuracy": 0.21397310346364976,
      "num_tokens": 127704775.0,
      "step": 63040
    },
    {
      "entropy": 5.849109935760498,
      "epoch": 5.509481779253692,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024187814900885992,
      "loss": 5.2118,
      "mean_token_accuracy": 0.21320080906152725,
      "num_tokens": 127715321.0,
      "step": 63045
    },
    {
      "entropy": 5.851075792312622,
      "epoch": 5.509918727606397,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024184732771243244,
      "loss": 5.1793,
      "mean_token_accuracy": 0.22398952394723892,
      "num_tokens": 127725180.0,
      "step": 63050
    },
    {
      "entropy": 5.85295934677124,
      "epoch": 5.510355675959102,
      "grad_norm": 1.046875,
      "learning_rate": 0.00024181650705189058,
      "loss": 5.1731,
      "mean_token_accuracy": 0.22133755832910537,
      "num_tokens": 127735131.0,
      "step": 63055
    },
    {
      "entropy": 5.863603305816651,
      "epoch": 5.510792624311806,
      "grad_norm": 1.125,
      "learning_rate": 0.0002417856870278256,
      "loss": 5.2638,
      "mean_token_accuracy": 0.22010251730680466,
      "num_tokens": 127744761.0,
      "step": 63060
    },
    {
      "entropy": 5.846759891510009,
      "epoch": 5.5112295726645115,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00024175486764082833,
      "loss": 5.2108,
      "mean_token_accuracy": 0.21834613680839537,
      "num_tokens": 127756316.0,
      "step": 63065
    },
    {
      "entropy": 5.8161193370819095,
      "epoch": 5.511666521017216,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024172404889149024,
      "loss": 5.1997,
      "mean_token_accuracy": 0.22261194586753846,
      "num_tokens": 127767625.0,
      "step": 63070
    },
    {
      "entropy": 5.9149498462677,
      "epoch": 5.5121034693699205,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024169323078040246,
      "loss": 5.1988,
      "mean_token_accuracy": 0.21739398539066315,
      "num_tokens": 127779170.0,
      "step": 63075
    },
    {
      "entropy": 5.899806022644043,
      "epoch": 5.512540417722625,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002416624133081558,
      "loss": 5.2039,
      "mean_token_accuracy": 0.21868074536323548,
      "num_tokens": 127790052.0,
      "step": 63080
    },
    {
      "entropy": 5.870931625366211,
      "epoch": 5.512977366075329,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00024163159647534155,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21142965853214263,
      "num_tokens": 127801153.0,
      "step": 63085
    },
    {
      "entropy": 5.782386398315429,
      "epoch": 5.513414314428035,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024160078028255083,
      "loss": 5.1308,
      "mean_token_accuracy": 0.22634016275405883,
      "num_tokens": 127811132.0,
      "step": 63090
    },
    {
      "entropy": 5.836400318145752,
      "epoch": 5.513851262780739,
      "grad_norm": 1.125,
      "learning_rate": 0.00024156996473037457,
      "loss": 5.167,
      "mean_token_accuracy": 0.22242638021707534,
      "num_tokens": 127820834.0,
      "step": 63095
    },
    {
      "entropy": 5.907649993896484,
      "epoch": 5.514288211133444,
      "grad_norm": 1.171875,
      "learning_rate": 0.000241539149819404,
      "loss": 5.2307,
      "mean_token_accuracy": 0.21553981602191924,
      "num_tokens": 127830291.0,
      "step": 63100
    },
    {
      "entropy": 5.948049306869507,
      "epoch": 5.514725159486149,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024150833555023,
      "loss": 5.2446,
      "mean_token_accuracy": 0.21234794706106186,
      "num_tokens": 127841154.0,
      "step": 63105
    },
    {
      "entropy": 5.936449337005615,
      "epoch": 5.5151621078388535,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024147752192344368,
      "loss": 5.2652,
      "mean_token_accuracy": 0.21313405632972718,
      "num_tokens": 127852246.0,
      "step": 63110
    },
    {
      "entropy": 5.86778244972229,
      "epoch": 5.515599056191558,
      "grad_norm": 1.015625,
      "learning_rate": 0.00024144670893963606,
      "loss": 5.2311,
      "mean_token_accuracy": 0.21480853855609894,
      "num_tokens": 127863980.0,
      "step": 63115
    },
    {
      "entropy": 5.90272216796875,
      "epoch": 5.516036004544263,
      "grad_norm": 1.25,
      "learning_rate": 0.0002414158965993981,
      "loss": 5.1694,
      "mean_token_accuracy": 0.2167251691222191,
      "num_tokens": 127873086.0,
      "step": 63120
    },
    {
      "entropy": 5.875883960723877,
      "epoch": 5.516472952896968,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024138508490332094,
      "loss": 5.2479,
      "mean_token_accuracy": 0.209800885617733,
      "num_tokens": 127884485.0,
      "step": 63125
    },
    {
      "entropy": 5.854984855651855,
      "epoch": 5.516909901249672,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024135427385199538,
      "loss": 5.1702,
      "mean_token_accuracy": 0.22051215618848802,
      "num_tokens": 127894199.0,
      "step": 63130
    },
    {
      "entropy": 5.791103219985962,
      "epoch": 5.517346849602377,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024132346344601252,
      "loss": 5.1019,
      "mean_token_accuracy": 0.22564944624900818,
      "num_tokens": 127903957.0,
      "step": 63135
    },
    {
      "entropy": 5.876287984848022,
      "epoch": 5.517783797955081,
      "grad_norm": 1.03125,
      "learning_rate": 0.00024129265368596326,
      "loss": 5.2177,
      "mean_token_accuracy": 0.21635696142911912,
      "num_tokens": 127915271.0,
      "step": 63140
    },
    {
      "entropy": 5.943780565261841,
      "epoch": 5.518220746307787,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024126184457243856,
      "loss": 5.265,
      "mean_token_accuracy": 0.21316516697406768,
      "num_tokens": 127925705.0,
      "step": 63145
    },
    {
      "entropy": 5.932550668716431,
      "epoch": 5.518657694660491,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024123103610602942,
      "loss": 5.2493,
      "mean_token_accuracy": 0.21134878993034362,
      "num_tokens": 127935368.0,
      "step": 63150
    },
    {
      "entropy": 5.903613901138305,
      "epoch": 5.5190946430131955,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024120022828732663,
      "loss": 5.2352,
      "mean_token_accuracy": 0.21306075602769853,
      "num_tokens": 127944629.0,
      "step": 63155
    },
    {
      "entropy": 5.832472276687622,
      "epoch": 5.519531591365901,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00024116942111692118,
      "loss": 5.1813,
      "mean_token_accuracy": 0.22218823283910752,
      "num_tokens": 127954073.0,
      "step": 63160
    },
    {
      "entropy": 5.791061353683472,
      "epoch": 5.519968539718605,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00024113861459540398,
      "loss": 5.1016,
      "mean_token_accuracy": 0.2230896234512329,
      "num_tokens": 127964344.0,
      "step": 63165
    },
    {
      "entropy": 5.844797134399414,
      "epoch": 5.52040548807131,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00024110780872336584,
      "loss": 5.132,
      "mean_token_accuracy": 0.2221521958708763,
      "num_tokens": 127974710.0,
      "step": 63170
    },
    {
      "entropy": 5.8382548809051515,
      "epoch": 5.520842436424015,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024107700350139782,
      "loss": 5.2055,
      "mean_token_accuracy": 0.2191100686788559,
      "num_tokens": 127986120.0,
      "step": 63175
    },
    {
      "entropy": 5.89903416633606,
      "epoch": 5.52127938477672,
      "grad_norm": 1.015625,
      "learning_rate": 0.00024104619893009055,
      "loss": 5.2411,
      "mean_token_accuracy": 0.21333095282316208,
      "num_tokens": 127996610.0,
      "step": 63180
    },
    {
      "entropy": 5.898902940750122,
      "epoch": 5.521716333129424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000241015395010035,
      "loss": 5.2801,
      "mean_token_accuracy": 0.2129032149910927,
      "num_tokens": 128005780.0,
      "step": 63185
    },
    {
      "entropy": 5.847370624542236,
      "epoch": 5.522153281482129,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002409845917418219,
      "loss": 5.1893,
      "mean_token_accuracy": 0.2136075094342232,
      "num_tokens": 128015266.0,
      "step": 63190
    },
    {
      "entropy": 5.826754951477051,
      "epoch": 5.522590229834833,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002409537891260423,
      "loss": 5.1372,
      "mean_token_accuracy": 0.21661849915981293,
      "num_tokens": 128024820.0,
      "step": 63195
    },
    {
      "entropy": 5.8963405132293705,
      "epoch": 5.523027178187538,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024092298716328682,
      "loss": 5.3008,
      "mean_token_accuracy": 0.2164151519536972,
      "num_tokens": 128035208.0,
      "step": 63200
    },
    {
      "entropy": 5.853041696548462,
      "epoch": 5.523464126540243,
      "grad_norm": 1.0625,
      "learning_rate": 0.00024089218585414624,
      "loss": 5.2078,
      "mean_token_accuracy": 0.21715140640735625,
      "num_tokens": 128045406.0,
      "step": 63205
    },
    {
      "entropy": 5.836138963699341,
      "epoch": 5.523901074892947,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002408613851992114,
      "loss": 5.2131,
      "mean_token_accuracy": 0.21652154326438905,
      "num_tokens": 128054543.0,
      "step": 63210
    },
    {
      "entropy": 5.885333728790283,
      "epoch": 5.524338023245653,
      "grad_norm": 1.046875,
      "learning_rate": 0.00024083058519907313,
      "loss": 5.1936,
      "mean_token_accuracy": 0.2168925568461418,
      "num_tokens": 128064843.0,
      "step": 63215
    },
    {
      "entropy": 5.8697291851043705,
      "epoch": 5.524774971598357,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024079978585432216,
      "loss": 5.2078,
      "mean_token_accuracy": 0.21506910026073456,
      "num_tokens": 128075169.0,
      "step": 63220
    },
    {
      "entropy": 5.847352313995361,
      "epoch": 5.525211919951062,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024076898716554918,
      "loss": 5.1822,
      "mean_token_accuracy": 0.21859415769577026,
      "num_tokens": 128085268.0,
      "step": 63225
    },
    {
      "entropy": 5.76501317024231,
      "epoch": 5.525648868303766,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024073818913334494,
      "loss": 5.111,
      "mean_token_accuracy": 0.22790595442056655,
      "num_tokens": 128096788.0,
      "step": 63230
    },
    {
      "entropy": 5.873884677886963,
      "epoch": 5.5260858166564715,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024070739175830025,
      "loss": 5.1732,
      "mean_token_accuracy": 0.21702795028686522,
      "num_tokens": 128105870.0,
      "step": 63235
    },
    {
      "entropy": 5.8678062915802,
      "epoch": 5.526522765009176,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002406765950410057,
      "loss": 5.2425,
      "mean_token_accuracy": 0.21853168904781342,
      "num_tokens": 128115861.0,
      "step": 63240
    },
    {
      "entropy": 5.867905378341675,
      "epoch": 5.52695971336188,
      "grad_norm": 1.125,
      "learning_rate": 0.00024064579898205212,
      "loss": 5.226,
      "mean_token_accuracy": 0.21383107006549834,
      "num_tokens": 128126876.0,
      "step": 63245
    },
    {
      "entropy": 5.838128328323364,
      "epoch": 5.527396661714585,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00024061500358203003,
      "loss": 5.173,
      "mean_token_accuracy": 0.2209769606590271,
      "num_tokens": 128136959.0,
      "step": 63250
    },
    {
      "entropy": 5.773875951766968,
      "epoch": 5.52783361006729,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024058420884153026,
      "loss": 5.1276,
      "mean_token_accuracy": 0.22562411278486252,
      "num_tokens": 128146777.0,
      "step": 63255
    },
    {
      "entropy": 5.681094026565551,
      "epoch": 5.528270558419995,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024055341476114333,
      "loss": 5.0431,
      "mean_token_accuracy": 0.2383737951517105,
      "num_tokens": 128156815.0,
      "step": 63260
    },
    {
      "entropy": 5.849923038482666,
      "epoch": 5.528707506772699,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024052262134145996,
      "loss": 5.2744,
      "mean_token_accuracy": 0.21796447336673735,
      "num_tokens": 128166687.0,
      "step": 63265
    },
    {
      "entropy": 5.872981691360474,
      "epoch": 5.5291444551254045,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024049182858307085,
      "loss": 5.1161,
      "mean_token_accuracy": 0.22243909984827043,
      "num_tokens": 128176534.0,
      "step": 63270
    },
    {
      "entropy": 5.88353419303894,
      "epoch": 5.529581403478109,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024046103648656648,
      "loss": 5.2115,
      "mean_token_accuracy": 0.21931845098733901,
      "num_tokens": 128186320.0,
      "step": 63275
    },
    {
      "entropy": 5.841065120697022,
      "epoch": 5.5300183518308135,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024043024505253753,
      "loss": 5.2046,
      "mean_token_accuracy": 0.22067221999168396,
      "num_tokens": 128195911.0,
      "step": 63280
    },
    {
      "entropy": 5.851698923110962,
      "epoch": 5.530455300183518,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024039945428157462,
      "loss": 5.1557,
      "mean_token_accuracy": 0.22219768315553665,
      "num_tokens": 128206413.0,
      "step": 63285
    },
    {
      "entropy": 5.8972588062286375,
      "epoch": 5.530892248536223,
      "grad_norm": 1.125,
      "learning_rate": 0.0002403686641742683,
      "loss": 5.2076,
      "mean_token_accuracy": 0.21704278290271758,
      "num_tokens": 128215822.0,
      "step": 63290
    },
    {
      "entropy": 5.84859185218811,
      "epoch": 5.531329196888928,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024033787473120915,
      "loss": 5.2144,
      "mean_token_accuracy": 0.21447362303733825,
      "num_tokens": 128225305.0,
      "step": 63295
    },
    {
      "entropy": 5.828509950637818,
      "epoch": 5.531766145241632,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024030708595298772,
      "loss": 5.145,
      "mean_token_accuracy": 0.22269807010889053,
      "num_tokens": 128235787.0,
      "step": 63300
    },
    {
      "entropy": 5.90304045677185,
      "epoch": 5.532203093594337,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024027629784019455,
      "loss": 5.2615,
      "mean_token_accuracy": 0.2120865523815155,
      "num_tokens": 128246855.0,
      "step": 63305
    },
    {
      "entropy": 5.815795040130615,
      "epoch": 5.532640041947042,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024024551039342025,
      "loss": 5.1811,
      "mean_token_accuracy": 0.21857484430074692,
      "num_tokens": 128258974.0,
      "step": 63310
    },
    {
      "entropy": 5.882314729690552,
      "epoch": 5.5330769902997465,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024021472361325519,
      "loss": 5.2035,
      "mean_token_accuracy": 0.21758367717266083,
      "num_tokens": 128269445.0,
      "step": 63315
    },
    {
      "entropy": 5.887620353698731,
      "epoch": 5.533513938652451,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024018393750029,
      "loss": 5.2597,
      "mean_token_accuracy": 0.21541216224431992,
      "num_tokens": 128279837.0,
      "step": 63320
    },
    {
      "entropy": 5.87345814704895,
      "epoch": 5.533950887005156,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002401531520551151,
      "loss": 5.1332,
      "mean_token_accuracy": 0.21689736992120742,
      "num_tokens": 128289607.0,
      "step": 63325
    },
    {
      "entropy": 5.870246839523316,
      "epoch": 5.534387835357861,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024012236727832104,
      "loss": 5.1928,
      "mean_token_accuracy": 0.21417636573314666,
      "num_tokens": 128300544.0,
      "step": 63330
    },
    {
      "entropy": 5.846699142456055,
      "epoch": 5.534824783710565,
      "grad_norm": 1.125,
      "learning_rate": 0.00024009158317049822,
      "loss": 5.182,
      "mean_token_accuracy": 0.2212500289082527,
      "num_tokens": 128311208.0,
      "step": 63335
    },
    {
      "entropy": 5.812546014785767,
      "epoch": 5.53526173206327,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024006079973223717,
      "loss": 5.1635,
      "mean_token_accuracy": 0.21825673133134843,
      "num_tokens": 128322050.0,
      "step": 63340
    },
    {
      "entropy": 5.876035737991333,
      "epoch": 5.535698680415975,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00024003001696412823,
      "loss": 5.2306,
      "mean_token_accuracy": 0.21870448291301728,
      "num_tokens": 128332774.0,
      "step": 63345
    },
    {
      "entropy": 5.82977523803711,
      "epoch": 5.53613562876868,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023999923486676196,
      "loss": 5.19,
      "mean_token_accuracy": 0.2179649904370308,
      "num_tokens": 128342840.0,
      "step": 63350
    },
    {
      "entropy": 5.770937871932984,
      "epoch": 5.536572577121384,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023996845344072864,
      "loss": 5.1132,
      "mean_token_accuracy": 0.22609151601791383,
      "num_tokens": 128351944.0,
      "step": 63355
    },
    {
      "entropy": 5.849227380752564,
      "epoch": 5.5370095254740885,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023993767268661875,
      "loss": 5.1983,
      "mean_token_accuracy": 0.2204376146197319,
      "num_tokens": 128362106.0,
      "step": 63360
    },
    {
      "entropy": 5.865212154388428,
      "epoch": 5.537446473826794,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023990689260502274,
      "loss": 5.2096,
      "mean_token_accuracy": 0.21726509630680085,
      "num_tokens": 128372354.0,
      "step": 63365
    },
    {
      "entropy": 5.885105419158935,
      "epoch": 5.537883422179498,
      "grad_norm": 1.125,
      "learning_rate": 0.0002398761131965308,
      "loss": 5.2081,
      "mean_token_accuracy": 0.2145190045237541,
      "num_tokens": 128381696.0,
      "step": 63370
    },
    {
      "entropy": 5.792347955703735,
      "epoch": 5.538320370532203,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023984533446173346,
      "loss": 5.0872,
      "mean_token_accuracy": 0.22778098434209823,
      "num_tokens": 128391340.0,
      "step": 63375
    },
    {
      "entropy": 5.75974063873291,
      "epoch": 5.538757318884908,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023981455640122108,
      "loss": 5.1365,
      "mean_token_accuracy": 0.23187617510557174,
      "num_tokens": 128401779.0,
      "step": 63380
    },
    {
      "entropy": 5.893540382385254,
      "epoch": 5.539194267237613,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002397837790155839,
      "loss": 5.2757,
      "mean_token_accuracy": 0.21315939724445343,
      "num_tokens": 128412797.0,
      "step": 63385
    },
    {
      "entropy": 5.876939344406128,
      "epoch": 5.539631215590317,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002397530023054123,
      "loss": 5.262,
      "mean_token_accuracy": 0.2077382206916809,
      "num_tokens": 128423873.0,
      "step": 63390
    },
    {
      "entropy": 5.892089939117431,
      "epoch": 5.540068163943022,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002397222262712966,
      "loss": 5.1858,
      "mean_token_accuracy": 0.22333684265613557,
      "num_tokens": 128433792.0,
      "step": 63395
    },
    {
      "entropy": 5.8652294158935545,
      "epoch": 5.540505112295727,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00023969145091382705,
      "loss": 5.1671,
      "mean_token_accuracy": 0.21557160168886186,
      "num_tokens": 128443742.0,
      "step": 63400
    },
    {
      "entropy": 5.877409505844116,
      "epoch": 5.540942060648431,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000239660676233594,
      "loss": 5.2595,
      "mean_token_accuracy": 0.21245871633291244,
      "num_tokens": 128453871.0,
      "step": 63405
    },
    {
      "entropy": 5.880755043029785,
      "epoch": 5.541379009001136,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002396299022311877,
      "loss": 5.2451,
      "mean_token_accuracy": 0.21249308735132216,
      "num_tokens": 128463889.0,
      "step": 63410
    },
    {
      "entropy": 5.868647718429566,
      "epoch": 5.54181595735384,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023959912890719837,
      "loss": 5.1553,
      "mean_token_accuracy": 0.22687858045101167,
      "num_tokens": 128473524.0,
      "step": 63415
    },
    {
      "entropy": 5.909854173660278,
      "epoch": 5.542252905706546,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023956835626221624,
      "loss": 5.2959,
      "mean_token_accuracy": 0.2029982849955559,
      "num_tokens": 128484209.0,
      "step": 63420
    },
    {
      "entropy": 5.8864133834838865,
      "epoch": 5.54268985405925,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023953758429683166,
      "loss": 5.2467,
      "mean_token_accuracy": 0.2186324715614319,
      "num_tokens": 128494380.0,
      "step": 63425
    },
    {
      "entropy": 5.922642374038697,
      "epoch": 5.543126802411955,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023950681301163486,
      "loss": 5.2795,
      "mean_token_accuracy": 0.20483413189649582,
      "num_tokens": 128505709.0,
      "step": 63430
    },
    {
      "entropy": 5.872191667556763,
      "epoch": 5.54356375076466,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023947604240721595,
      "loss": 5.2306,
      "mean_token_accuracy": 0.21749051958322524,
      "num_tokens": 128516082.0,
      "step": 63435
    },
    {
      "entropy": 5.844697189331055,
      "epoch": 5.5440006991173645,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023944527248416514,
      "loss": 5.1761,
      "mean_token_accuracy": 0.22105972319841385,
      "num_tokens": 128526519.0,
      "step": 63440
    },
    {
      "entropy": 5.8680100440979,
      "epoch": 5.544437647470069,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023941450324307267,
      "loss": 5.2072,
      "mean_token_accuracy": 0.22478398233652114,
      "num_tokens": 128536821.0,
      "step": 63445
    },
    {
      "entropy": 5.8976844310760494,
      "epoch": 5.544874595822773,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023938373468452862,
      "loss": 5.2253,
      "mean_token_accuracy": 0.2145645275712013,
      "num_tokens": 128547089.0,
      "step": 63450
    },
    {
      "entropy": 5.81972222328186,
      "epoch": 5.545311544175479,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002393529668091233,
      "loss": 5.1591,
      "mean_token_accuracy": 0.22082050740718842,
      "num_tokens": 128556763.0,
      "step": 63455
    },
    {
      "entropy": 5.904683923721313,
      "epoch": 5.545748492528183,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023932219961744666,
      "loss": 5.2829,
      "mean_token_accuracy": 0.20758814513683319,
      "num_tokens": 128566490.0,
      "step": 63460
    },
    {
      "entropy": 5.895102214813233,
      "epoch": 5.546185440880888,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023929143311008893,
      "loss": 5.1437,
      "mean_token_accuracy": 0.21747440695762635,
      "num_tokens": 128575616.0,
      "step": 63465
    },
    {
      "entropy": 5.830424356460571,
      "epoch": 5.546622389233592,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002392606672876403,
      "loss": 5.165,
      "mean_token_accuracy": 0.2250490590929985,
      "num_tokens": 128585358.0,
      "step": 63470
    },
    {
      "entropy": 5.779691934585571,
      "epoch": 5.5470593375862975,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023922990215069067,
      "loss": 5.1365,
      "mean_token_accuracy": 0.22769861668348312,
      "num_tokens": 128595718.0,
      "step": 63475
    },
    {
      "entropy": 5.800051355361939,
      "epoch": 5.547496285939002,
      "grad_norm": 1.046875,
      "learning_rate": 0.00023919913769983043,
      "loss": 5.1796,
      "mean_token_accuracy": 0.22510749995708465,
      "num_tokens": 128605555.0,
      "step": 63480
    },
    {
      "entropy": 5.90748348236084,
      "epoch": 5.5479332342917065,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002391683739356494,
      "loss": 5.2316,
      "mean_token_accuracy": 0.20973506867885588,
      "num_tokens": 128616871.0,
      "step": 63485
    },
    {
      "entropy": 5.940604257583618,
      "epoch": 5.548370182644412,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002391376108587377,
      "loss": 5.2528,
      "mean_token_accuracy": 0.21066011637449264,
      "num_tokens": 128626883.0,
      "step": 63490
    },
    {
      "entropy": 5.844066572189331,
      "epoch": 5.548807130997116,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023910684846968546,
      "loss": 5.1499,
      "mean_token_accuracy": 0.22259829044342042,
      "num_tokens": 128636705.0,
      "step": 63495
    },
    {
      "entropy": 5.79863429069519,
      "epoch": 5.549244079349821,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023907608676908265,
      "loss": 5.1384,
      "mean_token_accuracy": 0.21884691715240479,
      "num_tokens": 128646994.0,
      "step": 63500
    },
    {
      "entropy": 5.826951360702514,
      "epoch": 5.549681027702525,
      "grad_norm": 0.984375,
      "learning_rate": 0.00023904532575751936,
      "loss": 5.2126,
      "mean_token_accuracy": 0.21728143692016602,
      "num_tokens": 128657250.0,
      "step": 63505
    },
    {
      "entropy": 5.831587171554565,
      "epoch": 5.550117976055231,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023901456543558552,
      "loss": 5.1683,
      "mean_token_accuracy": 0.22536282539367675,
      "num_tokens": 128667114.0,
      "step": 63510
    },
    {
      "entropy": 5.823401737213135,
      "epoch": 5.550554924407935,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023898380580387118,
      "loss": 5.1805,
      "mean_token_accuracy": 0.2192960873246193,
      "num_tokens": 128676769.0,
      "step": 63515
    },
    {
      "entropy": 5.887738275527954,
      "epoch": 5.5509918727606395,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002389530468629663,
      "loss": 5.1896,
      "mean_token_accuracy": 0.22117000967264175,
      "num_tokens": 128688024.0,
      "step": 63520
    },
    {
      "entropy": 5.875977802276611,
      "epoch": 5.551428821113344,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023892228861346093,
      "loss": 5.2277,
      "mean_token_accuracy": 0.21917243897914887,
      "num_tokens": 128698725.0,
      "step": 63525
    },
    {
      "entropy": 5.854404449462891,
      "epoch": 5.551865769466049,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002388915310559449,
      "loss": 5.1572,
      "mean_token_accuracy": 0.21371169835329057,
      "num_tokens": 128708226.0,
      "step": 63530
    },
    {
      "entropy": 5.881818342208862,
      "epoch": 5.552302717818754,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00023886077419100826,
      "loss": 5.226,
      "mean_token_accuracy": 0.2101417988538742,
      "num_tokens": 128717949.0,
      "step": 63535
    },
    {
      "entropy": 5.856348943710327,
      "epoch": 5.552739666171458,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023883001801924087,
      "loss": 5.203,
      "mean_token_accuracy": 0.21698809862136842,
      "num_tokens": 128727813.0,
      "step": 63540
    },
    {
      "entropy": 5.850809574127197,
      "epoch": 5.553176614524164,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023879926254123275,
      "loss": 5.1585,
      "mean_token_accuracy": 0.22260628640651703,
      "num_tokens": 128737664.0,
      "step": 63545
    },
    {
      "entropy": 5.917582702636719,
      "epoch": 5.553613562876868,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023876850775757374,
      "loss": 5.2146,
      "mean_token_accuracy": 0.21519850343465804,
      "num_tokens": 128747602.0,
      "step": 63550
    },
    {
      "entropy": 5.901100063323975,
      "epoch": 5.554050511229573,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023873775366885369,
      "loss": 5.2534,
      "mean_token_accuracy": 0.21481017172336578,
      "num_tokens": 128757529.0,
      "step": 63555
    },
    {
      "entropy": 5.882526493072509,
      "epoch": 5.554487459582277,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023870700027566255,
      "loss": 5.2717,
      "mean_token_accuracy": 0.20926604717969893,
      "num_tokens": 128767999.0,
      "step": 63560
    },
    {
      "entropy": 5.834909200668335,
      "epoch": 5.554924407934982,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023867624757859015,
      "loss": 5.2122,
      "mean_token_accuracy": 0.22087569236755372,
      "num_tokens": 128778167.0,
      "step": 63565
    },
    {
      "entropy": 5.935021686553955,
      "epoch": 5.555361356287687,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023864549557822635,
      "loss": 5.2379,
      "mean_token_accuracy": 0.20765876024961472,
      "num_tokens": 128787938.0,
      "step": 63570
    },
    {
      "entropy": 5.893155574798584,
      "epoch": 5.555798304640391,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023861474427516106,
      "loss": 5.1409,
      "mean_token_accuracy": 0.2246684953570366,
      "num_tokens": 128798501.0,
      "step": 63575
    },
    {
      "entropy": 5.833809804916382,
      "epoch": 5.556235252993096,
      "grad_norm": 1.015625,
      "learning_rate": 0.00023858399366998394,
      "loss": 5.2377,
      "mean_token_accuracy": 0.2149544343352318,
      "num_tokens": 128809712.0,
      "step": 63580
    },
    {
      "entropy": 5.87627477645874,
      "epoch": 5.556672201345801,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023855324376328497,
      "loss": 5.2014,
      "mean_token_accuracy": 0.21839278787374497,
      "num_tokens": 128820237.0,
      "step": 63585
    },
    {
      "entropy": 5.931636953353882,
      "epoch": 5.557109149698506,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023852249455565383,
      "loss": 5.2482,
      "mean_token_accuracy": 0.21636037677526473,
      "num_tokens": 128829667.0,
      "step": 63590
    },
    {
      "entropy": 5.826652240753174,
      "epoch": 5.55754609805121,
      "grad_norm": 1.125,
      "learning_rate": 0.00023849174604768036,
      "loss": 5.1572,
      "mean_token_accuracy": 0.22551700621843337,
      "num_tokens": 128840910.0,
      "step": 63595
    },
    {
      "entropy": 5.842716884613037,
      "epoch": 5.5579830464039155,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023846099823995442,
      "loss": 5.1915,
      "mean_token_accuracy": 0.21744650602340698,
      "num_tokens": 128851229.0,
      "step": 63600
    },
    {
      "entropy": 5.778213596343994,
      "epoch": 5.55841999475662,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002384302511330656,
      "loss": 5.1377,
      "mean_token_accuracy": 0.22589361369609834,
      "num_tokens": 128861505.0,
      "step": 63605
    },
    {
      "entropy": 5.844314098358154,
      "epoch": 5.558856943109324,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023839950472760374,
      "loss": 5.1575,
      "mean_token_accuracy": 0.21735706478357314,
      "num_tokens": 128870894.0,
      "step": 63610
    },
    {
      "entropy": 5.869061517715454,
      "epoch": 5.559293891462029,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023836875902415855,
      "loss": 5.2571,
      "mean_token_accuracy": 0.21829288452863693,
      "num_tokens": 128881412.0,
      "step": 63615
    },
    {
      "entropy": 5.93348650932312,
      "epoch": 5.559730839814734,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023833801402331977,
      "loss": 5.231,
      "mean_token_accuracy": 0.21440369784832,
      "num_tokens": 128892153.0,
      "step": 63620
    },
    {
      "entropy": 5.856330108642578,
      "epoch": 5.560167788167439,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023830726972567712,
      "loss": 5.2267,
      "mean_token_accuracy": 0.21965937465429305,
      "num_tokens": 128901762.0,
      "step": 63625
    },
    {
      "entropy": 5.875362730026245,
      "epoch": 5.560604736520143,
      "grad_norm": 1.015625,
      "learning_rate": 0.00023827652613182026,
      "loss": 5.2899,
      "mean_token_accuracy": 0.2133667066693306,
      "num_tokens": 128912357.0,
      "step": 63630
    },
    {
      "entropy": 5.855581474304199,
      "epoch": 5.561041684872848,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023824578324233882,
      "loss": 5.1311,
      "mean_token_accuracy": 0.21673226803541185,
      "num_tokens": 128922153.0,
      "step": 63635
    },
    {
      "entropy": 5.799210834503174,
      "epoch": 5.561478633225553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023821504105782254,
      "loss": 5.139,
      "mean_token_accuracy": 0.22430564165115358,
      "num_tokens": 128932477.0,
      "step": 63640
    },
    {
      "entropy": 5.832653141021728,
      "epoch": 5.5619155815782575,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023818429957886113,
      "loss": 5.1835,
      "mean_token_accuracy": 0.21981499642133712,
      "num_tokens": 128943640.0,
      "step": 63645
    },
    {
      "entropy": 5.90344500541687,
      "epoch": 5.562352529930962,
      "grad_norm": 1.15625,
      "learning_rate": 0.000238153558806044,
      "loss": 5.2327,
      "mean_token_accuracy": 0.22023664712905883,
      "num_tokens": 128953593.0,
      "step": 63650
    },
    {
      "entropy": 5.878356552124023,
      "epoch": 5.562789478283667,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000238122818739961,
      "loss": 5.1771,
      "mean_token_accuracy": 0.21972069293260574,
      "num_tokens": 128963213.0,
      "step": 63655
    },
    {
      "entropy": 5.796129274368286,
      "epoch": 5.563226426636372,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002380920793812016,
      "loss": 5.1309,
      "mean_token_accuracy": 0.2291587382555008,
      "num_tokens": 128973178.0,
      "step": 63660
    },
    {
      "entropy": 5.787599420547485,
      "epoch": 5.563663374989076,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023806134073035552,
      "loss": 5.1645,
      "mean_token_accuracy": 0.2196995958685875,
      "num_tokens": 128983368.0,
      "step": 63665
    },
    {
      "entropy": 5.860134220123291,
      "epoch": 5.564100323341781,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023803060278801232,
      "loss": 5.2603,
      "mean_token_accuracy": 0.21175330579280854,
      "num_tokens": 128993570.0,
      "step": 63670
    },
    {
      "entropy": 5.871928119659424,
      "epoch": 5.564537271694486,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023799986555476143,
      "loss": 5.2184,
      "mean_token_accuracy": 0.21302065700292588,
      "num_tokens": 129004282.0,
      "step": 63675
    },
    {
      "entropy": 5.955371332168579,
      "epoch": 5.564974220047191,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023796912903119256,
      "loss": 5.2854,
      "mean_token_accuracy": 0.21087818145751952,
      "num_tokens": 129013541.0,
      "step": 63680
    },
    {
      "entropy": 5.922638845443726,
      "epoch": 5.565411168399895,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002379383932178952,
      "loss": 5.2857,
      "mean_token_accuracy": 0.2081062138080597,
      "num_tokens": 129023890.0,
      "step": 63685
    },
    {
      "entropy": 5.938042163848877,
      "epoch": 5.5658481167525995,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023790765811545884,
      "loss": 5.3047,
      "mean_token_accuracy": 0.20927495658397674,
      "num_tokens": 129033760.0,
      "step": 63690
    },
    {
      "entropy": 5.918340396881104,
      "epoch": 5.566285065105305,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023787692372447306,
      "loss": 5.233,
      "mean_token_accuracy": 0.21543728411197663,
      "num_tokens": 129043993.0,
      "step": 63695
    },
    {
      "entropy": 5.813448333740235,
      "epoch": 5.566722013458009,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023784619004552738,
      "loss": 5.1203,
      "mean_token_accuracy": 0.22250189483165742,
      "num_tokens": 129053516.0,
      "step": 63700
    },
    {
      "entropy": 5.7649860858917235,
      "epoch": 5.567158961810714,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023781545707921115,
      "loss": 5.097,
      "mean_token_accuracy": 0.2181764855980873,
      "num_tokens": 129063137.0,
      "step": 63705
    },
    {
      "entropy": 5.782045078277588,
      "epoch": 5.567595910163419,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023778472482611395,
      "loss": 5.1161,
      "mean_token_accuracy": 0.220230832695961,
      "num_tokens": 129072503.0,
      "step": 63710
    },
    {
      "entropy": 5.799971055984497,
      "epoch": 5.568032858516124,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002377539932868253,
      "loss": 5.1619,
      "mean_token_accuracy": 0.21856678724288942,
      "num_tokens": 129081545.0,
      "step": 63715
    },
    {
      "entropy": 5.798859643936157,
      "epoch": 5.568469806868828,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023772326246193455,
      "loss": 5.1023,
      "mean_token_accuracy": 0.21763259023427964,
      "num_tokens": 129091838.0,
      "step": 63720
    },
    {
      "entropy": 5.856218671798706,
      "epoch": 5.568906755221533,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023769253235203115,
      "loss": 5.2457,
      "mean_token_accuracy": 0.21839949041604995,
      "num_tokens": 129102340.0,
      "step": 63725
    },
    {
      "entropy": 5.923123693466186,
      "epoch": 5.569343703574237,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023766180295770452,
      "loss": 5.2184,
      "mean_token_accuracy": 0.21291692852973937,
      "num_tokens": 129113476.0,
      "step": 63730
    },
    {
      "entropy": 5.9015766143798825,
      "epoch": 5.569780651926942,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023763107427954405,
      "loss": 5.231,
      "mean_token_accuracy": 0.2150472432374954,
      "num_tokens": 129123240.0,
      "step": 63735
    },
    {
      "entropy": 5.862364768981934,
      "epoch": 5.570217600279647,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002376003463181392,
      "loss": 5.1597,
      "mean_token_accuracy": 0.2230610281229019,
      "num_tokens": 129133241.0,
      "step": 63740
    },
    {
      "entropy": 5.824510097503662,
      "epoch": 5.570654548632351,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002375696190740793,
      "loss": 5.2098,
      "mean_token_accuracy": 0.21936869919300078,
      "num_tokens": 129144887.0,
      "step": 63745
    },
    {
      "entropy": 5.869297885894776,
      "epoch": 5.571091496985057,
      "grad_norm": 1.125,
      "learning_rate": 0.00023753889254795373,
      "loss": 5.2451,
      "mean_token_accuracy": 0.21654494404792785,
      "num_tokens": 129154802.0,
      "step": 63750
    },
    {
      "entropy": 5.866836833953857,
      "epoch": 5.571528445337761,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002375081667403518,
      "loss": 5.2185,
      "mean_token_accuracy": 0.21249566674232484,
      "num_tokens": 129164799.0,
      "step": 63755
    },
    {
      "entropy": 5.82460470199585,
      "epoch": 5.571965393690466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023747744165186286,
      "loss": 5.1806,
      "mean_token_accuracy": 0.22166521102190018,
      "num_tokens": 129174945.0,
      "step": 63760
    },
    {
      "entropy": 5.8028209686279295,
      "epoch": 5.572402342043171,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023744671728307627,
      "loss": 5.1159,
      "mean_token_accuracy": 0.22143416255712509,
      "num_tokens": 129184301.0,
      "step": 63765
    },
    {
      "entropy": 5.883843421936035,
      "epoch": 5.5728392903958754,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002374159936345814,
      "loss": 5.2171,
      "mean_token_accuracy": 0.21786772310733796,
      "num_tokens": 129194322.0,
      "step": 63770
    },
    {
      "entropy": 5.877766704559326,
      "epoch": 5.57327623874858,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002373852707069673,
      "loss": 5.1614,
      "mean_token_accuracy": 0.2206292688846588,
      "num_tokens": 129204829.0,
      "step": 63775
    },
    {
      "entropy": 5.795277547836304,
      "epoch": 5.573713187101284,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002373545485008235,
      "loss": 5.1025,
      "mean_token_accuracy": 0.23047892600297928,
      "num_tokens": 129214829.0,
      "step": 63780
    },
    {
      "entropy": 5.882835817337036,
      "epoch": 5.574150135453989,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023732382701673922,
      "loss": 5.262,
      "mean_token_accuracy": 0.21445826739072799,
      "num_tokens": 129225341.0,
      "step": 63785
    },
    {
      "entropy": 5.871273803710937,
      "epoch": 5.574587083806694,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023729310625530365,
      "loss": 5.1793,
      "mean_token_accuracy": 0.2218847781419754,
      "num_tokens": 129234905.0,
      "step": 63790
    },
    {
      "entropy": 5.872131967544556,
      "epoch": 5.575024032159399,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023726238621710606,
      "loss": 5.1598,
      "mean_token_accuracy": 0.22456476986408233,
      "num_tokens": 129244548.0,
      "step": 63795
    },
    {
      "entropy": 5.807434701919556,
      "epoch": 5.575460980512103,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023723166690273566,
      "loss": 5.1584,
      "mean_token_accuracy": 0.21925043761730195,
      "num_tokens": 129255202.0,
      "step": 63800
    },
    {
      "entropy": 5.857636165618897,
      "epoch": 5.5758979288648085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023720094831278178,
      "loss": 5.2316,
      "mean_token_accuracy": 0.21442125290632247,
      "num_tokens": 129265406.0,
      "step": 63805
    },
    {
      "entropy": 5.92410717010498,
      "epoch": 5.576334877217513,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023717023044783343,
      "loss": 5.2492,
      "mean_token_accuracy": 0.21572352647781373,
      "num_tokens": 129275956.0,
      "step": 63810
    },
    {
      "entropy": 5.919098901748657,
      "epoch": 5.5767718255702174,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023713951330847982,
      "loss": 5.2448,
      "mean_token_accuracy": 0.21053614169359208,
      "num_tokens": 129286049.0,
      "step": 63815
    },
    {
      "entropy": 5.84486141204834,
      "epoch": 5.577208773922923,
      "grad_norm": 1.125,
      "learning_rate": 0.00023710879689531027,
      "loss": 5.1455,
      "mean_token_accuracy": 0.22326355129480363,
      "num_tokens": 129294925.0,
      "step": 63820
    },
    {
      "entropy": 5.815016031265259,
      "epoch": 5.577645722275627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023707808120891383,
      "loss": 5.1662,
      "mean_token_accuracy": 0.21838375329971313,
      "num_tokens": 129304450.0,
      "step": 63825
    },
    {
      "entropy": 5.8815826892852785,
      "epoch": 5.578082670628332,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002370473662498797,
      "loss": 5.2641,
      "mean_token_accuracy": 0.21536433100700378,
      "num_tokens": 129314405.0,
      "step": 63830
    },
    {
      "entropy": 5.783610725402832,
      "epoch": 5.578519618981036,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023701665201879697,
      "loss": 5.1938,
      "mean_token_accuracy": 0.22398670762777328,
      "num_tokens": 129324809.0,
      "step": 63835
    },
    {
      "entropy": 5.862096071243286,
      "epoch": 5.578956567333741,
      "grad_norm": 1.125,
      "learning_rate": 0.00023698593851625472,
      "loss": 5.2265,
      "mean_token_accuracy": 0.21587184816598892,
      "num_tokens": 129334596.0,
      "step": 63840
    },
    {
      "entropy": 5.904763078689575,
      "epoch": 5.579393515686446,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00023695522574284207,
      "loss": 5.2235,
      "mean_token_accuracy": 0.21479777991771698,
      "num_tokens": 129345262.0,
      "step": 63845
    },
    {
      "entropy": 5.896980428695679,
      "epoch": 5.5798304640391505,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00023692451369914814,
      "loss": 5.1591,
      "mean_token_accuracy": 0.2230005905032158,
      "num_tokens": 129355061.0,
      "step": 63850
    },
    {
      "entropy": 5.841945838928223,
      "epoch": 5.580267412391855,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023689380238576204,
      "loss": 5.2028,
      "mean_token_accuracy": 0.2141129955649376,
      "num_tokens": 129365067.0,
      "step": 63855
    },
    {
      "entropy": 5.759049987792968,
      "epoch": 5.58070436074456,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00023686309180327275,
      "loss": 5.1474,
      "mean_token_accuracy": 0.22424988001585006,
      "num_tokens": 129375555.0,
      "step": 63860
    },
    {
      "entropy": 5.906939506530762,
      "epoch": 5.581141309097265,
      "grad_norm": 1.03125,
      "learning_rate": 0.00023683238195226932,
      "loss": 5.2268,
      "mean_token_accuracy": 0.21450549960136414,
      "num_tokens": 129385537.0,
      "step": 63865
    },
    {
      "entropy": 5.794642591476441,
      "epoch": 5.581578257449969,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002368016728333408,
      "loss": 5.0677,
      "mean_token_accuracy": 0.22789825797080993,
      "num_tokens": 129394734.0,
      "step": 63870
    },
    {
      "entropy": 5.769634008407593,
      "epoch": 5.582015205802674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00023677096444707618,
      "loss": 5.1156,
      "mean_token_accuracy": 0.2183430701494217,
      "num_tokens": 129405282.0,
      "step": 63875
    },
    {
      "entropy": 5.8334815979003904,
      "epoch": 5.582452154155379,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023674025679406462,
      "loss": 5.2308,
      "mean_token_accuracy": 0.2168410077691078,
      "num_tokens": 129416570.0,
      "step": 63880
    },
    {
      "entropy": 5.859946823120117,
      "epoch": 5.582889102508084,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002367095498748949,
      "loss": 5.2441,
      "mean_token_accuracy": 0.20942078083753585,
      "num_tokens": 129427760.0,
      "step": 63885
    },
    {
      "entropy": 5.856885766983032,
      "epoch": 5.583326050860788,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023667884369015608,
      "loss": 5.1863,
      "mean_token_accuracy": 0.21639930605888366,
      "num_tokens": 129437337.0,
      "step": 63890
    },
    {
      "entropy": 5.932519817352295,
      "epoch": 5.5837629992134925,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023664813824043708,
      "loss": 5.2371,
      "mean_token_accuracy": 0.21090643852949142,
      "num_tokens": 129447867.0,
      "step": 63895
    },
    {
      "entropy": 5.866823720932007,
      "epoch": 5.584199947566198,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023661743352632693,
      "loss": 5.1702,
      "mean_token_accuracy": 0.22449432015419007,
      "num_tokens": 129457403.0,
      "step": 63900
    },
    {
      "entropy": 5.830970621109008,
      "epoch": 5.584636895918902,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023658672954841455,
      "loss": 5.1893,
      "mean_token_accuracy": 0.2231113374233246,
      "num_tokens": 129467956.0,
      "step": 63905
    },
    {
      "entropy": 5.830210447311401,
      "epoch": 5.585073844271607,
      "grad_norm": 1.125,
      "learning_rate": 0.00023655602630728878,
      "loss": 5.172,
      "mean_token_accuracy": 0.21522524505853652,
      "num_tokens": 129478545.0,
      "step": 63910
    },
    {
      "entropy": 5.849656248092652,
      "epoch": 5.585510792624312,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023652532380353858,
      "loss": 5.2799,
      "mean_token_accuracy": 0.21446481943130494,
      "num_tokens": 129490164.0,
      "step": 63915
    },
    {
      "entropy": 5.810576868057251,
      "epoch": 5.585947740977017,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023649462203775278,
      "loss": 5.1698,
      "mean_token_accuracy": 0.22622318267822267,
      "num_tokens": 129500708.0,
      "step": 63920
    },
    {
      "entropy": 5.881253576278686,
      "epoch": 5.586384689329721,
      "grad_norm": 1.125,
      "learning_rate": 0.00023646392101052033,
      "loss": 5.208,
      "mean_token_accuracy": 0.22195768505334854,
      "num_tokens": 129511315.0,
      "step": 63925
    },
    {
      "entropy": 5.919298839569092,
      "epoch": 5.586821637682426,
      "grad_norm": 1.03125,
      "learning_rate": 0.00023643322072243017,
      "loss": 5.2802,
      "mean_token_accuracy": 0.2127888724207878,
      "num_tokens": 129523267.0,
      "step": 63930
    },
    {
      "entropy": 5.901036500930786,
      "epoch": 5.587258586035131,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023640252117407097,
      "loss": 5.1792,
      "mean_token_accuracy": 0.22347622662782668,
      "num_tokens": 129533356.0,
      "step": 63935
    },
    {
      "entropy": 5.812704038619995,
      "epoch": 5.587695534387835,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002363718223660316,
      "loss": 5.1785,
      "mean_token_accuracy": 0.22426970005035402,
      "num_tokens": 129543274.0,
      "step": 63940
    },
    {
      "entropy": 5.897261667251587,
      "epoch": 5.58813248274054,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023634112429890092,
      "loss": 5.2795,
      "mean_token_accuracy": 0.21339603811502456,
      "num_tokens": 129553949.0,
      "step": 63945
    },
    {
      "entropy": 5.918239116668701,
      "epoch": 5.588569431093244,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023631042697326782,
      "loss": 5.2591,
      "mean_token_accuracy": 0.21063552647829056,
      "num_tokens": 129563710.0,
      "step": 63950
    },
    {
      "entropy": 5.873857164382935,
      "epoch": 5.58900637944595,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002362797303897209,
      "loss": 5.2099,
      "mean_token_accuracy": 0.21990311592817308,
      "num_tokens": 129573384.0,
      "step": 63955
    },
    {
      "entropy": 5.829197788238526,
      "epoch": 5.589443327798654,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023624903454884911,
      "loss": 5.1759,
      "mean_token_accuracy": 0.2195124238729477,
      "num_tokens": 129582084.0,
      "step": 63960
    },
    {
      "entropy": 5.864110517501831,
      "epoch": 5.589880276151359,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002362183394512411,
      "loss": 5.2257,
      "mean_token_accuracy": 0.2215159922838211,
      "num_tokens": 129593124.0,
      "step": 63965
    },
    {
      "entropy": 5.9010358333587645,
      "epoch": 5.590317224504064,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002361876450974857,
      "loss": 5.2777,
      "mean_token_accuracy": 0.20915793627500534,
      "num_tokens": 129604097.0,
      "step": 63970
    },
    {
      "entropy": 5.9013360977172855,
      "epoch": 5.5907541728567685,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023615695148817157,
      "loss": 5.1723,
      "mean_token_accuracy": 0.22133202254772186,
      "num_tokens": 129613396.0,
      "step": 63975
    },
    {
      "entropy": 5.892117357254028,
      "epoch": 5.591191121209473,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002361262586238875,
      "loss": 5.3027,
      "mean_token_accuracy": 0.20809357315301896,
      "num_tokens": 129624373.0,
      "step": 63980
    },
    {
      "entropy": 5.813512945175171,
      "epoch": 5.591628069562177,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023609556650522212,
      "loss": 5.1037,
      "mean_token_accuracy": 0.2265833616256714,
      "num_tokens": 129634862.0,
      "step": 63985
    },
    {
      "entropy": 5.784246015548706,
      "epoch": 5.592065017914883,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002360648751327642,
      "loss": 5.1446,
      "mean_token_accuracy": 0.22038004845380782,
      "num_tokens": 129645055.0,
      "step": 63990
    },
    {
      "entropy": 5.859124183654785,
      "epoch": 5.592501966267587,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023603418450710234,
      "loss": 5.2214,
      "mean_token_accuracy": 0.21204987466335296,
      "num_tokens": 129655004.0,
      "step": 63995
    },
    {
      "entropy": 5.856294012069702,
      "epoch": 5.592938914620292,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023600349462882527,
      "loss": 5.1662,
      "mean_token_accuracy": 0.22164520919322966,
      "num_tokens": 129666249.0,
      "step": 64000
    },
    {
      "entropy": 5.941412448883057,
      "epoch": 5.593375862972996,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023597280549852157,
      "loss": 5.3105,
      "mean_token_accuracy": 0.19944567531347274,
      "num_tokens": 129677635.0,
      "step": 64005
    },
    {
      "entropy": 5.8265846252441404,
      "epoch": 5.5938128113257015,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00023594211711677992,
      "loss": 5.1601,
      "mean_token_accuracy": 0.22144413292407988,
      "num_tokens": 129687701.0,
      "step": 64010
    },
    {
      "entropy": 5.889531660079956,
      "epoch": 5.594249759678406,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023591142948418893,
      "loss": 5.2718,
      "mean_token_accuracy": 0.20955138504505158,
      "num_tokens": 129698256.0,
      "step": 64015
    },
    {
      "entropy": 5.917569208145141,
      "epoch": 5.5946867080311105,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023588074260133725,
      "loss": 5.2871,
      "mean_token_accuracy": 0.20992430597543715,
      "num_tokens": 129707520.0,
      "step": 64020
    },
    {
      "entropy": 5.839009094238281,
      "epoch": 5.595123656383816,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023585005646881342,
      "loss": 5.2143,
      "mean_token_accuracy": 0.21925257444381713,
      "num_tokens": 129716805.0,
      "step": 64025
    },
    {
      "entropy": 5.864015817642212,
      "epoch": 5.59556060473652,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023581937108720598,
      "loss": 5.2417,
      "mean_token_accuracy": 0.21288644522428513,
      "num_tokens": 129726273.0,
      "step": 64030
    },
    {
      "entropy": 5.781790637969971,
      "epoch": 5.595997553089225,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023578868645710355,
      "loss": 5.0553,
      "mean_token_accuracy": 0.22847054600715638,
      "num_tokens": 129735540.0,
      "step": 64035
    },
    {
      "entropy": 5.795917797088623,
      "epoch": 5.596434501441929,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023575800257909463,
      "loss": 5.1089,
      "mean_token_accuracy": 0.22774634808301925,
      "num_tokens": 129745220.0,
      "step": 64040
    },
    {
      "entropy": 5.795451498031616,
      "epoch": 5.596871449794635,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002357273194537678,
      "loss": 5.137,
      "mean_token_accuracy": 0.22339977473020553,
      "num_tokens": 129754458.0,
      "step": 64045
    },
    {
      "entropy": 5.78691873550415,
      "epoch": 5.597308398147339,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023569663708171158,
      "loss": 5.1443,
      "mean_token_accuracy": 0.2173999160528183,
      "num_tokens": 129763208.0,
      "step": 64050
    },
    {
      "entropy": 5.8222438335418705,
      "epoch": 5.5977453465000435,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023566595546351443,
      "loss": 5.1638,
      "mean_token_accuracy": 0.22322543263435363,
      "num_tokens": 129772578.0,
      "step": 64055
    },
    {
      "entropy": 5.821260070800781,
      "epoch": 5.598182294852748,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023563527459976493,
      "loss": 5.1097,
      "mean_token_accuracy": 0.22688814699649812,
      "num_tokens": 129783504.0,
      "step": 64060
    },
    {
      "entropy": 5.7912298202514645,
      "epoch": 5.598619243205453,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023560459449105137,
      "loss": 5.0753,
      "mean_token_accuracy": 0.22744579166173934,
      "num_tokens": 129792906.0,
      "step": 64065
    },
    {
      "entropy": 5.896031379699707,
      "epoch": 5.599056191558158,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002355739151379624,
      "loss": 5.258,
      "mean_token_accuracy": 0.20848041474819184,
      "num_tokens": 129802763.0,
      "step": 64070
    },
    {
      "entropy": 5.966924142837525,
      "epoch": 5.599493139910862,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023554323654108645,
      "loss": 5.3926,
      "mean_token_accuracy": 0.1986069455742836,
      "num_tokens": 129813954.0,
      "step": 64075
    },
    {
      "entropy": 5.908002614974976,
      "epoch": 5.599930088263568,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023551255870101178,
      "loss": 5.1977,
      "mean_token_accuracy": 0.21435171365737915,
      "num_tokens": 129824167.0,
      "step": 64080
    },
    {
      "entropy": 5.786529684066773,
      "epoch": 5.600367036616272,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000235481881618327,
      "loss": 5.135,
      "mean_token_accuracy": 0.22074001282453537,
      "num_tokens": 129834280.0,
      "step": 64085
    },
    {
      "entropy": 5.797125625610351,
      "epoch": 5.600803984968977,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023545120529362047,
      "loss": 5.1346,
      "mean_token_accuracy": 0.23032351583242416,
      "num_tokens": 129843801.0,
      "step": 64090
    },
    {
      "entropy": 5.836624431610107,
      "epoch": 5.601240933321681,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023542052972748046,
      "loss": 5.2329,
      "mean_token_accuracy": 0.21114908009767533,
      "num_tokens": 129854091.0,
      "step": 64095
    },
    {
      "entropy": 5.872021865844727,
      "epoch": 5.601677881674386,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023538985492049553,
      "loss": 5.241,
      "mean_token_accuracy": 0.22048087120056153,
      "num_tokens": 129864849.0,
      "step": 64100
    },
    {
      "entropy": 5.990739202499389,
      "epoch": 5.602114830027091,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00023535918087325385,
      "loss": 5.3108,
      "mean_token_accuracy": 0.20660182386636733,
      "num_tokens": 129877560.0,
      "step": 64105
    },
    {
      "entropy": 5.920786952972412,
      "epoch": 5.602551778379795,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002353285075863439,
      "loss": 5.2755,
      "mean_token_accuracy": 0.2129400208592415,
      "num_tokens": 129888565.0,
      "step": 64110
    },
    {
      "entropy": 5.871281957626342,
      "epoch": 5.6029887267325,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002352978350603539,
      "loss": 5.3171,
      "mean_token_accuracy": 0.21319198459386826,
      "num_tokens": 129898662.0,
      "step": 64115
    },
    {
      "entropy": 5.800588178634643,
      "epoch": 5.603425675085205,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023526716329587234,
      "loss": 5.1057,
      "mean_token_accuracy": 0.23019563257694245,
      "num_tokens": 129909197.0,
      "step": 64120
    },
    {
      "entropy": 5.925140905380249,
      "epoch": 5.60386262343791,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002352364922934873,
      "loss": 5.306,
      "mean_token_accuracy": 0.21078669279813766,
      "num_tokens": 129919113.0,
      "step": 64125
    },
    {
      "entropy": 5.847978925704956,
      "epoch": 5.604299571790614,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002352058220537871,
      "loss": 5.1104,
      "mean_token_accuracy": 0.22676288038492204,
      "num_tokens": 129928136.0,
      "step": 64130
    },
    {
      "entropy": 5.824286699295044,
      "epoch": 5.6047365201433195,
      "grad_norm": 1.03125,
      "learning_rate": 0.00023517515257736026,
      "loss": 5.1642,
      "mean_token_accuracy": 0.21564094126224517,
      "num_tokens": 129937567.0,
      "step": 64135
    },
    {
      "entropy": 5.826114892959595,
      "epoch": 5.605173468496024,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023514448386479476,
      "loss": 5.1858,
      "mean_token_accuracy": 0.21792010813951493,
      "num_tokens": 129947546.0,
      "step": 64140
    },
    {
      "entropy": 5.845522546768189,
      "epoch": 5.605610416848728,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023511381591667896,
      "loss": 5.1796,
      "mean_token_accuracy": 0.21625043153762818,
      "num_tokens": 129957983.0,
      "step": 64145
    },
    {
      "entropy": 5.8665502071380615,
      "epoch": 5.606047365201433,
      "grad_norm": 1.125,
      "learning_rate": 0.0002350831487336011,
      "loss": 5.2218,
      "mean_token_accuracy": 0.2106500446796417,
      "num_tokens": 129967136.0,
      "step": 64150
    },
    {
      "entropy": 5.852910423278809,
      "epoch": 5.606484313554138,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023505248231614934,
      "loss": 5.2676,
      "mean_token_accuracy": 0.21893969923257828,
      "num_tokens": 129977536.0,
      "step": 64155
    },
    {
      "entropy": 5.873051977157592,
      "epoch": 5.606921261906843,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023502181666491202,
      "loss": 5.2008,
      "mean_token_accuracy": 0.2144257679581642,
      "num_tokens": 129987737.0,
      "step": 64160
    },
    {
      "entropy": 5.893047571182251,
      "epoch": 5.607358210259547,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023499115178047704,
      "loss": 5.2362,
      "mean_token_accuracy": 0.2189510703086853,
      "num_tokens": 129997698.0,
      "step": 64165
    },
    {
      "entropy": 5.867412185668945,
      "epoch": 5.607795158612252,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023496048766343282,
      "loss": 5.2138,
      "mean_token_accuracy": 0.2162206679582596,
      "num_tokens": 130008727.0,
      "step": 64170
    },
    {
      "entropy": 5.847812080383301,
      "epoch": 5.608232106964957,
      "grad_norm": 1.25,
      "learning_rate": 0.00023492982431436737,
      "loss": 5.1433,
      "mean_token_accuracy": 0.22271924912929536,
      "num_tokens": 130018410.0,
      "step": 64175
    },
    {
      "entropy": 5.896568298339844,
      "epoch": 5.6086690553176615,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023489916173386895,
      "loss": 5.2297,
      "mean_token_accuracy": 0.2115826889872551,
      "num_tokens": 130027436.0,
      "step": 64180
    },
    {
      "entropy": 5.915721845626831,
      "epoch": 5.609106003670366,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023486849992252558,
      "loss": 5.2603,
      "mean_token_accuracy": 0.21309852451086045,
      "num_tokens": 130038366.0,
      "step": 64185
    },
    {
      "entropy": 5.817821073532104,
      "epoch": 5.609542952023071,
      "grad_norm": 1.25,
      "learning_rate": 0.00023483783888092542,
      "loss": 5.193,
      "mean_token_accuracy": 0.22268728911876678,
      "num_tokens": 130047819.0,
      "step": 64190
    },
    {
      "entropy": 5.840653944015503,
      "epoch": 5.609979900375776,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023480717860965657,
      "loss": 5.2369,
      "mean_token_accuracy": 0.2167237415909767,
      "num_tokens": 130058345.0,
      "step": 64195
    },
    {
      "entropy": 5.831245565414429,
      "epoch": 5.61041684872848,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023477651910930702,
      "loss": 5.0789,
      "mean_token_accuracy": 0.22561460584402085,
      "num_tokens": 130068194.0,
      "step": 64200
    },
    {
      "entropy": 5.874300289154053,
      "epoch": 5.610853797081185,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023474586038046497,
      "loss": 5.2036,
      "mean_token_accuracy": 0.22183808833360671,
      "num_tokens": 130078082.0,
      "step": 64205
    },
    {
      "entropy": 5.832611799240112,
      "epoch": 5.61129074543389,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023471520242371847,
      "loss": 5.2341,
      "mean_token_accuracy": 0.213267919421196,
      "num_tokens": 130087912.0,
      "step": 64210
    },
    {
      "entropy": 5.827436780929565,
      "epoch": 5.6117276937865945,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023468454523965543,
      "loss": 5.1784,
      "mean_token_accuracy": 0.22629671692848205,
      "num_tokens": 130098766.0,
      "step": 64215
    },
    {
      "entropy": 5.908880949020386,
      "epoch": 5.612164642139299,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023465388882886395,
      "loss": 5.1999,
      "mean_token_accuracy": 0.21626883298158645,
      "num_tokens": 130109375.0,
      "step": 64220
    },
    {
      "entropy": 5.789808368682861,
      "epoch": 5.6126015904920035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000234623233191932,
      "loss": 5.1485,
      "mean_token_accuracy": 0.21868052035570146,
      "num_tokens": 130119666.0,
      "step": 64225
    },
    {
      "entropy": 5.817589282989502,
      "epoch": 5.613038538844709,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023459257832944758,
      "loss": 5.1255,
      "mean_token_accuracy": 0.22245331704616547,
      "num_tokens": 130130408.0,
      "step": 64230
    },
    {
      "entropy": 5.89793643951416,
      "epoch": 5.613475487197413,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023456192424199874,
      "loss": 5.2297,
      "mean_token_accuracy": 0.2157675087451935,
      "num_tokens": 130140910.0,
      "step": 64235
    },
    {
      "entropy": 5.858690166473389,
      "epoch": 5.613912435550118,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002345312709301734,
      "loss": 5.2325,
      "mean_token_accuracy": 0.21120957732200624,
      "num_tokens": 130151367.0,
      "step": 64240
    },
    {
      "entropy": 5.822938728332519,
      "epoch": 5.614349383902823,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002345006183945594,
      "loss": 5.163,
      "mean_token_accuracy": 0.220268552005291,
      "num_tokens": 130161630.0,
      "step": 64245
    },
    {
      "entropy": 5.818091583251953,
      "epoch": 5.614786332255528,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00023446996663574476,
      "loss": 5.1983,
      "mean_token_accuracy": 0.21236056387424468,
      "num_tokens": 130171117.0,
      "step": 64250
    },
    {
      "entropy": 5.889695739746093,
      "epoch": 5.615223280608232,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023443931565431754,
      "loss": 5.2349,
      "mean_token_accuracy": 0.2158350318670273,
      "num_tokens": 130180007.0,
      "step": 64255
    },
    {
      "entropy": 5.8815264225006105,
      "epoch": 5.6156602289609365,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023440866545086536,
      "loss": 5.3093,
      "mean_token_accuracy": 0.20532086491584778,
      "num_tokens": 130190590.0,
      "step": 64260
    },
    {
      "entropy": 5.914891386032105,
      "epoch": 5.616097177313642,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023437801602597624,
      "loss": 5.2592,
      "mean_token_accuracy": 0.21040077060461043,
      "num_tokens": 130200319.0,
      "step": 64265
    },
    {
      "entropy": 5.848422718048096,
      "epoch": 5.616534125666346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023434736738023814,
      "loss": 5.1776,
      "mean_token_accuracy": 0.22133075445890427,
      "num_tokens": 130210430.0,
      "step": 64270
    },
    {
      "entropy": 5.8921041011810305,
      "epoch": 5.616971074019051,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023431671951423878,
      "loss": 5.2157,
      "mean_token_accuracy": 0.21264859586954116,
      "num_tokens": 130220197.0,
      "step": 64275
    },
    {
      "entropy": 5.849622917175293,
      "epoch": 5.617408022371755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023428607242856616,
      "loss": 5.2239,
      "mean_token_accuracy": 0.22084927707910537,
      "num_tokens": 130231312.0,
      "step": 64280
    },
    {
      "entropy": 5.8811835765838625,
      "epoch": 5.617844970724461,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002342554261238079,
      "loss": 5.1879,
      "mean_token_accuracy": 0.22542700469493865,
      "num_tokens": 130240613.0,
      "step": 64285
    },
    {
      "entropy": 5.9482221603393555,
      "epoch": 5.618281919077165,
      "grad_norm": 1.046875,
      "learning_rate": 0.00023422478060055192,
      "loss": 5.2944,
      "mean_token_accuracy": 0.20223444998264312,
      "num_tokens": 130250305.0,
      "step": 64290
    },
    {
      "entropy": 5.87253589630127,
      "epoch": 5.61871886742987,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00023419413585938604,
      "loss": 5.2017,
      "mean_token_accuracy": 0.2148710682988167,
      "num_tokens": 130260413.0,
      "step": 64295
    },
    {
      "entropy": 5.884545612335205,
      "epoch": 5.619155815782575,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023416349190089799,
      "loss": 5.2198,
      "mean_token_accuracy": 0.21633283644914628,
      "num_tokens": 130271430.0,
      "step": 64300
    },
    {
      "entropy": 5.866492700576782,
      "epoch": 5.619592764135279,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00023413284872567558,
      "loss": 5.2106,
      "mean_token_accuracy": 0.2170715183019638,
      "num_tokens": 130282139.0,
      "step": 64305
    },
    {
      "entropy": 5.861152648925781,
      "epoch": 5.620029712487984,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00023410220633430658,
      "loss": 5.1242,
      "mean_token_accuracy": 0.2234819784760475,
      "num_tokens": 130292113.0,
      "step": 64310
    },
    {
      "entropy": 5.838694095611572,
      "epoch": 5.620466660840688,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002340715647273786,
      "loss": 5.169,
      "mean_token_accuracy": 0.21756714582443237,
      "num_tokens": 130302117.0,
      "step": 64315
    },
    {
      "entropy": 5.797450590133667,
      "epoch": 5.620903609193394,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023404092390547955,
      "loss": 5.1178,
      "mean_token_accuracy": 0.22449225485324859,
      "num_tokens": 130313373.0,
      "step": 64320
    },
    {
      "entropy": 5.856086874008179,
      "epoch": 5.621340557546098,
      "grad_norm": 1.15625,
      "learning_rate": 0.000234010283869197,
      "loss": 5.1641,
      "mean_token_accuracy": 0.21921956688165664,
      "num_tokens": 130323166.0,
      "step": 64325
    },
    {
      "entropy": 5.918017339706421,
      "epoch": 5.621777505898803,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023397964461911868,
      "loss": 5.1757,
      "mean_token_accuracy": 0.2163962796330452,
      "num_tokens": 130333405.0,
      "step": 64330
    },
    {
      "entropy": 5.78508882522583,
      "epoch": 5.622214454251507,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002339490061558323,
      "loss": 5.204,
      "mean_token_accuracy": 0.22078492641448974,
      "num_tokens": 130343113.0,
      "step": 64335
    },
    {
      "entropy": 5.873683929443359,
      "epoch": 5.6226514026042125,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023391836847992548,
      "loss": 5.2057,
      "mean_token_accuracy": 0.21683093160390854,
      "num_tokens": 130352749.0,
      "step": 64340
    },
    {
      "entropy": 5.916307258605957,
      "epoch": 5.623088350956917,
      "grad_norm": 1.125,
      "learning_rate": 0.00023388773159198584,
      "loss": 5.1841,
      "mean_token_accuracy": 0.22169528305530548,
      "num_tokens": 130363556.0,
      "step": 64345
    },
    {
      "entropy": 5.897096824645996,
      "epoch": 5.623525299309621,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002338570954926011,
      "loss": 5.295,
      "mean_token_accuracy": 0.20620214343070983,
      "num_tokens": 130372855.0,
      "step": 64350
    },
    {
      "entropy": 5.848867416381836,
      "epoch": 5.623962247662327,
      "grad_norm": 1.125,
      "learning_rate": 0.00023382646018235882,
      "loss": 5.1289,
      "mean_token_accuracy": 0.22391206324100493,
      "num_tokens": 130382756.0,
      "step": 64355
    },
    {
      "entropy": 5.801083278656006,
      "epoch": 5.624399196015031,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023379582566184659,
      "loss": 5.1143,
      "mean_token_accuracy": 0.22700486034154893,
      "num_tokens": 130393464.0,
      "step": 64360
    },
    {
      "entropy": 5.769684886932373,
      "epoch": 5.624836144367736,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023376519193165203,
      "loss": 5.1408,
      "mean_token_accuracy": 0.22471458464860916,
      "num_tokens": 130404400.0,
      "step": 64365
    },
    {
      "entropy": 5.84031662940979,
      "epoch": 5.62527309272044,
      "grad_norm": 1.125,
      "learning_rate": 0.0002337345589923627,
      "loss": 5.1894,
      "mean_token_accuracy": 0.21675549000501632,
      "num_tokens": 130414357.0,
      "step": 64370
    },
    {
      "entropy": 5.863948726654053,
      "epoch": 5.6257100410731455,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023370392684456608,
      "loss": 5.1121,
      "mean_token_accuracy": 0.22605366557836531,
      "num_tokens": 130423830.0,
      "step": 64375
    },
    {
      "entropy": 5.878417682647705,
      "epoch": 5.62614698942585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002336732954888499,
      "loss": 5.1744,
      "mean_token_accuracy": 0.2214088335633278,
      "num_tokens": 130433777.0,
      "step": 64380
    },
    {
      "entropy": 5.948243713378906,
      "epoch": 5.6265839377785545,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002336426649258015,
      "loss": 5.2929,
      "mean_token_accuracy": 0.2076518639922142,
      "num_tokens": 130444076.0,
      "step": 64385
    },
    {
      "entropy": 5.836037826538086,
      "epoch": 5.627020886131259,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002336120351560085,
      "loss": 5.1232,
      "mean_token_accuracy": 0.21992242634296416,
      "num_tokens": 130454397.0,
      "step": 64390
    },
    {
      "entropy": 5.814848756790161,
      "epoch": 5.627457834483964,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002335814061800583,
      "loss": 5.1157,
      "mean_token_accuracy": 0.22524047046899795,
      "num_tokens": 130464721.0,
      "step": 64395
    },
    {
      "entropy": 5.815730857849121,
      "epoch": 5.627894782836669,
      "grad_norm": 1.03125,
      "learning_rate": 0.00023355077799853846,
      "loss": 5.1788,
      "mean_token_accuracy": 0.2222041219472885,
      "num_tokens": 130476168.0,
      "step": 64400
    },
    {
      "entropy": 5.870305442810059,
      "epoch": 5.628331731189373,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002335201506120364,
      "loss": 5.1324,
      "mean_token_accuracy": 0.22146920412778853,
      "num_tokens": 130485059.0,
      "step": 64405
    },
    {
      "entropy": 5.806872940063476,
      "epoch": 5.628768679542079,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002334895240211396,
      "loss": 5.1133,
      "mean_token_accuracy": 0.22634293884038925,
      "num_tokens": 130496063.0,
      "step": 64410
    },
    {
      "entropy": 5.804260635375977,
      "epoch": 5.629205627894783,
      "grad_norm": 1.015625,
      "learning_rate": 0.00023345889822643545,
      "loss": 5.1865,
      "mean_token_accuracy": 0.2209795117378235,
      "num_tokens": 130506567.0,
      "step": 64415
    },
    {
      "entropy": 5.876440858840942,
      "epoch": 5.6296425762474875,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002334282732285114,
      "loss": 5.1961,
      "mean_token_accuracy": 0.2180885523557663,
      "num_tokens": 130517401.0,
      "step": 64420
    },
    {
      "entropy": 5.765341472625733,
      "epoch": 5.630079524600192,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002333976490279549,
      "loss": 5.0224,
      "mean_token_accuracy": 0.23933792859315872,
      "num_tokens": 130526758.0,
      "step": 64425
    },
    {
      "entropy": 5.815204858779907,
      "epoch": 5.6305164729528965,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023336702562535327,
      "loss": 5.1432,
      "mean_token_accuracy": 0.22079358845949174,
      "num_tokens": 130535955.0,
      "step": 64430
    },
    {
      "entropy": 5.8568041801452635,
      "epoch": 5.630953421305602,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00023333640302129394,
      "loss": 5.2652,
      "mean_token_accuracy": 0.21333255022764205,
      "num_tokens": 130546140.0,
      "step": 64435
    },
    {
      "entropy": 5.78113899230957,
      "epoch": 5.631390369658306,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023330578121636414,
      "loss": 5.1202,
      "mean_token_accuracy": 0.230213525891304,
      "num_tokens": 130556211.0,
      "step": 64440
    },
    {
      "entropy": 5.897710990905762,
      "epoch": 5.631827318011011,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023327516021115136,
      "loss": 5.2429,
      "mean_token_accuracy": 0.22148039191961288,
      "num_tokens": 130567113.0,
      "step": 64445
    },
    {
      "entropy": 5.8548869609832765,
      "epoch": 5.632264266363716,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002332445400062429,
      "loss": 5.1856,
      "mean_token_accuracy": 0.2275754228234291,
      "num_tokens": 130578087.0,
      "step": 64450
    },
    {
      "entropy": 5.9363480567932125,
      "epoch": 5.632701214716421,
      "grad_norm": 1.15625,
      "learning_rate": 0.000233213920602226,
      "loss": 5.2826,
      "mean_token_accuracy": 0.2115580067038536,
      "num_tokens": 130587930.0,
      "step": 64455
    },
    {
      "entropy": 5.921266984939575,
      "epoch": 5.633138163069125,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000233183301999688,
      "loss": 5.2581,
      "mean_token_accuracy": 0.21860289871692656,
      "num_tokens": 130598085.0,
      "step": 64460
    },
    {
      "entropy": 5.790771484375,
      "epoch": 5.63357511142183,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002331526841992162,
      "loss": 5.1025,
      "mean_token_accuracy": 0.2300743877887726,
      "num_tokens": 130607422.0,
      "step": 64465
    },
    {
      "entropy": 5.867926454544067,
      "epoch": 5.634012059774535,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023312206720139784,
      "loss": 5.2299,
      "mean_token_accuracy": 0.22072605639696122,
      "num_tokens": 130617917.0,
      "step": 64470
    },
    {
      "entropy": 5.8709437370300295,
      "epoch": 5.634449008127239,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023309145100682016,
      "loss": 5.1875,
      "mean_token_accuracy": 0.2255846917629242,
      "num_tokens": 130628338.0,
      "step": 64475
    },
    {
      "entropy": 5.889898157119751,
      "epoch": 5.634885956479944,
      "grad_norm": 1.125,
      "learning_rate": 0.00023306083561607045,
      "loss": 5.243,
      "mean_token_accuracy": 0.21713427156209947,
      "num_tokens": 130638033.0,
      "step": 64480
    },
    {
      "entropy": 5.841726541519165,
      "epoch": 5.635322904832648,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002330302210297357,
      "loss": 5.2454,
      "mean_token_accuracy": 0.20703228414058686,
      "num_tokens": 130647817.0,
      "step": 64485
    },
    {
      "entropy": 5.837736415863037,
      "epoch": 5.635759853185354,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023299960724840354,
      "loss": 5.1418,
      "mean_token_accuracy": 0.2276274472475052,
      "num_tokens": 130657933.0,
      "step": 64490
    },
    {
      "entropy": 5.8489484786987305,
      "epoch": 5.636196801538058,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023296899427266082,
      "loss": 5.0781,
      "mean_token_accuracy": 0.22336495369672776,
      "num_tokens": 130668742.0,
      "step": 64495
    },
    {
      "entropy": 5.842227792739868,
      "epoch": 5.636633749890763,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023293838210309482,
      "loss": 5.2135,
      "mean_token_accuracy": 0.2162829101085663,
      "num_tokens": 130678383.0,
      "step": 64500
    },
    {
      "entropy": 5.841608190536499,
      "epoch": 5.637070698243468,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023290777074029269,
      "loss": 5.1217,
      "mean_token_accuracy": 0.22070808857679367,
      "num_tokens": 130687283.0,
      "step": 64505
    },
    {
      "entropy": 5.786999893188477,
      "epoch": 5.637507646596172,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023287716018484156,
      "loss": 5.2222,
      "mean_token_accuracy": 0.22435532212257386,
      "num_tokens": 130698357.0,
      "step": 64510
    },
    {
      "entropy": 5.821756172180176,
      "epoch": 5.637944594948877,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023284655043732865,
      "loss": 5.1631,
      "mean_token_accuracy": 0.22097801864147187,
      "num_tokens": 130708040.0,
      "step": 64515
    },
    {
      "entropy": 5.930950546264649,
      "epoch": 5.638381543301581,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002328159414983409,
      "loss": 5.2531,
      "mean_token_accuracy": 0.21218764483928682,
      "num_tokens": 130717770.0,
      "step": 64520
    },
    {
      "entropy": 5.8899595737457275,
      "epoch": 5.638818491654287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002327853333684656,
      "loss": 5.2023,
      "mean_token_accuracy": 0.21068876683712007,
      "num_tokens": 130728108.0,
      "step": 64525
    },
    {
      "entropy": 5.883148193359375,
      "epoch": 5.639255440006991,
      "grad_norm": 1.125,
      "learning_rate": 0.00023275472604828963,
      "loss": 5.2379,
      "mean_token_accuracy": 0.21114400029182434,
      "num_tokens": 130738828.0,
      "step": 64530
    },
    {
      "entropy": 5.890517377853394,
      "epoch": 5.639692388359696,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023272411953840022,
      "loss": 5.3044,
      "mean_token_accuracy": 0.2049052596092224,
      "num_tokens": 130749458.0,
      "step": 64535
    },
    {
      "entropy": 5.866079616546631,
      "epoch": 5.6401293367124,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023269351383938425,
      "loss": 5.1957,
      "mean_token_accuracy": 0.22230200320482255,
      "num_tokens": 130759125.0,
      "step": 64540
    },
    {
      "entropy": 5.901029396057129,
      "epoch": 5.6405662850651055,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023266290895182897,
      "loss": 5.2397,
      "mean_token_accuracy": 0.2157977283000946,
      "num_tokens": 130769148.0,
      "step": 64545
    },
    {
      "entropy": 5.886897039413452,
      "epoch": 5.64100323341781,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023263230487632113,
      "loss": 5.2642,
      "mean_token_accuracy": 0.21013475954532623,
      "num_tokens": 130780561.0,
      "step": 64550
    },
    {
      "entropy": 5.7652887344360355,
      "epoch": 5.641440181770514,
      "grad_norm": 1.046875,
      "learning_rate": 0.000232601701613448,
      "loss": 5.13,
      "mean_token_accuracy": 0.2343839719891548,
      "num_tokens": 130791819.0,
      "step": 64555
    },
    {
      "entropy": 5.893352127075195,
      "epoch": 5.64187713012322,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002325710991637965,
      "loss": 5.1198,
      "mean_token_accuracy": 0.23059655874967575,
      "num_tokens": 130801365.0,
      "step": 64560
    },
    {
      "entropy": 5.809466075897217,
      "epoch": 5.642314078475924,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023254049752795343,
      "loss": 5.1245,
      "mean_token_accuracy": 0.22137253731489182,
      "num_tokens": 130812142.0,
      "step": 64565
    },
    {
      "entropy": 5.868739128112793,
      "epoch": 5.642751026828629,
      "grad_norm": 1.125,
      "learning_rate": 0.00023250989670650592,
      "loss": 5.2154,
      "mean_token_accuracy": 0.21998028308153153,
      "num_tokens": 130822264.0,
      "step": 64570
    },
    {
      "entropy": 5.908426189422608,
      "epoch": 5.643187975181333,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023247929670004082,
      "loss": 5.238,
      "mean_token_accuracy": 0.21551730185747148,
      "num_tokens": 130833137.0,
      "step": 64575
    },
    {
      "entropy": 5.84945821762085,
      "epoch": 5.6436249235340386,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002324486975091451,
      "loss": 5.194,
      "mean_token_accuracy": 0.21664062291383743,
      "num_tokens": 130843114.0,
      "step": 64580
    },
    {
      "entropy": 5.841030263900757,
      "epoch": 5.644061871886743,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023241809913440576,
      "loss": 5.2,
      "mean_token_accuracy": 0.21760984063148497,
      "num_tokens": 130854515.0,
      "step": 64585
    },
    {
      "entropy": 5.918975067138672,
      "epoch": 5.6444988202394475,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023238750157640947,
      "loss": 5.278,
      "mean_token_accuracy": 0.2133583053946495,
      "num_tokens": 130864143.0,
      "step": 64590
    },
    {
      "entropy": 5.861532735824585,
      "epoch": 5.644935768592152,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00023235690483574322,
      "loss": 5.2364,
      "mean_token_accuracy": 0.22455519884824754,
      "num_tokens": 130875433.0,
      "step": 64595
    },
    {
      "entropy": 5.8458646774292,
      "epoch": 5.645372716944857,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023232630891299394,
      "loss": 5.1623,
      "mean_token_accuracy": 0.22143013030290604,
      "num_tokens": 130885838.0,
      "step": 64600
    },
    {
      "entropy": 5.88093638420105,
      "epoch": 5.645809665297562,
      "grad_norm": 1.25,
      "learning_rate": 0.00023229571380874842,
      "loss": 5.2524,
      "mean_token_accuracy": 0.21391735821962357,
      "num_tokens": 130897120.0,
      "step": 64605
    },
    {
      "entropy": 5.857578086853027,
      "epoch": 5.646246613650266,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023226511952359347,
      "loss": 5.1715,
      "mean_token_accuracy": 0.21604464650154115,
      "num_tokens": 130907174.0,
      "step": 64610
    },
    {
      "entropy": 5.851419162750244,
      "epoch": 5.646683562002972,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002322345260581159,
      "loss": 5.1151,
      "mean_token_accuracy": 0.226036374270916,
      "num_tokens": 130917325.0,
      "step": 64615
    },
    {
      "entropy": 5.889623355865479,
      "epoch": 5.647120510355676,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023220393341290258,
      "loss": 5.178,
      "mean_token_accuracy": 0.21838627457618714,
      "num_tokens": 130926629.0,
      "step": 64620
    },
    {
      "entropy": 5.800858068466186,
      "epoch": 5.6475574587083806,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002321733415885402,
      "loss": 5.1419,
      "mean_token_accuracy": 0.22409908026456832,
      "num_tokens": 130936007.0,
      "step": 64625
    },
    {
      "entropy": 5.826116895675659,
      "epoch": 5.647994407061085,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002321427505856155,
      "loss": 5.2224,
      "mean_token_accuracy": 0.21952870935201646,
      "num_tokens": 130946485.0,
      "step": 64630
    },
    {
      "entropy": 5.867189741134643,
      "epoch": 5.64843135541379,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023211216040471545,
      "loss": 5.2845,
      "mean_token_accuracy": 0.2059147521853447,
      "num_tokens": 130957324.0,
      "step": 64635
    },
    {
      "entropy": 5.922948932647705,
      "epoch": 5.648868303766495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023208157104642657,
      "loss": 5.1747,
      "mean_token_accuracy": 0.21827703416347505,
      "num_tokens": 130967061.0,
      "step": 64640
    },
    {
      "entropy": 5.887616729736328,
      "epoch": 5.649305252119199,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002320509825113356,
      "loss": 5.1801,
      "mean_token_accuracy": 0.21684327572584153,
      "num_tokens": 130977318.0,
      "step": 64645
    },
    {
      "entropy": 5.819640970230102,
      "epoch": 5.649742200471904,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00023202039480002936,
      "loss": 5.1447,
      "mean_token_accuracy": 0.22517427504062654,
      "num_tokens": 130987478.0,
      "step": 64650
    },
    {
      "entropy": 5.848132085800171,
      "epoch": 5.650179148824609,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023198980791309443,
      "loss": 5.2102,
      "mean_token_accuracy": 0.21412178128957748,
      "num_tokens": 130997276.0,
      "step": 64655
    },
    {
      "entropy": 5.811236619949341,
      "epoch": 5.650616097177314,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023195922185111756,
      "loss": 5.1591,
      "mean_token_accuracy": 0.21898550540208817,
      "num_tokens": 131007156.0,
      "step": 64660
    },
    {
      "entropy": 5.869980669021606,
      "epoch": 5.651053045530018,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023192863661468532,
      "loss": 5.1969,
      "mean_token_accuracy": 0.21773593723773957,
      "num_tokens": 131016485.0,
      "step": 64665
    },
    {
      "entropy": 5.836806726455689,
      "epoch": 5.651489993882723,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023189805220438442,
      "loss": 5.2142,
      "mean_token_accuracy": 0.21065278202295304,
      "num_tokens": 131026215.0,
      "step": 64670
    },
    {
      "entropy": 5.8476887226104735,
      "epoch": 5.651926942235428,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023186746862080145,
      "loss": 5.1973,
      "mean_token_accuracy": 0.22148939669132234,
      "num_tokens": 131036590.0,
      "step": 64675
    },
    {
      "entropy": 5.875887298583985,
      "epoch": 5.652363890588132,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000231836885864523,
      "loss": 5.167,
      "mean_token_accuracy": 0.2174825355410576,
      "num_tokens": 131046198.0,
      "step": 64680
    },
    {
      "entropy": 5.846655035018921,
      "epoch": 5.652800838940837,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023180630393613576,
      "loss": 5.2003,
      "mean_token_accuracy": 0.22145725190639495,
      "num_tokens": 131056312.0,
      "step": 64685
    },
    {
      "entropy": 5.844111967086792,
      "epoch": 5.653237787293542,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023177572283622621,
      "loss": 5.2094,
      "mean_token_accuracy": 0.21627501398324966,
      "num_tokens": 131066389.0,
      "step": 64690
    },
    {
      "entropy": 5.880157947540283,
      "epoch": 5.653674735646247,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000231745142565381,
      "loss": 5.2396,
      "mean_token_accuracy": 0.22180154025554658,
      "num_tokens": 131076064.0,
      "step": 64695
    },
    {
      "entropy": 5.94751238822937,
      "epoch": 5.654111683998951,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023171456312418655,
      "loss": 5.2726,
      "mean_token_accuracy": 0.2155255302786827,
      "num_tokens": 131087299.0,
      "step": 64700
    },
    {
      "entropy": 5.957293891906739,
      "epoch": 5.654548632351656,
      "grad_norm": 1.046875,
      "learning_rate": 0.00023168398451322954,
      "loss": 5.3452,
      "mean_token_accuracy": 0.20758261531591415,
      "num_tokens": 131097065.0,
      "step": 64705
    },
    {
      "entropy": 5.877597332000732,
      "epoch": 5.654985580704361,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023165340673309633,
      "loss": 5.2386,
      "mean_token_accuracy": 0.21140304058790207,
      "num_tokens": 131107948.0,
      "step": 64710
    },
    {
      "entropy": 5.809462070465088,
      "epoch": 5.655422529057065,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023162282978437355,
      "loss": 5.1469,
      "mean_token_accuracy": 0.22220998555421828,
      "num_tokens": 131117963.0,
      "step": 64715
    },
    {
      "entropy": 5.8351599216461185,
      "epoch": 5.65585947740977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023159225366764752,
      "loss": 5.173,
      "mean_token_accuracy": 0.21304525434970856,
      "num_tokens": 131127284.0,
      "step": 64720
    },
    {
      "entropy": 5.8499266624450685,
      "epoch": 5.656296425762475,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002315616783835049,
      "loss": 5.2266,
      "mean_token_accuracy": 0.2152102068066597,
      "num_tokens": 131138112.0,
      "step": 64725
    },
    {
      "entropy": 5.949414539337158,
      "epoch": 5.65673337411518,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023153110393253209,
      "loss": 5.2356,
      "mean_token_accuracy": 0.20712544918060302,
      "num_tokens": 131146368.0,
      "step": 64730
    },
    {
      "entropy": 5.957181119918824,
      "epoch": 5.657170322467884,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023150053031531536,
      "loss": 5.3126,
      "mean_token_accuracy": 0.2105634793639183,
      "num_tokens": 131156784.0,
      "step": 64735
    },
    {
      "entropy": 5.871842622756958,
      "epoch": 5.657607270820589,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023146995753244132,
      "loss": 5.1935,
      "mean_token_accuracy": 0.223556225001812,
      "num_tokens": 131167898.0,
      "step": 64740
    },
    {
      "entropy": 5.835054779052735,
      "epoch": 5.658044219173294,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023143938558449628,
      "loss": 5.1818,
      "mean_token_accuracy": 0.2209285870194435,
      "num_tokens": 131178935.0,
      "step": 64745
    },
    {
      "entropy": 5.875084114074707,
      "epoch": 5.6584811675259985,
      "grad_norm": 1.125,
      "learning_rate": 0.0002314088144720667,
      "loss": 5.1784,
      "mean_token_accuracy": 0.2149837166070938,
      "num_tokens": 131188772.0,
      "step": 64750
    },
    {
      "entropy": 5.881157445907593,
      "epoch": 5.658918115878703,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023137824419573887,
      "loss": 5.1872,
      "mean_token_accuracy": 0.22096271961927413,
      "num_tokens": 131198360.0,
      "step": 64755
    },
    {
      "entropy": 5.8208534717559814,
      "epoch": 5.659355064231407,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023134767475609919,
      "loss": 5.1725,
      "mean_token_accuracy": 0.21640975326299666,
      "num_tokens": 131208288.0,
      "step": 64760
    },
    {
      "entropy": 5.886798095703125,
      "epoch": 5.659792012584113,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023131710615373398,
      "loss": 5.2008,
      "mean_token_accuracy": 0.21507121324539186,
      "num_tokens": 131218999.0,
      "step": 64765
    },
    {
      "entropy": 5.89106240272522,
      "epoch": 5.660228960936817,
      "grad_norm": 1.125,
      "learning_rate": 0.00023128653838922953,
      "loss": 5.283,
      "mean_token_accuracy": 0.2092982679605484,
      "num_tokens": 131230087.0,
      "step": 64770
    },
    {
      "entropy": 5.876003789901733,
      "epoch": 5.660665909289522,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023125597146317217,
      "loss": 5.1747,
      "mean_token_accuracy": 0.21763169020414352,
      "num_tokens": 131240573.0,
      "step": 64775
    },
    {
      "entropy": 5.844267129898071,
      "epoch": 5.661102857642227,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023122540537614828,
      "loss": 5.2089,
      "mean_token_accuracy": 0.21834307610988618,
      "num_tokens": 131250480.0,
      "step": 64780
    },
    {
      "entropy": 5.81480393409729,
      "epoch": 5.661539805994932,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023119484012874397,
      "loss": 5.1868,
      "mean_token_accuracy": 0.21540746837854385,
      "num_tokens": 131261202.0,
      "step": 64785
    },
    {
      "entropy": 5.870804977416992,
      "epoch": 5.661976754347636,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023116427572154557,
      "loss": 5.227,
      "mean_token_accuracy": 0.20880988240242004,
      "num_tokens": 131271810.0,
      "step": 64790
    },
    {
      "entropy": 5.868727540969848,
      "epoch": 5.6624137027003405,
      "grad_norm": 1.125,
      "learning_rate": 0.0002311337121551394,
      "loss": 5.1803,
      "mean_token_accuracy": 0.22162711024284362,
      "num_tokens": 131282691.0,
      "step": 64795
    },
    {
      "entropy": 5.82639102935791,
      "epoch": 5.662850651053046,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023110314943011157,
      "loss": 5.1489,
      "mean_token_accuracy": 0.22540630996227265,
      "num_tokens": 131293022.0,
      "step": 64800
    },
    {
      "entropy": 5.857341957092285,
      "epoch": 5.66328759940575,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002310725875470484,
      "loss": 5.1232,
      "mean_token_accuracy": 0.22453880608081817,
      "num_tokens": 131302694.0,
      "step": 64805
    },
    {
      "entropy": 5.780948781967163,
      "epoch": 5.663724547758455,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023104202650653594,
      "loss": 5.129,
      "mean_token_accuracy": 0.2242789536714554,
      "num_tokens": 131312667.0,
      "step": 64810
    },
    {
      "entropy": 5.798467588424683,
      "epoch": 5.664161496111159,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002310114663091605,
      "loss": 5.1545,
      "mean_token_accuracy": 0.21949863731861113,
      "num_tokens": 131323156.0,
      "step": 64815
    },
    {
      "entropy": 5.838244342803955,
      "epoch": 5.664598444463865,
      "grad_norm": 1.125,
      "learning_rate": 0.00023098090695550816,
      "loss": 5.2319,
      "mean_token_accuracy": 0.22100342065095901,
      "num_tokens": 131333749.0,
      "step": 64820
    },
    {
      "entropy": 5.858516883850098,
      "epoch": 5.665035392816569,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023095034844616502,
      "loss": 5.171,
      "mean_token_accuracy": 0.2243659421801567,
      "num_tokens": 131344613.0,
      "step": 64825
    },
    {
      "entropy": 5.857271194458008,
      "epoch": 5.665472341169274,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023091979078171743,
      "loss": 5.2009,
      "mean_token_accuracy": 0.21629970520734787,
      "num_tokens": 131356250.0,
      "step": 64830
    },
    {
      "entropy": 5.877870559692383,
      "epoch": 5.665909289521979,
      "grad_norm": 1.046875,
      "learning_rate": 0.00023088923396275124,
      "loss": 5.1859,
      "mean_token_accuracy": 0.21792474389076233,
      "num_tokens": 131367021.0,
      "step": 64835
    },
    {
      "entropy": 5.833713483810425,
      "epoch": 5.666346237874683,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023085867798985262,
      "loss": 5.1815,
      "mean_token_accuracy": 0.21636978834867476,
      "num_tokens": 131376872.0,
      "step": 64840
    },
    {
      "entropy": 5.8467662811279295,
      "epoch": 5.666783186227388,
      "grad_norm": 1.125,
      "learning_rate": 0.00023082812286360778,
      "loss": 5.1671,
      "mean_token_accuracy": 0.2230023190379143,
      "num_tokens": 131387428.0,
      "step": 64845
    },
    {
      "entropy": 5.911711263656616,
      "epoch": 5.667220134580092,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023079756858460267,
      "loss": 5.2672,
      "mean_token_accuracy": 0.21065792739391326,
      "num_tokens": 131397496.0,
      "step": 64850
    },
    {
      "entropy": 5.887935304641724,
      "epoch": 5.667657082932798,
      "grad_norm": 1.0,
      "learning_rate": 0.0002307670151534233,
      "loss": 5.1772,
      "mean_token_accuracy": 0.22135951220989228,
      "num_tokens": 131409208.0,
      "step": 64855
    },
    {
      "entropy": 5.878822755813599,
      "epoch": 5.668094031285502,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002307364625706558,
      "loss": 5.1642,
      "mean_token_accuracy": 0.21983003467321396,
      "num_tokens": 131419598.0,
      "step": 64860
    },
    {
      "entropy": 5.906772041320801,
      "epoch": 5.668530979638207,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002307059108368862,
      "loss": 5.2209,
      "mean_token_accuracy": 0.21558733135461808,
      "num_tokens": 131429075.0,
      "step": 64865
    },
    {
      "entropy": 5.841970062255859,
      "epoch": 5.668967927990911,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002306753599527004,
      "loss": 5.1854,
      "mean_token_accuracy": 0.21955386251211167,
      "num_tokens": 131439118.0,
      "step": 64870
    },
    {
      "entropy": 5.796212959289551,
      "epoch": 5.6694048763436165,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023064480991868437,
      "loss": 5.1896,
      "mean_token_accuracy": 0.22358029931783677,
      "num_tokens": 131449803.0,
      "step": 64875
    },
    {
      "entropy": 5.869312429428101,
      "epoch": 5.669841824696321,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023061426073542413,
      "loss": 5.2447,
      "mean_token_accuracy": 0.21149420291185378,
      "num_tokens": 131459535.0,
      "step": 64880
    },
    {
      "entropy": 5.837359142303467,
      "epoch": 5.670278773049025,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023058371240350568,
      "loss": 5.1768,
      "mean_token_accuracy": 0.22202336341142653,
      "num_tokens": 131469429.0,
      "step": 64885
    },
    {
      "entropy": 5.814764833450317,
      "epoch": 5.670715721401731,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002305531649235149,
      "loss": 5.1713,
      "mean_token_accuracy": 0.2251880943775177,
      "num_tokens": 131479425.0,
      "step": 64890
    },
    {
      "entropy": 5.872002220153808,
      "epoch": 5.671152669754435,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002305226182960377,
      "loss": 5.181,
      "mean_token_accuracy": 0.2229581281542778,
      "num_tokens": 131488961.0,
      "step": 64895
    },
    {
      "entropy": 5.849387073516846,
      "epoch": 5.67158961810714,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023049207252165993,
      "loss": 5.1857,
      "mean_token_accuracy": 0.2161399468779564,
      "num_tokens": 131498325.0,
      "step": 64900
    },
    {
      "entropy": 5.878541564941406,
      "epoch": 5.672026566459844,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002304615276009676,
      "loss": 5.2463,
      "mean_token_accuracy": 0.21480497419834138,
      "num_tokens": 131508458.0,
      "step": 64905
    },
    {
      "entropy": 5.876201009750366,
      "epoch": 5.6724635148125495,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023043098353454645,
      "loss": 5.216,
      "mean_token_accuracy": 0.22090312838554382,
      "num_tokens": 131519538.0,
      "step": 64910
    },
    {
      "entropy": 5.818258428573609,
      "epoch": 5.672900463165254,
      "grad_norm": 1.125,
      "learning_rate": 0.00023040044032298252,
      "loss": 5.0978,
      "mean_token_accuracy": 0.2277380034327507,
      "num_tokens": 131528941.0,
      "step": 64915
    },
    {
      "entropy": 5.86895637512207,
      "epoch": 5.6733374115179585,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023036989796686137,
      "loss": 5.1498,
      "mean_token_accuracy": 0.21506713330745697,
      "num_tokens": 131538925.0,
      "step": 64920
    },
    {
      "entropy": 5.88787956237793,
      "epoch": 5.673774359870663,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000230339356466769,
      "loss": 5.2752,
      "mean_token_accuracy": 0.21637689024209977,
      "num_tokens": 131549642.0,
      "step": 64925
    },
    {
      "entropy": 5.7998762130737305,
      "epoch": 5.674211308223368,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023030881582329117,
      "loss": 5.1987,
      "mean_token_accuracy": 0.2193143680691719,
      "num_tokens": 131560417.0,
      "step": 64930
    },
    {
      "entropy": 5.855131769180298,
      "epoch": 5.674648256576073,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002302782760370137,
      "loss": 5.0676,
      "mean_token_accuracy": 0.2288338854908943,
      "num_tokens": 131569792.0,
      "step": 64935
    },
    {
      "entropy": 5.815120792388916,
      "epoch": 5.675085204928777,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023024773710852238,
      "loss": 5.1417,
      "mean_token_accuracy": 0.22168996632099153,
      "num_tokens": 131581889.0,
      "step": 64940
    },
    {
      "entropy": 5.778819370269775,
      "epoch": 5.675522153281483,
      "grad_norm": 1.125,
      "learning_rate": 0.00023021719903840282,
      "loss": 5.0908,
      "mean_token_accuracy": 0.2309970512986183,
      "num_tokens": 131591989.0,
      "step": 64945
    },
    {
      "entropy": 5.792534351348877,
      "epoch": 5.675959101634187,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023018666182724086,
      "loss": 5.1576,
      "mean_token_accuracy": 0.2170906737446785,
      "num_tokens": 131601732.0,
      "step": 64950
    },
    {
      "entropy": 5.856941795349121,
      "epoch": 5.6763960499868915,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002301561254756222,
      "loss": 5.2003,
      "mean_token_accuracy": 0.22251104563474655,
      "num_tokens": 131611773.0,
      "step": 64955
    },
    {
      "entropy": 5.878403520584106,
      "epoch": 5.676832998339596,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002301255899841326,
      "loss": 5.1735,
      "mean_token_accuracy": 0.2179231494665146,
      "num_tokens": 131621670.0,
      "step": 64960
    },
    {
      "entropy": 5.830177068710327,
      "epoch": 5.677269946692301,
      "grad_norm": 1.125,
      "learning_rate": 0.00023009505535335768,
      "loss": 5.2055,
      "mean_token_accuracy": 0.22325861155986787,
      "num_tokens": 131632402.0,
      "step": 64965
    },
    {
      "entropy": 5.8148847103118895,
      "epoch": 5.677706895045006,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023006452158388314,
      "loss": 5.126,
      "mean_token_accuracy": 0.22823969423770904,
      "num_tokens": 131642680.0,
      "step": 64970
    },
    {
      "entropy": 5.821472549438477,
      "epoch": 5.67814384339771,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023003398867629454,
      "loss": 5.157,
      "mean_token_accuracy": 0.2182398706674576,
      "num_tokens": 131652059.0,
      "step": 64975
    },
    {
      "entropy": 5.790825748443604,
      "epoch": 5.678580791750415,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023000345663117767,
      "loss": 5.1085,
      "mean_token_accuracy": 0.22313546985387803,
      "num_tokens": 131663216.0,
      "step": 64980
    },
    {
      "entropy": 5.826331853866577,
      "epoch": 5.67901774010312,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022997292544911808,
      "loss": 5.1831,
      "mean_token_accuracy": 0.21815484762191772,
      "num_tokens": 131674258.0,
      "step": 64985
    },
    {
      "entropy": 5.832058000564575,
      "epoch": 5.679454688455825,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022994239513070142,
      "loss": 5.2132,
      "mean_token_accuracy": 0.2217940330505371,
      "num_tokens": 131684674.0,
      "step": 64990
    },
    {
      "entropy": 5.900372076034546,
      "epoch": 5.679891636808529,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022991186567651318,
      "loss": 5.1772,
      "mean_token_accuracy": 0.21846129149198532,
      "num_tokens": 131694421.0,
      "step": 64995
    },
    {
      "entropy": 5.85906491279602,
      "epoch": 5.680328585161234,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022988133708713894,
      "loss": 5.1586,
      "mean_token_accuracy": 0.2277519091963768,
      "num_tokens": 131703961.0,
      "step": 65000
    },
    {
      "entropy": 5.76245231628418,
      "epoch": 5.680765533513939,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022985080936316433,
      "loss": 5.0892,
      "mean_token_accuracy": 0.23339730948209764,
      "num_tokens": 131713269.0,
      "step": 65005
    },
    {
      "entropy": 5.840978002548217,
      "epoch": 5.681202481866643,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002298202825051749,
      "loss": 5.2166,
      "mean_token_accuracy": 0.21800187081098557,
      "num_tokens": 131723131.0,
      "step": 65010
    },
    {
      "entropy": 5.899763488769532,
      "epoch": 5.681639430219348,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022978975651375606,
      "loss": 5.2839,
      "mean_token_accuracy": 0.21170962154865264,
      "num_tokens": 131734592.0,
      "step": 65015
    },
    {
      "entropy": 5.8475425243377686,
      "epoch": 5.682076378572053,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002297592313894934,
      "loss": 5.1562,
      "mean_token_accuracy": 0.21997249722480774,
      "num_tokens": 131744810.0,
      "step": 65020
    },
    {
      "entropy": 5.854995059967041,
      "epoch": 5.682513326924758,
      "grad_norm": 1.03125,
      "learning_rate": 0.00022972870713297238,
      "loss": 5.1886,
      "mean_token_accuracy": 0.22282300740480424,
      "num_tokens": 131755909.0,
      "step": 65025
    },
    {
      "entropy": 5.878053522109985,
      "epoch": 5.682950275277462,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022969818374477853,
      "loss": 5.1532,
      "mean_token_accuracy": 0.21487950831651687,
      "num_tokens": 131766495.0,
      "step": 65030
    },
    {
      "entropy": 5.883507204055786,
      "epoch": 5.683387223630167,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002296676612254972,
      "loss": 5.2553,
      "mean_token_accuracy": 0.2085601046681404,
      "num_tokens": 131776799.0,
      "step": 65035
    },
    {
      "entropy": 5.745315217971802,
      "epoch": 5.683824171982872,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022963713957571393,
      "loss": 5.0981,
      "mean_token_accuracy": 0.23117981851100922,
      "num_tokens": 131787214.0,
      "step": 65040
    },
    {
      "entropy": 5.860649967193604,
      "epoch": 5.684261120335576,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022960661879601403,
      "loss": 5.2286,
      "mean_token_accuracy": 0.21814724206924438,
      "num_tokens": 131797909.0,
      "step": 65045
    },
    {
      "entropy": 5.859850215911865,
      "epoch": 5.684698068688281,
      "grad_norm": 1.21875,
      "learning_rate": 0.000229576098886983,
      "loss": 5.1563,
      "mean_token_accuracy": 0.22608205080032348,
      "num_tokens": 131808231.0,
      "step": 65050
    },
    {
      "entropy": 5.845982360839844,
      "epoch": 5.685135017040986,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022954557984920622,
      "loss": 5.1849,
      "mean_token_accuracy": 0.21642366349697112,
      "num_tokens": 131817712.0,
      "step": 65055
    },
    {
      "entropy": 5.762848329544068,
      "epoch": 5.685571965393691,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022951506168326902,
      "loss": 5.1441,
      "mean_token_accuracy": 0.23042793124914168,
      "num_tokens": 131828448.0,
      "step": 65060
    },
    {
      "entropy": 5.823200035095215,
      "epoch": 5.686008913746395,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022948454438975676,
      "loss": 5.2174,
      "mean_token_accuracy": 0.22229501456022263,
      "num_tokens": 131838533.0,
      "step": 65065
    },
    {
      "entropy": 5.860029649734497,
      "epoch": 5.6864458620991,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022945402796925474,
      "loss": 5.2138,
      "mean_token_accuracy": 0.2201746091246605,
      "num_tokens": 131849553.0,
      "step": 65070
    },
    {
      "entropy": 5.855438899993897,
      "epoch": 5.686882810451804,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022942351242234837,
      "loss": 5.1747,
      "mean_token_accuracy": 0.21310724765062333,
      "num_tokens": 131858522.0,
      "step": 65075
    },
    {
      "entropy": 5.788983345031738,
      "epoch": 5.6873197588045095,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022939299774962291,
      "loss": 5.1141,
      "mean_token_accuracy": 0.2300764203071594,
      "num_tokens": 131868318.0,
      "step": 65080
    },
    {
      "entropy": 5.847006511688233,
      "epoch": 5.687756707157214,
      "grad_norm": 1.125,
      "learning_rate": 0.0002293624839516637,
      "loss": 5.1902,
      "mean_token_accuracy": 0.22001306265592574,
      "num_tokens": 131878738.0,
      "step": 65085
    },
    {
      "entropy": 5.875554323196411,
      "epoch": 5.688193655509918,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002293319710290559,
      "loss": 5.2363,
      "mean_token_accuracy": 0.2173408940434456,
      "num_tokens": 131890189.0,
      "step": 65090
    },
    {
      "entropy": 5.866769695281983,
      "epoch": 5.688630603862624,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022930145898238485,
      "loss": 5.1075,
      "mean_token_accuracy": 0.22568917274475098,
      "num_tokens": 131898711.0,
      "step": 65095
    },
    {
      "entropy": 5.906430339813232,
      "epoch": 5.689067552215328,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022927094781223574,
      "loss": 5.2753,
      "mean_token_accuracy": 0.20896569192409514,
      "num_tokens": 131909957.0,
      "step": 65100
    },
    {
      "entropy": 5.84233865737915,
      "epoch": 5.689504500568033,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022924043751919383,
      "loss": 5.1962,
      "mean_token_accuracy": 0.21888684332370759,
      "num_tokens": 131919667.0,
      "step": 65105
    },
    {
      "entropy": 5.824660015106201,
      "epoch": 5.689941448920738,
      "grad_norm": 1.125,
      "learning_rate": 0.0002292099281038444,
      "loss": 5.1651,
      "mean_token_accuracy": 0.22555751949548722,
      "num_tokens": 131929806.0,
      "step": 65110
    },
    {
      "entropy": 5.8658100128173825,
      "epoch": 5.6903783972734425,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022917941956677242,
      "loss": 5.1567,
      "mean_token_accuracy": 0.22214357405900956,
      "num_tokens": 131940175.0,
      "step": 65115
    },
    {
      "entropy": 5.780007076263428,
      "epoch": 5.690815345626147,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022914891190856318,
      "loss": 5.1561,
      "mean_token_accuracy": 0.21954728811979293,
      "num_tokens": 131950658.0,
      "step": 65120
    },
    {
      "entropy": 5.916654586791992,
      "epoch": 5.6912522939788515,
      "grad_norm": 1.0625,
      "learning_rate": 0.00022911840512980186,
      "loss": 5.2706,
      "mean_token_accuracy": 0.2097219094634056,
      "num_tokens": 131961224.0,
      "step": 65125
    },
    {
      "entropy": 5.821924877166748,
      "epoch": 5.691689242331556,
      "grad_norm": 1.125,
      "learning_rate": 0.0002290878992310736,
      "loss": 5.1463,
      "mean_token_accuracy": 0.22630221396684647,
      "num_tokens": 131971237.0,
      "step": 65130
    },
    {
      "entropy": 5.8914634704589846,
      "epoch": 5.692126190684261,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002290573942129635,
      "loss": 5.2163,
      "mean_token_accuracy": 0.22441780865192412,
      "num_tokens": 131980410.0,
      "step": 65135
    },
    {
      "entropy": 5.834230661392212,
      "epoch": 5.692563139036966,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002290268900760566,
      "loss": 5.1355,
      "mean_token_accuracy": 0.221695414185524,
      "num_tokens": 131991039.0,
      "step": 65140
    },
    {
      "entropy": 5.796200323104858,
      "epoch": 5.69300008738967,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002289963868209381,
      "loss": 5.1274,
      "mean_token_accuracy": 0.22950029075145723,
      "num_tokens": 132000834.0,
      "step": 65145
    },
    {
      "entropy": 5.88184118270874,
      "epoch": 5.693437035742376,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022896588444819293,
      "loss": 5.1699,
      "mean_token_accuracy": 0.22768033742904664,
      "num_tokens": 132010142.0,
      "step": 65150
    },
    {
      "entropy": 5.8712140083312985,
      "epoch": 5.69387398409508,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022893538295840627,
      "loss": 5.2666,
      "mean_token_accuracy": 0.21515318006277084,
      "num_tokens": 132019951.0,
      "step": 65155
    },
    {
      "entropy": 5.8403733253479,
      "epoch": 5.6943109324477845,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002289048823521631,
      "loss": 5.0955,
      "mean_token_accuracy": 0.22532910704612732,
      "num_tokens": 132029361.0,
      "step": 65160
    },
    {
      "entropy": 5.882469606399536,
      "epoch": 5.69474788080049,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002288743826300484,
      "loss": 5.294,
      "mean_token_accuracy": 0.21131110787391663,
      "num_tokens": 132041225.0,
      "step": 65165
    },
    {
      "entropy": 5.8671534061431885,
      "epoch": 5.695184829153194,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002288438837926472,
      "loss": 5.2204,
      "mean_token_accuracy": 0.2176422655582428,
      "num_tokens": 132051127.0,
      "step": 65170
    },
    {
      "entropy": 5.860711145401001,
      "epoch": 5.695621777505899,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022881338584054457,
      "loss": 5.2037,
      "mean_token_accuracy": 0.2177181750535965,
      "num_tokens": 132060380.0,
      "step": 65175
    },
    {
      "entropy": 5.827767753601075,
      "epoch": 5.696058725858603,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002287828887743254,
      "loss": 5.0938,
      "mean_token_accuracy": 0.22596347481012344,
      "num_tokens": 132071200.0,
      "step": 65180
    },
    {
      "entropy": 5.803839063644409,
      "epoch": 5.696495674211308,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002287523925945746,
      "loss": 5.1194,
      "mean_token_accuracy": 0.22580587565898896,
      "num_tokens": 132079619.0,
      "step": 65185
    },
    {
      "entropy": 5.816324090957641,
      "epoch": 5.696932622564013,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022872189730187704,
      "loss": 5.1329,
      "mean_token_accuracy": 0.21895758360624312,
      "num_tokens": 132088953.0,
      "step": 65190
    },
    {
      "entropy": 5.845720815658569,
      "epoch": 5.697369570916718,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002286914028968179,
      "loss": 5.1311,
      "mean_token_accuracy": 0.22569118738174437,
      "num_tokens": 132099773.0,
      "step": 65195
    },
    {
      "entropy": 5.865204620361328,
      "epoch": 5.697806519269422,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022866090937998185,
      "loss": 5.2591,
      "mean_token_accuracy": 0.2139899030327797,
      "num_tokens": 132109123.0,
      "step": 65200
    },
    {
      "entropy": 5.873080444335938,
      "epoch": 5.698243467622127,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022863041675195379,
      "loss": 5.1481,
      "mean_token_accuracy": 0.2313572123646736,
      "num_tokens": 132119082.0,
      "step": 65205
    },
    {
      "entropy": 5.872237634658814,
      "epoch": 5.698680415974832,
      "grad_norm": 1.125,
      "learning_rate": 0.00022859992501331872,
      "loss": 5.1748,
      "mean_token_accuracy": 0.2214652955532074,
      "num_tokens": 132128690.0,
      "step": 65210
    },
    {
      "entropy": 5.86948971748352,
      "epoch": 5.699117364327536,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022856943416466136,
      "loss": 5.2309,
      "mean_token_accuracy": 0.2106015205383301,
      "num_tokens": 132139324.0,
      "step": 65215
    },
    {
      "entropy": 5.811659574508667,
      "epoch": 5.699554312680241,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002285389442065666,
      "loss": 5.2064,
      "mean_token_accuracy": 0.22248747199773788,
      "num_tokens": 132149342.0,
      "step": 65220
    },
    {
      "entropy": 5.903577518463135,
      "epoch": 5.699991261032946,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022850845513961926,
      "loss": 5.2308,
      "mean_token_accuracy": 0.21743931174278258,
      "num_tokens": 132159568.0,
      "step": 65225
    },
    {
      "entropy": 5.831060409545898,
      "epoch": 5.700428209385651,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002284779669644041,
      "loss": 5.1554,
      "mean_token_accuracy": 0.22731627523899078,
      "num_tokens": 132169818.0,
      "step": 65230
    },
    {
      "entropy": 5.897351312637329,
      "epoch": 5.700865157738355,
      "grad_norm": 1.25,
      "learning_rate": 0.00022844747968150586,
      "loss": 5.2101,
      "mean_token_accuracy": 0.2173832029104233,
      "num_tokens": 132179890.0,
      "step": 65235
    },
    {
      "entropy": 5.839784049987793,
      "epoch": 5.70130210609106,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022841699329150938,
      "loss": 5.1743,
      "mean_token_accuracy": 0.2201608180999756,
      "num_tokens": 132189513.0,
      "step": 65240
    },
    {
      "entropy": 5.836958742141723,
      "epoch": 5.701739054443765,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022838650779499943,
      "loss": 5.1856,
      "mean_token_accuracy": 0.22063399106264114,
      "num_tokens": 132200072.0,
      "step": 65245
    },
    {
      "entropy": 5.893290185928345,
      "epoch": 5.702176002796469,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022835602319256064,
      "loss": 5.2361,
      "mean_token_accuracy": 0.21073865443468093,
      "num_tokens": 132210263.0,
      "step": 65250
    },
    {
      "entropy": 5.850589036941528,
      "epoch": 5.702612951149174,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002283255394847778,
      "loss": 5.2516,
      "mean_token_accuracy": 0.21068027168512343,
      "num_tokens": 132220653.0,
      "step": 65255
    },
    {
      "entropy": 5.856373453140259,
      "epoch": 5.703049899501879,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002282950566722355,
      "loss": 5.2916,
      "mean_token_accuracy": 0.21409098505973817,
      "num_tokens": 132230063.0,
      "step": 65260
    },
    {
      "entropy": 5.885357189178467,
      "epoch": 5.703486847854584,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022826457475551854,
      "loss": 5.2127,
      "mean_token_accuracy": 0.21689701527357103,
      "num_tokens": 132239405.0,
      "step": 65265
    },
    {
      "entropy": 5.921557950973511,
      "epoch": 5.703923796207288,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022823409373521152,
      "loss": 5.2422,
      "mean_token_accuracy": 0.21270377337932586,
      "num_tokens": 132250015.0,
      "step": 65270
    },
    {
      "entropy": 5.967662048339844,
      "epoch": 5.704360744559993,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022820361361189907,
      "loss": 5.1688,
      "mean_token_accuracy": 0.21332550197839736,
      "num_tokens": 132259137.0,
      "step": 65275
    },
    {
      "entropy": 5.870868158340454,
      "epoch": 5.704797692912698,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022817313438616582,
      "loss": 5.1086,
      "mean_token_accuracy": 0.22945889681577683,
      "num_tokens": 132269292.0,
      "step": 65280
    },
    {
      "entropy": 5.918927621841431,
      "epoch": 5.7052346412654025,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022814265605859647,
      "loss": 5.302,
      "mean_token_accuracy": 0.21252593398094177,
      "num_tokens": 132279582.0,
      "step": 65285
    },
    {
      "entropy": 5.859821414947509,
      "epoch": 5.705671589618107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022811217862977545,
      "loss": 5.1839,
      "mean_token_accuracy": 0.21870081275701522,
      "num_tokens": 132289996.0,
      "step": 65290
    },
    {
      "entropy": 5.832814073562622,
      "epoch": 5.706108537970811,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022808170210028745,
      "loss": 5.1549,
      "mean_token_accuracy": 0.22301093339920045,
      "num_tokens": 132299861.0,
      "step": 65295
    },
    {
      "entropy": 5.920694065093994,
      "epoch": 5.706545486323517,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022805122647071697,
      "loss": 5.1703,
      "mean_token_accuracy": 0.21650572419166564,
      "num_tokens": 132310773.0,
      "step": 65300
    },
    {
      "entropy": 5.846638345718384,
      "epoch": 5.706982434676221,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022802075174164853,
      "loss": 5.2259,
      "mean_token_accuracy": 0.21476281881332399,
      "num_tokens": 132320863.0,
      "step": 65305
    },
    {
      "entropy": 5.874865961074829,
      "epoch": 5.707419383028926,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022799027791366673,
      "loss": 5.1783,
      "mean_token_accuracy": 0.22509964406490326,
      "num_tokens": 132330635.0,
      "step": 65310
    },
    {
      "entropy": 5.862723350524902,
      "epoch": 5.707856331381631,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002279598049873561,
      "loss": 5.2039,
      "mean_token_accuracy": 0.2160668268799782,
      "num_tokens": 132340732.0,
      "step": 65315
    },
    {
      "entropy": 5.865488004684448,
      "epoch": 5.7082932797343355,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022792933296330093,
      "loss": 5.2344,
      "mean_token_accuracy": 0.21385512351989747,
      "num_tokens": 132351830.0,
      "step": 65320
    },
    {
      "entropy": 5.843263244628906,
      "epoch": 5.70873022808704,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002278988618420859,
      "loss": 5.1398,
      "mean_token_accuracy": 0.23117070496082306,
      "num_tokens": 132361851.0,
      "step": 65325
    },
    {
      "entropy": 5.878447580337524,
      "epoch": 5.7091671764397445,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022786839162429535,
      "loss": 5.2865,
      "mean_token_accuracy": 0.21376174837350845,
      "num_tokens": 132372387.0,
      "step": 65330
    },
    {
      "entropy": 5.895508480072022,
      "epoch": 5.70960412479245,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002278379223105137,
      "loss": 5.2265,
      "mean_token_accuracy": 0.21402270048856736,
      "num_tokens": 132382596.0,
      "step": 65335
    },
    {
      "entropy": 5.886287784576416,
      "epoch": 5.710041073145154,
      "grad_norm": 1.125,
      "learning_rate": 0.00022780745390132552,
      "loss": 5.2445,
      "mean_token_accuracy": 0.21527181267738343,
      "num_tokens": 132393517.0,
      "step": 65340
    },
    {
      "entropy": 5.8294683456420895,
      "epoch": 5.710478021497859,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022777698639731502,
      "loss": 5.1085,
      "mean_token_accuracy": 0.21794841438531876,
      "num_tokens": 132403797.0,
      "step": 65345
    },
    {
      "entropy": 5.872380685806275,
      "epoch": 5.710914969850563,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002277465197990667,
      "loss": 5.1758,
      "mean_token_accuracy": 0.2250344917178154,
      "num_tokens": 132413863.0,
      "step": 65350
    },
    {
      "entropy": 5.857454061508179,
      "epoch": 5.711351918203269,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002277160541071649,
      "loss": 5.1684,
      "mean_token_accuracy": 0.22557390183210374,
      "num_tokens": 132424169.0,
      "step": 65355
    },
    {
      "entropy": 5.843824625015259,
      "epoch": 5.711788866555973,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022768558932219392,
      "loss": 5.211,
      "mean_token_accuracy": 0.21831227093935013,
      "num_tokens": 132433594.0,
      "step": 65360
    },
    {
      "entropy": 5.862755060195923,
      "epoch": 5.7122258149086775,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002276551254447382,
      "loss": 5.2523,
      "mean_token_accuracy": 0.21209479719400406,
      "num_tokens": 132443266.0,
      "step": 65365
    },
    {
      "entropy": 5.9036643505096436,
      "epoch": 5.712662763261383,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022762466247538193,
      "loss": 5.2192,
      "mean_token_accuracy": 0.21840358525514603,
      "num_tokens": 132454283.0,
      "step": 65370
    },
    {
      "entropy": 5.865748119354248,
      "epoch": 5.713099711614087,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002275942004147094,
      "loss": 5.1271,
      "mean_token_accuracy": 0.22464263141155244,
      "num_tokens": 132463747.0,
      "step": 65375
    },
    {
      "entropy": 5.825870180130005,
      "epoch": 5.713536659966792,
      "grad_norm": 1.0625,
      "learning_rate": 0.00022756373926330505,
      "loss": 5.1551,
      "mean_token_accuracy": 0.22560169398784638,
      "num_tokens": 132475135.0,
      "step": 65380
    },
    {
      "entropy": 5.810513591766357,
      "epoch": 5.713973608319496,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00022753327902175297,
      "loss": 5.1351,
      "mean_token_accuracy": 0.22443547248840331,
      "num_tokens": 132484749.0,
      "step": 65385
    },
    {
      "entropy": 5.890503311157227,
      "epoch": 5.714410556672202,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022750281969063753,
      "loss": 5.2307,
      "mean_token_accuracy": 0.2118915170431137,
      "num_tokens": 132494583.0,
      "step": 65390
    },
    {
      "entropy": 5.815858602523804,
      "epoch": 5.714847505024906,
      "grad_norm": 1.125,
      "learning_rate": 0.00022747236127054283,
      "loss": 5.1606,
      "mean_token_accuracy": 0.21931374669075013,
      "num_tokens": 132504900.0,
      "step": 65395
    },
    {
      "entropy": 5.876162910461426,
      "epoch": 5.715284453377611,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002274419037620532,
      "loss": 5.2316,
      "mean_token_accuracy": 0.21166356652975082,
      "num_tokens": 132515753.0,
      "step": 65400
    },
    {
      "entropy": 5.893593692779541,
      "epoch": 5.715721401730315,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022741144716575273,
      "loss": 5.2514,
      "mean_token_accuracy": 0.2185954585671425,
      "num_tokens": 132526097.0,
      "step": 65405
    },
    {
      "entropy": 5.856105613708496,
      "epoch": 5.71615835008302,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022738099148222567,
      "loss": 5.2085,
      "mean_token_accuracy": 0.2233396515250206,
      "num_tokens": 132535617.0,
      "step": 65410
    },
    {
      "entropy": 5.901138162612915,
      "epoch": 5.716595298435725,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002273505367120562,
      "loss": 5.2622,
      "mean_token_accuracy": 0.21386627852916718,
      "num_tokens": 132547409.0,
      "step": 65415
    },
    {
      "entropy": 5.846056604385376,
      "epoch": 5.717032246788429,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022732008285582828,
      "loss": 5.1823,
      "mean_token_accuracy": 0.22105204612016677,
      "num_tokens": 132557691.0,
      "step": 65420
    },
    {
      "entropy": 5.817932891845703,
      "epoch": 5.717469195141135,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022728962991412622,
      "loss": 5.0869,
      "mean_token_accuracy": 0.22213232517242432,
      "num_tokens": 132568045.0,
      "step": 65425
    },
    {
      "entropy": 5.795180940628052,
      "epoch": 5.717906143493839,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022725917788753402,
      "loss": 5.0981,
      "mean_token_accuracy": 0.22614047676324844,
      "num_tokens": 132577577.0,
      "step": 65430
    },
    {
      "entropy": 5.821645736694336,
      "epoch": 5.718343091846544,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022722872677663586,
      "loss": 5.1333,
      "mean_token_accuracy": 0.21803822219371796,
      "num_tokens": 132588804.0,
      "step": 65435
    },
    {
      "entropy": 5.787672996520996,
      "epoch": 5.718780040199248,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022719827658201574,
      "loss": 5.0947,
      "mean_token_accuracy": 0.23186817169189453,
      "num_tokens": 132599417.0,
      "step": 65440
    },
    {
      "entropy": 5.915970516204834,
      "epoch": 5.7192169885519535,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022716782730425772,
      "loss": 5.2751,
      "mean_token_accuracy": 0.20268092900514603,
      "num_tokens": 132609594.0,
      "step": 65445
    },
    {
      "entropy": 5.903883457183838,
      "epoch": 5.719653936904658,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002271373789439458,
      "loss": 5.2566,
      "mean_token_accuracy": 0.21137389838695525,
      "num_tokens": 132618686.0,
      "step": 65450
    },
    {
      "entropy": 5.908148956298828,
      "epoch": 5.720090885257362,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022710693150166406,
      "loss": 5.2781,
      "mean_token_accuracy": 0.20984719693660736,
      "num_tokens": 132628442.0,
      "step": 65455
    },
    {
      "entropy": 5.86840853691101,
      "epoch": 5.720527833610067,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022707648497799656,
      "loss": 5.2036,
      "mean_token_accuracy": 0.217788627743721,
      "num_tokens": 132637279.0,
      "step": 65460
    },
    {
      "entropy": 5.829380750656128,
      "epoch": 5.720964781962772,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022704603937352707,
      "loss": 5.1482,
      "mean_token_accuracy": 0.21925080418586732,
      "num_tokens": 132646377.0,
      "step": 65465
    },
    {
      "entropy": 5.803254890441894,
      "epoch": 5.721401730315477,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002270155946888397,
      "loss": 5.14,
      "mean_token_accuracy": 0.22938555479049683,
      "num_tokens": 132656614.0,
      "step": 65470
    },
    {
      "entropy": 5.838298082351685,
      "epoch": 5.721838678668181,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022698515092451832,
      "loss": 5.1886,
      "mean_token_accuracy": 0.22408915609121322,
      "num_tokens": 132667032.0,
      "step": 65475
    },
    {
      "entropy": 5.798021125793457,
      "epoch": 5.7222756270208865,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000226954708081147,
      "loss": 5.1191,
      "mean_token_accuracy": 0.225356125831604,
      "num_tokens": 132676614.0,
      "step": 65480
    },
    {
      "entropy": 5.833185815811158,
      "epoch": 5.722712575373591,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002269242661593095,
      "loss": 5.1565,
      "mean_token_accuracy": 0.22191443741321565,
      "num_tokens": 132686532.0,
      "step": 65485
    },
    {
      "entropy": 5.831837797164917,
      "epoch": 5.7231495237262955,
      "grad_norm": 1.0625,
      "learning_rate": 0.00022689382515958978,
      "loss": 5.169,
      "mean_token_accuracy": 0.2199942409992218,
      "num_tokens": 132697462.0,
      "step": 65490
    },
    {
      "entropy": 5.925391006469726,
      "epoch": 5.723586472079,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002268633850825717,
      "loss": 5.2685,
      "mean_token_accuracy": 0.21518985331058502,
      "num_tokens": 132708079.0,
      "step": 65495
    },
    {
      "entropy": 5.910814666748047,
      "epoch": 5.724023420431705,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022683294592883908,
      "loss": 5.2448,
      "mean_token_accuracy": 0.21406330168247223,
      "num_tokens": 132718059.0,
      "step": 65500
    },
    {
      "entropy": 5.86916561126709,
      "epoch": 5.72446036878441,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002268025076989758,
      "loss": 5.2289,
      "mean_token_accuracy": 0.21140081882476808,
      "num_tokens": 132727835.0,
      "step": 65505
    },
    {
      "entropy": 5.837273454666137,
      "epoch": 5.724897317137114,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002267720703935658,
      "loss": 5.1514,
      "mean_token_accuracy": 0.22424449175596237,
      "num_tokens": 132738042.0,
      "step": 65510
    },
    {
      "entropy": 5.776043558120728,
      "epoch": 5.725334265489819,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022674163401319264,
      "loss": 5.1227,
      "mean_token_accuracy": 0.22932112216949463,
      "num_tokens": 132747436.0,
      "step": 65515
    },
    {
      "entropy": 5.856878805160522,
      "epoch": 5.725771213842524,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022671119855844024,
      "loss": 5.1958,
      "mean_token_accuracy": 0.21545734405517578,
      "num_tokens": 132758400.0,
      "step": 65520
    },
    {
      "entropy": 5.824603605270386,
      "epoch": 5.7262081621952285,
      "grad_norm": 1.015625,
      "learning_rate": 0.00022668076402989235,
      "loss": 5.145,
      "mean_token_accuracy": 0.22076449394226075,
      "num_tokens": 132768187.0,
      "step": 65525
    },
    {
      "entropy": 5.852774524688721,
      "epoch": 5.726645110547933,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022665033042813272,
      "loss": 5.1859,
      "mean_token_accuracy": 0.21936444491147994,
      "num_tokens": 132778064.0,
      "step": 65530
    },
    {
      "entropy": 5.891811323165894,
      "epoch": 5.727082058900638,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022661989775374518,
      "loss": 5.2338,
      "mean_token_accuracy": 0.21055000871419907,
      "num_tokens": 132788059.0,
      "step": 65535
    },
    {
      "entropy": 5.873929071426391,
      "epoch": 5.727519007253343,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00022658946600731322,
      "loss": 5.2269,
      "mean_token_accuracy": 0.21563052386045456,
      "num_tokens": 132798693.0,
      "step": 65540
    },
    {
      "entropy": 5.867721700668335,
      "epoch": 5.727955955606047,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00022655903518942067,
      "loss": 5.1826,
      "mean_token_accuracy": 0.2166870892047882,
      "num_tokens": 132809186.0,
      "step": 65545
    },
    {
      "entropy": 5.832508897781372,
      "epoch": 5.728392903958752,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022652860530065133,
      "loss": 5.1455,
      "mean_token_accuracy": 0.22062283754348755,
      "num_tokens": 132820171.0,
      "step": 65550
    },
    {
      "entropy": 5.843926239013672,
      "epoch": 5.728829852311457,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00022649817634158866,
      "loss": 5.2503,
      "mean_token_accuracy": 0.22261011749505996,
      "num_tokens": 132831140.0,
      "step": 65555
    },
    {
      "entropy": 5.897799253463745,
      "epoch": 5.729266800664162,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00022646774831281636,
      "loss": 5.2505,
      "mean_token_accuracy": 0.21022696942090988,
      "num_tokens": 132841591.0,
      "step": 65560
    },
    {
      "entropy": 5.879287433624268,
      "epoch": 5.729703749016866,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022643732121491812,
      "loss": 5.1818,
      "mean_token_accuracy": 0.21669285148382186,
      "num_tokens": 132850930.0,
      "step": 65565
    },
    {
      "entropy": 5.844901132583618,
      "epoch": 5.7301406973695705,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022640689504847751,
      "loss": 5.1742,
      "mean_token_accuracy": 0.22043840289115907,
      "num_tokens": 132861169.0,
      "step": 65570
    },
    {
      "entropy": 5.863872623443603,
      "epoch": 5.730577645722276,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002263764698140781,
      "loss": 5.2113,
      "mean_token_accuracy": 0.21050239503383636,
      "num_tokens": 132871002.0,
      "step": 65575
    },
    {
      "entropy": 5.828380107879639,
      "epoch": 5.73101459407498,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022634604551230347,
      "loss": 5.1792,
      "mean_token_accuracy": 0.2205774173140526,
      "num_tokens": 132880897.0,
      "step": 65580
    },
    {
      "entropy": 5.8081457138061525,
      "epoch": 5.731451542427685,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022631562214373718,
      "loss": 5.0736,
      "mean_token_accuracy": 0.22893522679805756,
      "num_tokens": 132890955.0,
      "step": 65585
    },
    {
      "entropy": 5.813956022262573,
      "epoch": 5.73188849078039,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022628519970896283,
      "loss": 5.081,
      "mean_token_accuracy": 0.22950173318386077,
      "num_tokens": 132900751.0,
      "step": 65590
    },
    {
      "entropy": 5.889012384414673,
      "epoch": 5.732325439133095,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022625477820856384,
      "loss": 5.2845,
      "mean_token_accuracy": 0.21571284383535386,
      "num_tokens": 132911699.0,
      "step": 65595
    },
    {
      "entropy": 5.802113103866577,
      "epoch": 5.732762387485799,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022622435764312372,
      "loss": 5.1714,
      "mean_token_accuracy": 0.22216500490903854,
      "num_tokens": 132922282.0,
      "step": 65600
    },
    {
      "entropy": 5.850289058685303,
      "epoch": 5.733199335838504,
      "grad_norm": 1.0625,
      "learning_rate": 0.000226193938013226,
      "loss": 5.1769,
      "mean_token_accuracy": 0.21123681962490082,
      "num_tokens": 132932443.0,
      "step": 65605
    },
    {
      "entropy": 5.897161674499512,
      "epoch": 5.733636284191209,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022616351931945412,
      "loss": 5.2236,
      "mean_token_accuracy": 0.21650906205177306,
      "num_tokens": 132941966.0,
      "step": 65610
    },
    {
      "entropy": 5.905399751663208,
      "epoch": 5.734073232543913,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022613310156239152,
      "loss": 5.2662,
      "mean_token_accuracy": 0.21297476142644883,
      "num_tokens": 132952931.0,
      "step": 65615
    },
    {
      "entropy": 5.8592523574829105,
      "epoch": 5.734510180896618,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022610268474262168,
      "loss": 5.1646,
      "mean_token_accuracy": 0.21798968613147734,
      "num_tokens": 132963845.0,
      "step": 65620
    },
    {
      "entropy": 5.856521940231323,
      "epoch": 5.734947129249322,
      "grad_norm": 1.125,
      "learning_rate": 0.0002260722688607279,
      "loss": 5.2328,
      "mean_token_accuracy": 0.2198347494006157,
      "num_tokens": 132973901.0,
      "step": 65625
    },
    {
      "entropy": 5.863951539993286,
      "epoch": 5.735384077602028,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002260418539172937,
      "loss": 5.2299,
      "mean_token_accuracy": 0.21850849092006683,
      "num_tokens": 132983914.0,
      "step": 65630
    },
    {
      "entropy": 5.858559942245483,
      "epoch": 5.735821025954732,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022601143991290236,
      "loss": 5.194,
      "mean_token_accuracy": 0.21952730268239976,
      "num_tokens": 132993823.0,
      "step": 65635
    },
    {
      "entropy": 5.842055320739746,
      "epoch": 5.736257974307437,
      "grad_norm": 1.125,
      "learning_rate": 0.00022598102684813726,
      "loss": 5.2382,
      "mean_token_accuracy": 0.21860575079917907,
      "num_tokens": 133003931.0,
      "step": 65640
    },
    {
      "entropy": 5.8900562763214115,
      "epoch": 5.736694922660142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022595061472358181,
      "loss": 5.2024,
      "mean_token_accuracy": 0.22107594460248947,
      "num_tokens": 133014823.0,
      "step": 65645
    },
    {
      "entropy": 5.8563508033752445,
      "epoch": 5.7371318710128465,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022592020353981923,
      "loss": 5.1451,
      "mean_token_accuracy": 0.22573865950107574,
      "num_tokens": 133024789.0,
      "step": 65650
    },
    {
      "entropy": 5.865795230865478,
      "epoch": 5.737568819365551,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022588979329743288,
      "loss": 5.1487,
      "mean_token_accuracy": 0.2188168376684189,
      "num_tokens": 133034134.0,
      "step": 65655
    },
    {
      "entropy": 5.746505403518677,
      "epoch": 5.738005767718255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000225859383997006,
      "loss": 5.0625,
      "mean_token_accuracy": 0.2312395989894867,
      "num_tokens": 133044935.0,
      "step": 65660
    },
    {
      "entropy": 5.874696969985962,
      "epoch": 5.738442716070961,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00022582897563912187,
      "loss": 5.2564,
      "mean_token_accuracy": 0.217007552087307,
      "num_tokens": 133054999.0,
      "step": 65665
    },
    {
      "entropy": 5.8564817905426025,
      "epoch": 5.738879664423665,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022579856822436384,
      "loss": 5.1995,
      "mean_token_accuracy": 0.21746398210525514,
      "num_tokens": 133064912.0,
      "step": 65670
    },
    {
      "entropy": 5.892544364929199,
      "epoch": 5.73931661277637,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000225768161753315,
      "loss": 5.2394,
      "mean_token_accuracy": 0.21861063092947006,
      "num_tokens": 133075594.0,
      "step": 65675
    },
    {
      "entropy": 5.834579229354858,
      "epoch": 5.739753561129074,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022573775622655856,
      "loss": 5.1809,
      "mean_token_accuracy": 0.22413267195224762,
      "num_tokens": 133086333.0,
      "step": 65680
    },
    {
      "entropy": 5.7382323265075685,
      "epoch": 5.74019050948178,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022570735164467783,
      "loss": 5.0361,
      "mean_token_accuracy": 0.23218736797571182,
      "num_tokens": 133096355.0,
      "step": 65685
    },
    {
      "entropy": 5.885757112503052,
      "epoch": 5.740627457834484,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002256769480082559,
      "loss": 5.2716,
      "mean_token_accuracy": 0.21096325814723968,
      "num_tokens": 133108096.0,
      "step": 65690
    },
    {
      "entropy": 5.888557863235474,
      "epoch": 5.7410644061871885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022564654531787598,
      "loss": 5.2359,
      "mean_token_accuracy": 0.21609672158956528,
      "num_tokens": 133117985.0,
      "step": 65695
    },
    {
      "entropy": 5.90753583908081,
      "epoch": 5.741501354539894,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002256161435741212,
      "loss": 5.2101,
      "mean_token_accuracy": 0.21602988243103027,
      "num_tokens": 133128800.0,
      "step": 65700
    },
    {
      "entropy": 5.854672622680664,
      "epoch": 5.741938302892598,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002255857427775746,
      "loss": 5.184,
      "mean_token_accuracy": 0.22600213438272476,
      "num_tokens": 133138392.0,
      "step": 65705
    },
    {
      "entropy": 5.875291061401367,
      "epoch": 5.742375251245303,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022555534292881935,
      "loss": 5.2195,
      "mean_token_accuracy": 0.2193059504032135,
      "num_tokens": 133148567.0,
      "step": 65710
    },
    {
      "entropy": 5.8141154766082765,
      "epoch": 5.742812199598007,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00022552494402843854,
      "loss": 5.1787,
      "mean_token_accuracy": 0.22290997803211213,
      "num_tokens": 133159741.0,
      "step": 65715
    },
    {
      "entropy": 5.896880626678467,
      "epoch": 5.743249147950712,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002254945460770153,
      "loss": 5.1844,
      "mean_token_accuracy": 0.21317808777093888,
      "num_tokens": 133169961.0,
      "step": 65720
    },
    {
      "entropy": 5.87728910446167,
      "epoch": 5.743686096303417,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002254641490751325,
      "loss": 5.1538,
      "mean_token_accuracy": 0.2164826825261116,
      "num_tokens": 133179903.0,
      "step": 65725
    },
    {
      "entropy": 5.889117097854614,
      "epoch": 5.744123044656122,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022543375302337337,
      "loss": 5.2284,
      "mean_token_accuracy": 0.21606453359127045,
      "num_tokens": 133191292.0,
      "step": 65730
    },
    {
      "entropy": 5.8681943893432615,
      "epoch": 5.744559993008826,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022540335792232074,
      "loss": 5.1779,
      "mean_token_accuracy": 0.2147350549697876,
      "num_tokens": 133201474.0,
      "step": 65735
    },
    {
      "entropy": 5.918263578414917,
      "epoch": 5.744996941361531,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022537296377255777,
      "loss": 5.2828,
      "mean_token_accuracy": 0.21438754498958587,
      "num_tokens": 133211470.0,
      "step": 65740
    },
    {
      "entropy": 5.794482946395874,
      "epoch": 5.745433889714236,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022534257057466733,
      "loss": 5.1648,
      "mean_token_accuracy": 0.22145593911409378,
      "num_tokens": 133221516.0,
      "step": 65745
    },
    {
      "entropy": 5.820031356811524,
      "epoch": 5.74587083806694,
      "grad_norm": 1.125,
      "learning_rate": 0.0002253121783292324,
      "loss": 5.175,
      "mean_token_accuracy": 0.21621614396572114,
      "num_tokens": 133232400.0,
      "step": 65750
    },
    {
      "entropy": 5.849058628082275,
      "epoch": 5.746307786419646,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022528178703683593,
      "loss": 5.2421,
      "mean_token_accuracy": 0.2101709544658661,
      "num_tokens": 133242823.0,
      "step": 65755
    },
    {
      "entropy": 5.882054233551026,
      "epoch": 5.74674473477235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022525139669806084,
      "loss": 5.1268,
      "mean_token_accuracy": 0.22298028767108918,
      "num_tokens": 133252289.0,
      "step": 65760
    },
    {
      "entropy": 5.9016828536987305,
      "epoch": 5.747181683125055,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022522100731349004,
      "loss": 5.1786,
      "mean_token_accuracy": 0.2213909924030304,
      "num_tokens": 133262506.0,
      "step": 65765
    },
    {
      "entropy": 5.8600085258483885,
      "epoch": 5.747618631477759,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002251906188837064,
      "loss": 5.2222,
      "mean_token_accuracy": 0.20607935190200805,
      "num_tokens": 133272993.0,
      "step": 65770
    },
    {
      "entropy": 5.8763954639434814,
      "epoch": 5.748055579830464,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022516023140929276,
      "loss": 5.282,
      "mean_token_accuracy": 0.20745983868837356,
      "num_tokens": 133283492.0,
      "step": 65775
    },
    {
      "entropy": 5.91414704322815,
      "epoch": 5.748492528183169,
      "grad_norm": 1.125,
      "learning_rate": 0.00022512984489083204,
      "loss": 5.2108,
      "mean_token_accuracy": 0.21829716861248016,
      "num_tokens": 133293164.0,
      "step": 65780
    },
    {
      "entropy": 5.821771097183228,
      "epoch": 5.748929476535873,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022509945932890698,
      "loss": 5.0692,
      "mean_token_accuracy": 0.2336355686187744,
      "num_tokens": 133303590.0,
      "step": 65785
    },
    {
      "entropy": 5.800416088104248,
      "epoch": 5.749366424888578,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002250690747241005,
      "loss": 5.0999,
      "mean_token_accuracy": 0.23185079097747802,
      "num_tokens": 133313716.0,
      "step": 65790
    },
    {
      "entropy": 5.776414299011231,
      "epoch": 5.749803373241283,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022503869107699532,
      "loss": 5.1061,
      "mean_token_accuracy": 0.22252476662397386,
      "num_tokens": 133323117.0,
      "step": 65795
    },
    {
      "entropy": 5.911239385604858,
      "epoch": 5.750240321593988,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022500830838817417,
      "loss": 5.3273,
      "mean_token_accuracy": 0.21112513393163682,
      "num_tokens": 133333185.0,
      "step": 65800
    },
    {
      "entropy": 5.875881052017212,
      "epoch": 5.750677269946692,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022497792665821987,
      "loss": 5.2424,
      "mean_token_accuracy": 0.2169290155172348,
      "num_tokens": 133343036.0,
      "step": 65805
    },
    {
      "entropy": 5.954621171951294,
      "epoch": 5.7511142182993975,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022494754588771516,
      "loss": 5.2723,
      "mean_token_accuracy": 0.2145085647702217,
      "num_tokens": 133353277.0,
      "step": 65810
    },
    {
      "entropy": 5.879582405090332,
      "epoch": 5.751551166652102,
      "grad_norm": 1.125,
      "learning_rate": 0.00022491716607724277,
      "loss": 5.2825,
      "mean_token_accuracy": 0.2133282706141472,
      "num_tokens": 133364600.0,
      "step": 65815
    },
    {
      "entropy": 5.917683696746826,
      "epoch": 5.7519881150048064,
      "grad_norm": 1.25,
      "learning_rate": 0.00022488678722738537,
      "loss": 5.2582,
      "mean_token_accuracy": 0.2143932119011879,
      "num_tokens": 133374819.0,
      "step": 65820
    },
    {
      "entropy": 5.901544046401978,
      "epoch": 5.752425063357511,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002248564093387256,
      "loss": 5.217,
      "mean_token_accuracy": 0.2236375629901886,
      "num_tokens": 133384756.0,
      "step": 65825
    },
    {
      "entropy": 5.830971765518188,
      "epoch": 5.752862011710215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00022482603241184618,
      "loss": 5.1947,
      "mean_token_accuracy": 0.22095291316509247,
      "num_tokens": 133395668.0,
      "step": 65830
    },
    {
      "entropy": 5.878186941146851,
      "epoch": 5.753298960062921,
      "grad_norm": 1.125,
      "learning_rate": 0.0002247956564473298,
      "loss": 5.2143,
      "mean_token_accuracy": 0.21391136050224305,
      "num_tokens": 133406513.0,
      "step": 65835
    },
    {
      "entropy": 5.895036458969116,
      "epoch": 5.753735908415625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002247652814457591,
      "loss": 5.2211,
      "mean_token_accuracy": 0.21384386718273163,
      "num_tokens": 133416543.0,
      "step": 65840
    },
    {
      "entropy": 5.892801332473755,
      "epoch": 5.75417285676833,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022473490740771652,
      "loss": 5.1936,
      "mean_token_accuracy": 0.21725016683340073,
      "num_tokens": 133427163.0,
      "step": 65845
    },
    {
      "entropy": 5.951254796981812,
      "epoch": 5.754609805121035,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022470453433378473,
      "loss": 5.2609,
      "mean_token_accuracy": 0.21043774634599685,
      "num_tokens": 133437746.0,
      "step": 65850
    },
    {
      "entropy": 5.857697534561157,
      "epoch": 5.7550467534737395,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002246741622245464,
      "loss": 5.1891,
      "mean_token_accuracy": 0.22173381298780442,
      "num_tokens": 133446876.0,
      "step": 65855
    },
    {
      "entropy": 5.858274984359741,
      "epoch": 5.755483701826444,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022464379108058397,
      "loss": 5.2384,
      "mean_token_accuracy": 0.21714155226945878,
      "num_tokens": 133456504.0,
      "step": 65860
    },
    {
      "entropy": 5.864192962646484,
      "epoch": 5.7559206501791484,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022461342090248005,
      "loss": 5.2125,
      "mean_token_accuracy": 0.213756163418293,
      "num_tokens": 133466997.0,
      "step": 65865
    },
    {
      "entropy": 5.942999696731567,
      "epoch": 5.756357598531854,
      "grad_norm": 1.0625,
      "learning_rate": 0.00022458305169081711,
      "loss": 5.2828,
      "mean_token_accuracy": 0.20567047595977783,
      "num_tokens": 133478558.0,
      "step": 65870
    },
    {
      "entropy": 5.941318082809448,
      "epoch": 5.756794546884558,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022455268344617762,
      "loss": 5.244,
      "mean_token_accuracy": 0.21414451599121093,
      "num_tokens": 133489356.0,
      "step": 65875
    },
    {
      "entropy": 5.8918627262115475,
      "epoch": 5.757231495237263,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002245223161691441,
      "loss": 5.2131,
      "mean_token_accuracy": 0.22331370413303375,
      "num_tokens": 133498556.0,
      "step": 65880
    },
    {
      "entropy": 5.809934616088867,
      "epoch": 5.757668443589967,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022449194986029903,
      "loss": 5.2284,
      "mean_token_accuracy": 0.219138203561306,
      "num_tokens": 133508928.0,
      "step": 65885
    },
    {
      "entropy": 5.85606107711792,
      "epoch": 5.758105391942673,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002244615845202248,
      "loss": 5.2163,
      "mean_token_accuracy": 0.219800141453743,
      "num_tokens": 133519677.0,
      "step": 65890
    },
    {
      "entropy": 5.925543212890625,
      "epoch": 5.758542340295377,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002244312201495039,
      "loss": 5.2183,
      "mean_token_accuracy": 0.21944824755191802,
      "num_tokens": 133530937.0,
      "step": 65895
    },
    {
      "entropy": 5.943887186050415,
      "epoch": 5.7589792886480815,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00022440085674871863,
      "loss": 5.2153,
      "mean_token_accuracy": 0.21769786179065703,
      "num_tokens": 133541531.0,
      "step": 65900
    },
    {
      "entropy": 5.852562618255615,
      "epoch": 5.759416237000787,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022437049431845146,
      "loss": 5.1949,
      "mean_token_accuracy": 0.222142818570137,
      "num_tokens": 133552214.0,
      "step": 65905
    },
    {
      "entropy": 5.78638014793396,
      "epoch": 5.759853185353491,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022434013285928478,
      "loss": 5.1059,
      "mean_token_accuracy": 0.23008139580488204,
      "num_tokens": 133562388.0,
      "step": 65910
    },
    {
      "entropy": 5.848232221603394,
      "epoch": 5.760290133706196,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022430977237180083,
      "loss": 5.1286,
      "mean_token_accuracy": 0.22227615714073182,
      "num_tokens": 133572655.0,
      "step": 65915
    },
    {
      "entropy": 5.893653106689453,
      "epoch": 5.7607270820589,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002242794128565821,
      "loss": 5.1713,
      "mean_token_accuracy": 0.2235310673713684,
      "num_tokens": 133582977.0,
      "step": 65920
    },
    {
      "entropy": 5.83491096496582,
      "epoch": 5.761164030411606,
      "grad_norm": 1.25,
      "learning_rate": 0.00022424905431421078,
      "loss": 5.19,
      "mean_token_accuracy": 0.21767399758100509,
      "num_tokens": 133592410.0,
      "step": 65925
    },
    {
      "entropy": 5.812216281890869,
      "epoch": 5.76160097876431,
      "grad_norm": 1.125,
      "learning_rate": 0.0002242186967452692,
      "loss": 5.1769,
      "mean_token_accuracy": 0.22394211441278458,
      "num_tokens": 133602588.0,
      "step": 65930
    },
    {
      "entropy": 5.80299744606018,
      "epoch": 5.762037927117015,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022418834015033957,
      "loss": 5.1227,
      "mean_token_accuracy": 0.23093242943286896,
      "num_tokens": 133612273.0,
      "step": 65935
    },
    {
      "entropy": 5.871256875991821,
      "epoch": 5.762474875469719,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022415798453000424,
      "loss": 5.2129,
      "mean_token_accuracy": 0.21360071152448654,
      "num_tokens": 133622631.0,
      "step": 65940
    },
    {
      "entropy": 5.889992713928223,
      "epoch": 5.762911823822424,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022412762988484536,
      "loss": 5.2067,
      "mean_token_accuracy": 0.21808809489011766,
      "num_tokens": 133632944.0,
      "step": 65945
    },
    {
      "entropy": 5.815762090682983,
      "epoch": 5.763348772175129,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002240972762154453,
      "loss": 5.112,
      "mean_token_accuracy": 0.22839616239070892,
      "num_tokens": 133642677.0,
      "step": 65950
    },
    {
      "entropy": 5.814058876037597,
      "epoch": 5.763785720527833,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022406692352238607,
      "loss": 5.1544,
      "mean_token_accuracy": 0.22142015993595124,
      "num_tokens": 133652628.0,
      "step": 65955
    },
    {
      "entropy": 5.847351741790772,
      "epoch": 5.764222668880539,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022403657180624997,
      "loss": 5.121,
      "mean_token_accuracy": 0.2271469309926033,
      "num_tokens": 133663271.0,
      "step": 65960
    },
    {
      "entropy": 5.866667366027832,
      "epoch": 5.764659617233243,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022400622106761908,
      "loss": 5.1865,
      "mean_token_accuracy": 0.21682780981063843,
      "num_tokens": 133673493.0,
      "step": 65965
    },
    {
      "entropy": 5.886945676803589,
      "epoch": 5.765096565585948,
      "grad_norm": 1.0625,
      "learning_rate": 0.00022397587130707565,
      "loss": 5.1757,
      "mean_token_accuracy": 0.21623408049345016,
      "num_tokens": 133684456.0,
      "step": 65970
    },
    {
      "entropy": 5.867978620529175,
      "epoch": 5.765533513938652,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022394552252520173,
      "loss": 5.2134,
      "mean_token_accuracy": 0.21653225719928743,
      "num_tokens": 133693455.0,
      "step": 65975
    },
    {
      "entropy": 5.877761125564575,
      "epoch": 5.7659704622913575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022391517472257944,
      "loss": 5.1702,
      "mean_token_accuracy": 0.2232026308774948,
      "num_tokens": 133702792.0,
      "step": 65980
    },
    {
      "entropy": 5.848413610458374,
      "epoch": 5.766407410644062,
      "grad_norm": 1.125,
      "learning_rate": 0.00022388482789979086,
      "loss": 5.2298,
      "mean_token_accuracy": 0.22332362681627274,
      "num_tokens": 133712982.0,
      "step": 65985
    },
    {
      "entropy": 5.884835577011108,
      "epoch": 5.766844358996766,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022385448205741805,
      "loss": 5.2425,
      "mean_token_accuracy": 0.21468753814697267,
      "num_tokens": 133723158.0,
      "step": 65990
    },
    {
      "entropy": 5.866214275360107,
      "epoch": 5.767281307349471,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022382413719604306,
      "loss": 5.1738,
      "mean_token_accuracy": 0.2216173380613327,
      "num_tokens": 133732478.0,
      "step": 65995
    },
    {
      "entropy": 5.831392383575439,
      "epoch": 5.767718255702176,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000223793793316248,
      "loss": 5.2057,
      "mean_token_accuracy": 0.21541159003973007,
      "num_tokens": 133743651.0,
      "step": 66000
    },
    {
      "epoch": 5.767718255702176,
      "eval_entropy": 5.58235353982446,
      "eval_loss": 5.283485412597656,
      "eval_mean_token_accuracy": 0.2223907921335237,
      "eval_num_tokens": 133743651.0,
      "eval_runtime": 27.8794,
      "eval_samples_per_second": 1327.647,
      "eval_steps_per_second": 165.965,
      "step": 66000
    },
    {
      "entropy": 5.864885044097901,
      "epoch": 5.768155204054881,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022376345041861474,
      "loss": 5.1418,
      "mean_token_accuracy": 0.22184867560863494,
      "num_tokens": 133754533.0,
      "step": 66005
    },
    {
      "entropy": 5.952494764328003,
      "epoch": 5.768592152407585,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022373310850372542,
      "loss": 5.2419,
      "mean_token_accuracy": 0.2132766991853714,
      "num_tokens": 133764569.0,
      "step": 66010
    },
    {
      "entropy": 5.900409412384033,
      "epoch": 5.7690291007602905,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022370276757216186,
      "loss": 5.2493,
      "mean_token_accuracy": 0.21112211644649506,
      "num_tokens": 133774822.0,
      "step": 66015
    },
    {
      "entropy": 5.863125658035278,
      "epoch": 5.769466049112995,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022367242762450613,
      "loss": 5.2111,
      "mean_token_accuracy": 0.21880098432302475,
      "num_tokens": 133784499.0,
      "step": 66020
    },
    {
      "entropy": 5.819790840148926,
      "epoch": 5.7699029974656995,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022364208866134023,
      "loss": 5.2088,
      "mean_token_accuracy": 0.22131007462739943,
      "num_tokens": 133794548.0,
      "step": 66025
    },
    {
      "entropy": 5.846718835830688,
      "epoch": 5.770339945818404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022361175068324584,
      "loss": 5.1226,
      "mean_token_accuracy": 0.22615820467472075,
      "num_tokens": 133803971.0,
      "step": 66030
    },
    {
      "entropy": 5.895432090759277,
      "epoch": 5.770776894171109,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002235814136908051,
      "loss": 5.2329,
      "mean_token_accuracy": 0.22349630445241928,
      "num_tokens": 133814012.0,
      "step": 66035
    },
    {
      "entropy": 5.826703023910523,
      "epoch": 5.771213842523814,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002235510776845997,
      "loss": 5.1159,
      "mean_token_accuracy": 0.22522959858179092,
      "num_tokens": 133823929.0,
      "step": 66040
    },
    {
      "entropy": 5.8080442428588865,
      "epoch": 5.771650790876518,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022352074266521166,
      "loss": 5.0777,
      "mean_token_accuracy": 0.2357552707195282,
      "num_tokens": 133834114.0,
      "step": 66045
    },
    {
      "entropy": 5.887954664230347,
      "epoch": 5.772087739229223,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022349040863322274,
      "loss": 5.1795,
      "mean_token_accuracy": 0.22373500764369963,
      "num_tokens": 133844165.0,
      "step": 66050
    },
    {
      "entropy": 5.892343044281006,
      "epoch": 5.772524687581928,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022346007558921477,
      "loss": 5.2148,
      "mean_token_accuracy": 0.22065313756465912,
      "num_tokens": 133854368.0,
      "step": 66055
    },
    {
      "entropy": 5.854540824890137,
      "epoch": 5.7729616359346325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002234297435337695,
      "loss": 5.2253,
      "mean_token_accuracy": 0.21190689355134965,
      "num_tokens": 133865372.0,
      "step": 66060
    },
    {
      "entropy": 5.856259870529175,
      "epoch": 5.773398584287337,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002233994124674688,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2128212943673134,
      "num_tokens": 133875579.0,
      "step": 66065
    },
    {
      "entropy": 5.877905082702637,
      "epoch": 5.773835532640042,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002233690823908945,
      "loss": 5.1837,
      "mean_token_accuracy": 0.22304349392652512,
      "num_tokens": 133886019.0,
      "step": 66070
    },
    {
      "entropy": 5.834945297241211,
      "epoch": 5.774272480992747,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022333875330462816,
      "loss": 5.158,
      "mean_token_accuracy": 0.2250032588839531,
      "num_tokens": 133895371.0,
      "step": 66075
    },
    {
      "entropy": 5.787657403945923,
      "epoch": 5.774709429345451,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002233084252092516,
      "loss": 5.1623,
      "mean_token_accuracy": 0.22570869326591492,
      "num_tokens": 133905270.0,
      "step": 66080
    },
    {
      "entropy": 5.885726118087769,
      "epoch": 5.775146377698156,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022327809810534656,
      "loss": 5.229,
      "mean_token_accuracy": 0.22025462687015535,
      "num_tokens": 133915506.0,
      "step": 66085
    },
    {
      "entropy": 5.925701236724853,
      "epoch": 5.775583326050861,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022324777199349467,
      "loss": 5.2422,
      "mean_token_accuracy": 0.21771016716957092,
      "num_tokens": 133925797.0,
      "step": 66090
    },
    {
      "entropy": 5.869666910171508,
      "epoch": 5.776020274403566,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002232174468742777,
      "loss": 5.1935,
      "mean_token_accuracy": 0.22766295522451402,
      "num_tokens": 133935065.0,
      "step": 66095
    },
    {
      "entropy": 5.89008264541626,
      "epoch": 5.77645722275627,
      "grad_norm": 1.125,
      "learning_rate": 0.00022318712274827714,
      "loss": 5.2443,
      "mean_token_accuracy": 0.21735630184412003,
      "num_tokens": 133945943.0,
      "step": 66100
    },
    {
      "entropy": 5.87345290184021,
      "epoch": 5.7768941711089745,
      "grad_norm": 1.125,
      "learning_rate": 0.00022315679961607476,
      "loss": 5.2874,
      "mean_token_accuracy": 0.2060530036687851,
      "num_tokens": 133957287.0,
      "step": 66105
    },
    {
      "entropy": 5.859213018417359,
      "epoch": 5.77733111946168,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002231264774782521,
      "loss": 5.1578,
      "mean_token_accuracy": 0.22344274818897247,
      "num_tokens": 133967048.0,
      "step": 66110
    },
    {
      "entropy": 5.869256114959716,
      "epoch": 5.777768067814384,
      "grad_norm": 1.125,
      "learning_rate": 0.00022309615633539083,
      "loss": 5.1822,
      "mean_token_accuracy": 0.22545687556266786,
      "num_tokens": 133976667.0,
      "step": 66115
    },
    {
      "entropy": 5.87899079322815,
      "epoch": 5.778205016167089,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022306583618807247,
      "loss": 5.1042,
      "mean_token_accuracy": 0.2282773733139038,
      "num_tokens": 133985476.0,
      "step": 66120
    },
    {
      "entropy": 5.864523839950562,
      "epoch": 5.778641964519794,
      "grad_norm": 1.03125,
      "learning_rate": 0.00022303551703687857,
      "loss": 5.2062,
      "mean_token_accuracy": 0.214934803545475,
      "num_tokens": 133996079.0,
      "step": 66125
    },
    {
      "entropy": 5.903233242034912,
      "epoch": 5.779078912872499,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022300519888239067,
      "loss": 5.217,
      "mean_token_accuracy": 0.2132893204689026,
      "num_tokens": 134006399.0,
      "step": 66130
    },
    {
      "entropy": 5.904192781448364,
      "epoch": 5.779515861225203,
      "grad_norm": 1.125,
      "learning_rate": 0.0002229748817251903,
      "loss": 5.2365,
      "mean_token_accuracy": 0.21325511187314988,
      "num_tokens": 134016799.0,
      "step": 66135
    },
    {
      "entropy": 5.81442699432373,
      "epoch": 5.779952809577908,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022294456556585897,
      "loss": 5.1298,
      "mean_token_accuracy": 0.22397446781396865,
      "num_tokens": 134027194.0,
      "step": 66140
    },
    {
      "entropy": 5.823262739181518,
      "epoch": 5.780389757930613,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002229142504049782,
      "loss": 5.0982,
      "mean_token_accuracy": 0.22927106022834778,
      "num_tokens": 134037950.0,
      "step": 66145
    },
    {
      "entropy": 5.954899835586548,
      "epoch": 5.780826706283317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022288393624312936,
      "loss": 5.2637,
      "mean_token_accuracy": 0.21892748773097992,
      "num_tokens": 134048725.0,
      "step": 66150
    },
    {
      "entropy": 5.8040934085845945,
      "epoch": 5.781263654636022,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022285362308089392,
      "loss": 5.1363,
      "mean_token_accuracy": 0.2258239597082138,
      "num_tokens": 134058747.0,
      "step": 66155
    },
    {
      "entropy": 5.813416576385498,
      "epoch": 5.781700602988726,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022282331091885333,
      "loss": 5.1826,
      "mean_token_accuracy": 0.2179713562130928,
      "num_tokens": 134068055.0,
      "step": 66160
    },
    {
      "entropy": 5.800970268249512,
      "epoch": 5.782137551341432,
      "grad_norm": 1.046875,
      "learning_rate": 0.00022279299975758892,
      "loss": 5.1896,
      "mean_token_accuracy": 0.2258324220776558,
      "num_tokens": 134079347.0,
      "step": 66165
    },
    {
      "entropy": 5.857674074172974,
      "epoch": 5.782574499694136,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002227626895976822,
      "loss": 5.1713,
      "mean_token_accuracy": 0.21985072493553162,
      "num_tokens": 134090045.0,
      "step": 66170
    },
    {
      "entropy": 5.839987802505493,
      "epoch": 5.783011448046841,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022273238043971443,
      "loss": 5.1704,
      "mean_token_accuracy": 0.22099724113941194,
      "num_tokens": 134099826.0,
      "step": 66175
    },
    {
      "entropy": 5.8395617485046385,
      "epoch": 5.783448396399546,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022270207228426703,
      "loss": 5.2418,
      "mean_token_accuracy": 0.20737125873565673,
      "num_tokens": 134110393.0,
      "step": 66180
    },
    {
      "entropy": 5.807950782775879,
      "epoch": 5.7838853447522505,
      "grad_norm": 1.25,
      "learning_rate": 0.00022267176513192122,
      "loss": 5.1573,
      "mean_token_accuracy": 0.22156886309385299,
      "num_tokens": 134120156.0,
      "step": 66185
    },
    {
      "entropy": 5.894956159591675,
      "epoch": 5.784322293104955,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00022264145898325845,
      "loss": 5.2075,
      "mean_token_accuracy": 0.21467987149953843,
      "num_tokens": 134130144.0,
      "step": 66190
    },
    {
      "entropy": 5.889321947097779,
      "epoch": 5.784759241457659,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002226111538388599,
      "loss": 5.1625,
      "mean_token_accuracy": 0.21628503054380416,
      "num_tokens": 134139924.0,
      "step": 66195
    },
    {
      "entropy": 5.9098536491394045,
      "epoch": 5.785196189810365,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022258084969930687,
      "loss": 5.293,
      "mean_token_accuracy": 0.21028897315263748,
      "num_tokens": 134149953.0,
      "step": 66200
    },
    {
      "entropy": 5.864300537109375,
      "epoch": 5.785633138163069,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022255054656518058,
      "loss": 5.1381,
      "mean_token_accuracy": 0.2272540733218193,
      "num_tokens": 134160251.0,
      "step": 66205
    },
    {
      "entropy": 5.891451406478882,
      "epoch": 5.786070086515774,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002225202444370623,
      "loss": 5.1812,
      "mean_token_accuracy": 0.21641992777585983,
      "num_tokens": 134170652.0,
      "step": 66210
    },
    {
      "entropy": 5.856587886810303,
      "epoch": 5.786507034868478,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002224899433155333,
      "loss": 5.1503,
      "mean_token_accuracy": 0.2304309144616127,
      "num_tokens": 134182260.0,
      "step": 66215
    },
    {
      "entropy": 5.8471709251403805,
      "epoch": 5.7869439832211835,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002224596432011747,
      "loss": 5.1485,
      "mean_token_accuracy": 0.22100187540054322,
      "num_tokens": 134191684.0,
      "step": 66220
    },
    {
      "entropy": 5.837886714935303,
      "epoch": 5.787380931573888,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022242934409456754,
      "loss": 5.1754,
      "mean_token_accuracy": 0.22136377394199372,
      "num_tokens": 134201329.0,
      "step": 66225
    },
    {
      "entropy": 5.85833511352539,
      "epoch": 5.7878178799265925,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022239904599629322,
      "loss": 5.205,
      "mean_token_accuracy": 0.21719232499599456,
      "num_tokens": 134211511.0,
      "step": 66230
    },
    {
      "entropy": 5.862509298324585,
      "epoch": 5.788254828279298,
      "grad_norm": 1.125,
      "learning_rate": 0.0002223687489069327,
      "loss": 5.2368,
      "mean_token_accuracy": 0.20965474396944045,
      "num_tokens": 134222442.0,
      "step": 66235
    },
    {
      "entropy": 5.937197303771972,
      "epoch": 5.788691776632002,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022233845282706722,
      "loss": 5.2587,
      "mean_token_accuracy": 0.2064770996570587,
      "num_tokens": 134232848.0,
      "step": 66240
    },
    {
      "entropy": 5.891340112686157,
      "epoch": 5.789128724984707,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002223081577572778,
      "loss": 5.1862,
      "mean_token_accuracy": 0.2219362735748291,
      "num_tokens": 134242537.0,
      "step": 66245
    },
    {
      "entropy": 5.886201000213623,
      "epoch": 5.789565673337411,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022227786369814545,
      "loss": 5.1892,
      "mean_token_accuracy": 0.22064715325832368,
      "num_tokens": 134252093.0,
      "step": 66250
    },
    {
      "entropy": 5.858938312530517,
      "epoch": 5.790002621690117,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002222475706502513,
      "loss": 5.206,
      "mean_token_accuracy": 0.21994608938694,
      "num_tokens": 134261949.0,
      "step": 66255
    },
    {
      "entropy": 5.825273132324218,
      "epoch": 5.790439570042821,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002222172786141764,
      "loss": 5.1696,
      "mean_token_accuracy": 0.22459787726402283,
      "num_tokens": 134272132.0,
      "step": 66260
    },
    {
      "entropy": 5.848517847061157,
      "epoch": 5.7908765183955255,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002221869875905018,
      "loss": 5.1743,
      "mean_token_accuracy": 0.21797502040863037,
      "num_tokens": 134281556.0,
      "step": 66265
    },
    {
      "entropy": 5.8868021965026855,
      "epoch": 5.79131346674823,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022215669757980844,
      "loss": 5.209,
      "mean_token_accuracy": 0.21462781876325607,
      "num_tokens": 134290941.0,
      "step": 66270
    },
    {
      "entropy": 5.8712695121765135,
      "epoch": 5.791750415100935,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002221264085826773,
      "loss": 5.1724,
      "mean_token_accuracy": 0.22148035764694213,
      "num_tokens": 134300296.0,
      "step": 66275
    },
    {
      "entropy": 5.8298111915588375,
      "epoch": 5.79218736345364,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002220961205996894,
      "loss": 5.169,
      "mean_token_accuracy": 0.22321338802576066,
      "num_tokens": 134310381.0,
      "step": 66280
    },
    {
      "entropy": 5.855704832077026,
      "epoch": 5.792624311806344,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022206583363142558,
      "loss": 5.2423,
      "mean_token_accuracy": 0.2136362001299858,
      "num_tokens": 134320024.0,
      "step": 66285
    },
    {
      "entropy": 5.8864326000213625,
      "epoch": 5.79306126015905,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022203554767846685,
      "loss": 5.1676,
      "mean_token_accuracy": 0.21985329389572145,
      "num_tokens": 134328510.0,
      "step": 66290
    },
    {
      "entropy": 5.928340053558349,
      "epoch": 5.793498208511754,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000222005262741394,
      "loss": 5.2441,
      "mean_token_accuracy": 0.2099695771932602,
      "num_tokens": 134339032.0,
      "step": 66295
    },
    {
      "entropy": 5.859893846511841,
      "epoch": 5.793935156864459,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022197497882078799,
      "loss": 5.1452,
      "mean_token_accuracy": 0.21957810074090958,
      "num_tokens": 134347654.0,
      "step": 66300
    },
    {
      "entropy": 5.960163736343384,
      "epoch": 5.794372105217163,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022194469591722976,
      "loss": 5.3128,
      "mean_token_accuracy": 0.20651943385601043,
      "num_tokens": 134357381.0,
      "step": 66305
    },
    {
      "entropy": 5.826956701278687,
      "epoch": 5.794809053569868,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022191441403129997,
      "loss": 5.1137,
      "mean_token_accuracy": 0.23096192628145218,
      "num_tokens": 134367438.0,
      "step": 66310
    },
    {
      "entropy": 5.878236722946167,
      "epoch": 5.795246001922573,
      "grad_norm": 1.125,
      "learning_rate": 0.00022188413316357952,
      "loss": 5.2329,
      "mean_token_accuracy": 0.21054086983203887,
      "num_tokens": 134377498.0,
      "step": 66315
    },
    {
      "entropy": 5.796097946166992,
      "epoch": 5.795682950275277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002218538533146493,
      "loss": 5.1505,
      "mean_token_accuracy": 0.22170721739530563,
      "num_tokens": 134386770.0,
      "step": 66320
    },
    {
      "entropy": 5.8151023387908936,
      "epoch": 5.796119898627982,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022182357448508999,
      "loss": 5.1461,
      "mean_token_accuracy": 0.22333384305238724,
      "num_tokens": 134396253.0,
      "step": 66325
    },
    {
      "entropy": 5.8061151027679445,
      "epoch": 5.796556846980687,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00022179329667548243,
      "loss": 5.1392,
      "mean_token_accuracy": 0.22362336665391921,
      "num_tokens": 134406824.0,
      "step": 66330
    },
    {
      "entropy": 5.813995361328125,
      "epoch": 5.796993795333392,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022176301988640723,
      "loss": 5.151,
      "mean_token_accuracy": 0.2247452199459076,
      "num_tokens": 134417348.0,
      "step": 66335
    },
    {
      "entropy": 5.8672503471374515,
      "epoch": 5.797430743686096,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022173274411844529,
      "loss": 5.2489,
      "mean_token_accuracy": 0.21562246531248092,
      "num_tokens": 134427812.0,
      "step": 66340
    },
    {
      "entropy": 5.812258195877075,
      "epoch": 5.7978676920388015,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022170246937217714,
      "loss": 5.0872,
      "mean_token_accuracy": 0.22979280203580857,
      "num_tokens": 134438145.0,
      "step": 66345
    },
    {
      "entropy": 5.895279598236084,
      "epoch": 5.798304640391506,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002216721956481837,
      "loss": 5.2098,
      "mean_token_accuracy": 0.21511407047510148,
      "num_tokens": 134447930.0,
      "step": 66350
    },
    {
      "entropy": 5.869379997253418,
      "epoch": 5.79874158874421,
      "grad_norm": 1.125,
      "learning_rate": 0.00022164192294704538,
      "loss": 5.2232,
      "mean_token_accuracy": 0.21058902591466905,
      "num_tokens": 134457350.0,
      "step": 66355
    },
    {
      "entropy": 5.946941137313843,
      "epoch": 5.799178537096915,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002216116512693429,
      "loss": 5.2785,
      "mean_token_accuracy": 0.2158254787325859,
      "num_tokens": 134468259.0,
      "step": 66360
    },
    {
      "entropy": 5.847870922088623,
      "epoch": 5.799615485449619,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022158138061565696,
      "loss": 5.1664,
      "mean_token_accuracy": 0.22236492186784745,
      "num_tokens": 134478939.0,
      "step": 66365
    },
    {
      "entropy": 5.832521295547485,
      "epoch": 5.800052433802325,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022155111098656817,
      "loss": 5.1813,
      "mean_token_accuracy": 0.22380824089050294,
      "num_tokens": 134489512.0,
      "step": 66370
    },
    {
      "entropy": 5.838075923919678,
      "epoch": 5.800489382155029,
      "grad_norm": 1.125,
      "learning_rate": 0.00022152084238265713,
      "loss": 5.1205,
      "mean_token_accuracy": 0.22437483221292495,
      "num_tokens": 134500231.0,
      "step": 66375
    },
    {
      "entropy": 5.841284132003784,
      "epoch": 5.800926330507734,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022149057480450423,
      "loss": 5.1965,
      "mean_token_accuracy": 0.21883343756198884,
      "num_tokens": 134510583.0,
      "step": 66380
    },
    {
      "entropy": 5.87030611038208,
      "epoch": 5.801363278860439,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002214603082526902,
      "loss": 5.2579,
      "mean_token_accuracy": 0.22052256762981415,
      "num_tokens": 134520626.0,
      "step": 66385
    },
    {
      "entropy": 5.84719557762146,
      "epoch": 5.8018002272131435,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022143004272779548,
      "loss": 5.1069,
      "mean_token_accuracy": 0.22615054994821548,
      "num_tokens": 134530413.0,
      "step": 66390
    },
    {
      "entropy": 5.899518299102783,
      "epoch": 5.802237175565848,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022139977823040064,
      "loss": 5.2545,
      "mean_token_accuracy": 0.21529655456542968,
      "num_tokens": 134540995.0,
      "step": 66395
    },
    {
      "entropy": 5.782318115234375,
      "epoch": 5.802674123918553,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022136951476108617,
      "loss": 5.0439,
      "mean_token_accuracy": 0.2284384027123451,
      "num_tokens": 134550041.0,
      "step": 66400
    },
    {
      "entropy": 5.8127296447753904,
      "epoch": 5.803111072271258,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002213392523204325,
      "loss": 5.185,
      "mean_token_accuracy": 0.21802657395601271,
      "num_tokens": 134559549.0,
      "step": 66405
    },
    {
      "entropy": 5.7608099460601805,
      "epoch": 5.803548020623962,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022130899090902002,
      "loss": 5.1382,
      "mean_token_accuracy": 0.224787175655365,
      "num_tokens": 134569056.0,
      "step": 66410
    },
    {
      "entropy": 5.8262331008911135,
      "epoch": 5.803984968976667,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002212787305274293,
      "loss": 5.1649,
      "mean_token_accuracy": 0.2188044771552086,
      "num_tokens": 134579007.0,
      "step": 66415
    },
    {
      "entropy": 5.878269290924072,
      "epoch": 5.804421917329371,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022124847117624064,
      "loss": 5.1718,
      "mean_token_accuracy": 0.21810310035943986,
      "num_tokens": 134588815.0,
      "step": 66420
    },
    {
      "entropy": 5.870835542678833,
      "epoch": 5.8048588656820765,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002212182128560345,
      "loss": 5.1904,
      "mean_token_accuracy": 0.2139526292681694,
      "num_tokens": 134599597.0,
      "step": 66425
    },
    {
      "entropy": 5.833804559707642,
      "epoch": 5.805295814034781,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002211879555673912,
      "loss": 5.1258,
      "mean_token_accuracy": 0.22427267730236053,
      "num_tokens": 134609721.0,
      "step": 66430
    },
    {
      "entropy": 5.8337634086608885,
      "epoch": 5.8057327623874855,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022115769931089108,
      "loss": 5.1135,
      "mean_token_accuracy": 0.22616951912641525,
      "num_tokens": 134619754.0,
      "step": 66435
    },
    {
      "entropy": 5.813305330276489,
      "epoch": 5.806169710740191,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00022112744408711455,
      "loss": 5.1253,
      "mean_token_accuracy": 0.2264800861477852,
      "num_tokens": 134630832.0,
      "step": 66440
    },
    {
      "entropy": 5.8348020076751705,
      "epoch": 5.806606659092895,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002210971898966418,
      "loss": 5.1554,
      "mean_token_accuracy": 0.22446725219488145,
      "num_tokens": 134640496.0,
      "step": 66445
    },
    {
      "entropy": 5.866026115417481,
      "epoch": 5.8070436074456,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002210669367400533,
      "loss": 5.2527,
      "mean_token_accuracy": 0.21608901619911194,
      "num_tokens": 134650771.0,
      "step": 66450
    },
    {
      "entropy": 5.851635885238648,
      "epoch": 5.807480555798305,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022103668461792914,
      "loss": 5.1783,
      "mean_token_accuracy": 0.22122657895088196,
      "num_tokens": 134661143.0,
      "step": 66455
    },
    {
      "entropy": 5.804690027236939,
      "epoch": 5.80791750415101,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022100643353084964,
      "loss": 5.1309,
      "mean_token_accuracy": 0.2253286749124527,
      "num_tokens": 134671198.0,
      "step": 66460
    },
    {
      "entropy": 5.86027684211731,
      "epoch": 5.808354452503714,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022097618347939507,
      "loss": 5.1299,
      "mean_token_accuracy": 0.2290608137845993,
      "num_tokens": 134682000.0,
      "step": 66465
    },
    {
      "entropy": 5.9431263446807865,
      "epoch": 5.8087914008564185,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022094593446414558,
      "loss": 5.2796,
      "mean_token_accuracy": 0.21564877182245254,
      "num_tokens": 134691251.0,
      "step": 66470
    },
    {
      "entropy": 5.9321465492248535,
      "epoch": 5.809228349209123,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022091568648568145,
      "loss": 5.2039,
      "mean_token_accuracy": 0.21521949023008347,
      "num_tokens": 134700782.0,
      "step": 66475
    },
    {
      "entropy": 5.79865665435791,
      "epoch": 5.809665297561828,
      "grad_norm": 1.125,
      "learning_rate": 0.00022088543954458273,
      "loss": 5.083,
      "mean_token_accuracy": 0.22777874171733856,
      "num_tokens": 134709668.0,
      "step": 66480
    },
    {
      "entropy": 5.857082366943359,
      "epoch": 5.810102245914533,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022085519364142963,
      "loss": 5.2389,
      "mean_token_accuracy": 0.21074821650981904,
      "num_tokens": 134719424.0,
      "step": 66485
    },
    {
      "entropy": 5.8726067543029785,
      "epoch": 5.810539194267237,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022082494877680232,
      "loss": 5.1798,
      "mean_token_accuracy": 0.21702405959367752,
      "num_tokens": 134729163.0,
      "step": 66490
    },
    {
      "entropy": 5.929634094238281,
      "epoch": 5.810976142619943,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002207947049512809,
      "loss": 5.285,
      "mean_token_accuracy": 0.20381457656621932,
      "num_tokens": 134738977.0,
      "step": 66495
    },
    {
      "entropy": 5.950778818130493,
      "epoch": 5.811413090972647,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022076446216544543,
      "loss": 5.2945,
      "mean_token_accuracy": 0.21412632167339324,
      "num_tokens": 134749193.0,
      "step": 66500
    },
    {
      "entropy": 5.870892524719238,
      "epoch": 5.811850039325352,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022073422041987595,
      "loss": 5.2823,
      "mean_token_accuracy": 0.21813719272613524,
      "num_tokens": 134760395.0,
      "step": 66505
    },
    {
      "entropy": 5.908199262619019,
      "epoch": 5.812286987678056,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022070397971515256,
      "loss": 5.1891,
      "mean_token_accuracy": 0.21361069679260253,
      "num_tokens": 134770294.0,
      "step": 66510
    },
    {
      "entropy": 5.924416446685791,
      "epoch": 5.812723936030761,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022067374005185536,
      "loss": 5.2639,
      "mean_token_accuracy": 0.2181227833032608,
      "num_tokens": 134781665.0,
      "step": 66515
    },
    {
      "entropy": 5.860230541229248,
      "epoch": 5.813160884383466,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022064350143056427,
      "loss": 5.0969,
      "mean_token_accuracy": 0.22801635265350342,
      "num_tokens": 134791562.0,
      "step": 66520
    },
    {
      "entropy": 5.866259479522705,
      "epoch": 5.81359783273617,
      "grad_norm": 1.125,
      "learning_rate": 0.00022061326385185926,
      "loss": 5.1929,
      "mean_token_accuracy": 0.22107226997613907,
      "num_tokens": 134801628.0,
      "step": 66525
    },
    {
      "entropy": 5.847078609466553,
      "epoch": 5.814034781088875,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022058302731632035,
      "loss": 5.2048,
      "mean_token_accuracy": 0.2206421136856079,
      "num_tokens": 134812685.0,
      "step": 66530
    },
    {
      "entropy": 5.892379665374756,
      "epoch": 5.81447172944158,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022055279182452752,
      "loss": 5.2645,
      "mean_token_accuracy": 0.20819865316152572,
      "num_tokens": 134822722.0,
      "step": 66535
    },
    {
      "entropy": 5.867629098892212,
      "epoch": 5.814908677794285,
      "grad_norm": 1.125,
      "learning_rate": 0.00022052255737706067,
      "loss": 5.1627,
      "mean_token_accuracy": 0.22177413254976272,
      "num_tokens": 134833230.0,
      "step": 66540
    },
    {
      "entropy": 5.8899373531341555,
      "epoch": 5.815345626146989,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022049232397449976,
      "loss": 5.1921,
      "mean_token_accuracy": 0.21904219388961793,
      "num_tokens": 134842844.0,
      "step": 66545
    },
    {
      "entropy": 5.926973342895508,
      "epoch": 5.8157825744996945,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022046209161742452,
      "loss": 5.236,
      "mean_token_accuracy": 0.21506583839654922,
      "num_tokens": 134853132.0,
      "step": 66550
    },
    {
      "entropy": 5.885177850723267,
      "epoch": 5.816219522852399,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022043186030641505,
      "loss": 5.1953,
      "mean_token_accuracy": 0.2180112525820732,
      "num_tokens": 134863690.0,
      "step": 66555
    },
    {
      "entropy": 5.8532459259033205,
      "epoch": 5.816656471205103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022040163004205098,
      "loss": 5.1835,
      "mean_token_accuracy": 0.2115626335144043,
      "num_tokens": 134875209.0,
      "step": 66560
    },
    {
      "entropy": 5.854822111129761,
      "epoch": 5.817093419557808,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022037140082491237,
      "loss": 5.1431,
      "mean_token_accuracy": 0.21463313102722167,
      "num_tokens": 134885031.0,
      "step": 66565
    },
    {
      "entropy": 5.806766223907471,
      "epoch": 5.817530367910513,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002203411726555789,
      "loss": 5.2224,
      "mean_token_accuracy": 0.21454026103019713,
      "num_tokens": 134894620.0,
      "step": 66570
    },
    {
      "entropy": 5.895751571655273,
      "epoch": 5.817967316263218,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002203109455346303,
      "loss": 5.1306,
      "mean_token_accuracy": 0.22149472683668137,
      "num_tokens": 134904937.0,
      "step": 66575
    },
    {
      "entropy": 5.92750506401062,
      "epoch": 5.818404264615922,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002202807194626465,
      "loss": 5.2523,
      "mean_token_accuracy": 0.21462429016828538,
      "num_tokens": 134915807.0,
      "step": 66580
    },
    {
      "entropy": 5.918323373794555,
      "epoch": 5.818841212968627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002202504944402071,
      "loss": 5.2588,
      "mean_token_accuracy": 0.22216661423444747,
      "num_tokens": 134925433.0,
      "step": 66585
    },
    {
      "entropy": 5.85350546836853,
      "epoch": 5.819278161321332,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000220220270467892,
      "loss": 5.1903,
      "mean_token_accuracy": 0.21920850723981858,
      "num_tokens": 134936634.0,
      "step": 66590
    },
    {
      "entropy": 5.877127552032471,
      "epoch": 5.8197151096740365,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00022019004754628081,
      "loss": 5.2212,
      "mean_token_accuracy": 0.22087670117616653,
      "num_tokens": 134948158.0,
      "step": 66595
    },
    {
      "entropy": 5.857545804977417,
      "epoch": 5.820152058026741,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022015982567595316,
      "loss": 5.1518,
      "mean_token_accuracy": 0.22094322443008424,
      "num_tokens": 134957693.0,
      "step": 66600
    },
    {
      "entropy": 5.793030738830566,
      "epoch": 5.820589006379446,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022012960485748878,
      "loss": 5.0899,
      "mean_token_accuracy": 0.22689830064773558,
      "num_tokens": 134967217.0,
      "step": 66605
    },
    {
      "entropy": 5.845946264266968,
      "epoch": 5.821025954732151,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002200993850914673,
      "loss": 5.1605,
      "mean_token_accuracy": 0.22306651175022124,
      "num_tokens": 134976237.0,
      "step": 66610
    },
    {
      "entropy": 5.799192667007446,
      "epoch": 5.821462903084855,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022006916637846847,
      "loss": 5.1945,
      "mean_token_accuracy": 0.21789292842149735,
      "num_tokens": 134985120.0,
      "step": 66615
    },
    {
      "entropy": 5.887028312683105,
      "epoch": 5.82189985143756,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022003894871907176,
      "loss": 5.1851,
      "mean_token_accuracy": 0.212926284968853,
      "num_tokens": 134995824.0,
      "step": 66620
    },
    {
      "entropy": 5.877435636520386,
      "epoch": 5.822336799790265,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022000873211385687,
      "loss": 5.2379,
      "mean_token_accuracy": 0.2217693731188774,
      "num_tokens": 135005754.0,
      "step": 66625
    },
    {
      "entropy": 5.813489198684692,
      "epoch": 5.8227737481429696,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021997851656340329,
      "loss": 5.1327,
      "mean_token_accuracy": 0.23425643593072892,
      "num_tokens": 135015915.0,
      "step": 66630
    },
    {
      "entropy": 5.87489595413208,
      "epoch": 5.823210696495674,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021994830206829058,
      "loss": 5.1913,
      "mean_token_accuracy": 0.22608540952205658,
      "num_tokens": 135025905.0,
      "step": 66635
    },
    {
      "entropy": 5.910023307800293,
      "epoch": 5.8236476448483785,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002199180886290983,
      "loss": 5.19,
      "mean_token_accuracy": 0.2195936843752861,
      "num_tokens": 135034601.0,
      "step": 66640
    },
    {
      "entropy": 5.794791412353516,
      "epoch": 5.824084593201084,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00021988787624640588,
      "loss": 5.1105,
      "mean_token_accuracy": 0.22964905500411986,
      "num_tokens": 135045489.0,
      "step": 66645
    },
    {
      "entropy": 5.8420069217681885,
      "epoch": 5.824521541553788,
      "grad_norm": 1.0625,
      "learning_rate": 0.00021985766492079296,
      "loss": 5.2561,
      "mean_token_accuracy": 0.21727876514196395,
      "num_tokens": 135056889.0,
      "step": 66650
    },
    {
      "entropy": 5.825519275665283,
      "epoch": 5.824958489906493,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021982745465283886,
      "loss": 5.174,
      "mean_token_accuracy": 0.2212582692503929,
      "num_tokens": 135066899.0,
      "step": 66655
    },
    {
      "entropy": 5.911979103088379,
      "epoch": 5.825395438259198,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00021979724544312314,
      "loss": 5.2508,
      "mean_token_accuracy": 0.21315400004386903,
      "num_tokens": 135076775.0,
      "step": 66660
    },
    {
      "entropy": 5.896006345748901,
      "epoch": 5.825832386611903,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021976703729222513,
      "loss": 5.147,
      "mean_token_accuracy": 0.22088779211044313,
      "num_tokens": 135086461.0,
      "step": 66665
    },
    {
      "entropy": 5.864131021499634,
      "epoch": 5.826269334964607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021973683020072433,
      "loss": 5.0987,
      "mean_token_accuracy": 0.21468560695648192,
      "num_tokens": 135095793.0,
      "step": 66670
    },
    {
      "entropy": 5.873029708862305,
      "epoch": 5.8267062833173116,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021970662416920006,
      "loss": 5.2418,
      "mean_token_accuracy": 0.21626275181770324,
      "num_tokens": 135106186.0,
      "step": 66675
    },
    {
      "entropy": 5.917072582244873,
      "epoch": 5.827143231670017,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002196764191982317,
      "loss": 5.2304,
      "mean_token_accuracy": 0.21116856932640077,
      "num_tokens": 135115602.0,
      "step": 66680
    },
    {
      "entropy": 5.884652090072632,
      "epoch": 5.827580180022721,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021964621528839862,
      "loss": 5.1685,
      "mean_token_accuracy": 0.21578267365694045,
      "num_tokens": 135126102.0,
      "step": 66685
    },
    {
      "entropy": 5.866587400436401,
      "epoch": 5.828017128375426,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021961601244028018,
      "loss": 5.1582,
      "mean_token_accuracy": 0.22229697704315185,
      "num_tokens": 135135937.0,
      "step": 66690
    },
    {
      "entropy": 5.824381303787232,
      "epoch": 5.82845407672813,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002195858106544556,
      "loss": 5.2172,
      "mean_token_accuracy": 0.21102405339479446,
      "num_tokens": 135146975.0,
      "step": 66695
    },
    {
      "entropy": 5.8215080261230465,
      "epoch": 5.828891025080836,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002195556099315042,
      "loss": 5.2225,
      "mean_token_accuracy": 0.21714119762182235,
      "num_tokens": 135158752.0,
      "step": 66700
    },
    {
      "entropy": 5.842640447616577,
      "epoch": 5.82932797343354,
      "grad_norm": 1.125,
      "learning_rate": 0.0002195254102720053,
      "loss": 5.1301,
      "mean_token_accuracy": 0.21830743700265884,
      "num_tokens": 135168914.0,
      "step": 66705
    },
    {
      "entropy": 5.802412223815918,
      "epoch": 5.829764921786245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021949521167653803,
      "loss": 5.117,
      "mean_token_accuracy": 0.23233585208654403,
      "num_tokens": 135178637.0,
      "step": 66710
    },
    {
      "entropy": 5.894646644592285,
      "epoch": 5.83020187013895,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002194650141456817,
      "loss": 5.2899,
      "mean_token_accuracy": 0.214915032684803,
      "num_tokens": 135188788.0,
      "step": 66715
    },
    {
      "entropy": 5.957788944244385,
      "epoch": 5.830638818491654,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002194348176800155,
      "loss": 5.2425,
      "mean_token_accuracy": 0.21538525819778442,
      "num_tokens": 135199110.0,
      "step": 66720
    },
    {
      "entropy": 5.8883991718292235,
      "epoch": 5.831075766844359,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021940462228011866,
      "loss": 5.1669,
      "mean_token_accuracy": 0.22899804562330245,
      "num_tokens": 135207381.0,
      "step": 66725
    },
    {
      "entropy": 5.8280534744262695,
      "epoch": 5.831512715197063,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021937442794657032,
      "loss": 5.1557,
      "mean_token_accuracy": 0.22309809923171997,
      "num_tokens": 135217589.0,
      "step": 66730
    },
    {
      "entropy": 5.766569471359253,
      "epoch": 5.831949663549769,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021934423467994952,
      "loss": 5.1077,
      "mean_token_accuracy": 0.22871006429195403,
      "num_tokens": 135228566.0,
      "step": 66735
    },
    {
      "entropy": 5.797220516204834,
      "epoch": 5.832386611902473,
      "grad_norm": 1.125,
      "learning_rate": 0.00021931404248083547,
      "loss": 5.1568,
      "mean_token_accuracy": 0.22268159687519073,
      "num_tokens": 135238553.0,
      "step": 66740
    },
    {
      "entropy": 5.894642448425293,
      "epoch": 5.832823560255178,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021928385134980727,
      "loss": 5.2016,
      "mean_token_accuracy": 0.2165094405412674,
      "num_tokens": 135248944.0,
      "step": 66745
    },
    {
      "entropy": 5.900688552856446,
      "epoch": 5.833260508607882,
      "grad_norm": 1.0625,
      "learning_rate": 0.000219253661287444,
      "loss": 5.1963,
      "mean_token_accuracy": 0.21638642698526384,
      "num_tokens": 135259664.0,
      "step": 66750
    },
    {
      "entropy": 5.8755653381347654,
      "epoch": 5.8336974569605875,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021922347229432477,
      "loss": 5.2576,
      "mean_token_accuracy": 0.2082836553454399,
      "num_tokens": 135269986.0,
      "step": 66755
    },
    {
      "entropy": 5.889786624908448,
      "epoch": 5.834134405313292,
      "grad_norm": 1.125,
      "learning_rate": 0.00021919328437102847,
      "loss": 5.1056,
      "mean_token_accuracy": 0.23222008347511292,
      "num_tokens": 135279236.0,
      "step": 66760
    },
    {
      "entropy": 5.826585102081299,
      "epoch": 5.834571353665996,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002191630975181343,
      "loss": 5.0878,
      "mean_token_accuracy": 0.22885318398475646,
      "num_tokens": 135288880.0,
      "step": 66765
    },
    {
      "entropy": 5.864100646972656,
      "epoch": 5.835008302018702,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021913291173622114,
      "loss": 5.2078,
      "mean_token_accuracy": 0.22009155452251433,
      "num_tokens": 135299016.0,
      "step": 66770
    },
    {
      "entropy": 5.812758445739746,
      "epoch": 5.835445250371406,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021910272702586797,
      "loss": 5.1491,
      "mean_token_accuracy": 0.22363606840372086,
      "num_tokens": 135309174.0,
      "step": 66775
    },
    {
      "entropy": 5.920878744125366,
      "epoch": 5.835882198724111,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021907254338765392,
      "loss": 5.2345,
      "mean_token_accuracy": 0.21932964026927948,
      "num_tokens": 135318832.0,
      "step": 66780
    },
    {
      "entropy": 5.885229063034058,
      "epoch": 5.836319147076815,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00021904236082215768,
      "loss": 5.2338,
      "mean_token_accuracy": 0.21614603847265243,
      "num_tokens": 135329144.0,
      "step": 66785
    },
    {
      "entropy": 5.936698722839355,
      "epoch": 5.836756095429521,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021901217932995826,
      "loss": 5.2696,
      "mean_token_accuracy": 0.21268495172262192,
      "num_tokens": 135339143.0,
      "step": 66790
    },
    {
      "entropy": 5.900057697296143,
      "epoch": 5.837193043782225,
      "grad_norm": 1.078125,
      "learning_rate": 0.00021898199891163462,
      "loss": 5.1676,
      "mean_token_accuracy": 0.2151379704475403,
      "num_tokens": 135349373.0,
      "step": 66795
    },
    {
      "entropy": 5.853840923309326,
      "epoch": 5.8376299921349295,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021895181956776562,
      "loss": 5.1733,
      "mean_token_accuracy": 0.21776266843080522,
      "num_tokens": 135358487.0,
      "step": 66800
    },
    {
      "entropy": 5.915089702606201,
      "epoch": 5.838066940487634,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002189216412989301,
      "loss": 5.2383,
      "mean_token_accuracy": 0.2095630332827568,
      "num_tokens": 135368259.0,
      "step": 66805
    },
    {
      "entropy": 5.875944137573242,
      "epoch": 5.838503888840339,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021889146410570683,
      "loss": 5.2463,
      "mean_token_accuracy": 0.21399417966604234,
      "num_tokens": 135378700.0,
      "step": 66810
    },
    {
      "entropy": 5.782761764526367,
      "epoch": 5.838940837193044,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00021886128798867467,
      "loss": 5.1506,
      "mean_token_accuracy": 0.22224618941545488,
      "num_tokens": 135388750.0,
      "step": 66815
    },
    {
      "entropy": 5.897074890136719,
      "epoch": 5.839377785545748,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021883111294841246,
      "loss": 5.2104,
      "mean_token_accuracy": 0.21726977378129958,
      "num_tokens": 135398459.0,
      "step": 66820
    },
    {
      "entropy": 5.872057580947876,
      "epoch": 5.839814733898454,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00021880093898549895,
      "loss": 5.167,
      "mean_token_accuracy": 0.218971648812294,
      "num_tokens": 135408407.0,
      "step": 66825
    },
    {
      "entropy": 5.877472972869873,
      "epoch": 5.840251682251158,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021877076610051283,
      "loss": 5.2184,
      "mean_token_accuracy": 0.2188398227095604,
      "num_tokens": 135419185.0,
      "step": 66830
    },
    {
      "entropy": 5.929921340942383,
      "epoch": 5.840688630603863,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002187405942940329,
      "loss": 5.2699,
      "mean_token_accuracy": 0.21628921031951903,
      "num_tokens": 135429741.0,
      "step": 66835
    },
    {
      "entropy": 5.89922308921814,
      "epoch": 5.841125578956567,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021871042356663784,
      "loss": 5.1647,
      "mean_token_accuracy": 0.2196866288781166,
      "num_tokens": 135440743.0,
      "step": 66840
    },
    {
      "entropy": 5.8520739555358885,
      "epoch": 5.841562527309272,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002186802539189064,
      "loss": 5.2244,
      "mean_token_accuracy": 0.22385917156934737,
      "num_tokens": 135451734.0,
      "step": 66845
    },
    {
      "entropy": 5.903974294662476,
      "epoch": 5.841999475661977,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021865008535141718,
      "loss": 5.3149,
      "mean_token_accuracy": 0.20728872418403627,
      "num_tokens": 135461665.0,
      "step": 66850
    },
    {
      "entropy": 5.872595167160034,
      "epoch": 5.842436424014681,
      "grad_norm": 1.125,
      "learning_rate": 0.00021861991786474883,
      "loss": 5.2014,
      "mean_token_accuracy": 0.22274935096502305,
      "num_tokens": 135471434.0,
      "step": 66855
    },
    {
      "entropy": 5.8776007175445555,
      "epoch": 5.842873372367386,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021858975145948,
      "loss": 5.1696,
      "mean_token_accuracy": 0.22022336721420288,
      "num_tokens": 135480971.0,
      "step": 66860
    },
    {
      "entropy": 5.847512054443359,
      "epoch": 5.843310320720091,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00021855958613618931,
      "loss": 5.1523,
      "mean_token_accuracy": 0.21867692321538926,
      "num_tokens": 135490669.0,
      "step": 66865
    },
    {
      "entropy": 5.888030385971069,
      "epoch": 5.843747269072796,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021852942189545528,
      "loss": 5.2245,
      "mean_token_accuracy": 0.21378173679113388,
      "num_tokens": 135500557.0,
      "step": 66870
    },
    {
      "entropy": 5.794460344314575,
      "epoch": 5.8441842174255,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002184992587378566,
      "loss": 5.1151,
      "mean_token_accuracy": 0.23678739219903946,
      "num_tokens": 135510625.0,
      "step": 66875
    },
    {
      "entropy": 5.889738845825195,
      "epoch": 5.8446211657782055,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021846909666397175,
      "loss": 5.1968,
      "mean_token_accuracy": 0.21926476955413818,
      "num_tokens": 135521215.0,
      "step": 66880
    },
    {
      "entropy": 5.869614839553833,
      "epoch": 5.84505811413091,
      "grad_norm": 1.03125,
      "learning_rate": 0.00021843893567437917,
      "loss": 5.2073,
      "mean_token_accuracy": 0.21389021426439286,
      "num_tokens": 135531060.0,
      "step": 66885
    },
    {
      "entropy": 5.905872011184693,
      "epoch": 5.845495062483614,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002184087757696575,
      "loss": 5.2048,
      "mean_token_accuracy": 0.2210041955113411,
      "num_tokens": 135541386.0,
      "step": 66890
    },
    {
      "entropy": 5.910528373718262,
      "epoch": 5.845932010836319,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021837861695038507,
      "loss": 5.2477,
      "mean_token_accuracy": 0.21372472047805785,
      "num_tokens": 135551996.0,
      "step": 66895
    },
    {
      "entropy": 5.96494312286377,
      "epoch": 5.846368959189024,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002183484592171406,
      "loss": 5.2385,
      "mean_token_accuracy": 0.21667050421237946,
      "num_tokens": 135561863.0,
      "step": 66900
    },
    {
      "entropy": 5.894104719161987,
      "epoch": 5.846805907541729,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021831830257050227,
      "loss": 5.1864,
      "mean_token_accuracy": 0.21889664232730865,
      "num_tokens": 135571934.0,
      "step": 66905
    },
    {
      "entropy": 5.856169843673706,
      "epoch": 5.847242855894433,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002182881470110486,
      "loss": 5.1423,
      "mean_token_accuracy": 0.21621956527233124,
      "num_tokens": 135582214.0,
      "step": 66910
    },
    {
      "entropy": 5.848018646240234,
      "epoch": 5.847679804247138,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021825799253935796,
      "loss": 5.2169,
      "mean_token_accuracy": 0.22045865803956985,
      "num_tokens": 135590934.0,
      "step": 66915
    },
    {
      "entropy": 5.8862368106842045,
      "epoch": 5.848116752599843,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002182278391560088,
      "loss": 5.2514,
      "mean_token_accuracy": 0.2158889129757881,
      "num_tokens": 135601597.0,
      "step": 66920
    },
    {
      "entropy": 5.864450550079345,
      "epoch": 5.8485537009525475,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002181976868615795,
      "loss": 5.2075,
      "mean_token_accuracy": 0.218508343398571,
      "num_tokens": 135612114.0,
      "step": 66925
    },
    {
      "entropy": 5.944826364517212,
      "epoch": 5.848990649305252,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002181675356566482,
      "loss": 5.2425,
      "mean_token_accuracy": 0.22092675268650055,
      "num_tokens": 135621192.0,
      "step": 66930
    },
    {
      "entropy": 5.81259822845459,
      "epoch": 5.849427597657957,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002181373855417934,
      "loss": 5.0905,
      "mean_token_accuracy": 0.22213209122419358,
      "num_tokens": 135630391.0,
      "step": 66935
    },
    {
      "entropy": 5.8868217945098875,
      "epoch": 5.849864546010662,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021810723651759334,
      "loss": 5.2433,
      "mean_token_accuracy": 0.22142367511987687,
      "num_tokens": 135640033.0,
      "step": 66940
    },
    {
      "entropy": 5.77677059173584,
      "epoch": 5.850301494363366,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021807708858462627,
      "loss": 5.0793,
      "mean_token_accuracy": 0.24324673265218735,
      "num_tokens": 135650193.0,
      "step": 66945
    },
    {
      "entropy": 5.864670276641846,
      "epoch": 5.850738442716071,
      "grad_norm": 1.25,
      "learning_rate": 0.00021804694174347057,
      "loss": 5.0998,
      "mean_token_accuracy": 0.23273678570985795,
      "num_tokens": 135659729.0,
      "step": 66950
    },
    {
      "entropy": 5.856444931030273,
      "epoch": 5.851175391068776,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021801679599470425,
      "loss": 5.1678,
      "mean_token_accuracy": 0.2243942677974701,
      "num_tokens": 135670104.0,
      "step": 66955
    },
    {
      "entropy": 5.865113115310669,
      "epoch": 5.8516123394214805,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021798665133890567,
      "loss": 5.2585,
      "mean_token_accuracy": 0.22032210528850554,
      "num_tokens": 135681584.0,
      "step": 66960
    },
    {
      "entropy": 5.863117837905884,
      "epoch": 5.852049287774185,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000217956507776653,
      "loss": 5.1354,
      "mean_token_accuracy": 0.226634281873703,
      "num_tokens": 135692494.0,
      "step": 66965
    },
    {
      "entropy": 5.841217947006226,
      "epoch": 5.8524862361268895,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021792636530852438,
      "loss": 5.1867,
      "mean_token_accuracy": 0.2192397966980934,
      "num_tokens": 135702743.0,
      "step": 66970
    },
    {
      "entropy": 5.822662591934204,
      "epoch": 5.852923184479595,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000217896223935098,
      "loss": 5.1326,
      "mean_token_accuracy": 0.22375082075595856,
      "num_tokens": 135713109.0,
      "step": 66975
    },
    {
      "entropy": 5.925880241394043,
      "epoch": 5.853360132832299,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021786608365695193,
      "loss": 5.212,
      "mean_token_accuracy": 0.21412104815244676,
      "num_tokens": 135722456.0,
      "step": 66980
    },
    {
      "entropy": 5.864151477813721,
      "epoch": 5.853797081185004,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00021783594447466443,
      "loss": 5.178,
      "mean_token_accuracy": 0.22144676595926285,
      "num_tokens": 135732096.0,
      "step": 66985
    },
    {
      "entropy": 5.833246946334839,
      "epoch": 5.854234029537709,
      "grad_norm": 1.125,
      "learning_rate": 0.00021780580638881337,
      "loss": 5.216,
      "mean_token_accuracy": 0.22197404503822327,
      "num_tokens": 135742434.0,
      "step": 66990
    },
    {
      "entropy": 5.843466091156006,
      "epoch": 5.854670977890414,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002177756693999769,
      "loss": 5.1984,
      "mean_token_accuracy": 0.21854297071695328,
      "num_tokens": 135752543.0,
      "step": 66995
    },
    {
      "entropy": 5.856702280044556,
      "epoch": 5.855107926243118,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021774553350873316,
      "loss": 5.1639,
      "mean_token_accuracy": 0.21896408051252364,
      "num_tokens": 135762100.0,
      "step": 67000
    },
    {
      "entropy": 5.929325962066651,
      "epoch": 5.8555448745958225,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021771539871566004,
      "loss": 5.2307,
      "mean_token_accuracy": 0.2137481912970543,
      "num_tokens": 135771443.0,
      "step": 67005
    },
    {
      "entropy": 5.893265581130981,
      "epoch": 5.855981822948528,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00021768526502133562,
      "loss": 5.2196,
      "mean_token_accuracy": 0.2161790430545807,
      "num_tokens": 135782377.0,
      "step": 67010
    },
    {
      "entropy": 5.8526218891143795,
      "epoch": 5.856418771301232,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021765513242633777,
      "loss": 5.1871,
      "mean_token_accuracy": 0.2185458391904831,
      "num_tokens": 135792927.0,
      "step": 67015
    },
    {
      "entropy": 5.7974082946777346,
      "epoch": 5.856855719653937,
      "grad_norm": 1.25,
      "learning_rate": 0.0002176250009312446,
      "loss": 5.1308,
      "mean_token_accuracy": 0.2247852459549904,
      "num_tokens": 135802743.0,
      "step": 67020
    },
    {
      "entropy": 5.865463733673096,
      "epoch": 5.857292668006641,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021759487053663395,
      "loss": 5.2111,
      "mean_token_accuracy": 0.21640412211418153,
      "num_tokens": 135813225.0,
      "step": 67025
    },
    {
      "entropy": 5.868959093093872,
      "epoch": 5.857729616359347,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021756474124308374,
      "loss": 5.1939,
      "mean_token_accuracy": 0.22047173976898193,
      "num_tokens": 135822455.0,
      "step": 67030
    },
    {
      "entropy": 5.875754737854004,
      "epoch": 5.858166564712051,
      "grad_norm": 1.03125,
      "learning_rate": 0.00021753461305117194,
      "loss": 5.2341,
      "mean_token_accuracy": 0.21982866674661636,
      "num_tokens": 135832523.0,
      "step": 67035
    },
    {
      "entropy": 5.878081846237182,
      "epoch": 5.858603513064756,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021750448596147636,
      "loss": 5.1588,
      "mean_token_accuracy": 0.21406962722539902,
      "num_tokens": 135841502.0,
      "step": 67040
    },
    {
      "entropy": 5.870607614517212,
      "epoch": 5.859040461417461,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021747435997457483,
      "loss": 5.1961,
      "mean_token_accuracy": 0.21963448822498322,
      "num_tokens": 135851089.0,
      "step": 67045
    },
    {
      "entropy": 5.831365346908569,
      "epoch": 5.859477409770165,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021744423509104528,
      "loss": 5.1472,
      "mean_token_accuracy": 0.22559956908226014,
      "num_tokens": 135861379.0,
      "step": 67050
    },
    {
      "entropy": 5.908458471298218,
      "epoch": 5.85991435812287,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002174141113114654,
      "loss": 5.2788,
      "mean_token_accuracy": 0.21402962058782576,
      "num_tokens": 135871517.0,
      "step": 67055
    },
    {
      "entropy": 5.896042966842652,
      "epoch": 5.860351306475574,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021738398863641305,
      "loss": 5.2705,
      "mean_token_accuracy": 0.2144300863146782,
      "num_tokens": 135881397.0,
      "step": 67060
    },
    {
      "entropy": 5.879998826980591,
      "epoch": 5.860788254828279,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021735386706646603,
      "loss": 5.1942,
      "mean_token_accuracy": 0.21703180223703383,
      "num_tokens": 135891979.0,
      "step": 67065
    },
    {
      "entropy": 5.8820545196533205,
      "epoch": 5.861225203180984,
      "grad_norm": 1.078125,
      "learning_rate": 0.000217323746602202,
      "loss": 5.2166,
      "mean_token_accuracy": 0.21687691211700438,
      "num_tokens": 135902293.0,
      "step": 67070
    },
    {
      "entropy": 5.873942708969116,
      "epoch": 5.861662151533689,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021729362724419877,
      "loss": 5.2343,
      "mean_token_accuracy": 0.2163105458021164,
      "num_tokens": 135912837.0,
      "step": 67075
    },
    {
      "entropy": 5.846223783493042,
      "epoch": 5.862099099886393,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021726350899303398,
      "loss": 5.1517,
      "mean_token_accuracy": 0.21819928735494615,
      "num_tokens": 135922325.0,
      "step": 67080
    },
    {
      "entropy": 5.918496465682983,
      "epoch": 5.8625360482390985,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021723339184928536,
      "loss": 5.2542,
      "mean_token_accuracy": 0.2096431538462639,
      "num_tokens": 135932411.0,
      "step": 67085
    },
    {
      "entropy": 5.867646360397339,
      "epoch": 5.862972996591803,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021720327581353056,
      "loss": 5.1611,
      "mean_token_accuracy": 0.22289588302373886,
      "num_tokens": 135942369.0,
      "step": 67090
    },
    {
      "entropy": 5.909434175491333,
      "epoch": 5.863409944944507,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021717316088634713,
      "loss": 5.2409,
      "mean_token_accuracy": 0.20890649557113647,
      "num_tokens": 135952543.0,
      "step": 67095
    },
    {
      "entropy": 5.866675758361817,
      "epoch": 5.863846893297213,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002171430470683129,
      "loss": 5.203,
      "mean_token_accuracy": 0.21782141029834748,
      "num_tokens": 135962115.0,
      "step": 67100
    },
    {
      "entropy": 5.882909488677979,
      "epoch": 5.864283841649917,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002171129343600053,
      "loss": 5.2086,
      "mean_token_accuracy": 0.21579495668411255,
      "num_tokens": 135972480.0,
      "step": 67105
    },
    {
      "entropy": 5.879274559020996,
      "epoch": 5.864720790002622,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00021708282276200198,
      "loss": 5.1961,
      "mean_token_accuracy": 0.21395135670900345,
      "num_tokens": 135982456.0,
      "step": 67110
    },
    {
      "entropy": 5.718268346786499,
      "epoch": 5.865157738355326,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021705271227488037,
      "loss": 4.9909,
      "mean_token_accuracy": 0.2409786954522133,
      "num_tokens": 135991977.0,
      "step": 67115
    },
    {
      "entropy": 5.828989839553833,
      "epoch": 5.865594686708031,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021702260289921816,
      "loss": 5.0847,
      "mean_token_accuracy": 0.22734851837158204,
      "num_tokens": 136001816.0,
      "step": 67120
    },
    {
      "entropy": 5.894415378570557,
      "epoch": 5.866031635060736,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002169924946355928,
      "loss": 5.2238,
      "mean_token_accuracy": 0.21873237043619156,
      "num_tokens": 136011834.0,
      "step": 67125
    },
    {
      "entropy": 5.87400336265564,
      "epoch": 5.8664685834134405,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021696238748458182,
      "loss": 5.2756,
      "mean_token_accuracy": 0.21091406047344208,
      "num_tokens": 136022209.0,
      "step": 67130
    },
    {
      "entropy": 5.914291429519653,
      "epoch": 5.866905531766145,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00021693228144676263,
      "loss": 5.2351,
      "mean_token_accuracy": 0.2160167410969734,
      "num_tokens": 136032914.0,
      "step": 67135
    },
    {
      "entropy": 5.898163223266602,
      "epoch": 5.86734248011885,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021690217652271273,
      "loss": 5.1117,
      "mean_token_accuracy": 0.21947490572929382,
      "num_tokens": 136042317.0,
      "step": 67140
    },
    {
      "entropy": 5.860033178329468,
      "epoch": 5.867779428471555,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021687207271300945,
      "loss": 5.1588,
      "mean_token_accuracy": 0.223305381834507,
      "num_tokens": 136051626.0,
      "step": 67145
    },
    {
      "entropy": 5.8637841701507565,
      "epoch": 5.868216376824259,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021684197001823035,
      "loss": 5.2128,
      "mean_token_accuracy": 0.21613923013210296,
      "num_tokens": 136060938.0,
      "step": 67150
    },
    {
      "entropy": 5.920667886734009,
      "epoch": 5.868653325176964,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021681186843895274,
      "loss": 5.292,
      "mean_token_accuracy": 0.20785573571920396,
      "num_tokens": 136071612.0,
      "step": 67155
    },
    {
      "entropy": 5.8758885860443115,
      "epoch": 5.869090273529669,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021678176797575394,
      "loss": 5.1803,
      "mean_token_accuracy": 0.22119362652301788,
      "num_tokens": 136082273.0,
      "step": 67160
    },
    {
      "entropy": 5.876877641677856,
      "epoch": 5.8695272218823735,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021675166862921135,
      "loss": 5.2955,
      "mean_token_accuracy": 0.21250657141208648,
      "num_tokens": 136095392.0,
      "step": 67165
    },
    {
      "entropy": 5.906102180480957,
      "epoch": 5.869964170235078,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021672157039990232,
      "loss": 5.1612,
      "mean_token_accuracy": 0.2223142385482788,
      "num_tokens": 136105192.0,
      "step": 67170
    },
    {
      "entropy": 5.902859449386597,
      "epoch": 5.8704011185877825,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021669147328840405,
      "loss": 5.2526,
      "mean_token_accuracy": 0.2182951331138611,
      "num_tokens": 136115975.0,
      "step": 67175
    },
    {
      "entropy": 5.900802135467529,
      "epoch": 5.870838066940488,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021666137729529396,
      "loss": 5.2687,
      "mean_token_accuracy": 0.21226902306079865,
      "num_tokens": 136126339.0,
      "step": 67180
    },
    {
      "entropy": 5.8968370914459225,
      "epoch": 5.871275015293192,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021663128242114916,
      "loss": 5.2215,
      "mean_token_accuracy": 0.21895421743392945,
      "num_tokens": 136136584.0,
      "step": 67185
    },
    {
      "entropy": 5.89301495552063,
      "epoch": 5.871711963645897,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021660118866654693,
      "loss": 5.1866,
      "mean_token_accuracy": 0.2225720316171646,
      "num_tokens": 136147619.0,
      "step": 67190
    },
    {
      "entropy": 5.8242486953735355,
      "epoch": 5.872148911998602,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021657109603206447,
      "loss": 5.1705,
      "mean_token_accuracy": 0.22676611691713333,
      "num_tokens": 136157886.0,
      "step": 67195
    },
    {
      "entropy": 5.758865642547607,
      "epoch": 5.872585860351307,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021654100451827906,
      "loss": 5.0906,
      "mean_token_accuracy": 0.23307599425315856,
      "num_tokens": 136168562.0,
      "step": 67200
    },
    {
      "entropy": 5.815044641494751,
      "epoch": 5.873022808704011,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021651091412576784,
      "loss": 5.0827,
      "mean_token_accuracy": 0.22583973109722139,
      "num_tokens": 136178091.0,
      "step": 67205
    },
    {
      "entropy": 5.8092926979064945,
      "epoch": 5.8734597570567155,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002164808248551079,
      "loss": 5.1653,
      "mean_token_accuracy": 0.21904280185699462,
      "num_tokens": 136187471.0,
      "step": 67210
    },
    {
      "entropy": 5.815232372283935,
      "epoch": 5.873896705409421,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021645073670687644,
      "loss": 5.1152,
      "mean_token_accuracy": 0.22124789655208588,
      "num_tokens": 136197345.0,
      "step": 67215
    },
    {
      "entropy": 5.843695020675659,
      "epoch": 5.874333653762125,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021642064968165048,
      "loss": 5.1905,
      "mean_token_accuracy": 0.22049548774957656,
      "num_tokens": 136208782.0,
      "step": 67220
    },
    {
      "entropy": 5.916498708724975,
      "epoch": 5.87477060211483,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021639056378000714,
      "loss": 5.2302,
      "mean_token_accuracy": 0.2082400918006897,
      "num_tokens": 136218250.0,
      "step": 67225
    },
    {
      "entropy": 5.94364104270935,
      "epoch": 5.875207550467534,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002163604790025236,
      "loss": 5.2336,
      "mean_token_accuracy": 0.21536957174539567,
      "num_tokens": 136228389.0,
      "step": 67230
    },
    {
      "entropy": 5.869941854476929,
      "epoch": 5.87564449882024,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021633039534977672,
      "loss": 5.1653,
      "mean_token_accuracy": 0.22552737444639206,
      "num_tokens": 136238287.0,
      "step": 67235
    },
    {
      "entropy": 5.804437112808228,
      "epoch": 5.876081447172944,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021630031282234358,
      "loss": 5.1688,
      "mean_token_accuracy": 0.22399812787771226,
      "num_tokens": 136248258.0,
      "step": 67240
    },
    {
      "entropy": 5.805457353591919,
      "epoch": 5.876518395525649,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002162702314208012,
      "loss": 5.1769,
      "mean_token_accuracy": 0.22475456893444062,
      "num_tokens": 136258524.0,
      "step": 67245
    },
    {
      "entropy": 5.83909797668457,
      "epoch": 5.876955343878354,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021624015114572666,
      "loss": 5.1804,
      "mean_token_accuracy": 0.22071931511163712,
      "num_tokens": 136268293.0,
      "step": 67250
    },
    {
      "entropy": 5.922709894180298,
      "epoch": 5.877392292231058,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021621007199769672,
      "loss": 5.1867,
      "mean_token_accuracy": 0.2220923736691475,
      "num_tokens": 136276993.0,
      "step": 67255
    },
    {
      "entropy": 5.845469617843628,
      "epoch": 5.877829240583763,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021617999397728838,
      "loss": 5.1715,
      "mean_token_accuracy": 0.21937833726406097,
      "num_tokens": 136286942.0,
      "step": 67260
    },
    {
      "entropy": 5.838899230957031,
      "epoch": 5.878266188936467,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00021614991708507865,
      "loss": 5.1258,
      "mean_token_accuracy": 0.2243460476398468,
      "num_tokens": 136297648.0,
      "step": 67265
    },
    {
      "entropy": 5.886485195159912,
      "epoch": 5.878703137289173,
      "grad_norm": 1.125,
      "learning_rate": 0.00021611984132164424,
      "loss": 5.2146,
      "mean_token_accuracy": 0.2131659582257271,
      "num_tokens": 136308781.0,
      "step": 67270
    },
    {
      "entropy": 5.82776951789856,
      "epoch": 5.879140085641877,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021608976668756224,
      "loss": 5.1335,
      "mean_token_accuracy": 0.22617264091968536,
      "num_tokens": 136318494.0,
      "step": 67275
    },
    {
      "entropy": 5.897763204574585,
      "epoch": 5.879577033994582,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021605969318340935,
      "loss": 5.2213,
      "mean_token_accuracy": 0.21384437382221222,
      "num_tokens": 136328724.0,
      "step": 67280
    },
    {
      "entropy": 5.874914979934692,
      "epoch": 5.880013982347286,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021602962080976242,
      "loss": 5.2921,
      "mean_token_accuracy": 0.219061978161335,
      "num_tokens": 136340630.0,
      "step": 67285
    },
    {
      "entropy": 5.906552362442016,
      "epoch": 5.8804509306999915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021599954956719826,
      "loss": 5.1176,
      "mean_token_accuracy": 0.22075773477554322,
      "num_tokens": 136349774.0,
      "step": 67290
    },
    {
      "entropy": 5.778850126266479,
      "epoch": 5.880887879052696,
      "grad_norm": 1.125,
      "learning_rate": 0.00021596947945629364,
      "loss": 5.1508,
      "mean_token_accuracy": 0.23075590431690216,
      "num_tokens": 136359901.0,
      "step": 67295
    },
    {
      "entropy": 5.85232515335083,
      "epoch": 5.8813248274054,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00021593941047762545,
      "loss": 5.1846,
      "mean_token_accuracy": 0.2224324032664299,
      "num_tokens": 136370572.0,
      "step": 67300
    },
    {
      "entropy": 5.882918453216552,
      "epoch": 5.881761775758106,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00021590934263177015,
      "loss": 5.2327,
      "mean_token_accuracy": 0.21956628412008286,
      "num_tokens": 136379966.0,
      "step": 67305
    },
    {
      "entropy": 5.844583463668823,
      "epoch": 5.88219872411081,
      "grad_norm": 1.125,
      "learning_rate": 0.00021587927591930467,
      "loss": 5.1529,
      "mean_token_accuracy": 0.21740369349718094,
      "num_tokens": 136390615.0,
      "step": 67310
    },
    {
      "entropy": 5.899348640441895,
      "epoch": 5.882635672463515,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00021584921034080573,
      "loss": 5.1537,
      "mean_token_accuracy": 0.22672442346811295,
      "num_tokens": 136401099.0,
      "step": 67315
    },
    {
      "entropy": 5.86808934211731,
      "epoch": 5.883072620816219,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021581914589684982,
      "loss": 5.1222,
      "mean_token_accuracy": 0.22761340141296388,
      "num_tokens": 136410656.0,
      "step": 67320
    },
    {
      "entropy": 5.854814577102661,
      "epoch": 5.8835095691689245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021578908258801372,
      "loss": 5.1771,
      "mean_token_accuracy": 0.22256945222616195,
      "num_tokens": 136421224.0,
      "step": 67325
    },
    {
      "entropy": 5.797152137756347,
      "epoch": 5.883946517521629,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021575902041487405,
      "loss": 5.1664,
      "mean_token_accuracy": 0.22884691506624222,
      "num_tokens": 136430324.0,
      "step": 67330
    },
    {
      "entropy": 5.860334491729736,
      "epoch": 5.8843834658743335,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021572895937800745,
      "loss": 5.1776,
      "mean_token_accuracy": 0.21932793110609056,
      "num_tokens": 136440496.0,
      "step": 67335
    },
    {
      "entropy": 5.897741174697876,
      "epoch": 5.884820414227038,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002156988994779905,
      "loss": 5.1825,
      "mean_token_accuracy": 0.21827392131090165,
      "num_tokens": 136451411.0,
      "step": 67340
    },
    {
      "entropy": 5.881001996994018,
      "epoch": 5.885257362579743,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021566884071539966,
      "loss": 5.1908,
      "mean_token_accuracy": 0.21547040939331055,
      "num_tokens": 136460877.0,
      "step": 67345
    },
    {
      "entropy": 5.93163161277771,
      "epoch": 5.885694310932448,
      "grad_norm": 1.0625,
      "learning_rate": 0.00021563878309081158,
      "loss": 5.2535,
      "mean_token_accuracy": 0.21062489300966264,
      "num_tokens": 136471681.0,
      "step": 67350
    },
    {
      "entropy": 5.9174713611602785,
      "epoch": 5.886131259285152,
      "grad_norm": 1.046875,
      "learning_rate": 0.00021560872660480275,
      "loss": 5.2724,
      "mean_token_accuracy": 0.2113678991794586,
      "num_tokens": 136481570.0,
      "step": 67355
    },
    {
      "entropy": 5.8830572128295895,
      "epoch": 5.886568207637858,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021557867125794966,
      "loss": 5.2085,
      "mean_token_accuracy": 0.2190649077296257,
      "num_tokens": 136492616.0,
      "step": 67360
    },
    {
      "entropy": 5.8876136302947994,
      "epoch": 5.887005155990562,
      "grad_norm": 1.125,
      "learning_rate": 0.00021554861705082879,
      "loss": 5.2402,
      "mean_token_accuracy": 0.2141732543706894,
      "num_tokens": 136502723.0,
      "step": 67365
    },
    {
      "entropy": 5.893055963516235,
      "epoch": 5.8874421043432665,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002155185639840166,
      "loss": 5.2089,
      "mean_token_accuracy": 0.220580030977726,
      "num_tokens": 136514194.0,
      "step": 67370
    },
    {
      "entropy": 5.875965881347656,
      "epoch": 5.887879052695971,
      "grad_norm": 1.078125,
      "learning_rate": 0.00021548851205808951,
      "loss": 5.1962,
      "mean_token_accuracy": 0.22198109924793244,
      "num_tokens": 136524740.0,
      "step": 67375
    },
    {
      "entropy": 5.891937685012818,
      "epoch": 5.888316001048676,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021545846127362395,
      "loss": 5.1652,
      "mean_token_accuracy": 0.2215964674949646,
      "num_tokens": 136534120.0,
      "step": 67380
    },
    {
      "entropy": 5.829614782333374,
      "epoch": 5.888752949401381,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00021542841163119632,
      "loss": 5.1454,
      "mean_token_accuracy": 0.22215994149446489,
      "num_tokens": 136545021.0,
      "step": 67385
    },
    {
      "entropy": 5.802795362472534,
      "epoch": 5.889189897754085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002153983631313829,
      "loss": 5.1222,
      "mean_token_accuracy": 0.22263740003108978,
      "num_tokens": 136554807.0,
      "step": 67390
    },
    {
      "entropy": 5.873703813552856,
      "epoch": 5.88962684610679,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021536831577476012,
      "loss": 5.1946,
      "mean_token_accuracy": 0.21737397611141204,
      "num_tokens": 136565065.0,
      "step": 67395
    },
    {
      "entropy": 5.817498302459716,
      "epoch": 5.890063794459495,
      "grad_norm": 1.03125,
      "learning_rate": 0.00021533826956190432,
      "loss": 5.1344,
      "mean_token_accuracy": 0.2198767051100731,
      "num_tokens": 136574708.0,
      "step": 67400
    },
    {
      "entropy": 5.793298196792603,
      "epoch": 5.8905007428122,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021530822449339178,
      "loss": 5.1465,
      "mean_token_accuracy": 0.23123454451560974,
      "num_tokens": 136586149.0,
      "step": 67405
    },
    {
      "entropy": 5.810852575302124,
      "epoch": 5.890937691164904,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002152781805697988,
      "loss": 5.1653,
      "mean_token_accuracy": 0.22052039057016373,
      "num_tokens": 136597103.0,
      "step": 67410
    },
    {
      "entropy": 5.870308494567871,
      "epoch": 5.891374639517609,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002152481377917016,
      "loss": 5.1497,
      "mean_token_accuracy": 0.22549151182174682,
      "num_tokens": 136607032.0,
      "step": 67415
    },
    {
      "entropy": 5.945936298370361,
      "epoch": 5.891811587870314,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00021521809615967637,
      "loss": 5.2315,
      "mean_token_accuracy": 0.22278264611959459,
      "num_tokens": 136617497.0,
      "step": 67420
    },
    {
      "entropy": 5.880619382858276,
      "epoch": 5.892248536223018,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021518805567429945,
      "loss": 5.2376,
      "mean_token_accuracy": 0.2101381927728653,
      "num_tokens": 136627500.0,
      "step": 67425
    },
    {
      "entropy": 5.870204019546509,
      "epoch": 5.892685484575723,
      "grad_norm": 1.125,
      "learning_rate": 0.0002151580163361469,
      "loss": 5.2254,
      "mean_token_accuracy": 0.2192816972732544,
      "num_tokens": 136637729.0,
      "step": 67430
    },
    {
      "entropy": 5.957834768295288,
      "epoch": 5.893122432928428,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021512797814579498,
      "loss": 5.2561,
      "mean_token_accuracy": 0.2126225396990776,
      "num_tokens": 136647871.0,
      "step": 67435
    },
    {
      "entropy": 5.784438180923462,
      "epoch": 5.893559381281133,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021509794110381985,
      "loss": 5.1456,
      "mean_token_accuracy": 0.22931046336889266,
      "num_tokens": 136657245.0,
      "step": 67440
    },
    {
      "entropy": 5.803116226196289,
      "epoch": 5.893996329633837,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021506790521079756,
      "loss": 5.1732,
      "mean_token_accuracy": 0.22255283296108247,
      "num_tokens": 136667391.0,
      "step": 67445
    },
    {
      "entropy": 5.838616752624512,
      "epoch": 5.894433277986542,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021503787046730423,
      "loss": 5.1795,
      "mean_token_accuracy": 0.21852242052555085,
      "num_tokens": 136676971.0,
      "step": 67450
    },
    {
      "entropy": 5.864603424072266,
      "epoch": 5.894870226339247,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021500783687391595,
      "loss": 5.1437,
      "mean_token_accuracy": 0.22683824747800826,
      "num_tokens": 136686479.0,
      "step": 67455
    },
    {
      "entropy": 5.885271120071411,
      "epoch": 5.895307174691951,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021497780443120886,
      "loss": 5.1524,
      "mean_token_accuracy": 0.22388266324996947,
      "num_tokens": 136696015.0,
      "step": 67460
    },
    {
      "entropy": 5.862126207351684,
      "epoch": 5.895744123044656,
      "grad_norm": 1.125,
      "learning_rate": 0.0002149477731397589,
      "loss": 5.1868,
      "mean_token_accuracy": 0.2197664886713028,
      "num_tokens": 136705414.0,
      "step": 67465
    },
    {
      "entropy": 5.807699203491211,
      "epoch": 5.896181071397361,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002149177430001421,
      "loss": 5.1023,
      "mean_token_accuracy": 0.2257761463522911,
      "num_tokens": 136714954.0,
      "step": 67470
    },
    {
      "entropy": 5.832133769989014,
      "epoch": 5.896618019750066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002148877140129345,
      "loss": 5.1705,
      "mean_token_accuracy": 0.2243797868490219,
      "num_tokens": 136724638.0,
      "step": 67475
    },
    {
      "entropy": 5.921682357788086,
      "epoch": 5.89705496810277,
      "grad_norm": 1.125,
      "learning_rate": 0.00021485768617871204,
      "loss": 5.2392,
      "mean_token_accuracy": 0.20869479030370713,
      "num_tokens": 136734944.0,
      "step": 67480
    },
    {
      "entropy": 5.877887296676636,
      "epoch": 5.897491916455475,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021482765949805072,
      "loss": 5.2067,
      "mean_token_accuracy": 0.22308150231838225,
      "num_tokens": 136745958.0,
      "step": 67485
    },
    {
      "entropy": 5.897628736495972,
      "epoch": 5.89792886480818,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002147976339715263,
      "loss": 5.2192,
      "mean_token_accuracy": 0.2188554361462593,
      "num_tokens": 136754915.0,
      "step": 67490
    },
    {
      "entropy": 5.86933217048645,
      "epoch": 5.8983658131608845,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00021476760959971493,
      "loss": 5.1272,
      "mean_token_accuracy": 0.22591274827718735,
      "num_tokens": 136764070.0,
      "step": 67495
    },
    {
      "entropy": 5.867020940780639,
      "epoch": 5.898802761513589,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002147375863831923,
      "loss": 5.1317,
      "mean_token_accuracy": 0.22809478193521499,
      "num_tokens": 136773732.0,
      "step": 67500
    },
    {
      "entropy": 5.849305057525635,
      "epoch": 5.899239709866293,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002147075643225343,
      "loss": 5.2233,
      "mean_token_accuracy": 0.2194196492433548,
      "num_tokens": 136782862.0,
      "step": 67505
    },
    {
      "entropy": 5.820456504821777,
      "epoch": 5.899676658218999,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021467754341831691,
      "loss": 5.1702,
      "mean_token_accuracy": 0.22200393825769424,
      "num_tokens": 136793195.0,
      "step": 67510
    },
    {
      "entropy": 5.8579246520996096,
      "epoch": 5.900113606571703,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002146475236711158,
      "loss": 5.2273,
      "mean_token_accuracy": 0.2157132849097252,
      "num_tokens": 136803500.0,
      "step": 67515
    },
    {
      "entropy": 5.866679382324219,
      "epoch": 5.900550554924408,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002146175050815068,
      "loss": 5.1191,
      "mean_token_accuracy": 0.22695374339818955,
      "num_tokens": 136813348.0,
      "step": 67520
    },
    {
      "entropy": 5.8722704410552975,
      "epoch": 5.900987503277113,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002145874876500657,
      "loss": 5.2511,
      "mean_token_accuracy": 0.21423248797655106,
      "num_tokens": 136824766.0,
      "step": 67525
    },
    {
      "entropy": 5.808745622634888,
      "epoch": 5.9014244516298175,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021455747137736825,
      "loss": 5.1629,
      "mean_token_accuracy": 0.23134966492652892,
      "num_tokens": 136835288.0,
      "step": 67530
    },
    {
      "entropy": 5.936719703674316,
      "epoch": 5.901861399982522,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021452745626399016,
      "loss": 5.2291,
      "mean_token_accuracy": 0.21072373986244203,
      "num_tokens": 136846491.0,
      "step": 67535
    },
    {
      "entropy": 5.8748668193817135,
      "epoch": 5.9022983483352265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021449744231050717,
      "loss": 5.1365,
      "mean_token_accuracy": 0.2243300825357437,
      "num_tokens": 136855743.0,
      "step": 67540
    },
    {
      "entropy": 5.881786203384399,
      "epoch": 5.902735296687932,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021446742951749493,
      "loss": 5.1798,
      "mean_token_accuracy": 0.2219702810049057,
      "num_tokens": 136865923.0,
      "step": 67545
    },
    {
      "entropy": 5.776490926742554,
      "epoch": 5.903172245040636,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021443741788552907,
      "loss": 5.1112,
      "mean_token_accuracy": 0.2311447963118553,
      "num_tokens": 136876284.0,
      "step": 67550
    },
    {
      "entropy": 5.898776578903198,
      "epoch": 5.903609193393341,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021440740741518534,
      "loss": 5.2477,
      "mean_token_accuracy": 0.21288086026906966,
      "num_tokens": 136886297.0,
      "step": 67555
    },
    {
      "entropy": 5.828553581237793,
      "epoch": 5.904046141746045,
      "grad_norm": 1.03125,
      "learning_rate": 0.00021437739810703922,
      "loss": 5.148,
      "mean_token_accuracy": 0.21725087314844133,
      "num_tokens": 136895946.0,
      "step": 67560
    },
    {
      "entropy": 5.827222394943237,
      "epoch": 5.904483090098751,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002143473899616664,
      "loss": 5.1584,
      "mean_token_accuracy": 0.2219487965106964,
      "num_tokens": 136905623.0,
      "step": 67565
    },
    {
      "entropy": 5.842682409286499,
      "epoch": 5.904920038451455,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021431738297964244,
      "loss": 5.2334,
      "mean_token_accuracy": 0.21479922384023667,
      "num_tokens": 136915735.0,
      "step": 67570
    },
    {
      "entropy": 5.872321081161499,
      "epoch": 5.9053569868041595,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021428737716154285,
      "loss": 5.1587,
      "mean_token_accuracy": 0.21916113793849945,
      "num_tokens": 136924737.0,
      "step": 67575
    },
    {
      "entropy": 5.88627667427063,
      "epoch": 5.905793935156865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002142573725079432,
      "loss": 5.1879,
      "mean_token_accuracy": 0.21905974298715591,
      "num_tokens": 136934476.0,
      "step": 67580
    },
    {
      "entropy": 5.898710918426514,
      "epoch": 5.906230883509569,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002142273690194189,
      "loss": 5.2251,
      "mean_token_accuracy": 0.21633452475070952,
      "num_tokens": 136944985.0,
      "step": 67585
    },
    {
      "entropy": 5.8278426170349125,
      "epoch": 5.906667831862274,
      "grad_norm": 1.078125,
      "learning_rate": 0.00021419736669654561,
      "loss": 5.1267,
      "mean_token_accuracy": 0.22470396906137466,
      "num_tokens": 136955023.0,
      "step": 67590
    },
    {
      "entropy": 5.849764108657837,
      "epoch": 5.907104780214978,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021416736553989864,
      "loss": 5.1428,
      "mean_token_accuracy": 0.21979418098926545,
      "num_tokens": 136964526.0,
      "step": 67595
    },
    {
      "entropy": 5.8477928161621096,
      "epoch": 5.907541728567684,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021413736555005343,
      "loss": 5.1351,
      "mean_token_accuracy": 0.2221284657716751,
      "num_tokens": 136975310.0,
      "step": 67600
    },
    {
      "entropy": 5.912347221374512,
      "epoch": 5.907978676920388,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00021410736672758553,
      "loss": 5.198,
      "mean_token_accuracy": 0.21966669857501983,
      "num_tokens": 136986085.0,
      "step": 67605
    },
    {
      "entropy": 5.802706289291382,
      "epoch": 5.908415625273093,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002140773690730702,
      "loss": 5.128,
      "mean_token_accuracy": 0.224717316031456,
      "num_tokens": 136996270.0,
      "step": 67610
    },
    {
      "entropy": 5.7925848960876465,
      "epoch": 5.908852573625797,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002140473725870828,
      "loss": 5.0852,
      "mean_token_accuracy": 0.2297326222062111,
      "num_tokens": 137006476.0,
      "step": 67615
    },
    {
      "entropy": 5.8824646949768065,
      "epoch": 5.909289521978502,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021401737727019877,
      "loss": 5.2392,
      "mean_token_accuracy": 0.21253261417150499,
      "num_tokens": 137016056.0,
      "step": 67620
    },
    {
      "entropy": 5.854554891586304,
      "epoch": 5.909726470331207,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021398738312299344,
      "loss": 5.1985,
      "mean_token_accuracy": 0.2175740644335747,
      "num_tokens": 137026081.0,
      "step": 67625
    },
    {
      "entropy": 5.838731145858764,
      "epoch": 5.910163418683911,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00021395739014604204,
      "loss": 5.2388,
      "mean_token_accuracy": 0.2171509876847267,
      "num_tokens": 137036525.0,
      "step": 67630
    },
    {
      "entropy": 5.814050340652466,
      "epoch": 5.910600367036617,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021392739833991987,
      "loss": 5.1126,
      "mean_token_accuracy": 0.22839071303606034,
      "num_tokens": 137046654.0,
      "step": 67635
    },
    {
      "entropy": 5.858284616470337,
      "epoch": 5.911037315389321,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021389740770520223,
      "loss": 5.1684,
      "mean_token_accuracy": 0.2217394158244133,
      "num_tokens": 137056929.0,
      "step": 67640
    },
    {
      "entropy": 5.7732891082763675,
      "epoch": 5.911474263742026,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002138674182424643,
      "loss": 5.084,
      "mean_token_accuracy": 0.23424223959445953,
      "num_tokens": 137066508.0,
      "step": 67645
    },
    {
      "entropy": 5.89617018699646,
      "epoch": 5.91191121209473,
      "grad_norm": 1.0625,
      "learning_rate": 0.00021383742995228133,
      "loss": 5.2448,
      "mean_token_accuracy": 0.213581445813179,
      "num_tokens": 137076836.0,
      "step": 67650
    },
    {
      "entropy": 5.844670581817627,
      "epoch": 5.9123481604474355,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021380744283522852,
      "loss": 5.1995,
      "mean_token_accuracy": 0.21009046584367752,
      "num_tokens": 137087736.0,
      "step": 67655
    },
    {
      "entropy": 5.885467672348023,
      "epoch": 5.91278510880014,
      "grad_norm": 1.09375,
      "learning_rate": 0.000213777456891881,
      "loss": 5.1914,
      "mean_token_accuracy": 0.21980078369379044,
      "num_tokens": 137098022.0,
      "step": 67660
    },
    {
      "entropy": 5.882741546630859,
      "epoch": 5.913222057152844,
      "grad_norm": 1.0625,
      "learning_rate": 0.000213747472122814,
      "loss": 5.2023,
      "mean_token_accuracy": 0.21419377028942108,
      "num_tokens": 137108523.0,
      "step": 67665
    },
    {
      "entropy": 5.885676097869873,
      "epoch": 5.913659005505549,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021371748852860246,
      "loss": 5.2254,
      "mean_token_accuracy": 0.22022032737731934,
      "num_tokens": 137120036.0,
      "step": 67670
    },
    {
      "entropy": 5.888956451416016,
      "epoch": 5.914095953858254,
      "grad_norm": 1.0,
      "learning_rate": 0.00021368750610982162,
      "loss": 5.2357,
      "mean_token_accuracy": 0.21812534630298613,
      "num_tokens": 137130371.0,
      "step": 67675
    },
    {
      "entropy": 5.881307363510132,
      "epoch": 5.914532902210959,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002136575248670466,
      "loss": 5.2295,
      "mean_token_accuracy": 0.21213730424642563,
      "num_tokens": 137140923.0,
      "step": 67680
    },
    {
      "entropy": 5.819354391098022,
      "epoch": 5.914969850563663,
      "grad_norm": 1.0625,
      "learning_rate": 0.00021362754480085233,
      "loss": 5.1041,
      "mean_token_accuracy": 0.2230118989944458,
      "num_tokens": 137151104.0,
      "step": 67685
    },
    {
      "entropy": 5.912252378463745,
      "epoch": 5.915406798916369,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021359756591181401,
      "loss": 5.2806,
      "mean_token_accuracy": 0.2127352014183998,
      "num_tokens": 137160549.0,
      "step": 67690
    },
    {
      "entropy": 5.908971166610717,
      "epoch": 5.915843747269073,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002135675882005065,
      "loss": 5.1943,
      "mean_token_accuracy": 0.22020630836486815,
      "num_tokens": 137170052.0,
      "step": 67695
    },
    {
      "entropy": 5.8411565780639645,
      "epoch": 5.9162806956217775,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00021353761166750484,
      "loss": 5.1154,
      "mean_token_accuracy": 0.23056040406227113,
      "num_tokens": 137179721.0,
      "step": 67700
    },
    {
      "entropy": 5.884853982925415,
      "epoch": 5.916717643974482,
      "grad_norm": 1.078125,
      "learning_rate": 0.000213507636313384,
      "loss": 5.2395,
      "mean_token_accuracy": 0.2122403398156166,
      "num_tokens": 137189890.0,
      "step": 67705
    },
    {
      "entropy": 5.782976675033569,
      "epoch": 5.917154592327186,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021347766213871888,
      "loss": 5.101,
      "mean_token_accuracy": 0.22517765313386917,
      "num_tokens": 137200256.0,
      "step": 67710
    },
    {
      "entropy": 5.900360727310181,
      "epoch": 5.917591540679892,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021344768914408445,
      "loss": 5.2509,
      "mean_token_accuracy": 0.21517229676246644,
      "num_tokens": 137211082.0,
      "step": 67715
    },
    {
      "entropy": 5.876407480239868,
      "epoch": 5.918028489032596,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021341771733005566,
      "loss": 5.2323,
      "mean_token_accuracy": 0.2204744577407837,
      "num_tokens": 137221570.0,
      "step": 67720
    },
    {
      "entropy": 5.857307720184326,
      "epoch": 5.918465437385301,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002133877466972073,
      "loss": 5.1627,
      "mean_token_accuracy": 0.22366515845060347,
      "num_tokens": 137231836.0,
      "step": 67725
    },
    {
      "entropy": 5.759245300292969,
      "epoch": 5.918902385738006,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021335777724611416,
      "loss": 5.1016,
      "mean_token_accuracy": 0.2364266261458397,
      "num_tokens": 137241523.0,
      "step": 67730
    },
    {
      "entropy": 5.938065767288208,
      "epoch": 5.919339334090711,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021332780897735123,
      "loss": 5.2533,
      "mean_token_accuracy": 0.20664129704236983,
      "num_tokens": 137251960.0,
      "step": 67735
    },
    {
      "entropy": 5.8744100570678714,
      "epoch": 5.919776282443415,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002132978418914932,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21060800403356553,
      "num_tokens": 137262089.0,
      "step": 67740
    },
    {
      "entropy": 5.872996520996094,
      "epoch": 5.92021323079612,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021326787598911496,
      "loss": 5.185,
      "mean_token_accuracy": 0.22080783993005754,
      "num_tokens": 137271955.0,
      "step": 67745
    },
    {
      "entropy": 5.84998631477356,
      "epoch": 5.920650179148825,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021323791127079113,
      "loss": 5.0953,
      "mean_token_accuracy": 0.21567561328411103,
      "num_tokens": 137281530.0,
      "step": 67750
    },
    {
      "entropy": 5.889038705825806,
      "epoch": 5.921087127501529,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021320794773709652,
      "loss": 5.2433,
      "mean_token_accuracy": 0.21804862916469575,
      "num_tokens": 137291347.0,
      "step": 67755
    },
    {
      "entropy": 5.812016344070434,
      "epoch": 5.921524075854234,
      "grad_norm": 1.125,
      "learning_rate": 0.00021317798538860589,
      "loss": 5.176,
      "mean_token_accuracy": 0.2228507488965988,
      "num_tokens": 137302047.0,
      "step": 67760
    },
    {
      "entropy": 5.9089268207550045,
      "epoch": 5.921961024206938,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021314802422589391,
      "loss": 5.2361,
      "mean_token_accuracy": 0.2127871572971344,
      "num_tokens": 137311636.0,
      "step": 67765
    },
    {
      "entropy": 5.8316529273986815,
      "epoch": 5.922397972559644,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021311806424953523,
      "loss": 5.1603,
      "mean_token_accuracy": 0.22273559421300887,
      "num_tokens": 137321964.0,
      "step": 67770
    },
    {
      "entropy": 5.86822543144226,
      "epoch": 5.922834920912348,
      "grad_norm": 1.125,
      "learning_rate": 0.0002130881054601045,
      "loss": 5.2199,
      "mean_token_accuracy": 0.21780480891466142,
      "num_tokens": 137332084.0,
      "step": 67775
    },
    {
      "entropy": 5.924166059494018,
      "epoch": 5.923271869265053,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002130581478581763,
      "loss": 5.2699,
      "mean_token_accuracy": 0.2139489933848381,
      "num_tokens": 137342138.0,
      "step": 67780
    },
    {
      "entropy": 5.877191686630249,
      "epoch": 5.923708817617758,
      "grad_norm": 1.046875,
      "learning_rate": 0.00021302819144432527,
      "loss": 5.1947,
      "mean_token_accuracy": 0.2207550972700119,
      "num_tokens": 137351886.0,
      "step": 67785
    },
    {
      "entropy": 5.880921649932861,
      "epoch": 5.924145765970462,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021299823621912612,
      "loss": 5.1877,
      "mean_token_accuracy": 0.22038525938987732,
      "num_tokens": 137361480.0,
      "step": 67790
    },
    {
      "entropy": 5.878343725204468,
      "epoch": 5.924582714323167,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021296828218315318,
      "loss": 5.2041,
      "mean_token_accuracy": 0.2223843440413475,
      "num_tokens": 137372770.0,
      "step": 67795
    },
    {
      "entropy": 5.910775423049927,
      "epoch": 5.925019662675871,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021293832933698116,
      "loss": 5.2736,
      "mean_token_accuracy": 0.21373497247695922,
      "num_tokens": 137383531.0,
      "step": 67800
    },
    {
      "entropy": 5.850192642211914,
      "epoch": 5.925456611028577,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021290837768118444,
      "loss": 5.1676,
      "mean_token_accuracy": 0.2262984961271286,
      "num_tokens": 137393290.0,
      "step": 67805
    },
    {
      "entropy": 5.828208971023559,
      "epoch": 5.925893559381281,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002128784272163376,
      "loss": 5.1112,
      "mean_token_accuracy": 0.22214192152023315,
      "num_tokens": 137403115.0,
      "step": 67810
    },
    {
      "entropy": 5.853261709213257,
      "epoch": 5.926330507733986,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002128484779430151,
      "loss": 5.2065,
      "mean_token_accuracy": 0.21495856046676637,
      "num_tokens": 137412343.0,
      "step": 67815
    },
    {
      "entropy": 5.884172344207764,
      "epoch": 5.92676745608669,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021281852986179133,
      "loss": 5.2458,
      "mean_token_accuracy": 0.21595387756824494,
      "num_tokens": 137422209.0,
      "step": 67820
    },
    {
      "entropy": 5.925195932388306,
      "epoch": 5.9272044044393954,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021278858297324078,
      "loss": 5.2574,
      "mean_token_accuracy": 0.21560018062591552,
      "num_tokens": 137431615.0,
      "step": 67825
    },
    {
      "entropy": 5.877206468582154,
      "epoch": 5.9276413527921,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021275863727793776,
      "loss": 5.1632,
      "mean_token_accuracy": 0.2238369643688202,
      "num_tokens": 137441850.0,
      "step": 67830
    },
    {
      "entropy": 5.794729280471802,
      "epoch": 5.928078301144804,
      "grad_norm": 1.078125,
      "learning_rate": 0.00021272869277645667,
      "loss": 5.0779,
      "mean_token_accuracy": 0.22890057265758515,
      "num_tokens": 137451906.0,
      "step": 67835
    },
    {
      "entropy": 5.810908460617066,
      "epoch": 5.92851524949751,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00021269874946937194,
      "loss": 5.1629,
      "mean_token_accuracy": 0.22670662254095078,
      "num_tokens": 137462478.0,
      "step": 67840
    },
    {
      "entropy": 5.878889369964599,
      "epoch": 5.928952197850214,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021266880735725785,
      "loss": 5.1764,
      "mean_token_accuracy": 0.2178562879562378,
      "num_tokens": 137473077.0,
      "step": 67845
    },
    {
      "entropy": 5.892953109741211,
      "epoch": 5.929389146202919,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021263886644068864,
      "loss": 5.2046,
      "mean_token_accuracy": 0.21329241693019868,
      "num_tokens": 137483239.0,
      "step": 67850
    },
    {
      "entropy": 5.856145763397217,
      "epoch": 5.929826094555623,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021260892672023867,
      "loss": 5.1226,
      "mean_token_accuracy": 0.2274843394756317,
      "num_tokens": 137494232.0,
      "step": 67855
    },
    {
      "entropy": 5.898625898361206,
      "epoch": 5.9302630429083285,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002125789881964822,
      "loss": 5.2022,
      "mean_token_accuracy": 0.21117056161165237,
      "num_tokens": 137503902.0,
      "step": 67860
    },
    {
      "entropy": 5.839788913726807,
      "epoch": 5.930699991261033,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021254905086999344,
      "loss": 5.2088,
      "mean_token_accuracy": 0.21925950199365615,
      "num_tokens": 137513061.0,
      "step": 67865
    },
    {
      "entropy": 5.810002326965332,
      "epoch": 5.9311369396137374,
      "grad_norm": 1.125,
      "learning_rate": 0.00021251911474134662,
      "loss": 5.1072,
      "mean_token_accuracy": 0.23036912232637405,
      "num_tokens": 137523192.0,
      "step": 67870
    },
    {
      "entropy": 5.914357852935791,
      "epoch": 5.931573887966442,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021248917981111582,
      "loss": 5.1548,
      "mean_token_accuracy": 0.217241133749485,
      "num_tokens": 137533051.0,
      "step": 67875
    },
    {
      "entropy": 5.771841430664063,
      "epoch": 5.932010836319147,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021245924607987538,
      "loss": 5.1011,
      "mean_token_accuracy": 0.2329208254814148,
      "num_tokens": 137543018.0,
      "step": 67880
    },
    {
      "entropy": 5.870952558517456,
      "epoch": 5.932447784671852,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021242931354819943,
      "loss": 5.2539,
      "mean_token_accuracy": 0.21342408359050752,
      "num_tokens": 137552888.0,
      "step": 67885
    },
    {
      "entropy": 5.806854104995727,
      "epoch": 5.932884733024556,
      "grad_norm": 1.0625,
      "learning_rate": 0.00021239938221666194,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21889517605304717,
      "num_tokens": 137562877.0,
      "step": 67890
    },
    {
      "entropy": 5.897919034957885,
      "epoch": 5.933321681377262,
      "grad_norm": 1.125,
      "learning_rate": 0.00021236945208583712,
      "loss": 5.2311,
      "mean_token_accuracy": 0.20180242657661437,
      "num_tokens": 137573375.0,
      "step": 67895
    },
    {
      "entropy": 5.857644319534302,
      "epoch": 5.933758629729966,
      "grad_norm": 1.078125,
      "learning_rate": 0.00021233952315629907,
      "loss": 5.1926,
      "mean_token_accuracy": 0.2212660312652588,
      "num_tokens": 137583190.0,
      "step": 67900
    },
    {
      "entropy": 5.969005060195923,
      "epoch": 5.9341955780826705,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002123095954286217,
      "loss": 5.2246,
      "mean_token_accuracy": 0.2110506549477577,
      "num_tokens": 137593283.0,
      "step": 67905
    },
    {
      "entropy": 5.825566625595092,
      "epoch": 5.934632526435375,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002122796689033793,
      "loss": 5.1383,
      "mean_token_accuracy": 0.22802665829658508,
      "num_tokens": 137603638.0,
      "step": 67910
    },
    {
      "entropy": 5.81635594367981,
      "epoch": 5.93506947478808,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021224974358114557,
      "loss": 5.1688,
      "mean_token_accuracy": 0.21575087457895278,
      "num_tokens": 137613928.0,
      "step": 67915
    },
    {
      "entropy": 5.831985902786255,
      "epoch": 5.935506423140785,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002122198194624947,
      "loss": 5.1647,
      "mean_token_accuracy": 0.21872459799051286,
      "num_tokens": 137624747.0,
      "step": 67920
    },
    {
      "entropy": 5.807035255432129,
      "epoch": 5.935943371493489,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021218989654800058,
      "loss": 5.1354,
      "mean_token_accuracy": 0.22689522802829742,
      "num_tokens": 137635599.0,
      "step": 67925
    },
    {
      "entropy": 5.83734712600708,
      "epoch": 5.936380319846194,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002121599748382372,
      "loss": 5.1675,
      "mean_token_accuracy": 0.2248977690935135,
      "num_tokens": 137646421.0,
      "step": 67930
    },
    {
      "entropy": 5.8591149806976315,
      "epoch": 5.936817268198899,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002121300543337784,
      "loss": 5.1099,
      "mean_token_accuracy": 0.23187205940485,
      "num_tokens": 137656130.0,
      "step": 67935
    },
    {
      "entropy": 5.851056146621704,
      "epoch": 5.937254216551604,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021210013503519804,
      "loss": 5.1201,
      "mean_token_accuracy": 0.2276336058974266,
      "num_tokens": 137666037.0,
      "step": 67940
    },
    {
      "entropy": 5.818784141540528,
      "epoch": 5.937691164904308,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002120702169430701,
      "loss": 5.1333,
      "mean_token_accuracy": 0.22412640303373338,
      "num_tokens": 137676434.0,
      "step": 67945
    },
    {
      "entropy": 5.758370876312256,
      "epoch": 5.938128113257013,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021204030005796836,
      "loss": 5.1113,
      "mean_token_accuracy": 0.22866294980049134,
      "num_tokens": 137686349.0,
      "step": 67950
    },
    {
      "entropy": 5.855712127685547,
      "epoch": 5.938565061609718,
      "grad_norm": 1.125,
      "learning_rate": 0.00021201038438046666,
      "loss": 5.1247,
      "mean_token_accuracy": 0.21915023177862167,
      "num_tokens": 137695996.0,
      "step": 67955
    },
    {
      "entropy": 5.845601177215576,
      "epoch": 5.939002009962422,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021198046991113885,
      "loss": 5.1952,
      "mean_token_accuracy": 0.21939148008823395,
      "num_tokens": 137704979.0,
      "step": 67960
    },
    {
      "entropy": 5.811288261413575,
      "epoch": 5.939438958315127,
      "grad_norm": 1.046875,
      "learning_rate": 0.00021195055665055863,
      "loss": 5.105,
      "mean_token_accuracy": 0.2209096997976303,
      "num_tokens": 137715167.0,
      "step": 67965
    },
    {
      "entropy": 5.773065996170044,
      "epoch": 5.939875906667832,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002119206445992997,
      "loss": 5.1215,
      "mean_token_accuracy": 0.22893617004156114,
      "num_tokens": 137724740.0,
      "step": 67970
    },
    {
      "entropy": 5.80446720123291,
      "epoch": 5.940312855020537,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021189073375793587,
      "loss": 5.1112,
      "mean_token_accuracy": 0.22279162108898162,
      "num_tokens": 137734424.0,
      "step": 67975
    },
    {
      "entropy": 5.854788875579834,
      "epoch": 5.940749803373241,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021186082412704088,
      "loss": 5.1365,
      "mean_token_accuracy": 0.21930664628744126,
      "num_tokens": 137744226.0,
      "step": 67980
    },
    {
      "entropy": 5.848227548599243,
      "epoch": 5.941186751725946,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002118309157071884,
      "loss": 5.1964,
      "mean_token_accuracy": 0.2208077445626259,
      "num_tokens": 137754320.0,
      "step": 67985
    },
    {
      "entropy": 5.822793054580688,
      "epoch": 5.941623700078651,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021180100849895202,
      "loss": 5.1404,
      "mean_token_accuracy": 0.22400965839624404,
      "num_tokens": 137765269.0,
      "step": 67990
    },
    {
      "entropy": 5.837355995178223,
      "epoch": 5.942060648431355,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002117711025029054,
      "loss": 5.1111,
      "mean_token_accuracy": 0.2220176488161087,
      "num_tokens": 137774907.0,
      "step": 67995
    },
    {
      "entropy": 5.840958547592163,
      "epoch": 5.94249759678406,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021174119771962213,
      "loss": 5.2615,
      "mean_token_accuracy": 0.22081296592950822,
      "num_tokens": 137785149.0,
      "step": 68000
    },
    {
      "entropy": 5.850549840927124,
      "epoch": 5.942934545136765,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021171129414967587,
      "loss": 5.1946,
      "mean_token_accuracy": 0.21825399100780488,
      "num_tokens": 137795251.0,
      "step": 68005
    },
    {
      "entropy": 5.926516914367676,
      "epoch": 5.94337149348947,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021168139179364022,
      "loss": 5.2917,
      "mean_token_accuracy": 0.21137077659368514,
      "num_tokens": 137806498.0,
      "step": 68010
    },
    {
      "entropy": 5.92105712890625,
      "epoch": 5.943808441842174,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00021165149065208855,
      "loss": 5.2228,
      "mean_token_accuracy": 0.21649890393018723,
      "num_tokens": 137818103.0,
      "step": 68015
    },
    {
      "entropy": 5.850191879272461,
      "epoch": 5.944245390194879,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002116215907255945,
      "loss": 5.096,
      "mean_token_accuracy": 0.23173731863498687,
      "num_tokens": 137827803.0,
      "step": 68020
    },
    {
      "entropy": 5.823595333099365,
      "epoch": 5.944682338547584,
      "grad_norm": 1.0625,
      "learning_rate": 0.00021159169201473156,
      "loss": 5.1802,
      "mean_token_accuracy": 0.21732815057039262,
      "num_tokens": 137838192.0,
      "step": 68025
    },
    {
      "entropy": 5.807587289810181,
      "epoch": 5.9451192869002885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002115617945200732,
      "loss": 5.22,
      "mean_token_accuracy": 0.2086254984140396,
      "num_tokens": 137848117.0,
      "step": 68030
    },
    {
      "entropy": 5.890899515151977,
      "epoch": 5.945556235252993,
      "grad_norm": 1.125,
      "learning_rate": 0.00021153189824219283,
      "loss": 5.2637,
      "mean_token_accuracy": 0.21313586682081223,
      "num_tokens": 137858534.0,
      "step": 68035
    },
    {
      "entropy": 5.910704278945923,
      "epoch": 5.945993183605697,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002115020031816639,
      "loss": 5.1928,
      "mean_token_accuracy": 0.21056549847126008,
      "num_tokens": 137869359.0,
      "step": 68040
    },
    {
      "entropy": 5.879097557067871,
      "epoch": 5.946430131958403,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021147210933905993,
      "loss": 5.2093,
      "mean_token_accuracy": 0.22080986499786376,
      "num_tokens": 137880108.0,
      "step": 68045
    },
    {
      "entropy": 5.834100580215454,
      "epoch": 5.946867080311107,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021144221671495411,
      "loss": 5.1195,
      "mean_token_accuracy": 0.22505532205104828,
      "num_tokens": 137891263.0,
      "step": 68050
    },
    {
      "entropy": 5.868061542510986,
      "epoch": 5.947304028663812,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021141232530991989,
      "loss": 5.1891,
      "mean_token_accuracy": 0.2204607293009758,
      "num_tokens": 137901320.0,
      "step": 68055
    },
    {
      "entropy": 5.8814794540405275,
      "epoch": 5.947740977016517,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021138243512453054,
      "loss": 5.2403,
      "mean_token_accuracy": 0.21802955716848374,
      "num_tokens": 137911650.0,
      "step": 68060
    },
    {
      "entropy": 5.91586709022522,
      "epoch": 5.9481779253692215,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002113525461593595,
      "loss": 5.2226,
      "mean_token_accuracy": 0.21688087433576583,
      "num_tokens": 137920492.0,
      "step": 68065
    },
    {
      "entropy": 5.817664670944214,
      "epoch": 5.948614873721926,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021132265841498,
      "loss": 5.1746,
      "mean_token_accuracy": 0.22144295871257783,
      "num_tokens": 137931048.0,
      "step": 68070
    },
    {
      "entropy": 5.853635597229004,
      "epoch": 5.9490518220746305,
      "grad_norm": 1.078125,
      "learning_rate": 0.00021129277189196522,
      "loss": 5.1819,
      "mean_token_accuracy": 0.2219742104411125,
      "num_tokens": 137940526.0,
      "step": 68075
    },
    {
      "entropy": 5.857830333709717,
      "epoch": 5.949488770427336,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021126288659088844,
      "loss": 5.156,
      "mean_token_accuracy": 0.2197437658905983,
      "num_tokens": 137950603.0,
      "step": 68080
    },
    {
      "entropy": 5.883630800247192,
      "epoch": 5.94992571878004,
      "grad_norm": 1.125,
      "learning_rate": 0.0002112330025123229,
      "loss": 5.2095,
      "mean_token_accuracy": 0.21603257656097413,
      "num_tokens": 137960418.0,
      "step": 68085
    },
    {
      "entropy": 5.801201963424683,
      "epoch": 5.950362667132745,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002112031196568418,
      "loss": 5.1311,
      "mean_token_accuracy": 0.23095417767763138,
      "num_tokens": 137970475.0,
      "step": 68090
    },
    {
      "entropy": 5.847604608535766,
      "epoch": 5.950799615485449,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002111732380250183,
      "loss": 5.2166,
      "mean_token_accuracy": 0.22366770803928376,
      "num_tokens": 137980140.0,
      "step": 68095
    },
    {
      "entropy": 5.8654601097106935,
      "epoch": 5.951236563838155,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021114335761742554,
      "loss": 5.1911,
      "mean_token_accuracy": 0.21888307183980943,
      "num_tokens": 137990928.0,
      "step": 68100
    },
    {
      "entropy": 5.848326206207275,
      "epoch": 5.951673512190859,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021111347843463663,
      "loss": 5.1537,
      "mean_token_accuracy": 0.22126141041517258,
      "num_tokens": 138000349.0,
      "step": 68105
    },
    {
      "entropy": 5.834854030609131,
      "epoch": 5.9521104605435635,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021108360047722468,
      "loss": 5.1437,
      "mean_token_accuracy": 0.22318968474864959,
      "num_tokens": 138010735.0,
      "step": 68110
    },
    {
      "entropy": 5.804487228393555,
      "epoch": 5.952547408896269,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021105372374576276,
      "loss": 5.19,
      "mean_token_accuracy": 0.2193005010485649,
      "num_tokens": 138020475.0,
      "step": 68115
    },
    {
      "entropy": 5.828789377212525,
      "epoch": 5.952984357248973,
      "grad_norm": 1.0625,
      "learning_rate": 0.00021102384824082398,
      "loss": 5.2097,
      "mean_token_accuracy": 0.21938058137893676,
      "num_tokens": 138031706.0,
      "step": 68120
    },
    {
      "entropy": 5.884333562850952,
      "epoch": 5.953421305601678,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021099397396298126,
      "loss": 5.2732,
      "mean_token_accuracy": 0.22331578582525252,
      "num_tokens": 138042491.0,
      "step": 68125
    },
    {
      "entropy": 5.907237243652344,
      "epoch": 5.953858253954382,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021096410091280766,
      "loss": 5.246,
      "mean_token_accuracy": 0.21588537991046905,
      "num_tokens": 138052421.0,
      "step": 68130
    },
    {
      "entropy": 5.886181306838989,
      "epoch": 5.954295202307088,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021093422909087618,
      "loss": 5.1779,
      "mean_token_accuracy": 0.21882028877735138,
      "num_tokens": 138062723.0,
      "step": 68135
    },
    {
      "entropy": 5.942178106307983,
      "epoch": 5.954732150659792,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021090435849775978,
      "loss": 5.1827,
      "mean_token_accuracy": 0.2230260744690895,
      "num_tokens": 138073059.0,
      "step": 68140
    },
    {
      "entropy": 5.82693920135498,
      "epoch": 5.955169099012497,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021087448913403135,
      "loss": 5.1355,
      "mean_token_accuracy": 0.2214043468236923,
      "num_tokens": 138082837.0,
      "step": 68145
    },
    {
      "entropy": 5.757692956924439,
      "epoch": 5.955606047365201,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021084462100026376,
      "loss": 5.1209,
      "mean_token_accuracy": 0.2320299118757248,
      "num_tokens": 138091772.0,
      "step": 68150
    },
    {
      "entropy": 5.783743953704834,
      "epoch": 5.956042995717906,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00021081475409703,
      "loss": 5.1732,
      "mean_token_accuracy": 0.22502320408821105,
      "num_tokens": 138101199.0,
      "step": 68155
    },
    {
      "entropy": 5.8203675746917725,
      "epoch": 5.956479944070611,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002107848884249029,
      "loss": 5.1867,
      "mean_token_accuracy": 0.22410277426242828,
      "num_tokens": 138110444.0,
      "step": 68160
    },
    {
      "entropy": 5.91816143989563,
      "epoch": 5.956916892423315,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002107550239844553,
      "loss": 5.2259,
      "mean_token_accuracy": 0.2135966882109642,
      "num_tokens": 138120882.0,
      "step": 68165
    },
    {
      "entropy": 5.9139241695404055,
      "epoch": 5.957353840776021,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00021072516077625997,
      "loss": 5.273,
      "mean_token_accuracy": 0.21468099802732468,
      "num_tokens": 138131669.0,
      "step": 68170
    },
    {
      "entropy": 5.901111793518067,
      "epoch": 5.957790789128725,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002106952988008897,
      "loss": 5.2061,
      "mean_token_accuracy": 0.21969050467014312,
      "num_tokens": 138142261.0,
      "step": 68175
    },
    {
      "entropy": 5.813044261932373,
      "epoch": 5.95822773748143,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021066543805891736,
      "loss": 5.1478,
      "mean_token_accuracy": 0.22617272287607193,
      "num_tokens": 138153587.0,
      "step": 68180
    },
    {
      "entropy": 5.838840389251709,
      "epoch": 5.958664685834134,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002106355785509156,
      "loss": 5.1795,
      "mean_token_accuracy": 0.22306962460279464,
      "num_tokens": 138163356.0,
      "step": 68185
    },
    {
      "entropy": 5.862901306152343,
      "epoch": 5.9591016341868395,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021060572027745722,
      "loss": 5.1358,
      "mean_token_accuracy": 0.22393228709697724,
      "num_tokens": 138173577.0,
      "step": 68190
    },
    {
      "entropy": 5.87113037109375,
      "epoch": 5.959538582539544,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021057586323911483,
      "loss": 5.2551,
      "mean_token_accuracy": 0.21549834161996842,
      "num_tokens": 138183511.0,
      "step": 68195
    },
    {
      "entropy": 5.889410972595215,
      "epoch": 5.959975530892248,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021054600743646113,
      "loss": 5.2277,
      "mean_token_accuracy": 0.22637003511190415,
      "num_tokens": 138193890.0,
      "step": 68200
    },
    {
      "entropy": 5.838451147079468,
      "epoch": 5.960412479244953,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021051615287006876,
      "loss": 5.1986,
      "mean_token_accuracy": 0.21482729613780976,
      "num_tokens": 138203756.0,
      "step": 68205
    },
    {
      "entropy": 5.859114694595337,
      "epoch": 5.960849427597658,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021048629954051036,
      "loss": 5.2244,
      "mean_token_accuracy": 0.21538358628749849,
      "num_tokens": 138214354.0,
      "step": 68210
    },
    {
      "entropy": 5.902188682556153,
      "epoch": 5.961286375950363,
      "grad_norm": 1.0625,
      "learning_rate": 0.00021045644744835862,
      "loss": 5.219,
      "mean_token_accuracy": 0.22128729969263078,
      "num_tokens": 138225051.0,
      "step": 68215
    },
    {
      "entropy": 5.929259634017944,
      "epoch": 5.961723324303067,
      "grad_norm": 1.1875,
      "learning_rate": 0.000210426596594186,
      "loss": 5.2835,
      "mean_token_accuracy": 0.21438988894224167,
      "num_tokens": 138236354.0,
      "step": 68220
    },
    {
      "entropy": 5.952331590652466,
      "epoch": 5.9621602726557725,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021039674697856509,
      "loss": 5.2845,
      "mean_token_accuracy": 0.21417704075574875,
      "num_tokens": 138245663.0,
      "step": 68225
    },
    {
      "entropy": 5.77581377029419,
      "epoch": 5.962597221008477,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021036689860206843,
      "loss": 5.0464,
      "mean_token_accuracy": 0.23018243312835693,
      "num_tokens": 138254919.0,
      "step": 68230
    },
    {
      "entropy": 5.78088231086731,
      "epoch": 5.9630341693611815,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021033705146526844,
      "loss": 5.112,
      "mean_token_accuracy": 0.2243779867887497,
      "num_tokens": 138265029.0,
      "step": 68235
    },
    {
      "entropy": 5.801622533798218,
      "epoch": 5.963471117713886,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021030720556873783,
      "loss": 5.1324,
      "mean_token_accuracy": 0.23044332563877107,
      "num_tokens": 138275083.0,
      "step": 68240
    },
    {
      "entropy": 5.9229988098144535,
      "epoch": 5.963908066066591,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021027736091304883,
      "loss": 5.2564,
      "mean_token_accuracy": 0.2132118746638298,
      "num_tokens": 138285685.0,
      "step": 68245
    },
    {
      "entropy": 5.845450925827026,
      "epoch": 5.964345014419296,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021024751749877397,
      "loss": 5.1536,
      "mean_token_accuracy": 0.21896968632936478,
      "num_tokens": 138296177.0,
      "step": 68250
    },
    {
      "entropy": 5.853399467468262,
      "epoch": 5.964781962772,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021021767532648566,
      "loss": 5.1551,
      "mean_token_accuracy": 0.22544949054718016,
      "num_tokens": 138305913.0,
      "step": 68255
    },
    {
      "entropy": 5.847826623916626,
      "epoch": 5.965218911124705,
      "grad_norm": 1.078125,
      "learning_rate": 0.00021018783439675625,
      "loss": 5.1381,
      "mean_token_accuracy": 0.2263333946466446,
      "num_tokens": 138315103.0,
      "step": 68260
    },
    {
      "entropy": 5.900194501876831,
      "epoch": 5.96565585947741,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00021015799471015822,
      "loss": 5.2916,
      "mean_token_accuracy": 0.21344302147626876,
      "num_tokens": 138325366.0,
      "step": 68265
    },
    {
      "entropy": 5.803937244415283,
      "epoch": 5.9660928078301145,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021012815626726377,
      "loss": 5.1906,
      "mean_token_accuracy": 0.21819845587015152,
      "num_tokens": 138335598.0,
      "step": 68270
    },
    {
      "entropy": 5.829917764663696,
      "epoch": 5.966529756182819,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021009831906864524,
      "loss": 5.2011,
      "mean_token_accuracy": 0.21727644950151442,
      "num_tokens": 138345197.0,
      "step": 68275
    },
    {
      "entropy": 5.896704149246216,
      "epoch": 5.966966704535524,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021006848311487503,
      "loss": 5.1936,
      "mean_token_accuracy": 0.2221965104341507,
      "num_tokens": 138354967.0,
      "step": 68280
    },
    {
      "entropy": 5.91747899055481,
      "epoch": 5.967403652888229,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021003864840652526,
      "loss": 5.2861,
      "mean_token_accuracy": 0.21190920174121858,
      "num_tokens": 138364785.0,
      "step": 68285
    },
    {
      "entropy": 5.9191741943359375,
      "epoch": 5.967840601240933,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021000881494416835,
      "loss": 5.2262,
      "mean_token_accuracy": 0.22070652693510057,
      "num_tokens": 138374466.0,
      "step": 68290
    },
    {
      "entropy": 5.794999599456787,
      "epoch": 5.968277549593638,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002099789827283764,
      "loss": 5.123,
      "mean_token_accuracy": 0.2370285466313362,
      "num_tokens": 138385274.0,
      "step": 68295
    },
    {
      "entropy": 5.8415055751800535,
      "epoch": 5.968714497946343,
      "grad_norm": 1.125,
      "learning_rate": 0.00020994915175972156,
      "loss": 5.1449,
      "mean_token_accuracy": 0.22030268758535385,
      "num_tokens": 138396068.0,
      "step": 68300
    },
    {
      "entropy": 5.852690649032593,
      "epoch": 5.969151446299048,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002099193220387761,
      "loss": 5.1578,
      "mean_token_accuracy": 0.22383374273777007,
      "num_tokens": 138405544.0,
      "step": 68305
    },
    {
      "entropy": 5.876817035675049,
      "epoch": 5.969588394651752,
      "grad_norm": 1.09375,
      "learning_rate": 0.00020988949356611227,
      "loss": 5.159,
      "mean_token_accuracy": 0.21930068135261535,
      "num_tokens": 138415592.0,
      "step": 68310
    },
    {
      "entropy": 5.816133308410644,
      "epoch": 5.9700253430044565,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002098596663423019,
      "loss": 5.1303,
      "mean_token_accuracy": 0.22455657422542571,
      "num_tokens": 138425172.0,
      "step": 68315
    },
    {
      "entropy": 5.8460362434387205,
      "epoch": 5.970462291357162,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002098298403679173,
      "loss": 5.1997,
      "mean_token_accuracy": 0.21827838569879532,
      "num_tokens": 138435363.0,
      "step": 68320
    },
    {
      "entropy": 5.885623168945313,
      "epoch": 5.970899239709866,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020980001564353058,
      "loss": 5.2425,
      "mean_token_accuracy": 0.21439995914697646,
      "num_tokens": 138446571.0,
      "step": 68325
    },
    {
      "entropy": 5.8667176246643065,
      "epoch": 5.971336188062571,
      "grad_norm": 1.078125,
      "learning_rate": 0.00020977019216971365,
      "loss": 5.1825,
      "mean_token_accuracy": 0.22541691958904267,
      "num_tokens": 138456709.0,
      "step": 68330
    },
    {
      "entropy": 5.9154744148254395,
      "epoch": 5.971773136415276,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020974036994703872,
      "loss": 5.1535,
      "mean_token_accuracy": 0.2180614575743675,
      "num_tokens": 138465854.0,
      "step": 68335
    },
    {
      "entropy": 5.897468900680542,
      "epoch": 5.972210084767981,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020971054897607762,
      "loss": 5.2442,
      "mean_token_accuracy": 0.207752788066864,
      "num_tokens": 138477017.0,
      "step": 68340
    },
    {
      "entropy": 5.873090600967407,
      "epoch": 5.972647033120685,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00020968072925740233,
      "loss": 5.2093,
      "mean_token_accuracy": 0.2168806552886963,
      "num_tokens": 138487702.0,
      "step": 68345
    },
    {
      "entropy": 5.863796949386597,
      "epoch": 5.97308398147339,
      "grad_norm": 1.125,
      "learning_rate": 0.000209650910791585,
      "loss": 5.1947,
      "mean_token_accuracy": 0.21658414155244826,
      "num_tokens": 138496974.0,
      "step": 68350
    },
    {
      "entropy": 5.869671249389649,
      "epoch": 5.973520929826094,
      "grad_norm": 1.109375,
      "learning_rate": 0.00020962109357919735,
      "loss": 5.1996,
      "mean_token_accuracy": 0.2183567300438881,
      "num_tokens": 138505762.0,
      "step": 68355
    },
    {
      "entropy": 5.865725040435791,
      "epoch": 5.973957878178799,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020959127762081148,
      "loss": 5.2347,
      "mean_token_accuracy": 0.21902348101139069,
      "num_tokens": 138515547.0,
      "step": 68360
    },
    {
      "entropy": 5.902436113357544,
      "epoch": 5.974394826531504,
      "grad_norm": 1.0625,
      "learning_rate": 0.00020956146291699912,
      "loss": 5.2375,
      "mean_token_accuracy": 0.21532029956579207,
      "num_tokens": 138525717.0,
      "step": 68365
    },
    {
      "entropy": 5.883460998535156,
      "epoch": 5.974831774884208,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002095316494683322,
      "loss": 5.1715,
      "mean_token_accuracy": 0.2199222683906555,
      "num_tokens": 138534849.0,
      "step": 68370
    },
    {
      "entropy": 5.850621175765991,
      "epoch": 5.975268723236914,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020950183727538253,
      "loss": 5.1172,
      "mean_token_accuracy": 0.2283562570810318,
      "num_tokens": 138543876.0,
      "step": 68375
    },
    {
      "entropy": 5.857935190200806,
      "epoch": 5.975705671589618,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020947202633872186,
      "loss": 5.1851,
      "mean_token_accuracy": 0.21830644309520722,
      "num_tokens": 138552994.0,
      "step": 68380
    },
    {
      "entropy": 5.875469589233399,
      "epoch": 5.976142619942323,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020944221665892223,
      "loss": 5.2268,
      "mean_token_accuracy": 0.21623689830303192,
      "num_tokens": 138562178.0,
      "step": 68385
    },
    {
      "entropy": 5.842540121078491,
      "epoch": 5.976579568295028,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002094124082365551,
      "loss": 5.2544,
      "mean_token_accuracy": 0.21846123039722443,
      "num_tokens": 138572763.0,
      "step": 68390
    },
    {
      "entropy": 5.903442335128784,
      "epoch": 5.9770165166477325,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020938260107219243,
      "loss": 5.195,
      "mean_token_accuracy": 0.21956282556056977,
      "num_tokens": 138582404.0,
      "step": 68395
    },
    {
      "entropy": 5.895185279846191,
      "epoch": 5.977453465000437,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002093527951664058,
      "loss": 5.2391,
      "mean_token_accuracy": 0.2180878296494484,
      "num_tokens": 138593915.0,
      "step": 68400
    },
    {
      "entropy": 5.970258331298828,
      "epoch": 5.977890413353141,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020932299051976705,
      "loss": 5.2863,
      "mean_token_accuracy": 0.2141244813799858,
      "num_tokens": 138606055.0,
      "step": 68405
    },
    {
      "entropy": 5.8636767864227295,
      "epoch": 5.978327361705846,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020929318713284767,
      "loss": 5.1352,
      "mean_token_accuracy": 0.22192932814359664,
      "num_tokens": 138615735.0,
      "step": 68410
    },
    {
      "entropy": 5.859955978393555,
      "epoch": 5.978764310058551,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020926338500621943,
      "loss": 5.1476,
      "mean_token_accuracy": 0.2288079082965851,
      "num_tokens": 138624666.0,
      "step": 68415
    },
    {
      "entropy": 5.890016412734985,
      "epoch": 5.979201258411256,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020923358414045386,
      "loss": 5.2752,
      "mean_token_accuracy": 0.21544211804866792,
      "num_tokens": 138634428.0,
      "step": 68420
    },
    {
      "entropy": 5.833053922653198,
      "epoch": 5.97963820676396,
      "grad_norm": 1.0,
      "learning_rate": 0.0002092037845361227,
      "loss": 5.171,
      "mean_token_accuracy": 0.22124452888965607,
      "num_tokens": 138645527.0,
      "step": 68425
    },
    {
      "entropy": 5.953887271881103,
      "epoch": 5.9800751551166655,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020917398619379736,
      "loss": 5.2487,
      "mean_token_accuracy": 0.21261955052614212,
      "num_tokens": 138655730.0,
      "step": 68430
    },
    {
      "entropy": 5.886365985870361,
      "epoch": 5.98051210346937,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020914418911404948,
      "loss": 5.2088,
      "mean_token_accuracy": 0.2231580838561058,
      "num_tokens": 138666678.0,
      "step": 68435
    },
    {
      "entropy": 5.851137018203735,
      "epoch": 5.9809490518220745,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020911439329745053,
      "loss": 5.1834,
      "mean_token_accuracy": 0.22162419259548188,
      "num_tokens": 138676330.0,
      "step": 68440
    },
    {
      "entropy": 5.882750606536865,
      "epoch": 5.98138600017478,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020908459874457208,
      "loss": 5.2011,
      "mean_token_accuracy": 0.21729222685098648,
      "num_tokens": 138685969.0,
      "step": 68445
    },
    {
      "entropy": 5.926346969604492,
      "epoch": 5.981822948527484,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020905480545598554,
      "loss": 5.224,
      "mean_token_accuracy": 0.21752817332744598,
      "num_tokens": 138695828.0,
      "step": 68450
    },
    {
      "entropy": 5.9095947265625,
      "epoch": 5.982259896880189,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020902501343226238,
      "loss": 5.2626,
      "mean_token_accuracy": 0.21933350563049317,
      "num_tokens": 138704645.0,
      "step": 68455
    },
    {
      "entropy": 5.905821084976196,
      "epoch": 5.982696845232893,
      "grad_norm": 1.125,
      "learning_rate": 0.00020899522267397408,
      "loss": 5.187,
      "mean_token_accuracy": 0.2186296060681343,
      "num_tokens": 138715777.0,
      "step": 68460
    },
    {
      "entropy": 5.766638660430909,
      "epoch": 5.983133793585598,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00020896543318169194,
      "loss": 5.0698,
      "mean_token_accuracy": 0.23284002840518953,
      "num_tokens": 138727413.0,
      "step": 68465
    },
    {
      "entropy": 5.853582525253296,
      "epoch": 5.983570741938303,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020893564495598742,
      "loss": 5.2022,
      "mean_token_accuracy": 0.21967993974685668,
      "num_tokens": 138737439.0,
      "step": 68470
    },
    {
      "entropy": 5.790260124206543,
      "epoch": 5.9840076902910075,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002089058579974319,
      "loss": 5.1372,
      "mean_token_accuracy": 0.22368603199720383,
      "num_tokens": 138747099.0,
      "step": 68475
    },
    {
      "entropy": 5.889055490493774,
      "epoch": 5.984444638643712,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020887607230659657,
      "loss": 5.2315,
      "mean_token_accuracy": 0.21388580948114394,
      "num_tokens": 138757243.0,
      "step": 68480
    },
    {
      "entropy": 5.887865114212036,
      "epoch": 5.984881586996417,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002088462878840528,
      "loss": 5.1655,
      "mean_token_accuracy": 0.21917466819286346,
      "num_tokens": 138768040.0,
      "step": 68485
    },
    {
      "entropy": 5.890075588226319,
      "epoch": 5.985318535349122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020881650473037201,
      "loss": 5.1528,
      "mean_token_accuracy": 0.21986482292413712,
      "num_tokens": 138777931.0,
      "step": 68490
    },
    {
      "entropy": 5.89318437576294,
      "epoch": 5.985755483701826,
      "grad_norm": 1.09375,
      "learning_rate": 0.00020878672284612527,
      "loss": 5.2044,
      "mean_token_accuracy": 0.21758547574281692,
      "num_tokens": 138788988.0,
      "step": 68495
    },
    {
      "entropy": 5.887037563323974,
      "epoch": 5.986192432054531,
      "grad_norm": 1.125,
      "learning_rate": 0.00020875694223188384,
      "loss": 5.2154,
      "mean_token_accuracy": 0.2194177970290184,
      "num_tokens": 138799914.0,
      "step": 68500
    },
    {
      "entropy": 5.795401287078858,
      "epoch": 5.986629380407236,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020872716288821902,
      "loss": 5.1104,
      "mean_token_accuracy": 0.2232319638133049,
      "num_tokens": 138809372.0,
      "step": 68505
    },
    {
      "entropy": 5.872664546966552,
      "epoch": 5.987066328759941,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020869738481570194,
      "loss": 5.2562,
      "mean_token_accuracy": 0.21384696811437606,
      "num_tokens": 138819466.0,
      "step": 68510
    },
    {
      "entropy": 5.859951782226562,
      "epoch": 5.987503277112645,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020866760801490375,
      "loss": 5.2091,
      "mean_token_accuracy": 0.2163074016571045,
      "num_tokens": 138829517.0,
      "step": 68515
    },
    {
      "entropy": 5.860038757324219,
      "epoch": 5.9879402254653495,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00020863783248639562,
      "loss": 5.2333,
      "mean_token_accuracy": 0.21955876052379608,
      "num_tokens": 138839883.0,
      "step": 68520
    },
    {
      "entropy": 5.904307842254639,
      "epoch": 5.988377173818055,
      "grad_norm": 1.078125,
      "learning_rate": 0.00020860805823074864,
      "loss": 5.2325,
      "mean_token_accuracy": 0.2137469083070755,
      "num_tokens": 138849906.0,
      "step": 68525
    },
    {
      "entropy": 5.872119665145874,
      "epoch": 5.988814122170759,
      "grad_norm": 1.09375,
      "learning_rate": 0.00020857828524853384,
      "loss": 5.1963,
      "mean_token_accuracy": 0.2215694382786751,
      "num_tokens": 138860567.0,
      "step": 68530
    },
    {
      "entropy": 5.828563404083252,
      "epoch": 5.989251070523464,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020854851354032235,
      "loss": 5.1675,
      "mean_token_accuracy": 0.2206422746181488,
      "num_tokens": 138870679.0,
      "step": 68535
    },
    {
      "entropy": 5.900813961029053,
      "epoch": 5.989688018876169,
      "grad_norm": 1.0625,
      "learning_rate": 0.00020851874310668522,
      "loss": 5.2523,
      "mean_token_accuracy": 0.21122182309627532,
      "num_tokens": 138880721.0,
      "step": 68540
    },
    {
      "entropy": 5.785097885131836,
      "epoch": 5.990124967228874,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00020848897394819344,
      "loss": 5.1061,
      "mean_token_accuracy": 0.2324810042977333,
      "num_tokens": 138890554.0,
      "step": 68545
    },
    {
      "entropy": 5.903443145751953,
      "epoch": 5.990561915581578,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020845920606541796,
      "loss": 5.1913,
      "mean_token_accuracy": 0.22036555260419846,
      "num_tokens": 138900717.0,
      "step": 68550
    },
    {
      "entropy": 5.847182416915894,
      "epoch": 5.990998863934283,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020842943945892977,
      "loss": 5.1482,
      "mean_token_accuracy": 0.2232143521308899,
      "num_tokens": 138910210.0,
      "step": 68555
    },
    {
      "entropy": 5.840483665466309,
      "epoch": 5.991435812286988,
      "grad_norm": 1.125,
      "learning_rate": 0.00020839967412929984,
      "loss": 5.2264,
      "mean_token_accuracy": 0.21732279658317566,
      "num_tokens": 138920425.0,
      "step": 68560
    },
    {
      "entropy": 5.874788045883179,
      "epoch": 5.991872760639692,
      "grad_norm": 1.234375,
      "learning_rate": 0.000208369910077099,
      "loss": 5.2243,
      "mean_token_accuracy": 0.21547735780477523,
      "num_tokens": 138931165.0,
      "step": 68565
    },
    {
      "entropy": 5.918040323257446,
      "epoch": 5.992309708992397,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002083401473028983,
      "loss": 5.1653,
      "mean_token_accuracy": 0.22409879714250563,
      "num_tokens": 138940469.0,
      "step": 68570
    },
    {
      "entropy": 5.88107705116272,
      "epoch": 5.992746657345101,
      "grad_norm": 1.109375,
      "learning_rate": 0.00020831038580726847,
      "loss": 5.1962,
      "mean_token_accuracy": 0.22720667123794555,
      "num_tokens": 138951064.0,
      "step": 68575
    },
    {
      "entropy": 5.751532030105591,
      "epoch": 5.993183605697807,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00020828062559078033,
      "loss": 5.0616,
      "mean_token_accuracy": 0.23857394158840178,
      "num_tokens": 138961855.0,
      "step": 68580
    },
    {
      "entropy": 5.827530288696289,
      "epoch": 5.993620554050511,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020825086665400477,
      "loss": 5.2001,
      "mean_token_accuracy": 0.21865104585886003,
      "num_tokens": 138972139.0,
      "step": 68585
    },
    {
      "entropy": 5.887578296661377,
      "epoch": 5.994057502403216,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020822110899751262,
      "loss": 5.2614,
      "mean_token_accuracy": 0.2127714604139328,
      "num_tokens": 138981947.0,
      "step": 68590
    },
    {
      "entropy": 5.918397521972656,
      "epoch": 5.994494450755921,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020819135262187456,
      "loss": 5.1557,
      "mean_token_accuracy": 0.21842389106750487,
      "num_tokens": 138992776.0,
      "step": 68595
    },
    {
      "entropy": 5.885498571395874,
      "epoch": 5.9949313991086255,
      "grad_norm": 1.125,
      "learning_rate": 0.00020816159752766128,
      "loss": 5.1899,
      "mean_token_accuracy": 0.21871695518493653,
      "num_tokens": 139003421.0,
      "step": 68600
    },
    {
      "entropy": 5.855163621902466,
      "epoch": 5.99536834746133,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020813184371544364,
      "loss": 5.2054,
      "mean_token_accuracy": 0.21190468817949296,
      "num_tokens": 139013630.0,
      "step": 68605
    },
    {
      "entropy": 5.839390468597412,
      "epoch": 5.995805295814034,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002081020911857923,
      "loss": 5.2267,
      "mean_token_accuracy": 0.21785213351249694,
      "num_tokens": 139024229.0,
      "step": 68610
    },
    {
      "entropy": 5.856097936630249,
      "epoch": 5.99624224416674,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020807233993927794,
      "loss": 5.2029,
      "mean_token_accuracy": 0.21667774319648742,
      "num_tokens": 139034447.0,
      "step": 68615
    },
    {
      "entropy": 5.875517416000366,
      "epoch": 5.996679192519444,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020804258997647113,
      "loss": 5.1555,
      "mean_token_accuracy": 0.21855561435222626,
      "num_tokens": 139043916.0,
      "step": 68620
    },
    {
      "entropy": 5.822917938232422,
      "epoch": 5.997116140872149,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002080128412979425,
      "loss": 5.1299,
      "mean_token_accuracy": 0.22732212543487548,
      "num_tokens": 139054071.0,
      "step": 68625
    },
    {
      "entropy": 5.85383882522583,
      "epoch": 5.997553089224853,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020798309390426267,
      "loss": 5.1602,
      "mean_token_accuracy": 0.22429479658603668,
      "num_tokens": 139064105.0,
      "step": 68630
    },
    {
      "entropy": 5.846964550018311,
      "epoch": 5.9979900375775586,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020795334779600224,
      "loss": 5.1627,
      "mean_token_accuracy": 0.22342219948768616,
      "num_tokens": 139074160.0,
      "step": 68635
    },
    {
      "entropy": 5.828006601333618,
      "epoch": 5.998426985930263,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020792360297373176,
      "loss": 5.1301,
      "mean_token_accuracy": 0.2319851502776146,
      "num_tokens": 139083223.0,
      "step": 68640
    },
    {
      "entropy": 5.899991893768311,
      "epoch": 5.9988639342829675,
      "grad_norm": 1.046875,
      "learning_rate": 0.00020789385943802168,
      "loss": 5.2424,
      "mean_token_accuracy": 0.21569312512874603,
      "num_tokens": 139093242.0,
      "step": 68645
    },
    {
      "entropy": 5.87273941040039,
      "epoch": 5.999300882635673,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002078641171894426,
      "loss": 5.2089,
      "mean_token_accuracy": 0.21761928349733353,
      "num_tokens": 139103789.0,
      "step": 68650
    },
    {
      "entropy": 5.830122232437134,
      "epoch": 5.999737830988377,
      "grad_norm": 1.078125,
      "learning_rate": 0.00020783437622856483,
      "loss": 5.1758,
      "mean_token_accuracy": 0.21992071568965912,
      "num_tokens": 139114172.0,
      "step": 68655
    },
    {
      "entropy": 5.8171882152557375,
      "epoch": 6.000174779341082,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000207804636555959,
      "loss": 5.1061,
      "mean_token_accuracy": 0.23139904886484147,
      "num_tokens": 139123521.0,
      "step": 68660
    },
    {
      "entropy": 5.928965044021607,
      "epoch": 6.000611727693786,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020777489817219542,
      "loss": 5.1932,
      "mean_token_accuracy": 0.2215207502245903,
      "num_tokens": 139132090.0,
      "step": 68665
    },
    {
      "entropy": 5.881282567977905,
      "epoch": 6.001048676046492,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020774516107784451,
      "loss": 5.136,
      "mean_token_accuracy": 0.22554513514041902,
      "num_tokens": 139141818.0,
      "step": 68670
    },
    {
      "entropy": 5.825534152984619,
      "epoch": 6.001485624399196,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002077154252734767,
      "loss": 5.0819,
      "mean_token_accuracy": 0.22914658039808272,
      "num_tokens": 139152013.0,
      "step": 68675
    },
    {
      "entropy": 5.853633975982666,
      "epoch": 6.0019225727519006,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020768569075966227,
      "loss": 5.2098,
      "mean_token_accuracy": 0.22343087196350098,
      "num_tokens": 139162860.0,
      "step": 68680
    },
    {
      "entropy": 5.821537065505981,
      "epoch": 6.002359521104605,
      "grad_norm": 1.078125,
      "learning_rate": 0.00020765595753697154,
      "loss": 5.1066,
      "mean_token_accuracy": 0.2223280504345894,
      "num_tokens": 139174301.0,
      "step": 68685
    },
    {
      "entropy": 5.845666837692261,
      "epoch": 6.00279646945731,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020762622560597483,
      "loss": 5.1576,
      "mean_token_accuracy": 0.22419662326574324,
      "num_tokens": 139184589.0,
      "step": 68690
    },
    {
      "entropy": 5.810250854492187,
      "epoch": 6.003233417810015,
      "grad_norm": 1.109375,
      "learning_rate": 0.00020759649496724243,
      "loss": 5.1118,
      "mean_token_accuracy": 0.2291295662522316,
      "num_tokens": 139194931.0,
      "step": 68695
    },
    {
      "entropy": 5.863856267929077,
      "epoch": 6.003670366162719,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020756676562134457,
      "loss": 5.211,
      "mean_token_accuracy": 0.220402529835701,
      "num_tokens": 139205714.0,
      "step": 68700
    },
    {
      "entropy": 5.822979927062988,
      "epoch": 6.004107314515425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020753703756885145,
      "loss": 5.061,
      "mean_token_accuracy": 0.2375776618719101,
      "num_tokens": 139215882.0,
      "step": 68705
    },
    {
      "entropy": 5.861194753646851,
      "epoch": 6.004544262868129,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020750731081033337,
      "loss": 5.0926,
      "mean_token_accuracy": 0.231016843020916,
      "num_tokens": 139224365.0,
      "step": 68710
    },
    {
      "entropy": 5.765335273742676,
      "epoch": 6.004981211220834,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002074775853463604,
      "loss": 5.0751,
      "mean_token_accuracy": 0.22953174859285355,
      "num_tokens": 139235221.0,
      "step": 68715
    },
    {
      "entropy": 5.858967065811157,
      "epoch": 6.005418159573538,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002074478611775027,
      "loss": 5.1429,
      "mean_token_accuracy": 0.22376932203769684,
      "num_tokens": 139245351.0,
      "step": 68720
    },
    {
      "entropy": 5.848507642745972,
      "epoch": 6.005855107926243,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00020741813830433043,
      "loss": 5.1267,
      "mean_token_accuracy": 0.2252267926931381,
      "num_tokens": 139255994.0,
      "step": 68725
    },
    {
      "entropy": 5.90514612197876,
      "epoch": 6.006292056278948,
      "grad_norm": 1.25,
      "learning_rate": 0.0002073884167274137,
      "loss": 5.1781,
      "mean_token_accuracy": 0.21689599454402925,
      "num_tokens": 139265569.0,
      "step": 68730
    },
    {
      "entropy": 5.862273502349853,
      "epoch": 6.006729004631652,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020735869644732246,
      "loss": 5.1559,
      "mean_token_accuracy": 0.21915487200021744,
      "num_tokens": 139275441.0,
      "step": 68735
    },
    {
      "entropy": 5.859245300292969,
      "epoch": 6.007165952984357,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00020732897746462698,
      "loss": 5.1555,
      "mean_token_accuracy": 0.22303904592990875,
      "num_tokens": 139287013.0,
      "step": 68740
    },
    {
      "entropy": 5.902216815948487,
      "epoch": 6.007602901337062,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020729925977989712,
      "loss": 5.2063,
      "mean_token_accuracy": 0.22155265808105468,
      "num_tokens": 139297355.0,
      "step": 68745
    },
    {
      "entropy": 5.868480157852173,
      "epoch": 6.008039849689767,
      "grad_norm": 1.125,
      "learning_rate": 0.00020726954339370302,
      "loss": 5.1637,
      "mean_token_accuracy": 0.21921158730983734,
      "num_tokens": 139306944.0,
      "step": 68750
    },
    {
      "entropy": 5.870715713500976,
      "epoch": 6.008476798042471,
      "grad_norm": 1.09375,
      "learning_rate": 0.00020723982830661453,
      "loss": 5.1641,
      "mean_token_accuracy": 0.2206232503056526,
      "num_tokens": 139317864.0,
      "step": 68755
    },
    {
      "entropy": 5.81795973777771,
      "epoch": 6.0089137463951765,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020721011451920157,
      "loss": 5.0866,
      "mean_token_accuracy": 0.2319038227200508,
      "num_tokens": 139328007.0,
      "step": 68760
    },
    {
      "entropy": 5.863208150863647,
      "epoch": 6.009350694747881,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002071804020320342,
      "loss": 5.1728,
      "mean_token_accuracy": 0.22032346129417418,
      "num_tokens": 139338263.0,
      "step": 68765
    },
    {
      "entropy": 5.906293869018555,
      "epoch": 6.009787643100585,
      "grad_norm": 1.046875,
      "learning_rate": 0.00020715069084568222,
      "loss": 5.2027,
      "mean_token_accuracy": 0.21909113973379135,
      "num_tokens": 139348941.0,
      "step": 68770
    },
    {
      "entropy": 5.841446161270142,
      "epoch": 6.01022459145329,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020712098096071563,
      "loss": 5.0975,
      "mean_token_accuracy": 0.23943955153226854,
      "num_tokens": 139358792.0,
      "step": 68775
    },
    {
      "entropy": 5.915570163726807,
      "epoch": 6.010661539805995,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002070912723777041,
      "loss": 5.1874,
      "mean_token_accuracy": 0.2165796786546707,
      "num_tokens": 139368228.0,
      "step": 68780
    },
    {
      "entropy": 5.8101723194122314,
      "epoch": 6.0110984881587,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020706156509721753,
      "loss": 5.0517,
      "mean_token_accuracy": 0.23421007096767427,
      "num_tokens": 139377293.0,
      "step": 68785
    },
    {
      "entropy": 5.864528942108154,
      "epoch": 6.011535436511404,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002070318591198258,
      "loss": 5.1239,
      "mean_token_accuracy": 0.22578848749399186,
      "num_tokens": 139388445.0,
      "step": 68790
    },
    {
      "entropy": 5.8736762523651125,
      "epoch": 6.011972384864109,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002070021544460986,
      "loss": 5.154,
      "mean_token_accuracy": 0.22663495242595671,
      "num_tokens": 139398111.0,
      "step": 68795
    },
    {
      "entropy": 5.8598462581634525,
      "epoch": 6.012409333216814,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002069724510766058,
      "loss": 5.2056,
      "mean_token_accuracy": 0.22250710874795915,
      "num_tokens": 139408076.0,
      "step": 68800
    },
    {
      "entropy": 5.972533702850342,
      "epoch": 6.0128462815695185,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020694274901191694,
      "loss": 5.2498,
      "mean_token_accuracy": 0.21274533718824387,
      "num_tokens": 139417887.0,
      "step": 68805
    },
    {
      "entropy": 5.919769906997681,
      "epoch": 6.013283229922223,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020691304825260176,
      "loss": 5.1538,
      "mean_token_accuracy": 0.22063722163438798,
      "num_tokens": 139428655.0,
      "step": 68810
    },
    {
      "entropy": 5.908668470382691,
      "epoch": 6.013720178274928,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002068833487992301,
      "loss": 5.1839,
      "mean_token_accuracy": 0.22350462079048156,
      "num_tokens": 139438629.0,
      "step": 68815
    },
    {
      "entropy": 5.801992416381836,
      "epoch": 6.014157126627633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020685365065237145,
      "loss": 5.148,
      "mean_token_accuracy": 0.22942440509796141,
      "num_tokens": 139448939.0,
      "step": 68820
    },
    {
      "entropy": 5.808496189117432,
      "epoch": 6.014594074980337,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020682395381259555,
      "loss": 5.1133,
      "mean_token_accuracy": 0.2242097809910774,
      "num_tokens": 139459035.0,
      "step": 68825
    },
    {
      "entropy": 5.893164587020874,
      "epoch": 6.015031023333042,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020679425828047188,
      "loss": 5.1453,
      "mean_token_accuracy": 0.21851285994052888,
      "num_tokens": 139469398.0,
      "step": 68830
    },
    {
      "entropy": 5.699171113967895,
      "epoch": 6.015467971685747,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020676456405657008,
      "loss": 4.9701,
      "mean_token_accuracy": 0.24790160953998566,
      "num_tokens": 139479627.0,
      "step": 68835
    },
    {
      "entropy": 5.810693168640137,
      "epoch": 6.015904920038452,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020673487114145973,
      "loss": 5.0767,
      "mean_token_accuracy": 0.22668546736240386,
      "num_tokens": 139488992.0,
      "step": 68840
    },
    {
      "entropy": 5.84248046875,
      "epoch": 6.016341868391156,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020670517953571023,
      "loss": 5.1286,
      "mean_token_accuracy": 0.22632875442504882,
      "num_tokens": 139499486.0,
      "step": 68845
    },
    {
      "entropy": 5.879194831848144,
      "epoch": 6.0167788167438605,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020667548923989127,
      "loss": 5.1978,
      "mean_token_accuracy": 0.218325175344944,
      "num_tokens": 139509494.0,
      "step": 68850
    },
    {
      "entropy": 5.900369548797608,
      "epoch": 6.017215765096566,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002066458002545722,
      "loss": 5.1954,
      "mean_token_accuracy": 0.21842906922101973,
      "num_tokens": 139520895.0,
      "step": 68855
    },
    {
      "entropy": 5.848164892196655,
      "epoch": 6.01765271344927,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020661611258032247,
      "loss": 5.1702,
      "mean_token_accuracy": 0.23106627613306047,
      "num_tokens": 139531089.0,
      "step": 68860
    },
    {
      "entropy": 5.960538530349732,
      "epoch": 6.018089661801975,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020658642621771157,
      "loss": 5.2463,
      "mean_token_accuracy": 0.2087053492665291,
      "num_tokens": 139541418.0,
      "step": 68865
    },
    {
      "entropy": 5.79078254699707,
      "epoch": 6.01852661015468,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020655674116730883,
      "loss": 5.0575,
      "mean_token_accuracy": 0.22555909305810928,
      "num_tokens": 139551028.0,
      "step": 68870
    },
    {
      "entropy": 5.873870086669922,
      "epoch": 6.018963558507385,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020652705742968375,
      "loss": 5.1275,
      "mean_token_accuracy": 0.22494687288999557,
      "num_tokens": 139560620.0,
      "step": 68875
    },
    {
      "entropy": 5.807058382034302,
      "epoch": 6.019400506860089,
      "grad_norm": 1.25,
      "learning_rate": 0.00020649737500540548,
      "loss": 5.1483,
      "mean_token_accuracy": 0.22664466947317125,
      "num_tokens": 139570011.0,
      "step": 68880
    },
    {
      "entropy": 5.809964990615844,
      "epoch": 6.019837455212794,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002064676938950435,
      "loss": 5.1048,
      "mean_token_accuracy": 0.2310781419277191,
      "num_tokens": 139579453.0,
      "step": 68885
    },
    {
      "entropy": 5.814675521850586,
      "epoch": 6.020274403565499,
      "grad_norm": 1.046875,
      "learning_rate": 0.00020643801409916702,
      "loss": 5.1784,
      "mean_token_accuracy": 0.22179269641637803,
      "num_tokens": 139591903.0,
      "step": 68890
    },
    {
      "entropy": 5.865139818191528,
      "epoch": 6.020711351918203,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020640833561834545,
      "loss": 5.16,
      "mean_token_accuracy": 0.218650783598423,
      "num_tokens": 139602049.0,
      "step": 68895
    },
    {
      "entropy": 5.864307451248169,
      "epoch": 6.021148300270908,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020637865845314786,
      "loss": 5.12,
      "mean_token_accuracy": 0.22113568037748338,
      "num_tokens": 139612182.0,
      "step": 68900
    },
    {
      "entropy": 5.912632608413697,
      "epoch": 6.021585248623612,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00020634898260414358,
      "loss": 5.3085,
      "mean_token_accuracy": 0.22068902254104614,
      "num_tokens": 139623024.0,
      "step": 68905
    },
    {
      "entropy": 5.861228942871094,
      "epoch": 6.022022196976318,
      "grad_norm": 1.078125,
      "learning_rate": 0.00020631930807190176,
      "loss": 5.1559,
      "mean_token_accuracy": 0.2181338220834732,
      "num_tokens": 139633382.0,
      "step": 68910
    },
    {
      "entropy": 5.8187438488006595,
      "epoch": 6.022459145329022,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002062896348569916,
      "loss": 5.112,
      "mean_token_accuracy": 0.22493701726198195,
      "num_tokens": 139643997.0,
      "step": 68915
    },
    {
      "entropy": 5.81480040550232,
      "epoch": 6.022896093681727,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020625996295998235,
      "loss": 5.0832,
      "mean_token_accuracy": 0.23058861792087554,
      "num_tokens": 139654183.0,
      "step": 68920
    },
    {
      "entropy": 5.910806798934937,
      "epoch": 6.023333042034431,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020623029238144293,
      "loss": 5.2207,
      "mean_token_accuracy": 0.21835335493087768,
      "num_tokens": 139663896.0,
      "step": 68925
    },
    {
      "entropy": 5.906390714645386,
      "epoch": 6.0237699903871365,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020620062312194255,
      "loss": 5.2168,
      "mean_token_accuracy": 0.21518912613391877,
      "num_tokens": 139674029.0,
      "step": 68930
    },
    {
      "entropy": 5.853833293914795,
      "epoch": 6.024206938739841,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020617095518205025,
      "loss": 5.2101,
      "mean_token_accuracy": 0.22251973748207093,
      "num_tokens": 139684004.0,
      "step": 68935
    },
    {
      "entropy": 5.908812952041626,
      "epoch": 6.024643887092545,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002061412885623351,
      "loss": 5.212,
      "mean_token_accuracy": 0.21729241758584977,
      "num_tokens": 139693565.0,
      "step": 68940
    },
    {
      "entropy": 5.937834358215332,
      "epoch": 6.025080835445251,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002061116232633662,
      "loss": 5.2014,
      "mean_token_accuracy": 0.22634048908948898,
      "num_tokens": 139704107.0,
      "step": 68945
    },
    {
      "entropy": 5.88965482711792,
      "epoch": 6.025517783797955,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020608195928571238,
      "loss": 5.2134,
      "mean_token_accuracy": 0.21788131147623063,
      "num_tokens": 139713583.0,
      "step": 68950
    },
    {
      "entropy": 5.851122522354126,
      "epoch": 6.02595473215066,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020605229662994275,
      "loss": 5.1817,
      "mean_token_accuracy": 0.2284438893198967,
      "num_tokens": 139725321.0,
      "step": 68955
    },
    {
      "entropy": 5.818893718719482,
      "epoch": 6.026391680503364,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020602263529662614,
      "loss": 5.0717,
      "mean_token_accuracy": 0.23283518701791764,
      "num_tokens": 139736208.0,
      "step": 68960
    },
    {
      "entropy": 5.776533031463623,
      "epoch": 6.0268286288560695,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002059929752863315,
      "loss": 5.1208,
      "mean_token_accuracy": 0.2231415629386902,
      "num_tokens": 139747202.0,
      "step": 68965
    },
    {
      "entropy": 5.79691367149353,
      "epoch": 6.027265577208774,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020596331659962787,
      "loss": 5.0585,
      "mean_token_accuracy": 0.2313605234026909,
      "num_tokens": 139756918.0,
      "step": 68970
    },
    {
      "entropy": 5.835877227783203,
      "epoch": 6.0277025255614785,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00020593365923708392,
      "loss": 5.0715,
      "mean_token_accuracy": 0.23277323096990585,
      "num_tokens": 139766523.0,
      "step": 68975
    },
    {
      "entropy": 5.803523683547974,
      "epoch": 6.028139473914183,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020590400319926862,
      "loss": 5.0703,
      "mean_token_accuracy": 0.2233869969844818,
      "num_tokens": 139776571.0,
      "step": 68980
    },
    {
      "entropy": 5.903566789627075,
      "epoch": 6.028576422266888,
      "grad_norm": 1.125,
      "learning_rate": 0.00020587434848675067,
      "loss": 5.181,
      "mean_token_accuracy": 0.2251462459564209,
      "num_tokens": 139786847.0,
      "step": 68985
    },
    {
      "entropy": 5.819308757781982,
      "epoch": 6.029013370619593,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020584469510009896,
      "loss": 5.1097,
      "mean_token_accuracy": 0.2243235781788826,
      "num_tokens": 139797985.0,
      "step": 68990
    },
    {
      "entropy": 5.888493967056275,
      "epoch": 6.029450318972297,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002058150430398823,
      "loss": 5.179,
      "mean_token_accuracy": 0.21977950930595397,
      "num_tokens": 139809207.0,
      "step": 68995
    },
    {
      "entropy": 5.90876259803772,
      "epoch": 6.029887267325003,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020578539230666938,
      "loss": 5.1477,
      "mean_token_accuracy": 0.22668506503105162,
      "num_tokens": 139819155.0,
      "step": 69000
    },
    {
      "epoch": 6.029887267325003,
      "eval_entropy": 5.548344746196608,
      "eval_loss": 5.272055625915527,
      "eval_mean_token_accuracy": 0.2243084922490868,
      "eval_num_tokens": 139819155.0,
      "eval_runtime": 27.9056,
      "eval_samples_per_second": 1326.401,
      "eval_steps_per_second": 165.809,
      "step": 69000
    },
    {
      "entropy": 5.874105453491211,
      "epoch": 6.030324215677707,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020575574290102882,
      "loss": 5.238,
      "mean_token_accuracy": 0.212441286444664,
      "num_tokens": 139829483.0,
      "step": 69005
    },
    {
      "entropy": 5.852885389328003,
      "epoch": 6.0307611640304115,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002057260948235294,
      "loss": 5.1567,
      "mean_token_accuracy": 0.22455537617206572,
      "num_tokens": 139839512.0,
      "step": 69010
    },
    {
      "entropy": 5.85319881439209,
      "epoch": 6.031198112383116,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002056964480747398,
      "loss": 5.1306,
      "mean_token_accuracy": 0.22649106085300447,
      "num_tokens": 139849320.0,
      "step": 69015
    },
    {
      "entropy": 5.802697229385376,
      "epoch": 6.031635060735821,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020566680265522868,
      "loss": 5.127,
      "mean_token_accuracy": 0.22101041227579116,
      "num_tokens": 139858813.0,
      "step": 69020
    },
    {
      "entropy": 5.824394989013672,
      "epoch": 6.032072009088526,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002056371585655646,
      "loss": 5.1264,
      "mean_token_accuracy": 0.22829947769641876,
      "num_tokens": 139868302.0,
      "step": 69025
    },
    {
      "entropy": 5.799734258651734,
      "epoch": 6.03250895744123,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002056075158063161,
      "loss": 5.1071,
      "mean_token_accuracy": 0.23372469544410707,
      "num_tokens": 139877838.0,
      "step": 69030
    },
    {
      "entropy": 5.786713552474976,
      "epoch": 6.032945905793935,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020557787437805187,
      "loss": 5.0894,
      "mean_token_accuracy": 0.2301474094390869,
      "num_tokens": 139887249.0,
      "step": 69035
    },
    {
      "entropy": 5.764718008041382,
      "epoch": 6.03338285414664,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020554823428134034,
      "loss": 5.1233,
      "mean_token_accuracy": 0.2268841505050659,
      "num_tokens": 139897382.0,
      "step": 69040
    },
    {
      "entropy": 5.813353490829468,
      "epoch": 6.033819802499345,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002055185955167501,
      "loss": 5.0443,
      "mean_token_accuracy": 0.22591757625341416,
      "num_tokens": 139907084.0,
      "step": 69045
    },
    {
      "entropy": 5.869820070266724,
      "epoch": 6.034256750852049,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020548895808484964,
      "loss": 5.1874,
      "mean_token_accuracy": 0.22644759267568587,
      "num_tokens": 139918017.0,
      "step": 69050
    },
    {
      "entropy": 5.851360893249511,
      "epoch": 6.034693699204754,
      "grad_norm": 1.125,
      "learning_rate": 0.00020545932198620732,
      "loss": 5.1663,
      "mean_token_accuracy": 0.22248625308275222,
      "num_tokens": 139928415.0,
      "step": 69055
    },
    {
      "entropy": 5.8214044094085695,
      "epoch": 6.035130647557459,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020542968722139166,
      "loss": 4.9814,
      "mean_token_accuracy": 0.23845770210027695,
      "num_tokens": 139937761.0,
      "step": 69060
    },
    {
      "entropy": 5.7918657779693605,
      "epoch": 6.035567595910163,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020540005379097109,
      "loss": 5.1375,
      "mean_token_accuracy": 0.22321654111146927,
      "num_tokens": 139947823.0,
      "step": 69065
    },
    {
      "entropy": 5.786837434768676,
      "epoch": 6.036004544262868,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020537042169551395,
      "loss": 5.1551,
      "mean_token_accuracy": 0.22967029213905335,
      "num_tokens": 139959013.0,
      "step": 69070
    },
    {
      "entropy": 5.81768946647644,
      "epoch": 6.036441492615573,
      "grad_norm": 1.125,
      "learning_rate": 0.0002053407909355886,
      "loss": 5.0463,
      "mean_token_accuracy": 0.23455275297164918,
      "num_tokens": 139968658.0,
      "step": 69075
    },
    {
      "entropy": 5.785214757919311,
      "epoch": 6.036878440968278,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020531116151176337,
      "loss": 5.0299,
      "mean_token_accuracy": 0.2344844877719879,
      "num_tokens": 139978636.0,
      "step": 69080
    },
    {
      "entropy": 5.877297830581665,
      "epoch": 6.037315389320982,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020528153342460654,
      "loss": 5.1478,
      "mean_token_accuracy": 0.22364976406097412,
      "num_tokens": 139989037.0,
      "step": 69085
    },
    {
      "entropy": 5.901794338226319,
      "epoch": 6.037752337673687,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020525190667468646,
      "loss": 5.1581,
      "mean_token_accuracy": 0.2203919470310211,
      "num_tokens": 139998865.0,
      "step": 69090
    },
    {
      "entropy": 5.827345323562622,
      "epoch": 6.038189286026392,
      "grad_norm": 1.25,
      "learning_rate": 0.00020522228126257136,
      "loss": 5.0992,
      "mean_token_accuracy": 0.22853575944900512,
      "num_tokens": 140008719.0,
      "step": 69095
    },
    {
      "entropy": 5.826257371902466,
      "epoch": 6.038626234379096,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002051926571888294,
      "loss": 5.2117,
      "mean_token_accuracy": 0.22132592350244523,
      "num_tokens": 140018463.0,
      "step": 69100
    },
    {
      "entropy": 5.767529296875,
      "epoch": 6.039063182731801,
      "grad_norm": 1.21875,
      "learning_rate": 0.000205163034454029,
      "loss": 5.0754,
      "mean_token_accuracy": 0.22659126818180084,
      "num_tokens": 140029191.0,
      "step": 69105
    },
    {
      "entropy": 5.878107595443725,
      "epoch": 6.039500131084506,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020513341305873812,
      "loss": 5.153,
      "mean_token_accuracy": 0.22594546526670456,
      "num_tokens": 140039308.0,
      "step": 69110
    },
    {
      "entropy": 5.902524709701538,
      "epoch": 6.039937079437211,
      "grad_norm": 1.125,
      "learning_rate": 0.000205103793003525,
      "loss": 5.1248,
      "mean_token_accuracy": 0.22063422352075576,
      "num_tokens": 140049861.0,
      "step": 69115
    },
    {
      "entropy": 5.875771141052246,
      "epoch": 6.040374027789915,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002050741742889577,
      "loss": 5.1457,
      "mean_token_accuracy": 0.22553696781396865,
      "num_tokens": 140061048.0,
      "step": 69120
    },
    {
      "entropy": 5.8495965003967285,
      "epoch": 6.04081097614262,
      "grad_norm": 1.125,
      "learning_rate": 0.0002050445569156044,
      "loss": 5.2078,
      "mean_token_accuracy": 0.22499210089445115,
      "num_tokens": 140071745.0,
      "step": 69125
    },
    {
      "entropy": 5.852827835083008,
      "epoch": 6.041247924495325,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020501494088403328,
      "loss": 5.2218,
      "mean_token_accuracy": 0.21581353545188903,
      "num_tokens": 140081258.0,
      "step": 69130
    },
    {
      "entropy": 5.877460145950318,
      "epoch": 6.0416848728480295,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020498532619481213,
      "loss": 5.1594,
      "mean_token_accuracy": 0.2259945496916771,
      "num_tokens": 140090794.0,
      "step": 69135
    },
    {
      "entropy": 5.811429500579834,
      "epoch": 6.042121821200734,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002049557128485091,
      "loss": 5.1033,
      "mean_token_accuracy": 0.22755440771579744,
      "num_tokens": 140101395.0,
      "step": 69140
    },
    {
      "entropy": 5.830364513397217,
      "epoch": 6.042558769553438,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002049261008456923,
      "loss": 5.1067,
      "mean_token_accuracy": 0.22228345423936843,
      "num_tokens": 140111175.0,
      "step": 69145
    },
    {
      "entropy": 5.881429815292359,
      "epoch": 6.042995717906144,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020489649018692952,
      "loss": 5.1481,
      "mean_token_accuracy": 0.21959500908851623,
      "num_tokens": 140121168.0,
      "step": 69150
    },
    {
      "entropy": 5.7713837146759035,
      "epoch": 6.043432666258848,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020486688087278883,
      "loss": 5.1134,
      "mean_token_accuracy": 0.22784795612096786,
      "num_tokens": 140130837.0,
      "step": 69155
    },
    {
      "entropy": 5.94601411819458,
      "epoch": 6.043869614611553,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020483727290383812,
      "loss": 5.2268,
      "mean_token_accuracy": 0.21557896286249162,
      "num_tokens": 140141201.0,
      "step": 69160
    },
    {
      "entropy": 5.780764579772949,
      "epoch": 6.044306562964258,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020480766628064523,
      "loss": 5.113,
      "mean_token_accuracy": 0.23178460747003554,
      "num_tokens": 140151683.0,
      "step": 69165
    },
    {
      "entropy": 5.882227182388306,
      "epoch": 6.0447435113169625,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020477806100377817,
      "loss": 5.1504,
      "mean_token_accuracy": 0.22761245518922807,
      "num_tokens": 140162929.0,
      "step": 69170
    },
    {
      "entropy": 5.826442670822144,
      "epoch": 6.045180459669667,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020474845707380463,
      "loss": 5.0804,
      "mean_token_accuracy": 0.22984902709722518,
      "num_tokens": 140172682.0,
      "step": 69175
    },
    {
      "entropy": 5.795203399658203,
      "epoch": 6.0456174080223715,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020471885449129253,
      "loss": 5.0964,
      "mean_token_accuracy": 0.22515202462673187,
      "num_tokens": 140183129.0,
      "step": 69180
    },
    {
      "entropy": 5.883539342880249,
      "epoch": 6.046054356375077,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002046892532568096,
      "loss": 5.2289,
      "mean_token_accuracy": 0.21374032646417618,
      "num_tokens": 140194788.0,
      "step": 69185
    },
    {
      "entropy": 5.858478116989136,
      "epoch": 6.046491304727781,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020465965337092363,
      "loss": 5.1365,
      "mean_token_accuracy": 0.22122201174497605,
      "num_tokens": 140205137.0,
      "step": 69190
    },
    {
      "entropy": 5.844442224502563,
      "epoch": 6.046928253080486,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002046300548342024,
      "loss": 5.0873,
      "mean_token_accuracy": 0.22865487188100814,
      "num_tokens": 140215073.0,
      "step": 69195
    },
    {
      "entropy": 5.759956932067871,
      "epoch": 6.04736520143319,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00020460045764721357,
      "loss": 5.1075,
      "mean_token_accuracy": 0.22863126695156097,
      "num_tokens": 140225280.0,
      "step": 69200
    },
    {
      "entropy": 5.857414627075196,
      "epoch": 6.047802149785896,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002045708618105248,
      "loss": 5.1337,
      "mean_token_accuracy": 0.22234255969524383,
      "num_tokens": 140234990.0,
      "step": 69205
    },
    {
      "entropy": 5.937241411209106,
      "epoch": 6.0482390981386,
      "grad_norm": 1.125,
      "learning_rate": 0.00020454126732470386,
      "loss": 5.1616,
      "mean_token_accuracy": 0.21882239878177642,
      "num_tokens": 140244333.0,
      "step": 69210
    },
    {
      "entropy": 5.842146635055542,
      "epoch": 6.0486760464913045,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020451167419031834,
      "loss": 5.1314,
      "mean_token_accuracy": 0.23018452227115632,
      "num_tokens": 140254436.0,
      "step": 69215
    },
    {
      "entropy": 5.822539138793945,
      "epoch": 6.049112994844009,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002044820824079358,
      "loss": 5.1771,
      "mean_token_accuracy": 0.23068441599607467,
      "num_tokens": 140264575.0,
      "step": 69220
    },
    {
      "entropy": 5.85261812210083,
      "epoch": 6.049549943196714,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020445249197812394,
      "loss": 5.1616,
      "mean_token_accuracy": 0.22998396456241607,
      "num_tokens": 140273734.0,
      "step": 69225
    },
    {
      "entropy": 5.95009412765503,
      "epoch": 6.049986891549419,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020442290290145022,
      "loss": 5.2219,
      "mean_token_accuracy": 0.21656394451856614,
      "num_tokens": 140283944.0,
      "step": 69230
    },
    {
      "entropy": 5.917208528518676,
      "epoch": 6.050423839902123,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020439331517848214,
      "loss": 5.1693,
      "mean_token_accuracy": 0.221559838950634,
      "num_tokens": 140293388.0,
      "step": 69235
    },
    {
      "entropy": 5.847663307189942,
      "epoch": 6.050860788254829,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020436372880978736,
      "loss": 5.1866,
      "mean_token_accuracy": 0.2238154634833336,
      "num_tokens": 140302488.0,
      "step": 69240
    },
    {
      "entropy": 5.827377557754517,
      "epoch": 6.051297736607533,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020433414379593317,
      "loss": 5.1489,
      "mean_token_accuracy": 0.22784710079431533,
      "num_tokens": 140313282.0,
      "step": 69245
    },
    {
      "entropy": 5.806141328811646,
      "epoch": 6.051734684960238,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020430456013748727,
      "loss": 5.1225,
      "mean_token_accuracy": 0.22563257813453674,
      "num_tokens": 140323976.0,
      "step": 69250
    },
    {
      "entropy": 5.853887557983398,
      "epoch": 6.052171633312942,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00020427497783501693,
      "loss": 5.139,
      "mean_token_accuracy": 0.22344034016132355,
      "num_tokens": 140334213.0,
      "step": 69255
    },
    {
      "entropy": 5.850074577331543,
      "epoch": 6.052608581665647,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002042453968890895,
      "loss": 5.0886,
      "mean_token_accuracy": 0.2269190862774849,
      "num_tokens": 140344061.0,
      "step": 69260
    },
    {
      "entropy": 5.845412015914917,
      "epoch": 6.053045530018352,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00020421581730027243,
      "loss": 5.1028,
      "mean_token_accuracy": 0.22954776436090468,
      "num_tokens": 140354040.0,
      "step": 69265
    },
    {
      "entropy": 5.813282489776611,
      "epoch": 6.053482478371056,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002041862390691331,
      "loss": 5.1107,
      "mean_token_accuracy": 0.2263941377401352,
      "num_tokens": 140363808.0,
      "step": 69270
    },
    {
      "entropy": 5.8127645492553714,
      "epoch": 6.053919426723761,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020415666219623886,
      "loss": 5.0954,
      "mean_token_accuracy": 0.23306984305381775,
      "num_tokens": 140373233.0,
      "step": 69275
    },
    {
      "entropy": 5.845934867858887,
      "epoch": 6.054356375076466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020412708668215691,
      "loss": 5.2255,
      "mean_token_accuracy": 0.20959712117910384,
      "num_tokens": 140383879.0,
      "step": 69280
    },
    {
      "entropy": 5.841794013977051,
      "epoch": 6.054793323429171,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020409751252745457,
      "loss": 5.1479,
      "mean_token_accuracy": 0.21974659711122513,
      "num_tokens": 140394046.0,
      "step": 69285
    },
    {
      "entropy": 5.8794975757598875,
      "epoch": 6.055230271781875,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020406793973269904,
      "loss": 5.2607,
      "mean_token_accuracy": 0.21913128942251206,
      "num_tokens": 140405039.0,
      "step": 69290
    },
    {
      "entropy": 5.950076866149902,
      "epoch": 6.0556672201345805,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020403836829845768,
      "loss": 5.2019,
      "mean_token_accuracy": 0.21779888421297072,
      "num_tokens": 140414521.0,
      "step": 69295
    },
    {
      "entropy": 5.874454164505005,
      "epoch": 6.056104168487285,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020400879822529762,
      "loss": 5.1856,
      "mean_token_accuracy": 0.2200072094798088,
      "num_tokens": 140425123.0,
      "step": 69300
    },
    {
      "entropy": 5.8299822330474855,
      "epoch": 6.056541116839989,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020397922951378595,
      "loss": 5.1428,
      "mean_token_accuracy": 0.2276284098625183,
      "num_tokens": 140434999.0,
      "step": 69305
    },
    {
      "entropy": 5.801126718521118,
      "epoch": 6.056978065192694,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020394966216448983,
      "loss": 5.1198,
      "mean_token_accuracy": 0.22796255499124526,
      "num_tokens": 140445469.0,
      "step": 69310
    },
    {
      "entropy": 5.849272680282593,
      "epoch": 6.057415013545399,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002039200961779765,
      "loss": 5.1436,
      "mean_token_accuracy": 0.21661657989025115,
      "num_tokens": 140454774.0,
      "step": 69315
    },
    {
      "entropy": 5.899533462524414,
      "epoch": 6.057851961898104,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020389053155481297,
      "loss": 5.1965,
      "mean_token_accuracy": 0.22264182865619658,
      "num_tokens": 140464093.0,
      "step": 69320
    },
    {
      "entropy": 5.833332824707031,
      "epoch": 6.058288910250808,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020386096829556628,
      "loss": 5.1276,
      "mean_token_accuracy": 0.2247079610824585,
      "num_tokens": 140475248.0,
      "step": 69325
    },
    {
      "entropy": 5.8452818393707275,
      "epoch": 6.058725858603513,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002038314064008035,
      "loss": 5.1076,
      "mean_token_accuracy": 0.2233192354440689,
      "num_tokens": 140485651.0,
      "step": 69330
    },
    {
      "entropy": 5.837257146835327,
      "epoch": 6.059162806956218,
      "grad_norm": 1.046875,
      "learning_rate": 0.00020380184587109163,
      "loss": 5.1959,
      "mean_token_accuracy": 0.2244327560067177,
      "num_tokens": 140498176.0,
      "step": 69335
    },
    {
      "entropy": 5.902980136871338,
      "epoch": 6.0595997553089225,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002037722867069976,
      "loss": 5.2264,
      "mean_token_accuracy": 0.21362577825784684,
      "num_tokens": 140508740.0,
      "step": 69340
    },
    {
      "entropy": 5.878762865066529,
      "epoch": 6.060036703661627,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020374272890908858,
      "loss": 5.1751,
      "mean_token_accuracy": 0.21712316274642945,
      "num_tokens": 140520419.0,
      "step": 69345
    },
    {
      "entropy": 5.764191150665283,
      "epoch": 6.060473652014332,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020371317247793125,
      "loss": 5.0203,
      "mean_token_accuracy": 0.24031205773353576,
      "num_tokens": 140530503.0,
      "step": 69350
    },
    {
      "entropy": 5.839783430099487,
      "epoch": 6.060910600367037,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020368361741409264,
      "loss": 5.1265,
      "mean_token_accuracy": 0.22041442692279817,
      "num_tokens": 140541140.0,
      "step": 69355
    },
    {
      "entropy": 5.815104389190674,
      "epoch": 6.061347548719741,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002036540637181396,
      "loss": 5.1454,
      "mean_token_accuracy": 0.22208609879016877,
      "num_tokens": 140550675.0,
      "step": 69360
    },
    {
      "entropy": 5.838600158691406,
      "epoch": 6.061784497072446,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020362451139063902,
      "loss": 5.0943,
      "mean_token_accuracy": 0.22154576033353807,
      "num_tokens": 140560908.0,
      "step": 69365
    },
    {
      "entropy": 5.817678070068359,
      "epoch": 6.062221445425151,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002035949604321578,
      "loss": 5.0483,
      "mean_token_accuracy": 0.22962642163038255,
      "num_tokens": 140571046.0,
      "step": 69370
    },
    {
      "entropy": 5.860207366943359,
      "epoch": 6.0626583937778555,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020356541084326257,
      "loss": 5.2068,
      "mean_token_accuracy": 0.21661909967660903,
      "num_tokens": 140580950.0,
      "step": 69375
    },
    {
      "entropy": 5.780282068252563,
      "epoch": 6.06309534213056,
      "grad_norm": 1.125,
      "learning_rate": 0.00020353586262452018,
      "loss": 5.0653,
      "mean_token_accuracy": 0.22879698723554612,
      "num_tokens": 140591414.0,
      "step": 69380
    },
    {
      "entropy": 5.809758710861206,
      "epoch": 6.0635322904832645,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020350631577649742,
      "loss": 5.136,
      "mean_token_accuracy": 0.22119158059358596,
      "num_tokens": 140600680.0,
      "step": 69385
    },
    {
      "entropy": 5.80311279296875,
      "epoch": 6.06396923883597,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020347677029976098,
      "loss": 5.0722,
      "mean_token_accuracy": 0.23648345917463304,
      "num_tokens": 140610181.0,
      "step": 69390
    },
    {
      "entropy": 5.812883615493774,
      "epoch": 6.064406187188674,
      "grad_norm": 1.25,
      "learning_rate": 0.00020344722619487754,
      "loss": 5.1904,
      "mean_token_accuracy": 0.22232812345027925,
      "num_tokens": 140620732.0,
      "step": 69395
    },
    {
      "entropy": 5.830493068695068,
      "epoch": 6.064843135541379,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020341768346241384,
      "loss": 5.1236,
      "mean_token_accuracy": 0.22012290358543396,
      "num_tokens": 140630560.0,
      "step": 69400
    },
    {
      "entropy": 5.79889030456543,
      "epoch": 6.065280083894084,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020338814210293644,
      "loss": 5.098,
      "mean_token_accuracy": 0.22733240574598312,
      "num_tokens": 140640578.0,
      "step": 69405
    },
    {
      "entropy": 5.780286693572998,
      "epoch": 6.065717032246789,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000203358602117012,
      "loss": 5.1201,
      "mean_token_accuracy": 0.22593388706445694,
      "num_tokens": 140651655.0,
      "step": 69410
    },
    {
      "entropy": 5.878146028518676,
      "epoch": 6.066153980599493,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002033290635052071,
      "loss": 5.211,
      "mean_token_accuracy": 0.21513136029243468,
      "num_tokens": 140662221.0,
      "step": 69415
    },
    {
      "entropy": 5.9173572063446045,
      "epoch": 6.0665909289521975,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002032995262680884,
      "loss": 5.2018,
      "mean_token_accuracy": 0.2195039853453636,
      "num_tokens": 140672114.0,
      "step": 69420
    },
    {
      "entropy": 5.802514362335205,
      "epoch": 6.067027877304903,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002032699904062223,
      "loss": 5.0229,
      "mean_token_accuracy": 0.2389114946126938,
      "num_tokens": 140681827.0,
      "step": 69425
    },
    {
      "entropy": 5.891010284423828,
      "epoch": 6.067464825657607,
      "grad_norm": 1.109375,
      "learning_rate": 0.00020324045592017532,
      "loss": 5.1597,
      "mean_token_accuracy": 0.21637391895055771,
      "num_tokens": 140691909.0,
      "step": 69430
    },
    {
      "entropy": 5.8652245044708256,
      "epoch": 6.067901774010312,
      "grad_norm": 1.015625,
      "learning_rate": 0.00020321092281051407,
      "loss": 5.2137,
      "mean_token_accuracy": 0.21204006373882295,
      "num_tokens": 140701875.0,
      "step": 69435
    },
    {
      "entropy": 5.81404972076416,
      "epoch": 6.068338722363016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020318139107780486,
      "loss": 5.1654,
      "mean_token_accuracy": 0.23527909368276595,
      "num_tokens": 140712407.0,
      "step": 69440
    },
    {
      "entropy": 5.870880794525147,
      "epoch": 6.068775670715722,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00020315186072261422,
      "loss": 5.1806,
      "mean_token_accuracy": 0.22037169337272644,
      "num_tokens": 140722384.0,
      "step": 69445
    },
    {
      "entropy": 5.878231525421143,
      "epoch": 6.069212619068426,
      "grad_norm": 1.125,
      "learning_rate": 0.00020312233174550853,
      "loss": 5.1178,
      "mean_token_accuracy": 0.22671087831258774,
      "num_tokens": 140732065.0,
      "step": 69450
    },
    {
      "entropy": 5.7929222106933596,
      "epoch": 6.069649567421131,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020309280414705422,
      "loss": 5.1262,
      "mean_token_accuracy": 0.23077486008405684,
      "num_tokens": 140741869.0,
      "step": 69455
    },
    {
      "entropy": 5.836217641830444,
      "epoch": 6.070086515773836,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020306327792781766,
      "loss": 5.1245,
      "mean_token_accuracy": 0.2299022451043129,
      "num_tokens": 140751535.0,
      "step": 69460
    },
    {
      "entropy": 5.830125141143799,
      "epoch": 6.07052346412654,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020303375308836508,
      "loss": 5.1101,
      "mean_token_accuracy": 0.2263118714094162,
      "num_tokens": 140762553.0,
      "step": 69465
    },
    {
      "entropy": 5.903000688552856,
      "epoch": 6.070960412479245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002030042296292628,
      "loss": 5.1801,
      "mean_token_accuracy": 0.2201431453227997,
      "num_tokens": 140771906.0,
      "step": 69470
    },
    {
      "entropy": 5.840651559829712,
      "epoch": 6.071397360831949,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002029747075510771,
      "loss": 5.0828,
      "mean_token_accuracy": 0.23064902573823928,
      "num_tokens": 140780661.0,
      "step": 69475
    },
    {
      "entropy": 5.846395635604859,
      "epoch": 6.071834309184655,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020294518685437434,
      "loss": 5.2031,
      "mean_token_accuracy": 0.22467246800661086,
      "num_tokens": 140791027.0,
      "step": 69480
    },
    {
      "entropy": 5.835890007019043,
      "epoch": 6.072271257537359,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020291566753972057,
      "loss": 5.108,
      "mean_token_accuracy": 0.23155177235603333,
      "num_tokens": 140802609.0,
      "step": 69485
    },
    {
      "entropy": 5.860843515396118,
      "epoch": 6.072708205890064,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002028861496076821,
      "loss": 5.0804,
      "mean_token_accuracy": 0.22876459658145903,
      "num_tokens": 140811759.0,
      "step": 69490
    },
    {
      "entropy": 5.8326891422271725,
      "epoch": 6.073145154242768,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020285663305882507,
      "loss": 5.1663,
      "mean_token_accuracy": 0.2233194276690483,
      "num_tokens": 140822277.0,
      "step": 69495
    },
    {
      "entropy": 5.847937107086182,
      "epoch": 6.0735821025954735,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002028271178937156,
      "loss": 5.1178,
      "mean_token_accuracy": 0.22548292875289916,
      "num_tokens": 140831960.0,
      "step": 69500
    },
    {
      "entropy": 5.833580160140992,
      "epoch": 6.074019050948178,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020279760411291993,
      "loss": 5.1702,
      "mean_token_accuracy": 0.22274278700351716,
      "num_tokens": 140842216.0,
      "step": 69505
    },
    {
      "entropy": 5.814882612228393,
      "epoch": 6.074455999300882,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020276809171700395,
      "loss": 5.1265,
      "mean_token_accuracy": 0.2245000422000885,
      "num_tokens": 140851201.0,
      "step": 69510
    },
    {
      "entropy": 5.837885093688965,
      "epoch": 6.074892947653588,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020273858070653377,
      "loss": 5.1073,
      "mean_token_accuracy": 0.22536493688821793,
      "num_tokens": 140862176.0,
      "step": 69515
    },
    {
      "entropy": 5.8920666694641115,
      "epoch": 6.075329896006292,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020270907108207553,
      "loss": 5.2331,
      "mean_token_accuracy": 0.21585543751716613,
      "num_tokens": 140872960.0,
      "step": 69520
    },
    {
      "entropy": 5.855195331573486,
      "epoch": 6.075766844358997,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002026795628441952,
      "loss": 5.1416,
      "mean_token_accuracy": 0.2242415204644203,
      "num_tokens": 140883542.0,
      "step": 69525
    },
    {
      "entropy": 5.808722019195557,
      "epoch": 6.076203792711701,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020265005599345875,
      "loss": 5.1389,
      "mean_token_accuracy": 0.220771624147892,
      "num_tokens": 140893150.0,
      "step": 69530
    },
    {
      "entropy": 5.820524740219116,
      "epoch": 6.0766407410644065,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020262055053043215,
      "loss": 5.0906,
      "mean_token_accuracy": 0.23051931709051132,
      "num_tokens": 140903199.0,
      "step": 69535
    },
    {
      "entropy": 5.850038480758667,
      "epoch": 6.077077689417111,
      "grad_norm": 1.125,
      "learning_rate": 0.00020259104645568127,
      "loss": 5.078,
      "mean_token_accuracy": 0.22694414407014846,
      "num_tokens": 140912832.0,
      "step": 69540
    },
    {
      "entropy": 5.86175045967102,
      "epoch": 6.0775146377698155,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002025615437697721,
      "loss": 5.072,
      "mean_token_accuracy": 0.22956817001104354,
      "num_tokens": 140922787.0,
      "step": 69545
    },
    {
      "entropy": 5.786328506469727,
      "epoch": 6.07795158612252,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020253204247327046,
      "loss": 5.1313,
      "mean_token_accuracy": 0.22731308192014693,
      "num_tokens": 140932726.0,
      "step": 69550
    },
    {
      "entropy": 5.790648698806763,
      "epoch": 6.078388534475225,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020250254256674223,
      "loss": 5.0565,
      "mean_token_accuracy": 0.23588847368955612,
      "num_tokens": 140942745.0,
      "step": 69555
    },
    {
      "entropy": 5.835395336151123,
      "epoch": 6.07882548282793,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020247304405075317,
      "loss": 5.1095,
      "mean_token_accuracy": 0.2229289561510086,
      "num_tokens": 140953363.0,
      "step": 69560
    },
    {
      "entropy": 5.878841543197632,
      "epoch": 6.079262431180634,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002024435469258692,
      "loss": 5.1084,
      "mean_token_accuracy": 0.22740317583084108,
      "num_tokens": 140963841.0,
      "step": 69565
    },
    {
      "entropy": 5.8150718212127686,
      "epoch": 6.07969937953334,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020241405119265598,
      "loss": 5.1178,
      "mean_token_accuracy": 0.23063139766454696,
      "num_tokens": 140974633.0,
      "step": 69570
    },
    {
      "entropy": 5.810128593444825,
      "epoch": 6.080136327886044,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020238455685167928,
      "loss": 5.0994,
      "mean_token_accuracy": 0.23233289271593094,
      "num_tokens": 140986270.0,
      "step": 69575
    },
    {
      "entropy": 5.80255765914917,
      "epoch": 6.0805732762387485,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002023550639035049,
      "loss": 5.0898,
      "mean_token_accuracy": 0.22847151160240173,
      "num_tokens": 140995928.0,
      "step": 69580
    },
    {
      "entropy": 5.802024221420288,
      "epoch": 6.081010224591453,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002023255723486984,
      "loss": 5.103,
      "mean_token_accuracy": 0.22960526198148729,
      "num_tokens": 141005265.0,
      "step": 69585
    },
    {
      "entropy": 5.7869720458984375,
      "epoch": 6.081447172944158,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020229608218782553,
      "loss": 5.0382,
      "mean_token_accuracy": 0.23639172464609146,
      "num_tokens": 141015029.0,
      "step": 69590
    },
    {
      "entropy": 5.819604682922363,
      "epoch": 6.081884121296863,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020226659342145186,
      "loss": 5.0917,
      "mean_token_accuracy": 0.23011133074760437,
      "num_tokens": 141023763.0,
      "step": 69595
    },
    {
      "entropy": 5.8900152206420895,
      "epoch": 6.082321069649567,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020223710605014305,
      "loss": 5.2862,
      "mean_token_accuracy": 0.21756176501512528,
      "num_tokens": 141034180.0,
      "step": 69600
    },
    {
      "entropy": 5.776221895217896,
      "epoch": 6.082758018002272,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020220762007446474,
      "loss": 5.1503,
      "mean_token_accuracy": 0.22476284354925155,
      "num_tokens": 141043817.0,
      "step": 69605
    },
    {
      "entropy": 5.835810708999634,
      "epoch": 6.083194966354977,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020217813549498233,
      "loss": 5.2222,
      "mean_token_accuracy": 0.21767432540655135,
      "num_tokens": 141053832.0,
      "step": 69610
    },
    {
      "entropy": 5.86129207611084,
      "epoch": 6.083631914707682,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020214865231226147,
      "loss": 5.1163,
      "mean_token_accuracy": 0.2265701100230217,
      "num_tokens": 141064334.0,
      "step": 69615
    },
    {
      "entropy": 5.885386610031128,
      "epoch": 6.084068863060386,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020211917052686762,
      "loss": 5.1621,
      "mean_token_accuracy": 0.22130893915891647,
      "num_tokens": 141073715.0,
      "step": 69620
    },
    {
      "entropy": 5.783041667938233,
      "epoch": 6.0845058114130905,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020208969013936633,
      "loss": 5.0316,
      "mean_token_accuracy": 0.2351677805185318,
      "num_tokens": 141084255.0,
      "step": 69625
    },
    {
      "entropy": 5.857237482070923,
      "epoch": 6.084942759765796,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020206021115032297,
      "loss": 5.2045,
      "mean_token_accuracy": 0.21577191948890687,
      "num_tokens": 141094597.0,
      "step": 69630
    },
    {
      "entropy": 5.815689373016357,
      "epoch": 6.0853797081185,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020203073356030293,
      "loss": 5.1594,
      "mean_token_accuracy": 0.21770806163549422,
      "num_tokens": 141104949.0,
      "step": 69635
    },
    {
      "entropy": 5.782705974578858,
      "epoch": 6.085816656471205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002020012573698717,
      "loss": 5.106,
      "mean_token_accuracy": 0.23259497731924056,
      "num_tokens": 141115478.0,
      "step": 69640
    },
    {
      "entropy": 5.838930225372314,
      "epoch": 6.08625360482391,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002019717825795946,
      "loss": 5.1279,
      "mean_token_accuracy": 0.223356494307518,
      "num_tokens": 141125517.0,
      "step": 69645
    },
    {
      "entropy": 5.80334620475769,
      "epoch": 6.086690553176615,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020194230919003698,
      "loss": 5.1466,
      "mean_token_accuracy": 0.2304024651646614,
      "num_tokens": 141135483.0,
      "step": 69650
    },
    {
      "entropy": 5.831393146514893,
      "epoch": 6.087127501529319,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020191283720176413,
      "loss": 5.0847,
      "mean_token_accuracy": 0.23612601906061173,
      "num_tokens": 141145631.0,
      "step": 69655
    },
    {
      "entropy": 5.881703090667725,
      "epoch": 6.087564449882024,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020188336661534128,
      "loss": 5.1966,
      "mean_token_accuracy": 0.2101586639881134,
      "num_tokens": 141155372.0,
      "step": 69660
    },
    {
      "entropy": 5.795291519165039,
      "epoch": 6.088001398234729,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020185389743133386,
      "loss": 5.122,
      "mean_token_accuracy": 0.22297335267066956,
      "num_tokens": 141165881.0,
      "step": 69665
    },
    {
      "entropy": 5.788751649856567,
      "epoch": 6.088438346587433,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000201824429650307,
      "loss": 5.0765,
      "mean_token_accuracy": 0.2233137771487236,
      "num_tokens": 141175890.0,
      "step": 69670
    },
    {
      "entropy": 5.831302738189697,
      "epoch": 6.088875294940138,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000201794963272826,
      "loss": 5.1355,
      "mean_token_accuracy": 0.2249202996492386,
      "num_tokens": 141185511.0,
      "step": 69675
    },
    {
      "entropy": 5.837126636505127,
      "epoch": 6.089312243292842,
      "grad_norm": 1.25,
      "learning_rate": 0.0002017654982994559,
      "loss": 5.2324,
      "mean_token_accuracy": 0.22185763865709304,
      "num_tokens": 141195017.0,
      "step": 69680
    },
    {
      "entropy": 5.902753829956055,
      "epoch": 6.089749191645548,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020173603473076185,
      "loss": 5.1475,
      "mean_token_accuracy": 0.22741297781467437,
      "num_tokens": 141204236.0,
      "step": 69685
    },
    {
      "entropy": 5.880736446380615,
      "epoch": 6.090186139998252,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002017065725673091,
      "loss": 5.1693,
      "mean_token_accuracy": 0.22821279615163803,
      "num_tokens": 141214172.0,
      "step": 69690
    },
    {
      "entropy": 5.89495735168457,
      "epoch": 6.090623088350957,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020167711180966275,
      "loss": 5.1254,
      "mean_token_accuracy": 0.2246581345796585,
      "num_tokens": 141223994.0,
      "step": 69695
    },
    {
      "entropy": 5.790171957015991,
      "epoch": 6.091060036703662,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020164765245838778,
      "loss": 5.1331,
      "mean_token_accuracy": 0.21876976191997527,
      "num_tokens": 141234621.0,
      "step": 69700
    },
    {
      "entropy": 5.819083118438721,
      "epoch": 6.0914969850563665,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020161819451404927,
      "loss": 5.0685,
      "mean_token_accuracy": 0.22658351063728333,
      "num_tokens": 141244263.0,
      "step": 69705
    },
    {
      "entropy": 5.83737850189209,
      "epoch": 6.091933933409071,
      "grad_norm": 1.125,
      "learning_rate": 0.00020158873797721216,
      "loss": 5.1819,
      "mean_token_accuracy": 0.22099213153123856,
      "num_tokens": 141254320.0,
      "step": 69710
    },
    {
      "entropy": 5.854991006851196,
      "epoch": 6.092370881761775,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002015592828484416,
      "loss": 5.1735,
      "mean_token_accuracy": 0.22743515968322753,
      "num_tokens": 141265226.0,
      "step": 69715
    },
    {
      "entropy": 5.849716091156006,
      "epoch": 6.092807830114481,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002015298291283025,
      "loss": 5.203,
      "mean_token_accuracy": 0.21932098865509034,
      "num_tokens": 141275320.0,
      "step": 69720
    },
    {
      "entropy": 5.909502792358398,
      "epoch": 6.093244778467185,
      "grad_norm": 1.03125,
      "learning_rate": 0.00020150037681735978,
      "loss": 5.182,
      "mean_token_accuracy": 0.2232228621840477,
      "num_tokens": 141285411.0,
      "step": 69725
    },
    {
      "entropy": 5.839595127105713,
      "epoch": 6.09368172681989,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020147092591617826,
      "loss": 5.0302,
      "mean_token_accuracy": 0.22912879288196564,
      "num_tokens": 141295164.0,
      "step": 69730
    },
    {
      "entropy": 5.8798681735992435,
      "epoch": 6.094118675172594,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020144147642532295,
      "loss": 5.1911,
      "mean_token_accuracy": 0.2170073390007019,
      "num_tokens": 141305345.0,
      "step": 69735
    },
    {
      "entropy": 5.812002086639405,
      "epoch": 6.0945556235253,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020141202834535866,
      "loss": 5.1323,
      "mean_token_accuracy": 0.22772397100925446,
      "num_tokens": 141315478.0,
      "step": 69740
    },
    {
      "entropy": 5.844466495513916,
      "epoch": 6.094992571878004,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020138258167685025,
      "loss": 5.2114,
      "mean_token_accuracy": 0.22334292083978652,
      "num_tokens": 141325896.0,
      "step": 69745
    },
    {
      "entropy": 5.823497581481933,
      "epoch": 6.0954295202307085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002013531364203625,
      "loss": 5.1221,
      "mean_token_accuracy": 0.2273145481944084,
      "num_tokens": 141336113.0,
      "step": 69750
    },
    {
      "entropy": 5.8582603454589846,
      "epoch": 6.095866468583414,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020132369257646015,
      "loss": 5.0688,
      "mean_token_accuracy": 0.23924387097358704,
      "num_tokens": 141346355.0,
      "step": 69755
    },
    {
      "entropy": 5.963270568847657,
      "epoch": 6.096303416936118,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00020129425014570802,
      "loss": 5.2608,
      "mean_token_accuracy": 0.2202950969338417,
      "num_tokens": 141357806.0,
      "step": 69760
    },
    {
      "entropy": 5.8985679149627686,
      "epoch": 6.096740365288823,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002012648091286708,
      "loss": 5.1751,
      "mean_token_accuracy": 0.21973987370729448,
      "num_tokens": 141367529.0,
      "step": 69765
    },
    {
      "entropy": 5.8123565196990965,
      "epoch": 6.097177313641527,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002012353695259131,
      "loss": 5.0708,
      "mean_token_accuracy": 0.23024348467588424,
      "num_tokens": 141378372.0,
      "step": 69770
    },
    {
      "entropy": 5.787427473068237,
      "epoch": 6.097614261994233,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020120593133799976,
      "loss": 5.0904,
      "mean_token_accuracy": 0.23172144889831542,
      "num_tokens": 141388372.0,
      "step": 69775
    },
    {
      "entropy": 5.731652498245239,
      "epoch": 6.098051210346937,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020117649456549531,
      "loss": 4.9789,
      "mean_token_accuracy": 0.2379856824874878,
      "num_tokens": 141398459.0,
      "step": 69780
    },
    {
      "entropy": 5.760429763793946,
      "epoch": 6.098488158699642,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002011470592089643,
      "loss": 5.0902,
      "mean_token_accuracy": 0.23287004083395005,
      "num_tokens": 141408975.0,
      "step": 69785
    },
    {
      "entropy": 5.831823110580444,
      "epoch": 6.098925107052346,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020111762526897142,
      "loss": 5.1801,
      "mean_token_accuracy": 0.22601098269224168,
      "num_tokens": 141419547.0,
      "step": 69790
    },
    {
      "entropy": 5.87305064201355,
      "epoch": 6.099362055405051,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020108819274608126,
      "loss": 5.171,
      "mean_token_accuracy": 0.2178846001625061,
      "num_tokens": 141428716.0,
      "step": 69795
    },
    {
      "entropy": 5.795052623748779,
      "epoch": 6.099799003757756,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020105876164085818,
      "loss": 5.1415,
      "mean_token_accuracy": 0.2277742862701416,
      "num_tokens": 141439636.0,
      "step": 69800
    },
    {
      "entropy": 5.831130838394165,
      "epoch": 6.10023595211046,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00020102933195386685,
      "loss": 5.1059,
      "mean_token_accuracy": 0.22811751812696457,
      "num_tokens": 141450672.0,
      "step": 69805
    },
    {
      "entropy": 5.900415277481079,
      "epoch": 6.100672900463166,
      "grad_norm": 1.25,
      "learning_rate": 0.00020099990368567178,
      "loss": 5.2617,
      "mean_token_accuracy": 0.2106873109936714,
      "num_tokens": 141460642.0,
      "step": 69810
    },
    {
      "entropy": 5.8879647731781,
      "epoch": 6.10110984881587,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002009704768368373,
      "loss": 5.1336,
      "mean_token_accuracy": 0.22890762388706207,
      "num_tokens": 141470062.0,
      "step": 69815
    },
    {
      "entropy": 5.857402896881103,
      "epoch": 6.101546797168575,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002009410514079278,
      "loss": 5.0797,
      "mean_token_accuracy": 0.2343340441584587,
      "num_tokens": 141479827.0,
      "step": 69820
    },
    {
      "entropy": 5.778765869140625,
      "epoch": 6.101983745521279,
      "grad_norm": 1.125,
      "learning_rate": 0.00020091162739950774,
      "loss": 5.0744,
      "mean_token_accuracy": 0.23660405725240707,
      "num_tokens": 141490327.0,
      "step": 69825
    },
    {
      "entropy": 5.784572696685791,
      "epoch": 6.1024206938739844,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020088220481214152,
      "loss": 5.1083,
      "mean_token_accuracy": 0.22770523130893708,
      "num_tokens": 141500150.0,
      "step": 69830
    },
    {
      "entropy": 5.841646909713745,
      "epoch": 6.102857642226689,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020085278364639347,
      "loss": 5.1582,
      "mean_token_accuracy": 0.22481673508882521,
      "num_tokens": 141509598.0,
      "step": 69835
    },
    {
      "entropy": 5.8350037097930905,
      "epoch": 6.103294590579393,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002008233639028279,
      "loss": 5.1511,
      "mean_token_accuracy": 0.22131529301404954,
      "num_tokens": 141519801.0,
      "step": 69840
    },
    {
      "entropy": 5.854510974884033,
      "epoch": 6.103731538932098,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020079394558200904,
      "loss": 5.1615,
      "mean_token_accuracy": 0.22111736238002777,
      "num_tokens": 141529987.0,
      "step": 69845
    },
    {
      "entropy": 5.847797679901123,
      "epoch": 6.104168487284803,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020076452868450118,
      "loss": 5.139,
      "mean_token_accuracy": 0.2191311925649643,
      "num_tokens": 141540947.0,
      "step": 69850
    },
    {
      "entropy": 5.830555152893067,
      "epoch": 6.104605435637508,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002007351132108685,
      "loss": 5.1703,
      "mean_token_accuracy": 0.22141456007957458,
      "num_tokens": 141552368.0,
      "step": 69855
    },
    {
      "entropy": 5.837300062179565,
      "epoch": 6.105042383990212,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002007056991616754,
      "loss": 5.1272,
      "mean_token_accuracy": 0.22504555732011794,
      "num_tokens": 141562505.0,
      "step": 69860
    },
    {
      "entropy": 5.829103755950928,
      "epoch": 6.105479332342917,
      "grad_norm": 1.109375,
      "learning_rate": 0.00020067628653748588,
      "loss": 5.1234,
      "mean_token_accuracy": 0.22551584541797637,
      "num_tokens": 141573713.0,
      "step": 69865
    },
    {
      "entropy": 5.919946908950806,
      "epoch": 6.105916280695622,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020064687533886404,
      "loss": 5.1688,
      "mean_token_accuracy": 0.21737061738967894,
      "num_tokens": 141584217.0,
      "step": 69870
    },
    {
      "entropy": 5.898239946365356,
      "epoch": 6.1063532290483264,
      "grad_norm": 1.109375,
      "learning_rate": 0.00020061746556637416,
      "loss": 5.162,
      "mean_token_accuracy": 0.2283315896987915,
      "num_tokens": 141594344.0,
      "step": 69875
    },
    {
      "entropy": 5.899674129486084,
      "epoch": 6.106790177401031,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020058805722058027,
      "loss": 5.1885,
      "mean_token_accuracy": 0.21959789842367172,
      "num_tokens": 141605927.0,
      "step": 69880
    },
    {
      "entropy": 5.868799352645874,
      "epoch": 6.107227125753736,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020055865030204646,
      "loss": 5.1579,
      "mean_token_accuracy": 0.223355969786644,
      "num_tokens": 141616438.0,
      "step": 69885
    },
    {
      "entropy": 5.86576828956604,
      "epoch": 6.107664074106441,
      "grad_norm": 1.125,
      "learning_rate": 0.00020052924481133667,
      "loss": 5.1408,
      "mean_token_accuracy": 0.2203812777996063,
      "num_tokens": 141627073.0,
      "step": 69890
    },
    {
      "entropy": 5.831465101242065,
      "epoch": 6.108101022459145,
      "grad_norm": 1.15625,
      "learning_rate": 0.000200499840749015,
      "loss": 5.1586,
      "mean_token_accuracy": 0.2196997031569481,
      "num_tokens": 141636572.0,
      "step": 69895
    },
    {
      "entropy": 5.840401363372803,
      "epoch": 6.10853797081185,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020047043811564547,
      "loss": 5.1707,
      "mean_token_accuracy": 0.2248268187046051,
      "num_tokens": 141646291.0,
      "step": 69900
    },
    {
      "entropy": 5.870792198181152,
      "epoch": 6.108974919164555,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00020044103691179194,
      "loss": 5.1827,
      "mean_token_accuracy": 0.21871624886989594,
      "num_tokens": 141657047.0,
      "step": 69905
    },
    {
      "entropy": 5.873771810531617,
      "epoch": 6.1094118675172595,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00020041163713801845,
      "loss": 5.1843,
      "mean_token_accuracy": 0.22345936447381973,
      "num_tokens": 141668620.0,
      "step": 69910
    },
    {
      "entropy": 5.854202508926392,
      "epoch": 6.109848815869964,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00020038223879488882,
      "loss": 5.1127,
      "mean_token_accuracy": 0.23239877820014954,
      "num_tokens": 141678885.0,
      "step": 69915
    },
    {
      "entropy": 5.801882266998291,
      "epoch": 6.1102857642226684,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020035284188296687,
      "loss": 5.1046,
      "mean_token_accuracy": 0.22733244001865388,
      "num_tokens": 141688400.0,
      "step": 69920
    },
    {
      "entropy": 5.799589490890503,
      "epoch": 6.110722712575374,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020032344640281662,
      "loss": 5.097,
      "mean_token_accuracy": 0.23597875237464905,
      "num_tokens": 141698851.0,
      "step": 69925
    },
    {
      "entropy": 5.833784866333008,
      "epoch": 6.111159660928078,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020029405235500177,
      "loss": 5.0958,
      "mean_token_accuracy": 0.23047683984041215,
      "num_tokens": 141708915.0,
      "step": 69930
    },
    {
      "entropy": 5.81254358291626,
      "epoch": 6.111596609280783,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020026465974008618,
      "loss": 5.0922,
      "mean_token_accuracy": 0.22862282991409302,
      "num_tokens": 141720300.0,
      "step": 69935
    },
    {
      "entropy": 5.866451692581177,
      "epoch": 6.112033557633488,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002002352685586335,
      "loss": 5.107,
      "mean_token_accuracy": 0.22587087452411653,
      "num_tokens": 141729798.0,
      "step": 69940
    },
    {
      "entropy": 5.880594110488891,
      "epoch": 6.112470505986193,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020020587881120755,
      "loss": 5.1953,
      "mean_token_accuracy": 0.22395160496234895,
      "num_tokens": 141740132.0,
      "step": 69945
    },
    {
      "entropy": 5.824875020980835,
      "epoch": 6.112907454338897,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002001764904983721,
      "loss": 5.1661,
      "mean_token_accuracy": 0.22378654778003693,
      "num_tokens": 141750231.0,
      "step": 69950
    },
    {
      "entropy": 5.826145839691162,
      "epoch": 6.1133444026916015,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020014710362069077,
      "loss": 5.1512,
      "mean_token_accuracy": 0.22392451763153076,
      "num_tokens": 141761385.0,
      "step": 69955
    },
    {
      "entropy": 5.908224582672119,
      "epoch": 6.113781351044307,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020011771817872722,
      "loss": 5.1202,
      "mean_token_accuracy": 0.23131305128335952,
      "num_tokens": 141770668.0,
      "step": 69960
    },
    {
      "entropy": 5.894895219802857,
      "epoch": 6.114218299397011,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020008833417304502,
      "loss": 5.1824,
      "mean_token_accuracy": 0.22267802506685258,
      "num_tokens": 141780259.0,
      "step": 69965
    },
    {
      "entropy": 5.920859718322754,
      "epoch": 6.114655247749716,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002000589516042078,
      "loss": 5.1491,
      "mean_token_accuracy": 0.220801243185997,
      "num_tokens": 141790266.0,
      "step": 69970
    },
    {
      "entropy": 5.909363889694214,
      "epoch": 6.11509219610242,
      "grad_norm": 1.125,
      "learning_rate": 0.0002000295704727792,
      "loss": 5.1574,
      "mean_token_accuracy": 0.22092786580324172,
      "num_tokens": 141800870.0,
      "step": 69975
    },
    {
      "entropy": 5.8031219959259035,
      "epoch": 6.115529144455126,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020000019077932274,
      "loss": 5.0323,
      "mean_token_accuracy": 0.237224780023098,
      "num_tokens": 141810373.0,
      "step": 69980
    },
    {
      "entropy": 5.807530546188355,
      "epoch": 6.11596609280783,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019997081252440193,
      "loss": 5.1072,
      "mean_token_accuracy": 0.2288241371512413,
      "num_tokens": 141819348.0,
      "step": 69985
    },
    {
      "entropy": 5.792418956756592,
      "epoch": 6.116403041160535,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019994143570858014,
      "loss": 5.0982,
      "mean_token_accuracy": 0.2324607789516449,
      "num_tokens": 141828862.0,
      "step": 69990
    },
    {
      "entropy": 5.839476871490478,
      "epoch": 6.11683998951324,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019991206033242103,
      "loss": 5.1133,
      "mean_token_accuracy": 0.22819588333368301,
      "num_tokens": 141837718.0,
      "step": 69995
    },
    {
      "entropy": 5.862972640991211,
      "epoch": 6.117276937865944,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019988268639648793,
      "loss": 5.1567,
      "mean_token_accuracy": 0.22153465896844865,
      "num_tokens": 141848470.0,
      "step": 70000
    },
    {
      "entropy": 5.916492176055908,
      "epoch": 6.117713886218649,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019985331390134432,
      "loss": 5.1661,
      "mean_token_accuracy": 0.22179060578346252,
      "num_tokens": 141858091.0,
      "step": 70005
    },
    {
      "entropy": 5.910352373123169,
      "epoch": 6.118150834571353,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019982394284755352,
      "loss": 5.1145,
      "mean_token_accuracy": 0.2333526447415352,
      "num_tokens": 141868469.0,
      "step": 70010
    },
    {
      "entropy": 5.874615049362182,
      "epoch": 6.118587782924059,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019979457323567878,
      "loss": 5.1583,
      "mean_token_accuracy": 0.22022066563367843,
      "num_tokens": 141877891.0,
      "step": 70015
    },
    {
      "entropy": 5.775642442703247,
      "epoch": 6.119024731276763,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019976520506628364,
      "loss": 5.0643,
      "mean_token_accuracy": 0.23153169453144073,
      "num_tokens": 141888251.0,
      "step": 70020
    },
    {
      "entropy": 5.887964248657227,
      "epoch": 6.119461679629468,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001997358383399312,
      "loss": 5.2236,
      "mean_token_accuracy": 0.21452661752700805,
      "num_tokens": 141898249.0,
      "step": 70025
    },
    {
      "entropy": 5.8800849437713625,
      "epoch": 6.119898627982172,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001997064730571849,
      "loss": 5.1577,
      "mean_token_accuracy": 0.22799917161464692,
      "num_tokens": 141908110.0,
      "step": 70030
    },
    {
      "entropy": 5.88116717338562,
      "epoch": 6.1203355763348775,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019967710921860795,
      "loss": 5.0857,
      "mean_token_accuracy": 0.22648982107639312,
      "num_tokens": 141918692.0,
      "step": 70035
    },
    {
      "entropy": 5.837185525894165,
      "epoch": 6.120772524687582,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001996477468247634,
      "loss": 5.1039,
      "mean_token_accuracy": 0.22989484965801238,
      "num_tokens": 141928205.0,
      "step": 70040
    },
    {
      "entropy": 5.871635580062867,
      "epoch": 6.121209473040286,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019961838587621457,
      "loss": 5.1154,
      "mean_token_accuracy": 0.22310865819454193,
      "num_tokens": 141937035.0,
      "step": 70045
    },
    {
      "entropy": 5.832102870941162,
      "epoch": 6.121646421392992,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001995890263735246,
      "loss": 5.0873,
      "mean_token_accuracy": 0.2319301411509514,
      "num_tokens": 141946656.0,
      "step": 70050
    },
    {
      "entropy": 5.818221807479858,
      "epoch": 6.122083369745696,
      "grad_norm": 1.109375,
      "learning_rate": 0.00019955966831725668,
      "loss": 5.1443,
      "mean_token_accuracy": 0.22773782312870025,
      "num_tokens": 141956698.0,
      "step": 70055
    },
    {
      "entropy": 5.8860899925231935,
      "epoch": 6.122520318098401,
      "grad_norm": 1.078125,
      "learning_rate": 0.00019953031170797376,
      "loss": 5.1849,
      "mean_token_accuracy": 0.22762616723775864,
      "num_tokens": 141968641.0,
      "step": 70060
    },
    {
      "entropy": 5.827550840377808,
      "epoch": 6.122957266451105,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019950095654623908,
      "loss": 5.1015,
      "mean_token_accuracy": 0.2282833054661751,
      "num_tokens": 141979133.0,
      "step": 70065
    },
    {
      "entropy": 5.904693555831909,
      "epoch": 6.1233942148038105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019947160283261556,
      "loss": 5.2059,
      "mean_token_accuracy": 0.2205070286989212,
      "num_tokens": 141989565.0,
      "step": 70070
    },
    {
      "entropy": 5.871506690979004,
      "epoch": 6.123831163156515,
      "grad_norm": 1.25,
      "learning_rate": 0.00019944225056766629,
      "loss": 5.0585,
      "mean_token_accuracy": 0.22796231061220168,
      "num_tokens": 141998821.0,
      "step": 70075
    },
    {
      "entropy": 5.836687803268433,
      "epoch": 6.1242681115092195,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019941289975195425,
      "loss": 5.1359,
      "mean_token_accuracy": 0.22714918106794357,
      "num_tokens": 142008308.0,
      "step": 70080
    },
    {
      "entropy": 5.8229512691497805,
      "epoch": 6.124705059861924,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019938355038604246,
      "loss": 5.1302,
      "mean_token_accuracy": 0.22470941096544267,
      "num_tokens": 142018155.0,
      "step": 70085
    },
    {
      "entropy": 5.838626861572266,
      "epoch": 6.125142008214629,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001993542024704937,
      "loss": 5.1531,
      "mean_token_accuracy": 0.2269817352294922,
      "num_tokens": 142028363.0,
      "step": 70090
    },
    {
      "entropy": 5.904847764968872,
      "epoch": 6.125578956567334,
      "grad_norm": 1.109375,
      "learning_rate": 0.00019932485600587097,
      "loss": 5.1136,
      "mean_token_accuracy": 0.22644663006067275,
      "num_tokens": 142038133.0,
      "step": 70095
    },
    {
      "entropy": 5.854980516433716,
      "epoch": 6.126015904920038,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001992955109927373,
      "loss": 5.1534,
      "mean_token_accuracy": 0.2208598345518112,
      "num_tokens": 142048599.0,
      "step": 70100
    },
    {
      "entropy": 5.781077337265015,
      "epoch": 6.126452853272744,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019926616743165526,
      "loss": 5.0342,
      "mean_token_accuracy": 0.23449489921331407,
      "num_tokens": 142058131.0,
      "step": 70105
    },
    {
      "entropy": 5.837939167022705,
      "epoch": 6.126889801625448,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019923682532318782,
      "loss": 5.1492,
      "mean_token_accuracy": 0.22389456480741501,
      "num_tokens": 142069129.0,
      "step": 70110
    },
    {
      "entropy": 5.84678840637207,
      "epoch": 6.1273267499781525,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001992074846678978,
      "loss": 5.1414,
      "mean_token_accuracy": 0.22775069177150725,
      "num_tokens": 142079652.0,
      "step": 70115
    },
    {
      "entropy": 5.84831805229187,
      "epoch": 6.127763698330857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019917814546634795,
      "loss": 5.1351,
      "mean_token_accuracy": 0.23158909976482392,
      "num_tokens": 142089547.0,
      "step": 70120
    },
    {
      "entropy": 5.866306304931641,
      "epoch": 6.128200646683562,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019914880771910098,
      "loss": 5.1897,
      "mean_token_accuracy": 0.2233928546309471,
      "num_tokens": 142100203.0,
      "step": 70125
    },
    {
      "entropy": 5.832566022872925,
      "epoch": 6.128637595036267,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019911947142671964,
      "loss": 5.1289,
      "mean_token_accuracy": 0.22670524567365646,
      "num_tokens": 142109650.0,
      "step": 70130
    },
    {
      "entropy": 5.862476778030396,
      "epoch": 6.129074543388971,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001990901365897666,
      "loss": 5.162,
      "mean_token_accuracy": 0.22431124597787858,
      "num_tokens": 142120409.0,
      "step": 70135
    },
    {
      "entropy": 5.84286789894104,
      "epoch": 6.129511491741676,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001990608032088045,
      "loss": 5.1477,
      "mean_token_accuracy": 0.22119169533252717,
      "num_tokens": 142131148.0,
      "step": 70140
    },
    {
      "entropy": 5.848607778549194,
      "epoch": 6.129948440094381,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019903147128439596,
      "loss": 5.1173,
      "mean_token_accuracy": 0.2302807629108429,
      "num_tokens": 142140998.0,
      "step": 70145
    },
    {
      "entropy": 5.928045034408569,
      "epoch": 6.130385388447086,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001990021408171037,
      "loss": 5.3159,
      "mean_token_accuracy": 0.2104293167591095,
      "num_tokens": 142152564.0,
      "step": 70150
    },
    {
      "entropy": 5.906856870651245,
      "epoch": 6.13082233679979,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019897281180749,
      "loss": 5.1755,
      "mean_token_accuracy": 0.2121259391307831,
      "num_tokens": 142162230.0,
      "step": 70155
    },
    {
      "entropy": 5.820883512496948,
      "epoch": 6.131259285152495,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001989434842561178,
      "loss": 5.0689,
      "mean_token_accuracy": 0.23028005808591842,
      "num_tokens": 142172405.0,
      "step": 70160
    },
    {
      "entropy": 5.870146465301514,
      "epoch": 6.1316962335052,
      "grad_norm": 1.09375,
      "learning_rate": 0.00019891415816354936,
      "loss": 5.163,
      "mean_token_accuracy": 0.22456743866205214,
      "num_tokens": 142182950.0,
      "step": 70165
    },
    {
      "entropy": 5.810049057006836,
      "epoch": 6.132133181857904,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019888483353034726,
      "loss": 5.1084,
      "mean_token_accuracy": 0.23056249767541886,
      "num_tokens": 142192731.0,
      "step": 70170
    },
    {
      "entropy": 5.880335426330566,
      "epoch": 6.132570130210609,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019885551035707394,
      "loss": 5.202,
      "mean_token_accuracy": 0.22615770101547242,
      "num_tokens": 142202425.0,
      "step": 70175
    },
    {
      "entropy": 5.8487396240234375,
      "epoch": 6.133007078563314,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019882618864429174,
      "loss": 5.0854,
      "mean_token_accuracy": 0.23148886859416962,
      "num_tokens": 142212525.0,
      "step": 70180
    },
    {
      "entropy": 5.818871879577637,
      "epoch": 6.133444026916019,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001987968683925633,
      "loss": 5.1525,
      "mean_token_accuracy": 0.2190559059381485,
      "num_tokens": 142222551.0,
      "step": 70185
    },
    {
      "entropy": 5.876437664031982,
      "epoch": 6.133880975268723,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001987675496024508,
      "loss": 5.2074,
      "mean_token_accuracy": 0.21929918974637985,
      "num_tokens": 142232050.0,
      "step": 70190
    },
    {
      "entropy": 5.838903474807739,
      "epoch": 6.134317923621428,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019873823227451666,
      "loss": 5.068,
      "mean_token_accuracy": 0.22676663398742675,
      "num_tokens": 142242432.0,
      "step": 70195
    },
    {
      "entropy": 5.893690586090088,
      "epoch": 6.134754871974133,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001987089164093231,
      "loss": 5.2045,
      "mean_token_accuracy": 0.21395236998796463,
      "num_tokens": 142252444.0,
      "step": 70200
    },
    {
      "entropy": 5.847588729858399,
      "epoch": 6.135191820326837,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019867960200743256,
      "loss": 5.1452,
      "mean_token_accuracy": 0.22501395493745804,
      "num_tokens": 142262938.0,
      "step": 70205
    },
    {
      "entropy": 5.80824875831604,
      "epoch": 6.135628768679542,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001986502890694072,
      "loss": 5.1093,
      "mean_token_accuracy": 0.22606107592582703,
      "num_tokens": 142271882.0,
      "step": 70210
    },
    {
      "entropy": 5.780454921722412,
      "epoch": 6.136065717032247,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001986209775958094,
      "loss": 5.0533,
      "mean_token_accuracy": 0.23259256184101104,
      "num_tokens": 142280822.0,
      "step": 70215
    },
    {
      "entropy": 5.8472260475158695,
      "epoch": 6.136502665384952,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001985916675872012,
      "loss": 5.1549,
      "mean_token_accuracy": 0.2221015214920044,
      "num_tokens": 142290428.0,
      "step": 70220
    },
    {
      "entropy": 5.81612639427185,
      "epoch": 6.136939613737656,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001985623590441449,
      "loss": 5.1659,
      "mean_token_accuracy": 0.22371008694171907,
      "num_tokens": 142300955.0,
      "step": 70225
    },
    {
      "entropy": 5.877065420150757,
      "epoch": 6.137376562090361,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001985330519672025,
      "loss": 5.2114,
      "mean_token_accuracy": 0.22074114680290222,
      "num_tokens": 142311759.0,
      "step": 70230
    },
    {
      "entropy": 5.848851156234741,
      "epoch": 6.137813510443066,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019850374635693634,
      "loss": 5.1329,
      "mean_token_accuracy": 0.22855197489261628,
      "num_tokens": 142321524.0,
      "step": 70235
    },
    {
      "entropy": 5.888760328292847,
      "epoch": 6.1382504587957705,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019847444221390837,
      "loss": 5.1292,
      "mean_token_accuracy": 0.2321377858519554,
      "num_tokens": 142331627.0,
      "step": 70240
    },
    {
      "entropy": 5.9147766590118405,
      "epoch": 6.138687407148475,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019844513953868071,
      "loss": 5.2012,
      "mean_token_accuracy": 0.21946034729480743,
      "num_tokens": 142341696.0,
      "step": 70245
    },
    {
      "entropy": 5.840054082870483,
      "epoch": 6.139124355501179,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019841583833181536,
      "loss": 5.0859,
      "mean_token_accuracy": 0.22946443259716034,
      "num_tokens": 142351655.0,
      "step": 70250
    },
    {
      "entropy": 5.859507703781128,
      "epoch": 6.139561303853885,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019838653859387435,
      "loss": 5.1733,
      "mean_token_accuracy": 0.22677263766527175,
      "num_tokens": 142361916.0,
      "step": 70255
    },
    {
      "entropy": 5.809124422073364,
      "epoch": 6.139998252206589,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001983572403254198,
      "loss": 5.1564,
      "mean_token_accuracy": 0.23009066730737687,
      "num_tokens": 142372766.0,
      "step": 70260
    },
    {
      "entropy": 5.898736810684204,
      "epoch": 6.140435200559294,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001983279435270134,
      "loss": 5.1901,
      "mean_token_accuracy": 0.22071880251169204,
      "num_tokens": 142382963.0,
      "step": 70265
    },
    {
      "entropy": 5.925455045700073,
      "epoch": 6.140872148911998,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019829864819921722,
      "loss": 5.1574,
      "mean_token_accuracy": 0.2256655737757683,
      "num_tokens": 142391882.0,
      "step": 70270
    },
    {
      "entropy": 5.8322882652282715,
      "epoch": 6.1413090972647035,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019826935434259319,
      "loss": 5.0441,
      "mean_token_accuracy": 0.22673601508140565,
      "num_tokens": 142402131.0,
      "step": 70275
    },
    {
      "entropy": 5.822557163238526,
      "epoch": 6.141746045617408,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019824006195770315,
      "loss": 5.1126,
      "mean_token_accuracy": 0.22842299342155456,
      "num_tokens": 142412199.0,
      "step": 70280
    },
    {
      "entropy": 5.790284204483032,
      "epoch": 6.1421829939701125,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000198210771045109,
      "loss": 5.063,
      "mean_token_accuracy": 0.2257719084620476,
      "num_tokens": 142422274.0,
      "step": 70285
    },
    {
      "entropy": 5.863672542572021,
      "epoch": 6.142619942322818,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019818148160537246,
      "loss": 5.1505,
      "mean_token_accuracy": 0.21880586445331573,
      "num_tokens": 142432593.0,
      "step": 70290
    },
    {
      "entropy": 5.775078153610229,
      "epoch": 6.143056890675522,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019815219363905533,
      "loss": 5.1305,
      "mean_token_accuracy": 0.22497561872005462,
      "num_tokens": 142444315.0,
      "step": 70295
    },
    {
      "entropy": 5.878016519546509,
      "epoch": 6.143493839028227,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019812290714671943,
      "loss": 5.1816,
      "mean_token_accuracy": 0.2224859580397606,
      "num_tokens": 142455754.0,
      "step": 70300
    },
    {
      "entropy": 5.91046929359436,
      "epoch": 6.143930787380931,
      "grad_norm": 1.25,
      "learning_rate": 0.0001980936221289264,
      "loss": 5.1476,
      "mean_token_accuracy": 0.2208872213959694,
      "num_tokens": 142466506.0,
      "step": 70305
    },
    {
      "entropy": 5.8255615234375,
      "epoch": 6.144367735733637,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019806433858623806,
      "loss": 5.1157,
      "mean_token_accuracy": 0.22725120186805725,
      "num_tokens": 142477765.0,
      "step": 70310
    },
    {
      "entropy": 5.779866456985474,
      "epoch": 6.144804684086341,
      "grad_norm": 1.28125,
      "learning_rate": 0.000198035056519216,
      "loss": 5.0554,
      "mean_token_accuracy": 0.23812561333179474,
      "num_tokens": 142487486.0,
      "step": 70315
    },
    {
      "entropy": 5.858245944976806,
      "epoch": 6.1452416324390455,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019800577592842184,
      "loss": 5.2073,
      "mean_token_accuracy": 0.21669902056455612,
      "num_tokens": 142498135.0,
      "step": 70320
    },
    {
      "entropy": 5.81091079711914,
      "epoch": 6.14567858079175,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019797649681441726,
      "loss": 5.0668,
      "mean_token_accuracy": 0.23209204524755478,
      "num_tokens": 142508211.0,
      "step": 70325
    },
    {
      "entropy": 5.838972425460815,
      "epoch": 6.146115529144455,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019794721917776388,
      "loss": 5.1051,
      "mean_token_accuracy": 0.22184862196445465,
      "num_tokens": 142517954.0,
      "step": 70330
    },
    {
      "entropy": 5.860883903503418,
      "epoch": 6.14655247749716,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001979179430190232,
      "loss": 5.1383,
      "mean_token_accuracy": 0.22956187427043914,
      "num_tokens": 142527595.0,
      "step": 70335
    },
    {
      "entropy": 5.860525894165039,
      "epoch": 6.146989425849864,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019788866833875675,
      "loss": 5.1528,
      "mean_token_accuracy": 0.22470872104167938,
      "num_tokens": 142538177.0,
      "step": 70340
    },
    {
      "entropy": 5.900362920761109,
      "epoch": 6.14742637420257,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000197859395137526,
      "loss": 5.1936,
      "mean_token_accuracy": 0.21829127818346022,
      "num_tokens": 142547972.0,
      "step": 70345
    },
    {
      "entropy": 5.84137806892395,
      "epoch": 6.147863322555274,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019783012341589252,
      "loss": 5.1263,
      "mean_token_accuracy": 0.2199208214879036,
      "num_tokens": 142558316.0,
      "step": 70350
    },
    {
      "entropy": 5.923776578903198,
      "epoch": 6.148300270907979,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019780085317441765,
      "loss": 5.2291,
      "mean_token_accuracy": 0.21613916009664536,
      "num_tokens": 142568824.0,
      "step": 70355
    },
    {
      "entropy": 5.872442388534546,
      "epoch": 6.148737219260683,
      "grad_norm": 1.125,
      "learning_rate": 0.00019777158441366299,
      "loss": 5.1217,
      "mean_token_accuracy": 0.22008032202720643,
      "num_tokens": 142578677.0,
      "step": 70360
    },
    {
      "entropy": 5.869805860519409,
      "epoch": 6.149174167613388,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019774231713418972,
      "loss": 5.1761,
      "mean_token_accuracy": 0.22277945578098296,
      "num_tokens": 142588865.0,
      "step": 70365
    },
    {
      "entropy": 5.804379558563232,
      "epoch": 6.149611115966093,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019771305133655932,
      "loss": 5.1278,
      "mean_token_accuracy": 0.2292131468653679,
      "num_tokens": 142599053.0,
      "step": 70370
    },
    {
      "entropy": 5.87567982673645,
      "epoch": 6.150048064318797,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019768378702133304,
      "loss": 5.1433,
      "mean_token_accuracy": 0.2250397101044655,
      "num_tokens": 142609099.0,
      "step": 70375
    },
    {
      "entropy": 5.889419174194336,
      "epoch": 6.150485012671502,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019765452418907226,
      "loss": 5.152,
      "mean_token_accuracy": 0.22700399309396743,
      "num_tokens": 142618420.0,
      "step": 70380
    },
    {
      "entropy": 5.966778945922852,
      "epoch": 6.150921961024207,
      "grad_norm": 1.09375,
      "learning_rate": 0.00019762526284033833,
      "loss": 5.2995,
      "mean_token_accuracy": 0.21442655920982362,
      "num_tokens": 142628841.0,
      "step": 70385
    },
    {
      "entropy": 5.845328617095947,
      "epoch": 6.151358909376912,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001975960029756923,
      "loss": 5.127,
      "mean_token_accuracy": 0.22789558619260789,
      "num_tokens": 142639032.0,
      "step": 70390
    },
    {
      "entropy": 5.876882457733155,
      "epoch": 6.151795857729616,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019756674459569547,
      "loss": 5.1406,
      "mean_token_accuracy": 0.22602452635765075,
      "num_tokens": 142648383.0,
      "step": 70395
    },
    {
      "entropy": 5.812997531890869,
      "epoch": 6.1522328060823215,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019753748770090908,
      "loss": 5.0997,
      "mean_token_accuracy": 0.2271941915154457,
      "num_tokens": 142658735.0,
      "step": 70400
    },
    {
      "entropy": 5.84255576133728,
      "epoch": 6.152669754435026,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019750823229189435,
      "loss": 5.0865,
      "mean_token_accuracy": 0.22949060201644897,
      "num_tokens": 142668633.0,
      "step": 70405
    },
    {
      "entropy": 5.798914051055908,
      "epoch": 6.15310670278773,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019747897836921226,
      "loss": 5.1055,
      "mean_token_accuracy": 0.2276270106434822,
      "num_tokens": 142678488.0,
      "step": 70410
    },
    {
      "entropy": 5.8052167892456055,
      "epoch": 6.153543651140435,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019744972593342398,
      "loss": 5.0863,
      "mean_token_accuracy": 0.2248993217945099,
      "num_tokens": 142688784.0,
      "step": 70415
    },
    {
      "entropy": 5.881355905532837,
      "epoch": 6.15398059949314,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00019742047498509058,
      "loss": 5.154,
      "mean_token_accuracy": 0.2242879316210747,
      "num_tokens": 142698651.0,
      "step": 70420
    },
    {
      "entropy": 5.850592231750488,
      "epoch": 6.154417547845845,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001973912255247731,
      "loss": 5.1508,
      "mean_token_accuracy": 0.22788804173469543,
      "num_tokens": 142708336.0,
      "step": 70425
    },
    {
      "entropy": 5.863472604751587,
      "epoch": 6.154854496198549,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001973619775530327,
      "loss": 5.1177,
      "mean_token_accuracy": 0.22658934742212294,
      "num_tokens": 142718720.0,
      "step": 70430
    },
    {
      "entropy": 5.859143495559692,
      "epoch": 6.155291444551254,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001973327310704302,
      "loss": 5.1926,
      "mean_token_accuracy": 0.2200826495885849,
      "num_tokens": 142729494.0,
      "step": 70435
    },
    {
      "entropy": 5.843260097503662,
      "epoch": 6.155728392903959,
      "grad_norm": 1.078125,
      "learning_rate": 0.00019730348607752658,
      "loss": 5.1074,
      "mean_token_accuracy": 0.21997429728507994,
      "num_tokens": 142739069.0,
      "step": 70440
    },
    {
      "entropy": 5.898543453216552,
      "epoch": 6.1561653412566635,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001972742425748828,
      "loss": 5.2244,
      "mean_token_accuracy": 0.2148798018693924,
      "num_tokens": 142749523.0,
      "step": 70445
    },
    {
      "entropy": 5.899386739730835,
      "epoch": 6.156602289609368,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019724500056305977,
      "loss": 5.1277,
      "mean_token_accuracy": 0.2173467606306076,
      "num_tokens": 142759235.0,
      "step": 70450
    },
    {
      "entropy": 5.833049297332764,
      "epoch": 6.157039237962073,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019721576004261843,
      "loss": 5.1167,
      "mean_token_accuracy": 0.2276535600423813,
      "num_tokens": 142768306.0,
      "step": 70455
    },
    {
      "entropy": 5.816689920425415,
      "epoch": 6.157476186314778,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019718652101411954,
      "loss": 5.146,
      "mean_token_accuracy": 0.22615491300821305,
      "num_tokens": 142778931.0,
      "step": 70460
    },
    {
      "entropy": 5.824364566802979,
      "epoch": 6.157913134667482,
      "grad_norm": 1.125,
      "learning_rate": 0.0001971572834781239,
      "loss": 5.1261,
      "mean_token_accuracy": 0.2307574063539505,
      "num_tokens": 142789474.0,
      "step": 70465
    },
    {
      "entropy": 5.792242956161499,
      "epoch": 6.158350083020187,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001971280474351924,
      "loss": 5.0894,
      "mean_token_accuracy": 0.22945586442947388,
      "num_tokens": 142799646.0,
      "step": 70470
    },
    {
      "entropy": 5.833582496643066,
      "epoch": 6.158787031372892,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019709881288588576,
      "loss": 5.0673,
      "mean_token_accuracy": 0.2329195037484169,
      "num_tokens": 142809501.0,
      "step": 70475
    },
    {
      "entropy": 5.878305196762085,
      "epoch": 6.1592239797255965,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019706957983076472,
      "loss": 5.2088,
      "mean_token_accuracy": 0.21602862030267717,
      "num_tokens": 142820023.0,
      "step": 70480
    },
    {
      "entropy": 5.837976408004761,
      "epoch": 6.159660928078301,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019704034827038998,
      "loss": 5.2074,
      "mean_token_accuracy": 0.2255576267838478,
      "num_tokens": 142830588.0,
      "step": 70485
    },
    {
      "entropy": 5.850547122955322,
      "epoch": 6.1600978764310055,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019701111820532213,
      "loss": 5.1318,
      "mean_token_accuracy": 0.2272825837135315,
      "num_tokens": 142840127.0,
      "step": 70490
    },
    {
      "entropy": 5.87806077003479,
      "epoch": 6.160534824783711,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019698188963612195,
      "loss": 5.1663,
      "mean_token_accuracy": 0.22249420434236528,
      "num_tokens": 142849509.0,
      "step": 70495
    },
    {
      "entropy": 5.785623455047608,
      "epoch": 6.160971773136415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019695266256335004,
      "loss": 5.0077,
      "mean_token_accuracy": 0.2371005818247795,
      "num_tokens": 142859081.0,
      "step": 70500
    },
    {
      "entropy": 5.79752287864685,
      "epoch": 6.16140872148912,
      "grad_norm": 1.125,
      "learning_rate": 0.00019692343698756698,
      "loss": 5.1476,
      "mean_token_accuracy": 0.22955053001642228,
      "num_tokens": 142870522.0,
      "step": 70505
    },
    {
      "entropy": 5.823758935928344,
      "epoch": 6.161845669841824,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019689421290933312,
      "loss": 5.1451,
      "mean_token_accuracy": 0.2309833660721779,
      "num_tokens": 142880403.0,
      "step": 70510
    },
    {
      "entropy": 5.857873773574829,
      "epoch": 6.16228261819453,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019686499032920934,
      "loss": 5.1738,
      "mean_token_accuracy": 0.2223501458764076,
      "num_tokens": 142890904.0,
      "step": 70515
    },
    {
      "entropy": 5.881691741943359,
      "epoch": 6.162719566547234,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019683576924775604,
      "loss": 5.1469,
      "mean_token_accuracy": 0.2299436777830124,
      "num_tokens": 142900947.0,
      "step": 70520
    },
    {
      "entropy": 5.813160848617554,
      "epoch": 6.1631565148999385,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019680654966553352,
      "loss": 5.0896,
      "mean_token_accuracy": 0.22936705201864244,
      "num_tokens": 142911985.0,
      "step": 70525
    },
    {
      "entropy": 5.880483818054199,
      "epoch": 6.163593463252644,
      "grad_norm": 1.25,
      "learning_rate": 0.0001967773315831024,
      "loss": 5.172,
      "mean_token_accuracy": 0.2209697499871254,
      "num_tokens": 142921386.0,
      "step": 70530
    },
    {
      "entropy": 5.865697145462036,
      "epoch": 6.164030411605348,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001967481150010231,
      "loss": 5.1169,
      "mean_token_accuracy": 0.22254625856876373,
      "num_tokens": 142931017.0,
      "step": 70535
    },
    {
      "entropy": 5.89517765045166,
      "epoch": 6.164467359958053,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019671889991985592,
      "loss": 5.1662,
      "mean_token_accuracy": 0.22012299597263335,
      "num_tokens": 142941594.0,
      "step": 70540
    },
    {
      "entropy": 5.888795566558838,
      "epoch": 6.164904308310757,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019668968634016122,
      "loss": 5.1654,
      "mean_token_accuracy": 0.22235660403966903,
      "num_tokens": 142951918.0,
      "step": 70545
    },
    {
      "entropy": 5.828051996231079,
      "epoch": 6.165341256663463,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001966604742624994,
      "loss": 5.1043,
      "mean_token_accuracy": 0.22110507041215896,
      "num_tokens": 142962298.0,
      "step": 70550
    },
    {
      "entropy": 5.885697555541992,
      "epoch": 6.165778205016167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019663126368743073,
      "loss": 5.1576,
      "mean_token_accuracy": 0.22134243696928024,
      "num_tokens": 142971443.0,
      "step": 70555
    },
    {
      "entropy": 5.85417628288269,
      "epoch": 6.166215153368872,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019660205461551544,
      "loss": 5.167,
      "mean_token_accuracy": 0.2263900876045227,
      "num_tokens": 142981469.0,
      "step": 70560
    },
    {
      "entropy": 5.778123903274536,
      "epoch": 6.166652101721576,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019657284704731394,
      "loss": 5.0209,
      "mean_token_accuracy": 0.23501022905111313,
      "num_tokens": 142990621.0,
      "step": 70565
    },
    {
      "entropy": 5.8260749816894535,
      "epoch": 6.167089050074281,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019654364098338622,
      "loss": 5.1656,
      "mean_token_accuracy": 0.21991800963878633,
      "num_tokens": 142999576.0,
      "step": 70570
    },
    {
      "entropy": 5.9041663646698,
      "epoch": 6.167525998426986,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019651443642429263,
      "loss": 5.2254,
      "mean_token_accuracy": 0.22346727699041366,
      "num_tokens": 143009658.0,
      "step": 70575
    },
    {
      "entropy": 5.8802203178405765,
      "epoch": 6.16796294677969,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019648523337059326,
      "loss": 5.1742,
      "mean_token_accuracy": 0.22659541964530944,
      "num_tokens": 143019683.0,
      "step": 70580
    },
    {
      "entropy": 5.8231024742126465,
      "epoch": 6.168399895132396,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019645603182284828,
      "loss": 5.0377,
      "mean_token_accuracy": 0.23523132503032684,
      "num_tokens": 143028971.0,
      "step": 70585
    },
    {
      "entropy": 5.921360874176026,
      "epoch": 6.1688368434851,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019642683178161774,
      "loss": 5.18,
      "mean_token_accuracy": 0.22344377636909485,
      "num_tokens": 143039616.0,
      "step": 70590
    },
    {
      "entropy": 5.858158016204834,
      "epoch": 6.169273791837805,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019639763324746175,
      "loss": 5.1631,
      "mean_token_accuracy": 0.21832109689712526,
      "num_tokens": 143049303.0,
      "step": 70595
    },
    {
      "entropy": 5.888166427612305,
      "epoch": 6.169710740190509,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001963684362209403,
      "loss": 5.2185,
      "mean_token_accuracy": 0.2260141223669052,
      "num_tokens": 143059635.0,
      "step": 70600
    },
    {
      "entropy": 5.882112789154053,
      "epoch": 6.1701476885432145,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019633924070261343,
      "loss": 5.2077,
      "mean_token_accuracy": 0.21988680958747864,
      "num_tokens": 143070831.0,
      "step": 70605
    },
    {
      "entropy": 5.99140887260437,
      "epoch": 6.170584636895919,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001963100466930412,
      "loss": 5.29,
      "mean_token_accuracy": 0.21077989637851716,
      "num_tokens": 143080871.0,
      "step": 70610
    },
    {
      "entropy": 5.874715375900268,
      "epoch": 6.171021585248623,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019628085419278357,
      "loss": 5.1252,
      "mean_token_accuracy": 0.2199105814099312,
      "num_tokens": 143090215.0,
      "step": 70615
    },
    {
      "entropy": 5.8366209983825685,
      "epoch": 6.171458533601328,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019625166320240035,
      "loss": 5.1627,
      "mean_token_accuracy": 0.2220013067126274,
      "num_tokens": 143100432.0,
      "step": 70620
    },
    {
      "entropy": 5.813926076889038,
      "epoch": 6.171895481954033,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019622247372245145,
      "loss": 5.0654,
      "mean_token_accuracy": 0.23028520345687867,
      "num_tokens": 143110581.0,
      "step": 70625
    },
    {
      "entropy": 5.808697128295899,
      "epoch": 6.172332430306738,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00019619328575349676,
      "loss": 5.1062,
      "mean_token_accuracy": 0.22478285878896714,
      "num_tokens": 143122100.0,
      "step": 70630
    },
    {
      "entropy": 5.8092187404632565,
      "epoch": 6.172769378659442,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001961640992960962,
      "loss": 5.0525,
      "mean_token_accuracy": 0.2253273293375969,
      "num_tokens": 143131699.0,
      "step": 70635
    },
    {
      "entropy": 5.883888816833496,
      "epoch": 6.1732063270121476,
      "grad_norm": 1.25,
      "learning_rate": 0.00019613491435080955,
      "loss": 5.1897,
      "mean_token_accuracy": 0.2233828201889992,
      "num_tokens": 143141256.0,
      "step": 70640
    },
    {
      "entropy": 5.8956618309021,
      "epoch": 6.173643275364852,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001961057309181965,
      "loss": 5.1624,
      "mean_token_accuracy": 0.217754465341568,
      "num_tokens": 143151966.0,
      "step": 70645
    },
    {
      "entropy": 5.877072715759278,
      "epoch": 6.1740802237175565,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001960765489988169,
      "loss": 5.0948,
      "mean_token_accuracy": 0.22340027391910552,
      "num_tokens": 143162085.0,
      "step": 70650
    },
    {
      "entropy": 5.781615161895752,
      "epoch": 6.174517172070261,
      "grad_norm": 1.25,
      "learning_rate": 0.00019604736859323046,
      "loss": 5.0541,
      "mean_token_accuracy": 0.23535882979631423,
      "num_tokens": 143171868.0,
      "step": 70655
    },
    {
      "entropy": 5.869507598876953,
      "epoch": 6.174954120422966,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019601818970199682,
      "loss": 5.2345,
      "mean_token_accuracy": 0.21825449168682098,
      "num_tokens": 143182021.0,
      "step": 70660
    },
    {
      "entropy": 5.820664310455323,
      "epoch": 6.175391068775671,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019598901232567578,
      "loss": 5.1605,
      "mean_token_accuracy": 0.22800150364637375,
      "num_tokens": 143192599.0,
      "step": 70665
    },
    {
      "entropy": 5.915396308898925,
      "epoch": 6.175828017128375,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001959598364648268,
      "loss": 5.1744,
      "mean_token_accuracy": 0.22196293622255325,
      "num_tokens": 143202549.0,
      "step": 70670
    },
    {
      "entropy": 5.887094974517822,
      "epoch": 6.17626496548108,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001959306621200096,
      "loss": 5.2171,
      "mean_token_accuracy": 0.21903410404920579,
      "num_tokens": 143212832.0,
      "step": 70675
    },
    {
      "entropy": 5.881810331344605,
      "epoch": 6.176701913833785,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001959014892917837,
      "loss": 5.1851,
      "mean_token_accuracy": 0.2173925220966339,
      "num_tokens": 143225321.0,
      "step": 70680
    },
    {
      "entropy": 5.8541618347167965,
      "epoch": 6.1771388621864896,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001958723179807087,
      "loss": 5.1248,
      "mean_token_accuracy": 0.2264427736401558,
      "num_tokens": 143234610.0,
      "step": 70685
    },
    {
      "entropy": 5.85370888710022,
      "epoch": 6.177575810539194,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019584314818734417,
      "loss": 5.0859,
      "mean_token_accuracy": 0.22755958884954453,
      "num_tokens": 143244076.0,
      "step": 70690
    },
    {
      "entropy": 5.851344585418701,
      "epoch": 6.178012758891899,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019581397991224946,
      "loss": 5.2672,
      "mean_token_accuracy": 0.21230929493904113,
      "num_tokens": 143254495.0,
      "step": 70695
    },
    {
      "entropy": 5.777307558059692,
      "epoch": 6.178449707244604,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019578481315598412,
      "loss": 5.0956,
      "mean_token_accuracy": 0.228841070830822,
      "num_tokens": 143264783.0,
      "step": 70700
    },
    {
      "entropy": 5.859214353561401,
      "epoch": 6.178886655597308,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019575564791910754,
      "loss": 5.1011,
      "mean_token_accuracy": 0.22192272841930388,
      "num_tokens": 143274821.0,
      "step": 70705
    },
    {
      "entropy": 5.830496406555175,
      "epoch": 6.179323603950013,
      "grad_norm": 1.25,
      "learning_rate": 0.00019572648420217926,
      "loss": 5.0622,
      "mean_token_accuracy": 0.22965992093086243,
      "num_tokens": 143284339.0,
      "step": 70710
    },
    {
      "entropy": 5.860066986083984,
      "epoch": 6.179760552302718,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019569732200575845,
      "loss": 5.2332,
      "mean_token_accuracy": 0.21966606825590135,
      "num_tokens": 143295004.0,
      "step": 70715
    },
    {
      "entropy": 5.821693849563599,
      "epoch": 6.180197500655423,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001956681613304046,
      "loss": 5.0885,
      "mean_token_accuracy": 0.23006187826395036,
      "num_tokens": 143305014.0,
      "step": 70720
    },
    {
      "entropy": 5.812578248977661,
      "epoch": 6.180634449008127,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019563900217667697,
      "loss": 5.0963,
      "mean_token_accuracy": 0.22986339926719665,
      "num_tokens": 143315429.0,
      "step": 70725
    },
    {
      "entropy": 5.813144493103027,
      "epoch": 6.1810713973608316,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019560984454513485,
      "loss": 5.1261,
      "mean_token_accuracy": 0.22577976286411286,
      "num_tokens": 143326325.0,
      "step": 70730
    },
    {
      "entropy": 5.801329565048218,
      "epoch": 6.181508345713537,
      "grad_norm": 1.125,
      "learning_rate": 0.00019558068843633758,
      "loss": 5.0637,
      "mean_token_accuracy": 0.2325591742992401,
      "num_tokens": 143335764.0,
      "step": 70735
    },
    {
      "entropy": 5.810430526733398,
      "epoch": 6.181945294066241,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019555153385084428,
      "loss": 5.1626,
      "mean_token_accuracy": 0.2191239982843399,
      "num_tokens": 143346430.0,
      "step": 70740
    },
    {
      "entropy": 5.861912679672241,
      "epoch": 6.182382242418946,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019552238078921418,
      "loss": 5.153,
      "mean_token_accuracy": 0.2304980933666229,
      "num_tokens": 143356235.0,
      "step": 70745
    },
    {
      "entropy": 5.806515884399414,
      "epoch": 6.182819190771651,
      "grad_norm": 1.125,
      "learning_rate": 0.00019549322925200646,
      "loss": 5.1585,
      "mean_token_accuracy": 0.22402188926935196,
      "num_tokens": 143366110.0,
      "step": 70750
    },
    {
      "entropy": 5.898409032821656,
      "epoch": 6.183256139124356,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019546407923978028,
      "loss": 5.1739,
      "mean_token_accuracy": 0.22278835028409957,
      "num_tokens": 143375338.0,
      "step": 70755
    },
    {
      "entropy": 5.847723817825317,
      "epoch": 6.18369308747706,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019543493075309477,
      "loss": 5.2053,
      "mean_token_accuracy": 0.2204931065440178,
      "num_tokens": 143386022.0,
      "step": 70760
    },
    {
      "entropy": 5.842398977279663,
      "epoch": 6.184130035829765,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001954057837925089,
      "loss": 5.125,
      "mean_token_accuracy": 0.2240934357047081,
      "num_tokens": 143396386.0,
      "step": 70765
    },
    {
      "entropy": 5.828464698791504,
      "epoch": 6.18456698418247,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019537663835858182,
      "loss": 5.0942,
      "mean_token_accuracy": 0.22511324733495713,
      "num_tokens": 143406472.0,
      "step": 70770
    },
    {
      "entropy": 5.873295879364013,
      "epoch": 6.185003932535174,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00019534749445187255,
      "loss": 5.1904,
      "mean_token_accuracy": 0.22455264627933502,
      "num_tokens": 143416787.0,
      "step": 70775
    },
    {
      "entropy": 5.7956780910491945,
      "epoch": 6.185440880887879,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019531835207294007,
      "loss": 5.0436,
      "mean_token_accuracy": 0.2360600784420967,
      "num_tokens": 143426425.0,
      "step": 70780
    },
    {
      "entropy": 5.909214687347412,
      "epoch": 6.185877829240583,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001952892112223434,
      "loss": 5.2309,
      "mean_token_accuracy": 0.2227615788578987,
      "num_tokens": 143437601.0,
      "step": 70785
    },
    {
      "entropy": 5.88171763420105,
      "epoch": 6.186314777593289,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001952600719006414,
      "loss": 5.1751,
      "mean_token_accuracy": 0.21663739532232285,
      "num_tokens": 143447314.0,
      "step": 70790
    },
    {
      "entropy": 5.870941972732544,
      "epoch": 6.186751725945993,
      "grad_norm": 1.203125,
      "learning_rate": 0.000195230934108393,
      "loss": 5.2006,
      "mean_token_accuracy": 0.2185673698782921,
      "num_tokens": 143457047.0,
      "step": 70795
    },
    {
      "entropy": 5.845612955093384,
      "epoch": 6.187188674298698,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019520179784615705,
      "loss": 5.0969,
      "mean_token_accuracy": 0.23328502625226974,
      "num_tokens": 143467143.0,
      "step": 70800
    },
    {
      "entropy": 5.827211809158325,
      "epoch": 6.187625622651403,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019517266311449246,
      "loss": 5.1114,
      "mean_token_accuracy": 0.22825891822576522,
      "num_tokens": 143476811.0,
      "step": 70805
    },
    {
      "entropy": 5.813522815704346,
      "epoch": 6.1880625710041075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019514352991395806,
      "loss": 5.1601,
      "mean_token_accuracy": 0.21992840468883515,
      "num_tokens": 143487272.0,
      "step": 70810
    },
    {
      "entropy": 5.833193159103393,
      "epoch": 6.188499519356812,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001951143982451125,
      "loss": 5.1358,
      "mean_token_accuracy": 0.22754602283239364,
      "num_tokens": 143498058.0,
      "step": 70815
    },
    {
      "entropy": 5.882606315612793,
      "epoch": 6.188936467709516,
      "grad_norm": 1.078125,
      "learning_rate": 0.00019508526810851473,
      "loss": 5.1015,
      "mean_token_accuracy": 0.22768177092075348,
      "num_tokens": 143508295.0,
      "step": 70820
    },
    {
      "entropy": 5.895022010803222,
      "epoch": 6.189373416062222,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019505613950472335,
      "loss": 5.1361,
      "mean_token_accuracy": 0.2213056743144989,
      "num_tokens": 143518109.0,
      "step": 70825
    },
    {
      "entropy": 5.807128095626831,
      "epoch": 6.189810364414926,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001950270124342971,
      "loss": 5.0925,
      "mean_token_accuracy": 0.22927333563566207,
      "num_tokens": 143527769.0,
      "step": 70830
    },
    {
      "entropy": 5.805783796310425,
      "epoch": 6.190247312767631,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019499788689779468,
      "loss": 5.1013,
      "mean_token_accuracy": 0.2261640816926956,
      "num_tokens": 143536619.0,
      "step": 70835
    },
    {
      "entropy": 5.875133943557739,
      "epoch": 6.190684261120335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001949687628957747,
      "loss": 5.2375,
      "mean_token_accuracy": 0.21660100817680358,
      "num_tokens": 143547674.0,
      "step": 70840
    },
    {
      "entropy": 5.799744081497193,
      "epoch": 6.191121209473041,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001949396404287958,
      "loss": 5.0499,
      "mean_token_accuracy": 0.23244986683130264,
      "num_tokens": 143557786.0,
      "step": 70845
    },
    {
      "entropy": 5.903755569458008,
      "epoch": 6.191558157825745,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001949105194974165,
      "loss": 5.1811,
      "mean_token_accuracy": 0.21850071102380753,
      "num_tokens": 143568055.0,
      "step": 70850
    },
    {
      "entropy": 5.851580095291138,
      "epoch": 6.1919951061784495,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001948814001021955,
      "loss": 5.1226,
      "mean_token_accuracy": 0.22471374720335008,
      "num_tokens": 143578327.0,
      "step": 70855
    },
    {
      "entropy": 5.788921356201172,
      "epoch": 6.192432054531155,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019485228224369117,
      "loss": 5.0882,
      "mean_token_accuracy": 0.23120368421077728,
      "num_tokens": 143588105.0,
      "step": 70860
    },
    {
      "entropy": 5.8248481273651125,
      "epoch": 6.192869002883859,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019482316592246191,
      "loss": 5.1314,
      "mean_token_accuracy": 0.22293531447649,
      "num_tokens": 143598125.0,
      "step": 70865
    },
    {
      "entropy": 5.797477388381958,
      "epoch": 6.193305951236564,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001947940511390665,
      "loss": 5.0573,
      "mean_token_accuracy": 0.23321965485811233,
      "num_tokens": 143608111.0,
      "step": 70870
    },
    {
      "entropy": 5.854147815704346,
      "epoch": 6.193742899589268,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001947649378940632,
      "loss": 5.1438,
      "mean_token_accuracy": 0.22590012848377228,
      "num_tokens": 143617666.0,
      "step": 70875
    },
    {
      "entropy": 5.860308265686035,
      "epoch": 6.194179847941974,
      "grad_norm": 1.125,
      "learning_rate": 0.00019473582618801045,
      "loss": 5.0939,
      "mean_token_accuracy": 0.22361280620098115,
      "num_tokens": 143627801.0,
      "step": 70880
    },
    {
      "entropy": 5.820841550827026,
      "epoch": 6.194616796294678,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019470671602146662,
      "loss": 5.0833,
      "mean_token_accuracy": 0.23280992358922958,
      "num_tokens": 143637682.0,
      "step": 70885
    },
    {
      "entropy": 5.801399612426758,
      "epoch": 6.195053744647383,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019467760739498997,
      "loss": 5.0852,
      "mean_token_accuracy": 0.23291227221488953,
      "num_tokens": 143647019.0,
      "step": 70890
    },
    {
      "entropy": 5.864256715774536,
      "epoch": 6.195490693000087,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000194648500309139,
      "loss": 5.2239,
      "mean_token_accuracy": 0.22208388894796371,
      "num_tokens": 143657335.0,
      "step": 70895
    },
    {
      "entropy": 5.923485326766968,
      "epoch": 6.195927641352792,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019461939476447189,
      "loss": 5.2469,
      "mean_token_accuracy": 0.21757477670907974,
      "num_tokens": 143667714.0,
      "step": 70900
    },
    {
      "entropy": 5.83145546913147,
      "epoch": 6.196364589705497,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019459029076154678,
      "loss": 5.073,
      "mean_token_accuracy": 0.23082434087991716,
      "num_tokens": 143677296.0,
      "step": 70905
    },
    {
      "entropy": 5.873388957977295,
      "epoch": 6.196801538058201,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019456118830092213,
      "loss": 5.1486,
      "mean_token_accuracy": 0.22738081961870193,
      "num_tokens": 143687023.0,
      "step": 70910
    },
    {
      "entropy": 5.903255987167358,
      "epoch": 6.197238486410907,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019453208738315602,
      "loss": 5.1959,
      "mean_token_accuracy": 0.21324385404586793,
      "num_tokens": 143697517.0,
      "step": 70915
    },
    {
      "entropy": 5.869281053543091,
      "epoch": 6.197675434763611,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019450298800880673,
      "loss": 5.152,
      "mean_token_accuracy": 0.22228823006153106,
      "num_tokens": 143706355.0,
      "step": 70920
    },
    {
      "entropy": 5.820754957199097,
      "epoch": 6.198112383116316,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019447389017843215,
      "loss": 5.0869,
      "mean_token_accuracy": 0.2320231854915619,
      "num_tokens": 143716153.0,
      "step": 70925
    },
    {
      "entropy": 5.860919666290283,
      "epoch": 6.19854933146902,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019444479389259063,
      "loss": 5.2026,
      "mean_token_accuracy": 0.21897656321525574,
      "num_tokens": 143726279.0,
      "step": 70930
    },
    {
      "entropy": 5.851670169830323,
      "epoch": 6.1989862798217255,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019441569915184016,
      "loss": 5.1317,
      "mean_token_accuracy": 0.22492740601301192,
      "num_tokens": 143736089.0,
      "step": 70935
    },
    {
      "entropy": 5.840753078460693,
      "epoch": 6.19942322817443,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019438660595673878,
      "loss": 5.1042,
      "mean_token_accuracy": 0.23023911863565444,
      "num_tokens": 143747444.0,
      "step": 70940
    },
    {
      "entropy": 5.767662143707275,
      "epoch": 6.199860176527134,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019435751430784464,
      "loss": 5.0791,
      "mean_token_accuracy": 0.23743632733821868,
      "num_tokens": 143757828.0,
      "step": 70945
    },
    {
      "entropy": 5.87288088798523,
      "epoch": 6.200297124879839,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019432842420571557,
      "loss": 5.1609,
      "mean_token_accuracy": 0.22194624692201614,
      "num_tokens": 143767353.0,
      "step": 70950
    },
    {
      "entropy": 5.810948324203491,
      "epoch": 6.200734073232544,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019429933565090956,
      "loss": 5.1587,
      "mean_token_accuracy": 0.2295766294002533,
      "num_tokens": 143777814.0,
      "step": 70955
    },
    {
      "entropy": 5.843735551834106,
      "epoch": 6.201171021585249,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019427024864398457,
      "loss": 5.12,
      "mean_token_accuracy": 0.22869102358818055,
      "num_tokens": 143788414.0,
      "step": 70960
    },
    {
      "entropy": 5.89139986038208,
      "epoch": 6.201607969937953,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019424116318549856,
      "loss": 5.1524,
      "mean_token_accuracy": 0.22452181875705718,
      "num_tokens": 143799072.0,
      "step": 70965
    },
    {
      "entropy": 5.79620132446289,
      "epoch": 6.202044918290658,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019421207927600937,
      "loss": 5.1618,
      "mean_token_accuracy": 0.22477699518203736,
      "num_tokens": 143810347.0,
      "step": 70970
    },
    {
      "entropy": 5.852422571182251,
      "epoch": 6.202481866643363,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019418299691607483,
      "loss": 5.1108,
      "mean_token_accuracy": 0.2276458203792572,
      "num_tokens": 143819790.0,
      "step": 70975
    },
    {
      "entropy": 5.890444898605347,
      "epoch": 6.2029188149960675,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001941539161062527,
      "loss": 5.1962,
      "mean_token_accuracy": 0.22744353115558624,
      "num_tokens": 143829499.0,
      "step": 70980
    },
    {
      "entropy": 5.872804164886475,
      "epoch": 6.203355763348772,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019412483684710087,
      "loss": 5.1851,
      "mean_token_accuracy": 0.22044666707515717,
      "num_tokens": 143839594.0,
      "step": 70985
    },
    {
      "entropy": 5.928125667572021,
      "epoch": 6.203792711701477,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019409575913917703,
      "loss": 5.1906,
      "mean_token_accuracy": 0.22110386341810226,
      "num_tokens": 143849731.0,
      "step": 70990
    },
    {
      "entropy": 5.827741479873657,
      "epoch": 6.204229660054182,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019406668298303892,
      "loss": 5.1633,
      "mean_token_accuracy": 0.2274952232837677,
      "num_tokens": 143860025.0,
      "step": 70995
    },
    {
      "entropy": 5.8480062007904055,
      "epoch": 6.204666608406886,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019403760837924427,
      "loss": 5.1103,
      "mean_token_accuracy": 0.2228046715259552,
      "num_tokens": 143869840.0,
      "step": 71000
    },
    {
      "entropy": 5.927457618713379,
      "epoch": 6.205103556759591,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019400853532835068,
      "loss": 5.192,
      "mean_token_accuracy": 0.21691412180662156,
      "num_tokens": 143879877.0,
      "step": 71005
    },
    {
      "entropy": 5.810088157653809,
      "epoch": 6.205540505112296,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001939794638309158,
      "loss": 5.1137,
      "mean_token_accuracy": 0.2324603393673897,
      "num_tokens": 143889271.0,
      "step": 71010
    },
    {
      "entropy": 5.8609686374664305,
      "epoch": 6.2059774534650005,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019395039388749735,
      "loss": 5.1655,
      "mean_token_accuracy": 0.22132503539323806,
      "num_tokens": 143899293.0,
      "step": 71015
    },
    {
      "entropy": 5.846691226959228,
      "epoch": 6.206414401817705,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019392132549865277,
      "loss": 5.0877,
      "mean_token_accuracy": 0.23054419308900834,
      "num_tokens": 143908439.0,
      "step": 71020
    },
    {
      "entropy": 5.816397047042846,
      "epoch": 6.2068513501704095,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001938922586649396,
      "loss": 5.1107,
      "mean_token_accuracy": 0.23103264421224595,
      "num_tokens": 143917497.0,
      "step": 71025
    },
    {
      "entropy": 5.84615387916565,
      "epoch": 6.207288298523115,
      "grad_norm": 1.109375,
      "learning_rate": 0.00019386319338691543,
      "loss": 5.1332,
      "mean_token_accuracy": 0.23229780048131943,
      "num_tokens": 143927303.0,
      "step": 71030
    },
    {
      "entropy": 5.909069442749024,
      "epoch": 6.207725246875819,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019383412966513775,
      "loss": 5.2234,
      "mean_token_accuracy": 0.21965792030096054,
      "num_tokens": 143939067.0,
      "step": 71035
    },
    {
      "entropy": 5.850472545623779,
      "epoch": 6.208162195228524,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019380506750016398,
      "loss": 5.1834,
      "mean_token_accuracy": 0.22155956476926802,
      "num_tokens": 143948942.0,
      "step": 71040
    },
    {
      "entropy": 5.850659894943237,
      "epoch": 6.208599143581229,
      "grad_norm": 1.125,
      "learning_rate": 0.0001937760068925516,
      "loss": 5.1911,
      "mean_token_accuracy": 0.22162858098745347,
      "num_tokens": 143959663.0,
      "step": 71045
    },
    {
      "entropy": 5.925589847564697,
      "epoch": 6.209036091933934,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001937469478428579,
      "loss": 5.19,
      "mean_token_accuracy": 0.21852268427610397,
      "num_tokens": 143969185.0,
      "step": 71050
    },
    {
      "entropy": 5.916302299499511,
      "epoch": 6.209473040286638,
      "grad_norm": 1.046875,
      "learning_rate": 0.00019371789035164033,
      "loss": 5.2384,
      "mean_token_accuracy": 0.2157490462064743,
      "num_tokens": 143980103.0,
      "step": 71055
    },
    {
      "entropy": 5.884988021850586,
      "epoch": 6.2099099886393425,
      "grad_norm": 1.0625,
      "learning_rate": 0.00019368883441945624,
      "loss": 5.1738,
      "mean_token_accuracy": 0.21959457248449327,
      "num_tokens": 143990692.0,
      "step": 71060
    },
    {
      "entropy": 5.866333293914795,
      "epoch": 6.210346936992048,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001936597800468629,
      "loss": 5.1871,
      "mean_token_accuracy": 0.22594084590673447,
      "num_tokens": 143999704.0,
      "step": 71065
    },
    {
      "entropy": 5.86130576133728,
      "epoch": 6.210783885344752,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001936307272344176,
      "loss": 5.1196,
      "mean_token_accuracy": 0.22857726663351058,
      "num_tokens": 144010266.0,
      "step": 71070
    },
    {
      "entropy": 5.846620845794678,
      "epoch": 6.211220833697457,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019360167598267758,
      "loss": 5.1083,
      "mean_token_accuracy": 0.22399005740880967,
      "num_tokens": 144019906.0,
      "step": 71075
    },
    {
      "entropy": 5.805566596984863,
      "epoch": 6.211657782050161,
      "grad_norm": 1.125,
      "learning_rate": 0.00019357262629220008,
      "loss": 5.1523,
      "mean_token_accuracy": 0.22649009376764298,
      "num_tokens": 144030233.0,
      "step": 71080
    },
    {
      "entropy": 5.805175256729126,
      "epoch": 6.212094730402867,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019354357816354228,
      "loss": 5.0567,
      "mean_token_accuracy": 0.23474544435739517,
      "num_tokens": 144039178.0,
      "step": 71085
    },
    {
      "entropy": 5.8053430080413815,
      "epoch": 6.212531678755571,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019351453159726135,
      "loss": 5.1087,
      "mean_token_accuracy": 0.22654493749141694,
      "num_tokens": 144049651.0,
      "step": 71090
    },
    {
      "entropy": 5.882725763320923,
      "epoch": 6.212968627108276,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001934854865939144,
      "loss": 5.1558,
      "mean_token_accuracy": 0.2201948806643486,
      "num_tokens": 144059358.0,
      "step": 71095
    },
    {
      "entropy": 5.850611543655395,
      "epoch": 6.213405575460981,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019345644315405848,
      "loss": 5.2138,
      "mean_token_accuracy": 0.21550416946411133,
      "num_tokens": 144069857.0,
      "step": 71100
    },
    {
      "entropy": 5.8697991371154785,
      "epoch": 6.213842523813685,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019342740127825076,
      "loss": 5.1404,
      "mean_token_accuracy": 0.2296293407678604,
      "num_tokens": 144080088.0,
      "step": 71105
    },
    {
      "entropy": 5.830302715301514,
      "epoch": 6.21427947216639,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00019339836096704822,
      "loss": 5.1537,
      "mean_token_accuracy": 0.22775769829750062,
      "num_tokens": 144091856.0,
      "step": 71110
    },
    {
      "entropy": 5.878991508483887,
      "epoch": 6.214716420519094,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001933693222210079,
      "loss": 5.1467,
      "mean_token_accuracy": 0.22619489282369615,
      "num_tokens": 144102482.0,
      "step": 71115
    },
    {
      "entropy": 5.851378583908081,
      "epoch": 6.2151533688718,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019334028504068668,
      "loss": 5.0961,
      "mean_token_accuracy": 0.23181291818618774,
      "num_tokens": 144112523.0,
      "step": 71120
    },
    {
      "entropy": 5.847527551651001,
      "epoch": 6.215590317224504,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019331124942664163,
      "loss": 5.1214,
      "mean_token_accuracy": 0.22004164308309554,
      "num_tokens": 144122733.0,
      "step": 71125
    },
    {
      "entropy": 5.810159635543823,
      "epoch": 6.216027265577209,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001932822153794296,
      "loss": 5.0959,
      "mean_token_accuracy": 0.23435844480991364,
      "num_tokens": 144133157.0,
      "step": 71130
    },
    {
      "entropy": 5.855932235717773,
      "epoch": 6.216464213929913,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019325318289960746,
      "loss": 5.1013,
      "mean_token_accuracy": 0.23031260669231415,
      "num_tokens": 144143061.0,
      "step": 71135
    },
    {
      "entropy": 5.82419376373291,
      "epoch": 6.2169011622826185,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001932241519877322,
      "loss": 5.086,
      "mean_token_accuracy": 0.23151866495609283,
      "num_tokens": 144153489.0,
      "step": 71140
    },
    {
      "entropy": 5.874068021774292,
      "epoch": 6.217338110635323,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019319512264436044,
      "loss": 5.1907,
      "mean_token_accuracy": 0.2204453319311142,
      "num_tokens": 144163099.0,
      "step": 71145
    },
    {
      "entropy": 5.821674776077271,
      "epoch": 6.217775058988027,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019316609487004922,
      "loss": 5.1182,
      "mean_token_accuracy": 0.22253739982843398,
      "num_tokens": 144173003.0,
      "step": 71150
    },
    {
      "entropy": 5.88300838470459,
      "epoch": 6.218212007340732,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019313706866535508,
      "loss": 5.2211,
      "mean_token_accuracy": 0.2203282505273819,
      "num_tokens": 144183170.0,
      "step": 71155
    },
    {
      "entropy": 5.913792276382447,
      "epoch": 6.218648955693437,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019310804403083494,
      "loss": 5.2291,
      "mean_token_accuracy": 0.2137565180659294,
      "num_tokens": 144193184.0,
      "step": 71160
    },
    {
      "entropy": 5.807673740386963,
      "epoch": 6.219085904046142,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019307902096704537,
      "loss": 5.0993,
      "mean_token_accuracy": 0.23211754113435745,
      "num_tokens": 144203920.0,
      "step": 71165
    },
    {
      "entropy": 5.782598161697388,
      "epoch": 6.219522852398846,
      "grad_norm": 1.125,
      "learning_rate": 0.00019304999947454316,
      "loss": 5.0415,
      "mean_token_accuracy": 0.23376505225896835,
      "num_tokens": 144213671.0,
      "step": 71170
    },
    {
      "entropy": 5.8682568073272705,
      "epoch": 6.2199598007515515,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001930209795538848,
      "loss": 5.1252,
      "mean_token_accuracy": 0.22903558611869812,
      "num_tokens": 144223765.0,
      "step": 71175
    },
    {
      "entropy": 5.8275433540344235,
      "epoch": 6.220396749104256,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019299196120562705,
      "loss": 5.101,
      "mean_token_accuracy": 0.23283799439668657,
      "num_tokens": 144233475.0,
      "step": 71180
    },
    {
      "entropy": 5.800694799423217,
      "epoch": 6.2208336974569605,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019296294443032653,
      "loss": 5.1073,
      "mean_token_accuracy": 0.23423585742712022,
      "num_tokens": 144243624.0,
      "step": 71185
    },
    {
      "entropy": 5.77419900894165,
      "epoch": 6.221270645809665,
      "grad_norm": 1.125,
      "learning_rate": 0.00019293392922853963,
      "loss": 5.0555,
      "mean_token_accuracy": 0.2339276283979416,
      "num_tokens": 144253534.0,
      "step": 71190
    },
    {
      "entropy": 5.865536737442016,
      "epoch": 6.22170759416237,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000192904915600823,
      "loss": 5.2469,
      "mean_token_accuracy": 0.2135173797607422,
      "num_tokens": 144263591.0,
      "step": 71195
    },
    {
      "entropy": 5.829912614822388,
      "epoch": 6.222144542515075,
      "grad_norm": 1.25,
      "learning_rate": 0.00019287590354773315,
      "loss": 5.1517,
      "mean_token_accuracy": 0.22421122789382936,
      "num_tokens": 144274437.0,
      "step": 71200
    },
    {
      "entropy": 5.877057933807373,
      "epoch": 6.222581490867779,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019284689306982644,
      "loss": 5.1808,
      "mean_token_accuracy": 0.22456511855125427,
      "num_tokens": 144284829.0,
      "step": 71205
    },
    {
      "entropy": 5.915535974502563,
      "epoch": 6.223018439220484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019281788416765944,
      "loss": 5.1823,
      "mean_token_accuracy": 0.22039780020713806,
      "num_tokens": 144295950.0,
      "step": 71210
    },
    {
      "entropy": 5.922739934921265,
      "epoch": 6.223455387573189,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001927888768417884,
      "loss": 5.1627,
      "mean_token_accuracy": 0.22248132526874542,
      "num_tokens": 144306421.0,
      "step": 71215
    },
    {
      "entropy": 5.874308109283447,
      "epoch": 6.2238923359258935,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019275987109276972,
      "loss": 5.1681,
      "mean_token_accuracy": 0.22261759489774705,
      "num_tokens": 144317219.0,
      "step": 71220
    },
    {
      "entropy": 5.848761940002442,
      "epoch": 6.224329284278598,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019273086692115992,
      "loss": 5.1391,
      "mean_token_accuracy": 0.22356442958116532,
      "num_tokens": 144326528.0,
      "step": 71225
    },
    {
      "entropy": 5.916641807556152,
      "epoch": 6.224766232631303,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001927018643275152,
      "loss": 5.2605,
      "mean_token_accuracy": 0.21153346449136734,
      "num_tokens": 144337906.0,
      "step": 71230
    },
    {
      "entropy": 5.849124240875244,
      "epoch": 6.225203180984008,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001926728633123918,
      "loss": 5.0857,
      "mean_token_accuracy": 0.23162260949611663,
      "num_tokens": 144347824.0,
      "step": 71235
    },
    {
      "entropy": 5.86025562286377,
      "epoch": 6.225640129336712,
      "grad_norm": 1.125,
      "learning_rate": 0.00019264386387634602,
      "loss": 5.1613,
      "mean_token_accuracy": 0.21707813143730165,
      "num_tokens": 144359435.0,
      "step": 71240
    },
    {
      "entropy": 5.816849708557129,
      "epoch": 6.226077077689417,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019261486601993417,
      "loss": 5.1167,
      "mean_token_accuracy": 0.2227466657757759,
      "num_tokens": 144369568.0,
      "step": 71245
    },
    {
      "entropy": 5.813018894195556,
      "epoch": 6.226514026042122,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00019258586974371233,
      "loss": 5.1208,
      "mean_token_accuracy": 0.22051941156387328,
      "num_tokens": 144379328.0,
      "step": 71250
    },
    {
      "entropy": 5.83128023147583,
      "epoch": 6.226950974394827,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019255687504823665,
      "loss": 5.1478,
      "mean_token_accuracy": 0.22193140536546707,
      "num_tokens": 144388709.0,
      "step": 71255
    },
    {
      "entropy": 5.858391141891479,
      "epoch": 6.227387922747531,
      "grad_norm": 1.25,
      "learning_rate": 0.00019252788193406328,
      "loss": 5.1453,
      "mean_token_accuracy": 0.2281363397836685,
      "num_tokens": 144399507.0,
      "step": 71260
    },
    {
      "entropy": 5.8527184009552,
      "epoch": 6.2278248711002355,
      "grad_norm": 1.078125,
      "learning_rate": 0.00019249889040174844,
      "loss": 5.1056,
      "mean_token_accuracy": 0.22970701903104782,
      "num_tokens": 144409873.0,
      "step": 71265
    },
    {
      "entropy": 5.900296640396118,
      "epoch": 6.228261819452941,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019246990045184803,
      "loss": 5.1663,
      "mean_token_accuracy": 0.21652802228927612,
      "num_tokens": 144419130.0,
      "step": 71270
    },
    {
      "entropy": 5.851378965377807,
      "epoch": 6.228698767805645,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019244091208491827,
      "loss": 5.1234,
      "mean_token_accuracy": 0.22678279727697373,
      "num_tokens": 144429160.0,
      "step": 71275
    },
    {
      "entropy": 5.8288130283355715,
      "epoch": 6.22913571615835,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019241192530151496,
      "loss": 5.1914,
      "mean_token_accuracy": 0.2232392117381096,
      "num_tokens": 144439551.0,
      "step": 71280
    },
    {
      "entropy": 5.851215457916259,
      "epoch": 6.229572664511055,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019238294010219425,
      "loss": 5.1876,
      "mean_token_accuracy": 0.21618647277355194,
      "num_tokens": 144449837.0,
      "step": 71285
    },
    {
      "entropy": 5.855642652511596,
      "epoch": 6.23000961286376,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019235395648751203,
      "loss": 5.1729,
      "mean_token_accuracy": 0.22525318562984467,
      "num_tokens": 144460285.0,
      "step": 71290
    },
    {
      "entropy": 5.954895925521851,
      "epoch": 6.230446561216464,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019232497445802427,
      "loss": 5.1483,
      "mean_token_accuracy": 0.22275362461805343,
      "num_tokens": 144469417.0,
      "step": 71295
    },
    {
      "entropy": 5.843650531768799,
      "epoch": 6.230883509569169,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019229599401428678,
      "loss": 5.1233,
      "mean_token_accuracy": 0.229972006380558,
      "num_tokens": 144479144.0,
      "step": 71300
    },
    {
      "entropy": 5.916242980957032,
      "epoch": 6.231320457921874,
      "grad_norm": 1.109375,
      "learning_rate": 0.00019226701515685551,
      "loss": 5.2788,
      "mean_token_accuracy": 0.20994717627763748,
      "num_tokens": 144489337.0,
      "step": 71305
    },
    {
      "entropy": 5.789400720596314,
      "epoch": 6.231757406274578,
      "grad_norm": 1.125,
      "learning_rate": 0.0001922380378862862,
      "loss": 5.1071,
      "mean_token_accuracy": 0.22975774109363556,
      "num_tokens": 144499675.0,
      "step": 71310
    },
    {
      "entropy": 5.822513628005981,
      "epoch": 6.232194354627283,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019220906220313467,
      "loss": 5.0995,
      "mean_token_accuracy": 0.22843005657196044,
      "num_tokens": 144509031.0,
      "step": 71315
    },
    {
      "entropy": 5.8938709735870365,
      "epoch": 6.232631302979987,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001921800881079568,
      "loss": 5.1611,
      "mean_token_accuracy": 0.21667154580354692,
      "num_tokens": 144518878.0,
      "step": 71320
    },
    {
      "entropy": 5.9141889095306395,
      "epoch": 6.233068251332693,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019215111560130815,
      "loss": 5.2999,
      "mean_token_accuracy": 0.21479005962610245,
      "num_tokens": 144529259.0,
      "step": 71325
    },
    {
      "entropy": 5.8641357421875,
      "epoch": 6.233505199685397,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001921221446837445,
      "loss": 5.1093,
      "mean_token_accuracy": 0.23326195478439332,
      "num_tokens": 144540409.0,
      "step": 71330
    },
    {
      "entropy": 5.903982448577881,
      "epoch": 6.233942148038102,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019209317535582156,
      "loss": 5.2277,
      "mean_token_accuracy": 0.217233145236969,
      "num_tokens": 144549804.0,
      "step": 71335
    },
    {
      "entropy": 5.834684181213379,
      "epoch": 6.234379096390807,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019206420761809495,
      "loss": 5.1268,
      "mean_token_accuracy": 0.23105103075504302,
      "num_tokens": 144560257.0,
      "step": 71340
    },
    {
      "entropy": 5.815328216552734,
      "epoch": 6.2348160447435115,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001920352414711204,
      "loss": 5.0709,
      "mean_token_accuracy": 0.23372989892959595,
      "num_tokens": 144570213.0,
      "step": 71345
    },
    {
      "entropy": 5.887032222747803,
      "epoch": 6.235252993096216,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019200627691545331,
      "loss": 5.1584,
      "mean_token_accuracy": 0.22482981383800507,
      "num_tokens": 144579800.0,
      "step": 71350
    },
    {
      "entropy": 5.850050783157348,
      "epoch": 6.23568994144892,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019197731395164931,
      "loss": 5.1123,
      "mean_token_accuracy": 0.2223457619547844,
      "num_tokens": 144588794.0,
      "step": 71355
    },
    {
      "entropy": 5.871359634399414,
      "epoch": 6.236126889801626,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019194835258026395,
      "loss": 5.1635,
      "mean_token_accuracy": 0.22567853778600694,
      "num_tokens": 144600068.0,
      "step": 71360
    },
    {
      "entropy": 5.8542052745819095,
      "epoch": 6.23656383815433,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019191939280185265,
      "loss": 5.0997,
      "mean_token_accuracy": 0.2289726987481117,
      "num_tokens": 144608955.0,
      "step": 71365
    },
    {
      "entropy": 5.897859668731689,
      "epoch": 6.237000786507035,
      "grad_norm": 1.125,
      "learning_rate": 0.00019189043461697103,
      "loss": 5.1839,
      "mean_token_accuracy": 0.22086766213178635,
      "num_tokens": 144620028.0,
      "step": 71370
    },
    {
      "entropy": 5.902985954284668,
      "epoch": 6.237437734859739,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019186147802617437,
      "loss": 5.1916,
      "mean_token_accuracy": 0.2213537275791168,
      "num_tokens": 144630601.0,
      "step": 71375
    },
    {
      "entropy": 5.839159727096558,
      "epoch": 6.2378746832124445,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019183252303001807,
      "loss": 5.0461,
      "mean_token_accuracy": 0.22808138132095337,
      "num_tokens": 144639768.0,
      "step": 71380
    },
    {
      "entropy": 5.864537048339844,
      "epoch": 6.238311631565149,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00019180356962905765,
      "loss": 5.1419,
      "mean_token_accuracy": 0.2205442652106285,
      "num_tokens": 144650459.0,
      "step": 71385
    },
    {
      "entropy": 5.835242700576782,
      "epoch": 6.2387485799178535,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001917746178238483,
      "loss": 5.0966,
      "mean_token_accuracy": 0.23188744187355043,
      "num_tokens": 144660757.0,
      "step": 71390
    },
    {
      "entropy": 5.850914239883423,
      "epoch": 6.239185528270559,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019174566761494545,
      "loss": 5.1097,
      "mean_token_accuracy": 0.22828275412321092,
      "num_tokens": 144671025.0,
      "step": 71395
    },
    {
      "entropy": 5.810758256912232,
      "epoch": 6.239622476623263,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019171671900290429,
      "loss": 5.1592,
      "mean_token_accuracy": 0.22147239446640016,
      "num_tokens": 144681509.0,
      "step": 71400
    },
    {
      "entropy": 5.923345375061035,
      "epoch": 6.240059424975968,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001916877719882801,
      "loss": 5.2232,
      "mean_token_accuracy": 0.21661533713340758,
      "num_tokens": 144691740.0,
      "step": 71405
    },
    {
      "entropy": 5.826278781890869,
      "epoch": 6.240496373328672,
      "grad_norm": 1.125,
      "learning_rate": 0.00019165882657162802,
      "loss": 5.0958,
      "mean_token_accuracy": 0.2391523689031601,
      "num_tokens": 144701323.0,
      "step": 71410
    },
    {
      "entropy": 5.903327560424804,
      "epoch": 6.240933321681378,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001916298827535034,
      "loss": 5.2159,
      "mean_token_accuracy": 0.215176123380661,
      "num_tokens": 144712368.0,
      "step": 71415
    },
    {
      "entropy": 5.894266080856323,
      "epoch": 6.241370270034082,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019160094053446132,
      "loss": 5.1418,
      "mean_token_accuracy": 0.2194230452179909,
      "num_tokens": 144722469.0,
      "step": 71420
    },
    {
      "entropy": 5.8773821830749515,
      "epoch": 6.2418072183867865,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019157199991505692,
      "loss": 5.1322,
      "mean_token_accuracy": 0.22914327085018157,
      "num_tokens": 144731557.0,
      "step": 71425
    },
    {
      "entropy": 5.83959813117981,
      "epoch": 6.242244166739491,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019154306089584522,
      "loss": 5.1817,
      "mean_token_accuracy": 0.22056097239255906,
      "num_tokens": 144741125.0,
      "step": 71430
    },
    {
      "entropy": 5.939187479019165,
      "epoch": 6.242681115092196,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019151412347738128,
      "loss": 5.2317,
      "mean_token_accuracy": 0.21612759530544282,
      "num_tokens": 144751198.0,
      "step": 71435
    },
    {
      "entropy": 5.8904365539550785,
      "epoch": 6.243118063444901,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019148518766022032,
      "loss": 5.0996,
      "mean_token_accuracy": 0.22917955368757248,
      "num_tokens": 144761183.0,
      "step": 71440
    },
    {
      "entropy": 5.922707891464233,
      "epoch": 6.243555011797605,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019145625344491724,
      "loss": 5.2011,
      "mean_token_accuracy": 0.21908003687858582,
      "num_tokens": 144771243.0,
      "step": 71445
    },
    {
      "entropy": 5.781618261337281,
      "epoch": 6.243991960150311,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019142732083202692,
      "loss": 5.1021,
      "mean_token_accuracy": 0.22944247126579284,
      "num_tokens": 144782651.0,
      "step": 71450
    },
    {
      "entropy": 5.889032173156738,
      "epoch": 6.244428908503015,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019139838982210438,
      "loss": 5.1796,
      "mean_token_accuracy": 0.22264713644981385,
      "num_tokens": 144792625.0,
      "step": 71455
    },
    {
      "entropy": 5.907199811935425,
      "epoch": 6.24486585685572,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019136946041570452,
      "loss": 5.2101,
      "mean_token_accuracy": 0.22106633931398392,
      "num_tokens": 144802524.0,
      "step": 71460
    },
    {
      "entropy": 5.846062564849854,
      "epoch": 6.245302805208424,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001913405326133823,
      "loss": 5.1063,
      "mean_token_accuracy": 0.22702829092741011,
      "num_tokens": 144812897.0,
      "step": 71465
    },
    {
      "entropy": 5.863259410858154,
      "epoch": 6.245739753561129,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001913116064156924,
      "loss": 5.0567,
      "mean_token_accuracy": 0.2292264923453331,
      "num_tokens": 144823257.0,
      "step": 71470
    },
    {
      "entropy": 5.845180892944336,
      "epoch": 6.246176701913834,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019128268182318982,
      "loss": 5.0623,
      "mean_token_accuracy": 0.22513317465782165,
      "num_tokens": 144832481.0,
      "step": 71475
    },
    {
      "entropy": 5.856839847564697,
      "epoch": 6.246613650266538,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001912537588364292,
      "loss": 5.2078,
      "mean_token_accuracy": 0.2137986660003662,
      "num_tokens": 144841736.0,
      "step": 71480
    },
    {
      "entropy": 5.847162914276123,
      "epoch": 6.247050598619243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019122483745596543,
      "loss": 5.1489,
      "mean_token_accuracy": 0.22170366793870927,
      "num_tokens": 144852469.0,
      "step": 71485
    },
    {
      "entropy": 5.88466157913208,
      "epoch": 6.247487546971948,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019119591768235322,
      "loss": 5.227,
      "mean_token_accuracy": 0.21479032933712006,
      "num_tokens": 144863720.0,
      "step": 71490
    },
    {
      "entropy": 5.870867586135864,
      "epoch": 6.247924495324653,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019116699951614714,
      "loss": 5.0981,
      "mean_token_accuracy": 0.22821618765592575,
      "num_tokens": 144873809.0,
      "step": 71495
    },
    {
      "entropy": 5.836747550964356,
      "epoch": 6.248361443677357,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019113808295790198,
      "loss": 5.0731,
      "mean_token_accuracy": 0.2304156467318535,
      "num_tokens": 144883927.0,
      "step": 71500
    },
    {
      "entropy": 5.840743112564087,
      "epoch": 6.2487983920300625,
      "grad_norm": 1.125,
      "learning_rate": 0.00019110916800817225,
      "loss": 5.1911,
      "mean_token_accuracy": 0.22505735009908676,
      "num_tokens": 144895535.0,
      "step": 71505
    },
    {
      "entropy": 5.845117044448853,
      "epoch": 6.249235340382767,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001910802546675127,
      "loss": 5.0977,
      "mean_token_accuracy": 0.23065923005342484,
      "num_tokens": 144905648.0,
      "step": 71510
    },
    {
      "entropy": 5.8914182662963865,
      "epoch": 6.249672288735471,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019105134293647785,
      "loss": 5.1829,
      "mean_token_accuracy": 0.220900496840477,
      "num_tokens": 144915569.0,
      "step": 71515
    },
    {
      "entropy": 5.860309267044068,
      "epoch": 6.250109237088176,
      "grad_norm": 1.25,
      "learning_rate": 0.0001910224328156222,
      "loss": 5.1603,
      "mean_token_accuracy": 0.22061746418476105,
      "num_tokens": 144925557.0,
      "step": 71520
    },
    {
      "entropy": 5.918118238449097,
      "epoch": 6.250546185440881,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019099352430550033,
      "loss": 5.1824,
      "mean_token_accuracy": 0.2208523139357567,
      "num_tokens": 144935832.0,
      "step": 71525
    },
    {
      "entropy": 5.921855115890503,
      "epoch": 6.250983133793586,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019096461740666665,
      "loss": 5.1662,
      "mean_token_accuracy": 0.21568928062915801,
      "num_tokens": 144945456.0,
      "step": 71530
    },
    {
      "entropy": 5.846384906768799,
      "epoch": 6.25142008214629,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019093571211967563,
      "loss": 5.1099,
      "mean_token_accuracy": 0.22619952112436295,
      "num_tokens": 144956475.0,
      "step": 71535
    },
    {
      "entropy": 5.846402549743653,
      "epoch": 6.251857030498995,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019090680844508179,
      "loss": 5.1676,
      "mean_token_accuracy": 0.22238553166389466,
      "num_tokens": 144967134.0,
      "step": 71540
    },
    {
      "entropy": 5.839578008651733,
      "epoch": 6.2522939788517,
      "grad_norm": 1.078125,
      "learning_rate": 0.00019087790638343933,
      "loss": 5.1268,
      "mean_token_accuracy": 0.22706668674945832,
      "num_tokens": 144977446.0,
      "step": 71545
    },
    {
      "entropy": 5.84370231628418,
      "epoch": 6.2527309272044045,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019084900593530275,
      "loss": 5.1586,
      "mean_token_accuracy": 0.2260437563061714,
      "num_tokens": 144989126.0,
      "step": 71550
    },
    {
      "entropy": 5.8132459163665775,
      "epoch": 6.253167875557109,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019082010710122626,
      "loss": 5.1099,
      "mean_token_accuracy": 0.22377352118492128,
      "num_tokens": 144999371.0,
      "step": 71555
    },
    {
      "entropy": 5.852130460739136,
      "epoch": 6.253604823909814,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019079120988176428,
      "loss": 5.0996,
      "mean_token_accuracy": 0.22645104080438613,
      "num_tokens": 145010446.0,
      "step": 71560
    },
    {
      "entropy": 5.897742700576782,
      "epoch": 6.254041772262519,
      "grad_norm": 1.125,
      "learning_rate": 0.00019076231427747105,
      "loss": 5.1882,
      "mean_token_accuracy": 0.2206544741988182,
      "num_tokens": 145021523.0,
      "step": 71565
    },
    {
      "entropy": 5.843159580230713,
      "epoch": 6.254478720615223,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019073342028890067,
      "loss": 5.088,
      "mean_token_accuracy": 0.23418018519878386,
      "num_tokens": 145030947.0,
      "step": 71570
    },
    {
      "entropy": 5.886388444900513,
      "epoch": 6.254915668967928,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019070452791660752,
      "loss": 5.1737,
      "mean_token_accuracy": 0.2218462571501732,
      "num_tokens": 145040597.0,
      "step": 71575
    },
    {
      "entropy": 5.826588582992554,
      "epoch": 6.255352617320633,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019067563716114566,
      "loss": 5.1499,
      "mean_token_accuracy": 0.22467397451400756,
      "num_tokens": 145051236.0,
      "step": 71580
    },
    {
      "entropy": 5.839041423797608,
      "epoch": 6.2557895656733375,
      "grad_norm": 1.09375,
      "learning_rate": 0.00019064674802306937,
      "loss": 5.1418,
      "mean_token_accuracy": 0.22756630480289458,
      "num_tokens": 145061986.0,
      "step": 71585
    },
    {
      "entropy": 5.845593976974487,
      "epoch": 6.256226514026042,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001906178605029325,
      "loss": 5.1277,
      "mean_token_accuracy": 0.23015473335981368,
      "num_tokens": 145072997.0,
      "step": 71590
    },
    {
      "entropy": 5.8732171058654785,
      "epoch": 6.2566634623787465,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019058897460128937,
      "loss": 5.2305,
      "mean_token_accuracy": 0.2166363015770912,
      "num_tokens": 145083584.0,
      "step": 71595
    },
    {
      "entropy": 5.853892612457275,
      "epoch": 6.257100410731452,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019056009031869398,
      "loss": 5.1503,
      "mean_token_accuracy": 0.23163308054208756,
      "num_tokens": 145093656.0,
      "step": 71600
    },
    {
      "entropy": 5.91475133895874,
      "epoch": 6.257537359084156,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019053120765570025,
      "loss": 5.1936,
      "mean_token_accuracy": 0.21856266707181932,
      "num_tokens": 145103586.0,
      "step": 71605
    },
    {
      "entropy": 5.735148239135742,
      "epoch": 6.257974307436861,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019050232661286222,
      "loss": 5.0176,
      "mean_token_accuracy": 0.2421151801943779,
      "num_tokens": 145113412.0,
      "step": 71610
    },
    {
      "entropy": 5.820935773849487,
      "epoch": 6.258411255789566,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019047344719073377,
      "loss": 5.0966,
      "mean_token_accuracy": 0.22493873983621598,
      "num_tokens": 145123012.0,
      "step": 71615
    },
    {
      "entropy": 5.861056804656982,
      "epoch": 6.258848204142271,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019044456938986898,
      "loss": 5.1811,
      "mean_token_accuracy": 0.22534924745559692,
      "num_tokens": 145132219.0,
      "step": 71620
    },
    {
      "entropy": 5.859162378311157,
      "epoch": 6.259285152494975,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019041569321082168,
      "loss": 5.1346,
      "mean_token_accuracy": 0.218079674243927,
      "num_tokens": 145141832.0,
      "step": 71625
    },
    {
      "entropy": 5.851041412353515,
      "epoch": 6.2597221008476795,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019038681865414566,
      "loss": 5.1265,
      "mean_token_accuracy": 0.22568719685077668,
      "num_tokens": 145152202.0,
      "step": 71630
    },
    {
      "entropy": 5.790966367721557,
      "epoch": 6.260159049200385,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019035794572039478,
      "loss": 5.0809,
      "mean_token_accuracy": 0.23636410236358643,
      "num_tokens": 145162377.0,
      "step": 71635
    },
    {
      "entropy": 5.86320013999939,
      "epoch": 6.260595997553089,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019032907441012288,
      "loss": 5.1624,
      "mean_token_accuracy": 0.2270297408103943,
      "num_tokens": 145172627.0,
      "step": 71640
    },
    {
      "entropy": 5.865035915374756,
      "epoch": 6.261032945905794,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019030020472388364,
      "loss": 5.1,
      "mean_token_accuracy": 0.2278255745768547,
      "num_tokens": 145182534.0,
      "step": 71645
    },
    {
      "entropy": 5.900242900848388,
      "epoch": 6.261469894258498,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019027133666223102,
      "loss": 5.1596,
      "mean_token_accuracy": 0.22127032428979873,
      "num_tokens": 145192886.0,
      "step": 71650
    },
    {
      "entropy": 5.823805141448974,
      "epoch": 6.261906842611204,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001902424702257184,
      "loss": 5.0934,
      "mean_token_accuracy": 0.23110245764255524,
      "num_tokens": 145203444.0,
      "step": 71655
    },
    {
      "entropy": 5.823578786849976,
      "epoch": 6.262343790963908,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001902136054148997,
      "loss": 5.1105,
      "mean_token_accuracy": 0.2335546374320984,
      "num_tokens": 145213576.0,
      "step": 71660
    },
    {
      "entropy": 5.845987939834595,
      "epoch": 6.262780739316613,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001901847422303284,
      "loss": 5.1369,
      "mean_token_accuracy": 0.2294697642326355,
      "num_tokens": 145224086.0,
      "step": 71665
    },
    {
      "entropy": 5.777046871185303,
      "epoch": 6.263217687669317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019015588067255825,
      "loss": 5.123,
      "mean_token_accuracy": 0.229355326294899,
      "num_tokens": 145234501.0,
      "step": 71670
    },
    {
      "entropy": 5.853376388549805,
      "epoch": 6.263654636022022,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001901270207421427,
      "loss": 5.1688,
      "mean_token_accuracy": 0.22138271182775499,
      "num_tokens": 145245462.0,
      "step": 71675
    },
    {
      "entropy": 5.977707958221435,
      "epoch": 6.264091584374727,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001900981624396354,
      "loss": 5.2379,
      "mean_token_accuracy": 0.2140301376581192,
      "num_tokens": 145255627.0,
      "step": 71680
    },
    {
      "entropy": 5.845815658569336,
      "epoch": 6.264528532727431,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019006930576558984,
      "loss": 5.0996,
      "mean_token_accuracy": 0.2272238925099373,
      "num_tokens": 145265737.0,
      "step": 71685
    },
    {
      "entropy": 5.807312297821045,
      "epoch": 6.264965481080137,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019004045072055947,
      "loss": 5.0738,
      "mean_token_accuracy": 0.22877012491226195,
      "num_tokens": 145275745.0,
      "step": 71690
    },
    {
      "entropy": 5.8300646305084225,
      "epoch": 6.265402429432841,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019001159730509782,
      "loss": 5.0561,
      "mean_token_accuracy": 0.2276218369603157,
      "num_tokens": 145285333.0,
      "step": 71695
    },
    {
      "entropy": 5.810372924804687,
      "epoch": 6.265839377785546,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018998274551975826,
      "loss": 5.0873,
      "mean_token_accuracy": 0.23160415440797805,
      "num_tokens": 145294422.0,
      "step": 71700
    },
    {
      "entropy": 5.790967750549316,
      "epoch": 6.26627632613825,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018995389536509416,
      "loss": 5.1181,
      "mean_token_accuracy": 0.23106375932693482,
      "num_tokens": 145304551.0,
      "step": 71705
    },
    {
      "entropy": 5.811009073257447,
      "epoch": 6.2667132744909555,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001899250468416589,
      "loss": 5.1061,
      "mean_token_accuracy": 0.2288027063012123,
      "num_tokens": 145314731.0,
      "step": 71710
    },
    {
      "entropy": 5.906192350387573,
      "epoch": 6.26715022284366,
      "grad_norm": 1.25,
      "learning_rate": 0.0001898961999500059,
      "loss": 5.183,
      "mean_token_accuracy": 0.2212224081158638,
      "num_tokens": 145324922.0,
      "step": 71715
    },
    {
      "entropy": 5.864826917648315,
      "epoch": 6.267587171196364,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018986735469068834,
      "loss": 5.1514,
      "mean_token_accuracy": 0.2251606196165085,
      "num_tokens": 145334654.0,
      "step": 71720
    },
    {
      "entropy": 5.818315172195435,
      "epoch": 6.268024119549069,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018983851106425952,
      "loss": 5.1218,
      "mean_token_accuracy": 0.23156283050775528,
      "num_tokens": 145344969.0,
      "step": 71725
    },
    {
      "entropy": 5.857831048965454,
      "epoch": 6.268461067901774,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018980966907127268,
      "loss": 5.133,
      "mean_token_accuracy": 0.2271430417895317,
      "num_tokens": 145354916.0,
      "step": 71730
    },
    {
      "entropy": 5.811461925506592,
      "epoch": 6.268898016254479,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018978082871228097,
      "loss": 5.0595,
      "mean_token_accuracy": 0.23581389635801314,
      "num_tokens": 145363826.0,
      "step": 71735
    },
    {
      "entropy": 5.844895744323731,
      "epoch": 6.269334964607183,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018975198998783766,
      "loss": 5.1828,
      "mean_token_accuracy": 0.22168124318122864,
      "num_tokens": 145373934.0,
      "step": 71740
    },
    {
      "entropy": 5.696545934677124,
      "epoch": 6.269771912959888,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001897231528984959,
      "loss": 5.0493,
      "mean_token_accuracy": 0.2441927656531334,
      "num_tokens": 145385004.0,
      "step": 71745
    },
    {
      "entropy": 5.825702667236328,
      "epoch": 6.270208861312593,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018969431744480868,
      "loss": 5.1233,
      "mean_token_accuracy": 0.23094096630811692,
      "num_tokens": 145394263.0,
      "step": 71750
    },
    {
      "entropy": 5.847896146774292,
      "epoch": 6.2706458096652975,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001896654836273292,
      "loss": 5.0578,
      "mean_token_accuracy": 0.23986643701791763,
      "num_tokens": 145403859.0,
      "step": 71755
    },
    {
      "entropy": 5.882591581344604,
      "epoch": 6.271082758018002,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001896366514466104,
      "loss": 5.083,
      "mean_token_accuracy": 0.22863281518220901,
      "num_tokens": 145412999.0,
      "step": 71760
    },
    {
      "entropy": 5.842592096328735,
      "epoch": 6.271519706370707,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018960782090320539,
      "loss": 5.1672,
      "mean_token_accuracy": 0.22144107669591903,
      "num_tokens": 145422717.0,
      "step": 71765
    },
    {
      "entropy": 5.7883172035217285,
      "epoch": 6.271956654723412,
      "grad_norm": 1.125,
      "learning_rate": 0.0001895789919976672,
      "loss": 5.0648,
      "mean_token_accuracy": 0.23346628695726396,
      "num_tokens": 145432889.0,
      "step": 71770
    },
    {
      "entropy": 5.812003087997437,
      "epoch": 6.272393603076116,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001895501647305486,
      "loss": 5.0734,
      "mean_token_accuracy": 0.22875899821519852,
      "num_tokens": 145441786.0,
      "step": 71775
    },
    {
      "entropy": 5.8158784866333,
      "epoch": 6.272830551428821,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018952133910240266,
      "loss": 5.2072,
      "mean_token_accuracy": 0.22111574411392212,
      "num_tokens": 145451912.0,
      "step": 71780
    },
    {
      "entropy": 5.834260368347168,
      "epoch": 6.273267499781526,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018949251511378218,
      "loss": 5.15,
      "mean_token_accuracy": 0.22673084884881972,
      "num_tokens": 145462301.0,
      "step": 71785
    },
    {
      "entropy": 5.879627275466919,
      "epoch": 6.273704448134231,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018946369276524007,
      "loss": 5.2257,
      "mean_token_accuracy": 0.22062134444713594,
      "num_tokens": 145473451.0,
      "step": 71790
    },
    {
      "entropy": 5.822649097442627,
      "epoch": 6.274141396486935,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018943487205732922,
      "loss": 5.0926,
      "mean_token_accuracy": 0.23067878186702728,
      "num_tokens": 145483574.0,
      "step": 71795
    },
    {
      "entropy": 5.893055772781372,
      "epoch": 6.2745783448396395,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001894060529906023,
      "loss": 5.2603,
      "mean_token_accuracy": 0.21454242169857024,
      "num_tokens": 145493382.0,
      "step": 71800
    },
    {
      "entropy": 5.857532596588134,
      "epoch": 6.275015293192345,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018937723556561214,
      "loss": 5.0862,
      "mean_token_accuracy": 0.23043402880430222,
      "num_tokens": 145503455.0,
      "step": 71805
    },
    {
      "entropy": 5.83367428779602,
      "epoch": 6.275452241545049,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018934841978291144,
      "loss": 5.1631,
      "mean_token_accuracy": 0.2275402620434761,
      "num_tokens": 145514565.0,
      "step": 71810
    },
    {
      "entropy": 5.817405891418457,
      "epoch": 6.275889189897754,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000189319605643053,
      "loss": 5.139,
      "mean_token_accuracy": 0.2216300278902054,
      "num_tokens": 145524224.0,
      "step": 71815
    },
    {
      "entropy": 5.736604356765747,
      "epoch": 6.276326138250459,
      "grad_norm": 1.125,
      "learning_rate": 0.00018929079314658942,
      "loss": 5.1445,
      "mean_token_accuracy": 0.2402306839823723,
      "num_tokens": 145535412.0,
      "step": 71820
    },
    {
      "entropy": 5.8942248821258545,
      "epoch": 6.276763086603164,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018926198229407331,
      "loss": 5.1192,
      "mean_token_accuracy": 0.22313011437654495,
      "num_tokens": 145544998.0,
      "step": 71825
    },
    {
      "entropy": 5.849314594268799,
      "epoch": 6.277200034955868,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001892331730860573,
      "loss": 5.1689,
      "mean_token_accuracy": 0.22085878849029542,
      "num_tokens": 145556093.0,
      "step": 71830
    },
    {
      "entropy": 5.873813819885254,
      "epoch": 6.277636983308573,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018920436552309394,
      "loss": 5.1457,
      "mean_token_accuracy": 0.227882082760334,
      "num_tokens": 145565284.0,
      "step": 71835
    },
    {
      "entropy": 5.808170318603516,
      "epoch": 6.278073931661278,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018917555960573586,
      "loss": 5.0625,
      "mean_token_accuracy": 0.22782938778400422,
      "num_tokens": 145575330.0,
      "step": 71840
    },
    {
      "entropy": 5.869514036178589,
      "epoch": 6.278510880013982,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001891467553345355,
      "loss": 5.2144,
      "mean_token_accuracy": 0.2197765588760376,
      "num_tokens": 145585653.0,
      "step": 71845
    },
    {
      "entropy": 5.882787466049194,
      "epoch": 6.278947828366687,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001891179527100454,
      "loss": 5.1323,
      "mean_token_accuracy": 0.2291311651468277,
      "num_tokens": 145595031.0,
      "step": 71850
    },
    {
      "entropy": 5.842732906341553,
      "epoch": 6.279384776719391,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00018908915173281788,
      "loss": 5.0953,
      "mean_token_accuracy": 0.22958822101354598,
      "num_tokens": 145605338.0,
      "step": 71855
    },
    {
      "entropy": 5.815436983108521,
      "epoch": 6.279821725072097,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018906035240340546,
      "loss": 5.0929,
      "mean_token_accuracy": 0.23088741302490234,
      "num_tokens": 145614865.0,
      "step": 71860
    },
    {
      "entropy": 5.808637523651123,
      "epoch": 6.280258673424801,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018903155472236056,
      "loss": 5.1209,
      "mean_token_accuracy": 0.21948884129524232,
      "num_tokens": 145624086.0,
      "step": 71865
    },
    {
      "entropy": 5.8721788883209225,
      "epoch": 6.280695621777506,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018900275869023555,
      "loss": 5.1889,
      "mean_token_accuracy": 0.2178674429655075,
      "num_tokens": 145635128.0,
      "step": 71870
    },
    {
      "entropy": 5.838678550720215,
      "epoch": 6.281132570130211,
      "grad_norm": 1.375,
      "learning_rate": 0.00018897396430758256,
      "loss": 5.1363,
      "mean_token_accuracy": 0.2277005910873413,
      "num_tokens": 145645831.0,
      "step": 71875
    },
    {
      "entropy": 5.845111846923828,
      "epoch": 6.2815695184829154,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018894517157495405,
      "loss": 5.1007,
      "mean_token_accuracy": 0.22652852833271026,
      "num_tokens": 145656058.0,
      "step": 71880
    },
    {
      "entropy": 5.769203948974609,
      "epoch": 6.28200646683562,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001889163804929022,
      "loss": 5.0397,
      "mean_token_accuracy": 0.2351369634270668,
      "num_tokens": 145666853.0,
      "step": 71885
    },
    {
      "entropy": 5.7419990539550785,
      "epoch": 6.282443415188324,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018888759106197939,
      "loss": 4.9908,
      "mean_token_accuracy": 0.24280763268470765,
      "num_tokens": 145677081.0,
      "step": 71890
    },
    {
      "entropy": 5.845042991638183,
      "epoch": 6.28288036354103,
      "grad_norm": 1.125,
      "learning_rate": 0.00018885880328273764,
      "loss": 5.1106,
      "mean_token_accuracy": 0.22615728825330733,
      "num_tokens": 145686459.0,
      "step": 71895
    },
    {
      "entropy": 5.854172325134277,
      "epoch": 6.283317311893734,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018883001715572922,
      "loss": 5.216,
      "mean_token_accuracy": 0.22386546581983566,
      "num_tokens": 145696378.0,
      "step": 71900
    },
    {
      "entropy": 5.79339189529419,
      "epoch": 6.283754260246439,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018880123268150617,
      "loss": 5.0784,
      "mean_token_accuracy": 0.2289997562766075,
      "num_tokens": 145705747.0,
      "step": 71905
    },
    {
      "entropy": 5.819409322738648,
      "epoch": 6.284191208599143,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018877244986062065,
      "loss": 5.0689,
      "mean_token_accuracy": 0.23391274958848954,
      "num_tokens": 145715708.0,
      "step": 71910
    },
    {
      "entropy": 5.864548301696777,
      "epoch": 6.2846281569518485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001887436686936248,
      "loss": 5.1785,
      "mean_token_accuracy": 0.22506771236658096,
      "num_tokens": 145727756.0,
      "step": 71915
    },
    {
      "entropy": 5.802159357070923,
      "epoch": 6.285065105304553,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001887148891810705,
      "loss": 5.102,
      "mean_token_accuracy": 0.2317645788192749,
      "num_tokens": 145736958.0,
      "step": 71920
    },
    {
      "entropy": 5.892162895202636,
      "epoch": 6.2855020536572574,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018868611132350977,
      "loss": 5.2384,
      "mean_token_accuracy": 0.2164470985531807,
      "num_tokens": 145747307.0,
      "step": 71925
    },
    {
      "entropy": 5.887689447402954,
      "epoch": 6.285939002009963,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018865733512149474,
      "loss": 5.1679,
      "mean_token_accuracy": 0.225517575442791,
      "num_tokens": 145757380.0,
      "step": 71930
    },
    {
      "entropy": 5.880665588378906,
      "epoch": 6.286375950362667,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018862856057557726,
      "loss": 5.1741,
      "mean_token_accuracy": 0.22205866873264313,
      "num_tokens": 145767308.0,
      "step": 71935
    },
    {
      "entropy": 5.8384541988372805,
      "epoch": 6.286812898715372,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018859978768630926,
      "loss": 5.0964,
      "mean_token_accuracy": 0.23798907697200775,
      "num_tokens": 145777449.0,
      "step": 71940
    },
    {
      "entropy": 5.847797060012818,
      "epoch": 6.287249847068076,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018857101645424263,
      "loss": 5.2171,
      "mean_token_accuracy": 0.22148824185132981,
      "num_tokens": 145787709.0,
      "step": 71945
    },
    {
      "entropy": 5.870982980728149,
      "epoch": 6.287686795420782,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001885422468799291,
      "loss": 5.2086,
      "mean_token_accuracy": 0.22089226841926574,
      "num_tokens": 145798644.0,
      "step": 71950
    },
    {
      "entropy": 5.843153047561645,
      "epoch": 6.288123743773486,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018851347896392068,
      "loss": 5.0947,
      "mean_token_accuracy": 0.22298056036233901,
      "num_tokens": 145808490.0,
      "step": 71955
    },
    {
      "entropy": 5.975449514389038,
      "epoch": 6.2885606921261905,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018848471270676903,
      "loss": 5.3233,
      "mean_token_accuracy": 0.20738816410303115,
      "num_tokens": 145818765.0,
      "step": 71960
    },
    {
      "entropy": 5.864915752410889,
      "epoch": 6.288997640478895,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018845594810902588,
      "loss": 5.088,
      "mean_token_accuracy": 0.23254332542419434,
      "num_tokens": 145828951.0,
      "step": 71965
    },
    {
      "entropy": 5.873586654663086,
      "epoch": 6.2894345888316,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018842718517124302,
      "loss": 5.1574,
      "mean_token_accuracy": 0.22622091621160506,
      "num_tokens": 145838766.0,
      "step": 71970
    },
    {
      "entropy": 5.75520429611206,
      "epoch": 6.289871537184305,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001883984238939721,
      "loss": 5.1007,
      "mean_token_accuracy": 0.2290012776851654,
      "num_tokens": 145848439.0,
      "step": 71975
    },
    {
      "entropy": 5.840938472747803,
      "epoch": 6.290308485537009,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018836966427776487,
      "loss": 5.1231,
      "mean_token_accuracy": 0.2276280030608177,
      "num_tokens": 145856932.0,
      "step": 71980
    },
    {
      "entropy": 5.894784688949585,
      "epoch": 6.290745433889715,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018834090632317274,
      "loss": 5.1346,
      "mean_token_accuracy": 0.22001306265592574,
      "num_tokens": 145866092.0,
      "step": 71985
    },
    {
      "entropy": 5.841274404525757,
      "epoch": 6.291182382242419,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018831215003074753,
      "loss": 5.0851,
      "mean_token_accuracy": 0.2296702519059181,
      "num_tokens": 145876149.0,
      "step": 71990
    },
    {
      "entropy": 5.821620988845825,
      "epoch": 6.291619330595124,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018828339540104066,
      "loss": 5.1011,
      "mean_token_accuracy": 0.22722122222185134,
      "num_tokens": 145885733.0,
      "step": 71995
    },
    {
      "entropy": 5.849638366699219,
      "epoch": 6.292056278947828,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018825464243460373,
      "loss": 5.1909,
      "mean_token_accuracy": 0.21989562809467317,
      "num_tokens": 145896824.0,
      "step": 72000
    },
    {
      "epoch": 6.292056278947828,
      "eval_entropy": 5.5843778809744675,
      "eval_loss": 5.259658336639404,
      "eval_mean_token_accuracy": 0.2254523584445661,
      "eval_num_tokens": 145896824.0,
      "eval_runtime": 27.8423,
      "eval_samples_per_second": 1329.418,
      "eval_steps_per_second": 166.186,
      "step": 72000
    },
    {
      "entropy": 5.841578626632691,
      "epoch": 6.292493227300533,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018822589113198822,
      "loss": 5.1284,
      "mean_token_accuracy": 0.22798730432987213,
      "num_tokens": 145907650.0,
      "step": 72005
    },
    {
      "entropy": 5.885219621658325,
      "epoch": 6.292930175653238,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018819714149374557,
      "loss": 5.0948,
      "mean_token_accuracy": 0.22891731709241867,
      "num_tokens": 145916735.0,
      "step": 72010
    },
    {
      "entropy": 5.8561756134033205,
      "epoch": 6.293367124005942,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001881683935204272,
      "loss": 5.0833,
      "mean_token_accuracy": 0.22907498329877854,
      "num_tokens": 145926531.0,
      "step": 72015
    },
    {
      "entropy": 5.7677703380584715,
      "epoch": 6.293804072358647,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001881396472125846,
      "loss": 5.1204,
      "mean_token_accuracy": 0.22837310433387756,
      "num_tokens": 145936922.0,
      "step": 72020
    },
    {
      "entropy": 5.741657304763794,
      "epoch": 6.294241020711352,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000188110902570769,
      "loss": 5.0785,
      "mean_token_accuracy": 0.23109312057495118,
      "num_tokens": 145947552.0,
      "step": 72025
    },
    {
      "entropy": 5.849215841293335,
      "epoch": 6.294677969064057,
      "grad_norm": 1.25,
      "learning_rate": 0.00018808215959553186,
      "loss": 5.1647,
      "mean_token_accuracy": 0.22531960755586625,
      "num_tokens": 145956656.0,
      "step": 72030
    },
    {
      "entropy": 5.855993843078613,
      "epoch": 6.295114917416761,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001880534182874245,
      "loss": 5.0817,
      "mean_token_accuracy": 0.23251558989286422,
      "num_tokens": 145966302.0,
      "step": 72035
    },
    {
      "entropy": 5.839532136917114,
      "epoch": 6.2955518657694665,
      "grad_norm": 1.125,
      "learning_rate": 0.00018802467864699807,
      "loss": 5.1054,
      "mean_token_accuracy": 0.22855718433856964,
      "num_tokens": 145975874.0,
      "step": 72040
    },
    {
      "entropy": 5.81556715965271,
      "epoch": 6.295988814122171,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018799594067480386,
      "loss": 5.0524,
      "mean_token_accuracy": 0.22866583615541458,
      "num_tokens": 145985354.0,
      "step": 72045
    },
    {
      "entropy": 5.8993736743927006,
      "epoch": 6.296425762474875,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001879672043713932,
      "loss": 5.2288,
      "mean_token_accuracy": 0.21978286802768707,
      "num_tokens": 145996132.0,
      "step": 72050
    },
    {
      "entropy": 5.82150616645813,
      "epoch": 6.29686271082758,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001879384697373171,
      "loss": 5.0934,
      "mean_token_accuracy": 0.2327631816267967,
      "num_tokens": 146006785.0,
      "step": 72055
    },
    {
      "entropy": 5.843734979629517,
      "epoch": 6.297299659180285,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018790973677312674,
      "loss": 5.0995,
      "mean_token_accuracy": 0.2343357428908348,
      "num_tokens": 146017398.0,
      "step": 72060
    },
    {
      "entropy": 5.7749895572662355,
      "epoch": 6.29773660753299,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00018788100547937326,
      "loss": 5.0569,
      "mean_token_accuracy": 0.23352739363908767,
      "num_tokens": 146027125.0,
      "step": 72065
    },
    {
      "entropy": 5.8781853199005125,
      "epoch": 6.298173555885694,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001878522758566078,
      "loss": 5.0883,
      "mean_token_accuracy": 0.23409549593925477,
      "num_tokens": 146036459.0,
      "step": 72070
    },
    {
      "entropy": 5.8500806331634525,
      "epoch": 6.298610504238399,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018782354790538138,
      "loss": 5.1909,
      "mean_token_accuracy": 0.22212437838315963,
      "num_tokens": 146046933.0,
      "step": 72075
    },
    {
      "entropy": 5.840065956115723,
      "epoch": 6.299047452591104,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001877948216262449,
      "loss": 5.1208,
      "mean_token_accuracy": 0.2274506703019142,
      "num_tokens": 146056263.0,
      "step": 72080
    },
    {
      "entropy": 5.7356390953063965,
      "epoch": 6.2994844009438085,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018776609701974946,
      "loss": 5.1093,
      "mean_token_accuracy": 0.2288315087556839,
      "num_tokens": 146066536.0,
      "step": 72085
    },
    {
      "entropy": 5.765359163284302,
      "epoch": 6.299921349296513,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018773737408644598,
      "loss": 5.1079,
      "mean_token_accuracy": 0.22236145436763763,
      "num_tokens": 146075431.0,
      "step": 72090
    },
    {
      "entropy": 5.816920423507691,
      "epoch": 6.300358297649218,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018770865282688543,
      "loss": 5.1533,
      "mean_token_accuracy": 0.22776558846235276,
      "num_tokens": 146086394.0,
      "step": 72095
    },
    {
      "entropy": 5.8259296894073485,
      "epoch": 6.300795246001923,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001876799332416187,
      "loss": 5.1775,
      "mean_token_accuracy": 0.22634806632995605,
      "num_tokens": 146096381.0,
      "step": 72100
    },
    {
      "entropy": 5.787042665481567,
      "epoch": 6.301232194354627,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018765121533119655,
      "loss": 5.0524,
      "mean_token_accuracy": 0.24094485491514206,
      "num_tokens": 146107107.0,
      "step": 72105
    },
    {
      "entropy": 5.897206974029541,
      "epoch": 6.301669142707332,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001876224990961698,
      "loss": 5.1394,
      "mean_token_accuracy": 0.22049384266138078,
      "num_tokens": 146117019.0,
      "step": 72110
    },
    {
      "entropy": 5.830308151245117,
      "epoch": 6.302106091060037,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018759378453708935,
      "loss": 5.1188,
      "mean_token_accuracy": 0.22529346793889998,
      "num_tokens": 146127964.0,
      "step": 72115
    },
    {
      "entropy": 5.864289617538452,
      "epoch": 6.3025430394127415,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00018756507165450592,
      "loss": 5.1501,
      "mean_token_accuracy": 0.22371819019317626,
      "num_tokens": 146139770.0,
      "step": 72120
    },
    {
      "entropy": 5.847931003570556,
      "epoch": 6.302979987765446,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018753636044897026,
      "loss": 5.1579,
      "mean_token_accuracy": 0.22292775213718413,
      "num_tokens": 146149405.0,
      "step": 72125
    },
    {
      "entropy": 5.889864921569824,
      "epoch": 6.3034169361181505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018750765092103306,
      "loss": 5.1714,
      "mean_token_accuracy": 0.2185973897576332,
      "num_tokens": 146160256.0,
      "step": 72130
    },
    {
      "entropy": 5.875790977478028,
      "epoch": 6.303853884470856,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018747894307124486,
      "loss": 5.1254,
      "mean_token_accuracy": 0.22125847786664962,
      "num_tokens": 146169960.0,
      "step": 72135
    },
    {
      "entropy": 5.780628776550293,
      "epoch": 6.30429083282356,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018745023690015645,
      "loss": 5.0125,
      "mean_token_accuracy": 0.2435020014643669,
      "num_tokens": 146180073.0,
      "step": 72140
    },
    {
      "entropy": 5.82453293800354,
      "epoch": 6.304727781176265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001874215324083184,
      "loss": 5.1955,
      "mean_token_accuracy": 0.22174403667449952,
      "num_tokens": 146190294.0,
      "step": 72145
    },
    {
      "entropy": 5.812459564208984,
      "epoch": 6.30516472952897,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018739282959628122,
      "loss": 5.0998,
      "mean_token_accuracy": 0.23061055988073348,
      "num_tokens": 146200705.0,
      "step": 72150
    },
    {
      "entropy": 5.82184362411499,
      "epoch": 6.305601677881675,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018736412846459544,
      "loss": 5.1805,
      "mean_token_accuracy": 0.22084500938653945,
      "num_tokens": 146210593.0,
      "step": 72155
    },
    {
      "entropy": 5.861445331573487,
      "epoch": 6.306038626234379,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018733542901381163,
      "loss": 5.1663,
      "mean_token_accuracy": 0.21895135939121246,
      "num_tokens": 146220896.0,
      "step": 72160
    },
    {
      "entropy": 5.879372644424438,
      "epoch": 6.3064755745870835,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001873067312444802,
      "loss": 5.1849,
      "mean_token_accuracy": 0.22508913278579712,
      "num_tokens": 146231381.0,
      "step": 72165
    },
    {
      "entropy": 5.863334655761719,
      "epoch": 6.306912522939789,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018727803515715164,
      "loss": 5.1325,
      "mean_token_accuracy": 0.22289154082536697,
      "num_tokens": 146241468.0,
      "step": 72170
    },
    {
      "entropy": 5.866286563873291,
      "epoch": 6.307349471292493,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018724934075237638,
      "loss": 5.2027,
      "mean_token_accuracy": 0.22281202673912048,
      "num_tokens": 146252042.0,
      "step": 72175
    },
    {
      "entropy": 5.893193244934082,
      "epoch": 6.307786419645198,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001872206480307047,
      "loss": 5.1264,
      "mean_token_accuracy": 0.22498160600662231,
      "num_tokens": 146261849.0,
      "step": 72180
    },
    {
      "entropy": 5.870346164703369,
      "epoch": 6.308223367997902,
      "grad_norm": 1.078125,
      "learning_rate": 0.00018719195699268694,
      "loss": 5.146,
      "mean_token_accuracy": 0.2258158341050148,
      "num_tokens": 146273009.0,
      "step": 72185
    },
    {
      "entropy": 5.88878173828125,
      "epoch": 6.308660316350608,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018716326763887348,
      "loss": 5.2433,
      "mean_token_accuracy": 0.2188969910144806,
      "num_tokens": 146282349.0,
      "step": 72190
    },
    {
      "entropy": 5.840032005310059,
      "epoch": 6.309097264703312,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018713457996981465,
      "loss": 5.1993,
      "mean_token_accuracy": 0.22573923617601394,
      "num_tokens": 146292556.0,
      "step": 72195
    },
    {
      "entropy": 5.857830667495728,
      "epoch": 6.309534213056017,
      "grad_norm": 1.078125,
      "learning_rate": 0.00018710589398606055,
      "loss": 5.1279,
      "mean_token_accuracy": 0.22596187740564347,
      "num_tokens": 146302117.0,
      "step": 72200
    },
    {
      "entropy": 5.880795669555664,
      "epoch": 6.309971161408722,
      "grad_norm": 1.078125,
      "learning_rate": 0.00018707720968816143,
      "loss": 5.1756,
      "mean_token_accuracy": 0.22617925107479095,
      "num_tokens": 146312572.0,
      "step": 72205
    },
    {
      "entropy": 5.896174955368042,
      "epoch": 6.310408109761426,
      "grad_norm": 1.109375,
      "learning_rate": 0.00018704852707666754,
      "loss": 5.1753,
      "mean_token_accuracy": 0.22061055153608322,
      "num_tokens": 146322395.0,
      "step": 72210
    },
    {
      "entropy": 5.831590223312378,
      "epoch": 6.310845058114131,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018701984615212896,
      "loss": 5.1655,
      "mean_token_accuracy": 0.2262759268283844,
      "num_tokens": 146331876.0,
      "step": 72215
    },
    {
      "entropy": 5.8317680835723875,
      "epoch": 6.311282006466835,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018699116691509588,
      "loss": 5.2013,
      "mean_token_accuracy": 0.21996092200279235,
      "num_tokens": 146341168.0,
      "step": 72220
    },
    {
      "entropy": 5.758568859100341,
      "epoch": 6.311718954819541,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001869624893661183,
      "loss": 5.1047,
      "mean_token_accuracy": 0.2299373596906662,
      "num_tokens": 146350550.0,
      "step": 72225
    },
    {
      "entropy": 5.851555061340332,
      "epoch": 6.312155903172245,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018693381350574635,
      "loss": 5.1342,
      "mean_token_accuracy": 0.22071764767169952,
      "num_tokens": 146359943.0,
      "step": 72230
    },
    {
      "entropy": 5.891003084182739,
      "epoch": 6.31259285152495,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001869051393345299,
      "loss": 5.1986,
      "mean_token_accuracy": 0.2242514669895172,
      "num_tokens": 146369778.0,
      "step": 72235
    },
    {
      "entropy": 5.86409068107605,
      "epoch": 6.313029799877654,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018687646685301912,
      "loss": 5.1437,
      "mean_token_accuracy": 0.22863148301839828,
      "num_tokens": 146380142.0,
      "step": 72240
    },
    {
      "entropy": 5.854613780975342,
      "epoch": 6.3134667482303595,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001868477960617639,
      "loss": 5.1058,
      "mean_token_accuracy": 0.2279454752802849,
      "num_tokens": 146390175.0,
      "step": 72245
    },
    {
      "entropy": 5.795762920379639,
      "epoch": 6.313903696583064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001868191269613141,
      "loss": 5.1446,
      "mean_token_accuracy": 0.22144603580236435,
      "num_tokens": 146401439.0,
      "step": 72250
    },
    {
      "entropy": 5.864608716964722,
      "epoch": 6.314340644935768,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018679045955221962,
      "loss": 5.1933,
      "mean_token_accuracy": 0.21332387775182723,
      "num_tokens": 146411175.0,
      "step": 72255
    },
    {
      "entropy": 5.865074253082275,
      "epoch": 6.314777593288474,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001867617938350304,
      "loss": 5.1319,
      "mean_token_accuracy": 0.22295069247484206,
      "num_tokens": 146419881.0,
      "step": 72260
    },
    {
      "entropy": 5.847608423233032,
      "epoch": 6.315214541641178,
      "grad_norm": 1.25,
      "learning_rate": 0.00018673312981029615,
      "loss": 5.112,
      "mean_token_accuracy": 0.23451610654592514,
      "num_tokens": 146429387.0,
      "step": 72265
    },
    {
      "entropy": 5.895270204544067,
      "epoch": 6.315651489993883,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018670446747856678,
      "loss": 5.1964,
      "mean_token_accuracy": 0.22264638245105745,
      "num_tokens": 146438627.0,
      "step": 72270
    },
    {
      "entropy": 5.739557313919067,
      "epoch": 6.316088438346587,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018667580684039205,
      "loss": 5.0025,
      "mean_token_accuracy": 0.24003732204437256,
      "num_tokens": 146449481.0,
      "step": 72275
    },
    {
      "entropy": 5.8272310256958,
      "epoch": 6.3165253866992925,
      "grad_norm": 1.125,
      "learning_rate": 0.00018664714789632143,
      "loss": 5.1324,
      "mean_token_accuracy": 0.2268952026963234,
      "num_tokens": 146460324.0,
      "step": 72280
    },
    {
      "entropy": 5.836705827713013,
      "epoch": 6.316962335051997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018661849064690494,
      "loss": 5.0691,
      "mean_token_accuracy": 0.2334319144487381,
      "num_tokens": 146469389.0,
      "step": 72285
    },
    {
      "entropy": 5.869317245483399,
      "epoch": 6.3173992834047015,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018658983509269213,
      "loss": 5.1218,
      "mean_token_accuracy": 0.22933177202939986,
      "num_tokens": 146478969.0,
      "step": 72290
    },
    {
      "entropy": 5.873444080352783,
      "epoch": 6.317836231757406,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018656118123423256,
      "loss": 5.1731,
      "mean_token_accuracy": 0.21448509842157365,
      "num_tokens": 146489902.0,
      "step": 72295
    },
    {
      "entropy": 5.847076511383056,
      "epoch": 6.318273180110111,
      "grad_norm": 1.03125,
      "learning_rate": 0.00018653252907207586,
      "loss": 5.1146,
      "mean_token_accuracy": 0.2294072687625885,
      "num_tokens": 146500176.0,
      "step": 72300
    },
    {
      "entropy": 5.833225393295288,
      "epoch": 6.318710128462816,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018650387860677165,
      "loss": 5.1119,
      "mean_token_accuracy": 0.23436062932014465,
      "num_tokens": 146510703.0,
      "step": 72305
    },
    {
      "entropy": 5.871707391738892,
      "epoch": 6.31914707681552,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018647522983886945,
      "loss": 5.2034,
      "mean_token_accuracy": 0.22085948139429093,
      "num_tokens": 146519973.0,
      "step": 72310
    },
    {
      "entropy": 5.862466430664062,
      "epoch": 6.319584025168225,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018644658276891868,
      "loss": 5.1427,
      "mean_token_accuracy": 0.22479356080293655,
      "num_tokens": 146529877.0,
      "step": 72315
    },
    {
      "entropy": 5.8292614936828615,
      "epoch": 6.32002097352093,
      "grad_norm": 1.125,
      "learning_rate": 0.00018641793739746887,
      "loss": 5.1215,
      "mean_token_accuracy": 0.22907039821147918,
      "num_tokens": 146539640.0,
      "step": 72320
    },
    {
      "entropy": 5.908474493026733,
      "epoch": 6.3204579218736345,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001863892937250694,
      "loss": 5.2227,
      "mean_token_accuracy": 0.21897317320108414,
      "num_tokens": 146550048.0,
      "step": 72325
    },
    {
      "entropy": 5.839481163024902,
      "epoch": 6.320894870226339,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018636065175226975,
      "loss": 5.1347,
      "mean_token_accuracy": 0.2232520878314972,
      "num_tokens": 146561407.0,
      "step": 72330
    },
    {
      "entropy": 5.935452270507812,
      "epoch": 6.321331818579044,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018633201147961925,
      "loss": 5.1638,
      "mean_token_accuracy": 0.21698663085699083,
      "num_tokens": 146570808.0,
      "step": 72335
    },
    {
      "entropy": 5.824620532989502,
      "epoch": 6.321768766931749,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018630337290766725,
      "loss": 5.1093,
      "mean_token_accuracy": 0.22705416828393937,
      "num_tokens": 146580671.0,
      "step": 72340
    },
    {
      "entropy": 5.811110544204712,
      "epoch": 6.322205715284453,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018627473603696299,
      "loss": 5.115,
      "mean_token_accuracy": 0.22992196977138518,
      "num_tokens": 146591422.0,
      "step": 72345
    },
    {
      "entropy": 5.821047115325928,
      "epoch": 6.322642663637158,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018624610086805576,
      "loss": 5.1448,
      "mean_token_accuracy": 0.2231715589761734,
      "num_tokens": 146601475.0,
      "step": 72350
    },
    {
      "entropy": 5.842620706558227,
      "epoch": 6.323079611989863,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018621746740149492,
      "loss": 5.1535,
      "mean_token_accuracy": 0.2239365756511688,
      "num_tokens": 146611845.0,
      "step": 72355
    },
    {
      "entropy": 5.815116024017334,
      "epoch": 6.323516560342568,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018618883563782952,
      "loss": 5.0816,
      "mean_token_accuracy": 0.23249044567346572,
      "num_tokens": 146621314.0,
      "step": 72360
    },
    {
      "entropy": 5.855613851547242,
      "epoch": 6.323953508695272,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018616020557760878,
      "loss": 5.1908,
      "mean_token_accuracy": 0.21956486105918885,
      "num_tokens": 146632021.0,
      "step": 72365
    },
    {
      "entropy": 5.929385948181152,
      "epoch": 6.3243904570479765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001861315772213818,
      "loss": 5.3034,
      "mean_token_accuracy": 0.21837970614433289,
      "num_tokens": 146644016.0,
      "step": 72370
    },
    {
      "entropy": 5.889007568359375,
      "epoch": 6.324827405400682,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018610295056969777,
      "loss": 5.1723,
      "mean_token_accuracy": 0.2281434118747711,
      "num_tokens": 146653827.0,
      "step": 72375
    },
    {
      "entropy": 5.916420888900757,
      "epoch": 6.325264353753386,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018607432562310578,
      "loss": 5.1984,
      "mean_token_accuracy": 0.218391315639019,
      "num_tokens": 146663662.0,
      "step": 72380
    },
    {
      "entropy": 5.906700468063354,
      "epoch": 6.325701302106091,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001860457023821548,
      "loss": 5.2236,
      "mean_token_accuracy": 0.2195701539516449,
      "num_tokens": 146674466.0,
      "step": 72385
    },
    {
      "entropy": 5.867724370956421,
      "epoch": 6.326138250458796,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018601708084739382,
      "loss": 5.1578,
      "mean_token_accuracy": 0.2266852602362633,
      "num_tokens": 146684091.0,
      "step": 72390
    },
    {
      "entropy": 5.879706573486328,
      "epoch": 6.326575198811501,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001859884610193719,
      "loss": 5.1406,
      "mean_token_accuracy": 0.22908557951450348,
      "num_tokens": 146694313.0,
      "step": 72395
    },
    {
      "entropy": 5.876324701309204,
      "epoch": 6.327012147164205,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018595984289863793,
      "loss": 5.2175,
      "mean_token_accuracy": 0.22679202258586884,
      "num_tokens": 146705576.0,
      "step": 72400
    },
    {
      "entropy": 5.887454652786255,
      "epoch": 6.32744909551691,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018593122648574085,
      "loss": 5.1612,
      "mean_token_accuracy": 0.2214813247323036,
      "num_tokens": 146715559.0,
      "step": 72405
    },
    {
      "entropy": 5.839494085311889,
      "epoch": 6.327886043869615,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018590261178122946,
      "loss": 5.1486,
      "mean_token_accuracy": 0.22317339181900026,
      "num_tokens": 146725374.0,
      "step": 72410
    },
    {
      "entropy": 5.803871536254883,
      "epoch": 6.328322992222319,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001858739987856527,
      "loss": 5.0364,
      "mean_token_accuracy": 0.23195461183786392,
      "num_tokens": 146734848.0,
      "step": 72415
    },
    {
      "entropy": 5.799620246887207,
      "epoch": 6.328759940575024,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018584538749955932,
      "loss": 5.1245,
      "mean_token_accuracy": 0.22314271181821824,
      "num_tokens": 146744115.0,
      "step": 72420
    },
    {
      "entropy": 5.768437337875366,
      "epoch": 6.329196888927728,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001858167779234981,
      "loss": 5.1192,
      "mean_token_accuracy": 0.22992843985557557,
      "num_tokens": 146756208.0,
      "step": 72425
    },
    {
      "entropy": 5.865057182312012,
      "epoch": 6.329633837280434,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001857881700580179,
      "loss": 5.0966,
      "mean_token_accuracy": 0.22466551810503005,
      "num_tokens": 146766711.0,
      "step": 72430
    },
    {
      "entropy": 5.832705783843994,
      "epoch": 6.330070785633138,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018575956390366727,
      "loss": 5.1032,
      "mean_token_accuracy": 0.23071803003549576,
      "num_tokens": 146777385.0,
      "step": 72435
    },
    {
      "entropy": 5.854583883285523,
      "epoch": 6.330507733985843,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000185730959460995,
      "loss": 5.2206,
      "mean_token_accuracy": 0.2209946855902672,
      "num_tokens": 146787670.0,
      "step": 72440
    },
    {
      "entropy": 5.855083990097046,
      "epoch": 6.330944682338547,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001857023567305497,
      "loss": 5.1618,
      "mean_token_accuracy": 0.2246311604976654,
      "num_tokens": 146797909.0,
      "step": 72445
    },
    {
      "entropy": 5.848271036148072,
      "epoch": 6.3313816306912525,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001856737557128799,
      "loss": 5.1209,
      "mean_token_accuracy": 0.2243381142616272,
      "num_tokens": 146809342.0,
      "step": 72450
    },
    {
      "entropy": 5.861229372024536,
      "epoch": 6.331818579043957,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018564515640853441,
      "loss": 5.1802,
      "mean_token_accuracy": 0.2221403256058693,
      "num_tokens": 146819325.0,
      "step": 72455
    },
    {
      "entropy": 5.815156507492065,
      "epoch": 6.332255527396661,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018561655881806161,
      "loss": 5.1133,
      "mean_token_accuracy": 0.23100340962409974,
      "num_tokens": 146829807.0,
      "step": 72460
    },
    {
      "entropy": 5.859108114242554,
      "epoch": 6.332692475749367,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018558796294201004,
      "loss": 5.1403,
      "mean_token_accuracy": 0.2284606710076332,
      "num_tokens": 146839523.0,
      "step": 72465
    },
    {
      "entropy": 5.786097288131714,
      "epoch": 6.333129424102071,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00018555936878092818,
      "loss": 5.0779,
      "mean_token_accuracy": 0.23323225677013398,
      "num_tokens": 146850221.0,
      "step": 72470
    },
    {
      "entropy": 5.840610456466675,
      "epoch": 6.333566372454776,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00018553077633536448,
      "loss": 5.1643,
      "mean_token_accuracy": 0.22332069277763367,
      "num_tokens": 146859790.0,
      "step": 72475
    },
    {
      "entropy": 5.88089656829834,
      "epoch": 6.33400332080748,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018550218560586736,
      "loss": 5.1524,
      "mean_token_accuracy": 0.22387775480747224,
      "num_tokens": 146869473.0,
      "step": 72480
    },
    {
      "entropy": 5.847297143936157,
      "epoch": 6.3344402691601855,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018547359659298524,
      "loss": 5.108,
      "mean_token_accuracy": 0.23111696690320968,
      "num_tokens": 146878576.0,
      "step": 72485
    },
    {
      "entropy": 5.84265832901001,
      "epoch": 6.33487721751289,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018544500929726642,
      "loss": 5.1557,
      "mean_token_accuracy": 0.23007678389549255,
      "num_tokens": 146888556.0,
      "step": 72490
    },
    {
      "entropy": 5.785219812393189,
      "epoch": 6.3353141658655945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00018541642371925927,
      "loss": 5.0806,
      "mean_token_accuracy": 0.23009790182113649,
      "num_tokens": 146898983.0,
      "step": 72495
    },
    {
      "entropy": 5.838924932479858,
      "epoch": 6.335751114218299,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018538783985951214,
      "loss": 5.1387,
      "mean_token_accuracy": 0.22210775762796403,
      "num_tokens": 146909272.0,
      "step": 72500
    },
    {
      "entropy": 5.832192659378052,
      "epoch": 6.336188062571004,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018535925771857308,
      "loss": 5.1177,
      "mean_token_accuracy": 0.22808535546064376,
      "num_tokens": 146919565.0,
      "step": 72505
    },
    {
      "entropy": 5.9214166641235355,
      "epoch": 6.336625010923709,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00018533067729699043,
      "loss": 5.2326,
      "mean_token_accuracy": 0.21980534344911576,
      "num_tokens": 146930906.0,
      "step": 72510
    },
    {
      "entropy": 5.900609302520752,
      "epoch": 6.337061959276413,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001853020985953124,
      "loss": 5.1689,
      "mean_token_accuracy": 0.21639546602964402,
      "num_tokens": 146941361.0,
      "step": 72515
    },
    {
      "entropy": 5.8283881664276125,
      "epoch": 6.337498907629119,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00018527352161408711,
      "loss": 5.0869,
      "mean_token_accuracy": 0.2291995108127594,
      "num_tokens": 146951417.0,
      "step": 72520
    },
    {
      "entropy": 5.874202728271484,
      "epoch": 6.337935855981823,
      "grad_norm": 1.25,
      "learning_rate": 0.00018524494635386275,
      "loss": 5.1698,
      "mean_token_accuracy": 0.22214027047157286,
      "num_tokens": 146961505.0,
      "step": 72525
    },
    {
      "entropy": 5.821188926696777,
      "epoch": 6.3383728043345275,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018521637281518734,
      "loss": 5.1078,
      "mean_token_accuracy": 0.23553853034973143,
      "num_tokens": 146971367.0,
      "step": 72530
    },
    {
      "entropy": 5.860066032409668,
      "epoch": 6.338809752687232,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018518780099860882,
      "loss": 5.1822,
      "mean_token_accuracy": 0.22315552830696106,
      "num_tokens": 146981589.0,
      "step": 72535
    },
    {
      "entropy": 5.860833692550659,
      "epoch": 6.339246701039937,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018515923090467546,
      "loss": 5.1672,
      "mean_token_accuracy": 0.22299739569425583,
      "num_tokens": 146991570.0,
      "step": 72540
    },
    {
      "entropy": 5.914301061630249,
      "epoch": 6.339683649392642,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018513066253393511,
      "loss": 5.2497,
      "mean_token_accuracy": 0.21716898083686828,
      "num_tokens": 147000939.0,
      "step": 72545
    },
    {
      "entropy": 5.884474992752075,
      "epoch": 6.340120597745346,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001851020958869357,
      "loss": 5.1821,
      "mean_token_accuracy": 0.216640867292881,
      "num_tokens": 147010292.0,
      "step": 72550
    },
    {
      "entropy": 5.842664051055908,
      "epoch": 6.340557546098051,
      "grad_norm": 1.09375,
      "learning_rate": 0.00018507353096422523,
      "loss": 5.1563,
      "mean_token_accuracy": 0.22008133232593535,
      "num_tokens": 147020728.0,
      "step": 72555
    },
    {
      "entropy": 5.865942764282226,
      "epoch": 6.340994494450756,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001850449677663516,
      "loss": 5.1676,
      "mean_token_accuracy": 0.22380214631557466,
      "num_tokens": 147030787.0,
      "step": 72560
    },
    {
      "entropy": 5.872980690002441,
      "epoch": 6.341431442803461,
      "grad_norm": 1.125,
      "learning_rate": 0.00018501640629386257,
      "loss": 5.2341,
      "mean_token_accuracy": 0.21945943534374238,
      "num_tokens": 147040297.0,
      "step": 72565
    },
    {
      "entropy": 5.8661461353302,
      "epoch": 6.341868391156165,
      "grad_norm": 1.203125,
      "learning_rate": 0.000184987846547306,
      "loss": 5.067,
      "mean_token_accuracy": 0.226376448571682,
      "num_tokens": 147050086.0,
      "step": 72570
    },
    {
      "entropy": 5.88127293586731,
      "epoch": 6.34230533950887,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018495928852722975,
      "loss": 5.1868,
      "mean_token_accuracy": 0.21754722148180008,
      "num_tokens": 147060875.0,
      "step": 72575
    },
    {
      "entropy": 5.8447190284729,
      "epoch": 6.342742287861575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001849307322341815,
      "loss": 5.0972,
      "mean_token_accuracy": 0.22790978401899337,
      "num_tokens": 147070533.0,
      "step": 72580
    },
    {
      "entropy": 5.842341375350952,
      "epoch": 6.343179236214279,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018490217766870898,
      "loss": 5.2121,
      "mean_token_accuracy": 0.21788451373577117,
      "num_tokens": 147081529.0,
      "step": 72585
    },
    {
      "entropy": 5.82121000289917,
      "epoch": 6.343616184566984,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018487362483135994,
      "loss": 5.0754,
      "mean_token_accuracy": 0.21884619295597077,
      "num_tokens": 147091988.0,
      "step": 72590
    },
    {
      "entropy": 5.873034381866455,
      "epoch": 6.344053132919689,
      "grad_norm": 1.15625,
      "learning_rate": 0.000184845073722682,
      "loss": 5.1184,
      "mean_token_accuracy": 0.22742220461368562,
      "num_tokens": 147101866.0,
      "step": 72595
    },
    {
      "entropy": 5.763231658935547,
      "epoch": 6.344490081272394,
      "grad_norm": 1.109375,
      "learning_rate": 0.00018481652434322276,
      "loss": 5.1138,
      "mean_token_accuracy": 0.23028398156166077,
      "num_tokens": 147111532.0,
      "step": 72600
    },
    {
      "entropy": 5.824702882766724,
      "epoch": 6.344927029625098,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018478797669352988,
      "loss": 5.1427,
      "mean_token_accuracy": 0.2209091603755951,
      "num_tokens": 147121337.0,
      "step": 72605
    },
    {
      "entropy": 5.7690459251403805,
      "epoch": 6.345363977977803,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018475943077415086,
      "loss": 5.0622,
      "mean_token_accuracy": 0.23238218426704407,
      "num_tokens": 147130490.0,
      "step": 72610
    },
    {
      "entropy": 5.849278211593628,
      "epoch": 6.345800926330508,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001847308865856332,
      "loss": 5.1385,
      "mean_token_accuracy": 0.22579171806573867,
      "num_tokens": 147139971.0,
      "step": 72615
    },
    {
      "entropy": 5.891011190414429,
      "epoch": 6.346237874683212,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018470234412852444,
      "loss": 5.1614,
      "mean_token_accuracy": 0.22316704988479613,
      "num_tokens": 147151720.0,
      "step": 72620
    },
    {
      "entropy": 5.816457223892212,
      "epoch": 6.346674823035917,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00018467380340337215,
      "loss": 5.0737,
      "mean_token_accuracy": 0.22512973546981813,
      "num_tokens": 147162336.0,
      "step": 72625
    },
    {
      "entropy": 5.842676639556885,
      "epoch": 6.347111771388622,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018464526441072356,
      "loss": 5.0863,
      "mean_token_accuracy": 0.23188547641038895,
      "num_tokens": 147171248.0,
      "step": 72630
    },
    {
      "entropy": 5.7748682498931885,
      "epoch": 6.347548719741327,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018461672715112604,
      "loss": 5.0369,
      "mean_token_accuracy": 0.23005905300378798,
      "num_tokens": 147181088.0,
      "step": 72635
    },
    {
      "entropy": 5.788130807876587,
      "epoch": 6.347985668094031,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018458819162512713,
      "loss": 5.1142,
      "mean_token_accuracy": 0.23305141627788545,
      "num_tokens": 147191360.0,
      "step": 72640
    },
    {
      "entropy": 5.886877489089966,
      "epoch": 6.348422616446736,
      "grad_norm": 1.25,
      "learning_rate": 0.00018455965783327412,
      "loss": 5.1536,
      "mean_token_accuracy": 0.2222736433148384,
      "num_tokens": 147202110.0,
      "step": 72645
    },
    {
      "entropy": 5.818548583984375,
      "epoch": 6.348859564799441,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00018453112577611425,
      "loss": 5.1109,
      "mean_token_accuracy": 0.23430984914302827,
      "num_tokens": 147211918.0,
      "step": 72650
    },
    {
      "entropy": 5.847478675842285,
      "epoch": 6.3492965131521455,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018450259545419473,
      "loss": 5.1048,
      "mean_token_accuracy": 0.23119520097970964,
      "num_tokens": 147220798.0,
      "step": 72655
    },
    {
      "entropy": 5.884366703033447,
      "epoch": 6.34973346150485,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018447406686806296,
      "loss": 5.2136,
      "mean_token_accuracy": 0.21951155066490174,
      "num_tokens": 147230908.0,
      "step": 72660
    },
    {
      "entropy": 5.844717741012573,
      "epoch": 6.350170409857554,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018444554001826596,
      "loss": 5.0864,
      "mean_token_accuracy": 0.22921257615089416,
      "num_tokens": 147241160.0,
      "step": 72665
    },
    {
      "entropy": 5.892133712768555,
      "epoch": 6.35060735821026,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018441701490535097,
      "loss": 5.2027,
      "mean_token_accuracy": 0.2249235212802887,
      "num_tokens": 147251168.0,
      "step": 72670
    },
    {
      "entropy": 5.872649621963501,
      "epoch": 6.351044306562964,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018438849152986502,
      "loss": 5.152,
      "mean_token_accuracy": 0.22296299934387206,
      "num_tokens": 147260465.0,
      "step": 72675
    },
    {
      "entropy": 5.84494686126709,
      "epoch": 6.351481254915669,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00018435996989235538,
      "loss": 5.1358,
      "mean_token_accuracy": 0.2232736676931381,
      "num_tokens": 147269704.0,
      "step": 72680
    },
    {
      "entropy": 5.810068655014038,
      "epoch": 6.351918203268374,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018433144999336905,
      "loss": 5.125,
      "mean_token_accuracy": 0.22694168835878373,
      "num_tokens": 147280411.0,
      "step": 72685
    },
    {
      "entropy": 5.869616222381592,
      "epoch": 6.3523551516210786,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018430293183345291,
      "loss": 5.1932,
      "mean_token_accuracy": 0.21956128180027007,
      "num_tokens": 147290877.0,
      "step": 72690
    },
    {
      "entropy": 5.83351469039917,
      "epoch": 6.352792099973783,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001842744154131541,
      "loss": 5.0697,
      "mean_token_accuracy": 0.23644603937864303,
      "num_tokens": 147301105.0,
      "step": 72695
    },
    {
      "entropy": 5.815585327148438,
      "epoch": 6.3532290483264875,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018424590073301957,
      "loss": 5.1855,
      "mean_token_accuracy": 0.22385303676128387,
      "num_tokens": 147310870.0,
      "step": 72700
    },
    {
      "entropy": 5.743127012252808,
      "epoch": 6.353665996679193,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018421738779359615,
      "loss": 5.0304,
      "mean_token_accuracy": 0.24384348839521408,
      "num_tokens": 147320376.0,
      "step": 72705
    },
    {
      "entropy": 5.769418478012085,
      "epoch": 6.354102945031897,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018418887659543088,
      "loss": 5.0498,
      "mean_token_accuracy": 0.23887379169464112,
      "num_tokens": 147330548.0,
      "step": 72710
    },
    {
      "entropy": 5.889265298843384,
      "epoch": 6.354539893384602,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018416036713907047,
      "loss": 5.1138,
      "mean_token_accuracy": 0.22313560098409652,
      "num_tokens": 147340316.0,
      "step": 72715
    },
    {
      "entropy": 5.899226474761963,
      "epoch": 6.354976841737306,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018413185942506188,
      "loss": 5.238,
      "mean_token_accuracy": 0.2154419466853142,
      "num_tokens": 147350281.0,
      "step": 72720
    },
    {
      "entropy": 5.8536975383758545,
      "epoch": 6.355413790090012,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018410335345395178,
      "loss": 5.1275,
      "mean_token_accuracy": 0.22568182051181793,
      "num_tokens": 147361330.0,
      "step": 72725
    },
    {
      "entropy": 5.848275423049927,
      "epoch": 6.355850738442716,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018407484922628703,
      "loss": 5.0787,
      "mean_token_accuracy": 0.23065268695354463,
      "num_tokens": 147370716.0,
      "step": 72730
    },
    {
      "entropy": 5.812117099761963,
      "epoch": 6.3562876867954206,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018404634674261432,
      "loss": 5.1869,
      "mean_token_accuracy": 0.22187321335077287,
      "num_tokens": 147381431.0,
      "step": 72735
    },
    {
      "entropy": 5.886304330825806,
      "epoch": 6.356724635148126,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018401784600348037,
      "loss": 5.1758,
      "mean_token_accuracy": 0.2238205000758171,
      "num_tokens": 147390618.0,
      "step": 72740
    },
    {
      "entropy": 5.827964830398559,
      "epoch": 6.35716158350083,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018398934700943176,
      "loss": 5.1508,
      "mean_token_accuracy": 0.22171950340270996,
      "num_tokens": 147400508.0,
      "step": 72745
    },
    {
      "entropy": 5.78629093170166,
      "epoch": 6.357598531853535,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018396084976101513,
      "loss": 5.0468,
      "mean_token_accuracy": 0.2341980069875717,
      "num_tokens": 147411166.0,
      "step": 72750
    },
    {
      "entropy": 5.874769306182861,
      "epoch": 6.358035480206239,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018393235425877714,
      "loss": 5.1285,
      "mean_token_accuracy": 0.22263016253709794,
      "num_tokens": 147420927.0,
      "step": 72755
    },
    {
      "entropy": 5.80640754699707,
      "epoch": 6.358472428558945,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018390386050326436,
      "loss": 5.1284,
      "mean_token_accuracy": 0.22704059928655623,
      "num_tokens": 147430829.0,
      "step": 72760
    },
    {
      "entropy": 5.795860004425049,
      "epoch": 6.358909376911649,
      "grad_norm": 1.25,
      "learning_rate": 0.0001838753684950232,
      "loss": 5.0619,
      "mean_token_accuracy": 0.23503130972385405,
      "num_tokens": 147440913.0,
      "step": 72765
    },
    {
      "entropy": 5.815934514999389,
      "epoch": 6.359346325264354,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018384687823460027,
      "loss": 5.0467,
      "mean_token_accuracy": 0.2377118095755577,
      "num_tokens": 147450681.0,
      "step": 72770
    },
    {
      "entropy": 5.872298908233643,
      "epoch": 6.359783273617058,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018381838972254195,
      "loss": 5.2154,
      "mean_token_accuracy": 0.22207053750753403,
      "num_tokens": 147460019.0,
      "step": 72775
    },
    {
      "entropy": 5.871546602249145,
      "epoch": 6.360220221969763,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018378990295939469,
      "loss": 5.1554,
      "mean_token_accuracy": 0.2195468783378601,
      "num_tokens": 147469995.0,
      "step": 72780
    },
    {
      "entropy": 5.887806701660156,
      "epoch": 6.360657170322468,
      "grad_norm": 1.0625,
      "learning_rate": 0.00018376141794570493,
      "loss": 5.2388,
      "mean_token_accuracy": 0.21572041511535645,
      "num_tokens": 147480458.0,
      "step": 72785
    },
    {
      "entropy": 5.836370611190796,
      "epoch": 6.361094118675172,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001837329346820189,
      "loss": 5.1525,
      "mean_token_accuracy": 0.22396825551986693,
      "num_tokens": 147490081.0,
      "step": 72790
    },
    {
      "entropy": 5.899908113479614,
      "epoch": 6.361531067027878,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018370445316888306,
      "loss": 5.177,
      "mean_token_accuracy": 0.22299468964338304,
      "num_tokens": 147499897.0,
      "step": 72795
    },
    {
      "entropy": 5.907278966903687,
      "epoch": 6.361968015380582,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018367597340684366,
      "loss": 5.1612,
      "mean_token_accuracy": 0.21623616814613342,
      "num_tokens": 147509576.0,
      "step": 72800
    },
    {
      "entropy": 5.787242746353149,
      "epoch": 6.362404963733287,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018364749539644693,
      "loss": 5.036,
      "mean_token_accuracy": 0.243058979511261,
      "num_tokens": 147519123.0,
      "step": 72805
    },
    {
      "entropy": 5.7995829582214355,
      "epoch": 6.362841912085991,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018361901913823913,
      "loss": 5.1202,
      "mean_token_accuracy": 0.23284627944231034,
      "num_tokens": 147530492.0,
      "step": 72810
    },
    {
      "entropy": 5.915546131134033,
      "epoch": 6.3632788604386965,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018359054463276637,
      "loss": 5.2409,
      "mean_token_accuracy": 0.21978560388088225,
      "num_tokens": 147541816.0,
      "step": 72815
    },
    {
      "entropy": 5.890714025497436,
      "epoch": 6.363715808791401,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018356207188057489,
      "loss": 5.1739,
      "mean_token_accuracy": 0.22334277927875518,
      "num_tokens": 147552772.0,
      "step": 72820
    },
    {
      "entropy": 5.870220422744751,
      "epoch": 6.364152757144105,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018353360088221076,
      "loss": 5.1613,
      "mean_token_accuracy": 0.217934513092041,
      "num_tokens": 147562960.0,
      "step": 72825
    },
    {
      "entropy": 5.874628067016602,
      "epoch": 6.36458970549681,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018350513163822014,
      "loss": 5.1805,
      "mean_token_accuracy": 0.22720730304718018,
      "num_tokens": 147574039.0,
      "step": 72830
    },
    {
      "entropy": 5.833862686157227,
      "epoch": 6.365026653849515,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018347666414914905,
      "loss": 5.1215,
      "mean_token_accuracy": 0.2277178093791008,
      "num_tokens": 147584476.0,
      "step": 72835
    },
    {
      "entropy": 5.699459791183472,
      "epoch": 6.36546360220222,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001834481984155434,
      "loss": 5.0269,
      "mean_token_accuracy": 0.24411601573228836,
      "num_tokens": 147593690.0,
      "step": 72840
    },
    {
      "entropy": 5.8814819812774655,
      "epoch": 6.365900550554924,
      "grad_norm": 1.125,
      "learning_rate": 0.00018341973443794935,
      "loss": 5.2101,
      "mean_token_accuracy": 0.2171971619129181,
      "num_tokens": 147603428.0,
      "step": 72845
    },
    {
      "entropy": 5.911051559448242,
      "epoch": 6.36633749890763,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018339127221691275,
      "loss": 5.2543,
      "mean_token_accuracy": 0.2188694015145302,
      "num_tokens": 147614206.0,
      "step": 72850
    },
    {
      "entropy": 5.822590446472168,
      "epoch": 6.366774447260334,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001833628117529796,
      "loss": 5.0323,
      "mean_token_accuracy": 0.2338779240846634,
      "num_tokens": 147624878.0,
      "step": 72855
    },
    {
      "entropy": 5.8099829196929935,
      "epoch": 6.3672113956130385,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018333435304669572,
      "loss": 5.0623,
      "mean_token_accuracy": 0.23728567063808442,
      "num_tokens": 147634843.0,
      "step": 72860
    },
    {
      "entropy": 5.865124225616455,
      "epoch": 6.367648343965743,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018330589609860692,
      "loss": 5.1568,
      "mean_token_accuracy": 0.22851618826389314,
      "num_tokens": 147645205.0,
      "step": 72865
    },
    {
      "entropy": 5.797825813293457,
      "epoch": 6.368085292318448,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018327744090925914,
      "loss": 5.1063,
      "mean_token_accuracy": 0.22743639200925828,
      "num_tokens": 147655087.0,
      "step": 72870
    },
    {
      "entropy": 5.779280948638916,
      "epoch": 6.368522240671153,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001832489874791981,
      "loss": 5.0544,
      "mean_token_accuracy": 0.23469415456056594,
      "num_tokens": 147665441.0,
      "step": 72875
    },
    {
      "entropy": 5.847301578521728,
      "epoch": 6.368959189023857,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001832205358089696,
      "loss": 5.1701,
      "mean_token_accuracy": 0.22473006844520568,
      "num_tokens": 147675928.0,
      "step": 72880
    },
    {
      "entropy": 5.926280164718628,
      "epoch": 6.369396137376562,
      "grad_norm": 1.09375,
      "learning_rate": 0.00018319208589911924,
      "loss": 5.2387,
      "mean_token_accuracy": 0.2247183218598366,
      "num_tokens": 147686781.0,
      "step": 72885
    },
    {
      "entropy": 5.914874267578125,
      "epoch": 6.369833085729267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0001831636377501928,
      "loss": 5.1324,
      "mean_token_accuracy": 0.2199099376797676,
      "num_tokens": 147696641.0,
      "step": 72890
    },
    {
      "entropy": 5.850332355499267,
      "epoch": 6.370270034081972,
      "grad_norm": 1.078125,
      "learning_rate": 0.00018313519136273587,
      "loss": 5.1402,
      "mean_token_accuracy": 0.22585152387619017,
      "num_tokens": 147706787.0,
      "step": 72895
    },
    {
      "entropy": 5.775746488571167,
      "epoch": 6.370706982434676,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018310674673729417,
      "loss": 5.0718,
      "mean_token_accuracy": 0.23276919424533843,
      "num_tokens": 147717208.0,
      "step": 72900
    },
    {
      "entropy": 5.880547285079956,
      "epoch": 6.371143930787381,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018307830387441322,
      "loss": 5.146,
      "mean_token_accuracy": 0.22425372004508973,
      "num_tokens": 147726600.0,
      "step": 72905
    },
    {
      "entropy": 5.889324474334717,
      "epoch": 6.371580879140086,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018304986277463853,
      "loss": 5.174,
      "mean_token_accuracy": 0.22747987359762192,
      "num_tokens": 147737344.0,
      "step": 72910
    },
    {
      "entropy": 5.875094938278198,
      "epoch": 6.37201782749279,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018302142343851562,
      "loss": 5.2057,
      "mean_token_accuracy": 0.2180796042084694,
      "num_tokens": 147747291.0,
      "step": 72915
    },
    {
      "entropy": 5.8378032684326175,
      "epoch": 6.372454775845495,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001829929858665901,
      "loss": 5.1258,
      "mean_token_accuracy": 0.2300889179110527,
      "num_tokens": 147757015.0,
      "step": 72920
    },
    {
      "entropy": 5.8195476055145265,
      "epoch": 6.3728917241982,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018296455005940726,
      "loss": 5.1442,
      "mean_token_accuracy": 0.22338288873434067,
      "num_tokens": 147767283.0,
      "step": 72925
    },
    {
      "entropy": 5.865281820297241,
      "epoch": 6.373328672550905,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018293611601751265,
      "loss": 5.0876,
      "mean_token_accuracy": 0.2321346715092659,
      "num_tokens": 147776982.0,
      "step": 72930
    },
    {
      "entropy": 5.944683408737182,
      "epoch": 6.373765620903609,
      "grad_norm": 1.125,
      "learning_rate": 0.00018290768374145145,
      "loss": 5.2461,
      "mean_token_accuracy": 0.21446063816547395,
      "num_tokens": 147786725.0,
      "step": 72935
    },
    {
      "entropy": 5.915924596786499,
      "epoch": 6.374202569256314,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018287925323176923,
      "loss": 5.2524,
      "mean_token_accuracy": 0.21005170196294784,
      "num_tokens": 147797818.0,
      "step": 72940
    },
    {
      "entropy": 5.837444543838501,
      "epoch": 6.374639517609019,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018285082448901113,
      "loss": 5.1264,
      "mean_token_accuracy": 0.2226758450269699,
      "num_tokens": 147807736.0,
      "step": 72945
    },
    {
      "entropy": 5.88768162727356,
      "epoch": 6.375076465961723,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001828223975137226,
      "loss": 5.1559,
      "mean_token_accuracy": 0.22369003295898438,
      "num_tokens": 147817034.0,
      "step": 72950
    },
    {
      "entropy": 5.872758054733277,
      "epoch": 6.375513414314428,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018279397230644873,
      "loss": 5.1597,
      "mean_token_accuracy": 0.2204527735710144,
      "num_tokens": 147828057.0,
      "step": 72955
    },
    {
      "entropy": 5.8093931674957275,
      "epoch": 6.375950362667132,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018276554886773478,
      "loss": 5.1336,
      "mean_token_accuracy": 0.22308487594127654,
      "num_tokens": 147838659.0,
      "step": 72960
    },
    {
      "entropy": 5.915434789657593,
      "epoch": 6.376387311019838,
      "grad_norm": 1.25,
      "learning_rate": 0.0001827371271981259,
      "loss": 5.2563,
      "mean_token_accuracy": 0.2147725462913513,
      "num_tokens": 147849873.0,
      "step": 72965
    },
    {
      "entropy": 5.896128225326538,
      "epoch": 6.376824259372542,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001827087072981673,
      "loss": 5.1902,
      "mean_token_accuracy": 0.2153070881962776,
      "num_tokens": 147860454.0,
      "step": 72970
    },
    {
      "entropy": 5.846531248092651,
      "epoch": 6.377261207725247,
      "grad_norm": 1.25,
      "learning_rate": 0.00018268028916840407,
      "loss": 5.1023,
      "mean_token_accuracy": 0.22640683501958847,
      "num_tokens": 147870584.0,
      "step": 72975
    },
    {
      "entropy": 5.850745630264282,
      "epoch": 6.377698156077952,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018265187280938128,
      "loss": 5.1455,
      "mean_token_accuracy": 0.22395666986703872,
      "num_tokens": 147881142.0,
      "step": 72980
    },
    {
      "entropy": 5.825744390487671,
      "epoch": 6.3781351044306565,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018262345822164395,
      "loss": 5.0793,
      "mean_token_accuracy": 0.23309891819953918,
      "num_tokens": 147890863.0,
      "step": 72985
    },
    {
      "entropy": 5.874249649047852,
      "epoch": 6.378572052783361,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018259504540573703,
      "loss": 5.178,
      "mean_token_accuracy": 0.2195145770907402,
      "num_tokens": 147901015.0,
      "step": 72990
    },
    {
      "entropy": 5.843266201019287,
      "epoch": 6.379009001136065,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018256663436220561,
      "loss": 5.1587,
      "mean_token_accuracy": 0.22390560656785966,
      "num_tokens": 147911774.0,
      "step": 72995
    },
    {
      "entropy": 5.842492723464966,
      "epoch": 6.379445949488771,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018253822509159458,
      "loss": 5.0892,
      "mean_token_accuracy": 0.23159415572881697,
      "num_tokens": 147921861.0,
      "step": 73000
    },
    {
      "entropy": 5.855422258377075,
      "epoch": 6.379882897841475,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001825098175944488,
      "loss": 5.2033,
      "mean_token_accuracy": 0.21753801703453063,
      "num_tokens": 147932333.0,
      "step": 73005
    },
    {
      "entropy": 5.870339584350586,
      "epoch": 6.38031984619418,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018248141187131317,
      "loss": 5.1833,
      "mean_token_accuracy": 0.22128180265426636,
      "num_tokens": 147942623.0,
      "step": 73010
    },
    {
      "entropy": 5.810753011703492,
      "epoch": 6.380756794546884,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018245300792273268,
      "loss": 5.0706,
      "mean_token_accuracy": 0.23581429421901703,
      "num_tokens": 147952176.0,
      "step": 73015
    },
    {
      "entropy": 5.9047496795654295,
      "epoch": 6.3811937428995895,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018242460574925185,
      "loss": 5.1797,
      "mean_token_accuracy": 0.21900341361761094,
      "num_tokens": 147962445.0,
      "step": 73020
    },
    {
      "entropy": 5.791550588607788,
      "epoch": 6.381630691252294,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018239620535141565,
      "loss": 5.0711,
      "mean_token_accuracy": 0.2355584755539894,
      "num_tokens": 147972885.0,
      "step": 73025
    },
    {
      "entropy": 5.888671016693115,
      "epoch": 6.3820676396049985,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018236780672976877,
      "loss": 5.1457,
      "mean_token_accuracy": 0.21968605667352675,
      "num_tokens": 147982131.0,
      "step": 73030
    },
    {
      "entropy": 5.813709735870361,
      "epoch": 6.382504587957704,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018233940988485588,
      "loss": 5.0658,
      "mean_token_accuracy": 0.23302856385707854,
      "num_tokens": 147990919.0,
      "step": 73035
    },
    {
      "entropy": 5.842710208892822,
      "epoch": 6.382941536310408,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018231101481722166,
      "loss": 5.1825,
      "mean_token_accuracy": 0.2253037691116333,
      "num_tokens": 148000714.0,
      "step": 73040
    },
    {
      "entropy": 5.884775829315186,
      "epoch": 6.383378484663113,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001822826215274107,
      "loss": 5.209,
      "mean_token_accuracy": 0.22178139090538024,
      "num_tokens": 148010236.0,
      "step": 73045
    },
    {
      "entropy": 5.849586915969849,
      "epoch": 6.383815433015817,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001822542300159677,
      "loss": 5.1195,
      "mean_token_accuracy": 0.22512341737747193,
      "num_tokens": 148019885.0,
      "step": 73050
    },
    {
      "entropy": 5.87825722694397,
      "epoch": 6.384252381368523,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018222584028343715,
      "loss": 5.1313,
      "mean_token_accuracy": 0.22947173416614533,
      "num_tokens": 148029335.0,
      "step": 73055
    },
    {
      "entropy": 5.806527709960937,
      "epoch": 6.384689329721227,
      "grad_norm": 1.125,
      "learning_rate": 0.00018219745233036356,
      "loss": 5.097,
      "mean_token_accuracy": 0.23244793117046356,
      "num_tokens": 148041399.0,
      "step": 73060
    },
    {
      "entropy": 5.851353549957276,
      "epoch": 6.3851262780739315,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018216906615729157,
      "loss": 5.1311,
      "mean_token_accuracy": 0.22427565455436707,
      "num_tokens": 148051042.0,
      "step": 73065
    },
    {
      "entropy": 5.833610534667969,
      "epoch": 6.385563226426636,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00018214068176476545,
      "loss": 5.1663,
      "mean_token_accuracy": 0.22365642637014388,
      "num_tokens": 148060976.0,
      "step": 73070
    },
    {
      "entropy": 5.855484294891357,
      "epoch": 6.386000174779341,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018211229915332972,
      "loss": 5.1606,
      "mean_token_accuracy": 0.22397490590810776,
      "num_tokens": 148070554.0,
      "step": 73075
    },
    {
      "entropy": 5.856350135803223,
      "epoch": 6.386437123132046,
      "grad_norm": 1.25,
      "learning_rate": 0.00018208391832352872,
      "loss": 5.135,
      "mean_token_accuracy": 0.2248913124203682,
      "num_tokens": 148081007.0,
      "step": 73080
    },
    {
      "entropy": 5.860214614868164,
      "epoch": 6.38687407148475,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018205553927590695,
      "loss": 5.0834,
      "mean_token_accuracy": 0.23201249688863754,
      "num_tokens": 148090629.0,
      "step": 73085
    },
    {
      "entropy": 5.826263523101806,
      "epoch": 6.387311019837455,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018202716201100856,
      "loss": 5.0422,
      "mean_token_accuracy": 0.2323952779173851,
      "num_tokens": 148099950.0,
      "step": 73090
    },
    {
      "entropy": 5.88429102897644,
      "epoch": 6.38774796819016,
      "grad_norm": 1.25,
      "learning_rate": 0.00018199878652937795,
      "loss": 5.2584,
      "mean_token_accuracy": 0.21590195894241332,
      "num_tokens": 148111197.0,
      "step": 73095
    },
    {
      "entropy": 5.8400434970855715,
      "epoch": 6.388184916542865,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018197041283155935,
      "loss": 5.196,
      "mean_token_accuracy": 0.22148296386003494,
      "num_tokens": 148122674.0,
      "step": 73100
    },
    {
      "entropy": 5.856236410140991,
      "epoch": 6.388621864895569,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018194204091809697,
      "loss": 5.1374,
      "mean_token_accuracy": 0.22857286781072617,
      "num_tokens": 148132325.0,
      "step": 73105
    },
    {
      "entropy": 5.883678770065307,
      "epoch": 6.389058813248274,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018191367078953502,
      "loss": 5.1284,
      "mean_token_accuracy": 0.22196541875600814,
      "num_tokens": 148141677.0,
      "step": 73110
    },
    {
      "entropy": 5.869571399688721,
      "epoch": 6.389495761600979,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001818853024464176,
      "loss": 5.1129,
      "mean_token_accuracy": 0.22942440956830978,
      "num_tokens": 148151529.0,
      "step": 73115
    },
    {
      "entropy": 5.874906730651856,
      "epoch": 6.389932709953683,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018185693588928893,
      "loss": 5.1945,
      "mean_token_accuracy": 0.21349408626556396,
      "num_tokens": 148161701.0,
      "step": 73120
    },
    {
      "entropy": 5.8843694686889645,
      "epoch": 6.390369658306388,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000181828571118693,
      "loss": 5.1934,
      "mean_token_accuracy": 0.22308941036462784,
      "num_tokens": 148171892.0,
      "step": 73125
    },
    {
      "entropy": 5.903471040725708,
      "epoch": 6.390806606659093,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001818002081351739,
      "loss": 5.1791,
      "mean_token_accuracy": 0.2180839627981186,
      "num_tokens": 148184270.0,
      "step": 73130
    },
    {
      "entropy": 5.7866191387176515,
      "epoch": 6.391243555011798,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001817718469392757,
      "loss": 5.1018,
      "mean_token_accuracy": 0.22981298416852952,
      "num_tokens": 148193821.0,
      "step": 73135
    },
    {
      "entropy": 5.864806461334228,
      "epoch": 6.391680503364502,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018174348753154223,
      "loss": 5.1113,
      "mean_token_accuracy": 0.2296646311879158,
      "num_tokens": 148203710.0,
      "step": 73140
    },
    {
      "entropy": 5.822215270996094,
      "epoch": 6.392117451717207,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018171512991251758,
      "loss": 5.1212,
      "mean_token_accuracy": 0.23304915875196458,
      "num_tokens": 148213374.0,
      "step": 73145
    },
    {
      "entropy": 5.886362218856812,
      "epoch": 6.392554400069912,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001816867740827456,
      "loss": 5.2085,
      "mean_token_accuracy": 0.2115652531385422,
      "num_tokens": 148224196.0,
      "step": 73150
    },
    {
      "entropy": 5.847718954086304,
      "epoch": 6.392991348422616,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018165842004277017,
      "loss": 5.0937,
      "mean_token_accuracy": 0.22885774970054626,
      "num_tokens": 148234230.0,
      "step": 73155
    },
    {
      "entropy": 5.902749013900757,
      "epoch": 6.393428296775321,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00018163006779313524,
      "loss": 5.1824,
      "mean_token_accuracy": 0.2211112290620804,
      "num_tokens": 148244716.0,
      "step": 73160
    },
    {
      "entropy": 5.816548633575439,
      "epoch": 6.393865245128026,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018160171733438448,
      "loss": 5.087,
      "mean_token_accuracy": 0.23575109243392944,
      "num_tokens": 148253985.0,
      "step": 73165
    },
    {
      "entropy": 5.947744607925415,
      "epoch": 6.394302193480731,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018157336866706175,
      "loss": 5.2826,
      "mean_token_accuracy": 0.21062003523111344,
      "num_tokens": 148263911.0,
      "step": 73170
    },
    {
      "entropy": 5.851012706756592,
      "epoch": 6.394739141833435,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018154502179171074,
      "loss": 5.1796,
      "mean_token_accuracy": 0.21765550673007966,
      "num_tokens": 148273939.0,
      "step": 73175
    },
    {
      "entropy": 5.768212699890137,
      "epoch": 6.39517609018614,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018151667670887516,
      "loss": 5.0915,
      "mean_token_accuracy": 0.2343534618616104,
      "num_tokens": 148284302.0,
      "step": 73180
    },
    {
      "entropy": 5.901895523071289,
      "epoch": 6.395613038538845,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018148833341909875,
      "loss": 5.19,
      "mean_token_accuracy": 0.22438288629055023,
      "num_tokens": 148294040.0,
      "step": 73185
    },
    {
      "entropy": 5.889430570602417,
      "epoch": 6.3960499868915495,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018145999192292506,
      "loss": 5.1459,
      "mean_token_accuracy": 0.22582899481058122,
      "num_tokens": 148305059.0,
      "step": 73190
    },
    {
      "entropy": 5.8121942520141605,
      "epoch": 6.396486935244254,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018143165222089775,
      "loss": 5.1606,
      "mean_token_accuracy": 0.22732058316469192,
      "num_tokens": 148314768.0,
      "step": 73195
    },
    {
      "entropy": 5.84719443321228,
      "epoch": 6.396923883596958,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018140331431356038,
      "loss": 5.156,
      "mean_token_accuracy": 0.21798517554998398,
      "num_tokens": 148324172.0,
      "step": 73200
    },
    {
      "entropy": 5.763055944442749,
      "epoch": 6.397360831949664,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018137497820145648,
      "loss": 5.0601,
      "mean_token_accuracy": 0.23641018867492675,
      "num_tokens": 148333844.0,
      "step": 73205
    },
    {
      "entropy": 5.8517101287841795,
      "epoch": 6.397797780302368,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001813466438851296,
      "loss": 5.1161,
      "mean_token_accuracy": 0.23054094463586808,
      "num_tokens": 148343496.0,
      "step": 73210
    },
    {
      "entropy": 5.830206680297851,
      "epoch": 6.398234728655073,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018131831136512317,
      "loss": 5.1181,
      "mean_token_accuracy": 0.22211918383836746,
      "num_tokens": 148353301.0,
      "step": 73215
    },
    {
      "entropy": 5.8902403831481935,
      "epoch": 6.398671677007778,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018128998064198055,
      "loss": 5.1465,
      "mean_token_accuracy": 0.2222375065088272,
      "num_tokens": 148363235.0,
      "step": 73220
    },
    {
      "entropy": 5.755382871627807,
      "epoch": 6.3991086253604825,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018126165171624526,
      "loss": 5.0088,
      "mean_token_accuracy": 0.2390403375029564,
      "num_tokens": 148373608.0,
      "step": 73225
    },
    {
      "entropy": 5.840422487258911,
      "epoch": 6.399545573713187,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018123332458846054,
      "loss": 5.133,
      "mean_token_accuracy": 0.22465772181749344,
      "num_tokens": 148383378.0,
      "step": 73230
    },
    {
      "entropy": 5.755222129821777,
      "epoch": 6.3999825220658915,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018120499925916984,
      "loss": 5.0394,
      "mean_token_accuracy": 0.2387675866484642,
      "num_tokens": 148393472.0,
      "step": 73235
    },
    {
      "entropy": 5.813883638381958,
      "epoch": 6.400419470418597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00018117667572891639,
      "loss": 5.1724,
      "mean_token_accuracy": 0.22035592794418335,
      "num_tokens": 148404706.0,
      "step": 73240
    },
    {
      "entropy": 5.857028770446777,
      "epoch": 6.400856418771301,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018114835399824354,
      "loss": 5.1711,
      "mean_token_accuracy": 0.22070628255605698,
      "num_tokens": 148414351.0,
      "step": 73245
    },
    {
      "entropy": 5.882517147064209,
      "epoch": 6.401293367124006,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018112003406769433,
      "loss": 5.1703,
      "mean_token_accuracy": 0.22129089683294295,
      "num_tokens": 148424127.0,
      "step": 73250
    },
    {
      "entropy": 5.843047332763672,
      "epoch": 6.40173031547671,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001810917159378121,
      "loss": 5.1111,
      "mean_token_accuracy": 0.22151120603084565,
      "num_tokens": 148435147.0,
      "step": 73255
    },
    {
      "entropy": 5.857938766479492,
      "epoch": 6.402167263829416,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018106339960913997,
      "loss": 5.1454,
      "mean_token_accuracy": 0.22312745451927185,
      "num_tokens": 148445697.0,
      "step": 73260
    },
    {
      "entropy": 5.834020519256592,
      "epoch": 6.40260421218212,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018103508508222104,
      "loss": 5.1502,
      "mean_token_accuracy": 0.23183072209358216,
      "num_tokens": 148455856.0,
      "step": 73265
    },
    {
      "entropy": 5.896197843551636,
      "epoch": 6.4030411605348245,
      "grad_norm": 1.09375,
      "learning_rate": 0.00018100677235759845,
      "loss": 5.2004,
      "mean_token_accuracy": 0.2219221770763397,
      "num_tokens": 148466147.0,
      "step": 73270
    },
    {
      "entropy": 5.911596488952637,
      "epoch": 6.40347810888753,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018097846143581514,
      "loss": 5.2412,
      "mean_token_accuracy": 0.2140851855278015,
      "num_tokens": 148476084.0,
      "step": 73275
    },
    {
      "entropy": 5.929143714904785,
      "epoch": 6.403915057240234,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018095015231741425,
      "loss": 5.1939,
      "mean_token_accuracy": 0.22424292862415313,
      "num_tokens": 148485588.0,
      "step": 73280
    },
    {
      "entropy": 5.77805347442627,
      "epoch": 6.404352005592939,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018092184500293873,
      "loss": 5.0914,
      "mean_token_accuracy": 0.2363268807530403,
      "num_tokens": 148495417.0,
      "step": 73285
    },
    {
      "entropy": 5.820903491973877,
      "epoch": 6.404788953945643,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018089353949293148,
      "loss": 5.079,
      "mean_token_accuracy": 0.23044343590736388,
      "num_tokens": 148506494.0,
      "step": 73290
    },
    {
      "entropy": 5.835593938827515,
      "epoch": 6.405225902298349,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00018086523578793556,
      "loss": 5.1251,
      "mean_token_accuracy": 0.2284395784139633,
      "num_tokens": 148517363.0,
      "step": 73295
    },
    {
      "entropy": 5.848186206817627,
      "epoch": 6.405662850651053,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018083693388849364,
      "loss": 5.2417,
      "mean_token_accuracy": 0.21294416189193727,
      "num_tokens": 148527120.0,
      "step": 73300
    },
    {
      "entropy": 5.8198154926300045,
      "epoch": 6.406099799003758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018080863379514874,
      "loss": 5.1037,
      "mean_token_accuracy": 0.22831031531095505,
      "num_tokens": 148537345.0,
      "step": 73305
    },
    {
      "entropy": 5.7998027324676515,
      "epoch": 6.406536747356462,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018078033550844352,
      "loss": 5.0358,
      "mean_token_accuracy": 0.23248265832662582,
      "num_tokens": 148547012.0,
      "step": 73310
    },
    {
      "entropy": 5.789921855926513,
      "epoch": 6.406973695709167,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001807520390289209,
      "loss": 5.0152,
      "mean_token_accuracy": 0.23265812695026397,
      "num_tokens": 148558156.0,
      "step": 73315
    },
    {
      "entropy": 5.8296832084655765,
      "epoch": 6.407410644061872,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018072374435712353,
      "loss": 5.116,
      "mean_token_accuracy": 0.23383793085813523,
      "num_tokens": 148568454.0,
      "step": 73320
    },
    {
      "entropy": 5.856864976882934,
      "epoch": 6.407847592414576,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018069545149359412,
      "loss": 5.1573,
      "mean_token_accuracy": 0.22758512794971467,
      "num_tokens": 148578640.0,
      "step": 73325
    },
    {
      "entropy": 5.894910526275635,
      "epoch": 6.408284540767282,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018066716043887537,
      "loss": 5.1607,
      "mean_token_accuracy": 0.22105743139982223,
      "num_tokens": 148588376.0,
      "step": 73330
    },
    {
      "entropy": 5.78793363571167,
      "epoch": 6.408721489119986,
      "grad_norm": 1.125,
      "learning_rate": 0.0001806388711935099,
      "loss": 5.166,
      "mean_token_accuracy": 0.2219899445772171,
      "num_tokens": 148599574.0,
      "step": 73335
    },
    {
      "entropy": 5.832223749160766,
      "epoch": 6.409158437472691,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018061058375804046,
      "loss": 5.0837,
      "mean_token_accuracy": 0.2311653822660446,
      "num_tokens": 148609266.0,
      "step": 73340
    },
    {
      "entropy": 5.7978010177612305,
      "epoch": 6.409595385825395,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018058229813300931,
      "loss": 5.0621,
      "mean_token_accuracy": 0.23762242048978804,
      "num_tokens": 148619546.0,
      "step": 73345
    },
    {
      "entropy": 5.8066771030426025,
      "epoch": 6.4100323341781005,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001805540143189592,
      "loss": 5.0915,
      "mean_token_accuracy": 0.22991767823696135,
      "num_tokens": 148630796.0,
      "step": 73350
    },
    {
      "entropy": 5.849533605575561,
      "epoch": 6.410469282530805,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001805257323164326,
      "loss": 5.1054,
      "mean_token_accuracy": 0.23067524880170823,
      "num_tokens": 148641896.0,
      "step": 73355
    },
    {
      "entropy": 5.8588066577911375,
      "epoch": 6.410906230883509,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018049745212597197,
      "loss": 5.1304,
      "mean_token_accuracy": 0.22555844038724898,
      "num_tokens": 148652880.0,
      "step": 73360
    },
    {
      "entropy": 5.822659349441528,
      "epoch": 6.411343179236214,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018046917374811973,
      "loss": 5.1102,
      "mean_token_accuracy": 0.21970403790473939,
      "num_tokens": 148664128.0,
      "step": 73365
    },
    {
      "entropy": 5.848092174530029,
      "epoch": 6.411780127588919,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018044089718341822,
      "loss": 5.1693,
      "mean_token_accuracy": 0.22252161502838136,
      "num_tokens": 148674317.0,
      "step": 73370
    },
    {
      "entropy": 5.860824775695801,
      "epoch": 6.412217075941624,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018041262243240991,
      "loss": 5.1084,
      "mean_token_accuracy": 0.22884957790374755,
      "num_tokens": 148683863.0,
      "step": 73375
    },
    {
      "entropy": 5.906922006607056,
      "epoch": 6.412654024294328,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018038434949563714,
      "loss": 5.1493,
      "mean_token_accuracy": 0.22531970739364623,
      "num_tokens": 148693892.0,
      "step": 73380
    },
    {
      "entropy": 5.809532976150512,
      "epoch": 6.4130909726470335,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018035607837364197,
      "loss": 5.1093,
      "mean_token_accuracy": 0.2314614847302437,
      "num_tokens": 148703563.0,
      "step": 73385
    },
    {
      "entropy": 5.819369173049926,
      "epoch": 6.413527920999738,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018032780906696694,
      "loss": 5.1249,
      "mean_token_accuracy": 0.22145611494779588,
      "num_tokens": 148713231.0,
      "step": 73390
    },
    {
      "entropy": 5.872687673568725,
      "epoch": 6.4139648693524425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018029954157615402,
      "loss": 5.2306,
      "mean_token_accuracy": 0.22662691920995712,
      "num_tokens": 148723470.0,
      "step": 73395
    },
    {
      "entropy": 5.823874378204346,
      "epoch": 6.414401817705147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001802712759017456,
      "loss": 5.0556,
      "mean_token_accuracy": 0.2333597421646118,
      "num_tokens": 148732359.0,
      "step": 73400
    },
    {
      "entropy": 5.862537908554077,
      "epoch": 6.414838766057852,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018024301204428366,
      "loss": 5.2136,
      "mean_token_accuracy": 0.21505222618579864,
      "num_tokens": 148743884.0,
      "step": 73405
    },
    {
      "entropy": 5.800106239318848,
      "epoch": 6.415275714410557,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018021475000431047,
      "loss": 5.1099,
      "mean_token_accuracy": 0.2316698580980301,
      "num_tokens": 148753962.0,
      "step": 73410
    },
    {
      "entropy": 5.7763416290283205,
      "epoch": 6.415712662763261,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018018648978236802,
      "loss": 5.0206,
      "mean_token_accuracy": 0.23071998208761216,
      "num_tokens": 148763400.0,
      "step": 73415
    },
    {
      "entropy": 5.9183385372161865,
      "epoch": 6.416149611115966,
      "grad_norm": 1.109375,
      "learning_rate": 0.00018015823137899835,
      "loss": 5.2106,
      "mean_token_accuracy": 0.2192715510725975,
      "num_tokens": 148773773.0,
      "step": 73420
    },
    {
      "entropy": 5.907947444915772,
      "epoch": 6.416586559468671,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018012997479474352,
      "loss": 5.1963,
      "mean_token_accuracy": 0.2188812792301178,
      "num_tokens": 148783245.0,
      "step": 73425
    },
    {
      "entropy": 5.828415966033935,
      "epoch": 6.4170235078213755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001801017200301456,
      "loss": 5.1088,
      "mean_token_accuracy": 0.22478103935718535,
      "num_tokens": 148793066.0,
      "step": 73430
    },
    {
      "entropy": 5.867594242095947,
      "epoch": 6.41746045617408,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001800734670857463,
      "loss": 5.2047,
      "mean_token_accuracy": 0.22018641829490662,
      "num_tokens": 148803265.0,
      "step": 73435
    },
    {
      "entropy": 5.867718172073364,
      "epoch": 6.417897404526785,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018004521596208766,
      "loss": 5.1559,
      "mean_token_accuracy": 0.22182943820953369,
      "num_tokens": 148813060.0,
      "step": 73440
    },
    {
      "entropy": 5.888051128387451,
      "epoch": 6.41833435287949,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018001696665971152,
      "loss": 5.2291,
      "mean_token_accuracy": 0.21918474435806273,
      "num_tokens": 148823553.0,
      "step": 73445
    },
    {
      "entropy": 5.79963002204895,
      "epoch": 6.418771301232194,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001799887191791597,
      "loss": 5.0515,
      "mean_token_accuracy": 0.23002394139766694,
      "num_tokens": 148832753.0,
      "step": 73450
    },
    {
      "entropy": 5.858261775970459,
      "epoch": 6.419208249584899,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017996047352097406,
      "loss": 5.1382,
      "mean_token_accuracy": 0.22740083634853364,
      "num_tokens": 148842236.0,
      "step": 73455
    },
    {
      "entropy": 5.733649444580078,
      "epoch": 6.419645197937604,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001799322296856964,
      "loss": 5.0242,
      "mean_token_accuracy": 0.24116024374961853,
      "num_tokens": 148852540.0,
      "step": 73460
    },
    {
      "entropy": 5.841505002975464,
      "epoch": 6.420082146290309,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017990398767386828,
      "loss": 5.1774,
      "mean_token_accuracy": 0.22464356422424317,
      "num_tokens": 148863009.0,
      "step": 73465
    },
    {
      "entropy": 5.820804452896118,
      "epoch": 6.420519094643013,
      "grad_norm": 1.125,
      "learning_rate": 0.00017987574748603154,
      "loss": 5.0962,
      "mean_token_accuracy": 0.23470994532108308,
      "num_tokens": 148873617.0,
      "step": 73470
    },
    {
      "entropy": 5.855910205841065,
      "epoch": 6.4209560429957175,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001798475091227279,
      "loss": 5.2119,
      "mean_token_accuracy": 0.218413844704628,
      "num_tokens": 148883233.0,
      "step": 73475
    },
    {
      "entropy": 5.801512956619263,
      "epoch": 6.421392991348423,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017981927258449882,
      "loss": 5.0259,
      "mean_token_accuracy": 0.2446288973093033,
      "num_tokens": 148893071.0,
      "step": 73480
    },
    {
      "entropy": 5.855356597900391,
      "epoch": 6.421829939701127,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000179791037871886,
      "loss": 5.099,
      "mean_token_accuracy": 0.22827094197273254,
      "num_tokens": 148902589.0,
      "step": 73485
    },
    {
      "entropy": 5.838666009902954,
      "epoch": 6.422266888053832,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001797628049854309,
      "loss": 5.0904,
      "mean_token_accuracy": 0.2286616012454033,
      "num_tokens": 148913109.0,
      "step": 73490
    },
    {
      "entropy": 5.823259735107422,
      "epoch": 6.422703836406537,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017973457392567516,
      "loss": 5.0764,
      "mean_token_accuracy": 0.23418951779603958,
      "num_tokens": 148922331.0,
      "step": 73495
    },
    {
      "entropy": 5.847050476074219,
      "epoch": 6.423140784759242,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017970634469316015,
      "loss": 5.196,
      "mean_token_accuracy": 0.2237717479467392,
      "num_tokens": 148932668.0,
      "step": 73500
    },
    {
      "entropy": 5.828180170059204,
      "epoch": 6.423577733111946,
      "grad_norm": 1.125,
      "learning_rate": 0.00017967811728842743,
      "loss": 5.1616,
      "mean_token_accuracy": 0.23077153712511062,
      "num_tokens": 148944423.0,
      "step": 73505
    },
    {
      "entropy": 5.922770214080811,
      "epoch": 6.424014681464651,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017964989171201834,
      "loss": 5.2051,
      "mean_token_accuracy": 0.2169150784611702,
      "num_tokens": 148954094.0,
      "step": 73510
    },
    {
      "entropy": 5.857075357437134,
      "epoch": 6.424451629817356,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017962166796447432,
      "loss": 5.1166,
      "mean_token_accuracy": 0.22925726175308228,
      "num_tokens": 148963018.0,
      "step": 73515
    },
    {
      "entropy": 5.837707757949829,
      "epoch": 6.42488857817006,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017959344604633664,
      "loss": 5.1357,
      "mean_token_accuracy": 0.2236726924777031,
      "num_tokens": 148972981.0,
      "step": 73520
    },
    {
      "entropy": 5.826418542861939,
      "epoch": 6.425325526522765,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017956522595814677,
      "loss": 5.1012,
      "mean_token_accuracy": 0.2310012772679329,
      "num_tokens": 148983926.0,
      "step": 73525
    },
    {
      "entropy": 5.9059216499328615,
      "epoch": 6.425762474875469,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001795370077004458,
      "loss": 5.1398,
      "mean_token_accuracy": 0.22632641047239305,
      "num_tokens": 148994376.0,
      "step": 73530
    },
    {
      "entropy": 5.829147291183472,
      "epoch": 6.426199423228175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000179508791273775,
      "loss": 5.1299,
      "mean_token_accuracy": 0.2301773026585579,
      "num_tokens": 149005195.0,
      "step": 73535
    },
    {
      "entropy": 5.875146579742432,
      "epoch": 6.426636371580879,
      "grad_norm": 1.25,
      "learning_rate": 0.0001794805766786757,
      "loss": 5.1902,
      "mean_token_accuracy": 0.2195151314139366,
      "num_tokens": 149014839.0,
      "step": 73540
    },
    {
      "entropy": 5.761015701293945,
      "epoch": 6.427073319933584,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017945236391568887,
      "loss": 5.0762,
      "mean_token_accuracy": 0.2295023426413536,
      "num_tokens": 149026555.0,
      "step": 73545
    },
    {
      "entropy": 5.799423599243164,
      "epoch": 6.427510268286289,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00017942415298535585,
      "loss": 5.1147,
      "mean_token_accuracy": 0.2267448455095291,
      "num_tokens": 149036598.0,
      "step": 73550
    },
    {
      "entropy": 5.868989706039429,
      "epoch": 6.4279472166389935,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001793959438882176,
      "loss": 5.144,
      "mean_token_accuracy": 0.22349563241004944,
      "num_tokens": 149046979.0,
      "step": 73555
    },
    {
      "entropy": 5.8180543899536135,
      "epoch": 6.428384164991698,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017936773662481536,
      "loss": 5.1007,
      "mean_token_accuracy": 0.23217946588993071,
      "num_tokens": 149056363.0,
      "step": 73560
    },
    {
      "entropy": 5.82881178855896,
      "epoch": 6.428821113344402,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017933953119568996,
      "loss": 5.1741,
      "mean_token_accuracy": 0.2180252492427826,
      "num_tokens": 149066605.0,
      "step": 73565
    },
    {
      "entropy": 5.813715934753418,
      "epoch": 6.429258061697108,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017931132760138247,
      "loss": 5.0907,
      "mean_token_accuracy": 0.2303234875202179,
      "num_tokens": 149076912.0,
      "step": 73570
    },
    {
      "entropy": 5.846907711029052,
      "epoch": 6.429695010049812,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017928312584243394,
      "loss": 5.1442,
      "mean_token_accuracy": 0.21734615415334702,
      "num_tokens": 149086573.0,
      "step": 73575
    },
    {
      "entropy": 5.876003837585449,
      "epoch": 6.430131958402517,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017925492591938514,
      "loss": 5.1993,
      "mean_token_accuracy": 0.22132749408483504,
      "num_tokens": 149096800.0,
      "step": 73580
    },
    {
      "entropy": 5.834008741378784,
      "epoch": 6.430568906755221,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017922672783277706,
      "loss": 5.2006,
      "mean_token_accuracy": 0.2195910930633545,
      "num_tokens": 149107660.0,
      "step": 73585
    },
    {
      "entropy": 5.845794916152954,
      "epoch": 6.4310058551079265,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017919853158315048,
      "loss": 5.0944,
      "mean_token_accuracy": 0.22880033999681473,
      "num_tokens": 149117531.0,
      "step": 73590
    },
    {
      "entropy": 5.889739894866944,
      "epoch": 6.431442803460631,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001791703371710463,
      "loss": 5.2147,
      "mean_token_accuracy": 0.21811861395835877,
      "num_tokens": 149128027.0,
      "step": 73595
    },
    {
      "entropy": 5.907227277755737,
      "epoch": 6.4318797518133355,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017914214459700524,
      "loss": 5.181,
      "mean_token_accuracy": 0.22135486155748368,
      "num_tokens": 149137148.0,
      "step": 73600
    },
    {
      "entropy": 5.92799391746521,
      "epoch": 6.43231670016604,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017911395386156808,
      "loss": 5.1904,
      "mean_token_accuracy": 0.21938680708408356,
      "num_tokens": 149147741.0,
      "step": 73605
    },
    {
      "entropy": 5.882166719436645,
      "epoch": 6.432753648518745,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001790857649652755,
      "loss": 5.1125,
      "mean_token_accuracy": 0.22481079697608947,
      "num_tokens": 149156924.0,
      "step": 73610
    },
    {
      "entropy": 5.872891759872436,
      "epoch": 6.43319059687145,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017905757790866823,
      "loss": 5.1263,
      "mean_token_accuracy": 0.22406568974256516,
      "num_tokens": 149166839.0,
      "step": 73615
    },
    {
      "entropy": 5.82074122428894,
      "epoch": 6.433627545224154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001790293926922869,
      "loss": 5.125,
      "mean_token_accuracy": 0.22519339472055436,
      "num_tokens": 149176607.0,
      "step": 73620
    },
    {
      "entropy": 5.88130259513855,
      "epoch": 6.43406449357686,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017900120931667218,
      "loss": 5.2309,
      "mean_token_accuracy": 0.2154600664973259,
      "num_tokens": 149186183.0,
      "step": 73625
    },
    {
      "entropy": 5.843174982070923,
      "epoch": 6.434501441929564,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001789730277823645,
      "loss": 5.0898,
      "mean_token_accuracy": 0.2287726655602455,
      "num_tokens": 149195769.0,
      "step": 73630
    },
    {
      "entropy": 5.8598250389099125,
      "epoch": 6.4349383902822685,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00017894484808990446,
      "loss": 5.1504,
      "mean_token_accuracy": 0.2217734545469284,
      "num_tokens": 149206045.0,
      "step": 73635
    },
    {
      "entropy": 5.826992082595825,
      "epoch": 6.435375338634973,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001789166702398326,
      "loss": 5.15,
      "mean_token_accuracy": 0.22874991446733475,
      "num_tokens": 149216825.0,
      "step": 73640
    },
    {
      "entropy": 5.832134675979614,
      "epoch": 6.435812286987678,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00017888849423268926,
      "loss": 5.1505,
      "mean_token_accuracy": 0.22162612527608871,
      "num_tokens": 149226928.0,
      "step": 73645
    },
    {
      "entropy": 5.876682281494141,
      "epoch": 6.436249235340383,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017886032006901503,
      "loss": 5.2059,
      "mean_token_accuracy": 0.21567688882350922,
      "num_tokens": 149236538.0,
      "step": 73650
    },
    {
      "entropy": 5.862291383743286,
      "epoch": 6.436686183693087,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0001788321477493502,
      "loss": 5.1885,
      "mean_token_accuracy": 0.2144342765212059,
      "num_tokens": 149247436.0,
      "step": 73655
    },
    {
      "entropy": 5.883228349685669,
      "epoch": 6.437123132045792,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001788039772742352,
      "loss": 5.19,
      "mean_token_accuracy": 0.21961985826492308,
      "num_tokens": 149257832.0,
      "step": 73660
    },
    {
      "entropy": 5.8593652725219725,
      "epoch": 6.437560080398497,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00017877580864421034,
      "loss": 5.1216,
      "mean_token_accuracy": 0.23033162355422973,
      "num_tokens": 149268266.0,
      "step": 73665
    },
    {
      "entropy": 5.894229078292847,
      "epoch": 6.437997028751202,
      "grad_norm": 1.125,
      "learning_rate": 0.00017874764185981596,
      "loss": 5.2428,
      "mean_token_accuracy": 0.21563371270895004,
      "num_tokens": 149278755.0,
      "step": 73670
    },
    {
      "entropy": 5.791153287887573,
      "epoch": 6.438433977103906,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017871947692159212,
      "loss": 5.1003,
      "mean_token_accuracy": 0.223319873213768,
      "num_tokens": 149288683.0,
      "step": 73675
    },
    {
      "entropy": 5.869689655303955,
      "epoch": 6.438870925456611,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001786913138300792,
      "loss": 5.1108,
      "mean_token_accuracy": 0.22634680271148683,
      "num_tokens": 149297603.0,
      "step": 73680
    },
    {
      "entropy": 5.861109447479248,
      "epoch": 6.439307873809316,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00017866315258581735,
      "loss": 5.1914,
      "mean_token_accuracy": 0.2201964631676674,
      "num_tokens": 149307557.0,
      "step": 73685
    },
    {
      "entropy": 5.860128879547119,
      "epoch": 6.43974482216202,
      "grad_norm": 1.0625,
      "learning_rate": 0.00017863499318934668,
      "loss": 5.1285,
      "mean_token_accuracy": 0.22616130262613296,
      "num_tokens": 149317649.0,
      "step": 73690
    },
    {
      "entropy": 5.82887487411499,
      "epoch": 6.440181770514725,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017860683564120734,
      "loss": 5.1684,
      "mean_token_accuracy": 0.22894023060798646,
      "num_tokens": 149328437.0,
      "step": 73695
    },
    {
      "entropy": 5.818018198013306,
      "epoch": 6.44061871886743,
      "grad_norm": 1.125,
      "learning_rate": 0.00017857867994193932,
      "loss": 5.1624,
      "mean_token_accuracy": 0.22561149746179582,
      "num_tokens": 149339014.0,
      "step": 73700
    },
    {
      "entropy": 5.851262426376342,
      "epoch": 6.441055667220135,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017855052609208282,
      "loss": 5.1361,
      "mean_token_accuracy": 0.2240991324186325,
      "num_tokens": 149349322.0,
      "step": 73705
    },
    {
      "entropy": 5.8650308609008786,
      "epoch": 6.441492615572839,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001785223740921779,
      "loss": 5.1696,
      "mean_token_accuracy": 0.2226854905486107,
      "num_tokens": 149359629.0,
      "step": 73710
    },
    {
      "entropy": 5.8884576797485355,
      "epoch": 6.441929563925544,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017849422394276428,
      "loss": 5.0937,
      "mean_token_accuracy": 0.22730125188827516,
      "num_tokens": 149369910.0,
      "step": 73715
    },
    {
      "entropy": 5.86160626411438,
      "epoch": 6.442366512278249,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000178466075644382,
      "loss": 5.1538,
      "mean_token_accuracy": 0.2227586939930916,
      "num_tokens": 149380197.0,
      "step": 73720
    },
    {
      "entropy": 5.817627048492431,
      "epoch": 6.442803460630953,
      "grad_norm": 1.25,
      "learning_rate": 0.000178437929197571,
      "loss": 5.1559,
      "mean_token_accuracy": 0.21795631051063538,
      "num_tokens": 149391230.0,
      "step": 73725
    },
    {
      "entropy": 5.811984348297119,
      "epoch": 6.443240408983658,
      "grad_norm": 1.25,
      "learning_rate": 0.0001784097846028711,
      "loss": 5.1454,
      "mean_token_accuracy": 0.2251102551817894,
      "num_tokens": 149401625.0,
      "step": 73730
    },
    {
      "entropy": 5.9078125,
      "epoch": 6.443677357336362,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017838164186082211,
      "loss": 5.2074,
      "mean_token_accuracy": 0.217262901365757,
      "num_tokens": 149411515.0,
      "step": 73735
    },
    {
      "entropy": 5.862288141250611,
      "epoch": 6.444114305689068,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001783535009719639,
      "loss": 5.1414,
      "mean_token_accuracy": 0.22551744431257248,
      "num_tokens": 149422269.0,
      "step": 73740
    },
    {
      "entropy": 5.874425554275513,
      "epoch": 6.444551254041772,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017832536193683617,
      "loss": 5.1544,
      "mean_token_accuracy": 0.21843355149030685,
      "num_tokens": 149432596.0,
      "step": 73745
    },
    {
      "entropy": 5.870477914810181,
      "epoch": 6.444988202394477,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001782972247559787,
      "loss": 5.0825,
      "mean_token_accuracy": 0.22909379601478577,
      "num_tokens": 149442051.0,
      "step": 73750
    },
    {
      "entropy": 5.845043897628784,
      "epoch": 6.445425150747182,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001782690894299311,
      "loss": 5.1399,
      "mean_token_accuracy": 0.2251943126320839,
      "num_tokens": 149451151.0,
      "step": 73755
    },
    {
      "entropy": 5.913668584823609,
      "epoch": 6.4458620990998865,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001782409559592331,
      "loss": 5.1742,
      "mean_token_accuracy": 0.22329632043838502,
      "num_tokens": 149461324.0,
      "step": 73760
    },
    {
      "entropy": 5.803378677368164,
      "epoch": 6.446299047452591,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001782128243444242,
      "loss": 5.1319,
      "mean_token_accuracy": 0.22488496601581573,
      "num_tokens": 149471413.0,
      "step": 73765
    },
    {
      "entropy": 5.884055519104004,
      "epoch": 6.446735995805295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017818469458604408,
      "loss": 5.2343,
      "mean_token_accuracy": 0.21854975074529648,
      "num_tokens": 149481164.0,
      "step": 73770
    },
    {
      "entropy": 5.879756784439087,
      "epoch": 6.447172944158001,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001781565666846322,
      "loss": 5.1428,
      "mean_token_accuracy": 0.2212679371237755,
      "num_tokens": 149491782.0,
      "step": 73775
    },
    {
      "entropy": 5.859331274032593,
      "epoch": 6.447609892510705,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017812844064072814,
      "loss": 5.1288,
      "mean_token_accuracy": 0.22496158480644227,
      "num_tokens": 149502118.0,
      "step": 73780
    },
    {
      "entropy": 5.875586080551147,
      "epoch": 6.44804684086341,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017810031645487136,
      "loss": 5.1926,
      "mean_token_accuracy": 0.22185714840888976,
      "num_tokens": 149511750.0,
      "step": 73785
    },
    {
      "entropy": 5.877994155883789,
      "epoch": 6.448483789216114,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001780721941276013,
      "loss": 5.0892,
      "mean_token_accuracy": 0.2290887340903282,
      "num_tokens": 149522342.0,
      "step": 73790
    },
    {
      "entropy": 5.797343015670776,
      "epoch": 6.44892073756882,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017804407365945728,
      "loss": 5.0856,
      "mean_token_accuracy": 0.23517339378595353,
      "num_tokens": 149532706.0,
      "step": 73795
    },
    {
      "entropy": 5.821302509307861,
      "epoch": 6.449357685921524,
      "grad_norm": 1.25,
      "learning_rate": 0.0001780159550509788,
      "loss": 5.1125,
      "mean_token_accuracy": 0.22385032773017882,
      "num_tokens": 149543076.0,
      "step": 73800
    },
    {
      "entropy": 5.855349349975586,
      "epoch": 6.4497946342742285,
      "grad_norm": 1.125,
      "learning_rate": 0.00017798783830270514,
      "loss": 5.0877,
      "mean_token_accuracy": 0.2279768243432045,
      "num_tokens": 149552902.0,
      "step": 73805
    },
    {
      "entropy": 5.8278497695922855,
      "epoch": 6.450231582626934,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017795972341517552,
      "loss": 5.0759,
      "mean_token_accuracy": 0.23185406774282455,
      "num_tokens": 149562286.0,
      "step": 73810
    },
    {
      "entropy": 5.8146891593933105,
      "epoch": 6.450668530979638,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001779316103889293,
      "loss": 5.1287,
      "mean_token_accuracy": 0.22860518246889114,
      "num_tokens": 149572837.0,
      "step": 73815
    },
    {
      "entropy": 5.928902816772461,
      "epoch": 6.451105479332343,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001779034992245056,
      "loss": 5.239,
      "mean_token_accuracy": 0.21931674033403398,
      "num_tokens": 149582321.0,
      "step": 73820
    },
    {
      "entropy": 5.826372909545898,
      "epoch": 6.451542427685047,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017787538992244367,
      "loss": 5.1015,
      "mean_token_accuracy": 0.23009946644306184,
      "num_tokens": 149592977.0,
      "step": 73825
    },
    {
      "entropy": 5.820595359802246,
      "epoch": 6.451979376037753,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00017784728248328264,
      "loss": 5.121,
      "mean_token_accuracy": 0.22575607895851135,
      "num_tokens": 149604183.0,
      "step": 73830
    },
    {
      "entropy": 5.834353733062744,
      "epoch": 6.452416324390457,
      "grad_norm": 1.25,
      "learning_rate": 0.00017781917690756172,
      "loss": 5.0897,
      "mean_token_accuracy": 0.23401864916086196,
      "num_tokens": 149614868.0,
      "step": 73835
    },
    {
      "entropy": 5.867247867584228,
      "epoch": 6.452853272743162,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017779107319581983,
      "loss": 5.1525,
      "mean_token_accuracy": 0.22386726289987563,
      "num_tokens": 149626540.0,
      "step": 73840
    },
    {
      "entropy": 5.830797719955444,
      "epoch": 6.453290221095866,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001777629713485961,
      "loss": 5.1984,
      "mean_token_accuracy": 0.2226854920387268,
      "num_tokens": 149637072.0,
      "step": 73845
    },
    {
      "entropy": 5.833781623840332,
      "epoch": 6.453727169448571,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001777348713664295,
      "loss": 5.1228,
      "mean_token_accuracy": 0.22842806875705718,
      "num_tokens": 149646916.0,
      "step": 73850
    },
    {
      "entropy": 5.852149581909179,
      "epoch": 6.454164117801276,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017770677324985919,
      "loss": 5.1704,
      "mean_token_accuracy": 0.22533014118671418,
      "num_tokens": 149657616.0,
      "step": 73855
    },
    {
      "entropy": 5.886824893951416,
      "epoch": 6.45460106615398,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017767867699942385,
      "loss": 5.1498,
      "mean_token_accuracy": 0.2260204941034317,
      "num_tokens": 149668018.0,
      "step": 73860
    },
    {
      "entropy": 5.85390272140503,
      "epoch": 6.455038014506686,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017765058261566248,
      "loss": 5.0972,
      "mean_token_accuracy": 0.22689555287361146,
      "num_tokens": 149677581.0,
      "step": 73865
    },
    {
      "entropy": 5.76492280960083,
      "epoch": 6.45547496285939,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001776224900991139,
      "loss": 5.0939,
      "mean_token_accuracy": 0.22598545253276825,
      "num_tokens": 149687547.0,
      "step": 73870
    },
    {
      "entropy": 5.797862243652344,
      "epoch": 6.455911911212095,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017759439945031703,
      "loss": 5.0952,
      "mean_token_accuracy": 0.22953420430421828,
      "num_tokens": 149698339.0,
      "step": 73875
    },
    {
      "entropy": 5.923425149917603,
      "epoch": 6.456348859564799,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001775663106698106,
      "loss": 5.1549,
      "mean_token_accuracy": 0.22147647887468339,
      "num_tokens": 149707455.0,
      "step": 73880
    },
    {
      "entropy": 5.869689702987671,
      "epoch": 6.4567858079175044,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001775382237581334,
      "loss": 5.2526,
      "mean_token_accuracy": 0.21460921168327332,
      "num_tokens": 149718130.0,
      "step": 73885
    },
    {
      "entropy": 5.873721027374268,
      "epoch": 6.457222756270209,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017751013871582412,
      "loss": 5.1848,
      "mean_token_accuracy": 0.22698510587215423,
      "num_tokens": 149727747.0,
      "step": 73890
    },
    {
      "entropy": 5.91201000213623,
      "epoch": 6.457659704622913,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017748205554342144,
      "loss": 5.1847,
      "mean_token_accuracy": 0.21793082356452942,
      "num_tokens": 149737371.0,
      "step": 73895
    },
    {
      "entropy": 5.975519466400146,
      "epoch": 6.458096652975618,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017745397424146408,
      "loss": 5.2405,
      "mean_token_accuracy": 0.2178151398897171,
      "num_tokens": 149747983.0,
      "step": 73900
    },
    {
      "entropy": 5.878432703018189,
      "epoch": 6.458533601328323,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001774258948104907,
      "loss": 5.1502,
      "mean_token_accuracy": 0.22971100062131883,
      "num_tokens": 149758175.0,
      "step": 73905
    },
    {
      "entropy": 5.825142908096313,
      "epoch": 6.458970549681028,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017739781725103964,
      "loss": 5.1124,
      "mean_token_accuracy": 0.23557743281126023,
      "num_tokens": 149767751.0,
      "step": 73910
    },
    {
      "entropy": 5.819654083251953,
      "epoch": 6.459407498033732,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00017736974156364965,
      "loss": 5.0756,
      "mean_token_accuracy": 0.23252890706062318,
      "num_tokens": 149777937.0,
      "step": 73915
    },
    {
      "entropy": 5.78355712890625,
      "epoch": 6.4598444463864375,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017734166774885919,
      "loss": 5.0633,
      "mean_token_accuracy": 0.23361152559518814,
      "num_tokens": 149787364.0,
      "step": 73920
    },
    {
      "entropy": 5.819217872619629,
      "epoch": 6.460281394739142,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017731359580720662,
      "loss": 5.1474,
      "mean_token_accuracy": 0.22433826923370362,
      "num_tokens": 149796333.0,
      "step": 73925
    },
    {
      "entropy": 5.8009966850280765,
      "epoch": 6.4607183430918464,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017728552573923056,
      "loss": 5.1587,
      "mean_token_accuracy": 0.2259762853384018,
      "num_tokens": 149806449.0,
      "step": 73930
    },
    {
      "entropy": 5.871656656265259,
      "epoch": 6.461155291444551,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017725745754546929,
      "loss": 5.0999,
      "mean_token_accuracy": 0.23023399710655212,
      "num_tokens": 149817909.0,
      "step": 73935
    },
    {
      "entropy": 5.896099424362182,
      "epoch": 6.461592239797256,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017722939122646113,
      "loss": 5.1797,
      "mean_token_accuracy": 0.21796709001064302,
      "num_tokens": 149827960.0,
      "step": 73940
    },
    {
      "entropy": 5.868229675292969,
      "epoch": 6.462029188149961,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001772013267827446,
      "loss": 5.1719,
      "mean_token_accuracy": 0.21696112751960756,
      "num_tokens": 149838004.0,
      "step": 73945
    },
    {
      "entropy": 5.885740900039673,
      "epoch": 6.462466136502665,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017717326421485784,
      "loss": 5.1485,
      "mean_token_accuracy": 0.2252076491713524,
      "num_tokens": 149847031.0,
      "step": 73950
    },
    {
      "entropy": 5.864166498184204,
      "epoch": 6.46290308485537,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017714520352333913,
      "loss": 5.1434,
      "mean_token_accuracy": 0.22439437061548234,
      "num_tokens": 149857538.0,
      "step": 73955
    },
    {
      "entropy": 5.826357889175415,
      "epoch": 6.463340033208075,
      "grad_norm": 1.125,
      "learning_rate": 0.0001771171447087267,
      "loss": 5.0909,
      "mean_token_accuracy": 0.2265279099345207,
      "num_tokens": 149867499.0,
      "step": 73960
    },
    {
      "entropy": 5.848671340942383,
      "epoch": 6.4637769815607795,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00017708908777155864,
      "loss": 5.0839,
      "mean_token_accuracy": 0.22673123180866242,
      "num_tokens": 149878008.0,
      "step": 73965
    },
    {
      "entropy": 5.870366525650025,
      "epoch": 6.464213929913484,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017706103271237323,
      "loss": 5.1668,
      "mean_token_accuracy": 0.2277296483516693,
      "num_tokens": 149889592.0,
      "step": 73970
    },
    {
      "entropy": 5.8731465339660645,
      "epoch": 6.464650878266189,
      "grad_norm": 1.03125,
      "learning_rate": 0.00017703297953170855,
      "loss": 5.2133,
      "mean_token_accuracy": 0.21848511695861816,
      "num_tokens": 149899634.0,
      "step": 73975
    },
    {
      "entropy": 5.857463264465332,
      "epoch": 6.465087826618894,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017700492823010262,
      "loss": 5.0631,
      "mean_token_accuracy": 0.23579303920269012,
      "num_tokens": 149909466.0,
      "step": 73980
    },
    {
      "entropy": 5.907362699508667,
      "epoch": 6.465524774971598,
      "grad_norm": 1.015625,
      "learning_rate": 0.00017697687880809354,
      "loss": 5.231,
      "mean_token_accuracy": 0.22116897106170655,
      "num_tokens": 149920805.0,
      "step": 73985
    },
    {
      "entropy": 5.832863140106201,
      "epoch": 6.465961723324303,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001769488312662193,
      "loss": 5.1579,
      "mean_token_accuracy": 0.22477604895830156,
      "num_tokens": 149931938.0,
      "step": 73990
    },
    {
      "entropy": 5.847085428237915,
      "epoch": 6.466398671677008,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017692078560501784,
      "loss": 5.1555,
      "mean_token_accuracy": 0.22469662129878998,
      "num_tokens": 149941568.0,
      "step": 73995
    },
    {
      "entropy": 5.881427335739136,
      "epoch": 6.466835620029713,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001768927418250272,
      "loss": 5.1529,
      "mean_token_accuracy": 0.2277890995144844,
      "num_tokens": 149951232.0,
      "step": 74000
    },
    {
      "entropy": 5.846122217178345,
      "epoch": 6.467272568382417,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017686469992678506,
      "loss": 5.0483,
      "mean_token_accuracy": 0.2288212448358536,
      "num_tokens": 149961771.0,
      "step": 74005
    },
    {
      "entropy": 5.820428085327149,
      "epoch": 6.4677095167351215,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017683665991082936,
      "loss": 5.1562,
      "mean_token_accuracy": 0.22384067624807358,
      "num_tokens": 149971112.0,
      "step": 74010
    },
    {
      "entropy": 5.8918290615081785,
      "epoch": 6.468146465087827,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017680862177769807,
      "loss": 5.259,
      "mean_token_accuracy": 0.2115975245833397,
      "num_tokens": 149980676.0,
      "step": 74015
    },
    {
      "entropy": 5.806964540481568,
      "epoch": 6.468583413440531,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001767805855279288,
      "loss": 5.081,
      "mean_token_accuracy": 0.2265932634472847,
      "num_tokens": 149990319.0,
      "step": 74020
    },
    {
      "entropy": 5.939627313613892,
      "epoch": 6.469020361793236,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017675255116205934,
      "loss": 5.2198,
      "mean_token_accuracy": 0.22018791139125823,
      "num_tokens": 150000983.0,
      "step": 74025
    },
    {
      "entropy": 5.8339276790618895,
      "epoch": 6.469457310145941,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001767245186806275,
      "loss": 5.0991,
      "mean_token_accuracy": 0.22591778337955476,
      "num_tokens": 150010397.0,
      "step": 74030
    },
    {
      "entropy": 5.846784210205078,
      "epoch": 6.469894258498646,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00017669648808417082,
      "loss": 5.1274,
      "mean_token_accuracy": 0.22465297132730483,
      "num_tokens": 150020225.0,
      "step": 74035
    },
    {
      "entropy": 5.814041090011597,
      "epoch": 6.47033120685135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017666845937322706,
      "loss": 5.1192,
      "mean_token_accuracy": 0.22791969925165176,
      "num_tokens": 150030298.0,
      "step": 74040
    },
    {
      "entropy": 5.838089227676392,
      "epoch": 6.470768155204055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00017664043254833378,
      "loss": 5.1012,
      "mean_token_accuracy": 0.23199611902236938,
      "num_tokens": 150040362.0,
      "step": 74045
    },
    {
      "entropy": 5.861462116241455,
      "epoch": 6.47120510355676,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017661240761002856,
      "loss": 5.1439,
      "mean_token_accuracy": 0.22660572677850724,
      "num_tokens": 150050030.0,
      "step": 74050
    },
    {
      "entropy": 5.8893355369567875,
      "epoch": 6.471642051909464,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017658438455884885,
      "loss": 5.0984,
      "mean_token_accuracy": 0.2236149564385414,
      "num_tokens": 150060114.0,
      "step": 74055
    },
    {
      "entropy": 5.893551588058472,
      "epoch": 6.472079000262169,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001765563633953322,
      "loss": 5.1617,
      "mean_token_accuracy": 0.2234485626220703,
      "num_tokens": 150069621.0,
      "step": 74060
    },
    {
      "entropy": 5.8262022018432615,
      "epoch": 6.472515948614873,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001765283441200161,
      "loss": 5.1131,
      "mean_token_accuracy": 0.22816023379564285,
      "num_tokens": 150080392.0,
      "step": 74065
    },
    {
      "entropy": 5.786530542373657,
      "epoch": 6.472952896967579,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017650032673343803,
      "loss": 5.0325,
      "mean_token_accuracy": 0.2359772726893425,
      "num_tokens": 150089701.0,
      "step": 74070
    },
    {
      "entropy": 5.878902769088745,
      "epoch": 6.473389845320283,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017647231123613523,
      "loss": 5.1198,
      "mean_token_accuracy": 0.2283893868327141,
      "num_tokens": 150098805.0,
      "step": 74075
    },
    {
      "entropy": 5.839437580108642,
      "epoch": 6.473826793672988,
      "grad_norm": 1.25,
      "learning_rate": 0.00017644429762864517,
      "loss": 5.0947,
      "mean_token_accuracy": 0.22511865794658661,
      "num_tokens": 150109009.0,
      "step": 74080
    },
    {
      "entropy": 5.7704626560211185,
      "epoch": 6.474263742025693,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017641628591150515,
      "loss": 5.1021,
      "mean_token_accuracy": 0.23656315207481385,
      "num_tokens": 150119153.0,
      "step": 74085
    },
    {
      "entropy": 5.819806528091431,
      "epoch": 6.4747006903783975,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017638827608525243,
      "loss": 5.0749,
      "mean_token_accuracy": 0.22706858962774276,
      "num_tokens": 150131790.0,
      "step": 74090
    },
    {
      "entropy": 5.836953783035279,
      "epoch": 6.475137638731102,
      "grad_norm": 1.25,
      "learning_rate": 0.00017636026815042417,
      "loss": 5.0985,
      "mean_token_accuracy": 0.2213662564754486,
      "num_tokens": 150141146.0,
      "step": 74095
    },
    {
      "entropy": 5.969625473022461,
      "epoch": 6.475574587083806,
      "grad_norm": 1.25,
      "learning_rate": 0.0001763322621075577,
      "loss": 5.2196,
      "mean_token_accuracy": 0.22049127668142318,
      "num_tokens": 150151677.0,
      "step": 74100
    },
    {
      "entropy": 5.853669357299805,
      "epoch": 6.476011535436512,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017630425795719013,
      "loss": 5.065,
      "mean_token_accuracy": 0.22650136053562164,
      "num_tokens": 150161739.0,
      "step": 74105
    },
    {
      "entropy": 5.817092990875244,
      "epoch": 6.476448483789216,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017627625569985856,
      "loss": 5.1472,
      "mean_token_accuracy": 0.22759625166654587,
      "num_tokens": 150171540.0,
      "step": 74110
    },
    {
      "entropy": 5.809889221191407,
      "epoch": 6.476885432141921,
      "grad_norm": 1.25,
      "learning_rate": 0.00017624825533610022,
      "loss": 5.0984,
      "mean_token_accuracy": 0.22959696501493454,
      "num_tokens": 150181924.0,
      "step": 74115
    },
    {
      "entropy": 5.883905649185181,
      "epoch": 6.477322380494625,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017622025686645204,
      "loss": 5.1947,
      "mean_token_accuracy": 0.2193683847784996,
      "num_tokens": 150192871.0,
      "step": 74120
    },
    {
      "entropy": 5.915045118331909,
      "epoch": 6.4777593288473305,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017619226029145108,
      "loss": 5.1946,
      "mean_token_accuracy": 0.2140173152089119,
      "num_tokens": 150202434.0,
      "step": 74125
    },
    {
      "entropy": 5.819415092468262,
      "epoch": 6.478196277200035,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001761642656116344,
      "loss": 5.033,
      "mean_token_accuracy": 0.22921958565711975,
      "num_tokens": 150212283.0,
      "step": 74130
    },
    {
      "entropy": 5.778509569168091,
      "epoch": 6.4786332255527395,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001761362728275389,
      "loss": 5.0462,
      "mean_token_accuracy": 0.23385699093341827,
      "num_tokens": 150222066.0,
      "step": 74135
    },
    {
      "entropy": 5.8230424404144285,
      "epoch": 6.479070173905445,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017610828193970147,
      "loss": 5.1031,
      "mean_token_accuracy": 0.22665468156337737,
      "num_tokens": 150231786.0,
      "step": 74140
    },
    {
      "entropy": 5.8533244132995605,
      "epoch": 6.479507122258149,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017608029294865896,
      "loss": 5.1469,
      "mean_token_accuracy": 0.2237139657139778,
      "num_tokens": 150242727.0,
      "step": 74145
    },
    {
      "entropy": 5.8920145511627195,
      "epoch": 6.479944070610854,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017605230585494836,
      "loss": 5.1693,
      "mean_token_accuracy": 0.2223154753446579,
      "num_tokens": 150252565.0,
      "step": 74150
    },
    {
      "entropy": 5.877965259552002,
      "epoch": 6.480381018963558,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017602432065910628,
      "loss": 5.143,
      "mean_token_accuracy": 0.2217748060822487,
      "num_tokens": 150263314.0,
      "step": 74155
    },
    {
      "entropy": 5.791022109985351,
      "epoch": 6.480817967316264,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017599633736166965,
      "loss": 5.0987,
      "mean_token_accuracy": 0.2340220257639885,
      "num_tokens": 150273381.0,
      "step": 74160
    },
    {
      "entropy": 5.8800629615783695,
      "epoch": 6.481254915668968,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017596835596317512,
      "loss": 5.1814,
      "mean_token_accuracy": 0.21330225169658662,
      "num_tokens": 150282724.0,
      "step": 74165
    },
    {
      "entropy": 5.843209886550904,
      "epoch": 6.4816918640216725,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017594037646415938,
      "loss": 5.1464,
      "mean_token_accuracy": 0.2169596180319786,
      "num_tokens": 150292873.0,
      "step": 74170
    },
    {
      "entropy": 5.85399751663208,
      "epoch": 6.482128812374377,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017591239886515918,
      "loss": 5.1892,
      "mean_token_accuracy": 0.2268635854125023,
      "num_tokens": 150303392.0,
      "step": 74175
    },
    {
      "entropy": 5.8007830619812015,
      "epoch": 6.482565760727082,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001758844231667111,
      "loss": 5.1086,
      "mean_token_accuracy": 0.22825663536787033,
      "num_tokens": 150313536.0,
      "step": 74180
    },
    {
      "entropy": 5.831011581420898,
      "epoch": 6.483002709079787,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017585644936935174,
      "loss": 5.085,
      "mean_token_accuracy": 0.22827102839946747,
      "num_tokens": 150323521.0,
      "step": 74185
    },
    {
      "entropy": 5.907744359970093,
      "epoch": 6.483439657432491,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017582847747361758,
      "loss": 5.2087,
      "mean_token_accuracy": 0.21609379053115846,
      "num_tokens": 150334185.0,
      "step": 74190
    },
    {
      "entropy": 5.895786094665527,
      "epoch": 6.483876605785197,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017580050748004523,
      "loss": 5.0943,
      "mean_token_accuracy": 0.22379720956087112,
      "num_tokens": 150343978.0,
      "step": 74195
    },
    {
      "entropy": 5.896799325942993,
      "epoch": 6.484313554137901,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017577253938917103,
      "loss": 5.2073,
      "mean_token_accuracy": 0.21823619604110717,
      "num_tokens": 150353732.0,
      "step": 74200
    },
    {
      "entropy": 5.869965410232544,
      "epoch": 6.484750502490606,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017574457320153165,
      "loss": 5.156,
      "mean_token_accuracy": 0.223934905230999,
      "num_tokens": 150363512.0,
      "step": 74205
    },
    {
      "entropy": 5.769422817230224,
      "epoch": 6.48518745084331,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017571660891766328,
      "loss": 5.0684,
      "mean_token_accuracy": 0.23959137201309205,
      "num_tokens": 150373677.0,
      "step": 74210
    },
    {
      "entropy": 5.844105339050293,
      "epoch": 6.485624399196015,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001756886465381024,
      "loss": 5.1776,
      "mean_token_accuracy": 0.22210632264614105,
      "num_tokens": 150384261.0,
      "step": 74215
    },
    {
      "entropy": 5.8377632141113285,
      "epoch": 6.48606134754872,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017566068606338526,
      "loss": 5.1,
      "mean_token_accuracy": 0.2307775855064392,
      "num_tokens": 150394159.0,
      "step": 74220
    },
    {
      "entropy": 5.861598062515259,
      "epoch": 6.486498295901424,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017563272749404824,
      "loss": 5.129,
      "mean_token_accuracy": 0.22829706221818924,
      "num_tokens": 150404089.0,
      "step": 74225
    },
    {
      "entropy": 5.8692464351654055,
      "epoch": 6.486935244254129,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001756047708306277,
      "loss": 5.0966,
      "mean_token_accuracy": 0.2275334358215332,
      "num_tokens": 150414234.0,
      "step": 74230
    },
    {
      "entropy": 5.912595653533936,
      "epoch": 6.487372192606834,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017557681607365967,
      "loss": 5.1975,
      "mean_token_accuracy": 0.22183555364608765,
      "num_tokens": 150424491.0,
      "step": 74235
    },
    {
      "entropy": 5.873813724517822,
      "epoch": 6.487809140959539,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017554886322368034,
      "loss": 5.1412,
      "mean_token_accuracy": 0.2255415990948677,
      "num_tokens": 150433928.0,
      "step": 74240
    },
    {
      "entropy": 5.8419801712036135,
      "epoch": 6.488246089312243,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017552091228122596,
      "loss": 5.0638,
      "mean_token_accuracy": 0.23510071337223054,
      "num_tokens": 150442950.0,
      "step": 74245
    },
    {
      "entropy": 5.860008955001831,
      "epoch": 6.4886830376649485,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001754929632468326,
      "loss": 5.1603,
      "mean_token_accuracy": 0.22641392201185226,
      "num_tokens": 150453973.0,
      "step": 74250
    },
    {
      "entropy": 5.8310152053833,
      "epoch": 6.489119986017653,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017546501612103634,
      "loss": 5.1584,
      "mean_token_accuracy": 0.23069673478603364,
      "num_tokens": 150464805.0,
      "step": 74255
    },
    {
      "entropy": 5.8104533672332765,
      "epoch": 6.489556934370357,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017543707090437321,
      "loss": 5.1269,
      "mean_token_accuracy": 0.2284701243042946,
      "num_tokens": 150474231.0,
      "step": 74260
    },
    {
      "entropy": 5.874285793304443,
      "epoch": 6.489993882723062,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017540912759737926,
      "loss": 5.1822,
      "mean_token_accuracy": 0.2221410021185875,
      "num_tokens": 150484060.0,
      "step": 74265
    },
    {
      "entropy": 5.855452585220337,
      "epoch": 6.490430831075767,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001753811862005904,
      "loss": 5.1464,
      "mean_token_accuracy": 0.21821054220199584,
      "num_tokens": 150494036.0,
      "step": 74270
    },
    {
      "entropy": 5.8959893703460695,
      "epoch": 6.490867779428472,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017535324671454262,
      "loss": 5.1695,
      "mean_token_accuracy": 0.2261888176202774,
      "num_tokens": 150503842.0,
      "step": 74275
    },
    {
      "entropy": 5.925,
      "epoch": 6.491304727781176,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001753253091397719,
      "loss": 5.182,
      "mean_token_accuracy": 0.2197259932756424,
      "num_tokens": 150513657.0,
      "step": 74280
    },
    {
      "entropy": 5.825108909606934,
      "epoch": 6.491741676133881,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017529737347681386,
      "loss": 5.1101,
      "mean_token_accuracy": 0.22742566466331482,
      "num_tokens": 150523193.0,
      "step": 74285
    },
    {
      "entropy": 5.854919910430908,
      "epoch": 6.492178624486586,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001752694397262044,
      "loss": 5.1939,
      "mean_token_accuracy": 0.2166520968079567,
      "num_tokens": 150533697.0,
      "step": 74290
    },
    {
      "entropy": 5.814464855194092,
      "epoch": 6.4926155728392905,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017524150788847937,
      "loss": 5.1192,
      "mean_token_accuracy": 0.23053910434246064,
      "num_tokens": 150542930.0,
      "step": 74295
    },
    {
      "entropy": 5.8926355838775635,
      "epoch": 6.493052521191995,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017521357796417448,
      "loss": 5.1758,
      "mean_token_accuracy": 0.22214075326919555,
      "num_tokens": 150552453.0,
      "step": 74300
    },
    {
      "entropy": 5.908994960784912,
      "epoch": 6.493489469544699,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017518564995382547,
      "loss": 5.2419,
      "mean_token_accuracy": 0.21469952166080475,
      "num_tokens": 150562045.0,
      "step": 74305
    },
    {
      "entropy": 5.886761093139649,
      "epoch": 6.493926417897405,
      "grad_norm": 1.265625,
      "learning_rate": 0.000175157723857968,
      "loss": 5.1433,
      "mean_token_accuracy": 0.22584627270698548,
      "num_tokens": 150571007.0,
      "step": 74310
    },
    {
      "entropy": 5.820954179763794,
      "epoch": 6.494363366250109,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001751297996771377,
      "loss": 5.1244,
      "mean_token_accuracy": 0.2295123517513275,
      "num_tokens": 150580378.0,
      "step": 74315
    },
    {
      "entropy": 5.82008547782898,
      "epoch": 6.494800314602814,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017510187741187017,
      "loss": 5.1474,
      "mean_token_accuracy": 0.2259451389312744,
      "num_tokens": 150590550.0,
      "step": 74320
    },
    {
      "entropy": 5.771988821029663,
      "epoch": 6.495237262955519,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017507395706270096,
      "loss": 5.0674,
      "mean_token_accuracy": 0.23445193320512772,
      "num_tokens": 150602223.0,
      "step": 74325
    },
    {
      "entropy": 5.800136852264404,
      "epoch": 6.4956742113082235,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001750460386301657,
      "loss": 5.0695,
      "mean_token_accuracy": 0.2337738499045372,
      "num_tokens": 150611412.0,
      "step": 74330
    },
    {
      "entropy": 5.830581712722778,
      "epoch": 6.496111159660928,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017501812211479978,
      "loss": 5.1387,
      "mean_token_accuracy": 0.2267576202750206,
      "num_tokens": 150622228.0,
      "step": 74335
    },
    {
      "entropy": 5.79988112449646,
      "epoch": 6.4965481080136325,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017499020751713857,
      "loss": 5.0925,
      "mean_token_accuracy": 0.22112250626087188,
      "num_tokens": 150631758.0,
      "step": 74340
    },
    {
      "entropy": 5.852042961120605,
      "epoch": 6.496985056366338,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017496229483771763,
      "loss": 5.1855,
      "mean_token_accuracy": 0.2217865452170372,
      "num_tokens": 150641537.0,
      "step": 74345
    },
    {
      "entropy": 5.72003583908081,
      "epoch": 6.497422004719042,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001749343840770723,
      "loss": 5.0265,
      "mean_token_accuracy": 0.24299768954515458,
      "num_tokens": 150650211.0,
      "step": 74350
    },
    {
      "entropy": 5.824460506439209,
      "epoch": 6.497858953071747,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017490647523573792,
      "loss": 5.0833,
      "mean_token_accuracy": 0.22893573045730592,
      "num_tokens": 150660487.0,
      "step": 74355
    },
    {
      "entropy": 5.836971044540405,
      "epoch": 6.498295901424451,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017487856831424977,
      "loss": 5.1138,
      "mean_token_accuracy": 0.22736020535230636,
      "num_tokens": 150670528.0,
      "step": 74360
    },
    {
      "entropy": 5.886629390716553,
      "epoch": 6.498732849777157,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017485066331314315,
      "loss": 5.1284,
      "mean_token_accuracy": 0.220489701628685,
      "num_tokens": 150680831.0,
      "step": 74365
    },
    {
      "entropy": 5.822115850448609,
      "epoch": 6.499169798129861,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001748227602329533,
      "loss": 5.1216,
      "mean_token_accuracy": 0.23436377495527266,
      "num_tokens": 150689985.0,
      "step": 74370
    },
    {
      "entropy": 5.811811828613282,
      "epoch": 6.4996067464825655,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017479485907421544,
      "loss": 5.0871,
      "mean_token_accuracy": 0.22832816541194917,
      "num_tokens": 150700184.0,
      "step": 74375
    },
    {
      "entropy": 5.764816665649414,
      "epoch": 6.50004369483527,
      "grad_norm": 1.125,
      "learning_rate": 0.00017476695983746466,
      "loss": 5.0906,
      "mean_token_accuracy": 0.23347988724708557,
      "num_tokens": 150710816.0,
      "step": 74380
    },
    {
      "entropy": 5.8247312068939205,
      "epoch": 6.500480643187975,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001747390625232361,
      "loss": 5.0509,
      "mean_token_accuracy": 0.23542152643203734,
      "num_tokens": 150720598.0,
      "step": 74385
    },
    {
      "entropy": 5.887665462493897,
      "epoch": 6.50091759154068,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001747111671320648,
      "loss": 5.2065,
      "mean_token_accuracy": 0.21954186856746674,
      "num_tokens": 150731717.0,
      "step": 74390
    },
    {
      "entropy": 5.911925315856934,
      "epoch": 6.501354539893384,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017468327366448595,
      "loss": 5.1534,
      "mean_token_accuracy": 0.22218178659677507,
      "num_tokens": 150741255.0,
      "step": 74395
    },
    {
      "entropy": 5.8146302700042725,
      "epoch": 6.50179148824609,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017465538212103443,
      "loss": 5.1384,
      "mean_token_accuracy": 0.22671164870262145,
      "num_tokens": 150751532.0,
      "step": 74400
    },
    {
      "entropy": 5.826131582260132,
      "epoch": 6.502228436598794,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017462749250224524,
      "loss": 5.1651,
      "mean_token_accuracy": 0.2249075099825859,
      "num_tokens": 150762609.0,
      "step": 74405
    },
    {
      "entropy": 5.802329349517822,
      "epoch": 6.502665384951499,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017459960480865327,
      "loss": 5.034,
      "mean_token_accuracy": 0.2264024645090103,
      "num_tokens": 150771437.0,
      "step": 74410
    },
    {
      "entropy": 5.834190750122071,
      "epoch": 6.503102333304203,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017457171904079372,
      "loss": 5.1305,
      "mean_token_accuracy": 0.2248489186167717,
      "num_tokens": 150781806.0,
      "step": 74415
    },
    {
      "entropy": 5.871825122833252,
      "epoch": 6.503539281656908,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00017454383519920107,
      "loss": 5.1821,
      "mean_token_accuracy": 0.22536755502223968,
      "num_tokens": 150792915.0,
      "step": 74420
    },
    {
      "entropy": 5.8411987781524655,
      "epoch": 6.503976230009613,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001745159532844103,
      "loss": 5.1945,
      "mean_token_accuracy": 0.2220282107591629,
      "num_tokens": 150802980.0,
      "step": 74425
    },
    {
      "entropy": 5.893220901489258,
      "epoch": 6.504413178362317,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017448807329695622,
      "loss": 5.1675,
      "mean_token_accuracy": 0.21322579085826873,
      "num_tokens": 150814360.0,
      "step": 74430
    },
    {
      "entropy": 5.915691661834717,
      "epoch": 6.504850126715022,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017446019523737356,
      "loss": 5.1738,
      "mean_token_accuracy": 0.22015644013881683,
      "num_tokens": 150824135.0,
      "step": 74435
    },
    {
      "entropy": 5.843747520446778,
      "epoch": 6.505287075067727,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017443231910619703,
      "loss": 5.1293,
      "mean_token_accuracy": 0.2282200962305069,
      "num_tokens": 150834843.0,
      "step": 74440
    },
    {
      "entropy": 5.878976631164551,
      "epoch": 6.505724023420432,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017440444490396135,
      "loss": 5.1386,
      "mean_token_accuracy": 0.22171110510826111,
      "num_tokens": 150844813.0,
      "step": 74445
    },
    {
      "entropy": 5.815541458129883,
      "epoch": 6.506160971773136,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017437657263120107,
      "loss": 5.0954,
      "mean_token_accuracy": 0.22922075688838958,
      "num_tokens": 150854828.0,
      "step": 74450
    },
    {
      "entropy": 5.814361238479615,
      "epoch": 6.5065979201258415,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001743487022884509,
      "loss": 5.1233,
      "mean_token_accuracy": 0.22637244164943696,
      "num_tokens": 150865803.0,
      "step": 74455
    },
    {
      "entropy": 5.789504051208496,
      "epoch": 6.507034868478546,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017432083387624531,
      "loss": 5.0495,
      "mean_token_accuracy": 0.23120368719100953,
      "num_tokens": 150875682.0,
      "step": 74460
    },
    {
      "entropy": 5.8935669422149655,
      "epoch": 6.50747181683125,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017429296739511903,
      "loss": 5.2642,
      "mean_token_accuracy": 0.2213817849755287,
      "num_tokens": 150887615.0,
      "step": 74465
    },
    {
      "entropy": 5.862496471405029,
      "epoch": 6.507908765183955,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00017426510284560627,
      "loss": 5.1547,
      "mean_token_accuracy": 0.22723651081323623,
      "num_tokens": 150897818.0,
      "step": 74470
    },
    {
      "entropy": 5.902993822097779,
      "epoch": 6.50834571353666,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00017423724022824167,
      "loss": 5.1501,
      "mean_token_accuracy": 0.22357550859451295,
      "num_tokens": 150908129.0,
      "step": 74475
    },
    {
      "entropy": 5.87047233581543,
      "epoch": 6.508782661889365,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017420937954355957,
      "loss": 5.1616,
      "mean_token_accuracy": 0.22047048658132554,
      "num_tokens": 150918267.0,
      "step": 74480
    },
    {
      "entropy": 5.830140447616577,
      "epoch": 6.509219610242069,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017418152079209435,
      "loss": 5.059,
      "mean_token_accuracy": 0.2354192316532135,
      "num_tokens": 150928255.0,
      "step": 74485
    },
    {
      "entropy": 5.815525722503662,
      "epoch": 6.509656558594774,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017415366397438044,
      "loss": 5.104,
      "mean_token_accuracy": 0.23422977775335313,
      "num_tokens": 150938103.0,
      "step": 74490
    },
    {
      "entropy": 5.869772434234619,
      "epoch": 6.510093506947479,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017412580909095205,
      "loss": 5.1688,
      "mean_token_accuracy": 0.2236022263765335,
      "num_tokens": 150949177.0,
      "step": 74495
    },
    {
      "entropy": 5.76165509223938,
      "epoch": 6.5105304553001835,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017409795614234353,
      "loss": 5.0267,
      "mean_token_accuracy": 0.23730306774377824,
      "num_tokens": 150958360.0,
      "step": 74500
    },
    {
      "entropy": 5.8455461978912355,
      "epoch": 6.510967403652888,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001740701051290891,
      "loss": 5.1054,
      "mean_token_accuracy": 0.2254016727209091,
      "num_tokens": 150968522.0,
      "step": 74505
    },
    {
      "entropy": 5.874509525299072,
      "epoch": 6.511404352005593,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017404225605172302,
      "loss": 5.1175,
      "mean_token_accuracy": 0.2308504283428192,
      "num_tokens": 150979308.0,
      "step": 74510
    },
    {
      "entropy": 5.884198713302612,
      "epoch": 6.511841300358298,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001740144089107793,
      "loss": 5.1554,
      "mean_token_accuracy": 0.22722212076187134,
      "num_tokens": 150989561.0,
      "step": 74515
    },
    {
      "entropy": 5.7322734832763675,
      "epoch": 6.512278248711002,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017398656370679204,
      "loss": 5.0263,
      "mean_token_accuracy": 0.24387008100748062,
      "num_tokens": 150999396.0,
      "step": 74520
    },
    {
      "entropy": 5.781173658370972,
      "epoch": 6.512715197063707,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017395872044029552,
      "loss": 5.0531,
      "mean_token_accuracy": 0.22562382966279984,
      "num_tokens": 151009879.0,
      "step": 74525
    },
    {
      "entropy": 5.814425230026245,
      "epoch": 6.513152145416412,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017393087911182366,
      "loss": 5.145,
      "mean_token_accuracy": 0.22555650174617767,
      "num_tokens": 151021142.0,
      "step": 74530
    },
    {
      "entropy": 5.805674934387207,
      "epoch": 6.5135890937691165,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017390303972191049,
      "loss": 5.0722,
      "mean_token_accuracy": 0.2392519235610962,
      "num_tokens": 151031040.0,
      "step": 74535
    },
    {
      "entropy": 5.849574089050293,
      "epoch": 6.514026042121821,
      "grad_norm": 1.125,
      "learning_rate": 0.00017387520227108997,
      "loss": 5.1225,
      "mean_token_accuracy": 0.22221139669418336,
      "num_tokens": 151041140.0,
      "step": 74540
    },
    {
      "entropy": 5.814621353149414,
      "epoch": 6.5144629904745255,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017384736675989608,
      "loss": 5.0764,
      "mean_token_accuracy": 0.23798331916332244,
      "num_tokens": 151050507.0,
      "step": 74545
    },
    {
      "entropy": 5.8531982421875,
      "epoch": 6.514899938827231,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017381953318886271,
      "loss": 5.1531,
      "mean_token_accuracy": 0.21888614743947982,
      "num_tokens": 151060376.0,
      "step": 74550
    },
    {
      "entropy": 5.895136213302612,
      "epoch": 6.515336887179935,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001737917015585237,
      "loss": 5.1498,
      "mean_token_accuracy": 0.22071198523044586,
      "num_tokens": 151069908.0,
      "step": 74555
    },
    {
      "entropy": 5.824333333969117,
      "epoch": 6.51577383553264,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017376387186941297,
      "loss": 5.085,
      "mean_token_accuracy": 0.23458793610334397,
      "num_tokens": 151080276.0,
      "step": 74560
    },
    {
      "entropy": 5.889244747161865,
      "epoch": 6.516210783885345,
      "grad_norm": 1.125,
      "learning_rate": 0.00017373604412206412,
      "loss": 5.2461,
      "mean_token_accuracy": 0.21091909259557723,
      "num_tokens": 151090183.0,
      "step": 74565
    },
    {
      "entropy": 5.897027492523193,
      "epoch": 6.51664773223805,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017370821831701105,
      "loss": 5.244,
      "mean_token_accuracy": 0.21950748264789582,
      "num_tokens": 151100635.0,
      "step": 74570
    },
    {
      "entropy": 5.869035005569458,
      "epoch": 6.517084680590754,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001736803944547874,
      "loss": 5.1633,
      "mean_token_accuracy": 0.22153975069522858,
      "num_tokens": 151111014.0,
      "step": 74575
    },
    {
      "entropy": 5.8802672863006595,
      "epoch": 6.5175216289434585,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001736525725359269,
      "loss": 5.2109,
      "mean_token_accuracy": 0.2187524601817131,
      "num_tokens": 151122153.0,
      "step": 74580
    },
    {
      "entropy": 5.916713047027588,
      "epoch": 6.517958577296164,
      "grad_norm": 1.125,
      "learning_rate": 0.00017362475256096313,
      "loss": 5.1729,
      "mean_token_accuracy": 0.22043536901473998,
      "num_tokens": 151132493.0,
      "step": 74585
    },
    {
      "entropy": 5.842700338363647,
      "epoch": 6.518395525648868,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017359693453042975,
      "loss": 5.1456,
      "mean_token_accuracy": 0.23110957741737365,
      "num_tokens": 151143698.0,
      "step": 74590
    },
    {
      "entropy": 5.869230794906616,
      "epoch": 6.518832474001573,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017356911844486027,
      "loss": 5.1366,
      "mean_token_accuracy": 0.22442597001791,
      "num_tokens": 151154003.0,
      "step": 74595
    },
    {
      "entropy": 5.848527097702027,
      "epoch": 6.519269422354277,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017354130430478827,
      "loss": 5.0817,
      "mean_token_accuracy": 0.23055584281682967,
      "num_tokens": 151164067.0,
      "step": 74600
    },
    {
      "entropy": 5.837478494644165,
      "epoch": 6.519706370706983,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001735134921107472,
      "loss": 5.1154,
      "mean_token_accuracy": 0.2234807401895523,
      "num_tokens": 151173820.0,
      "step": 74605
    },
    {
      "entropy": 5.86117582321167,
      "epoch": 6.520143319059687,
      "grad_norm": 1.25,
      "learning_rate": 0.00017348568186327065,
      "loss": 5.0269,
      "mean_token_accuracy": 0.22544922977685927,
      "num_tokens": 151183697.0,
      "step": 74610
    },
    {
      "entropy": 5.882255172729492,
      "epoch": 6.520580267412392,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017345787356289183,
      "loss": 5.1664,
      "mean_token_accuracy": 0.22452846318483352,
      "num_tokens": 151194294.0,
      "step": 74615
    },
    {
      "entropy": 5.815440940856933,
      "epoch": 6.521017215765097,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017343006721014419,
      "loss": 5.1123,
      "mean_token_accuracy": 0.2301260992884636,
      "num_tokens": 151203965.0,
      "step": 74620
    },
    {
      "entropy": 5.8543257236480715,
      "epoch": 6.521454164117801,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017340226280556105,
      "loss": 5.1198,
      "mean_token_accuracy": 0.22625983208417894,
      "num_tokens": 151214105.0,
      "step": 74625
    },
    {
      "entropy": 5.892944765090943,
      "epoch": 6.521891112470506,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017337446034967573,
      "loss": 5.2134,
      "mean_token_accuracy": 0.21847623735666274,
      "num_tokens": 151224698.0,
      "step": 74630
    },
    {
      "entropy": 5.827250623703003,
      "epoch": 6.52232806082321,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001733466598430216,
      "loss": 5.1388,
      "mean_token_accuracy": 0.21796779632568358,
      "num_tokens": 151236125.0,
      "step": 74635
    },
    {
      "entropy": 5.805305862426758,
      "epoch": 6.522765009175916,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001733188612861317,
      "loss": 5.1266,
      "mean_token_accuracy": 0.22514290362596512,
      "num_tokens": 151246338.0,
      "step": 74640
    },
    {
      "entropy": 5.856364774703979,
      "epoch": 6.52320195752862,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017329106467953937,
      "loss": 5.1862,
      "mean_token_accuracy": 0.2321103036403656,
      "num_tokens": 151256765.0,
      "step": 74645
    },
    {
      "entropy": 5.9299263000488285,
      "epoch": 6.523638905881325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017326327002377772,
      "loss": 5.2444,
      "mean_token_accuracy": 0.21527425050735474,
      "num_tokens": 151267067.0,
      "step": 74650
    },
    {
      "entropy": 5.823887014389038,
      "epoch": 6.524075854234029,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017323547731937987,
      "loss": 5.0341,
      "mean_token_accuracy": 0.2350343719124794,
      "num_tokens": 151277128.0,
      "step": 74655
    },
    {
      "entropy": 5.889476680755616,
      "epoch": 6.5245128025867345,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017320768656687896,
      "loss": 5.1988,
      "mean_token_accuracy": 0.22057972550392152,
      "num_tokens": 151286253.0,
      "step": 74660
    },
    {
      "entropy": 5.792010974884033,
      "epoch": 6.524949750939439,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017317989776680787,
      "loss": 5.0886,
      "mean_token_accuracy": 0.23055867999792098,
      "num_tokens": 151296320.0,
      "step": 74665
    },
    {
      "entropy": 5.813745784759521,
      "epoch": 6.525386699292143,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00017315211091969973,
      "loss": 5.0993,
      "mean_token_accuracy": 0.22631691992282868,
      "num_tokens": 151306663.0,
      "step": 74670
    },
    {
      "entropy": 5.8171412467956545,
      "epoch": 6.525823647644849,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001731243260260874,
      "loss": 5.1237,
      "mean_token_accuracy": 0.22977530658245088,
      "num_tokens": 151316641.0,
      "step": 74675
    },
    {
      "entropy": 5.855081796646118,
      "epoch": 6.526260595997553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017309654308650396,
      "loss": 5.1087,
      "mean_token_accuracy": 0.221814326941967,
      "num_tokens": 151327584.0,
      "step": 74680
    },
    {
      "entropy": 5.768422603607178,
      "epoch": 6.526697544350258,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017306876210148214,
      "loss": 5.0489,
      "mean_token_accuracy": 0.24200357645750045,
      "num_tokens": 151337262.0,
      "step": 74685
    },
    {
      "entropy": 5.833947944641113,
      "epoch": 6.527134492702962,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017304098307155492,
      "loss": 5.1579,
      "mean_token_accuracy": 0.22321096658706666,
      "num_tokens": 151347090.0,
      "step": 74690
    },
    {
      "entropy": 5.832805156707764,
      "epoch": 6.527571441055668,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017301320599725503,
      "loss": 5.1343,
      "mean_token_accuracy": 0.22723657935857772,
      "num_tokens": 151356982.0,
      "step": 74695
    },
    {
      "entropy": 5.891718006134033,
      "epoch": 6.528008389408372,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017298543087911533,
      "loss": 5.1938,
      "mean_token_accuracy": 0.2249947115778923,
      "num_tokens": 151366914.0,
      "step": 74700
    },
    {
      "entropy": 5.804526853561401,
      "epoch": 6.5284453377610765,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017295765771766864,
      "loss": 5.098,
      "mean_token_accuracy": 0.2255772054195404,
      "num_tokens": 151376341.0,
      "step": 74705
    },
    {
      "entropy": 5.84339017868042,
      "epoch": 6.528882286113781,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017292988651344737,
      "loss": 5.1659,
      "mean_token_accuracy": 0.21957187056541444,
      "num_tokens": 151387347.0,
      "step": 74710
    },
    {
      "entropy": 5.807166957855225,
      "epoch": 6.529319234466486,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017290211726698443,
      "loss": 5.0793,
      "mean_token_accuracy": 0.2293060764670372,
      "num_tokens": 151396858.0,
      "step": 74715
    },
    {
      "entropy": 5.848456859588623,
      "epoch": 6.529756182819191,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017287434997881234,
      "loss": 5.0915,
      "mean_token_accuracy": 0.23124003559350967,
      "num_tokens": 151406680.0,
      "step": 74720
    },
    {
      "entropy": 5.717595386505127,
      "epoch": 6.530193131171895,
      "grad_norm": 1.03125,
      "learning_rate": 0.00017284658464946376,
      "loss": 4.963,
      "mean_token_accuracy": 0.24860082864761351,
      "num_tokens": 151417163.0,
      "step": 74725
    },
    {
      "entropy": 5.800181722640991,
      "epoch": 6.530630079524601,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017281882127947117,
      "loss": 5.1075,
      "mean_token_accuracy": 0.23309026807546615,
      "num_tokens": 151427389.0,
      "step": 74730
    },
    {
      "entropy": 5.898140287399292,
      "epoch": 6.531067027877305,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001727910598693672,
      "loss": 5.1687,
      "mean_token_accuracy": 0.2226202517747879,
      "num_tokens": 151437057.0,
      "step": 74735
    },
    {
      "entropy": 5.9013991355896,
      "epoch": 6.53150397623001,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017276330041968418,
      "loss": 5.1718,
      "mean_token_accuracy": 0.22484539747238158,
      "num_tokens": 151447536.0,
      "step": 74740
    },
    {
      "entropy": 5.883897829055786,
      "epoch": 6.531940924582714,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017273554293095468,
      "loss": 5.1385,
      "mean_token_accuracy": 0.22295950800180436,
      "num_tokens": 151457063.0,
      "step": 74745
    },
    {
      "entropy": 5.83137469291687,
      "epoch": 6.532377872935419,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017270778740371105,
      "loss": 5.0598,
      "mean_token_accuracy": 0.2373654842376709,
      "num_tokens": 151466571.0,
      "step": 74750
    },
    {
      "entropy": 5.852616167068481,
      "epoch": 6.532814821288124,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001726800338384857,
      "loss": 5.1037,
      "mean_token_accuracy": 0.22904598861932754,
      "num_tokens": 151477030.0,
      "step": 74755
    },
    {
      "entropy": 5.858462238311768,
      "epoch": 6.533251769640828,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001726522822358108,
      "loss": 5.1775,
      "mean_token_accuracy": 0.21872851699590684,
      "num_tokens": 151487045.0,
      "step": 74760
    },
    {
      "entropy": 5.791012525558472,
      "epoch": 6.533688717993533,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001726245325962188,
      "loss": 5.1429,
      "mean_token_accuracy": 0.22813717424869537,
      "num_tokens": 151497824.0,
      "step": 74765
    },
    {
      "entropy": 5.858695793151855,
      "epoch": 6.534125666346238,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017259678492024193,
      "loss": 5.1782,
      "mean_token_accuracy": 0.21917569488286973,
      "num_tokens": 151507291.0,
      "step": 74770
    },
    {
      "entropy": 5.823984384536743,
      "epoch": 6.534562614698943,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001725690392084124,
      "loss": 5.1811,
      "mean_token_accuracy": 0.21906374543905258,
      "num_tokens": 151518108.0,
      "step": 74775
    },
    {
      "entropy": 5.863893651962281,
      "epoch": 6.534999563051647,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00017254129546126235,
      "loss": 5.1371,
      "mean_token_accuracy": 0.22021412700414658,
      "num_tokens": 151529043.0,
      "step": 74780
    },
    {
      "entropy": 5.858407640457154,
      "epoch": 6.5354365114043524,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017251355367932394,
      "loss": 5.1547,
      "mean_token_accuracy": 0.22327273041009904,
      "num_tokens": 151538825.0,
      "step": 74785
    },
    {
      "entropy": 5.869676733016968,
      "epoch": 6.535873459757057,
      "grad_norm": 1.125,
      "learning_rate": 0.0001724858138631293,
      "loss": 5.1234,
      "mean_token_accuracy": 0.2231223925948143,
      "num_tokens": 151549183.0,
      "step": 74790
    },
    {
      "entropy": 5.920793914794922,
      "epoch": 6.536310408109761,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00017245807601321052,
      "loss": 5.2098,
      "mean_token_accuracy": 0.2200920209288597,
      "num_tokens": 151559888.0,
      "step": 74795
    },
    {
      "entropy": 5.863823890686035,
      "epoch": 6.536747356462466,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001724303401300995,
      "loss": 5.1642,
      "mean_token_accuracy": 0.23081364184617997,
      "num_tokens": 151570997.0,
      "step": 74800
    },
    {
      "entropy": 5.818921995162964,
      "epoch": 6.537184304815171,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017240260621432827,
      "loss": 5.0942,
      "mean_token_accuracy": 0.2283409893512726,
      "num_tokens": 151581563.0,
      "step": 74805
    },
    {
      "entropy": 5.883257293701172,
      "epoch": 6.537621253167876,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017237487426642885,
      "loss": 5.136,
      "mean_token_accuracy": 0.22998303174972534,
      "num_tokens": 151591438.0,
      "step": 74810
    },
    {
      "entropy": 5.850209665298462,
      "epoch": 6.53805820152058,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001723471442869331,
      "loss": 5.1136,
      "mean_token_accuracy": 0.21646336019039153,
      "num_tokens": 151600493.0,
      "step": 74815
    },
    {
      "entropy": 5.771051216125488,
      "epoch": 6.538495149873285,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001723194162763729,
      "loss": 5.1281,
      "mean_token_accuracy": 0.2261558100581169,
      "num_tokens": 151609697.0,
      "step": 74820
    },
    {
      "entropy": 5.866156482696534,
      "epoch": 6.53893209822599,
      "grad_norm": 1.125,
      "learning_rate": 0.00017229169023528013,
      "loss": 5.1544,
      "mean_token_accuracy": 0.22519274801015854,
      "num_tokens": 151619869.0,
      "step": 74825
    },
    {
      "entropy": 5.860780143737793,
      "epoch": 6.5393690465786944,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017226396616418655,
      "loss": 5.174,
      "mean_token_accuracy": 0.22791684567928314,
      "num_tokens": 151630457.0,
      "step": 74830
    },
    {
      "entropy": 5.811419057846069,
      "epoch": 6.539805994931399,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017223624406362392,
      "loss": 5.0772,
      "mean_token_accuracy": 0.2346037894487381,
      "num_tokens": 151640980.0,
      "step": 74835
    },
    {
      "entropy": 5.79986538887024,
      "epoch": 6.540242943284104,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001722085239341241,
      "loss": 5.0443,
      "mean_token_accuracy": 0.2273050755262375,
      "num_tokens": 151652134.0,
      "step": 74840
    },
    {
      "entropy": 5.807658290863037,
      "epoch": 6.540679891636809,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017218080577621853,
      "loss": 5.1355,
      "mean_token_accuracy": 0.22145283818244935,
      "num_tokens": 151662643.0,
      "step": 74845
    },
    {
      "entropy": 5.845382738113403,
      "epoch": 6.541116839989513,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017215308959043897,
      "loss": 5.1266,
      "mean_token_accuracy": 0.23658183217048645,
      "num_tokens": 151672795.0,
      "step": 74850
    },
    {
      "entropy": 5.870053052902222,
      "epoch": 6.541553788342218,
      "grad_norm": 1.125,
      "learning_rate": 0.00017212537537731706,
      "loss": 5.1813,
      "mean_token_accuracy": 0.22180319130420684,
      "num_tokens": 151683449.0,
      "step": 74855
    },
    {
      "entropy": 5.809915161132812,
      "epoch": 6.541990736694923,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001720976631373844,
      "loss": 5.0876,
      "mean_token_accuracy": 0.23586752265691757,
      "num_tokens": 151693576.0,
      "step": 74860
    },
    {
      "entropy": 5.826575708389282,
      "epoch": 6.5424276850476275,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001720699528711724,
      "loss": 5.1424,
      "mean_token_accuracy": 0.23026684671640396,
      "num_tokens": 151703387.0,
      "step": 74865
    },
    {
      "entropy": 5.879665517807007,
      "epoch": 6.542864633400332,
      "grad_norm": 1.25,
      "learning_rate": 0.0001720422445792127,
      "loss": 5.1503,
      "mean_token_accuracy": 0.22808550149202347,
      "num_tokens": 151713421.0,
      "step": 74870
    },
    {
      "entropy": 5.742947769165039,
      "epoch": 6.5433015817530364,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017201453826203668,
      "loss": 5.0345,
      "mean_token_accuracy": 0.23224849998950958,
      "num_tokens": 151723492.0,
      "step": 74875
    },
    {
      "entropy": 5.829877710342407,
      "epoch": 6.543738530105742,
      "grad_norm": 1.078125,
      "learning_rate": 0.00017198683392017578,
      "loss": 5.1449,
      "mean_token_accuracy": 0.2267685130238533,
      "num_tokens": 151734132.0,
      "step": 74880
    },
    {
      "entropy": 5.837067413330078,
      "epoch": 6.544175478458446,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017195913155416135,
      "loss": 5.1105,
      "mean_token_accuracy": 0.23504483848810195,
      "num_tokens": 151743750.0,
      "step": 74885
    },
    {
      "entropy": 5.8591132164001465,
      "epoch": 6.544612426811151,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017193143116452488,
      "loss": 5.161,
      "mean_token_accuracy": 0.22628505229949952,
      "num_tokens": 151754369.0,
      "step": 74890
    },
    {
      "entropy": 5.885539960861206,
      "epoch": 6.545049375163856,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001719037327517975,
      "loss": 5.1531,
      "mean_token_accuracy": 0.22996583878993987,
      "num_tokens": 151765896.0,
      "step": 74895
    },
    {
      "entropy": 5.824945974349975,
      "epoch": 6.545486323516561,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017187603631651051,
      "loss": 5.1195,
      "mean_token_accuracy": 0.23091184943914414,
      "num_tokens": 151776017.0,
      "step": 74900
    },
    {
      "entropy": 5.817766094207764,
      "epoch": 6.545923271869265,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001718483418591952,
      "loss": 5.1415,
      "mean_token_accuracy": 0.23361337631940843,
      "num_tokens": 151784944.0,
      "step": 74905
    },
    {
      "entropy": 5.826651668548584,
      "epoch": 6.5463602202219695,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017182064938038276,
      "loss": 5.1307,
      "mean_token_accuracy": 0.23282610923051833,
      "num_tokens": 151796669.0,
      "step": 74910
    },
    {
      "entropy": 5.923692464828491,
      "epoch": 6.546797168574674,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017179295888060432,
      "loss": 5.1332,
      "mean_token_accuracy": 0.22433491051197052,
      "num_tokens": 151806099.0,
      "step": 74915
    },
    {
      "entropy": 5.877835702896118,
      "epoch": 6.547234116927379,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017176527036039102,
      "loss": 5.1732,
      "mean_token_accuracy": 0.22288728803396224,
      "num_tokens": 151815821.0,
      "step": 74920
    },
    {
      "entropy": 5.949569177627564,
      "epoch": 6.547671065280084,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017173758382027388,
      "loss": 5.2609,
      "mean_token_accuracy": 0.21795069873332978,
      "num_tokens": 151826415.0,
      "step": 74925
    },
    {
      "entropy": 5.833185243606567,
      "epoch": 6.548108013632788,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000171709899260784,
      "loss": 5.1152,
      "mean_token_accuracy": 0.22954798638820648,
      "num_tokens": 151836444.0,
      "step": 74930
    },
    {
      "entropy": 5.841530609130859,
      "epoch": 6.548544961985494,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017168221668245243,
      "loss": 5.0727,
      "mean_token_accuracy": 0.22535610496997832,
      "num_tokens": 151846591.0,
      "step": 74935
    },
    {
      "entropy": 5.8287022590637205,
      "epoch": 6.548981910338198,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017165453608581016,
      "loss": 5.1297,
      "mean_token_accuracy": 0.21785818338394164,
      "num_tokens": 151856012.0,
      "step": 74940
    },
    {
      "entropy": 5.811017036437988,
      "epoch": 6.549418858690903,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001716268574713879,
      "loss": 5.0917,
      "mean_token_accuracy": 0.23118188977241516,
      "num_tokens": 151865993.0,
      "step": 74945
    },
    {
      "entropy": 5.850158834457398,
      "epoch": 6.549855807043608,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017159918083971672,
      "loss": 5.1071,
      "mean_token_accuracy": 0.2286472350358963,
      "num_tokens": 151875929.0,
      "step": 74950
    },
    {
      "entropy": 6.00029730796814,
      "epoch": 6.550292755396312,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017157150619132743,
      "loss": 5.3428,
      "mean_token_accuracy": 0.20873293727636338,
      "num_tokens": 151886996.0,
      "step": 74955
    },
    {
      "entropy": 5.910454225540161,
      "epoch": 6.550729703749017,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017154383352675084,
      "loss": 5.1947,
      "mean_token_accuracy": 0.2222296491265297,
      "num_tokens": 151897858.0,
      "step": 74960
    },
    {
      "entropy": 5.861771202087402,
      "epoch": 6.551166652101721,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017151616284651774,
      "loss": 5.0966,
      "mean_token_accuracy": 0.2293229728937149,
      "num_tokens": 151907685.0,
      "step": 74965
    },
    {
      "entropy": 5.793583106994629,
      "epoch": 6.551603600454426,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017148849415115887,
      "loss": 5.0873,
      "mean_token_accuracy": 0.23766630440950393,
      "num_tokens": 151917401.0,
      "step": 74970
    },
    {
      "entropy": 5.84120807647705,
      "epoch": 6.552040548807131,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001714608274412049,
      "loss": 5.1114,
      "mean_token_accuracy": 0.23126125186681748,
      "num_tokens": 151927386.0,
      "step": 74975
    },
    {
      "entropy": 5.868842601776123,
      "epoch": 6.552477497159836,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001714331627171865,
      "loss": 5.108,
      "mean_token_accuracy": 0.2225910633802414,
      "num_tokens": 151937457.0,
      "step": 74980
    },
    {
      "entropy": 5.832253408432007,
      "epoch": 6.55291444551254,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017140549997963445,
      "loss": 5.1257,
      "mean_token_accuracy": 0.22524378299713135,
      "num_tokens": 151947928.0,
      "step": 74985
    },
    {
      "entropy": 5.832045030593872,
      "epoch": 6.5533513938652455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017137783922907903,
      "loss": 5.167,
      "mean_token_accuracy": 0.22690285742282867,
      "num_tokens": 151957327.0,
      "step": 74990
    },
    {
      "entropy": 5.81109585762024,
      "epoch": 6.55378834221795,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000171350180466051,
      "loss": 5.1229,
      "mean_token_accuracy": 0.2289107546210289,
      "num_tokens": 151967298.0,
      "step": 74995
    },
    {
      "entropy": 5.780852937698365,
      "epoch": 6.554225290570654,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001713225236910808,
      "loss": 5.0198,
      "mean_token_accuracy": 0.23356935977935792,
      "num_tokens": 151978003.0,
      "step": 75000
    },
    {
      "epoch": 6.554225290570654,
      "eval_entropy": 5.585022057883034,
      "eval_loss": 5.249141216278076,
      "eval_mean_token_accuracy": 0.22691678676391128,
      "eval_num_tokens": 151978003.0,
      "eval_runtime": 27.9139,
      "eval_samples_per_second": 1326.006,
      "eval_steps_per_second": 165.76,
      "step": 75000
    },
    {
      "entropy": 5.849746990203857,
      "epoch": 6.55466223892336,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017129486890469898,
      "loss": 5.1373,
      "mean_token_accuracy": 0.23536056280136108,
      "num_tokens": 151988547.0,
      "step": 75005
    },
    {
      "entropy": 5.87942795753479,
      "epoch": 6.555099187276064,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017126721610743583,
      "loss": 5.1775,
      "mean_token_accuracy": 0.22524345368146897,
      "num_tokens": 151997701.0,
      "step": 75010
    },
    {
      "entropy": 5.84477744102478,
      "epoch": 6.555536135628769,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017123956529982188,
      "loss": 5.1561,
      "mean_token_accuracy": 0.22882001399993895,
      "num_tokens": 152008238.0,
      "step": 75015
    },
    {
      "entropy": 5.866059589385986,
      "epoch": 6.555973083981473,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001712119164823874,
      "loss": 5.1615,
      "mean_token_accuracy": 0.22268118262290953,
      "num_tokens": 152018167.0,
      "step": 75020
    },
    {
      "entropy": 5.831728076934814,
      "epoch": 6.556410032334178,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001711842696556628,
      "loss": 5.078,
      "mean_token_accuracy": 0.22663387060165405,
      "num_tokens": 152027181.0,
      "step": 75025
    },
    {
      "entropy": 5.81962628364563,
      "epoch": 6.556846980686883,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017115662482017823,
      "loss": 5.1408,
      "mean_token_accuracy": 0.22577926963567735,
      "num_tokens": 152037819.0,
      "step": 75030
    },
    {
      "entropy": 5.815863561630249,
      "epoch": 6.5572839290395875,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00017112898197646416,
      "loss": 5.0716,
      "mean_token_accuracy": 0.22983201295137407,
      "num_tokens": 152048335.0,
      "step": 75035
    },
    {
      "entropy": 5.854509401321411,
      "epoch": 6.557720877392292,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017110134112505054,
      "loss": 5.1259,
      "mean_token_accuracy": 0.22195638716220856,
      "num_tokens": 152057043.0,
      "step": 75040
    },
    {
      "entropy": 5.786817264556885,
      "epoch": 6.558157825744997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017107370226646765,
      "loss": 5.0196,
      "mean_token_accuracy": 0.2340904414653778,
      "num_tokens": 152067473.0,
      "step": 75045
    },
    {
      "entropy": 5.812383460998535,
      "epoch": 6.558594774097702,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00017104606540124562,
      "loss": 5.169,
      "mean_token_accuracy": 0.21908900290727615,
      "num_tokens": 152077304.0,
      "step": 75050
    },
    {
      "entropy": 5.878760719299317,
      "epoch": 6.559031722450406,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017101843052991448,
      "loss": 5.2034,
      "mean_token_accuracy": 0.22114488035440444,
      "num_tokens": 152088297.0,
      "step": 75055
    },
    {
      "entropy": 5.835913944244385,
      "epoch": 6.559468670803111,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017099079765300437,
      "loss": 5.1914,
      "mean_token_accuracy": 0.2183254137635231,
      "num_tokens": 152098975.0,
      "step": 75060
    },
    {
      "entropy": 5.928298997879028,
      "epoch": 6.559905619155816,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017096316677104524,
      "loss": 5.1777,
      "mean_token_accuracy": 0.2251997634768486,
      "num_tokens": 152108641.0,
      "step": 75065
    },
    {
      "entropy": 5.851318359375,
      "epoch": 6.5603425675085205,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017093553788456712,
      "loss": 5.1198,
      "mean_token_accuracy": 0.2286348283290863,
      "num_tokens": 152118746.0,
      "step": 75070
    },
    {
      "entropy": 5.899266338348388,
      "epoch": 6.560779515861225,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017090791099409983,
      "loss": 5.2348,
      "mean_token_accuracy": 0.21328892558813095,
      "num_tokens": 152129511.0,
      "step": 75075
    },
    {
      "entropy": 5.863244962692261,
      "epoch": 6.5612164642139295,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001708802861001734,
      "loss": 5.1611,
      "mean_token_accuracy": 0.22293821722269058,
      "num_tokens": 152139963.0,
      "step": 75080
    },
    {
      "entropy": 5.7994788646697994,
      "epoch": 6.561653412566635,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001708526632033177,
      "loss": 5.0611,
      "mean_token_accuracy": 0.2318690910935402,
      "num_tokens": 152149975.0,
      "step": 75085
    },
    {
      "entropy": 5.819249773025513,
      "epoch": 6.562090360919339,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017082504230406248,
      "loss": 5.0759,
      "mean_token_accuracy": 0.22885710299015044,
      "num_tokens": 152160176.0,
      "step": 75090
    },
    {
      "entropy": 5.92138147354126,
      "epoch": 6.562527309272044,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017079742340293748,
      "loss": 5.213,
      "mean_token_accuracy": 0.21621890962123871,
      "num_tokens": 152170359.0,
      "step": 75095
    },
    {
      "entropy": 5.823499488830566,
      "epoch": 6.562964257624749,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00017076980650047253,
      "loss": 5.1364,
      "mean_token_accuracy": 0.2253219246864319,
      "num_tokens": 152181601.0,
      "step": 75100
    },
    {
      "entropy": 5.889943933486938,
      "epoch": 6.563401205977454,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00017074219159719727,
      "loss": 5.18,
      "mean_token_accuracy": 0.2216773271560669,
      "num_tokens": 152192035.0,
      "step": 75105
    },
    {
      "entropy": 5.860951995849609,
      "epoch": 6.563838154330158,
      "grad_norm": 1.25,
      "learning_rate": 0.00017071457869364143,
      "loss": 5.1156,
      "mean_token_accuracy": 0.23080596923828126,
      "num_tokens": 152201205.0,
      "step": 75110
    },
    {
      "entropy": 5.826960134506225,
      "epoch": 6.5642751026828625,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001706869677903345,
      "loss": 5.1268,
      "mean_token_accuracy": 0.2278878852725029,
      "num_tokens": 152212034.0,
      "step": 75115
    },
    {
      "entropy": 5.8302816390991214,
      "epoch": 6.564712051035568,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017065935888780636,
      "loss": 5.1727,
      "mean_token_accuracy": 0.22298354804515838,
      "num_tokens": 152220831.0,
      "step": 75120
    },
    {
      "entropy": 5.857802486419677,
      "epoch": 6.565148999388272,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017063175198658636,
      "loss": 5.2026,
      "mean_token_accuracy": 0.22700258195400239,
      "num_tokens": 152231648.0,
      "step": 75125
    },
    {
      "entropy": 5.8659648418426515,
      "epoch": 6.565585947740977,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000170604147087204,
      "loss": 5.1454,
      "mean_token_accuracy": 0.22523136734962462,
      "num_tokens": 152241401.0,
      "step": 75130
    },
    {
      "entropy": 5.872737693786621,
      "epoch": 6.566022896093681,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017057654419018882,
      "loss": 5.1414,
      "mean_token_accuracy": 0.22392525374889374,
      "num_tokens": 152251101.0,
      "step": 75135
    },
    {
      "entropy": 5.866718435287476,
      "epoch": 6.566459844446387,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00017054894329607024,
      "loss": 5.1292,
      "mean_token_accuracy": 0.23331964612007142,
      "num_tokens": 152261867.0,
      "step": 75140
    },
    {
      "entropy": 5.8649036407470705,
      "epoch": 6.566896792799091,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017052134440537762,
      "loss": 5.1139,
      "mean_token_accuracy": 0.22913839519023896,
      "num_tokens": 152271478.0,
      "step": 75145
    },
    {
      "entropy": 5.838688039779663,
      "epoch": 6.567333741151796,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017049374751864038,
      "loss": 5.1937,
      "mean_token_accuracy": 0.2163303956389427,
      "num_tokens": 152282668.0,
      "step": 75150
    },
    {
      "entropy": 5.875179052352905,
      "epoch": 6.567770689504501,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017046615263638788,
      "loss": 5.1861,
      "mean_token_accuracy": 0.2234261155128479,
      "num_tokens": 152292897.0,
      "step": 75155
    },
    {
      "entropy": 5.849946546554565,
      "epoch": 6.568207637857205,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001704385597591493,
      "loss": 5.1382,
      "mean_token_accuracy": 0.23127735257148743,
      "num_tokens": 152302746.0,
      "step": 75160
    },
    {
      "entropy": 5.824396085739136,
      "epoch": 6.56864458620991,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001704109688874539,
      "loss": 5.075,
      "mean_token_accuracy": 0.23012714236974716,
      "num_tokens": 152312967.0,
      "step": 75165
    },
    {
      "entropy": 5.8140449047088625,
      "epoch": 6.569081534562614,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017038338002183103,
      "loss": 5.0026,
      "mean_token_accuracy": 0.23362064063549043,
      "num_tokens": 152322489.0,
      "step": 75170
    },
    {
      "entropy": 5.870145320892334,
      "epoch": 6.56951848291532,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001703557931628097,
      "loss": 5.1458,
      "mean_token_accuracy": 0.21637202203273773,
      "num_tokens": 152332936.0,
      "step": 75175
    },
    {
      "entropy": 5.853695440292358,
      "epoch": 6.569955431268024,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017032820831091905,
      "loss": 5.1732,
      "mean_token_accuracy": 0.22169762551784516,
      "num_tokens": 152344625.0,
      "step": 75180
    },
    {
      "entropy": 5.9069725513458256,
      "epoch": 6.570392379620729,
      "grad_norm": 1.03125,
      "learning_rate": 0.00017030062546668823,
      "loss": 5.2012,
      "mean_token_accuracy": 0.22201840281486512,
      "num_tokens": 152355263.0,
      "step": 75185
    },
    {
      "entropy": 5.795952606201172,
      "epoch": 6.570829327973433,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017027304463064624,
      "loss": 5.0896,
      "mean_token_accuracy": 0.2286880850791931,
      "num_tokens": 152365096.0,
      "step": 75190
    },
    {
      "entropy": 5.7944550037384035,
      "epoch": 6.5712662763261385,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001702454658033222,
      "loss": 5.122,
      "mean_token_accuracy": 0.22486895471811294,
      "num_tokens": 152376273.0,
      "step": 75195
    },
    {
      "entropy": 5.921448707580566,
      "epoch": 6.571703224678843,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000170217888985245,
      "loss": 5.2432,
      "mean_token_accuracy": 0.21099198758602142,
      "num_tokens": 152385327.0,
      "step": 75200
    },
    {
      "entropy": 5.91284408569336,
      "epoch": 6.572140173031547,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017019031417694352,
      "loss": 5.1223,
      "mean_token_accuracy": 0.22681857347488404,
      "num_tokens": 152394378.0,
      "step": 75205
    },
    {
      "entropy": 5.816957235336304,
      "epoch": 6.572577121384253,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017016274137894677,
      "loss": 5.0936,
      "mean_token_accuracy": 0.2316028282046318,
      "num_tokens": 152404061.0,
      "step": 75210
    },
    {
      "entropy": 5.84780478477478,
      "epoch": 6.573014069736957,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017013517059178362,
      "loss": 5.072,
      "mean_token_accuracy": 0.2308264657855034,
      "num_tokens": 152414667.0,
      "step": 75215
    },
    {
      "entropy": 5.835027837753296,
      "epoch": 6.573451018089662,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001701076018159829,
      "loss": 5.0854,
      "mean_token_accuracy": 0.2194909229874611,
      "num_tokens": 152424710.0,
      "step": 75220
    },
    {
      "entropy": 5.902813720703125,
      "epoch": 6.573887966442366,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017008003505207325,
      "loss": 5.1538,
      "mean_token_accuracy": 0.22211713194847107,
      "num_tokens": 152434358.0,
      "step": 75225
    },
    {
      "entropy": 5.802985191345215,
      "epoch": 6.5743249147950715,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017005247030058352,
      "loss": 5.0303,
      "mean_token_accuracy": 0.23939485251903533,
      "num_tokens": 152443518.0,
      "step": 75230
    },
    {
      "entropy": 5.715892791748047,
      "epoch": 6.574761863147776,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017002490756204237,
      "loss": 4.9824,
      "mean_token_accuracy": 0.24137175530195237,
      "num_tokens": 152454196.0,
      "step": 75235
    },
    {
      "entropy": 5.745715570449829,
      "epoch": 6.5751988115004805,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016999734683697854,
      "loss": 5.0649,
      "mean_token_accuracy": 0.23444895297288895,
      "num_tokens": 152463926.0,
      "step": 75240
    },
    {
      "entropy": 5.833428764343262,
      "epoch": 6.575635759853185,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016996978812592058,
      "loss": 5.1593,
      "mean_token_accuracy": 0.22333595752716065,
      "num_tokens": 152473785.0,
      "step": 75245
    },
    {
      "entropy": 5.875739574432373,
      "epoch": 6.57607270820589,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016994223142939712,
      "loss": 5.1286,
      "mean_token_accuracy": 0.2242927670478821,
      "num_tokens": 152483711.0,
      "step": 75250
    },
    {
      "entropy": 5.839529800415039,
      "epoch": 6.576509656558595,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016991467674793676,
      "loss": 5.0522,
      "mean_token_accuracy": 0.24549976885318756,
      "num_tokens": 152492836.0,
      "step": 75255
    },
    {
      "entropy": 5.849619722366333,
      "epoch": 6.576946604911299,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001698871240820679,
      "loss": 5.1246,
      "mean_token_accuracy": 0.2305953934788704,
      "num_tokens": 152503773.0,
      "step": 75260
    },
    {
      "entropy": 5.869789075851441,
      "epoch": 6.577383553264005,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016985957343231922,
      "loss": 5.2049,
      "mean_token_accuracy": 0.21920028030872346,
      "num_tokens": 152513353.0,
      "step": 75265
    },
    {
      "entropy": 5.805629301071167,
      "epoch": 6.577820501616709,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001698320247992189,
      "loss": 5.0941,
      "mean_token_accuracy": 0.22802896052598953,
      "num_tokens": 152523567.0,
      "step": 75270
    },
    {
      "entropy": 5.843802070617675,
      "epoch": 6.5782574499694135,
      "grad_norm": 1.25,
      "learning_rate": 0.00016980447818329547,
      "loss": 5.1316,
      "mean_token_accuracy": 0.22540153414011002,
      "num_tokens": 152533610.0,
      "step": 75275
    },
    {
      "entropy": 5.655295515060425,
      "epoch": 6.578694398322118,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016977693358507723,
      "loss": 4.9555,
      "mean_token_accuracy": 0.2482635945081711,
      "num_tokens": 152543186.0,
      "step": 75280
    },
    {
      "entropy": 5.77034854888916,
      "epoch": 6.579131346674823,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001697493910050926,
      "loss": 5.059,
      "mean_token_accuracy": 0.23173180222511292,
      "num_tokens": 152554264.0,
      "step": 75285
    },
    {
      "entropy": 5.846754932403565,
      "epoch": 6.579568295027528,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016972185044386974,
      "loss": 5.1078,
      "mean_token_accuracy": 0.22634333819150926,
      "num_tokens": 152564290.0,
      "step": 75290
    },
    {
      "entropy": 5.891306924819946,
      "epoch": 6.580005243380232,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016969431190193697,
      "loss": 5.1765,
      "mean_token_accuracy": 0.21906857192516327,
      "num_tokens": 152574348.0,
      "step": 75295
    },
    {
      "entropy": 5.784286880493164,
      "epoch": 6.580442191732937,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001696667753798225,
      "loss": 5.131,
      "mean_token_accuracy": 0.2256764680147171,
      "num_tokens": 152584728.0,
      "step": 75300
    },
    {
      "entropy": 5.794996452331543,
      "epoch": 6.580879140085642,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016963924087805448,
      "loss": 5.0576,
      "mean_token_accuracy": 0.2363900661468506,
      "num_tokens": 152595324.0,
      "step": 75305
    },
    {
      "entropy": 5.867015743255616,
      "epoch": 6.581316088438347,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016961170839716096,
      "loss": 5.1509,
      "mean_token_accuracy": 0.22717494368553162,
      "num_tokens": 152607473.0,
      "step": 75310
    },
    {
      "entropy": 5.808022832870483,
      "epoch": 6.581753036791051,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016958417793767026,
      "loss": 4.965,
      "mean_token_accuracy": 0.24381362199783324,
      "num_tokens": 152616999.0,
      "step": 75315
    },
    {
      "entropy": 5.773401880264283,
      "epoch": 6.582189985143756,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016955664950011012,
      "loss": 5.0642,
      "mean_token_accuracy": 0.23196402043104172,
      "num_tokens": 152626769.0,
      "step": 75320
    },
    {
      "entropy": 5.864603424072266,
      "epoch": 6.582626933496461,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016952912308500873,
      "loss": 5.1453,
      "mean_token_accuracy": 0.22526508271694184,
      "num_tokens": 152637083.0,
      "step": 75325
    },
    {
      "entropy": 5.817083311080933,
      "epoch": 6.583063881849165,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016950159869289402,
      "loss": 5.2068,
      "mean_token_accuracy": 0.22449062764644623,
      "num_tokens": 152648486.0,
      "step": 75330
    },
    {
      "entropy": 5.7916725158691404,
      "epoch": 6.58350083020187,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016947407632429391,
      "loss": 5.0771,
      "mean_token_accuracy": 0.22196094840765,
      "num_tokens": 152658956.0,
      "step": 75335
    },
    {
      "entropy": 5.868988561630249,
      "epoch": 6.583937778554575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016944655597973636,
      "loss": 5.2287,
      "mean_token_accuracy": 0.229411543905735,
      "num_tokens": 152669753.0,
      "step": 75340
    },
    {
      "entropy": 5.885572385787964,
      "epoch": 6.58437472690728,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016941903765974912,
      "loss": 5.1185,
      "mean_token_accuracy": 0.2223971113562584,
      "num_tokens": 152678512.0,
      "step": 75345
    },
    {
      "entropy": 5.841171169281006,
      "epoch": 6.584811675259984,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016939152136486008,
      "loss": 5.1142,
      "mean_token_accuracy": 0.2281951680779457,
      "num_tokens": 152687875.0,
      "step": 75350
    },
    {
      "entropy": 5.864794540405273,
      "epoch": 6.585248623612689,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016936400709559695,
      "loss": 5.1553,
      "mean_token_accuracy": 0.22507307678461075,
      "num_tokens": 152697247.0,
      "step": 75355
    },
    {
      "entropy": 5.838809537887573,
      "epoch": 6.585685571965394,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016933649485248762,
      "loss": 5.1126,
      "mean_token_accuracy": 0.22777288407087326,
      "num_tokens": 152706895.0,
      "step": 75360
    },
    {
      "entropy": 5.783006000518799,
      "epoch": 6.586122520318098,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001693089846360597,
      "loss": 5.0638,
      "mean_token_accuracy": 0.23591352999210358,
      "num_tokens": 152716742.0,
      "step": 75365
    },
    {
      "entropy": 5.840488529205322,
      "epoch": 6.586559468670803,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016928147644684077,
      "loss": 5.1951,
      "mean_token_accuracy": 0.2194969728589058,
      "num_tokens": 152725578.0,
      "step": 75370
    },
    {
      "entropy": 5.8226593971252445,
      "epoch": 6.586996417023508,
      "grad_norm": 1.25,
      "learning_rate": 0.0001692539702853585,
      "loss": 5.1648,
      "mean_token_accuracy": 0.2294822797179222,
      "num_tokens": 152736644.0,
      "step": 75375
    },
    {
      "entropy": 5.772392082214355,
      "epoch": 6.587433365376213,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001692264661521405,
      "loss": 5.0038,
      "mean_token_accuracy": 0.23462138175964356,
      "num_tokens": 152746171.0,
      "step": 75380
    },
    {
      "entropy": 5.864650774002075,
      "epoch": 6.587870313728917,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016919896404771434,
      "loss": 5.1368,
      "mean_token_accuracy": 0.22396611273288727,
      "num_tokens": 152756239.0,
      "step": 75385
    },
    {
      "entropy": 5.859780883789062,
      "epoch": 6.588307262081622,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016917146397260743,
      "loss": 5.1602,
      "mean_token_accuracy": 0.22328702360391617,
      "num_tokens": 152766378.0,
      "step": 75390
    },
    {
      "entropy": 5.840550374984741,
      "epoch": 6.588744210434327,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016914396592734732,
      "loss": 5.2119,
      "mean_token_accuracy": 0.21746103316545487,
      "num_tokens": 152776409.0,
      "step": 75395
    },
    {
      "entropy": 5.765323829650879,
      "epoch": 6.5891811587870315,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016911646991246143,
      "loss": 4.9985,
      "mean_token_accuracy": 0.2384354963898659,
      "num_tokens": 152787122.0,
      "step": 75400
    },
    {
      "entropy": 5.855018568038941,
      "epoch": 6.589618107139736,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016908897592847707,
      "loss": 5.1311,
      "mean_token_accuracy": 0.22348955869674683,
      "num_tokens": 152797742.0,
      "step": 75405
    },
    {
      "entropy": 5.870284605026245,
      "epoch": 6.59005505549244,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016906148397592173,
      "loss": 5.1432,
      "mean_token_accuracy": 0.2223815679550171,
      "num_tokens": 152807989.0,
      "step": 75410
    },
    {
      "entropy": 5.834733390808106,
      "epoch": 6.590492003845146,
      "grad_norm": 1.25,
      "learning_rate": 0.00016903399405532267,
      "loss": 5.1578,
      "mean_token_accuracy": 0.2274997502565384,
      "num_tokens": 152818470.0,
      "step": 75415
    },
    {
      "entropy": 5.854251766204834,
      "epoch": 6.59092895219785,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016900650616720705,
      "loss": 5.1586,
      "mean_token_accuracy": 0.22558449804782868,
      "num_tokens": 152829142.0,
      "step": 75420
    },
    {
      "entropy": 5.878128099441528,
      "epoch": 6.591365900550555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001689790203121022,
      "loss": 5.1851,
      "mean_token_accuracy": 0.21769107282161712,
      "num_tokens": 152839299.0,
      "step": 75425
    },
    {
      "entropy": 5.8813127994537355,
      "epoch": 6.59180284890326,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001689515364905353,
      "loss": 5.0987,
      "mean_token_accuracy": 0.23214342892169954,
      "num_tokens": 152848481.0,
      "step": 75430
    },
    {
      "entropy": 5.857917928695679,
      "epoch": 6.5922397972559645,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016892405470303346,
      "loss": 5.1551,
      "mean_token_accuracy": 0.21929638385772704,
      "num_tokens": 152858762.0,
      "step": 75435
    },
    {
      "entropy": 5.781488370895386,
      "epoch": 6.592676745608669,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016889657495012384,
      "loss": 5.1169,
      "mean_token_accuracy": 0.23032900542020798,
      "num_tokens": 152869989.0,
      "step": 75440
    },
    {
      "entropy": 5.858526229858398,
      "epoch": 6.5931136939613735,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016886909723233358,
      "loss": 5.1983,
      "mean_token_accuracy": 0.22541554421186447,
      "num_tokens": 152880715.0,
      "step": 75445
    },
    {
      "entropy": 5.847015476226806,
      "epoch": 6.593550642314079,
      "grad_norm": 1.25,
      "learning_rate": 0.00016884162155018955,
      "loss": 5.1153,
      "mean_token_accuracy": 0.22542761713266374,
      "num_tokens": 152892691.0,
      "step": 75450
    },
    {
      "entropy": 5.842359304428101,
      "epoch": 6.593987590666783,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001688141479042189,
      "loss": 5.0858,
      "mean_token_accuracy": 0.23223237693309784,
      "num_tokens": 152903321.0,
      "step": 75455
    },
    {
      "entropy": 5.797276544570923,
      "epoch": 6.594424539019488,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001687866762949486,
      "loss": 5.0813,
      "mean_token_accuracy": 0.23109334856271743,
      "num_tokens": 152913871.0,
      "step": 75460
    },
    {
      "entropy": 5.82809362411499,
      "epoch": 6.594861487372192,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001687592067229054,
      "loss": 5.1312,
      "mean_token_accuracy": 0.22849484384059907,
      "num_tokens": 152924153.0,
      "step": 75465
    },
    {
      "entropy": 5.863054895401001,
      "epoch": 6.595298435724898,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016873173918861622,
      "loss": 5.187,
      "mean_token_accuracy": 0.22127790302038192,
      "num_tokens": 152934444.0,
      "step": 75470
    },
    {
      "entropy": 5.804449558258057,
      "epoch": 6.595735384077602,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016870427369260804,
      "loss": 5.0594,
      "mean_token_accuracy": 0.2366070941090584,
      "num_tokens": 152944258.0,
      "step": 75475
    },
    {
      "entropy": 5.843425369262695,
      "epoch": 6.5961723324303065,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001686768102354076,
      "loss": 5.1241,
      "mean_token_accuracy": 0.2314798966050148,
      "num_tokens": 152953775.0,
      "step": 75480
    },
    {
      "entropy": 5.946057224273682,
      "epoch": 6.596609280783012,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016864934881754168,
      "loss": 5.2362,
      "mean_token_accuracy": 0.21293796598911285,
      "num_tokens": 152963162.0,
      "step": 75485
    },
    {
      "entropy": 5.854734945297241,
      "epoch": 6.597046229135716,
      "grad_norm": 1.09375,
      "learning_rate": 0.00016862188943953695,
      "loss": 5.1131,
      "mean_token_accuracy": 0.22961823791265487,
      "num_tokens": 152973465.0,
      "step": 75490
    },
    {
      "entropy": 5.8009570121765135,
      "epoch": 6.597483177488421,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001685944321019201,
      "loss": 5.0807,
      "mean_token_accuracy": 0.23261951357126237,
      "num_tokens": 152983679.0,
      "step": 75495
    },
    {
      "entropy": 5.824099349975586,
      "epoch": 6.597920125841125,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016856697680521794,
      "loss": 5.1363,
      "mean_token_accuracy": 0.22918105870485306,
      "num_tokens": 152993541.0,
      "step": 75500
    },
    {
      "entropy": 5.88744740486145,
      "epoch": 6.598357074193831,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016853952354995677,
      "loss": 5.1185,
      "mean_token_accuracy": 0.22766793966293336,
      "num_tokens": 153003653.0,
      "step": 75505
    },
    {
      "entropy": 5.743768739700317,
      "epoch": 6.598794022546535,
      "grad_norm": 1.375,
      "learning_rate": 0.00016851207233666338,
      "loss": 5.0025,
      "mean_token_accuracy": 0.242754527926445,
      "num_tokens": 153012900.0,
      "step": 75510
    },
    {
      "entropy": 5.771023082733154,
      "epoch": 6.59923097089924,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016848462316586428,
      "loss": 5.1282,
      "mean_token_accuracy": 0.2231583908200264,
      "num_tokens": 153022797.0,
      "step": 75515
    },
    {
      "entropy": 5.8208984375,
      "epoch": 6.599667919251944,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016845717603808583,
      "loss": 5.1286,
      "mean_token_accuracy": 0.2283940613269806,
      "num_tokens": 153032091.0,
      "step": 75520
    },
    {
      "entropy": 5.870505619049072,
      "epoch": 6.600104867604649,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001684297309538546,
      "loss": 5.224,
      "mean_token_accuracy": 0.21124505549669265,
      "num_tokens": 153043178.0,
      "step": 75525
    },
    {
      "entropy": 5.879352664947509,
      "epoch": 6.600541815957354,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016840228791369698,
      "loss": 5.1281,
      "mean_token_accuracy": 0.22369484305381776,
      "num_tokens": 153052792.0,
      "step": 75530
    },
    {
      "entropy": 5.779606723785401,
      "epoch": 6.600978764310058,
      "grad_norm": 1.25,
      "learning_rate": 0.00016837484691813938,
      "loss": 5.0528,
      "mean_token_accuracy": 0.24264339357614517,
      "num_tokens": 153063778.0,
      "step": 75535
    },
    {
      "entropy": 5.815979909896851,
      "epoch": 6.601415712662764,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016834740796770802,
      "loss": 5.1397,
      "mean_token_accuracy": 0.23398369997739793,
      "num_tokens": 153073421.0,
      "step": 75540
    },
    {
      "entropy": 5.841549348831177,
      "epoch": 6.601852661015468,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016831997106292928,
      "loss": 5.0951,
      "mean_token_accuracy": 0.23279035687446595,
      "num_tokens": 153084114.0,
      "step": 75545
    },
    {
      "entropy": 5.889780330657959,
      "epoch": 6.602289609368173,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016829253620432948,
      "loss": 5.2348,
      "mean_token_accuracy": 0.2195084124803543,
      "num_tokens": 153095267.0,
      "step": 75550
    },
    {
      "entropy": 5.881812620162964,
      "epoch": 6.602726557720877,
      "grad_norm": 1.125,
      "learning_rate": 0.00016826510339243466,
      "loss": 5.1336,
      "mean_token_accuracy": 0.22098810970783234,
      "num_tokens": 153105155.0,
      "step": 75555
    },
    {
      "entropy": 5.792625856399536,
      "epoch": 6.6031635060735825,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016823767262777112,
      "loss": 5.0274,
      "mean_token_accuracy": 0.23408040702342986,
      "num_tokens": 153114435.0,
      "step": 75560
    },
    {
      "entropy": 5.819870185852051,
      "epoch": 6.603600454426287,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016821024391086494,
      "loss": 5.1662,
      "mean_token_accuracy": 0.22253812700510026,
      "num_tokens": 153124165.0,
      "step": 75565
    },
    {
      "entropy": 5.740944051742554,
      "epoch": 6.604037402778991,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016818281724224223,
      "loss": 5.0533,
      "mean_token_accuracy": 0.23345385640859603,
      "num_tokens": 153134144.0,
      "step": 75570
    },
    {
      "entropy": 5.828731203079224,
      "epoch": 6.604474351131696,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016815539262242908,
      "loss": 5.0866,
      "mean_token_accuracy": 0.23789045810699463,
      "num_tokens": 153145203.0,
      "step": 75575
    },
    {
      "entropy": 5.890122509002685,
      "epoch": 6.604911299484401,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016812797005195146,
      "loss": 5.1724,
      "mean_token_accuracy": 0.21901921331882476,
      "num_tokens": 153155623.0,
      "step": 75580
    },
    {
      "entropy": 5.771741151809692,
      "epoch": 6.605348247837106,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016810054953133536,
      "loss": 5.0869,
      "mean_token_accuracy": 0.23558472990989685,
      "num_tokens": 153166289.0,
      "step": 75585
    },
    {
      "entropy": 5.77825288772583,
      "epoch": 6.60578519618981,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016807313106110677,
      "loss": 5.1232,
      "mean_token_accuracy": 0.23170148581266403,
      "num_tokens": 153176693.0,
      "step": 75590
    },
    {
      "entropy": 5.821775007247925,
      "epoch": 6.6062221445425156,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016804571464179162,
      "loss": 5.1466,
      "mean_token_accuracy": 0.22438505440950393,
      "num_tokens": 153187146.0,
      "step": 75595
    },
    {
      "entropy": 5.84680142402649,
      "epoch": 6.60665909289522,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016801830027391563,
      "loss": 5.1058,
      "mean_token_accuracy": 0.2257787197828293,
      "num_tokens": 153196127.0,
      "step": 75600
    },
    {
      "entropy": 5.759702539443969,
      "epoch": 6.6070960412479245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016799088795800467,
      "loss": 5.119,
      "mean_token_accuracy": 0.23579706400632858,
      "num_tokens": 153206166.0,
      "step": 75605
    },
    {
      "entropy": 5.867849588394165,
      "epoch": 6.607532989600629,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001679634776945846,
      "loss": 5.1632,
      "mean_token_accuracy": 0.22207497507333757,
      "num_tokens": 153215961.0,
      "step": 75610
    },
    {
      "entropy": 5.871264266967773,
      "epoch": 6.607969937953333,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001679360694841811,
      "loss": 5.1956,
      "mean_token_accuracy": 0.21976365000009537,
      "num_tokens": 153226685.0,
      "step": 75615
    },
    {
      "entropy": 5.8598109722137455,
      "epoch": 6.608406886306039,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016790866332731986,
      "loss": 5.121,
      "mean_token_accuracy": 0.22935534417629241,
      "num_tokens": 153236808.0,
      "step": 75620
    },
    {
      "entropy": 5.831113338470459,
      "epoch": 6.608843834658743,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016788125922452657,
      "loss": 5.1034,
      "mean_token_accuracy": 0.2288370743393898,
      "num_tokens": 153247719.0,
      "step": 75625
    },
    {
      "entropy": 5.811720085144043,
      "epoch": 6.609280783011448,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001678538571763269,
      "loss": 5.0636,
      "mean_token_accuracy": 0.2320365458726883,
      "num_tokens": 153258202.0,
      "step": 75630
    },
    {
      "entropy": 5.870866012573242,
      "epoch": 6.609717731364153,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001678264571832464,
      "loss": 5.2075,
      "mean_token_accuracy": 0.2152174010872841,
      "num_tokens": 153268084.0,
      "step": 75635
    },
    {
      "entropy": 5.78386869430542,
      "epoch": 6.6101546797168576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016779905924581057,
      "loss": 5.1102,
      "mean_token_accuracy": 0.23145315796136856,
      "num_tokens": 153278348.0,
      "step": 75640
    },
    {
      "entropy": 5.833589506149292,
      "epoch": 6.610591628069562,
      "grad_norm": 1.25,
      "learning_rate": 0.00016777166336454503,
      "loss": 5.0991,
      "mean_token_accuracy": 0.2330515429377556,
      "num_tokens": 153289332.0,
      "step": 75645
    },
    {
      "entropy": 5.892413139343262,
      "epoch": 6.611028576422267,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016774426953997514,
      "loss": 5.215,
      "mean_token_accuracy": 0.2184402599930763,
      "num_tokens": 153299103.0,
      "step": 75650
    },
    {
      "entropy": 5.824222183227539,
      "epoch": 6.611465524774972,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016771687777262628,
      "loss": 5.1104,
      "mean_token_accuracy": 0.22406276166439057,
      "num_tokens": 153308807.0,
      "step": 75655
    },
    {
      "entropy": 5.87707200050354,
      "epoch": 6.611902473127676,
      "grad_norm": 1.21875,
      "learning_rate": 0.000167689488063024,
      "loss": 5.0887,
      "mean_token_accuracy": 0.2269663080573082,
      "num_tokens": 153318401.0,
      "step": 75660
    },
    {
      "entropy": 5.855170965194702,
      "epoch": 6.612339421480381,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016766210041169356,
      "loss": 5.1831,
      "mean_token_accuracy": 0.22684279978275299,
      "num_tokens": 153329016.0,
      "step": 75665
    },
    {
      "entropy": 5.838120841979981,
      "epoch": 6.612776369833085,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016763471481916025,
      "loss": 5.1051,
      "mean_token_accuracy": 0.22401154041290283,
      "num_tokens": 153338388.0,
      "step": 75670
    },
    {
      "entropy": 5.811215829849243,
      "epoch": 6.613213318185791,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016760733128594934,
      "loss": 5.1188,
      "mean_token_accuracy": 0.2328306704759598,
      "num_tokens": 153348884.0,
      "step": 75675
    },
    {
      "entropy": 5.847482538223266,
      "epoch": 6.613650266538495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016757994981258614,
      "loss": 5.12,
      "mean_token_accuracy": 0.2278086945414543,
      "num_tokens": 153358304.0,
      "step": 75680
    },
    {
      "entropy": 5.8340599060058596,
      "epoch": 6.6140872148911996,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016755257039959579,
      "loss": 5.0968,
      "mean_token_accuracy": 0.22842076420783997,
      "num_tokens": 153368400.0,
      "step": 75685
    },
    {
      "entropy": 5.851229000091553,
      "epoch": 6.614524163243905,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001675251930475034,
      "loss": 5.1352,
      "mean_token_accuracy": 0.22507814168930054,
      "num_tokens": 153378189.0,
      "step": 75690
    },
    {
      "entropy": 5.844403600692749,
      "epoch": 6.614961111596609,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016749781775683427,
      "loss": 5.1783,
      "mean_token_accuracy": 0.22365273535251617,
      "num_tokens": 153389318.0,
      "step": 75695
    },
    {
      "entropy": 5.845790243148803,
      "epoch": 6.615398059949314,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016747044452811315,
      "loss": 5.1179,
      "mean_token_accuracy": 0.23110420554876326,
      "num_tokens": 153399561.0,
      "step": 75700
    },
    {
      "entropy": 5.873381519317627,
      "epoch": 6.615835008302018,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016744307336186533,
      "loss": 5.1398,
      "mean_token_accuracy": 0.22553424835205077,
      "num_tokens": 153409356.0,
      "step": 75705
    },
    {
      "entropy": 5.830557823181152,
      "epoch": 6.616271956654724,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016741570425861572,
      "loss": 5.0747,
      "mean_token_accuracy": 0.2283741280436516,
      "num_tokens": 153419180.0,
      "step": 75710
    },
    {
      "entropy": 5.812517213821411,
      "epoch": 6.616708905007428,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001673883372188892,
      "loss": 5.0911,
      "mean_token_accuracy": 0.23461777120828628,
      "num_tokens": 153429311.0,
      "step": 75715
    },
    {
      "entropy": 5.856284475326538,
      "epoch": 6.617145853360133,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016736097224321085,
      "loss": 5.1914,
      "mean_token_accuracy": 0.22305579632520675,
      "num_tokens": 153439074.0,
      "step": 75720
    },
    {
      "entropy": 5.8102545738220215,
      "epoch": 6.617582801712837,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016733360933210542,
      "loss": 5.1633,
      "mean_token_accuracy": 0.2239438161253929,
      "num_tokens": 153449221.0,
      "step": 75725
    },
    {
      "entropy": 5.811808109283447,
      "epoch": 6.618019750065542,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016730624848609776,
      "loss": 5.0569,
      "mean_token_accuracy": 0.23296922594308853,
      "num_tokens": 153459287.0,
      "step": 75730
    },
    {
      "entropy": 5.848714208602905,
      "epoch": 6.618456698418247,
      "grad_norm": 1.125,
      "learning_rate": 0.0001672788897057127,
      "loss": 5.1756,
      "mean_token_accuracy": 0.22655766308307648,
      "num_tokens": 153470245.0,
      "step": 75735
    },
    {
      "entropy": 5.881117296218872,
      "epoch": 6.618893646770951,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016725153299147505,
      "loss": 5.1307,
      "mean_token_accuracy": 0.2230602577328682,
      "num_tokens": 153479815.0,
      "step": 75740
    },
    {
      "entropy": 5.822728824615479,
      "epoch": 6.619330595123657,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001672241783439094,
      "loss": 5.1197,
      "mean_token_accuracy": 0.2309379607439041,
      "num_tokens": 153489550.0,
      "step": 75745
    },
    {
      "entropy": 5.898336458206177,
      "epoch": 6.619767543476361,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016719682576354046,
      "loss": 5.1522,
      "mean_token_accuracy": 0.22537255883216858,
      "num_tokens": 153498804.0,
      "step": 75750
    },
    {
      "entropy": 5.845671796798706,
      "epoch": 6.620204491829066,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016716947525089292,
      "loss": 5.1386,
      "mean_token_accuracy": 0.22010375559329987,
      "num_tokens": 153508779.0,
      "step": 75755
    },
    {
      "entropy": 5.865339040756226,
      "epoch": 6.62064144018177,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016714212680649129,
      "loss": 5.2246,
      "mean_token_accuracy": 0.22495743930339812,
      "num_tokens": 153520083.0,
      "step": 75760
    },
    {
      "entropy": 5.848446893692016,
      "epoch": 6.6210783885344755,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016711478043086026,
      "loss": 5.104,
      "mean_token_accuracy": 0.22869090735912323,
      "num_tokens": 153530366.0,
      "step": 75765
    },
    {
      "entropy": 5.867528915405273,
      "epoch": 6.62151533688718,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016708743612452423,
      "loss": 5.0486,
      "mean_token_accuracy": 0.23394273817539216,
      "num_tokens": 153539977.0,
      "step": 75770
    },
    {
      "entropy": 5.878913164138794,
      "epoch": 6.621952285239884,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001670600938880077,
      "loss": 5.2226,
      "mean_token_accuracy": 0.21964834481477738,
      "num_tokens": 153550099.0,
      "step": 75775
    },
    {
      "entropy": 5.866553211212159,
      "epoch": 6.622389233592589,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016703275372183517,
      "loss": 5.1588,
      "mean_token_accuracy": 0.2241806760430336,
      "num_tokens": 153559225.0,
      "step": 75780
    },
    {
      "entropy": 5.791719675064087,
      "epoch": 6.622826181945294,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016700541562653096,
      "loss": 5.0218,
      "mean_token_accuracy": 0.23537472635507584,
      "num_tokens": 153568120.0,
      "step": 75785
    },
    {
      "entropy": 5.853184223175049,
      "epoch": 6.623263130297999,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016697807960261958,
      "loss": 5.1174,
      "mean_token_accuracy": 0.23194495737552642,
      "num_tokens": 153578449.0,
      "step": 75790
    },
    {
      "entropy": 5.840583944320679,
      "epoch": 6.623700078650703,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001669507456506251,
      "loss": 5.1284,
      "mean_token_accuracy": 0.22840810418128968,
      "num_tokens": 153589367.0,
      "step": 75795
    },
    {
      "entropy": 5.795636796951294,
      "epoch": 6.624137027003409,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000166923413771072,
      "loss": 5.124,
      "mean_token_accuracy": 0.2304065078496933,
      "num_tokens": 153600079.0,
      "step": 75800
    },
    {
      "entropy": 5.863008785247803,
      "epoch": 6.624573975356113,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016689608396448444,
      "loss": 5.1429,
      "mean_token_accuracy": 0.2279445931315422,
      "num_tokens": 153609584.0,
      "step": 75805
    },
    {
      "entropy": 5.842927408218384,
      "epoch": 6.6250109237088175,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016686875623138665,
      "loss": 5.0857,
      "mean_token_accuracy": 0.23437412977218627,
      "num_tokens": 153619330.0,
      "step": 75810
    },
    {
      "entropy": 5.860196733474732,
      "epoch": 6.625447872061522,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016684143057230273,
      "loss": 5.1366,
      "mean_token_accuracy": 0.2291480004787445,
      "num_tokens": 153628659.0,
      "step": 75815
    },
    {
      "entropy": 5.817116546630859,
      "epoch": 6.625884820414227,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016681410698775686,
      "loss": 5.1547,
      "mean_token_accuracy": 0.22398795932531357,
      "num_tokens": 153638432.0,
      "step": 75820
    },
    {
      "entropy": 5.852712869644165,
      "epoch": 6.626321768766932,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016678678547827304,
      "loss": 5.1859,
      "mean_token_accuracy": 0.22077534347772598,
      "num_tokens": 153648441.0,
      "step": 75825
    },
    {
      "entropy": 5.843874502182007,
      "epoch": 6.626758717119636,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016675946604437553,
      "loss": 5.1878,
      "mean_token_accuracy": 0.22367178052663803,
      "num_tokens": 153658620.0,
      "step": 75830
    },
    {
      "entropy": 5.90068359375,
      "epoch": 6.627195665472341,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016673214868658808,
      "loss": 5.1917,
      "mean_token_accuracy": 0.21678659319877625,
      "num_tokens": 153668403.0,
      "step": 75835
    },
    {
      "entropy": 5.869166040420533,
      "epoch": 6.627632613825046,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001667048334054348,
      "loss": 5.1808,
      "mean_token_accuracy": 0.22436856627464294,
      "num_tokens": 153678588.0,
      "step": 75840
    },
    {
      "entropy": 5.816032791137696,
      "epoch": 6.628069562177751,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001666775202014395,
      "loss": 5.1124,
      "mean_token_accuracy": 0.22761025428771972,
      "num_tokens": 153688093.0,
      "step": 75845
    },
    {
      "entropy": 5.799421167373657,
      "epoch": 6.628506510530455,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016665020907512617,
      "loss": 5.1032,
      "mean_token_accuracy": 0.23075727224349976,
      "num_tokens": 153697640.0,
      "step": 75850
    },
    {
      "entropy": 5.832125949859619,
      "epoch": 6.62894345888316,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001666229000270186,
      "loss": 5.1244,
      "mean_token_accuracy": 0.23020346462726593,
      "num_tokens": 153707198.0,
      "step": 75855
    },
    {
      "entropy": 5.877263116836548,
      "epoch": 6.629380407235865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001665955930576406,
      "loss": 5.2096,
      "mean_token_accuracy": 0.2175368160009384,
      "num_tokens": 153718097.0,
      "step": 75860
    },
    {
      "entropy": 5.8881494998931885,
      "epoch": 6.629817355588569,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001665682881675159,
      "loss": 5.1633,
      "mean_token_accuracy": 0.2181318700313568,
      "num_tokens": 153728892.0,
      "step": 75865
    },
    {
      "entropy": 5.831302118301392,
      "epoch": 6.630254303941274,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016654098535716827,
      "loss": 5.1498,
      "mean_token_accuracy": 0.22589985132217408,
      "num_tokens": 153739313.0,
      "step": 75870
    },
    {
      "entropy": 5.884886884689331,
      "epoch": 6.630691252293979,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016651368462712146,
      "loss": 5.153,
      "mean_token_accuracy": 0.21861194521188737,
      "num_tokens": 153749168.0,
      "step": 75875
    },
    {
      "entropy": 5.840649557113648,
      "epoch": 6.631128200646684,
      "grad_norm": 1.125,
      "learning_rate": 0.00016648638597789895,
      "loss": 5.1236,
      "mean_token_accuracy": 0.2269766852259636,
      "num_tokens": 153759233.0,
      "step": 75880
    },
    {
      "entropy": 5.808142423629761,
      "epoch": 6.631565148999388,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016645908941002435,
      "loss": 5.1322,
      "mean_token_accuracy": 0.23211078643798827,
      "num_tokens": 153769291.0,
      "step": 75885
    },
    {
      "entropy": 5.7906230926513675,
      "epoch": 6.632002097352093,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016643179492402134,
      "loss": 5.0135,
      "mean_token_accuracy": 0.23929688781499864,
      "num_tokens": 153778787.0,
      "step": 75890
    },
    {
      "entropy": 5.8703361511230465,
      "epoch": 6.632439045704798,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001664045025204134,
      "loss": 5.1992,
      "mean_token_accuracy": 0.22406268119812012,
      "num_tokens": 153789756.0,
      "step": 75895
    },
    {
      "entropy": 5.841914129257202,
      "epoch": 6.632875994057502,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016637721219972402,
      "loss": 5.1266,
      "mean_token_accuracy": 0.2215671494603157,
      "num_tokens": 153800110.0,
      "step": 75900
    },
    {
      "entropy": 5.875002908706665,
      "epoch": 6.633312942410207,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016634992396247656,
      "loss": 5.2172,
      "mean_token_accuracy": 0.22211963683366776,
      "num_tokens": 153811029.0,
      "step": 75905
    },
    {
      "entropy": 5.85824408531189,
      "epoch": 6.633749890762912,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016632263780919454,
      "loss": 5.1742,
      "mean_token_accuracy": 0.21920957416296005,
      "num_tokens": 153821786.0,
      "step": 75910
    },
    {
      "entropy": 5.873841667175293,
      "epoch": 6.634186839115617,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001662953537404013,
      "loss": 5.1171,
      "mean_token_accuracy": 0.22630738466978073,
      "num_tokens": 153832536.0,
      "step": 75915
    },
    {
      "entropy": 5.828742933273316,
      "epoch": 6.634623787468321,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016626807175662006,
      "loss": 5.0845,
      "mean_token_accuracy": 0.22935486435890198,
      "num_tokens": 153842549.0,
      "step": 75920
    },
    {
      "entropy": 5.919301319122314,
      "epoch": 6.635060735821026,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016624079185837432,
      "loss": 5.2281,
      "mean_token_accuracy": 0.21296763122081758,
      "num_tokens": 153852700.0,
      "step": 75925
    },
    {
      "entropy": 5.818543767929077,
      "epoch": 6.635497684173731,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016621351404618706,
      "loss": 5.0748,
      "mean_token_accuracy": 0.2365559458732605,
      "num_tokens": 153864004.0,
      "step": 75930
    },
    {
      "entropy": 5.831252241134644,
      "epoch": 6.6359346325264355,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016618623832058166,
      "loss": 5.0269,
      "mean_token_accuracy": 0.24157373756170272,
      "num_tokens": 153873733.0,
      "step": 75935
    },
    {
      "entropy": 5.789722537994384,
      "epoch": 6.63637158087914,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016615896468208117,
      "loss": 5.1177,
      "mean_token_accuracy": 0.22497524619102477,
      "num_tokens": 153884643.0,
      "step": 75940
    },
    {
      "entropy": 5.858077669143677,
      "epoch": 6.636808529231844,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016613169313120879,
      "loss": 5.1766,
      "mean_token_accuracy": 0.22322926819324493,
      "num_tokens": 153894809.0,
      "step": 75945
    },
    {
      "entropy": 5.8610511302948,
      "epoch": 6.63724547758455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001661044236684876,
      "loss": 5.1907,
      "mean_token_accuracy": 0.22478016316890717,
      "num_tokens": 153905558.0,
      "step": 75950
    },
    {
      "entropy": 5.8540068626403805,
      "epoch": 6.637682425937254,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001660771562944406,
      "loss": 5.1096,
      "mean_token_accuracy": 0.23208070546388626,
      "num_tokens": 153915744.0,
      "step": 75955
    },
    {
      "entropy": 5.804800415039063,
      "epoch": 6.638119374289959,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001660498910095908,
      "loss": 5.1066,
      "mean_token_accuracy": 0.23255475759506225,
      "num_tokens": 153926914.0,
      "step": 75960
    },
    {
      "entropy": 5.760906028747558,
      "epoch": 6.638556322642664,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016602262781446122,
      "loss": 5.0919,
      "mean_token_accuracy": 0.23476502299308777,
      "num_tokens": 153936177.0,
      "step": 75965
    },
    {
      "entropy": 5.871738290786743,
      "epoch": 6.6389932709953685,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001659953667095747,
      "loss": 5.2053,
      "mean_token_accuracy": 0.2212546780705452,
      "num_tokens": 153945645.0,
      "step": 75970
    },
    {
      "entropy": 5.858593702316284,
      "epoch": 6.639430219348073,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001659681076954543,
      "loss": 5.112,
      "mean_token_accuracy": 0.22753990888595582,
      "num_tokens": 153955564.0,
      "step": 75975
    },
    {
      "entropy": 5.825626087188721,
      "epoch": 6.6398671677007775,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001659408507726226,
      "loss": 5.0849,
      "mean_token_accuracy": 0.23002519458532333,
      "num_tokens": 153964837.0,
      "step": 75980
    },
    {
      "entropy": 5.818421173095703,
      "epoch": 6.640304116053483,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016591359594160255,
      "loss": 5.0427,
      "mean_token_accuracy": 0.2324089899659157,
      "num_tokens": 153975937.0,
      "step": 75985
    },
    {
      "entropy": 5.806035757064819,
      "epoch": 6.640741064406187,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001658863432029169,
      "loss": 5.145,
      "mean_token_accuracy": 0.2307773619890213,
      "num_tokens": 153987318.0,
      "step": 75990
    },
    {
      "entropy": 5.805276584625244,
      "epoch": 6.641178012758892,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016585909255708832,
      "loss": 5.134,
      "mean_token_accuracy": 0.2245781421661377,
      "num_tokens": 153998609.0,
      "step": 75995
    },
    {
      "entropy": 5.808856201171875,
      "epoch": 6.641614961111596,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016583184400463958,
      "loss": 5.1276,
      "mean_token_accuracy": 0.23097978979349137,
      "num_tokens": 154008831.0,
      "step": 76000
    },
    {
      "entropy": 5.849027013778686,
      "epoch": 6.642051909464302,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016580459754609323,
      "loss": 5.1284,
      "mean_token_accuracy": 0.22416057288646699,
      "num_tokens": 154018699.0,
      "step": 76005
    },
    {
      "entropy": 5.814638996124268,
      "epoch": 6.642488857817006,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016577735318197196,
      "loss": 5.1252,
      "mean_token_accuracy": 0.22935560643672942,
      "num_tokens": 154029672.0,
      "step": 76010
    },
    {
      "entropy": 5.83831672668457,
      "epoch": 6.6429258061697105,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016575011091279825,
      "loss": 5.1442,
      "mean_token_accuracy": 0.23154085576534272,
      "num_tokens": 154040076.0,
      "step": 76015
    },
    {
      "entropy": 5.857762384414673,
      "epoch": 6.643362754522416,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016572287073909474,
      "loss": 5.1076,
      "mean_token_accuracy": 0.2333894744515419,
      "num_tokens": 154050951.0,
      "step": 76020
    },
    {
      "entropy": 5.853385305404663,
      "epoch": 6.64379970287512,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001656956326613838,
      "loss": 5.1232,
      "mean_token_accuracy": 0.23000362068414687,
      "num_tokens": 154061035.0,
      "step": 76025
    },
    {
      "entropy": 5.876396512985229,
      "epoch": 6.644236651227825,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016566839668018786,
      "loss": 5.1694,
      "mean_token_accuracy": 0.21980541348457336,
      "num_tokens": 154071717.0,
      "step": 76030
    },
    {
      "entropy": 5.825916290283203,
      "epoch": 6.644673599580529,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016564116279602938,
      "loss": 5.1342,
      "mean_token_accuracy": 0.23652184754610062,
      "num_tokens": 154081945.0,
      "step": 76035
    },
    {
      "entropy": 5.899185371398926,
      "epoch": 6.645110547933235,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016561393100943068,
      "loss": 5.256,
      "mean_token_accuracy": 0.21387521624565126,
      "num_tokens": 154092086.0,
      "step": 76040
    },
    {
      "entropy": 5.894190311431885,
      "epoch": 6.645547496285939,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016558670132091409,
      "loss": 5.1248,
      "mean_token_accuracy": 0.2299377754330635,
      "num_tokens": 154102074.0,
      "step": 76045
    },
    {
      "entropy": 5.902597427368164,
      "epoch": 6.645984444638644,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001655594737310019,
      "loss": 5.1508,
      "mean_token_accuracy": 0.22721511721611024,
      "num_tokens": 154112563.0,
      "step": 76050
    },
    {
      "entropy": 5.801961898803711,
      "epoch": 6.646421392991348,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001655322482402164,
      "loss": 5.0994,
      "mean_token_accuracy": 0.23508742898702623,
      "num_tokens": 154122825.0,
      "step": 76055
    },
    {
      "entropy": 5.7677394390106205,
      "epoch": 6.646858341344053,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001655050248490797,
      "loss": 5.074,
      "mean_token_accuracy": 0.2324768826365471,
      "num_tokens": 154132705.0,
      "step": 76060
    },
    {
      "entropy": 5.835625076293946,
      "epoch": 6.647295289696758,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016547780355811398,
      "loss": 5.1563,
      "mean_token_accuracy": 0.22273299992084503,
      "num_tokens": 154142572.0,
      "step": 76065
    },
    {
      "entropy": 5.820410108566284,
      "epoch": 6.647732238049462,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016545058436784153,
      "loss": 5.0777,
      "mean_token_accuracy": 0.23421822786331176,
      "num_tokens": 154151725.0,
      "step": 76070
    },
    {
      "entropy": 5.862186241149902,
      "epoch": 6.648169186402168,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001654233672787841,
      "loss": 5.1709,
      "mean_token_accuracy": 0.22648879885673523,
      "num_tokens": 154161730.0,
      "step": 76075
    },
    {
      "entropy": 5.882863140106201,
      "epoch": 6.648606134754872,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000165396152291464,
      "loss": 5.1886,
      "mean_token_accuracy": 0.21721122711896895,
      "num_tokens": 154172056.0,
      "step": 76080
    },
    {
      "entropy": 5.767967557907104,
      "epoch": 6.649043083107577,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016536893940640308,
      "loss": 5.0411,
      "mean_token_accuracy": 0.2370420962572098,
      "num_tokens": 154182650.0,
      "step": 76085
    },
    {
      "entropy": 5.81458740234375,
      "epoch": 6.649480031460281,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016534172862412333,
      "loss": 5.1009,
      "mean_token_accuracy": 0.2352811187505722,
      "num_tokens": 154193014.0,
      "step": 76090
    },
    {
      "entropy": 5.792021226882935,
      "epoch": 6.6499169798129865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016531451994514668,
      "loss": 5.066,
      "mean_token_accuracy": 0.23294543474912643,
      "num_tokens": 154203044.0,
      "step": 76095
    },
    {
      "entropy": 5.865421915054322,
      "epoch": 6.650353928165691,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016528731336999504,
      "loss": 5.1244,
      "mean_token_accuracy": 0.22186523526906968,
      "num_tokens": 154212344.0,
      "step": 76100
    },
    {
      "entropy": 5.844140863418579,
      "epoch": 6.650790876518395,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001652601088991902,
      "loss": 5.2354,
      "mean_token_accuracy": 0.2168536052107811,
      "num_tokens": 154223972.0,
      "step": 76105
    },
    {
      "entropy": 5.7363814353942875,
      "epoch": 6.6512278248711,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000165232906533254,
      "loss": 5.0228,
      "mean_token_accuracy": 0.24270727783441542,
      "num_tokens": 154235239.0,
      "step": 76110
    },
    {
      "entropy": 5.842800855636597,
      "epoch": 6.651664773223805,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016520570627270815,
      "loss": 5.1475,
      "mean_token_accuracy": 0.21714411824941635,
      "num_tokens": 154245098.0,
      "step": 76115
    },
    {
      "entropy": 5.865471601486206,
      "epoch": 6.65210172157651,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001651785081180745,
      "loss": 5.0784,
      "mean_token_accuracy": 0.23177859038114548,
      "num_tokens": 154256552.0,
      "step": 76120
    },
    {
      "entropy": 5.880640935897827,
      "epoch": 6.652538669929214,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001651513120698745,
      "loss": 5.1385,
      "mean_token_accuracy": 0.23197562396526336,
      "num_tokens": 154268294.0,
      "step": 76125
    },
    {
      "entropy": 5.855361700057983,
      "epoch": 6.6529756182819195,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016512411812862987,
      "loss": 5.0576,
      "mean_token_accuracy": 0.23250431269407273,
      "num_tokens": 154279120.0,
      "step": 76130
    },
    {
      "entropy": 5.776500511169433,
      "epoch": 6.653412566634624,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016509692629486229,
      "loss": 4.9956,
      "mean_token_accuracy": 0.2388294205069542,
      "num_tokens": 154288286.0,
      "step": 76135
    },
    {
      "entropy": 5.720143508911133,
      "epoch": 6.6538495149873285,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016506973656909318,
      "loss": 5.0934,
      "mean_token_accuracy": 0.23739144802093506,
      "num_tokens": 154299346.0,
      "step": 76140
    },
    {
      "entropy": 5.875834989547729,
      "epoch": 6.654286463340033,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016504254895184417,
      "loss": 5.1764,
      "mean_token_accuracy": 0.22293404638767242,
      "num_tokens": 154309505.0,
      "step": 76145
    },
    {
      "entropy": 5.881800985336303,
      "epoch": 6.654723411692738,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001650153634436367,
      "loss": 5.0796,
      "mean_token_accuracy": 0.2313835456967354,
      "num_tokens": 154318746.0,
      "step": 76150
    },
    {
      "entropy": 5.876924705505371,
      "epoch": 6.655160360045443,
      "grad_norm": 1.25,
      "learning_rate": 0.00016498818004499213,
      "loss": 5.181,
      "mean_token_accuracy": 0.2175787314772606,
      "num_tokens": 154329772.0,
      "step": 76155
    },
    {
      "entropy": 5.751776170730591,
      "epoch": 6.655597308398147,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016496099875643196,
      "loss": 5.0515,
      "mean_token_accuracy": 0.24028928875923156,
      "num_tokens": 154338890.0,
      "step": 76160
    },
    {
      "entropy": 5.852940607070923,
      "epoch": 6.656034256750852,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016493381957847751,
      "loss": 5.1105,
      "mean_token_accuracy": 0.22926846891641617,
      "num_tokens": 154349716.0,
      "step": 76165
    },
    {
      "entropy": 5.770843362808227,
      "epoch": 6.656471205103557,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001649066425116501,
      "loss": 5.0279,
      "mean_token_accuracy": 0.22855314910411834,
      "num_tokens": 154359796.0,
      "step": 76170
    },
    {
      "entropy": 5.850842905044556,
      "epoch": 6.6569081534562615,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016487946755647093,
      "loss": 5.1244,
      "mean_token_accuracy": 0.22290916442871095,
      "num_tokens": 154370476.0,
      "step": 76175
    },
    {
      "entropy": 5.882400703430176,
      "epoch": 6.657345101808966,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001648522947134612,
      "loss": 5.2244,
      "mean_token_accuracy": 0.22142947018146514,
      "num_tokens": 154380376.0,
      "step": 76180
    },
    {
      "entropy": 5.813851308822632,
      "epoch": 6.657782050161671,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016482512398314226,
      "loss": 5.0894,
      "mean_token_accuracy": 0.23478900045156478,
      "num_tokens": 154389909.0,
      "step": 76185
    },
    {
      "entropy": 5.808371686935425,
      "epoch": 6.658218998514376,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016479795536603514,
      "loss": 5.0786,
      "mean_token_accuracy": 0.2336548835039139,
      "num_tokens": 154399523.0,
      "step": 76190
    },
    {
      "entropy": 5.925292730331421,
      "epoch": 6.65865594686708,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000164770788862661,
      "loss": 5.2845,
      "mean_token_accuracy": 0.21191813796758652,
      "num_tokens": 154409063.0,
      "step": 76195
    },
    {
      "entropy": 5.873245906829834,
      "epoch": 6.659092895219785,
      "grad_norm": 1.09375,
      "learning_rate": 0.00016474362447354085,
      "loss": 5.169,
      "mean_token_accuracy": 0.22334379553794861,
      "num_tokens": 154420351.0,
      "step": 76200
    },
    {
      "entropy": 5.94016547203064,
      "epoch": 6.65952984357249,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016471646219919588,
      "loss": 5.2526,
      "mean_token_accuracy": 0.21889689415693284,
      "num_tokens": 154431345.0,
      "step": 76205
    },
    {
      "entropy": 5.914586448669434,
      "epoch": 6.659966791925195,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016468930204014686,
      "loss": 5.1921,
      "mean_token_accuracy": 0.2256295293569565,
      "num_tokens": 154442129.0,
      "step": 76210
    },
    {
      "entropy": 5.873364400863648,
      "epoch": 6.660403740277899,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016466214399691486,
      "loss": 5.1435,
      "mean_token_accuracy": 0.2295995369553566,
      "num_tokens": 154451710.0,
      "step": 76215
    },
    {
      "entropy": 5.868114614486695,
      "epoch": 6.6608406886306035,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001646349880700207,
      "loss": 5.1649,
      "mean_token_accuracy": 0.22978851050138474,
      "num_tokens": 154462045.0,
      "step": 76220
    },
    {
      "entropy": 5.887681484222412,
      "epoch": 6.661277636983309,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016460783425998533,
      "loss": 5.1698,
      "mean_token_accuracy": 0.2216881349682808,
      "num_tokens": 154471428.0,
      "step": 76225
    },
    {
      "entropy": 5.903207397460937,
      "epoch": 6.661714585336013,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016458068256732956,
      "loss": 5.1955,
      "mean_token_accuracy": 0.2242863431572914,
      "num_tokens": 154481219.0,
      "step": 76230
    },
    {
      "entropy": 5.842860221862793,
      "epoch": 6.662151533688718,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016455353299257413,
      "loss": 5.0344,
      "mean_token_accuracy": 0.24020253419876098,
      "num_tokens": 154490592.0,
      "step": 76235
    },
    {
      "entropy": 5.888642454147339,
      "epoch": 6.662588482041423,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016452638553623983,
      "loss": 5.1818,
      "mean_token_accuracy": 0.21473369151353836,
      "num_tokens": 154500124.0,
      "step": 76240
    },
    {
      "entropy": 5.851993465423584,
      "epoch": 6.663025430394128,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016449924019884733,
      "loss": 5.1328,
      "mean_token_accuracy": 0.2303017944097519,
      "num_tokens": 154510248.0,
      "step": 76245
    },
    {
      "entropy": 5.821728277206421,
      "epoch": 6.663462378746832,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016447209698091728,
      "loss": 5.0533,
      "mean_token_accuracy": 0.2336779624223709,
      "num_tokens": 154519934.0,
      "step": 76250
    },
    {
      "entropy": 5.842702865600586,
      "epoch": 6.663899327099537,
      "grad_norm": 1.25,
      "learning_rate": 0.00016444495588297043,
      "loss": 5.0874,
      "mean_token_accuracy": 0.22858930379152298,
      "num_tokens": 154529802.0,
      "step": 76255
    },
    {
      "entropy": 5.854711675643921,
      "epoch": 6.664336275452241,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001644178169055272,
      "loss": 5.173,
      "mean_token_accuracy": 0.22362025529146196,
      "num_tokens": 154540194.0,
      "step": 76260
    },
    {
      "entropy": 5.889368391036987,
      "epoch": 6.664773223804946,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016439068004910816,
      "loss": 5.1672,
      "mean_token_accuracy": 0.21985167264938354,
      "num_tokens": 154549892.0,
      "step": 76265
    },
    {
      "entropy": 5.885042476654053,
      "epoch": 6.665210172157651,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016436354531423382,
      "loss": 5.1978,
      "mean_token_accuracy": 0.22497051507234572,
      "num_tokens": 154560359.0,
      "step": 76270
    },
    {
      "entropy": 5.817539119720459,
      "epoch": 6.665647120510355,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001643364127014247,
      "loss": 5.118,
      "mean_token_accuracy": 0.2312390238046646,
      "num_tokens": 154570563.0,
      "step": 76275
    },
    {
      "entropy": 5.922878932952881,
      "epoch": 6.666084068863061,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016430928221120112,
      "loss": 5.2185,
      "mean_token_accuracy": 0.21823899894952775,
      "num_tokens": 154580656.0,
      "step": 76280
    },
    {
      "entropy": 5.892114353179932,
      "epoch": 6.666521017215765,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016428215384408352,
      "loss": 5.1308,
      "mean_token_accuracy": 0.22671397477388383,
      "num_tokens": 154591498.0,
      "step": 76285
    },
    {
      "entropy": 5.856242322921753,
      "epoch": 6.66695796556847,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016425502760059225,
      "loss": 5.1284,
      "mean_token_accuracy": 0.21990089118480682,
      "num_tokens": 154602139.0,
      "step": 76290
    },
    {
      "entropy": 5.907528448104858,
      "epoch": 6.667394913921175,
      "grad_norm": 1.125,
      "learning_rate": 0.00016422790348124755,
      "loss": 5.1708,
      "mean_token_accuracy": 0.22323531210422515,
      "num_tokens": 154613518.0,
      "step": 76295
    },
    {
      "entropy": 5.820871114730835,
      "epoch": 6.6678318622738795,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001642007814865697,
      "loss": 5.1024,
      "mean_token_accuracy": 0.2269277900457382,
      "num_tokens": 154622963.0,
      "step": 76300
    },
    {
      "entropy": 5.860576820373535,
      "epoch": 6.668268810626584,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016417366161707906,
      "loss": 5.1778,
      "mean_token_accuracy": 0.2217840000987053,
      "num_tokens": 154633540.0,
      "step": 76305
    },
    {
      "entropy": 5.8683980941772464,
      "epoch": 6.668705758979288,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001641465438732955,
      "loss": 5.1633,
      "mean_token_accuracy": 0.22647972851991655,
      "num_tokens": 154644067.0,
      "step": 76310
    },
    {
      "entropy": 5.86145510673523,
      "epoch": 6.669142707331993,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001641194282557394,
      "loss": 5.1065,
      "mean_token_accuracy": 0.2362605392932892,
      "num_tokens": 154654107.0,
      "step": 76315
    },
    {
      "entropy": 5.847645235061646,
      "epoch": 6.669579655684698,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001640923147649307,
      "loss": 5.1547,
      "mean_token_accuracy": 0.2271491453051567,
      "num_tokens": 154663998.0,
      "step": 76320
    },
    {
      "entropy": 5.835273218154907,
      "epoch": 6.670016604037403,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016406520340138953,
      "loss": 5.0876,
      "mean_token_accuracy": 0.22758108973503113,
      "num_tokens": 154673381.0,
      "step": 76325
    },
    {
      "entropy": 5.82731671333313,
      "epoch": 6.670453552390107,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016403809416563588,
      "loss": 5.1317,
      "mean_token_accuracy": 0.22596031427383423,
      "num_tokens": 154682889.0,
      "step": 76330
    },
    {
      "entropy": 5.825682973861694,
      "epoch": 6.6708905007428125,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001640109870581898,
      "loss": 5.1474,
      "mean_token_accuracy": 0.22716057002544404,
      "num_tokens": 154692242.0,
      "step": 76335
    },
    {
      "entropy": 5.789054250717163,
      "epoch": 6.671327449095517,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016398388207957107,
      "loss": 5.0225,
      "mean_token_accuracy": 0.23782321214675903,
      "num_tokens": 154701949.0,
      "step": 76340
    },
    {
      "entropy": 5.873665857315063,
      "epoch": 6.6717643974482215,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016395677923029963,
      "loss": 5.2002,
      "mean_token_accuracy": 0.22152767926454545,
      "num_tokens": 154711945.0,
      "step": 76345
    },
    {
      "entropy": 5.834450435638428,
      "epoch": 6.672201345800926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001639296785108955,
      "loss": 5.1788,
      "mean_token_accuracy": 0.23263135999441148,
      "num_tokens": 154722549.0,
      "step": 76350
    },
    {
      "entropy": 5.773073339462281,
      "epoch": 6.672638294153631,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016390257992187824,
      "loss": 5.0183,
      "mean_token_accuracy": 0.23695361167192458,
      "num_tokens": 154732001.0,
      "step": 76355
    },
    {
      "entropy": 5.8615443229675295,
      "epoch": 6.673075242506336,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016387548346376768,
      "loss": 5.1929,
      "mean_token_accuracy": 0.22727195620536805,
      "num_tokens": 154742715.0,
      "step": 76360
    },
    {
      "entropy": 5.896287822723389,
      "epoch": 6.67351219085904,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001638483891370836,
      "loss": 5.1607,
      "mean_token_accuracy": 0.22572814524173737,
      "num_tokens": 154752606.0,
      "step": 76365
    },
    {
      "entropy": 5.824819087982178,
      "epoch": 6.673949139211745,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016382129694234566,
      "loss": 5.12,
      "mean_token_accuracy": 0.2276083916425705,
      "num_tokens": 154762519.0,
      "step": 76370
    },
    {
      "entropy": 5.934785795211792,
      "epoch": 6.67438608756445,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001637942068800735,
      "loss": 5.1647,
      "mean_token_accuracy": 0.22035557627677918,
      "num_tokens": 154772263.0,
      "step": 76375
    },
    {
      "entropy": 5.874913120269776,
      "epoch": 6.6748230359171545,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001637671189507867,
      "loss": 5.0742,
      "mean_token_accuracy": 0.22316476851701736,
      "num_tokens": 154781223.0,
      "step": 76380
    },
    {
      "entropy": 5.813022947311401,
      "epoch": 6.675259984269859,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001637400331550049,
      "loss": 5.1157,
      "mean_token_accuracy": 0.23030615150928496,
      "num_tokens": 154790257.0,
      "step": 76385
    },
    {
      "entropy": 5.843983221054077,
      "epoch": 6.675696932622564,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001637129494932475,
      "loss": 5.1136,
      "mean_token_accuracy": 0.22882840484380723,
      "num_tokens": 154799888.0,
      "step": 76390
    },
    {
      "entropy": 5.840944910049439,
      "epoch": 6.676133880975269,
      "grad_norm": 1.25,
      "learning_rate": 0.00016368586796603406,
      "loss": 5.0986,
      "mean_token_accuracy": 0.2351212576031685,
      "num_tokens": 154810208.0,
      "step": 76395
    },
    {
      "entropy": 5.839734029769898,
      "epoch": 6.676570829327973,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001636587885738841,
      "loss": 5.1327,
      "mean_token_accuracy": 0.22569168955087662,
      "num_tokens": 154820439.0,
      "step": 76400
    },
    {
      "entropy": 5.865071773529053,
      "epoch": 6.677007777680678,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016363171131731684,
      "loss": 5.1826,
      "mean_token_accuracy": 0.22119292616844177,
      "num_tokens": 154831808.0,
      "step": 76405
    },
    {
      "entropy": 5.884115171432495,
      "epoch": 6.677444726033383,
      "grad_norm": 1.09375,
      "learning_rate": 0.00016360463619685168,
      "loss": 5.155,
      "mean_token_accuracy": 0.2263089656829834,
      "num_tokens": 154842064.0,
      "step": 76410
    },
    {
      "entropy": 5.8195226192474365,
      "epoch": 6.677881674386088,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016357756321300798,
      "loss": 5.1382,
      "mean_token_accuracy": 0.22483410090208053,
      "num_tokens": 154852173.0,
      "step": 76415
    },
    {
      "entropy": 5.880508327484131,
      "epoch": 6.678318622738792,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016355049236630503,
      "loss": 5.1057,
      "mean_token_accuracy": 0.22756544202566148,
      "num_tokens": 154861903.0,
      "step": 76420
    },
    {
      "entropy": 5.858910942077637,
      "epoch": 6.6787555710914965,
      "grad_norm": 1.125,
      "learning_rate": 0.00016352342365726197,
      "loss": 5.1908,
      "mean_token_accuracy": 0.22098191380500792,
      "num_tokens": 154872008.0,
      "step": 76425
    },
    {
      "entropy": 5.809266471862793,
      "epoch": 6.679192519444202,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016349635708639814,
      "loss": 5.0782,
      "mean_token_accuracy": 0.23201775997877122,
      "num_tokens": 154882330.0,
      "step": 76430
    },
    {
      "entropy": 5.844366550445557,
      "epoch": 6.679629467796906,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016346929265423254,
      "loss": 5.0659,
      "mean_token_accuracy": 0.2315283328294754,
      "num_tokens": 154891434.0,
      "step": 76435
    },
    {
      "entropy": 5.899958324432373,
      "epoch": 6.680066416149611,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016344223036128438,
      "loss": 5.2034,
      "mean_token_accuracy": 0.21604759246110916,
      "num_tokens": 154901580.0,
      "step": 76440
    },
    {
      "entropy": 5.944929361343384,
      "epoch": 6.680503364502316,
      "grad_norm": 1.25,
      "learning_rate": 0.0001634151702080727,
      "loss": 5.2484,
      "mean_token_accuracy": 0.21274553388357162,
      "num_tokens": 154912196.0,
      "step": 76445
    },
    {
      "entropy": 5.917916965484619,
      "epoch": 6.680940312855021,
      "grad_norm": 1.09375,
      "learning_rate": 0.00016338811219511663,
      "loss": 5.219,
      "mean_token_accuracy": 0.21565340608358383,
      "num_tokens": 154922283.0,
      "step": 76450
    },
    {
      "entropy": 5.822020387649536,
      "epoch": 6.681377261207725,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016336105632293492,
      "loss": 5.0833,
      "mean_token_accuracy": 0.23961593061685563,
      "num_tokens": 154932671.0,
      "step": 76455
    },
    {
      "entropy": 5.667372417449951,
      "epoch": 6.68181420956043,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016333400259204667,
      "loss": 4.9357,
      "mean_token_accuracy": 0.24306222200393676,
      "num_tokens": 154942245.0,
      "step": 76460
    },
    {
      "entropy": 5.8490331172943115,
      "epoch": 6.682251157913135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016330695100297076,
      "loss": 5.1238,
      "mean_token_accuracy": 0.22452393174171448,
      "num_tokens": 154952209.0,
      "step": 76465
    },
    {
      "entropy": 5.807170772552491,
      "epoch": 6.682688106265839,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016327990155622606,
      "loss": 5.1858,
      "mean_token_accuracy": 0.22772806882858276,
      "num_tokens": 154962423.0,
      "step": 76470
    },
    {
      "entropy": 5.8883734226226805,
      "epoch": 6.683125054618544,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016325285425233135,
      "loss": 5.161,
      "mean_token_accuracy": 0.21271687895059585,
      "num_tokens": 154971581.0,
      "step": 76475
    },
    {
      "entropy": 5.8662168979644775,
      "epoch": 6.683562002971248,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001632258090918055,
      "loss": 5.1548,
      "mean_token_accuracy": 0.2298563912510872,
      "num_tokens": 154981517.0,
      "step": 76480
    },
    {
      "entropy": 5.842780685424804,
      "epoch": 6.683998951323954,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001631987660751672,
      "loss": 5.081,
      "mean_token_accuracy": 0.23104832023382188,
      "num_tokens": 154991775.0,
      "step": 76485
    },
    {
      "entropy": 5.880695867538452,
      "epoch": 6.684435899676658,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001631717252029351,
      "loss": 5.1266,
      "mean_token_accuracy": 0.2291541025042534,
      "num_tokens": 155001860.0,
      "step": 76490
    },
    {
      "entropy": 5.869053173065185,
      "epoch": 6.684872848029363,
      "grad_norm": 1.09375,
      "learning_rate": 0.00016314468647562805,
      "loss": 5.1965,
      "mean_token_accuracy": 0.21633443534374236,
      "num_tokens": 155013342.0,
      "step": 76495
    },
    {
      "entropy": 5.825735998153687,
      "epoch": 6.685309796382068,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001631176498937644,
      "loss": 5.0716,
      "mean_token_accuracy": 0.22913531661033631,
      "num_tokens": 155023010.0,
      "step": 76500
    },
    {
      "entropy": 5.803959512710572,
      "epoch": 6.6857467447347725,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016309061545786285,
      "loss": 5.1103,
      "mean_token_accuracy": 0.23021149188280104,
      "num_tokens": 155034392.0,
      "step": 76505
    },
    {
      "entropy": 5.817869710922241,
      "epoch": 6.686183693087477,
      "grad_norm": 1.171875,
      "learning_rate": 0.000163063583168442,
      "loss": 5.1109,
      "mean_token_accuracy": 0.22840025573968886,
      "num_tokens": 155044100.0,
      "step": 76510
    },
    {
      "entropy": 5.844702816009521,
      "epoch": 6.686620641440181,
      "grad_norm": 1.09375,
      "learning_rate": 0.00016303655302602022,
      "loss": 5.1413,
      "mean_token_accuracy": 0.22409642487764359,
      "num_tokens": 155055126.0,
      "step": 76515
    },
    {
      "entropy": 5.931470489501953,
      "epoch": 6.687057589792887,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016300952503111603,
      "loss": 5.2481,
      "mean_token_accuracy": 0.21892748028039932,
      "num_tokens": 155065704.0,
      "step": 76520
    },
    {
      "entropy": 5.899197578430176,
      "epoch": 6.687494538145591,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016298249918424786,
      "loss": 5.143,
      "mean_token_accuracy": 0.22028518766164779,
      "num_tokens": 155075432.0,
      "step": 76525
    },
    {
      "entropy": 5.862089443206787,
      "epoch": 6.687931486498296,
      "grad_norm": 1.171875,
      "learning_rate": 0.000162955475485934,
      "loss": 5.1269,
      "mean_token_accuracy": 0.22082828730344772,
      "num_tokens": 155085037.0,
      "step": 76530
    },
    {
      "entropy": 5.798416614532471,
      "epoch": 6.688368434851,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016292845393669288,
      "loss": 5.0326,
      "mean_token_accuracy": 0.23384902328252793,
      "num_tokens": 155094281.0,
      "step": 76535
    },
    {
      "entropy": 5.864529180526733,
      "epoch": 6.6888053832037055,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001629014345370427,
      "loss": 5.1797,
      "mean_token_accuracy": 0.2255120947957039,
      "num_tokens": 155104976.0,
      "step": 76540
    },
    {
      "entropy": 5.859778213500976,
      "epoch": 6.68924233155641,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016287441728750175,
      "loss": 5.1253,
      "mean_token_accuracy": 0.2336387887597084,
      "num_tokens": 155114606.0,
      "step": 76545
    },
    {
      "entropy": 5.782042932510376,
      "epoch": 6.6896792799091145,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016284740218858824,
      "loss": 5.0872,
      "mean_token_accuracy": 0.23717406243085862,
      "num_tokens": 155124807.0,
      "step": 76550
    },
    {
      "entropy": 5.796314382553101,
      "epoch": 6.69011622826182,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001628203892408203,
      "loss": 5.0719,
      "mean_token_accuracy": 0.23876355439424515,
      "num_tokens": 155134917.0,
      "step": 76555
    },
    {
      "entropy": 5.880048179626465,
      "epoch": 6.690553176614524,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001627933784447161,
      "loss": 5.1644,
      "mean_token_accuracy": 0.22964314222335816,
      "num_tokens": 155145247.0,
      "step": 76560
    },
    {
      "entropy": 5.802233362197876,
      "epoch": 6.690990124967229,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016276636980079367,
      "loss": 5.0834,
      "mean_token_accuracy": 0.23397710025310517,
      "num_tokens": 155154860.0,
      "step": 76565
    },
    {
      "entropy": 5.8697089672088625,
      "epoch": 6.691427073319933,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001627393633095711,
      "loss": 5.1165,
      "mean_token_accuracy": 0.23247601091861725,
      "num_tokens": 155164193.0,
      "step": 76570
    },
    {
      "entropy": 5.857525205612182,
      "epoch": 6.691864021672639,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001627123589715664,
      "loss": 5.1622,
      "mean_token_accuracy": 0.22766944020986557,
      "num_tokens": 155174871.0,
      "step": 76575
    },
    {
      "entropy": 5.848358011245727,
      "epoch": 6.692300970025343,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016268535678729747,
      "loss": 5.1603,
      "mean_token_accuracy": 0.23020731061697006,
      "num_tokens": 155185420.0,
      "step": 76580
    },
    {
      "entropy": 5.858381080627441,
      "epoch": 6.6927379183780475,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001626583567572823,
      "loss": 5.1292,
      "mean_token_accuracy": 0.22690737396478652,
      "num_tokens": 155195655.0,
      "step": 76585
    },
    {
      "entropy": 5.871741533279419,
      "epoch": 6.693174866730752,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016263135888203867,
      "loss": 5.1103,
      "mean_token_accuracy": 0.23015436828136443,
      "num_tokens": 155205164.0,
      "step": 76590
    },
    {
      "entropy": 5.825143527984619,
      "epoch": 6.693611815083457,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016260436316208445,
      "loss": 5.1285,
      "mean_token_accuracy": 0.23037141263484956,
      "num_tokens": 155215289.0,
      "step": 76595
    },
    {
      "entropy": 5.899353933334351,
      "epoch": 6.694048763436162,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001625773695979374,
      "loss": 5.1615,
      "mean_token_accuracy": 0.22597172260284423,
      "num_tokens": 155224490.0,
      "step": 76600
    },
    {
      "entropy": 5.889973783493042,
      "epoch": 6.694485711788866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016255037819011534,
      "loss": 5.1875,
      "mean_token_accuracy": 0.22183503210544586,
      "num_tokens": 155235055.0,
      "step": 76605
    },
    {
      "entropy": 5.807663917541504,
      "epoch": 6.694922660141572,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016252338893913592,
      "loss": 5.0644,
      "mean_token_accuracy": 0.23282941281795502,
      "num_tokens": 155245801.0,
      "step": 76610
    },
    {
      "entropy": 5.862449407577515,
      "epoch": 6.695359608494276,
      "grad_norm": 1.25,
      "learning_rate": 0.00016249640184551686,
      "loss": 5.1652,
      "mean_token_accuracy": 0.21460903584957122,
      "num_tokens": 155255942.0,
      "step": 76615
    },
    {
      "entropy": 5.851031923294068,
      "epoch": 6.695796556846981,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001624694169097758,
      "loss": 5.1564,
      "mean_token_accuracy": 0.22728344351053237,
      "num_tokens": 155264966.0,
      "step": 76620
    },
    {
      "entropy": 5.851796054840088,
      "epoch": 6.696233505199685,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016244243413243023,
      "loss": 5.0855,
      "mean_token_accuracy": 0.23156467825174332,
      "num_tokens": 155274706.0,
      "step": 76625
    },
    {
      "entropy": 5.900090980529785,
      "epoch": 6.69667045355239,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016241545351399785,
      "loss": 5.1715,
      "mean_token_accuracy": 0.22235766053199768,
      "num_tokens": 155285799.0,
      "step": 76630
    },
    {
      "entropy": 5.83395676612854,
      "epoch": 6.697107401905095,
      "grad_norm": 1.109375,
      "learning_rate": 0.000162388475054996,
      "loss": 5.1259,
      "mean_token_accuracy": 0.22392433285713195,
      "num_tokens": 155296456.0,
      "step": 76635
    },
    {
      "entropy": 5.877989292144775,
      "epoch": 6.697544350257799,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016236149875594223,
      "loss": 5.1891,
      "mean_token_accuracy": 0.21646265387535096,
      "num_tokens": 155307927.0,
      "step": 76640
    },
    {
      "entropy": 5.8682701110839846,
      "epoch": 6.697981298610504,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016233452461735388,
      "loss": 5.1682,
      "mean_token_accuracy": 0.22794336825609207,
      "num_tokens": 155317759.0,
      "step": 76645
    },
    {
      "entropy": 5.8794745922088625,
      "epoch": 6.698418246963209,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016230755263974838,
      "loss": 5.1584,
      "mean_token_accuracy": 0.22486845552921295,
      "num_tokens": 155327998.0,
      "step": 76650
    },
    {
      "entropy": 5.91175684928894,
      "epoch": 6.698855195315914,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001622805828236431,
      "loss": 5.1263,
      "mean_token_accuracy": 0.2305226743221283,
      "num_tokens": 155337198.0,
      "step": 76655
    },
    {
      "entropy": 5.907952117919922,
      "epoch": 6.699292143668618,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016225361516955533,
      "loss": 5.168,
      "mean_token_accuracy": 0.2228784069418907,
      "num_tokens": 155346397.0,
      "step": 76660
    },
    {
      "entropy": 5.871472549438477,
      "epoch": 6.6997290920213235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00016222664967800227,
      "loss": 5.1747,
      "mean_token_accuracy": 0.2214667320251465,
      "num_tokens": 155356756.0,
      "step": 76665
    },
    {
      "entropy": 5.855054712295532,
      "epoch": 6.700166040374028,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016219968634950116,
      "loss": 5.1626,
      "mean_token_accuracy": 0.22766497880220413,
      "num_tokens": 155366024.0,
      "step": 76670
    },
    {
      "entropy": 5.875545167922974,
      "epoch": 6.700602988726732,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016217272518456917,
      "loss": 5.1716,
      "mean_token_accuracy": 0.22038629055023193,
      "num_tokens": 155376241.0,
      "step": 76675
    },
    {
      "entropy": 5.864537620544434,
      "epoch": 6.701039937079437,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016214576618372352,
      "loss": 5.1509,
      "mean_token_accuracy": 0.21992084234952927,
      "num_tokens": 155386744.0,
      "step": 76680
    },
    {
      "entropy": 5.891005516052246,
      "epoch": 6.701476885432142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016211880934748115,
      "loss": 5.1674,
      "mean_token_accuracy": 0.21902995109558104,
      "num_tokens": 155397116.0,
      "step": 76685
    },
    {
      "entropy": 5.856457090377807,
      "epoch": 6.701913833784847,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016209185467635912,
      "loss": 5.1469,
      "mean_token_accuracy": 0.2239812508225441,
      "num_tokens": 155407651.0,
      "step": 76690
    },
    {
      "entropy": 5.8214842796325685,
      "epoch": 6.702350782137551,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001620649021708745,
      "loss": 5.0804,
      "mean_token_accuracy": 0.22949042469263076,
      "num_tokens": 155417051.0,
      "step": 76695
    },
    {
      "entropy": 5.813632678985596,
      "epoch": 6.702787730490256,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016203795183154424,
      "loss": 5.0897,
      "mean_token_accuracy": 0.22960612028837205,
      "num_tokens": 155426230.0,
      "step": 76700
    },
    {
      "entropy": 5.817945194244385,
      "epoch": 6.703224678842961,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016201100365888525,
      "loss": 5.143,
      "mean_token_accuracy": 0.2236471727490425,
      "num_tokens": 155437858.0,
      "step": 76705
    },
    {
      "entropy": 5.790648651123047,
      "epoch": 6.7036616271956655,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016198405765341443,
      "loss": 5.0421,
      "mean_token_accuracy": 0.237511545419693,
      "num_tokens": 155447309.0,
      "step": 76710
    },
    {
      "entropy": 5.811426115036011,
      "epoch": 6.70409857554837,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001619571138156486,
      "loss": 5.0792,
      "mean_token_accuracy": 0.2374952554702759,
      "num_tokens": 155459088.0,
      "step": 76715
    },
    {
      "entropy": 5.803419685363769,
      "epoch": 6.704535523901075,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00016193017214610452,
      "loss": 5.0911,
      "mean_token_accuracy": 0.23041474521160127,
      "num_tokens": 155469854.0,
      "step": 76720
    },
    {
      "entropy": 5.7990357875823975,
      "epoch": 6.70497247225378,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00016190323264529902,
      "loss": 5.0794,
      "mean_token_accuracy": 0.23376063555479049,
      "num_tokens": 155480495.0,
      "step": 76725
    },
    {
      "entropy": 5.825274562835693,
      "epoch": 6.705409420606484,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016187629531374886,
      "loss": 5.1293,
      "mean_token_accuracy": 0.22851891666650773,
      "num_tokens": 155490549.0,
      "step": 76730
    },
    {
      "entropy": 5.86010775566101,
      "epoch": 6.705846368959189,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016184936015197054,
      "loss": 5.1327,
      "mean_token_accuracy": 0.2233770325779915,
      "num_tokens": 155501638.0,
      "step": 76735
    },
    {
      "entropy": 5.87330117225647,
      "epoch": 6.706283317311894,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016182242716048078,
      "loss": 5.0889,
      "mean_token_accuracy": 0.23473575562238694,
      "num_tokens": 155511745.0,
      "step": 76740
    },
    {
      "entropy": 5.886266422271729,
      "epoch": 6.706720265664599,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001617954963397962,
      "loss": 5.155,
      "mean_token_accuracy": 0.22527038753032685,
      "num_tokens": 155522992.0,
      "step": 76745
    },
    {
      "entropy": 5.796104049682617,
      "epoch": 6.707157214017303,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001617685676904333,
      "loss": 5.0409,
      "mean_token_accuracy": 0.23718371391296386,
      "num_tokens": 155532288.0,
      "step": 76750
    },
    {
      "entropy": 5.844786739349365,
      "epoch": 6.7075941623700075,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001617416412129086,
      "loss": 5.1175,
      "mean_token_accuracy": 0.22824536710977555,
      "num_tokens": 155542388.0,
      "step": 76755
    },
    {
      "entropy": 5.879264831542969,
      "epoch": 6.708031110722713,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016171471690773858,
      "loss": 5.1748,
      "mean_token_accuracy": 0.22518169581890107,
      "num_tokens": 155552506.0,
      "step": 76760
    },
    {
      "entropy": 5.79171404838562,
      "epoch": 6.708468059075417,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016168779477543965,
      "loss": 5.0705,
      "mean_token_accuracy": 0.23473797589540482,
      "num_tokens": 155563537.0,
      "step": 76765
    },
    {
      "entropy": 5.8232685089111325,
      "epoch": 6.708905007428122,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016166087481652816,
      "loss": 5.1389,
      "mean_token_accuracy": 0.22687864452600479,
      "num_tokens": 155574050.0,
      "step": 76770
    },
    {
      "entropy": 5.795460605621338,
      "epoch": 6.709341955780827,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016163395703152063,
      "loss": 5.0374,
      "mean_token_accuracy": 0.23346542567014694,
      "num_tokens": 155583617.0,
      "step": 76775
    },
    {
      "entropy": 5.818985176086426,
      "epoch": 6.709778904133532,
      "grad_norm": 1.203125,
      "learning_rate": 0.000161607041420933,
      "loss": 5.1136,
      "mean_token_accuracy": 0.22476432472467422,
      "num_tokens": 155593448.0,
      "step": 76780
    },
    {
      "entropy": 5.8788261890411375,
      "epoch": 6.710215852486236,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016158012798528186,
      "loss": 5.1303,
      "mean_token_accuracy": 0.2211723282933235,
      "num_tokens": 155603358.0,
      "step": 76785
    },
    {
      "entropy": 5.885024070739746,
      "epoch": 6.710652800838941,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016155321672508326,
      "loss": 5.1136,
      "mean_token_accuracy": 0.22686949819326402,
      "num_tokens": 155613419.0,
      "step": 76790
    },
    {
      "entropy": 5.876690864562988,
      "epoch": 6.711089749191646,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001615263076408534,
      "loss": 5.1425,
      "mean_token_accuracy": 0.22428224384784698,
      "num_tokens": 155624112.0,
      "step": 76795
    },
    {
      "entropy": 5.849758768081665,
      "epoch": 6.71152669754435,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001614994007331084,
      "loss": 5.0872,
      "mean_token_accuracy": 0.22541537433862685,
      "num_tokens": 155634065.0,
      "step": 76800
    },
    {
      "entropy": 5.837621879577637,
      "epoch": 6.711963645897055,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016147249600236436,
      "loss": 5.1494,
      "mean_token_accuracy": 0.23177018761634827,
      "num_tokens": 155644517.0,
      "step": 76805
    },
    {
      "entropy": 5.826697301864624,
      "epoch": 6.712400594249759,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016144559344913738,
      "loss": 5.0991,
      "mean_token_accuracy": 0.2311225190758705,
      "num_tokens": 155654042.0,
      "step": 76810
    },
    {
      "entropy": 5.8707812309265135,
      "epoch": 6.712837542602465,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016141869307394335,
      "loss": 5.1738,
      "mean_token_accuracy": 0.2275741755962372,
      "num_tokens": 155664749.0,
      "step": 76815
    },
    {
      "entropy": 5.899528169631958,
      "epoch": 6.713274490955169,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016139179487729838,
      "loss": 5.2282,
      "mean_token_accuracy": 0.2183544933795929,
      "num_tokens": 155675377.0,
      "step": 76820
    },
    {
      "entropy": 5.874784994125366,
      "epoch": 6.713711439307874,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016136489885971837,
      "loss": 5.2036,
      "mean_token_accuracy": 0.22108560055494308,
      "num_tokens": 155686832.0,
      "step": 76825
    },
    {
      "entropy": 5.864764404296875,
      "epoch": 6.714148387660579,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016133800502171906,
      "loss": 5.1494,
      "mean_token_accuracy": 0.2185822308063507,
      "num_tokens": 155696606.0,
      "step": 76830
    },
    {
      "entropy": 5.856470155715942,
      "epoch": 6.7145853360132834,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016131111336381638,
      "loss": 5.0757,
      "mean_token_accuracy": 0.22769321799278258,
      "num_tokens": 155707050.0,
      "step": 76835
    },
    {
      "entropy": 5.884257173538208,
      "epoch": 6.715022284365988,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016128422388652608,
      "loss": 5.1791,
      "mean_token_accuracy": 0.2244916543364525,
      "num_tokens": 155716591.0,
      "step": 76840
    },
    {
      "entropy": 5.752255487442016,
      "epoch": 6.715459232718692,
      "grad_norm": 1.28125,
      "learning_rate": 0.000161257336590364,
      "loss": 5.1188,
      "mean_token_accuracy": 0.23221437633037567,
      "num_tokens": 155726816.0,
      "step": 76845
    },
    {
      "entropy": 5.8433764457702635,
      "epoch": 6.715896181071398,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016123045147584575,
      "loss": 5.1242,
      "mean_token_accuracy": 0.2251667082309723,
      "num_tokens": 155737366.0,
      "step": 76850
    },
    {
      "entropy": 5.855607557296753,
      "epoch": 6.716333129424102,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016120356854348712,
      "loss": 5.0958,
      "mean_token_accuracy": 0.22409184128046036,
      "num_tokens": 155747473.0,
      "step": 76855
    },
    {
      "entropy": 5.817062759399414,
      "epoch": 6.716770077776807,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016117668779380362,
      "loss": 5.1032,
      "mean_token_accuracy": 0.22447466254234313,
      "num_tokens": 155756858.0,
      "step": 76860
    },
    {
      "entropy": 5.88867220878601,
      "epoch": 6.717207026129511,
      "grad_norm": 1.125,
      "learning_rate": 0.00016114980922731092,
      "loss": 5.1883,
      "mean_token_accuracy": 0.21939126700162886,
      "num_tokens": 155767610.0,
      "step": 76865
    },
    {
      "entropy": 5.804067468643188,
      "epoch": 6.7176439744822165,
      "grad_norm": 1.125,
      "learning_rate": 0.0001611229328445245,
      "loss": 5.1077,
      "mean_token_accuracy": 0.23179561495780945,
      "num_tokens": 155777849.0,
      "step": 76870
    },
    {
      "entropy": 5.879244041442871,
      "epoch": 6.718080922834921,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016109605864596005,
      "loss": 5.195,
      "mean_token_accuracy": 0.21223336607217788,
      "num_tokens": 155787951.0,
      "step": 76875
    },
    {
      "entropy": 5.873669099807739,
      "epoch": 6.7185178711876254,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016106918663213277,
      "loss": 5.1132,
      "mean_token_accuracy": 0.22974129617214203,
      "num_tokens": 155798454.0,
      "step": 76880
    },
    {
      "entropy": 5.820954751968384,
      "epoch": 6.718954819540331,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016104231680355813,
      "loss": 5.0414,
      "mean_token_accuracy": 0.2358689472079277,
      "num_tokens": 155807856.0,
      "step": 76885
    },
    {
      "entropy": 5.783722257614135,
      "epoch": 6.719391767893035,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016101544916075156,
      "loss": 5.053,
      "mean_token_accuracy": 0.23509815335273743,
      "num_tokens": 155819102.0,
      "step": 76890
    },
    {
      "entropy": 5.821679306030274,
      "epoch": 6.71982871624574,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001609885837042284,
      "loss": 5.0809,
      "mean_token_accuracy": 0.23362398892641068,
      "num_tokens": 155828768.0,
      "step": 76895
    },
    {
      "entropy": 5.790878629684448,
      "epoch": 6.720265664598444,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000160961720434504,
      "loss": 5.0829,
      "mean_token_accuracy": 0.23276403844356536,
      "num_tokens": 155839754.0,
      "step": 76900
    },
    {
      "entropy": 5.824597692489624,
      "epoch": 6.720702612951149,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016093485935209356,
      "loss": 5.153,
      "mean_token_accuracy": 0.22466842979192733,
      "num_tokens": 155849114.0,
      "step": 76905
    },
    {
      "entropy": 5.801271390914917,
      "epoch": 6.721139561303854,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016090800045751242,
      "loss": 5.1224,
      "mean_token_accuracy": 0.22715928554534912,
      "num_tokens": 155858849.0,
      "step": 76910
    },
    {
      "entropy": 5.852819585800171,
      "epoch": 6.7215765096565585,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016088114375127543,
      "loss": 5.1197,
      "mean_token_accuracy": 0.22934303879737855,
      "num_tokens": 155869392.0,
      "step": 76915
    },
    {
      "entropy": 5.784466218948364,
      "epoch": 6.722013458009263,
      "grad_norm": 1.25,
      "learning_rate": 0.00016085428923389797,
      "loss": 5.074,
      "mean_token_accuracy": 0.235605151951313,
      "num_tokens": 155879803.0,
      "step": 76920
    },
    {
      "entropy": 5.8658689022064205,
      "epoch": 6.722450406361968,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016082743690589503,
      "loss": 5.1058,
      "mean_token_accuracy": 0.22927676141262054,
      "num_tokens": 155889814.0,
      "step": 76925
    },
    {
      "entropy": 5.8402259349823,
      "epoch": 6.722887354714673,
      "grad_norm": 1.125,
      "learning_rate": 0.00016080058676778167,
      "loss": 5.1243,
      "mean_token_accuracy": 0.2280411571264267,
      "num_tokens": 155899212.0,
      "step": 76930
    },
    {
      "entropy": 5.871709728240967,
      "epoch": 6.723324303067377,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001607737388200729,
      "loss": 5.0948,
      "mean_token_accuracy": 0.22805726826190947,
      "num_tokens": 155910628.0,
      "step": 76935
    },
    {
      "entropy": 5.8200843811035154,
      "epoch": 6.723761251420083,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001607468930632837,
      "loss": 5.12,
      "mean_token_accuracy": 0.2306300699710846,
      "num_tokens": 155920602.0,
      "step": 76940
    },
    {
      "entropy": 5.807301473617554,
      "epoch": 6.724198199772787,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001607200494979289,
      "loss": 5.1412,
      "mean_token_accuracy": 0.2231146812438965,
      "num_tokens": 155931463.0,
      "step": 76945
    },
    {
      "entropy": 5.846483278274536,
      "epoch": 6.724635148125492,
      "grad_norm": 1.125,
      "learning_rate": 0.00016069320812452348,
      "loss": 5.1902,
      "mean_token_accuracy": 0.2174411818385124,
      "num_tokens": 155942010.0,
      "step": 76950
    },
    {
      "entropy": 5.838056564331055,
      "epoch": 6.725072096478196,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001606663689435822,
      "loss": 5.143,
      "mean_token_accuracy": 0.23066183179616928,
      "num_tokens": 155952272.0,
      "step": 76955
    },
    {
      "entropy": 5.850044536590576,
      "epoch": 6.7255090448309005,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016063953195561998,
      "loss": 5.0665,
      "mean_token_accuracy": 0.230635467171669,
      "num_tokens": 155962346.0,
      "step": 76960
    },
    {
      "entropy": 5.872679567337036,
      "epoch": 6.725945993183606,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016061269716115135,
      "loss": 5.1285,
      "mean_token_accuracy": 0.22391907572746278,
      "num_tokens": 155971834.0,
      "step": 76965
    },
    {
      "entropy": 5.854733037948608,
      "epoch": 6.72638294153631,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001605858645606912,
      "loss": 5.15,
      "mean_token_accuracy": 0.22659264355897904,
      "num_tokens": 155982509.0,
      "step": 76970
    },
    {
      "entropy": 5.829850435256958,
      "epoch": 6.726819889889015,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016055903415475404,
      "loss": 5.1546,
      "mean_token_accuracy": 0.22720773071050643,
      "num_tokens": 155991987.0,
      "step": 76975
    },
    {
      "entropy": 5.853207588195801,
      "epoch": 6.72725683824172,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001605322059438546,
      "loss": 5.1575,
      "mean_token_accuracy": 0.22917945683002472,
      "num_tokens": 156002802.0,
      "step": 76980
    },
    {
      "entropy": 5.7007428169250485,
      "epoch": 6.727693786594425,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016050537992850738,
      "loss": 4.9757,
      "mean_token_accuracy": 0.2450285956263542,
      "num_tokens": 156012122.0,
      "step": 76985
    },
    {
      "entropy": 5.879536390304565,
      "epoch": 6.728130734947129,
      "grad_norm": 1.234375,
      "learning_rate": 0.000160478556109227,
      "loss": 5.1723,
      "mean_token_accuracy": 0.2197947695851326,
      "num_tokens": 156022027.0,
      "step": 76990
    },
    {
      "entropy": 5.813219022750855,
      "epoch": 6.7285676832998345,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016045173448652795,
      "loss": 5.0866,
      "mean_token_accuracy": 0.2318895310163498,
      "num_tokens": 156032346.0,
      "step": 76995
    },
    {
      "entropy": 5.861036682128907,
      "epoch": 6.729004631652539,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00016042491506092462,
      "loss": 5.1851,
      "mean_token_accuracy": 0.2269242912530899,
      "num_tokens": 156043397.0,
      "step": 77000
    },
    {
      "entropy": 5.800433015823364,
      "epoch": 6.729441580005243,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016039809783293147,
      "loss": 5.0481,
      "mean_token_accuracy": 0.23676410764455796,
      "num_tokens": 156054480.0,
      "step": 77005
    },
    {
      "entropy": 5.871887826919556,
      "epoch": 6.729878528357948,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016037128280306295,
      "loss": 5.1718,
      "mean_token_accuracy": 0.22556950002908707,
      "num_tokens": 156065428.0,
      "step": 77010
    },
    {
      "entropy": 5.886036539077759,
      "epoch": 6.730315476710652,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016034446997183321,
      "loss": 5.1799,
      "mean_token_accuracy": 0.22714847773313523,
      "num_tokens": 156075386.0,
      "step": 77015
    },
    {
      "entropy": 5.838928890228272,
      "epoch": 6.730752425063358,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001603176593397566,
      "loss": 5.1106,
      "mean_token_accuracy": 0.22248125523328782,
      "num_tokens": 156084986.0,
      "step": 77020
    },
    {
      "entropy": 5.849467182159424,
      "epoch": 6.731189373416062,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016029085090734738,
      "loss": 5.1006,
      "mean_token_accuracy": 0.23141608536243438,
      "num_tokens": 156095744.0,
      "step": 77025
    },
    {
      "entropy": 5.8572096824646,
      "epoch": 6.731626321768767,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016026404467511975,
      "loss": 5.1483,
      "mean_token_accuracy": 0.23193225115537644,
      "num_tokens": 156106219.0,
      "step": 77030
    },
    {
      "entropy": 5.840574741363525,
      "epoch": 6.732063270121472,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016023724064358785,
      "loss": 5.1617,
      "mean_token_accuracy": 0.22866207361221313,
      "num_tokens": 156115960.0,
      "step": 77035
    },
    {
      "entropy": 5.910040998458863,
      "epoch": 6.7325002184741765,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001602104388132658,
      "loss": 5.2556,
      "mean_token_accuracy": 0.21222716122865676,
      "num_tokens": 156127849.0,
      "step": 77040
    },
    {
      "entropy": 5.812426567077637,
      "epoch": 6.732937166826881,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016018363918466773,
      "loss": 5.1367,
      "mean_token_accuracy": 0.23133531212806702,
      "num_tokens": 156137515.0,
      "step": 77045
    },
    {
      "entropy": 5.8545811653137205,
      "epoch": 6.733374115179585,
      "grad_norm": 1.375,
      "learning_rate": 0.0001601568417583076,
      "loss": 5.1454,
      "mean_token_accuracy": 0.22552601397037506,
      "num_tokens": 156146701.0,
      "step": 77050
    },
    {
      "entropy": 5.859399223327637,
      "epoch": 6.733811063532291,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016013004653469938,
      "loss": 5.1288,
      "mean_token_accuracy": 0.22284646928310395,
      "num_tokens": 156156510.0,
      "step": 77055
    },
    {
      "entropy": 5.847551488876343,
      "epoch": 6.734248011884995,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001601032535143572,
      "loss": 5.167,
      "mean_token_accuracy": 0.2239221766591072,
      "num_tokens": 156167132.0,
      "step": 77060
    },
    {
      "entropy": 5.86423978805542,
      "epoch": 6.7346849602377,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016007646269779473,
      "loss": 5.21,
      "mean_token_accuracy": 0.2258903130888939,
      "num_tokens": 156176670.0,
      "step": 77065
    },
    {
      "entropy": 5.818339824676514,
      "epoch": 6.735121908590404,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001600496740855259,
      "loss": 5.1161,
      "mean_token_accuracy": 0.22997816354036332,
      "num_tokens": 156187381.0,
      "step": 77070
    },
    {
      "entropy": 5.906961488723755,
      "epoch": 6.7355588569431095,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016002288767806455,
      "loss": 5.1459,
      "mean_token_accuracy": 0.22434128671884537,
      "num_tokens": 156197139.0,
      "step": 77075
    },
    {
      "entropy": 5.9080219745635985,
      "epoch": 6.735995805295814,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015999610347592447,
      "loss": 5.2328,
      "mean_token_accuracy": 0.21176410466432571,
      "num_tokens": 156206705.0,
      "step": 77080
    },
    {
      "entropy": 5.847257041931153,
      "epoch": 6.7364327536485185,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015996932147961941,
      "loss": 5.1686,
      "mean_token_accuracy": 0.22929138839244842,
      "num_tokens": 156218287.0,
      "step": 77085
    },
    {
      "entropy": 5.798780488967895,
      "epoch": 6.736869702001224,
      "grad_norm": 1.28125,
      "learning_rate": 0.000159942541689663,
      "loss": 5.0586,
      "mean_token_accuracy": 0.22556622922420502,
      "num_tokens": 156228551.0,
      "step": 77090
    },
    {
      "entropy": 5.889741849899292,
      "epoch": 6.737306650353928,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015991576410656893,
      "loss": 5.1169,
      "mean_token_accuracy": 0.22133747935295106,
      "num_tokens": 156237939.0,
      "step": 77095
    },
    {
      "entropy": 5.944281148910522,
      "epoch": 6.737743598706633,
      "grad_norm": 1.125,
      "learning_rate": 0.00015988898873085084,
      "loss": 5.2329,
      "mean_token_accuracy": 0.21853212118148804,
      "num_tokens": 156247620.0,
      "step": 77100
    },
    {
      "entropy": 5.893162107467651,
      "epoch": 6.738180547059337,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015986221556302233,
      "loss": 5.2421,
      "mean_token_accuracy": 0.21868721544742584,
      "num_tokens": 156257013.0,
      "step": 77105
    },
    {
      "entropy": 5.816336536407471,
      "epoch": 6.738617495412043,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015983544460359677,
      "loss": 5.1982,
      "mean_token_accuracy": 0.2253308966755867,
      "num_tokens": 156267606.0,
      "step": 77110
    },
    {
      "entropy": 5.859693002700806,
      "epoch": 6.739054443764747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001598086758530878,
      "loss": 5.1715,
      "mean_token_accuracy": 0.2300535961985588,
      "num_tokens": 156278331.0,
      "step": 77115
    },
    {
      "entropy": 5.921296405792236,
      "epoch": 6.7394913921174515,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001597819093120087,
      "loss": 5.1572,
      "mean_token_accuracy": 0.2193126693367958,
      "num_tokens": 156287179.0,
      "step": 77120
    },
    {
      "entropy": 5.867377519607544,
      "epoch": 6.739928340470156,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015975514498087296,
      "loss": 5.118,
      "mean_token_accuracy": 0.22433046102523804,
      "num_tokens": 156296995.0,
      "step": 77125
    },
    {
      "entropy": 5.782844209671021,
      "epoch": 6.740365288822861,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015972838286019397,
      "loss": 5.0939,
      "mean_token_accuracy": 0.22982180416584014,
      "num_tokens": 156307721.0,
      "step": 77130
    },
    {
      "entropy": 5.817913436889649,
      "epoch": 6.740802237175566,
      "grad_norm": 1.09375,
      "learning_rate": 0.00015970162295048495,
      "loss": 5.1484,
      "mean_token_accuracy": 0.22558743357658387,
      "num_tokens": 156317924.0,
      "step": 77135
    },
    {
      "entropy": 5.852777099609375,
      "epoch": 6.74123918552827,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015967486525225922,
      "loss": 5.1196,
      "mean_token_accuracy": 0.22502638697624205,
      "num_tokens": 156327782.0,
      "step": 77140
    },
    {
      "entropy": 5.870080280303955,
      "epoch": 6.741676133880976,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015964810976603003,
      "loss": 5.1005,
      "mean_token_accuracy": 0.22816767990589143,
      "num_tokens": 156337943.0,
      "step": 77145
    },
    {
      "entropy": 5.871315908432007,
      "epoch": 6.74211308223368,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015962135649231053,
      "loss": 5.079,
      "mean_token_accuracy": 0.2280589759349823,
      "num_tokens": 156346851.0,
      "step": 77150
    },
    {
      "entropy": 5.806773614883423,
      "epoch": 6.742550030586385,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015959460543161393,
      "loss": 5.1349,
      "mean_token_accuracy": 0.22604529112577437,
      "num_tokens": 156355863.0,
      "step": 77155
    },
    {
      "entropy": 5.853447532653808,
      "epoch": 6.742986978939089,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015956785658445324,
      "loss": 5.2102,
      "mean_token_accuracy": 0.22458236813545226,
      "num_tokens": 156365710.0,
      "step": 77160
    },
    {
      "entropy": 5.8822150230407715,
      "epoch": 6.743423927291794,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001595411099513415,
      "loss": 5.212,
      "mean_token_accuracy": 0.21867192089557647,
      "num_tokens": 156376992.0,
      "step": 77165
    },
    {
      "entropy": 5.836557865142822,
      "epoch": 6.743860875644499,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001595143655327918,
      "loss": 5.1347,
      "mean_token_accuracy": 0.22358663827180864,
      "num_tokens": 156386921.0,
      "step": 77170
    },
    {
      "entropy": 5.888787317276001,
      "epoch": 6.744297823997203,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015948762332931708,
      "loss": 5.1835,
      "mean_token_accuracy": 0.22629116773605346,
      "num_tokens": 156398661.0,
      "step": 77175
    },
    {
      "entropy": 5.856169366836548,
      "epoch": 6.744734772349908,
      "grad_norm": 1.25,
      "learning_rate": 0.00015946088334143027,
      "loss": 5.1081,
      "mean_token_accuracy": 0.22886897027492523,
      "num_tokens": 156408839.0,
      "step": 77180
    },
    {
      "entropy": 5.831162643432617,
      "epoch": 6.745171720702613,
      "grad_norm": 1.0625,
      "learning_rate": 0.00015943414556964422,
      "loss": 5.1423,
      "mean_token_accuracy": 0.23587448745965958,
      "num_tokens": 156419599.0,
      "step": 77185
    },
    {
      "entropy": 5.8477943420410154,
      "epoch": 6.745608669055318,
      "grad_norm": 1.125,
      "learning_rate": 0.00015940741001447184,
      "loss": 5.129,
      "mean_token_accuracy": 0.23065298646688462,
      "num_tokens": 156430153.0,
      "step": 77190
    },
    {
      "entropy": 5.809762382507325,
      "epoch": 6.746045617408022,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015938067667642592,
      "loss": 5.0731,
      "mean_token_accuracy": 0.23639069199562074,
      "num_tokens": 156440425.0,
      "step": 77195
    },
    {
      "entropy": 5.860148620605469,
      "epoch": 6.7464825657607275,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001593539455560192,
      "loss": 5.1596,
      "mean_token_accuracy": 0.22692964375019073,
      "num_tokens": 156450289.0,
      "step": 77200
    },
    {
      "entropy": 5.852351903915405,
      "epoch": 6.746919514113432,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015932721665376447,
      "loss": 5.1477,
      "mean_token_accuracy": 0.22652677744626998,
      "num_tokens": 156460271.0,
      "step": 77205
    },
    {
      "entropy": 5.85827112197876,
      "epoch": 6.747356462466136,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015930048997017427,
      "loss": 5.161,
      "mean_token_accuracy": 0.22438655495643617,
      "num_tokens": 156471085.0,
      "step": 77210
    },
    {
      "entropy": 5.797822284698486,
      "epoch": 6.747793410818841,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015927376550576127,
      "loss": 5.0519,
      "mean_token_accuracy": 0.23181175589561462,
      "num_tokens": 156480667.0,
      "step": 77215
    },
    {
      "entropy": 5.847319650650024,
      "epoch": 6.748230359171546,
      "grad_norm": 1.25,
      "learning_rate": 0.00015924704326103812,
      "loss": 5.1697,
      "mean_token_accuracy": 0.22913493365049362,
      "num_tokens": 156492033.0,
      "step": 77220
    },
    {
      "entropy": 5.887995290756225,
      "epoch": 6.748667307524251,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015922032323651732,
      "loss": 5.1533,
      "mean_token_accuracy": 0.2277242660522461,
      "num_tokens": 156502666.0,
      "step": 77225
    },
    {
      "entropy": 5.861710691452027,
      "epoch": 6.749104255876955,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001591936054327114,
      "loss": 5.0773,
      "mean_token_accuracy": 0.23288158625364302,
      "num_tokens": 156512701.0,
      "step": 77230
    },
    {
      "entropy": 5.867279672622681,
      "epoch": 6.74954120422966,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001591668898501328,
      "loss": 5.1256,
      "mean_token_accuracy": 0.22789098918437958,
      "num_tokens": 156521300.0,
      "step": 77235
    },
    {
      "entropy": 5.808447408676147,
      "epoch": 6.749978152582365,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001591401764892939,
      "loss": 5.0699,
      "mean_token_accuracy": 0.22716027051210402,
      "num_tokens": 156531117.0,
      "step": 77240
    },
    {
      "entropy": 5.785601663589477,
      "epoch": 6.7504151009350695,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001591134653507072,
      "loss": 5.0668,
      "mean_token_accuracy": 0.2332886889576912,
      "num_tokens": 156540932.0,
      "step": 77245
    },
    {
      "entropy": 5.84414267539978,
      "epoch": 6.750852049287774,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001590867564348849,
      "loss": 5.1239,
      "mean_token_accuracy": 0.22976354211568834,
      "num_tokens": 156551122.0,
      "step": 77250
    },
    {
      "entropy": 5.827552366256714,
      "epoch": 6.751288997640479,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001590600497423394,
      "loss": 5.1606,
      "mean_token_accuracy": 0.2246357262134552,
      "num_tokens": 156561816.0,
      "step": 77255
    },
    {
      "entropy": 5.80054087638855,
      "epoch": 6.751725945993184,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015903334527358288,
      "loss": 5.1359,
      "mean_token_accuracy": 0.22644857466220855,
      "num_tokens": 156572002.0,
      "step": 77260
    },
    {
      "entropy": 5.832675170898438,
      "epoch": 6.752162894345888,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015900664302912753,
      "loss": 5.1567,
      "mean_token_accuracy": 0.226629276573658,
      "num_tokens": 156582883.0,
      "step": 77265
    },
    {
      "entropy": 5.80743727684021,
      "epoch": 6.752599842698593,
      "grad_norm": 1.15625,
      "learning_rate": 0.00015897994300948559,
      "loss": 5.0631,
      "mean_token_accuracy": 0.23581713140010835,
      "num_tokens": 156591938.0,
      "step": 77270
    },
    {
      "entropy": 5.846425914764405,
      "epoch": 6.753036791051298,
      "grad_norm": 1.125,
      "learning_rate": 0.00015895324521516914,
      "loss": 5.1401,
      "mean_token_accuracy": 0.2262797772884369,
      "num_tokens": 156601977.0,
      "step": 77275
    },
    {
      "entropy": 5.765338611602783,
      "epoch": 6.7534737394040025,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015892654964669024,
      "loss": 5.0232,
      "mean_token_accuracy": 0.23972522765398024,
      "num_tokens": 156610736.0,
      "step": 77280
    },
    {
      "entropy": 5.875652027130127,
      "epoch": 6.753910687756707,
      "grad_norm": 1.25,
      "learning_rate": 0.00015889985630456092,
      "loss": 5.1949,
      "mean_token_accuracy": 0.21783262193202974,
      "num_tokens": 156620686.0,
      "step": 77285
    },
    {
      "entropy": 5.858463764190674,
      "epoch": 6.7543476361094115,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015887316518929328,
      "loss": 5.1711,
      "mean_token_accuracy": 0.21911636292934417,
      "num_tokens": 156630036.0,
      "step": 77290
    },
    {
      "entropy": 5.872606420516968,
      "epoch": 6.754784584462117,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001588464763013991,
      "loss": 5.1236,
      "mean_token_accuracy": 0.2226647987961769,
      "num_tokens": 156640627.0,
      "step": 77295
    },
    {
      "entropy": 5.862147808074951,
      "epoch": 6.755221532814821,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015881978964139037,
      "loss": 5.2304,
      "mean_token_accuracy": 0.21141040921211243,
      "num_tokens": 156651266.0,
      "step": 77300
    },
    {
      "entropy": 5.854714965820312,
      "epoch": 6.755658481167526,
      "grad_norm": 1.25,
      "learning_rate": 0.00015879310520977897,
      "loss": 5.1544,
      "mean_token_accuracy": 0.22096445560455322,
      "num_tokens": 156660492.0,
      "step": 77305
    },
    {
      "entropy": 5.763400459289551,
      "epoch": 6.756095429520231,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001587664230070767,
      "loss": 5.0419,
      "mean_token_accuracy": 0.2341139018535614,
      "num_tokens": 156670830.0,
      "step": 77310
    },
    {
      "entropy": 5.865266180038452,
      "epoch": 6.756532377872936,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015873974303379533,
      "loss": 5.1164,
      "mean_token_accuracy": 0.22842632234096527,
      "num_tokens": 156680532.0,
      "step": 77315
    },
    {
      "entropy": 5.78402009010315,
      "epoch": 6.75696932622564,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001587130652904466,
      "loss": 5.0838,
      "mean_token_accuracy": 0.23102737218141556,
      "num_tokens": 156690675.0,
      "step": 77320
    },
    {
      "entropy": 5.821981000900268,
      "epoch": 6.7574062745783445,
      "grad_norm": 1.109375,
      "learning_rate": 0.00015868638977754225,
      "loss": 5.1108,
      "mean_token_accuracy": 0.2314704641699791,
      "num_tokens": 156702173.0,
      "step": 77325
    },
    {
      "entropy": 5.78776912689209,
      "epoch": 6.75784322293105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015865971649559385,
      "loss": 5.0595,
      "mean_token_accuracy": 0.2390546515583992,
      "num_tokens": 156713247.0,
      "step": 77330
    },
    {
      "entropy": 5.86389389038086,
      "epoch": 6.758280171283754,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015863304544511304,
      "loss": 5.0977,
      "mean_token_accuracy": 0.22943912893533708,
      "num_tokens": 156723664.0,
      "step": 77335
    },
    {
      "entropy": 5.8579872131347654,
      "epoch": 6.758717119636459,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015860637662661154,
      "loss": 5.1825,
      "mean_token_accuracy": 0.22210433781147004,
      "num_tokens": 156733241.0,
      "step": 77340
    },
    {
      "entropy": 5.819449377059937,
      "epoch": 6.759154067989163,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001585797100406006,
      "loss": 5.1688,
      "mean_token_accuracy": 0.21691033244132996,
      "num_tokens": 156743428.0,
      "step": 77345
    },
    {
      "entropy": 5.867942810058594,
      "epoch": 6.759591016341869,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015855304568759184,
      "loss": 5.1769,
      "mean_token_accuracy": 0.22418480813503266,
      "num_tokens": 156754948.0,
      "step": 77350
    },
    {
      "entropy": 5.946239376068116,
      "epoch": 6.760027964694573,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015852638356809662,
      "loss": 5.2543,
      "mean_token_accuracy": 0.21841851025819778,
      "num_tokens": 156765828.0,
      "step": 77355
    },
    {
      "entropy": 5.859562349319458,
      "epoch": 6.760464913047278,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015849972368262643,
      "loss": 5.1218,
      "mean_token_accuracy": 0.23096997290849686,
      "num_tokens": 156775398.0,
      "step": 77360
    },
    {
      "entropy": 5.868956899642944,
      "epoch": 6.760901861399983,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015847306603169263,
      "loss": 5.178,
      "mean_token_accuracy": 0.21912411451339722,
      "num_tokens": 156784943.0,
      "step": 77365
    },
    {
      "entropy": 5.824765968322754,
      "epoch": 6.761338809752687,
      "grad_norm": 1.046875,
      "learning_rate": 0.00015844641061580646,
      "loss": 5.0737,
      "mean_token_accuracy": 0.23545915335416795,
      "num_tokens": 156795389.0,
      "step": 77370
    },
    {
      "entropy": 5.806158018112183,
      "epoch": 6.761775758105392,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001584197574354792,
      "loss": 5.0302,
      "mean_token_accuracy": 0.2330608546733856,
      "num_tokens": 156804374.0,
      "step": 77375
    },
    {
      "entropy": 5.8532062530517575,
      "epoch": 6.762212706458096,
      "grad_norm": 1.25,
      "learning_rate": 0.00015839310649122204,
      "loss": 5.1591,
      "mean_token_accuracy": 0.22847369760274888,
      "num_tokens": 156814031.0,
      "step": 77380
    },
    {
      "entropy": 5.926404142379761,
      "epoch": 6.762649654810802,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015836645778354635,
      "loss": 5.1648,
      "mean_token_accuracy": 0.2253152385354042,
      "num_tokens": 156823651.0,
      "step": 77385
    },
    {
      "entropy": 5.872392797470093,
      "epoch": 6.763086603163506,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015833981131296298,
      "loss": 5.1449,
      "mean_token_accuracy": 0.22348810285329818,
      "num_tokens": 156833907.0,
      "step": 77390
    },
    {
      "entropy": 5.840192365646362,
      "epoch": 6.763523551516211,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015831316707998316,
      "loss": 5.1258,
      "mean_token_accuracy": 0.22722065597772598,
      "num_tokens": 156844731.0,
      "step": 77395
    },
    {
      "entropy": 5.758130979537964,
      "epoch": 6.763960499868915,
      "grad_norm": 1.171875,
      "learning_rate": 0.000158286525085118,
      "loss": 5.0882,
      "mean_token_accuracy": 0.22820698618888854,
      "num_tokens": 156855705.0,
      "step": 77400
    },
    {
      "entropy": 5.863936901092529,
      "epoch": 6.7643974482216205,
      "grad_norm": 1.125,
      "learning_rate": 0.00015825988532887838,
      "loss": 5.211,
      "mean_token_accuracy": 0.2216736704111099,
      "num_tokens": 156865823.0,
      "step": 77405
    },
    {
      "entropy": 5.850612926483154,
      "epoch": 6.764834396574325,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015823324781177533,
      "loss": 5.1334,
      "mean_token_accuracy": 0.22928674817085265,
      "num_tokens": 156875151.0,
      "step": 77410
    },
    {
      "entropy": 5.852293157577515,
      "epoch": 6.765271344927029,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015820661253431983,
      "loss": 5.1121,
      "mean_token_accuracy": 0.22256760746240617,
      "num_tokens": 156884957.0,
      "step": 77415
    },
    {
      "entropy": 5.828814888000489,
      "epoch": 6.765708293279735,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015817997949702265,
      "loss": 5.1833,
      "mean_token_accuracy": 0.22675254493951796,
      "num_tokens": 156896230.0,
      "step": 77420
    },
    {
      "entropy": 5.905418920516968,
      "epoch": 6.766145241632439,
      "grad_norm": 1.125,
      "learning_rate": 0.0001581533487003947,
      "loss": 5.1753,
      "mean_token_accuracy": 0.22333962321281434,
      "num_tokens": 156906997.0,
      "step": 77425
    },
    {
      "entropy": 5.823601245880127,
      "epoch": 6.766582189985144,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001581267201449467,
      "loss": 5.0032,
      "mean_token_accuracy": 0.23754018396139145,
      "num_tokens": 156917150.0,
      "step": 77430
    },
    {
      "entropy": 5.897792625427246,
      "epoch": 6.767019138337848,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001581000938311896,
      "loss": 5.1158,
      "mean_token_accuracy": 0.2228675901889801,
      "num_tokens": 156926282.0,
      "step": 77435
    },
    {
      "entropy": 5.819175052642822,
      "epoch": 6.7674560866905535,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015807346975963386,
      "loss": 5.1064,
      "mean_token_accuracy": 0.23088180422782897,
      "num_tokens": 156936945.0,
      "step": 77440
    },
    {
      "entropy": 5.8075599193573,
      "epoch": 6.767893035043258,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001580468479307902,
      "loss": 5.0917,
      "mean_token_accuracy": 0.22618455588817596,
      "num_tokens": 156946705.0,
      "step": 77445
    },
    {
      "entropy": 5.848680114746093,
      "epoch": 6.7683299833959625,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015802022834516933,
      "loss": 5.1649,
      "mean_token_accuracy": 0.21929074823856354,
      "num_tokens": 156957484.0,
      "step": 77450
    },
    {
      "entropy": 5.75747389793396,
      "epoch": 6.768766931748667,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015799361100328173,
      "loss": 5.0636,
      "mean_token_accuracy": 0.23050119280815123,
      "num_tokens": 156966967.0,
      "step": 77455
    },
    {
      "entropy": 5.8644989967346195,
      "epoch": 6.769203880101372,
      "grad_norm": 1.078125,
      "learning_rate": 0.00015796699590563798,
      "loss": 5.1703,
      "mean_token_accuracy": 0.2247754529118538,
      "num_tokens": 156977700.0,
      "step": 77460
    },
    {
      "entropy": 5.8482043743133545,
      "epoch": 6.769640828454077,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015794038305274861,
      "loss": 5.1841,
      "mean_token_accuracy": 0.22637748271226882,
      "num_tokens": 156987615.0,
      "step": 77465
    },
    {
      "entropy": 5.907417249679566,
      "epoch": 6.770077776806781,
      "grad_norm": 1.21875,
      "learning_rate": 0.000157913772445124,
      "loss": 5.1798,
      "mean_token_accuracy": 0.22515495866537094,
      "num_tokens": 156997990.0,
      "step": 77470
    },
    {
      "entropy": 5.854130554199219,
      "epoch": 6.770514725159487,
      "grad_norm": 1.125,
      "learning_rate": 0.0001578871640832746,
      "loss": 5.1204,
      "mean_token_accuracy": 0.2238435834646225,
      "num_tokens": 157007935.0,
      "step": 77475
    },
    {
      "entropy": 5.8361741065979,
      "epoch": 6.770951673512191,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015786055796771075,
      "loss": 5.1148,
      "mean_token_accuracy": 0.2288148134946823,
      "num_tokens": 157018211.0,
      "step": 77480
    },
    {
      "entropy": 5.807733488082886,
      "epoch": 6.7713886218648955,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015783395409894288,
      "loss": 5.0833,
      "mean_token_accuracy": 0.23348670303821564,
      "num_tokens": 157027965.0,
      "step": 77485
    },
    {
      "entropy": 5.848285293579101,
      "epoch": 6.7718255702176,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015780735247748108,
      "loss": 5.142,
      "mean_token_accuracy": 0.22686953842639923,
      "num_tokens": 157038409.0,
      "step": 77490
    },
    {
      "entropy": 5.843658399581909,
      "epoch": 6.772262518570305,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015778075310383567,
      "loss": 5.1522,
      "mean_token_accuracy": 0.2254299059510231,
      "num_tokens": 157048087.0,
      "step": 77495
    },
    {
      "entropy": 5.766572904586792,
      "epoch": 6.77269946692301,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015775415597851683,
      "loss": 5.0561,
      "mean_token_accuracy": 0.23991589099168778,
      "num_tokens": 157057206.0,
      "step": 77500
    },
    {
      "entropy": 5.862856388092041,
      "epoch": 6.773136415275714,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001577275611020347,
      "loss": 5.1487,
      "mean_token_accuracy": 0.2266971915960312,
      "num_tokens": 157066327.0,
      "step": 77505
    },
    {
      "entropy": 5.768247175216675,
      "epoch": 6.773573363628419,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001577009684748994,
      "loss": 5.0626,
      "mean_token_accuracy": 0.23119314163923263,
      "num_tokens": 157077550.0,
      "step": 77510
    },
    {
      "entropy": 5.863418865203857,
      "epoch": 6.774010311981124,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000157674378097621,
      "loss": 5.1784,
      "mean_token_accuracy": 0.2167935401201248,
      "num_tokens": 157087804.0,
      "step": 77515
    },
    {
      "entropy": 5.8859700679779055,
      "epoch": 6.774447260333829,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001576477899707095,
      "loss": 5.1759,
      "mean_token_accuracy": 0.21806812286376953,
      "num_tokens": 157097722.0,
      "step": 77520
    },
    {
      "entropy": 5.9053701877594,
      "epoch": 6.774884208686533,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001576212040946749,
      "loss": 5.237,
      "mean_token_accuracy": 0.2197875425219536,
      "num_tokens": 157107826.0,
      "step": 77525
    },
    {
      "entropy": 5.8437329769134525,
      "epoch": 6.775321157039238,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015759462047002716,
      "loss": 5.0664,
      "mean_token_accuracy": 0.23099480867385863,
      "num_tokens": 157116976.0,
      "step": 77530
    },
    {
      "entropy": 5.798198270797729,
      "epoch": 6.775758105391943,
      "grad_norm": 1.15625,
      "learning_rate": 0.00015756803909727606,
      "loss": 5.0523,
      "mean_token_accuracy": 0.2302216425538063,
      "num_tokens": 157126340.0,
      "step": 77535
    },
    {
      "entropy": 5.84760046005249,
      "epoch": 6.776195053744647,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001575414599769315,
      "loss": 5.1121,
      "mean_token_accuracy": 0.2375106006860733,
      "num_tokens": 157136013.0,
      "step": 77540
    },
    {
      "entropy": 5.858787250518799,
      "epoch": 6.776632002097352,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015751488310950326,
      "loss": 5.1643,
      "mean_token_accuracy": 0.2223638713359833,
      "num_tokens": 157145903.0,
      "step": 77545
    },
    {
      "entropy": 5.875908088684082,
      "epoch": 6.777068950450056,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00015748830849550112,
      "loss": 5.2184,
      "mean_token_accuracy": 0.21587470173835754,
      "num_tokens": 157156742.0,
      "step": 77550
    },
    {
      "entropy": 5.8739830493927006,
      "epoch": 6.777505898802762,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015746173613543485,
      "loss": 5.1498,
      "mean_token_accuracy": 0.22191282957792283,
      "num_tokens": 157165998.0,
      "step": 77555
    },
    {
      "entropy": 5.8491530418396,
      "epoch": 6.777942847155466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015743516602981407,
      "loss": 5.1781,
      "mean_token_accuracy": 0.22243682146072388,
      "num_tokens": 157176016.0,
      "step": 77560
    },
    {
      "entropy": 5.885302591323852,
      "epoch": 6.778379795508171,
      "grad_norm": 1.15625,
      "learning_rate": 0.00015740859817914833,
      "loss": 5.2076,
      "mean_token_accuracy": 0.22035008817911148,
      "num_tokens": 157186017.0,
      "step": 77565
    },
    {
      "entropy": 5.888958168029785,
      "epoch": 6.778816743860876,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015738203258394736,
      "loss": 5.1536,
      "mean_token_accuracy": 0.22956814616918564,
      "num_tokens": 157196730.0,
      "step": 77570
    },
    {
      "entropy": 5.843463373184204,
      "epoch": 6.77925369221358,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015735546924472064,
      "loss": 5.0676,
      "mean_token_accuracy": 0.2293557345867157,
      "num_tokens": 157205868.0,
      "step": 77575
    },
    {
      "entropy": 5.897561836242676,
      "epoch": 6.779690640566285,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001573289081619777,
      "loss": 5.1463,
      "mean_token_accuracy": 0.22535615265369416,
      "num_tokens": 157215869.0,
      "step": 77580
    },
    {
      "entropy": 5.909316492080689,
      "epoch": 6.78012758891899,
      "grad_norm": 1.25,
      "learning_rate": 0.00015730234933622792,
      "loss": 5.1916,
      "mean_token_accuracy": 0.2230176955461502,
      "num_tokens": 157226312.0,
      "step": 77585
    },
    {
      "entropy": 5.831169605255127,
      "epoch": 6.780564537271695,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015727579276798073,
      "loss": 5.0851,
      "mean_token_accuracy": 0.23809460699558258,
      "num_tokens": 157235943.0,
      "step": 77590
    },
    {
      "entropy": 5.841069650650025,
      "epoch": 6.781001485624399,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015724923845774548,
      "loss": 5.0811,
      "mean_token_accuracy": 0.22465719729661943,
      "num_tokens": 157246211.0,
      "step": 77595
    },
    {
      "entropy": 5.8310174465179445,
      "epoch": 6.781438433977104,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015722268640603153,
      "loss": 5.1398,
      "mean_token_accuracy": 0.21937483698129653,
      "num_tokens": 157256353.0,
      "step": 77600
    },
    {
      "entropy": 5.810041284561157,
      "epoch": 6.781875382329808,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015719613661334826,
      "loss": 5.1345,
      "mean_token_accuracy": 0.22740747034549713,
      "num_tokens": 157267399.0,
      "step": 77605
    },
    {
      "entropy": 5.856828260421753,
      "epoch": 6.7823123306825135,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015716958908020478,
      "loss": 5.1455,
      "mean_token_accuracy": 0.22552742511034013,
      "num_tokens": 157277772.0,
      "step": 77610
    },
    {
      "entropy": 5.8344334125518795,
      "epoch": 6.782749279035218,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001571430438071103,
      "loss": 5.1561,
      "mean_token_accuracy": 0.2215922012925148,
      "num_tokens": 157287745.0,
      "step": 77615
    },
    {
      "entropy": 5.888423871994019,
      "epoch": 6.783186227387922,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015711650079457414,
      "loss": 5.1766,
      "mean_token_accuracy": 0.2265632778406143,
      "num_tokens": 157297507.0,
      "step": 77620
    },
    {
      "entropy": 5.82788314819336,
      "epoch": 6.783623175740628,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001570899600431051,
      "loss": 5.0923,
      "mean_token_accuracy": 0.23129785060882568,
      "num_tokens": 157307071.0,
      "step": 77625
    },
    {
      "entropy": 5.833160448074341,
      "epoch": 6.784060124093332,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015706342155321246,
      "loss": 5.1086,
      "mean_token_accuracy": 0.23440664410591125,
      "num_tokens": 157317375.0,
      "step": 77630
    },
    {
      "entropy": 5.799065208435058,
      "epoch": 6.784497072446037,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001570368853254052,
      "loss": 5.0515,
      "mean_token_accuracy": 0.2274632841348648,
      "num_tokens": 157327388.0,
      "step": 77635
    },
    {
      "entropy": 5.790326261520386,
      "epoch": 6.784934020798742,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015701035136019226,
      "loss": 5.0283,
      "mean_token_accuracy": 0.23869227916002272,
      "num_tokens": 157337465.0,
      "step": 77640
    },
    {
      "entropy": 5.814453554153443,
      "epoch": 6.7853709691514466,
      "grad_norm": 1.25,
      "learning_rate": 0.00015698381965808267,
      "loss": 5.0833,
      "mean_token_accuracy": 0.22569558024406433,
      "num_tokens": 157348230.0,
      "step": 77645
    },
    {
      "entropy": 5.827569818496704,
      "epoch": 6.785807917504151,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001569572902195852,
      "loss": 5.0616,
      "mean_token_accuracy": 0.23446974009275437,
      "num_tokens": 157357580.0,
      "step": 77650
    },
    {
      "entropy": 5.751363706588745,
      "epoch": 6.7862448658568555,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015693076304520878,
      "loss": 4.9568,
      "mean_token_accuracy": 0.24412390738725662,
      "num_tokens": 157367335.0,
      "step": 77655
    },
    {
      "entropy": 5.825822401046753,
      "epoch": 6.78668181420956,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015690423813546217,
      "loss": 5.1528,
      "mean_token_accuracy": 0.22924420684576036,
      "num_tokens": 157378955.0,
      "step": 77660
    },
    {
      "entropy": 5.868398714065552,
      "epoch": 6.787118762562265,
      "grad_norm": 1.109375,
      "learning_rate": 0.00015687771549085426,
      "loss": 5.1614,
      "mean_token_accuracy": 0.21914084255695343,
      "num_tokens": 157390023.0,
      "step": 77665
    },
    {
      "entropy": 5.874795293807983,
      "epoch": 6.78755571091497,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015685119511189355,
      "loss": 5.1388,
      "mean_token_accuracy": 0.22577503472566604,
      "num_tokens": 157400403.0,
      "step": 77670
    },
    {
      "entropy": 5.805052328109741,
      "epoch": 6.787992659267674,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00015682467699908886,
      "loss": 5.0854,
      "mean_token_accuracy": 0.23006168603897095,
      "num_tokens": 157412322.0,
      "step": 77675
    },
    {
      "entropy": 5.820244073867798,
      "epoch": 6.78842960762038,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001567981611529488,
      "loss": 5.1607,
      "mean_token_accuracy": 0.227692349255085,
      "num_tokens": 157423574.0,
      "step": 77680
    },
    {
      "entropy": 5.834670877456665,
      "epoch": 6.788866555973084,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001567716475739819,
      "loss": 5.1181,
      "mean_token_accuracy": 0.22672139406204223,
      "num_tokens": 157433209.0,
      "step": 77685
    },
    {
      "entropy": 5.864950799942017,
      "epoch": 6.7893035043257886,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015674513626269675,
      "loss": 5.0949,
      "mean_token_accuracy": 0.22413283437490464,
      "num_tokens": 157442827.0,
      "step": 77690
    },
    {
      "entropy": 5.8340935707092285,
      "epoch": 6.789740452678493,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015671862721960189,
      "loss": 5.077,
      "mean_token_accuracy": 0.229316246509552,
      "num_tokens": 157452198.0,
      "step": 77695
    },
    {
      "entropy": 5.89455132484436,
      "epoch": 6.790177401031198,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015669212044520575,
      "loss": 5.1324,
      "mean_token_accuracy": 0.22165191024541855,
      "num_tokens": 157462062.0,
      "step": 77700
    },
    {
      "entropy": 5.87935357093811,
      "epoch": 6.790614349383903,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015666561594001666,
      "loss": 5.2344,
      "mean_token_accuracy": 0.22422745376825332,
      "num_tokens": 157472181.0,
      "step": 77705
    },
    {
      "entropy": 5.834589290618896,
      "epoch": 6.791051297736607,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015663911370454312,
      "loss": 5.0915,
      "mean_token_accuracy": 0.23211159110069274,
      "num_tokens": 157482446.0,
      "step": 77710
    },
    {
      "entropy": 5.8741356372833256,
      "epoch": 6.791488246089312,
      "grad_norm": 1.15625,
      "learning_rate": 0.00015661261373929347,
      "loss": 5.2064,
      "mean_token_accuracy": 0.22345095723867417,
      "num_tokens": 157493024.0,
      "step": 77715
    },
    {
      "entropy": 5.877835035324097,
      "epoch": 6.791925194442017,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015658611604477585,
      "loss": 5.168,
      "mean_token_accuracy": 0.21801366806030273,
      "num_tokens": 157502357.0,
      "step": 77720
    },
    {
      "entropy": 5.822451448440551,
      "epoch": 6.792362142794722,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015655962062149855,
      "loss": 5.0965,
      "mean_token_accuracy": 0.2323528453707695,
      "num_tokens": 157511938.0,
      "step": 77725
    },
    {
      "entropy": 5.874196624755859,
      "epoch": 6.792799091147426,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001565331274699698,
      "loss": 5.1721,
      "mean_token_accuracy": 0.22312053591012954,
      "num_tokens": 157523317.0,
      "step": 77730
    },
    {
      "entropy": 5.775043582916259,
      "epoch": 6.793236039500131,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015650663659069774,
      "loss": 5.0635,
      "mean_token_accuracy": 0.23378423750400543,
      "num_tokens": 157534079.0,
      "step": 77735
    },
    {
      "entropy": 5.825995683670044,
      "epoch": 6.793672987852836,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015648014798419046,
      "loss": 5.0774,
      "mean_token_accuracy": 0.2272651880979538,
      "num_tokens": 157543700.0,
      "step": 77740
    },
    {
      "entropy": 5.890455818176269,
      "epoch": 6.79410993620554,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015645366165095608,
      "loss": 5.147,
      "mean_token_accuracy": 0.21900539398193358,
      "num_tokens": 157553722.0,
      "step": 77745
    },
    {
      "entropy": 5.871105146408081,
      "epoch": 6.794546884558245,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015642717759150256,
      "loss": 5.1344,
      "mean_token_accuracy": 0.22614166289567947,
      "num_tokens": 157563379.0,
      "step": 77750
    },
    {
      "entropy": 5.833161687850952,
      "epoch": 6.79498383291095,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015640069580633793,
      "loss": 5.1563,
      "mean_token_accuracy": 0.22599392980337143,
      "num_tokens": 157572687.0,
      "step": 77755
    },
    {
      "entropy": 5.846201324462891,
      "epoch": 6.795420781263655,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001563742162959701,
      "loss": 5.1509,
      "mean_token_accuracy": 0.2230842739343643,
      "num_tokens": 157582297.0,
      "step": 77760
    },
    {
      "entropy": 5.829386043548584,
      "epoch": 6.795857729616359,
      "grad_norm": 1.109375,
      "learning_rate": 0.000156347739060907,
      "loss": 5.0711,
      "mean_token_accuracy": 0.2376772090792656,
      "num_tokens": 157592931.0,
      "step": 77765
    },
    {
      "entropy": 5.788301420211792,
      "epoch": 6.796294677969064,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001563212641016564,
      "loss": 5.1135,
      "mean_token_accuracy": 0.22289087921380996,
      "num_tokens": 157601928.0,
      "step": 77770
    },
    {
      "entropy": 5.8186437606811525,
      "epoch": 6.796731626321769,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001562947914187261,
      "loss": 5.0961,
      "mean_token_accuracy": 0.22832984775304793,
      "num_tokens": 157611727.0,
      "step": 77775
    },
    {
      "entropy": 5.844958543777466,
      "epoch": 6.797168574674473,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015626832101262394,
      "loss": 5.1878,
      "mean_token_accuracy": 0.22166377902030945,
      "num_tokens": 157621645.0,
      "step": 77780
    },
    {
      "entropy": 5.831177377700806,
      "epoch": 6.797605523027178,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015624185288385758,
      "loss": 5.1506,
      "mean_token_accuracy": 0.2292197197675705,
      "num_tokens": 157632366.0,
      "step": 77785
    },
    {
      "entropy": 5.873817014694214,
      "epoch": 6.798042471379883,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001562153870329347,
      "loss": 5.1289,
      "mean_token_accuracy": 0.22216730564832687,
      "num_tokens": 157641080.0,
      "step": 77790
    },
    {
      "entropy": 5.906739854812622,
      "epoch": 6.798479419732588,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015618892346036295,
      "loss": 5.1791,
      "mean_token_accuracy": 0.22526113390922547,
      "num_tokens": 157650933.0,
      "step": 77795
    },
    {
      "entropy": 5.86976580619812,
      "epoch": 6.798916368085292,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015616246216664992,
      "loss": 5.1237,
      "mean_token_accuracy": 0.22412583231925964,
      "num_tokens": 157661405.0,
      "step": 77800
    },
    {
      "entropy": 5.8927994728088375,
      "epoch": 6.799353316437997,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001561360031523031,
      "loss": 5.1213,
      "mean_token_accuracy": 0.2245055243372917,
      "num_tokens": 157672841.0,
      "step": 77805
    },
    {
      "entropy": 5.795952033996582,
      "epoch": 6.799790264790702,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015610954641783007,
      "loss": 5.0173,
      "mean_token_accuracy": 0.23921581357717514,
      "num_tokens": 157684417.0,
      "step": 77810
    },
    {
      "entropy": 5.7637341022491455,
      "epoch": 6.8002272131434065,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015608309196373828,
      "loss": 5.0197,
      "mean_token_accuracy": 0.2386419489979744,
      "num_tokens": 157695046.0,
      "step": 77815
    },
    {
      "entropy": 5.791245698928833,
      "epoch": 6.800664161496111,
      "grad_norm": 1.25,
      "learning_rate": 0.00015605663979053502,
      "loss": 5.0849,
      "mean_token_accuracy": 0.22614708989858628,
      "num_tokens": 157705354.0,
      "step": 77820
    },
    {
      "entropy": 5.802765703201294,
      "epoch": 6.801101109848815,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015603018989872775,
      "loss": 5.157,
      "mean_token_accuracy": 0.2297256335616112,
      "num_tokens": 157715128.0,
      "step": 77825
    },
    {
      "entropy": 5.804295063018799,
      "epoch": 6.801538058201521,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00015600374228882377,
      "loss": 5.0938,
      "mean_token_accuracy": 0.23254389613866805,
      "num_tokens": 157725390.0,
      "step": 77830
    },
    {
      "entropy": 5.83885498046875,
      "epoch": 6.801975006554225,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015597729696133038,
      "loss": 5.1298,
      "mean_token_accuracy": 0.22804169803857804,
      "num_tokens": 157735898.0,
      "step": 77835
    },
    {
      "entropy": 5.883774995803833,
      "epoch": 6.80241195490693,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001559508539167548,
      "loss": 5.164,
      "mean_token_accuracy": 0.21801635324954988,
      "num_tokens": 157745846.0,
      "step": 77840
    },
    {
      "entropy": 5.906780862808228,
      "epoch": 6.802848903259635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015592441315560418,
      "loss": 5.1609,
      "mean_token_accuracy": 0.22058830708265303,
      "num_tokens": 157756550.0,
      "step": 77845
    },
    {
      "entropy": 5.90163025856018,
      "epoch": 6.80328585161234,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015589797467838575,
      "loss": 5.1692,
      "mean_token_accuracy": 0.2207098752260208,
      "num_tokens": 157766751.0,
      "step": 77850
    },
    {
      "entropy": 5.8200562477111815,
      "epoch": 6.803722799965044,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015587153848560653,
      "loss": 5.1341,
      "mean_token_accuracy": 0.22446669787168502,
      "num_tokens": 157777312.0,
      "step": 77855
    },
    {
      "entropy": 5.88658561706543,
      "epoch": 6.8041597483177485,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015584510457777376,
      "loss": 5.1947,
      "mean_token_accuracy": 0.22494601607322692,
      "num_tokens": 157786980.0,
      "step": 77860
    },
    {
      "entropy": 5.9156081199646,
      "epoch": 6.804596696670454,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015581867295539424,
      "loss": 5.1846,
      "mean_token_accuracy": 0.22318530231714248,
      "num_tokens": 157798256.0,
      "step": 77865
    },
    {
      "entropy": 5.896460676193238,
      "epoch": 6.805033645023158,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015579224361897498,
      "loss": 5.1239,
      "mean_token_accuracy": 0.2261381134390831,
      "num_tokens": 157807723.0,
      "step": 77870
    },
    {
      "entropy": 5.8945239067077635,
      "epoch": 6.805470593375863,
      "grad_norm": 1.078125,
      "learning_rate": 0.00015576581656902297,
      "loss": 5.1729,
      "mean_token_accuracy": 0.22272953540086746,
      "num_tokens": 157818019.0,
      "step": 77875
    },
    {
      "entropy": 5.801969861984253,
      "epoch": 6.805907541728567,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015573939180604508,
      "loss": 5.0558,
      "mean_token_accuracy": 0.24040584564208983,
      "num_tokens": 157827652.0,
      "step": 77880
    },
    {
      "entropy": 5.818512487411499,
      "epoch": 6.806344490081273,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015571296933054816,
      "loss": 5.1447,
      "mean_token_accuracy": 0.22515862435102463,
      "num_tokens": 157838007.0,
      "step": 77885
    },
    {
      "entropy": 5.8145081996917725,
      "epoch": 6.806781438433977,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015568654914303897,
      "loss": 5.0824,
      "mean_token_accuracy": 0.22696779519319535,
      "num_tokens": 157848704.0,
      "step": 77890
    },
    {
      "entropy": 5.826266527175903,
      "epoch": 6.807218386786682,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001556601312440243,
      "loss": 5.1397,
      "mean_token_accuracy": 0.2248929798603058,
      "num_tokens": 157858923.0,
      "step": 77895
    },
    {
      "entropy": 5.867834281921387,
      "epoch": 6.807655335139387,
      "grad_norm": 1.09375,
      "learning_rate": 0.00015563371563401086,
      "loss": 5.1212,
      "mean_token_accuracy": 0.23003881573677062,
      "num_tokens": 157870751.0,
      "step": 77900
    },
    {
      "entropy": 5.840557098388672,
      "epoch": 6.808092283492091,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015560730231350527,
      "loss": 5.117,
      "mean_token_accuracy": 0.23676239252090453,
      "num_tokens": 157881139.0,
      "step": 77905
    },
    {
      "entropy": 5.804522895812989,
      "epoch": 6.808529231844796,
      "grad_norm": 1.125,
      "learning_rate": 0.00015558089128301427,
      "loss": 5.0499,
      "mean_token_accuracy": 0.24003324955701827,
      "num_tokens": 157890921.0,
      "step": 77910
    },
    {
      "entropy": 5.823206281661987,
      "epoch": 6.8089661801975,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015555448254304428,
      "loss": 5.0535,
      "mean_token_accuracy": 0.22885964810848236,
      "num_tokens": 157902181.0,
      "step": 77915
    },
    {
      "entropy": 5.838962841033935,
      "epoch": 6.809403128550206,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001555280760941019,
      "loss": 5.0903,
      "mean_token_accuracy": 0.22815290093421936,
      "num_tokens": 157911400.0,
      "step": 77920
    },
    {
      "entropy": 5.849265909194946,
      "epoch": 6.80984007690291,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015550167193669362,
      "loss": 5.175,
      "mean_token_accuracy": 0.2199573665857315,
      "num_tokens": 157921069.0,
      "step": 77925
    },
    {
      "entropy": 5.860534811019898,
      "epoch": 6.810277025255615,
      "grad_norm": 1.25,
      "learning_rate": 0.0001554752700713259,
      "loss": 5.1101,
      "mean_token_accuracy": 0.22896616011857987,
      "num_tokens": 157931516.0,
      "step": 77930
    },
    {
      "entropy": 5.796429586410523,
      "epoch": 6.810713973608319,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001554488704985051,
      "loss": 5.1016,
      "mean_token_accuracy": 0.22502542585134505,
      "num_tokens": 157941914.0,
      "step": 77935
    },
    {
      "entropy": 5.881075763702393,
      "epoch": 6.8111509219610245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015542247321873764,
      "loss": 5.154,
      "mean_token_accuracy": 0.2234971418976784,
      "num_tokens": 157951530.0,
      "step": 77940
    },
    {
      "entropy": 5.875495576858521,
      "epoch": 6.811587870313729,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015539607823252978,
      "loss": 5.1372,
      "mean_token_accuracy": 0.22797675281763077,
      "num_tokens": 157962230.0,
      "step": 77945
    },
    {
      "entropy": 5.805218648910523,
      "epoch": 6.812024818666433,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015536968554038784,
      "loss": 5.0206,
      "mean_token_accuracy": 0.23634438961744308,
      "num_tokens": 157971185.0,
      "step": 77950
    },
    {
      "entropy": 5.874067592620849,
      "epoch": 6.812461767019139,
      "grad_norm": 1.265625,
      "learning_rate": 0.000155343295142818,
      "loss": 5.1321,
      "mean_token_accuracy": 0.2122834727168083,
      "num_tokens": 157980173.0,
      "step": 77955
    },
    {
      "entropy": 5.868576622009277,
      "epoch": 6.812898715371843,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001553169070403264,
      "loss": 5.2017,
      "mean_token_accuracy": 0.225213123857975,
      "num_tokens": 157989440.0,
      "step": 77960
    },
    {
      "entropy": 5.869773626327515,
      "epoch": 6.813335663724548,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001552905212334193,
      "loss": 5.1772,
      "mean_token_accuracy": 0.22477304339408874,
      "num_tokens": 158000752.0,
      "step": 77965
    },
    {
      "entropy": 5.803919267654419,
      "epoch": 6.813772612077252,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015526413772260273,
      "loss": 5.0773,
      "mean_token_accuracy": 0.22794874906539916,
      "num_tokens": 158010906.0,
      "step": 77970
    },
    {
      "entropy": 5.764393949508667,
      "epoch": 6.8142095604299575,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015523775650838273,
      "loss": 5.065,
      "mean_token_accuracy": 0.23782073706388474,
      "num_tokens": 158020849.0,
      "step": 77975
    },
    {
      "entropy": 5.940878295898438,
      "epoch": 6.814646508782662,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015521137759126532,
      "loss": 5.2148,
      "mean_token_accuracy": 0.2163599669933319,
      "num_tokens": 158031367.0,
      "step": 77980
    },
    {
      "entropy": 5.8792235374450685,
      "epoch": 6.8150834571353665,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015518500097175643,
      "loss": 5.1532,
      "mean_token_accuracy": 0.2229325294494629,
      "num_tokens": 158041066.0,
      "step": 77985
    },
    {
      "entropy": 5.902747917175293,
      "epoch": 6.815520405488071,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015515862665036207,
      "loss": 5.2477,
      "mean_token_accuracy": 0.21717484891414643,
      "num_tokens": 158050701.0,
      "step": 77990
    },
    {
      "entropy": 5.878462600708008,
      "epoch": 6.815957353840776,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015513225462758808,
      "loss": 5.1443,
      "mean_token_accuracy": 0.22678849697113038,
      "num_tokens": 158060294.0,
      "step": 77995
    },
    {
      "entropy": 5.894906044006348,
      "epoch": 6.816394302193481,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015510588490394014,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21557270735502243,
      "num_tokens": 158069814.0,
      "step": 78000
    },
    {
      "epoch": 6.816394302193481,
      "eval_entropy": 5.589371754640324,
      "eval_loss": 5.239040851593018,
      "eval_mean_token_accuracy": 0.2279790382593168,
      "eval_num_tokens": 158069814.0,
      "eval_runtime": 27.9138,
      "eval_samples_per_second": 1326.013,
      "eval_steps_per_second": 165.761,
      "step": 78000
    },
    {
      "entropy": 5.895366859436035,
      "epoch": 6.816831250546185,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015507951747992422,
      "loss": 5.1782,
      "mean_token_accuracy": 0.22453705668449403,
      "num_tokens": 158080054.0,
      "step": 78005
    },
    {
      "entropy": 5.885518503189087,
      "epoch": 6.817268198898891,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015505315235604593,
      "loss": 5.1421,
      "mean_token_accuracy": 0.2313608020544052,
      "num_tokens": 158089471.0,
      "step": 78010
    },
    {
      "entropy": 5.7947550296783445,
      "epoch": 6.817705147251595,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015502678953281107,
      "loss": 5.0641,
      "mean_token_accuracy": 0.23552024513483047,
      "num_tokens": 158100160.0,
      "step": 78015
    },
    {
      "entropy": 5.859159183502197,
      "epoch": 6.8181420956042995,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015500042901072522,
      "loss": 5.1465,
      "mean_token_accuracy": 0.2214503437280655,
      "num_tokens": 158110739.0,
      "step": 78020
    },
    {
      "entropy": 5.783699989318848,
      "epoch": 6.818579043957004,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001549740707902941,
      "loss": 5.0697,
      "mean_token_accuracy": 0.23483711183071138,
      "num_tokens": 158121055.0,
      "step": 78025
    },
    {
      "entropy": 5.746672010421753,
      "epoch": 6.819015992309709,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00015494771487202307,
      "loss": 5.0267,
      "mean_token_accuracy": 0.2378452330827713,
      "num_tokens": 158131171.0,
      "step": 78030
    },
    {
      "entropy": 5.858074140548706,
      "epoch": 6.819452940662414,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015492136125641788,
      "loss": 5.1251,
      "mean_token_accuracy": 0.22538240998983383,
      "num_tokens": 158141226.0,
      "step": 78035
    },
    {
      "entropy": 5.8094981670379635,
      "epoch": 6.819889889015118,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015489500994398384,
      "loss": 5.1509,
      "mean_token_accuracy": 0.2219283401966095,
      "num_tokens": 158150454.0,
      "step": 78040
    },
    {
      "entropy": 5.840328979492187,
      "epoch": 6.820326837367823,
      "grad_norm": 1.109375,
      "learning_rate": 0.00015486866093522656,
      "loss": 5.1365,
      "mean_token_accuracy": 0.22216875851154327,
      "num_tokens": 158161310.0,
      "step": 78045
    },
    {
      "entropy": 5.8825750827789305,
      "epoch": 6.820763785720528,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015484231423065122,
      "loss": 5.1706,
      "mean_token_accuracy": 0.22206990718841552,
      "num_tokens": 158171708.0,
      "step": 78050
    },
    {
      "entropy": 5.82245945930481,
      "epoch": 6.821200734073233,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015481596983076324,
      "loss": 5.0998,
      "mean_token_accuracy": 0.2331772655248642,
      "num_tokens": 158181734.0,
      "step": 78055
    },
    {
      "entropy": 5.845136070251465,
      "epoch": 6.821637682425937,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015478962773606797,
      "loss": 5.1302,
      "mean_token_accuracy": 0.23084196746349334,
      "num_tokens": 158190988.0,
      "step": 78060
    },
    {
      "entropy": 5.791448926925659,
      "epoch": 6.822074630778642,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015476328794707057,
      "loss": 5.0964,
      "mean_token_accuracy": 0.23357478082180022,
      "num_tokens": 158200305.0,
      "step": 78065
    },
    {
      "entropy": 5.903768014907837,
      "epoch": 6.822511579131347,
      "grad_norm": 1.25,
      "learning_rate": 0.0001547369504642764,
      "loss": 5.2082,
      "mean_token_accuracy": 0.22270992249250413,
      "num_tokens": 158210225.0,
      "step": 78070
    },
    {
      "entropy": 5.824400281906128,
      "epoch": 6.822948527484051,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015471061528819042,
      "loss": 5.12,
      "mean_token_accuracy": 0.2290481746196747,
      "num_tokens": 158219969.0,
      "step": 78075
    },
    {
      "entropy": 5.898038291931153,
      "epoch": 6.823385475836756,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015468428241931795,
      "loss": 5.1975,
      "mean_token_accuracy": 0.21977114975452422,
      "num_tokens": 158230308.0,
      "step": 78080
    },
    {
      "entropy": 5.8401233673095705,
      "epoch": 6.823822424189461,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015465795185816395,
      "loss": 5.1217,
      "mean_token_accuracy": 0.2293207585811615,
      "num_tokens": 158240104.0,
      "step": 78085
    },
    {
      "entropy": 5.812691164016724,
      "epoch": 6.824259372542166,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015463162360523352,
      "loss": 5.0961,
      "mean_token_accuracy": 0.2306613251566887,
      "num_tokens": 158249652.0,
      "step": 78090
    },
    {
      "entropy": 5.884314346313476,
      "epoch": 6.82469632089487,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015460529766103172,
      "loss": 5.1508,
      "mean_token_accuracy": 0.22083186656236647,
      "num_tokens": 158261474.0,
      "step": 78095
    },
    {
      "entropy": 5.862444591522217,
      "epoch": 6.825133269247575,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015457897402606328,
      "loss": 5.1313,
      "mean_token_accuracy": 0.2314580500125885,
      "num_tokens": 158272192.0,
      "step": 78100
    },
    {
      "entropy": 5.874751329421997,
      "epoch": 6.82557021760028,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015455265270083318,
      "loss": 5.1791,
      "mean_token_accuracy": 0.2210135877132416,
      "num_tokens": 158281913.0,
      "step": 78105
    },
    {
      "entropy": 5.837364387512207,
      "epoch": 6.826007165952984,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001545263336858464,
      "loss": 5.1709,
      "mean_token_accuracy": 0.2241182282567024,
      "num_tokens": 158291626.0,
      "step": 78110
    },
    {
      "entropy": 5.873158264160156,
      "epoch": 6.826444114305689,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001545000169816076,
      "loss": 5.2699,
      "mean_token_accuracy": 0.2174067184329033,
      "num_tokens": 158303065.0,
      "step": 78115
    },
    {
      "entropy": 5.8529270648956295,
      "epoch": 6.826881062658394,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015447370258862164,
      "loss": 5.115,
      "mean_token_accuracy": 0.2298390045762062,
      "num_tokens": 158313097.0,
      "step": 78120
    },
    {
      "entropy": 5.849100875854492,
      "epoch": 6.827318011011099,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015444739050739323,
      "loss": 5.147,
      "mean_token_accuracy": 0.22816194146871566,
      "num_tokens": 158323107.0,
      "step": 78125
    },
    {
      "entropy": 5.8732294082641605,
      "epoch": 6.827754959363803,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015442108073842703,
      "loss": 5.0573,
      "mean_token_accuracy": 0.24138352274894714,
      "num_tokens": 158333375.0,
      "step": 78130
    },
    {
      "entropy": 5.8012631893157955,
      "epoch": 6.828191907716508,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015439477328222767,
      "loss": 5.1126,
      "mean_token_accuracy": 0.2285103067755699,
      "num_tokens": 158342982.0,
      "step": 78135
    },
    {
      "entropy": 5.849631500244141,
      "epoch": 6.828628856069213,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015436846813929986,
      "loss": 5.1795,
      "mean_token_accuracy": 0.2269661918282509,
      "num_tokens": 158352570.0,
      "step": 78140
    },
    {
      "entropy": 5.868309259414673,
      "epoch": 6.8290658044219175,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015434216531014797,
      "loss": 5.2007,
      "mean_token_accuracy": 0.21589485853910445,
      "num_tokens": 158363645.0,
      "step": 78145
    },
    {
      "entropy": 5.849125146865845,
      "epoch": 6.829502752774622,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015431586479527654,
      "loss": 5.1237,
      "mean_token_accuracy": 0.2258938357234001,
      "num_tokens": 158373903.0,
      "step": 78150
    },
    {
      "entropy": 5.923571538925171,
      "epoch": 6.829939701127326,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001542895665951901,
      "loss": 5.1691,
      "mean_token_accuracy": 0.21705779433250427,
      "num_tokens": 158383740.0,
      "step": 78155
    },
    {
      "entropy": 5.865497016906739,
      "epoch": 6.830376649480032,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015426327071039298,
      "loss": 5.1658,
      "mean_token_accuracy": 0.22584390193223952,
      "num_tokens": 158393913.0,
      "step": 78160
    },
    {
      "entropy": 5.888759899139404,
      "epoch": 6.830813597832736,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015423697714138962,
      "loss": 5.1817,
      "mean_token_accuracy": 0.21667279303073883,
      "num_tokens": 158404050.0,
      "step": 78165
    },
    {
      "entropy": 5.922099876403808,
      "epoch": 6.831250546185441,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001542106858886843,
      "loss": 5.1594,
      "mean_token_accuracy": 0.22735550105571747,
      "num_tokens": 158413524.0,
      "step": 78170
    },
    {
      "entropy": 5.815118312835693,
      "epoch": 6.831687494538146,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015418439695278133,
      "loss": 5.064,
      "mean_token_accuracy": 0.2340245723724365,
      "num_tokens": 158423798.0,
      "step": 78175
    },
    {
      "entropy": 5.857838773727417,
      "epoch": 6.8321244428908505,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015415811033418493,
      "loss": 5.15,
      "mean_token_accuracy": 0.22210167944431305,
      "num_tokens": 158433662.0,
      "step": 78180
    },
    {
      "entropy": 5.853578042984009,
      "epoch": 6.832561391243555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015413182603339927,
      "loss": 5.1577,
      "mean_token_accuracy": 0.22633693665266036,
      "num_tokens": 158442940.0,
      "step": 78185
    },
    {
      "entropy": 5.876735639572144,
      "epoch": 6.8329983395962595,
      "grad_norm": 1.15625,
      "learning_rate": 0.00015410554405092863,
      "loss": 5.129,
      "mean_token_accuracy": 0.23233738243579866,
      "num_tokens": 158453408.0,
      "step": 78190
    },
    {
      "entropy": 5.800921392440796,
      "epoch": 6.833435287948965,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015407926438727694,
      "loss": 5.1223,
      "mean_token_accuracy": 0.23125743567943574,
      "num_tokens": 158463903.0,
      "step": 78195
    },
    {
      "entropy": 5.81786847114563,
      "epoch": 6.833872236301669,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015405298704294828,
      "loss": 5.1102,
      "mean_token_accuracy": 0.23151327818632125,
      "num_tokens": 158473251.0,
      "step": 78200
    },
    {
      "entropy": 5.844295883178711,
      "epoch": 6.834309184654374,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001540267120184467,
      "loss": 5.0667,
      "mean_token_accuracy": 0.23691743463277817,
      "num_tokens": 158482840.0,
      "step": 78205
    },
    {
      "entropy": 5.819185447692871,
      "epoch": 6.834746133007078,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001540004393142762,
      "loss": 5.1115,
      "mean_token_accuracy": 0.2302337870001793,
      "num_tokens": 158492753.0,
      "step": 78210
    },
    {
      "entropy": 5.879928731918335,
      "epoch": 6.835183081359784,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001539741689309407,
      "loss": 5.2366,
      "mean_token_accuracy": 0.21829414665699004,
      "num_tokens": 158503365.0,
      "step": 78215
    },
    {
      "entropy": 5.831743812561035,
      "epoch": 6.835620029712488,
      "grad_norm": 1.296875,
      "learning_rate": 0.000153947900868944,
      "loss": 5.0913,
      "mean_token_accuracy": 0.2297557607293129,
      "num_tokens": 158513212.0,
      "step": 78220
    },
    {
      "entropy": 5.816415596008301,
      "epoch": 6.8360569780651925,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015392163512879003,
      "loss": 5.1235,
      "mean_token_accuracy": 0.23074107617139816,
      "num_tokens": 158522679.0,
      "step": 78225
    },
    {
      "entropy": 5.854680824279785,
      "epoch": 6.836493926417898,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001538953717109825,
      "loss": 5.1539,
      "mean_token_accuracy": 0.22553349137306214,
      "num_tokens": 158533532.0,
      "step": 78230
    },
    {
      "entropy": 5.8410125255584715,
      "epoch": 6.836930874770602,
      "grad_norm": 1.15625,
      "learning_rate": 0.00015386911061602524,
      "loss": 5.0981,
      "mean_token_accuracy": 0.23768271058797835,
      "num_tokens": 158544812.0,
      "step": 78235
    },
    {
      "entropy": 5.78130521774292,
      "epoch": 6.837367823123307,
      "grad_norm": 1.296875,
      "learning_rate": 0.000153842851844422,
      "loss": 5.0479,
      "mean_token_accuracy": 0.23248264640569688,
      "num_tokens": 158554943.0,
      "step": 78240
    },
    {
      "entropy": 5.813252019882202,
      "epoch": 6.837804771476011,
      "grad_norm": 1.15625,
      "learning_rate": 0.00015381659539667625,
      "loss": 5.0975,
      "mean_token_accuracy": 0.23237717449665068,
      "num_tokens": 158565274.0,
      "step": 78245
    },
    {
      "entropy": 5.784118700027466,
      "epoch": 6.838241719828716,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015379034127329167,
      "loss": 5.058,
      "mean_token_accuracy": 0.2349325254559517,
      "num_tokens": 158576495.0,
      "step": 78250
    },
    {
      "entropy": 5.842232084274292,
      "epoch": 6.838678668181421,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015376408947477188,
      "loss": 5.1558,
      "mean_token_accuracy": 0.22912642657756804,
      "num_tokens": 158587573.0,
      "step": 78255
    },
    {
      "entropy": 5.889782571792603,
      "epoch": 6.839115616534126,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015373784000162038,
      "loss": 5.1829,
      "mean_token_accuracy": 0.22180276960134507,
      "num_tokens": 158596884.0,
      "step": 78260
    },
    {
      "entropy": 5.8914941310882565,
      "epoch": 6.83955256488683,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015371159285434066,
      "loss": 5.1622,
      "mean_token_accuracy": 0.2275950089097023,
      "num_tokens": 158606132.0,
      "step": 78265
    },
    {
      "entropy": 5.933161306381225,
      "epoch": 6.839989513239535,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015368534803343614,
      "loss": 5.2257,
      "mean_token_accuracy": 0.21896380484104155,
      "num_tokens": 158616949.0,
      "step": 78270
    },
    {
      "entropy": 5.790334177017212,
      "epoch": 6.84042646159224,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015365910553941017,
      "loss": 5.0192,
      "mean_token_accuracy": 0.24098784029483794,
      "num_tokens": 158626289.0,
      "step": 78275
    },
    {
      "entropy": 5.755899095535279,
      "epoch": 6.840863409944944,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015363286537276618,
      "loss": 5.0394,
      "mean_token_accuracy": 0.23452406227588654,
      "num_tokens": 158637199.0,
      "step": 78280
    },
    {
      "entropy": 5.872763538360596,
      "epoch": 6.84130035829765,
      "grad_norm": 1.078125,
      "learning_rate": 0.00015360662753400754,
      "loss": 5.1867,
      "mean_token_accuracy": 0.2202085092663765,
      "num_tokens": 158647877.0,
      "step": 78285
    },
    {
      "entropy": 5.822349786758423,
      "epoch": 6.841737306650354,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015358039202363727,
      "loss": 5.0873,
      "mean_token_accuracy": 0.23613365590572358,
      "num_tokens": 158657807.0,
      "step": 78290
    },
    {
      "entropy": 5.859901285171508,
      "epoch": 6.842174255003059,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015355415884215867,
      "loss": 5.0828,
      "mean_token_accuracy": 0.23331008702516556,
      "num_tokens": 158667779.0,
      "step": 78295
    },
    {
      "entropy": 5.882628393173218,
      "epoch": 6.842611203355763,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00015352792799007498,
      "loss": 5.1551,
      "mean_token_accuracy": 0.21992477923631668,
      "num_tokens": 158678315.0,
      "step": 78300
    },
    {
      "entropy": 5.784864521026611,
      "epoch": 6.843048151708468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015350169946788921,
      "loss": 5.0355,
      "mean_token_accuracy": 0.23471533507108688,
      "num_tokens": 158687910.0,
      "step": 78305
    },
    {
      "entropy": 5.887207460403443,
      "epoch": 6.843485100061173,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001534754732761046,
      "loss": 5.2025,
      "mean_token_accuracy": 0.21888171434402465,
      "num_tokens": 158699066.0,
      "step": 78310
    },
    {
      "entropy": 5.815325355529785,
      "epoch": 6.843922048413877,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015344924941522404,
      "loss": 5.1466,
      "mean_token_accuracy": 0.2280951112508774,
      "num_tokens": 158709193.0,
      "step": 78315
    },
    {
      "entropy": 5.889065313339233,
      "epoch": 6.844358996766582,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001534230278857506,
      "loss": 5.297,
      "mean_token_accuracy": 0.2115326926112175,
      "num_tokens": 158720323.0,
      "step": 78320
    },
    {
      "entropy": 5.84193983078003,
      "epoch": 6.844795945119287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015339680868818728,
      "loss": 5.0494,
      "mean_token_accuracy": 0.23917797356843948,
      "num_tokens": 158729647.0,
      "step": 78325
    },
    {
      "entropy": 5.850230407714844,
      "epoch": 6.845232893471992,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015337059182303682,
      "loss": 5.1332,
      "mean_token_accuracy": 0.22819488495588303,
      "num_tokens": 158740060.0,
      "step": 78330
    },
    {
      "entropy": 5.8966387748718265,
      "epoch": 6.845669841824696,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001533443772908021,
      "loss": 5.1421,
      "mean_token_accuracy": 0.22397412359714508,
      "num_tokens": 158749811.0,
      "step": 78335
    },
    {
      "entropy": 5.848373222351074,
      "epoch": 6.846106790177401,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015331816509198598,
      "loss": 5.0861,
      "mean_token_accuracy": 0.22697376161813737,
      "num_tokens": 158758489.0,
      "step": 78340
    },
    {
      "entropy": 5.809828567504883,
      "epoch": 6.846543738530106,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015329195522709123,
      "loss": 5.1166,
      "mean_token_accuracy": 0.22251713424921035,
      "num_tokens": 158769183.0,
      "step": 78345
    },
    {
      "entropy": 5.748132801055908,
      "epoch": 6.8469806868828105,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015326574769662052,
      "loss": 5.0596,
      "mean_token_accuracy": 0.2377693772315979,
      "num_tokens": 158779556.0,
      "step": 78350
    },
    {
      "entropy": 5.799893474578857,
      "epoch": 6.847417635235515,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015323954250107661,
      "loss": 5.0998,
      "mean_token_accuracy": 0.23907267898321152,
      "num_tokens": 158789763.0,
      "step": 78355
    },
    {
      "entropy": 5.867489004135132,
      "epoch": 6.847854583588219,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000153213339640962,
      "loss": 5.1208,
      "mean_token_accuracy": 0.22607425451278687,
      "num_tokens": 158800180.0,
      "step": 78360
    },
    {
      "entropy": 5.828309679031372,
      "epoch": 6.848291531940925,
      "grad_norm": 1.109375,
      "learning_rate": 0.00015318713911677943,
      "loss": 5.0737,
      "mean_token_accuracy": 0.22336316853761673,
      "num_tokens": 158810579.0,
      "step": 78365
    },
    {
      "entropy": 5.880766534805298,
      "epoch": 6.848728480293629,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015316094092903134,
      "loss": 5.1705,
      "mean_token_accuracy": 0.22209844887256622,
      "num_tokens": 158819596.0,
      "step": 78370
    },
    {
      "entropy": 5.867542505264282,
      "epoch": 6.849165428646334,
      "grad_norm": 1.25,
      "learning_rate": 0.0001531347450782203,
      "loss": 5.1748,
      "mean_token_accuracy": 0.22620669305324553,
      "num_tokens": 158828684.0,
      "step": 78375
    },
    {
      "entropy": 5.872406816482544,
      "epoch": 6.849602376999039,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015310855156484864,
      "loss": 5.0956,
      "mean_token_accuracy": 0.22932066023349762,
      "num_tokens": 158838254.0,
      "step": 78380
    },
    {
      "entropy": 5.838248443603516,
      "epoch": 6.8500393253517435,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001530823603894188,
      "loss": 5.1334,
      "mean_token_accuracy": 0.22436415106058122,
      "num_tokens": 158847993.0,
      "step": 78385
    },
    {
      "entropy": 5.772061681747436,
      "epoch": 6.850476273704448,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015305617155243322,
      "loss": 5.0854,
      "mean_token_accuracy": 0.22964550405740738,
      "num_tokens": 158858575.0,
      "step": 78390
    },
    {
      "entropy": 5.852140235900879,
      "epoch": 6.8509132220571525,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00015302998505439417,
      "loss": 5.1547,
      "mean_token_accuracy": 0.21961232125759125,
      "num_tokens": 158868696.0,
      "step": 78395
    },
    {
      "entropy": 5.8844867706298825,
      "epoch": 6.851350170409858,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001530038008958039,
      "loss": 5.2216,
      "mean_token_accuracy": 0.21634309738874435,
      "num_tokens": 158878968.0,
      "step": 78400
    },
    {
      "entropy": 5.865823650360108,
      "epoch": 6.851787118762562,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001529776190771646,
      "loss": 5.1425,
      "mean_token_accuracy": 0.2316720589995384,
      "num_tokens": 158889495.0,
      "step": 78405
    },
    {
      "entropy": 5.9338123321533205,
      "epoch": 6.852224067115267,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001529514395989786,
      "loss": 5.1929,
      "mean_token_accuracy": 0.21982474625110626,
      "num_tokens": 158899412.0,
      "step": 78410
    },
    {
      "entropy": 5.883478736877441,
      "epoch": 6.852661015467971,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015292526246174788,
      "loss": 5.1935,
      "mean_token_accuracy": 0.21036970466375352,
      "num_tokens": 158909918.0,
      "step": 78415
    },
    {
      "entropy": 5.834931039810181,
      "epoch": 6.853097963820677,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001528990876659747,
      "loss": 5.1075,
      "mean_token_accuracy": 0.22395556420087814,
      "num_tokens": 158919506.0,
      "step": 78420
    },
    {
      "entropy": 5.742763328552246,
      "epoch": 6.853534912173381,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001528729152121609,
      "loss": 5.0033,
      "mean_token_accuracy": 0.23812780976295472,
      "num_tokens": 158929925.0,
      "step": 78425
    },
    {
      "entropy": 5.803689289093017,
      "epoch": 6.8539718605260855,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001528467451008086,
      "loss": 5.0798,
      "mean_token_accuracy": 0.22830545157194138,
      "num_tokens": 158938707.0,
      "step": 78430
    },
    {
      "entropy": 5.811156415939331,
      "epoch": 6.854408808878791,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001528205773324197,
      "loss": 5.0402,
      "mean_token_accuracy": 0.23738347440958024,
      "num_tokens": 158948086.0,
      "step": 78435
    },
    {
      "entropy": 5.833845520019532,
      "epoch": 6.854845757231495,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001527944119074962,
      "loss": 5.186,
      "mean_token_accuracy": 0.22391602545976638,
      "num_tokens": 158958427.0,
      "step": 78440
    },
    {
      "entropy": 5.855818557739258,
      "epoch": 6.8552827055842,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001527682488265399,
      "loss": 5.0818,
      "mean_token_accuracy": 0.23414502888917924,
      "num_tokens": 158966925.0,
      "step": 78445
    },
    {
      "entropy": 5.883692741394043,
      "epoch": 6.855719653936904,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001527420880900526,
      "loss": 5.1952,
      "mean_token_accuracy": 0.21766960620880127,
      "num_tokens": 158976063.0,
      "step": 78450
    },
    {
      "entropy": 5.812950563430786,
      "epoch": 6.85615660228961,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015271592969853614,
      "loss": 5.0806,
      "mean_token_accuracy": 0.23597290515899658,
      "num_tokens": 158986005.0,
      "step": 78455
    },
    {
      "entropy": 5.8740006446838375,
      "epoch": 6.856593550642314,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001526897736524922,
      "loss": 5.1689,
      "mean_token_accuracy": 0.2294640451669693,
      "num_tokens": 158995667.0,
      "step": 78460
    },
    {
      "entropy": 5.836291933059693,
      "epoch": 6.857030498995019,
      "grad_norm": 1.125,
      "learning_rate": 0.00015266361995242253,
      "loss": 5.152,
      "mean_token_accuracy": 0.2280134841799736,
      "num_tokens": 159005533.0,
      "step": 78465
    },
    {
      "entropy": 5.883401250839233,
      "epoch": 6.857467447347723,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00015263746859882878,
      "loss": 5.1947,
      "mean_token_accuracy": 0.22253188639879226,
      "num_tokens": 159015620.0,
      "step": 78470
    },
    {
      "entropy": 5.8760679244995115,
      "epoch": 6.857904395700428,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015261131959221246,
      "loss": 5.2074,
      "mean_token_accuracy": 0.2179595112800598,
      "num_tokens": 159026029.0,
      "step": 78475
    },
    {
      "entropy": 5.835489511489868,
      "epoch": 6.858341344053133,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015258517293307512,
      "loss": 5.0746,
      "mean_token_accuracy": 0.23270718455314637,
      "num_tokens": 159035027.0,
      "step": 78480
    },
    {
      "entropy": 5.851118516921997,
      "epoch": 6.858778292405837,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001525590286219183,
      "loss": 5.0557,
      "mean_token_accuracy": 0.23899878114461898,
      "num_tokens": 159045563.0,
      "step": 78485
    },
    {
      "entropy": 5.923153686523437,
      "epoch": 6.859215240758543,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001525328866592435,
      "loss": 5.178,
      "mean_token_accuracy": 0.22032850831747056,
      "num_tokens": 159056027.0,
      "step": 78490
    },
    {
      "entropy": 5.880445289611816,
      "epoch": 6.859652189111247,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001525067470455521,
      "loss": 5.0912,
      "mean_token_accuracy": 0.22642794251441956,
      "num_tokens": 159065304.0,
      "step": 78495
    },
    {
      "entropy": 5.824985408782959,
      "epoch": 6.860089137463952,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015248060978134546,
      "loss": 5.1172,
      "mean_token_accuracy": 0.2286025434732437,
      "num_tokens": 159076255.0,
      "step": 78500
    },
    {
      "entropy": 5.783015441894531,
      "epoch": 6.860526085816656,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00015245447486712493,
      "loss": 5.1115,
      "mean_token_accuracy": 0.2310975417494774,
      "num_tokens": 159086175.0,
      "step": 78505
    },
    {
      "entropy": 5.905102252960205,
      "epoch": 6.8609630341693615,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015242834230339182,
      "loss": 5.1958,
      "mean_token_accuracy": 0.21654991507530214,
      "num_tokens": 159096211.0,
      "step": 78510
    },
    {
      "entropy": 5.878716945648193,
      "epoch": 6.861399982522066,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015240221209064726,
      "loss": 5.1256,
      "mean_token_accuracy": 0.23097098618745804,
      "num_tokens": 159107036.0,
      "step": 78515
    },
    {
      "entropy": 5.849926567077636,
      "epoch": 6.86183693087477,
      "grad_norm": 1.125,
      "learning_rate": 0.00015237608422939266,
      "loss": 5.123,
      "mean_token_accuracy": 0.22564293295145035,
      "num_tokens": 159116619.0,
      "step": 78520
    },
    {
      "entropy": 5.8608832359313965,
      "epoch": 6.862273879227475,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015234995872012892,
      "loss": 5.166,
      "mean_token_accuracy": 0.21685556024312974,
      "num_tokens": 159127353.0,
      "step": 78525
    },
    {
      "entropy": 5.887146091461181,
      "epoch": 6.86271082758018,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015232383556335722,
      "loss": 5.1223,
      "mean_token_accuracy": 0.22077332884073259,
      "num_tokens": 159137444.0,
      "step": 78530
    },
    {
      "entropy": 5.814521217346192,
      "epoch": 6.863147775932885,
      "grad_norm": 1.109375,
      "learning_rate": 0.00015229771475957864,
      "loss": 5.1027,
      "mean_token_accuracy": 0.2265629917383194,
      "num_tokens": 159147884.0,
      "step": 78535
    },
    {
      "entropy": 5.881497430801391,
      "epoch": 6.863584724285589,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015227159630929423,
      "loss": 5.1378,
      "mean_token_accuracy": 0.23177792876958847,
      "num_tokens": 159158394.0,
      "step": 78540
    },
    {
      "entropy": 5.849803829193116,
      "epoch": 6.8640216726382945,
      "grad_norm": 1.25,
      "learning_rate": 0.00015224548021300482,
      "loss": 5.0909,
      "mean_token_accuracy": 0.22381671965122224,
      "num_tokens": 159168375.0,
      "step": 78545
    },
    {
      "entropy": 5.860710668563843,
      "epoch": 6.864458620990999,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001522193664712115,
      "loss": 5.2107,
      "mean_token_accuracy": 0.21800239384174347,
      "num_tokens": 159178310.0,
      "step": 78550
    },
    {
      "entropy": 5.8562477111816404,
      "epoch": 6.8648955693437035,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015219325508441505,
      "loss": 5.1334,
      "mean_token_accuracy": 0.2222789317369461,
      "num_tokens": 159188740.0,
      "step": 78555
    },
    {
      "entropy": 5.892991971969605,
      "epoch": 6.865332517696408,
      "grad_norm": 1.109375,
      "learning_rate": 0.00015216714605311633,
      "loss": 5.1912,
      "mean_token_accuracy": 0.22038948833942412,
      "num_tokens": 159198707.0,
      "step": 78560
    },
    {
      "entropy": 5.918051910400391,
      "epoch": 6.865769466049113,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001521410393778161,
      "loss": 5.1887,
      "mean_token_accuracy": 0.21572543382644654,
      "num_tokens": 159207474.0,
      "step": 78565
    },
    {
      "entropy": 5.829990243911743,
      "epoch": 6.866206414401818,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015211493505901515,
      "loss": 5.1356,
      "mean_token_accuracy": 0.23023246228694916,
      "num_tokens": 159217573.0,
      "step": 78570
    },
    {
      "entropy": 5.788445043563843,
      "epoch": 6.866643362754522,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015208883309721408,
      "loss": 5.1275,
      "mean_token_accuracy": 0.22781231850385666,
      "num_tokens": 159227086.0,
      "step": 78575
    },
    {
      "entropy": 5.76255955696106,
      "epoch": 6.867080311107227,
      "grad_norm": 1.125,
      "learning_rate": 0.00015206273349291363,
      "loss": 5.05,
      "mean_token_accuracy": 0.23739366978406906,
      "num_tokens": 159238163.0,
      "step": 78580
    },
    {
      "entropy": 5.819547033309936,
      "epoch": 6.867517259459932,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015203663624661435,
      "loss": 5.054,
      "mean_token_accuracy": 0.2348923623561859,
      "num_tokens": 159248127.0,
      "step": 78585
    },
    {
      "entropy": 5.813530349731446,
      "epoch": 6.8679542078126365,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001520105413588168,
      "loss": 5.0587,
      "mean_token_accuracy": 0.23947688043117524,
      "num_tokens": 159258404.0,
      "step": 78590
    },
    {
      "entropy": 5.9126828670501705,
      "epoch": 6.868391156165341,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015198444883002154,
      "loss": 5.2393,
      "mean_token_accuracy": 0.2183230087161064,
      "num_tokens": 159268134.0,
      "step": 78595
    },
    {
      "entropy": 5.867569875717163,
      "epoch": 6.868828104518046,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000151958358660729,
      "loss": 5.0788,
      "mean_token_accuracy": 0.234795181453228,
      "num_tokens": 159277089.0,
      "step": 78600
    },
    {
      "entropy": 5.811229991912842,
      "epoch": 6.869265052870751,
      "grad_norm": 1.125,
      "learning_rate": 0.0001519322708514396,
      "loss": 5.0732,
      "mean_token_accuracy": 0.23385856300592422,
      "num_tokens": 159286919.0,
      "step": 78605
    },
    {
      "entropy": 5.849855661392212,
      "epoch": 6.869702001223455,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015190618540265373,
      "loss": 5.1695,
      "mean_token_accuracy": 0.22135651111602783,
      "num_tokens": 159296259.0,
      "step": 78610
    },
    {
      "entropy": 5.781433391571045,
      "epoch": 6.87013894957616,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015188010231487181,
      "loss": 5.0834,
      "mean_token_accuracy": 0.23256431072950362,
      "num_tokens": 159306595.0,
      "step": 78615
    },
    {
      "entropy": 5.885308456420899,
      "epoch": 6.870575897928865,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015185402158859396,
      "loss": 5.1486,
      "mean_token_accuracy": 0.22559240460395813,
      "num_tokens": 159316756.0,
      "step": 78620
    },
    {
      "entropy": 5.822170162200928,
      "epoch": 6.87101284628157,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015182794322432053,
      "loss": 5.1394,
      "mean_token_accuracy": 0.23080401122570038,
      "num_tokens": 159326245.0,
      "step": 78625
    },
    {
      "entropy": 5.862877988815308,
      "epoch": 6.871449794634274,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015180186722255168,
      "loss": 5.1762,
      "mean_token_accuracy": 0.22028230875730515,
      "num_tokens": 159336180.0,
      "step": 78630
    },
    {
      "entropy": 5.8590264320373535,
      "epoch": 6.8718867429869785,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015177579358378757,
      "loss": 5.1092,
      "mean_token_accuracy": 0.2254078522324562,
      "num_tokens": 159345746.0,
      "step": 78635
    },
    {
      "entropy": 5.872349405288697,
      "epoch": 6.872323691339684,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001517497223085283,
      "loss": 5.1551,
      "mean_token_accuracy": 0.22744329869747162,
      "num_tokens": 159357171.0,
      "step": 78640
    },
    {
      "entropy": 5.778968858718872,
      "epoch": 6.872760639692388,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015172365339727396,
      "loss": 4.965,
      "mean_token_accuracy": 0.24240185171365738,
      "num_tokens": 159367319.0,
      "step": 78645
    },
    {
      "entropy": 5.767581415176392,
      "epoch": 6.873197588045093,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015169758685052448,
      "loss": 5.0982,
      "mean_token_accuracy": 0.23204938173294068,
      "num_tokens": 159376705.0,
      "step": 78650
    },
    {
      "entropy": 5.797119283676148,
      "epoch": 6.873634536397798,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015167152266877993,
      "loss": 5.066,
      "mean_token_accuracy": 0.23118557035923004,
      "num_tokens": 159387068.0,
      "step": 78655
    },
    {
      "entropy": 5.864565372467041,
      "epoch": 6.874071484750503,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015164546085254021,
      "loss": 5.1251,
      "mean_token_accuracy": 0.22411636561155318,
      "num_tokens": 159397738.0,
      "step": 78660
    },
    {
      "entropy": 5.8608321189880375,
      "epoch": 6.874508433103207,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015161940140230517,
      "loss": 5.1622,
      "mean_token_accuracy": 0.228010793030262,
      "num_tokens": 159408040.0,
      "step": 78665
    },
    {
      "entropy": 5.879010725021362,
      "epoch": 6.874945381455912,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015159334431857468,
      "loss": 5.1281,
      "mean_token_accuracy": 0.22518561780452728,
      "num_tokens": 159417419.0,
      "step": 78670
    },
    {
      "entropy": 5.872149991989136,
      "epoch": 6.875382329808617,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015156728960184852,
      "loss": 5.1454,
      "mean_token_accuracy": 0.2164447784423828,
      "num_tokens": 159427488.0,
      "step": 78675
    },
    {
      "entropy": 5.895752429962158,
      "epoch": 6.875819278161321,
      "grad_norm": 1.15625,
      "learning_rate": 0.00015154123725262642,
      "loss": 5.1369,
      "mean_token_accuracy": 0.22860851436853408,
      "num_tokens": 159437885.0,
      "step": 78680
    },
    {
      "entropy": 5.803361129760742,
      "epoch": 6.876256226514026,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015151518727140806,
      "loss": 5.1012,
      "mean_token_accuracy": 0.23408263623714448,
      "num_tokens": 159449190.0,
      "step": 78685
    },
    {
      "entropy": 5.849271249771118,
      "epoch": 6.87669317486673,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001514891396586931,
      "loss": 5.1383,
      "mean_token_accuracy": 0.22481104135513305,
      "num_tokens": 159459168.0,
      "step": 78690
    },
    {
      "entropy": 5.7817626953125,
      "epoch": 6.877130123219436,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001514630944149812,
      "loss": 5.0794,
      "mean_token_accuracy": 0.22751373946666717,
      "num_tokens": 159469192.0,
      "step": 78695
    },
    {
      "entropy": 5.870654010772705,
      "epoch": 6.87756707157214,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015143705154077187,
      "loss": 5.1837,
      "mean_token_accuracy": 0.22028105109930038,
      "num_tokens": 159478850.0,
      "step": 78700
    },
    {
      "entropy": 5.826450300216675,
      "epoch": 6.878004019924845,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015141101103656468,
      "loss": 5.0773,
      "mean_token_accuracy": 0.22726802080869674,
      "num_tokens": 159488218.0,
      "step": 78705
    },
    {
      "entropy": 5.843746376037598,
      "epoch": 6.87844096827755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000151384972902859,
      "loss": 5.1397,
      "mean_token_accuracy": 0.22740211039781572,
      "num_tokens": 159498679.0,
      "step": 78710
    },
    {
      "entropy": 5.851777791976929,
      "epoch": 6.8788779166302545,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015135893714015433,
      "loss": 5.1282,
      "mean_token_accuracy": 0.23284651786088945,
      "num_tokens": 159508181.0,
      "step": 78715
    },
    {
      "entropy": 5.802233123779297,
      "epoch": 6.879314864982959,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015133290374895003,
      "loss": 5.0993,
      "mean_token_accuracy": 0.2313896581530571,
      "num_tokens": 159517819.0,
      "step": 78720
    },
    {
      "entropy": 5.808415365219116,
      "epoch": 6.879751813335663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00015130687272974544,
      "loss": 5.0924,
      "mean_token_accuracy": 0.2320666268467903,
      "num_tokens": 159529337.0,
      "step": 78725
    },
    {
      "entropy": 5.811566114425659,
      "epoch": 6.880188761688369,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015128084408303982,
      "loss": 5.0972,
      "mean_token_accuracy": 0.22843187898397446,
      "num_tokens": 159540149.0,
      "step": 78730
    },
    {
      "entropy": 5.83530740737915,
      "epoch": 6.880625710041073,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015125481780933246,
      "loss": 5.0468,
      "mean_token_accuracy": 0.23729778677225113,
      "num_tokens": 159550866.0,
      "step": 78735
    },
    {
      "entropy": 5.8866219997406,
      "epoch": 6.881062658393778,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00015122879390912252,
      "loss": 5.2057,
      "mean_token_accuracy": 0.21808618009090425,
      "num_tokens": 159561230.0,
      "step": 78740
    },
    {
      "entropy": 5.875377416610718,
      "epoch": 6.881499606746482,
      "grad_norm": 1.25,
      "learning_rate": 0.0001512027723829092,
      "loss": 5.1721,
      "mean_token_accuracy": 0.22447361797094345,
      "num_tokens": 159570287.0,
      "step": 78745
    },
    {
      "entropy": 5.789941835403442,
      "epoch": 6.881936555099188,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015117675323119163,
      "loss": 5.0301,
      "mean_token_accuracy": 0.23561719357967376,
      "num_tokens": 159581125.0,
      "step": 78750
    },
    {
      "entropy": 5.877577495574951,
      "epoch": 6.882373503451892,
      "grad_norm": 1.25,
      "learning_rate": 0.0001511507364544687,
      "loss": 5.2682,
      "mean_token_accuracy": 0.21907155811786652,
      "num_tokens": 159591912.0,
      "step": 78755
    },
    {
      "entropy": 5.821523284912109,
      "epoch": 6.8828104518045965,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001511247220532396,
      "loss": 5.0928,
      "mean_token_accuracy": 0.22882066816091537,
      "num_tokens": 159602370.0,
      "step": 78760
    },
    {
      "entropy": 5.846452093124389,
      "epoch": 6.883247400157302,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015109871002800323,
      "loss": 5.0933,
      "mean_token_accuracy": 0.23077302873134614,
      "num_tokens": 159613388.0,
      "step": 78765
    },
    {
      "entropy": 5.85982575416565,
      "epoch": 6.883684348510006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015107270037925853,
      "loss": 5.1416,
      "mean_token_accuracy": 0.22300254106521605,
      "num_tokens": 159623955.0,
      "step": 78770
    },
    {
      "entropy": 5.798479318618774,
      "epoch": 6.884121296862711,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015104669310750434,
      "loss": 5.0902,
      "mean_token_accuracy": 0.23276909291744233,
      "num_tokens": 159632606.0,
      "step": 78775
    },
    {
      "entropy": 5.7915863513946535,
      "epoch": 6.884558245215415,
      "grad_norm": 1.109375,
      "learning_rate": 0.00015102068821323954,
      "loss": 5.0729,
      "mean_token_accuracy": 0.23880105912685395,
      "num_tokens": 159642176.0,
      "step": 78780
    },
    {
      "entropy": 5.913403415679932,
      "epoch": 6.884995193568121,
      "grad_norm": 1.25,
      "learning_rate": 0.00015099468569696289,
      "loss": 5.2237,
      "mean_token_accuracy": 0.21546831727027893,
      "num_tokens": 159652740.0,
      "step": 78785
    },
    {
      "entropy": 5.816539192199707,
      "epoch": 6.885432141920825,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015096868555917317,
      "loss": 5.0476,
      "mean_token_accuracy": 0.24147094041109085,
      "num_tokens": 159662641.0,
      "step": 78790
    },
    {
      "entropy": 5.857751798629761,
      "epoch": 6.88586909027353,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015094268780036906,
      "loss": 5.131,
      "mean_token_accuracy": 0.22104094177484512,
      "num_tokens": 159672385.0,
      "step": 78795
    },
    {
      "entropy": 5.889488887786865,
      "epoch": 6.886306038626234,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015091669242104928,
      "loss": 5.1885,
      "mean_token_accuracy": 0.22504245340824128,
      "num_tokens": 159683410.0,
      "step": 78800
    },
    {
      "entropy": 5.885727119445801,
      "epoch": 6.886742986978939,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015089069942171224,
      "loss": 5.1871,
      "mean_token_accuracy": 0.22632149904966353,
      "num_tokens": 159694196.0,
      "step": 78805
    },
    {
      "entropy": 5.844967937469482,
      "epoch": 6.887179935331644,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015086470880285664,
      "loss": 5.0827,
      "mean_token_accuracy": 0.236009781062603,
      "num_tokens": 159703191.0,
      "step": 78810
    },
    {
      "entropy": 5.849854803085327,
      "epoch": 6.887616883684348,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015083872056498098,
      "loss": 5.11,
      "mean_token_accuracy": 0.22965752184391022,
      "num_tokens": 159712532.0,
      "step": 78815
    },
    {
      "entropy": 5.889562320709229,
      "epoch": 6.888053832037054,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00015081273470858372,
      "loss": 5.1571,
      "mean_token_accuracy": 0.22195309847593309,
      "num_tokens": 159722721.0,
      "step": 78820
    },
    {
      "entropy": 5.890866088867187,
      "epoch": 6.888490780389758,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001507867512341633,
      "loss": 5.2015,
      "mean_token_accuracy": 0.2159147948026657,
      "num_tokens": 159732402.0,
      "step": 78825
    },
    {
      "entropy": 5.886023998260498,
      "epoch": 6.888927728742463,
      "grad_norm": 1.25,
      "learning_rate": 0.00015076077014221804,
      "loss": 5.2183,
      "mean_token_accuracy": 0.21640091240406037,
      "num_tokens": 159742388.0,
      "step": 78830
    },
    {
      "entropy": 5.864715242385865,
      "epoch": 6.889364677095167,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001507347914332463,
      "loss": 5.1618,
      "mean_token_accuracy": 0.2257368192076683,
      "num_tokens": 159752115.0,
      "step": 78835
    },
    {
      "entropy": 5.851391839981079,
      "epoch": 6.8898016254478724,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015070881510774643,
      "loss": 5.1317,
      "mean_token_accuracy": 0.2287936732172966,
      "num_tokens": 159762369.0,
      "step": 78840
    },
    {
      "entropy": 5.912846326828003,
      "epoch": 6.890238573800577,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001506828411662166,
      "loss": 5.187,
      "mean_token_accuracy": 0.22104996144771577,
      "num_tokens": 159772969.0,
      "step": 78845
    },
    {
      "entropy": 5.910997724533081,
      "epoch": 6.890675522153281,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00015065686960915507,
      "loss": 5.2336,
      "mean_token_accuracy": 0.21908450573682786,
      "num_tokens": 159784001.0,
      "step": 78850
    },
    {
      "entropy": 5.860372543334961,
      "epoch": 6.891112470505986,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015063090043705986,
      "loss": 5.1055,
      "mean_token_accuracy": 0.23397770375013352,
      "num_tokens": 159793890.0,
      "step": 78855
    },
    {
      "entropy": 5.899259090423584,
      "epoch": 6.891549418858691,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015060493365042911,
      "loss": 5.1698,
      "mean_token_accuracy": 0.21908364444971085,
      "num_tokens": 159803154.0,
      "step": 78860
    },
    {
      "entropy": 5.847819375991821,
      "epoch": 6.891986367211396,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000150578969249761,
      "loss": 5.1756,
      "mean_token_accuracy": 0.2290249362587929,
      "num_tokens": 159813964.0,
      "step": 78865
    },
    {
      "entropy": 5.863004207611084,
      "epoch": 6.8924233155641,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015055300723555343,
      "loss": 5.1336,
      "mean_token_accuracy": 0.22431600689888,
      "num_tokens": 159824154.0,
      "step": 78870
    },
    {
      "entropy": 5.8573204517364506,
      "epoch": 6.8928602639168055,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015052704760830433,
      "loss": 5.1731,
      "mean_token_accuracy": 0.22744101285934448,
      "num_tokens": 159834107.0,
      "step": 78875
    },
    {
      "entropy": 5.899953413009643,
      "epoch": 6.89329721226951,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015050109036851173,
      "loss": 5.1497,
      "mean_token_accuracy": 0.23122986555099487,
      "num_tokens": 159843950.0,
      "step": 78880
    },
    {
      "entropy": 5.8714025020599365,
      "epoch": 6.8937341606222144,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015047513551667347,
      "loss": 5.1585,
      "mean_token_accuracy": 0.22471745014190675,
      "num_tokens": 159854487.0,
      "step": 78885
    },
    {
      "entropy": 5.90092830657959,
      "epoch": 6.894171108974919,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001504491830532873,
      "loss": 5.2284,
      "mean_token_accuracy": 0.22357404381036758,
      "num_tokens": 159864925.0,
      "step": 78890
    },
    {
      "entropy": 5.820229387283325,
      "epoch": 6.894608057327623,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001504232329788512,
      "loss": 5.0893,
      "mean_token_accuracy": 0.22738387882709504,
      "num_tokens": 159874882.0,
      "step": 78895
    },
    {
      "entropy": 5.742080736160278,
      "epoch": 6.895045005680329,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015039728529386265,
      "loss": 5.0315,
      "mean_token_accuracy": 0.24470418840646743,
      "num_tokens": 159885074.0,
      "step": 78900
    },
    {
      "entropy": 5.7667077541351315,
      "epoch": 6.895481954033033,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015037133999881946,
      "loss": 5.031,
      "mean_token_accuracy": 0.237014102935791,
      "num_tokens": 159894677.0,
      "step": 78905
    },
    {
      "entropy": 5.814021015167237,
      "epoch": 6.895918902385738,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015034539709421932,
      "loss": 5.0698,
      "mean_token_accuracy": 0.23408328890800476,
      "num_tokens": 159904274.0,
      "step": 78910
    },
    {
      "entropy": 5.821523761749267,
      "epoch": 6.896355850738443,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015031945658055976,
      "loss": 5.1848,
      "mean_token_accuracy": 0.22114245146512984,
      "num_tokens": 159913826.0,
      "step": 78915
    },
    {
      "entropy": 5.800604915618896,
      "epoch": 6.8967927990911475,
      "grad_norm": 1.0625,
      "learning_rate": 0.00015029351845833832,
      "loss": 5.1394,
      "mean_token_accuracy": 0.2282139390707016,
      "num_tokens": 159924867.0,
      "step": 78920
    },
    {
      "entropy": 5.8359626770019535,
      "epoch": 6.897229747443852,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015026758272805257,
      "loss": 5.1174,
      "mean_token_accuracy": 0.22646504044532775,
      "num_tokens": 159934885.0,
      "step": 78925
    },
    {
      "entropy": 5.8689634799957275,
      "epoch": 6.897666695796557,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015024164939019997,
      "loss": 5.0735,
      "mean_token_accuracy": 0.23036430329084395,
      "num_tokens": 159944959.0,
      "step": 78930
    },
    {
      "entropy": 5.873261022567749,
      "epoch": 6.898103644149262,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015021571844527786,
      "loss": 5.1398,
      "mean_token_accuracy": 0.23093374073505402,
      "num_tokens": 159954021.0,
      "step": 78935
    },
    {
      "entropy": 5.844510269165039,
      "epoch": 6.898540592501966,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001501897898937837,
      "loss": 5.1161,
      "mean_token_accuracy": 0.22769259214401244,
      "num_tokens": 159963321.0,
      "step": 78940
    },
    {
      "entropy": 5.80727186203003,
      "epoch": 6.898977540854671,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001501638637362148,
      "loss": 5.0988,
      "mean_token_accuracy": 0.2275252804160118,
      "num_tokens": 159973181.0,
      "step": 78945
    },
    {
      "entropy": 5.837068223953247,
      "epoch": 6.899414489207375,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015013793997306835,
      "loss": 5.1444,
      "mean_token_accuracy": 0.22634527385234832,
      "num_tokens": 159982652.0,
      "step": 78950
    },
    {
      "entropy": 5.781807613372803,
      "epoch": 6.899851437560081,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015011201860484164,
      "loss": 5.1443,
      "mean_token_accuracy": 0.22032323032617568,
      "num_tokens": 159993274.0,
      "step": 78955
    },
    {
      "entropy": 5.8713384628295895,
      "epoch": 6.900288385912785,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015008609963203186,
      "loss": 5.2081,
      "mean_token_accuracy": 0.22820194959640502,
      "num_tokens": 160003416.0,
      "step": 78960
    },
    {
      "entropy": 5.832105016708374,
      "epoch": 6.9007253342654895,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015006018305513614,
      "loss": 5.0798,
      "mean_token_accuracy": 0.23412182182073593,
      "num_tokens": 160013327.0,
      "step": 78965
    },
    {
      "entropy": 5.762368106842041,
      "epoch": 6.901162282618195,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015003426887465155,
      "loss": 5.0175,
      "mean_token_accuracy": 0.24116799831390381,
      "num_tokens": 160021799.0,
      "step": 78970
    },
    {
      "entropy": 5.828119993209839,
      "epoch": 6.901599230970899,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015000835709107517,
      "loss": 5.1468,
      "mean_token_accuracy": 0.22154489159584045,
      "num_tokens": 160031279.0,
      "step": 78975
    },
    {
      "entropy": 5.8329675674438475,
      "epoch": 6.902036179323604,
      "grad_norm": 1.21875,
      "learning_rate": 0.000149982447704904,
      "loss": 5.085,
      "mean_token_accuracy": 0.22917586714029312,
      "num_tokens": 160041310.0,
      "step": 78980
    },
    {
      "entropy": 5.765676355361938,
      "epoch": 6.902473127676308,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000149956540716635,
      "loss": 5.0402,
      "mean_token_accuracy": 0.2391710638999939,
      "num_tokens": 160051370.0,
      "step": 78985
    },
    {
      "entropy": 5.873416996002197,
      "epoch": 6.902910076029014,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014993063612676508,
      "loss": 5.1696,
      "mean_token_accuracy": 0.22687036097049712,
      "num_tokens": 160061375.0,
      "step": 78990
    },
    {
      "entropy": 5.8184943199157715,
      "epoch": 6.903347024381718,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014990473393579118,
      "loss": 5.0967,
      "mean_token_accuracy": 0.2303204506635666,
      "num_tokens": 160070520.0,
      "step": 78995
    },
    {
      "entropy": 5.7978719711303714,
      "epoch": 6.903783972734423,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014987883414420994,
      "loss": 5.1032,
      "mean_token_accuracy": 0.22917671799659728,
      "num_tokens": 160081037.0,
      "step": 79000
    },
    {
      "entropy": 5.824172925949097,
      "epoch": 6.904220921087127,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001498529367525182,
      "loss": 5.054,
      "mean_token_accuracy": 0.23782475888729096,
      "num_tokens": 160090494.0,
      "step": 79005
    },
    {
      "entropy": 5.856429672241211,
      "epoch": 6.904657869439832,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014982704176121273,
      "loss": 5.1092,
      "mean_token_accuracy": 0.23364093154668808,
      "num_tokens": 160100455.0,
      "step": 79010
    },
    {
      "entropy": 5.844235134124756,
      "epoch": 6.905094817792537,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014980114917079014,
      "loss": 5.1053,
      "mean_token_accuracy": 0.22837853580713272,
      "num_tokens": 160110371.0,
      "step": 79015
    },
    {
      "entropy": 5.7845264911651615,
      "epoch": 6.905531766145241,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014977525898174712,
      "loss": 5.0675,
      "mean_token_accuracy": 0.2256306916475296,
      "num_tokens": 160120449.0,
      "step": 79020
    },
    {
      "entropy": 5.9332051277160645,
      "epoch": 6.905968714497947,
      "grad_norm": 1.09375,
      "learning_rate": 0.00014974937119458033,
      "loss": 5.2424,
      "mean_token_accuracy": 0.21684314757585527,
      "num_tokens": 160131166.0,
      "step": 79025
    },
    {
      "entropy": 5.873702335357666,
      "epoch": 6.906405662850651,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00014972348580978624,
      "loss": 5.068,
      "mean_token_accuracy": 0.2275787904858589,
      "num_tokens": 160141639.0,
      "step": 79030
    },
    {
      "entropy": 5.810317707061768,
      "epoch": 6.906842611203356,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001496976028278613,
      "loss": 5.0876,
      "mean_token_accuracy": 0.23426608443260194,
      "num_tokens": 160151441.0,
      "step": 79035
    },
    {
      "entropy": 5.809685182571411,
      "epoch": 6.90727955955606,
      "grad_norm": 1.15625,
      "learning_rate": 0.000149671722249302,
      "loss": 5.1192,
      "mean_token_accuracy": 0.22541120797395706,
      "num_tokens": 160161547.0,
      "step": 79040
    },
    {
      "entropy": 5.894552373886109,
      "epoch": 6.9077165079087655,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014964584407460468,
      "loss": 5.1525,
      "mean_token_accuracy": 0.21990722566843032,
      "num_tokens": 160171472.0,
      "step": 79045
    },
    {
      "entropy": 5.903398704528809,
      "epoch": 6.90815345626147,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001496199683042658,
      "loss": 5.1663,
      "mean_token_accuracy": 0.21840483248233794,
      "num_tokens": 160181466.0,
      "step": 79050
    },
    {
      "entropy": 5.897211790084839,
      "epoch": 6.908590404614174,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014959409493878163,
      "loss": 5.1287,
      "mean_token_accuracy": 0.22320446819067002,
      "num_tokens": 160191096.0,
      "step": 79055
    },
    {
      "entropy": 5.83978419303894,
      "epoch": 6.909027352966879,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014956822397864844,
      "loss": 5.2019,
      "mean_token_accuracy": 0.21990121454000472,
      "num_tokens": 160201432.0,
      "step": 79060
    },
    {
      "entropy": 5.82241005897522,
      "epoch": 6.909464301319584,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001495423554243624,
      "loss": 5.1167,
      "mean_token_accuracy": 0.22970918715000152,
      "num_tokens": 160211437.0,
      "step": 79065
    },
    {
      "entropy": 5.8520361423492435,
      "epoch": 6.909901249672289,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014951648927641976,
      "loss": 5.1333,
      "mean_token_accuracy": 0.2304375559091568,
      "num_tokens": 160221384.0,
      "step": 79070
    },
    {
      "entropy": 5.857079362869262,
      "epoch": 6.910338198024993,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00014949062553531656,
      "loss": 5.1203,
      "mean_token_accuracy": 0.22739203423261642,
      "num_tokens": 160230684.0,
      "step": 79075
    },
    {
      "entropy": 5.826854944229126,
      "epoch": 6.9107751463776985,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014946476420154905,
      "loss": 5.019,
      "mean_token_accuracy": 0.23708447962999343,
      "num_tokens": 160239893.0,
      "step": 79080
    },
    {
      "entropy": 5.805125093460083,
      "epoch": 6.911212094730403,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014943890527561305,
      "loss": 5.065,
      "mean_token_accuracy": 0.23592210859060286,
      "num_tokens": 160249616.0,
      "step": 79085
    },
    {
      "entropy": 5.883057451248169,
      "epoch": 6.9116490430831075,
      "grad_norm": 1.09375,
      "learning_rate": 0.00014941304875800465,
      "loss": 5.2334,
      "mean_token_accuracy": 0.2194941148161888,
      "num_tokens": 160260755.0,
      "step": 79090
    },
    {
      "entropy": 5.858260631561279,
      "epoch": 6.912085991435812,
      "grad_norm": 1.25,
      "learning_rate": 0.0001493871946492198,
      "loss": 5.1519,
      "mean_token_accuracy": 0.22859666496515274,
      "num_tokens": 160271023.0,
      "step": 79095
    },
    {
      "entropy": 5.854550457000732,
      "epoch": 6.912522939788517,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014936134294975435,
      "loss": 5.0989,
      "mean_token_accuracy": 0.22443705499172212,
      "num_tokens": 160280497.0,
      "step": 79100
    },
    {
      "entropy": 5.734835195541382,
      "epoch": 6.912959888141222,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014933549366010412,
      "loss": 5.0209,
      "mean_token_accuracy": 0.2412225693464279,
      "num_tokens": 160290627.0,
      "step": 79105
    },
    {
      "entropy": 5.816523599624634,
      "epoch": 6.913396836493926,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014930964678076502,
      "loss": 5.0968,
      "mean_token_accuracy": 0.2289571210741997,
      "num_tokens": 160300211.0,
      "step": 79110
    },
    {
      "entropy": 5.8115335464477536,
      "epoch": 6.913833784846631,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014928380231223282,
      "loss": 5.0872,
      "mean_token_accuracy": 0.23214104473590852,
      "num_tokens": 160310339.0,
      "step": 79115
    },
    {
      "entropy": 5.866864585876465,
      "epoch": 6.914270733199336,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00014925796025500306,
      "loss": 5.1533,
      "mean_token_accuracy": 0.22727746069431304,
      "num_tokens": 160321450.0,
      "step": 79120
    },
    {
      "entropy": 5.778533029556274,
      "epoch": 6.9147076815520405,
      "grad_norm": 1.125,
      "learning_rate": 0.00014923212060957158,
      "loss": 5.0682,
      "mean_token_accuracy": 0.23086895495653154,
      "num_tokens": 160332095.0,
      "step": 79125
    },
    {
      "entropy": 5.847651863098145,
      "epoch": 6.915144629904745,
      "grad_norm": 1.125,
      "learning_rate": 0.000149206283376434,
      "loss": 5.1379,
      "mean_token_accuracy": 0.23198436796665192,
      "num_tokens": 160342859.0,
      "step": 79130
    },
    {
      "entropy": 5.831871080398559,
      "epoch": 6.91558157825745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001491804485560857,
      "loss": 5.0153,
      "mean_token_accuracy": 0.2403448835015297,
      "num_tokens": 160352726.0,
      "step": 79135
    },
    {
      "entropy": 5.871451807022095,
      "epoch": 6.916018526610155,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014915461614902238,
      "loss": 5.1945,
      "mean_token_accuracy": 0.21141496747732164,
      "num_tokens": 160362192.0,
      "step": 79140
    },
    {
      "entropy": 5.874239206314087,
      "epoch": 6.916455474962859,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014912878615573938,
      "loss": 5.1714,
      "mean_token_accuracy": 0.21960816532373428,
      "num_tokens": 160371680.0,
      "step": 79145
    },
    {
      "entropy": 5.868332624435425,
      "epoch": 6.916892423315564,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00014910295857673228,
      "loss": 5.1218,
      "mean_token_accuracy": 0.22929338812828065,
      "num_tokens": 160381613.0,
      "step": 79150
    },
    {
      "entropy": 5.81124439239502,
      "epoch": 6.917329371668269,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001490771334124964,
      "loss": 5.0543,
      "mean_token_accuracy": 0.23514668345451356,
      "num_tokens": 160392062.0,
      "step": 79155
    },
    {
      "entropy": 5.8303412914276125,
      "epoch": 6.917766320020974,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014905131066352708,
      "loss": 5.1495,
      "mean_token_accuracy": 0.22951315194368363,
      "num_tokens": 160402838.0,
      "step": 79160
    },
    {
      "entropy": 5.867450714111328,
      "epoch": 6.918203268373678,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001490254903303196,
      "loss": 5.1703,
      "mean_token_accuracy": 0.22802277207374572,
      "num_tokens": 160412868.0,
      "step": 79165
    },
    {
      "entropy": 5.844403266906738,
      "epoch": 6.9186402167263825,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014899967241336926,
      "loss": 5.0757,
      "mean_token_accuracy": 0.22870834916830063,
      "num_tokens": 160422927.0,
      "step": 79170
    },
    {
      "entropy": 5.864924716949463,
      "epoch": 6.919077165079088,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014897385691317128,
      "loss": 5.1148,
      "mean_token_accuracy": 0.23153140097856523,
      "num_tokens": 160433250.0,
      "step": 79175
    },
    {
      "entropy": 5.887337398529053,
      "epoch": 6.919514113431792,
      "grad_norm": 1.125,
      "learning_rate": 0.00014894804383022071,
      "loss": 5.1462,
      "mean_token_accuracy": 0.22500489056110382,
      "num_tokens": 160443749.0,
      "step": 79180
    },
    {
      "entropy": 5.840299034118653,
      "epoch": 6.919951061784497,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014892223316501267,
      "loss": 5.0798,
      "mean_token_accuracy": 0.23331968933343888,
      "num_tokens": 160453395.0,
      "step": 79185
    },
    {
      "entropy": 5.8315928936004635,
      "epoch": 6.920388010137202,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014889642491804227,
      "loss": 5.1711,
      "mean_token_accuracy": 0.22646752893924713,
      "num_tokens": 160462607.0,
      "step": 79190
    },
    {
      "entropy": 5.853482341766357,
      "epoch": 6.920824958489907,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014887061908980452,
      "loss": 5.1492,
      "mean_token_accuracy": 0.23134395480155945,
      "num_tokens": 160473051.0,
      "step": 79195
    },
    {
      "entropy": 5.811080551147461,
      "epoch": 6.921261906842611,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001488448156807944,
      "loss": 5.1588,
      "mean_token_accuracy": 0.22841639816761017,
      "num_tokens": 160483532.0,
      "step": 79200
    },
    {
      "entropy": 5.845499753952026,
      "epoch": 6.921698855195316,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001488190146915068,
      "loss": 5.1592,
      "mean_token_accuracy": 0.22746254205703736,
      "num_tokens": 160493922.0,
      "step": 79205
    },
    {
      "entropy": 5.871728563308716,
      "epoch": 6.922135803548021,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014879321612243663,
      "loss": 5.1901,
      "mean_token_accuracy": 0.21620431095361708,
      "num_tokens": 160504243.0,
      "step": 79210
    },
    {
      "entropy": 5.817148447036743,
      "epoch": 6.922572751900725,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014876741997407871,
      "loss": 5.0919,
      "mean_token_accuracy": 0.23196225315332414,
      "num_tokens": 160513767.0,
      "step": 79215
    },
    {
      "entropy": 5.854953956604004,
      "epoch": 6.92300970025343,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014874162624692783,
      "loss": 5.1344,
      "mean_token_accuracy": 0.2240314930677414,
      "num_tokens": 160523722.0,
      "step": 79220
    },
    {
      "entropy": 5.860129070281983,
      "epoch": 6.923446648606134,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014871583494147878,
      "loss": 5.0971,
      "mean_token_accuracy": 0.22906588166952133,
      "num_tokens": 160533674.0,
      "step": 79225
    },
    {
      "entropy": 5.895994663238525,
      "epoch": 6.92388359695884,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014869004605822612,
      "loss": 5.1236,
      "mean_token_accuracy": 0.2250472292304039,
      "num_tokens": 160543968.0,
      "step": 79230
    },
    {
      "entropy": 5.9242119789123535,
      "epoch": 6.924320545311544,
      "grad_norm": 1.25,
      "learning_rate": 0.00014866425959766456,
      "loss": 5.2013,
      "mean_token_accuracy": 0.2171635612845421,
      "num_tokens": 160553339.0,
      "step": 79235
    },
    {
      "entropy": 5.792721176147461,
      "epoch": 6.924757493664249,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00014863847556028864,
      "loss": 5.0465,
      "mean_token_accuracy": 0.23442402184009553,
      "num_tokens": 160564338.0,
      "step": 79240
    },
    {
      "entropy": 5.847083806991577,
      "epoch": 6.925194442016954,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014861269394659304,
      "loss": 5.1552,
      "mean_token_accuracy": 0.22609825283288956,
      "num_tokens": 160574540.0,
      "step": 79245
    },
    {
      "entropy": 5.831756973266602,
      "epoch": 6.9256313903696585,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014858691475707218,
      "loss": 5.1114,
      "mean_token_accuracy": 0.23167545199394227,
      "num_tokens": 160584243.0,
      "step": 79250
    },
    {
      "entropy": 5.853668689727783,
      "epoch": 6.926068338722363,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001485611379922205,
      "loss": 5.1085,
      "mean_token_accuracy": 0.22391507029533386,
      "num_tokens": 160593962.0,
      "step": 79255
    },
    {
      "entropy": 5.804416179656982,
      "epoch": 6.926505287075067,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014853536365253247,
      "loss": 5.1012,
      "mean_token_accuracy": 0.22495632916688918,
      "num_tokens": 160604050.0,
      "step": 79260
    },
    {
      "entropy": 5.808465814590454,
      "epoch": 6.926942235427773,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014850959173850247,
      "loss": 5.129,
      "mean_token_accuracy": 0.23310912698507308,
      "num_tokens": 160613466.0,
      "step": 79265
    },
    {
      "entropy": 5.852941036224365,
      "epoch": 6.927379183780477,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014848382225062475,
      "loss": 5.1756,
      "mean_token_accuracy": 0.22232395708560942,
      "num_tokens": 160624034.0,
      "step": 79270
    },
    {
      "entropy": 5.853905820846558,
      "epoch": 6.927816132133182,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001484580551893937,
      "loss": 5.1879,
      "mean_token_accuracy": 0.21835365891456604,
      "num_tokens": 160634945.0,
      "step": 79275
    },
    {
      "entropy": 5.8489429473876955,
      "epoch": 6.928253080485886,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001484322905553034,
      "loss": 5.149,
      "mean_token_accuracy": 0.22790893763303757,
      "num_tokens": 160645271.0,
      "step": 79280
    },
    {
      "entropy": 5.848199892044067,
      "epoch": 6.9286900288385915,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014840652834884803,
      "loss": 5.1371,
      "mean_token_accuracy": 0.23252401053905486,
      "num_tokens": 160655686.0,
      "step": 79285
    },
    {
      "entropy": 5.843095827102661,
      "epoch": 6.929126977191296,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001483807685705218,
      "loss": 5.1128,
      "mean_token_accuracy": 0.22649357318878174,
      "num_tokens": 160664891.0,
      "step": 79290
    },
    {
      "entropy": 5.792616605758667,
      "epoch": 6.9295639255440005,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014835501122081876,
      "loss": 5.0239,
      "mean_token_accuracy": 0.23422226458787918,
      "num_tokens": 160674353.0,
      "step": 79295
    },
    {
      "entropy": 5.8421824932098385,
      "epoch": 6.930000873896706,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014832925630023304,
      "loss": 5.1798,
      "mean_token_accuracy": 0.22597991526126862,
      "num_tokens": 160684476.0,
      "step": 79300
    },
    {
      "entropy": 5.834081983566284,
      "epoch": 6.93043782224941,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001483035038092585,
      "loss": 5.0249,
      "mean_token_accuracy": 0.234979809820652,
      "num_tokens": 160694392.0,
      "step": 79305
    },
    {
      "entropy": 5.8434031963348385,
      "epoch": 6.930874770602115,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014827775374838913,
      "loss": 5.2097,
      "mean_token_accuracy": 0.22118836790323257,
      "num_tokens": 160704014.0,
      "step": 79310
    },
    {
      "entropy": 5.8090612411499025,
      "epoch": 6.931311718954819,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001482520061181189,
      "loss": 5.0041,
      "mean_token_accuracy": 0.23798396289348603,
      "num_tokens": 160712637.0,
      "step": 79315
    },
    {
      "entropy": 5.887007856369019,
      "epoch": 6.931748667307525,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014822626091894154,
      "loss": 5.14,
      "mean_token_accuracy": 0.22123675495386125,
      "num_tokens": 160722617.0,
      "step": 79320
    },
    {
      "entropy": 5.787435054779053,
      "epoch": 6.932185615660229,
      "grad_norm": 1.109375,
      "learning_rate": 0.00014820051815135106,
      "loss": 5.024,
      "mean_token_accuracy": 0.2416036143898964,
      "num_tokens": 160733515.0,
      "step": 79325
    },
    {
      "entropy": 5.820637941360474,
      "epoch": 6.9326225640129335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014817477781584097,
      "loss": 5.1456,
      "mean_token_accuracy": 0.22707034647464752,
      "num_tokens": 160743349.0,
      "step": 79330
    },
    {
      "entropy": 5.771912956237793,
      "epoch": 6.933059512365638,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001481490399129051,
      "loss": 5.0955,
      "mean_token_accuracy": 0.2340279132127762,
      "num_tokens": 160753987.0,
      "step": 79335
    },
    {
      "entropy": 5.801831293106079,
      "epoch": 6.933496460718343,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001481233044430371,
      "loss": 5.0953,
      "mean_token_accuracy": 0.23036432564258574,
      "num_tokens": 160764041.0,
      "step": 79340
    },
    {
      "entropy": 5.818323040008545,
      "epoch": 6.933933409071048,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014809757140673066,
      "loss": 5.0958,
      "mean_token_accuracy": 0.23331868797540664,
      "num_tokens": 160774521.0,
      "step": 79345
    },
    {
      "entropy": 5.807750034332275,
      "epoch": 6.934370357423752,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014807184080447926,
      "loss": 5.0567,
      "mean_token_accuracy": 0.2307167634367943,
      "num_tokens": 160783919.0,
      "step": 79350
    },
    {
      "entropy": 5.857725381851196,
      "epoch": 6.934807305776458,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014804611263677646,
      "loss": 5.1778,
      "mean_token_accuracy": 0.2267763704061508,
      "num_tokens": 160793354.0,
      "step": 79355
    },
    {
      "entropy": 5.80457615852356,
      "epoch": 6.935244254129162,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014802038690411574,
      "loss": 5.1328,
      "mean_token_accuracy": 0.2247879236936569,
      "num_tokens": 160803663.0,
      "step": 79360
    },
    {
      "entropy": 5.786164093017578,
      "epoch": 6.935681202481867,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014799466360699055,
      "loss": 5.0738,
      "mean_token_accuracy": 0.22970558404922486,
      "num_tokens": 160814977.0,
      "step": 79365
    },
    {
      "entropy": 5.787948369979858,
      "epoch": 6.936118150834571,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014796894274589427,
      "loss": 5.0817,
      "mean_token_accuracy": 0.24036647379398346,
      "num_tokens": 160825073.0,
      "step": 79370
    },
    {
      "entropy": 5.859035396575928,
      "epoch": 6.936555099187276,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014794322432132022,
      "loss": 5.095,
      "mean_token_accuracy": 0.22611771672964096,
      "num_tokens": 160834726.0,
      "step": 79375
    },
    {
      "entropy": 5.876909446716309,
      "epoch": 6.936992047539981,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014791750833376165,
      "loss": 5.1656,
      "mean_token_accuracy": 0.2244714692234993,
      "num_tokens": 160845498.0,
      "step": 79380
    },
    {
      "entropy": 5.8343085765838625,
      "epoch": 6.937428995892685,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014789179478371194,
      "loss": 5.0846,
      "mean_token_accuracy": 0.2305150732398033,
      "num_tokens": 160855085.0,
      "step": 79385
    },
    {
      "entropy": 5.8314549922943115,
      "epoch": 6.93786594424539,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014786608367166418,
      "loss": 5.1033,
      "mean_token_accuracy": 0.23223934769630433,
      "num_tokens": 160865006.0,
      "step": 79390
    },
    {
      "entropy": 5.814495277404785,
      "epoch": 6.938302892598095,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014784037499811155,
      "loss": 5.1242,
      "mean_token_accuracy": 0.22961210161447526,
      "num_tokens": 160874730.0,
      "step": 79395
    },
    {
      "entropy": 5.822061824798584,
      "epoch": 6.9387398409508,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001478146687635472,
      "loss": 5.1302,
      "mean_token_accuracy": 0.2313792809844017,
      "num_tokens": 160885254.0,
      "step": 79400
    },
    {
      "entropy": 5.798679494857788,
      "epoch": 6.939176789303504,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001477889649684641,
      "loss": 5.0702,
      "mean_token_accuracy": 0.2374751940369606,
      "num_tokens": 160895578.0,
      "step": 79405
    },
    {
      "entropy": 5.863584852218628,
      "epoch": 6.9396137376562095,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014776326361335538,
      "loss": 5.1562,
      "mean_token_accuracy": 0.2258582293987274,
      "num_tokens": 160904927.0,
      "step": 79410
    },
    {
      "entropy": 5.843097257614136,
      "epoch": 6.940050686008914,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001477375646987139,
      "loss": 5.1546,
      "mean_token_accuracy": 0.21967836022377013,
      "num_tokens": 160914978.0,
      "step": 79415
    },
    {
      "entropy": 5.887016105651855,
      "epoch": 6.940487634361618,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014771186822503258,
      "loss": 5.1878,
      "mean_token_accuracy": 0.22105126827955246,
      "num_tokens": 160925242.0,
      "step": 79420
    },
    {
      "entropy": 5.846437263488769,
      "epoch": 6.940924582714323,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014768617419280431,
      "loss": 5.0578,
      "mean_token_accuracy": 0.2304573491215706,
      "num_tokens": 160934348.0,
      "step": 79425
    },
    {
      "entropy": 5.759840774536133,
      "epoch": 6.941361531067028,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014766048260252197,
      "loss": 5.0424,
      "mean_token_accuracy": 0.2405494898557663,
      "num_tokens": 160944780.0,
      "step": 79430
    },
    {
      "entropy": 5.795964765548706,
      "epoch": 6.941798479419733,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001476347934546783,
      "loss": 5.1419,
      "mean_token_accuracy": 0.22214934527873992,
      "num_tokens": 160954882.0,
      "step": 79435
    },
    {
      "entropy": 5.761165285110474,
      "epoch": 6.942235427772437,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014760910674976597,
      "loss": 5.0898,
      "mean_token_accuracy": 0.23070093840360642,
      "num_tokens": 160964852.0,
      "step": 79440
    },
    {
      "entropy": 5.865723133087158,
      "epoch": 6.942672376125142,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014758342248827777,
      "loss": 5.1113,
      "mean_token_accuracy": 0.22442458868026732,
      "num_tokens": 160974806.0,
      "step": 79445
    },
    {
      "entropy": 5.802331972122192,
      "epoch": 6.943109324477847,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00014755774067070626,
      "loss": 5.0176,
      "mean_token_accuracy": 0.23737716674804688,
      "num_tokens": 160984962.0,
      "step": 79450
    },
    {
      "entropy": 5.873596239089966,
      "epoch": 6.9435462728305515,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001475320612975441,
      "loss": 5.1165,
      "mean_token_accuracy": 0.2234298124909401,
      "num_tokens": 160996060.0,
      "step": 79455
    },
    {
      "entropy": 5.826400518417358,
      "epoch": 6.943983221183256,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014750638436928387,
      "loss": 5.1177,
      "mean_token_accuracy": 0.22348145991563798,
      "num_tokens": 161006393.0,
      "step": 79460
    },
    {
      "entropy": 5.813255882263183,
      "epoch": 6.944420169535961,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00014748070988641792,
      "loss": 5.1326,
      "mean_token_accuracy": 0.23057890683412552,
      "num_tokens": 161016862.0,
      "step": 79465
    },
    {
      "entropy": 5.819608163833618,
      "epoch": 6.944857117888666,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014745503784943876,
      "loss": 5.15,
      "mean_token_accuracy": 0.22428754568099976,
      "num_tokens": 161027593.0,
      "step": 79470
    },
    {
      "entropy": 5.8477894306182865,
      "epoch": 6.94529406624137,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014742936825883882,
      "loss": 5.0816,
      "mean_token_accuracy": 0.22879349142313005,
      "num_tokens": 161036435.0,
      "step": 79475
    },
    {
      "entropy": 5.877334070205689,
      "epoch": 6.945731014594075,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001474037011151104,
      "loss": 5.1163,
      "mean_token_accuracy": 0.22994970679283142,
      "num_tokens": 161047002.0,
      "step": 79480
    },
    {
      "entropy": 5.837790012359619,
      "epoch": 6.94616796294678,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00014737803641874593,
      "loss": 5.1737,
      "mean_token_accuracy": 0.22807820588350297,
      "num_tokens": 161057437.0,
      "step": 79485
    },
    {
      "entropy": 5.849143552780151,
      "epoch": 6.9466049112994845,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014735237417023753,
      "loss": 5.1475,
      "mean_token_accuracy": 0.22295186817646026,
      "num_tokens": 161066288.0,
      "step": 79490
    },
    {
      "entropy": 5.848408603668213,
      "epoch": 6.947041859652189,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001473267143700775,
      "loss": 5.141,
      "mean_token_accuracy": 0.2238974541425705,
      "num_tokens": 161075727.0,
      "step": 79495
    },
    {
      "entropy": 5.841929864883423,
      "epoch": 6.9474788080048935,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014730105701875795,
      "loss": 5.0853,
      "mean_token_accuracy": 0.2299799233675003,
      "num_tokens": 161086138.0,
      "step": 79500
    },
    {
      "entropy": 5.812687397003174,
      "epoch": 6.947915756357599,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001472754021167712,
      "loss": 5.1018,
      "mean_token_accuracy": 0.23517228811979293,
      "num_tokens": 161096246.0,
      "step": 79505
    },
    {
      "entropy": 5.789494800567627,
      "epoch": 6.948352704710303,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014724974966460902,
      "loss": 5.091,
      "mean_token_accuracy": 0.2287714570760727,
      "num_tokens": 161107179.0,
      "step": 79510
    },
    {
      "entropy": 5.864368629455567,
      "epoch": 6.948789653063008,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001472240996627636,
      "loss": 5.1356,
      "mean_token_accuracy": 0.22132529616355895,
      "num_tokens": 161117573.0,
      "step": 79515
    },
    {
      "entropy": 5.8781962394714355,
      "epoch": 6.949226601415713,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001471984521117269,
      "loss": 5.1889,
      "mean_token_accuracy": 0.21427354216575623,
      "num_tokens": 161127289.0,
      "step": 79520
    },
    {
      "entropy": 5.828091287612915,
      "epoch": 6.949663549768418,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014717280701199086,
      "loss": 5.0146,
      "mean_token_accuracy": 0.24646969884634018,
      "num_tokens": 161136581.0,
      "step": 79525
    },
    {
      "entropy": 5.718423700332641,
      "epoch": 6.950100498121122,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014714716436404736,
      "loss": 5.0235,
      "mean_token_accuracy": 0.24822816848754883,
      "num_tokens": 161145839.0,
      "step": 79530
    },
    {
      "entropy": 5.84046630859375,
      "epoch": 6.9505374464738265,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014712152416838825,
      "loss": 5.1847,
      "mean_token_accuracy": 0.22437905371189118,
      "num_tokens": 161156504.0,
      "step": 79535
    },
    {
      "entropy": 5.867692470550537,
      "epoch": 6.950974394826531,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014709588642550532,
      "loss": 5.0861,
      "mean_token_accuracy": 0.22961564213037491,
      "num_tokens": 161165264.0,
      "step": 79540
    },
    {
      "entropy": 5.82673487663269,
      "epoch": 6.951411343179236,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014707025113589028,
      "loss": 5.1777,
      "mean_token_accuracy": 0.22216662466526033,
      "num_tokens": 161175959.0,
      "step": 79545
    },
    {
      "entropy": 5.850141811370849,
      "epoch": 6.951848291531941,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014704461830003485,
      "loss": 5.1172,
      "mean_token_accuracy": 0.2313404843211174,
      "num_tokens": 161186129.0,
      "step": 79550
    },
    {
      "entropy": 5.780645704269409,
      "epoch": 6.952285239884645,
      "grad_norm": 1.140625,
      "learning_rate": 0.00014701898791843083,
      "loss": 5.0058,
      "mean_token_accuracy": 0.23897765576839447,
      "num_tokens": 161196005.0,
      "step": 79555
    },
    {
      "entropy": 5.891268062591553,
      "epoch": 6.952722188237351,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014699335999156956,
      "loss": 5.1854,
      "mean_token_accuracy": 0.22176852226257324,
      "num_tokens": 161205441.0,
      "step": 79560
    },
    {
      "entropy": 5.763437938690186,
      "epoch": 6.953159136590055,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014696773451994272,
      "loss": 5.0659,
      "mean_token_accuracy": 0.23881304264068604,
      "num_tokens": 161215420.0,
      "step": 79565
    },
    {
      "entropy": 5.769033050537109,
      "epoch": 6.95359608494276,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014694211150404182,
      "loss": 5.0765,
      "mean_token_accuracy": 0.22236306071281434,
      "num_tokens": 161226744.0,
      "step": 79570
    },
    {
      "entropy": 5.836979055404663,
      "epoch": 6.954033033295465,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014691649094435833,
      "loss": 5.1282,
      "mean_token_accuracy": 0.23339296579360963,
      "num_tokens": 161236935.0,
      "step": 79575
    },
    {
      "entropy": 5.851547479629517,
      "epoch": 6.954469981648169,
      "grad_norm": 1.125,
      "learning_rate": 0.00014689087284138364,
      "loss": 5.135,
      "mean_token_accuracy": 0.22802358269691467,
      "num_tokens": 161245864.0,
      "step": 79580
    },
    {
      "entropy": 5.886496925354004,
      "epoch": 6.954906930000874,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014686525719560917,
      "loss": 5.1197,
      "mean_token_accuracy": 0.2258894622325897,
      "num_tokens": 161256051.0,
      "step": 79585
    },
    {
      "entropy": 5.804137802124023,
      "epoch": 6.955343878353578,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001468396440075262,
      "loss": 5.1487,
      "mean_token_accuracy": 0.22907410115003585,
      "num_tokens": 161266513.0,
      "step": 79590
    },
    {
      "entropy": 5.8224279403686525,
      "epoch": 6.955780826706283,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014681403327762598,
      "loss": 5.0872,
      "mean_token_accuracy": 0.23104315102100373,
      "num_tokens": 161277035.0,
      "step": 79595
    },
    {
      "entropy": 5.861011075973511,
      "epoch": 6.956217775058988,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014678842500639977,
      "loss": 5.1841,
      "mean_token_accuracy": 0.22595782577991486,
      "num_tokens": 161287263.0,
      "step": 79600
    },
    {
      "entropy": 5.856982946395874,
      "epoch": 6.956654723411693,
      "grad_norm": 1.140625,
      "learning_rate": 0.00014676281919433887,
      "loss": 5.1421,
      "mean_token_accuracy": 0.22692172825336457,
      "num_tokens": 161297690.0,
      "step": 79605
    },
    {
      "entropy": 5.875106287002564,
      "epoch": 6.957091671764397,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00014673721584193423,
      "loss": 5.1353,
      "mean_token_accuracy": 0.22338371574878693,
      "num_tokens": 161309289.0,
      "step": 79610
    },
    {
      "entropy": 5.888733291625977,
      "epoch": 6.9575286201171025,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014671161494967695,
      "loss": 5.2043,
      "mean_token_accuracy": 0.21934226304292678,
      "num_tokens": 161320213.0,
      "step": 79615
    },
    {
      "entropy": 5.861879491806031,
      "epoch": 6.957965568469807,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014668601651805812,
      "loss": 5.1263,
      "mean_token_accuracy": 0.22988396883010864,
      "num_tokens": 161331196.0,
      "step": 79620
    },
    {
      "entropy": 5.837471961975098,
      "epoch": 6.958402516822511,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00014666042054756873,
      "loss": 5.1099,
      "mean_token_accuracy": 0.2253287762403488,
      "num_tokens": 161341529.0,
      "step": 79625
    },
    {
      "entropy": 5.897751951217652,
      "epoch": 6.958839465175217,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014663482703869976,
      "loss": 5.2427,
      "mean_token_accuracy": 0.2106556251645088,
      "num_tokens": 161352864.0,
      "step": 79630
    },
    {
      "entropy": 5.811904525756836,
      "epoch": 6.959276413527921,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014660923599194204,
      "loss": 5.1051,
      "mean_token_accuracy": 0.23201262801885605,
      "num_tokens": 161362911.0,
      "step": 79635
    },
    {
      "entropy": 5.884804916381836,
      "epoch": 6.959713361880626,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014658364740778647,
      "loss": 5.1701,
      "mean_token_accuracy": 0.22929092049598693,
      "num_tokens": 161372667.0,
      "step": 79640
    },
    {
      "entropy": 5.813298320770263,
      "epoch": 6.96015031023333,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014655806128672377,
      "loss": 5.1145,
      "mean_token_accuracy": 0.22766423970460892,
      "num_tokens": 161382521.0,
      "step": 79645
    },
    {
      "entropy": 5.830141019821167,
      "epoch": 6.960587258586035,
      "grad_norm": 1.140625,
      "learning_rate": 0.00014653247762924488,
      "loss": 5.1032,
      "mean_token_accuracy": 0.22928989231586455,
      "num_tokens": 161392336.0,
      "step": 79650
    },
    {
      "entropy": 5.857625246047974,
      "epoch": 6.96102420693874,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001465068964358403,
      "loss": 5.0965,
      "mean_token_accuracy": 0.22632303833961487,
      "num_tokens": 161401817.0,
      "step": 79655
    },
    {
      "entropy": 5.780534648895264,
      "epoch": 6.9614611552914445,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001464813177070008,
      "loss": 5.091,
      "mean_token_accuracy": 0.23151164203882219,
      "num_tokens": 161412520.0,
      "step": 79660
    },
    {
      "entropy": 5.797271728515625,
      "epoch": 6.961898103644149,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014645574144321694,
      "loss": 5.042,
      "mean_token_accuracy": 0.2392318829894066,
      "num_tokens": 161422494.0,
      "step": 79665
    },
    {
      "entropy": 5.859785890579223,
      "epoch": 6.962335051996854,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014643016764497935,
      "loss": 5.1146,
      "mean_token_accuracy": 0.22942466139793397,
      "num_tokens": 161431444.0,
      "step": 79670
    },
    {
      "entropy": 5.807456970214844,
      "epoch": 6.962772000349559,
      "grad_norm": 1.125,
      "learning_rate": 0.00014640459631277847,
      "loss": 5.1095,
      "mean_token_accuracy": 0.2216830551624298,
      "num_tokens": 161441932.0,
      "step": 79675
    },
    {
      "entropy": 5.780512714385987,
      "epoch": 6.963208948702263,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014637902744710484,
      "loss": 5.1009,
      "mean_token_accuracy": 0.2344687968492508,
      "num_tokens": 161451794.0,
      "step": 79680
    },
    {
      "entropy": 5.822278690338135,
      "epoch": 6.963645897054968,
      "grad_norm": 1.125,
      "learning_rate": 0.00014635346104844888,
      "loss": 5.1498,
      "mean_token_accuracy": 0.218816314637661,
      "num_tokens": 161462777.0,
      "step": 79685
    },
    {
      "entropy": 5.85160231590271,
      "epoch": 6.964082845407673,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014632789711730093,
      "loss": 5.1035,
      "mean_token_accuracy": 0.22096503973007203,
      "num_tokens": 161473204.0,
      "step": 79690
    },
    {
      "entropy": 5.903257369995117,
      "epoch": 6.9645197937603776,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001463023356541513,
      "loss": 5.1775,
      "mean_token_accuracy": 0.22865863740444184,
      "num_tokens": 161484334.0,
      "step": 79695
    },
    {
      "entropy": 5.873321104049682,
      "epoch": 6.964956742113082,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014627677665949044,
      "loss": 5.1119,
      "mean_token_accuracy": 0.22841818183660506,
      "num_tokens": 161494400.0,
      "step": 79700
    },
    {
      "entropy": 5.83343539237976,
      "epoch": 6.9653936904657865,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014625122013380836,
      "loss": 5.118,
      "mean_token_accuracy": 0.23303980231285096,
      "num_tokens": 161504088.0,
      "step": 79705
    },
    {
      "entropy": 5.870010566711426,
      "epoch": 6.965830638818492,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001462256660775953,
      "loss": 5.124,
      "mean_token_accuracy": 0.22502999305725097,
      "num_tokens": 161514191.0,
      "step": 79710
    },
    {
      "entropy": 5.840807771682739,
      "epoch": 6.966267587171196,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014620011449134149,
      "loss": 5.1079,
      "mean_token_accuracy": 0.22647328078746795,
      "num_tokens": 161523675.0,
      "step": 79715
    },
    {
      "entropy": 5.803449249267578,
      "epoch": 6.966704535523901,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014617456537553697,
      "loss": 5.1464,
      "mean_token_accuracy": 0.22832464128732682,
      "num_tokens": 161534130.0,
      "step": 79720
    },
    {
      "entropy": 5.843276453018189,
      "epoch": 6.967141483876606,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001461490187306717,
      "loss": 5.1599,
      "mean_token_accuracy": 0.22617266029119493,
      "num_tokens": 161543326.0,
      "step": 79725
    },
    {
      "entropy": 5.804865550994873,
      "epoch": 6.967578432229311,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014612347455723586,
      "loss": 5.072,
      "mean_token_accuracy": 0.2305322453379631,
      "num_tokens": 161553372.0,
      "step": 79730
    },
    {
      "entropy": 5.785096216201782,
      "epoch": 6.968015380582015,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014609793285571927,
      "loss": 4.974,
      "mean_token_accuracy": 0.23538341671228408,
      "num_tokens": 161563569.0,
      "step": 79735
    },
    {
      "entropy": 5.790729713439942,
      "epoch": 6.9684523289347196,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014607239362661198,
      "loss": 5.0983,
      "mean_token_accuracy": 0.22884358316659928,
      "num_tokens": 161573156.0,
      "step": 79740
    },
    {
      "entropy": 5.846676826477051,
      "epoch": 6.968889277287425,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014604685687040364,
      "loss": 5.0911,
      "mean_token_accuracy": 0.22945598214864732,
      "num_tokens": 161582699.0,
      "step": 79745
    },
    {
      "entropy": 5.840509510040283,
      "epoch": 6.969326225640129,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014602132258758417,
      "loss": 5.0903,
      "mean_token_accuracy": 0.22997875958681108,
      "num_tokens": 161593002.0,
      "step": 79750
    },
    {
      "entropy": 5.9409425258636475,
      "epoch": 6.969763173992834,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001459957907786433,
      "loss": 5.2251,
      "mean_token_accuracy": 0.21949010193347931,
      "num_tokens": 161603125.0,
      "step": 79755
    },
    {
      "entropy": 5.850868463516235,
      "epoch": 6.970200122345538,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001459702614440708,
      "loss": 5.1294,
      "mean_token_accuracy": 0.23080655336380004,
      "num_tokens": 161612215.0,
      "step": 79760
    },
    {
      "entropy": 5.772415447235107,
      "epoch": 6.970637070698244,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014594473458435624,
      "loss": 5.0074,
      "mean_token_accuracy": 0.23659236282110213,
      "num_tokens": 161621657.0,
      "step": 79765
    },
    {
      "entropy": 5.779409313201905,
      "epoch": 6.971074019050948,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014591921019998933,
      "loss": 5.0618,
      "mean_token_accuracy": 0.2264306753873825,
      "num_tokens": 161631753.0,
      "step": 79770
    },
    {
      "entropy": 5.794551277160645,
      "epoch": 6.971510967403653,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014589368829145963,
      "loss": 5.1594,
      "mean_token_accuracy": 0.22930776178836823,
      "num_tokens": 161640909.0,
      "step": 79775
    },
    {
      "entropy": 5.803687953948975,
      "epoch": 6.971947915756358,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014586816885925658,
      "loss": 5.0792,
      "mean_token_accuracy": 0.23521366864442825,
      "num_tokens": 161651137.0,
      "step": 79780
    },
    {
      "entropy": 5.845685386657715,
      "epoch": 6.972384864109062,
      "grad_norm": 1.125,
      "learning_rate": 0.00014584265190386987,
      "loss": 5.1292,
      "mean_token_accuracy": 0.22155859172344208,
      "num_tokens": 161661770.0,
      "step": 79785
    },
    {
      "entropy": 5.81899061203003,
      "epoch": 6.972821812461767,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014581713742578862,
      "loss": 5.0982,
      "mean_token_accuracy": 0.23174673020839692,
      "num_tokens": 161671010.0,
      "step": 79790
    },
    {
      "entropy": 5.838154268264771,
      "epoch": 6.973258760814471,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001457916254255024,
      "loss": 5.1784,
      "mean_token_accuracy": 0.22157174050807954,
      "num_tokens": 161680913.0,
      "step": 79795
    },
    {
      "entropy": 5.830982208251953,
      "epoch": 6.973695709167177,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014576611590350047,
      "loss": 5.0852,
      "mean_token_accuracy": 0.2290141612291336,
      "num_tokens": 161690971.0,
      "step": 79800
    },
    {
      "entropy": 5.8419411182403564,
      "epoch": 6.974132657519881,
      "grad_norm": 1.125,
      "learning_rate": 0.00014574060886027223,
      "loss": 5.1586,
      "mean_token_accuracy": 0.2314160004258156,
      "num_tokens": 161701495.0,
      "step": 79805
    },
    {
      "entropy": 5.9083778858184814,
      "epoch": 6.974569605872586,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014571510429630676,
      "loss": 5.2008,
      "mean_token_accuracy": 0.216614006459713,
      "num_tokens": 161711227.0,
      "step": 79810
    },
    {
      "entropy": 5.870453023910523,
      "epoch": 6.97500655422529,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001456896022120934,
      "loss": 5.1212,
      "mean_token_accuracy": 0.22800654917955399,
      "num_tokens": 161721962.0,
      "step": 79815
    },
    {
      "entropy": 5.834214162826538,
      "epoch": 6.9754435025779955,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00014566410260812118,
      "loss": 5.1227,
      "mean_token_accuracy": 0.23631850332021714,
      "num_tokens": 161732537.0,
      "step": 79820
    },
    {
      "entropy": 5.843114995956421,
      "epoch": 6.9758804509307,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001456386054848793,
      "loss": 5.125,
      "mean_token_accuracy": 0.2320607528090477,
      "num_tokens": 161742938.0,
      "step": 79825
    },
    {
      "entropy": 5.942194271087646,
      "epoch": 6.976317399283404,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00014561311084285673,
      "loss": 5.2061,
      "mean_token_accuracy": 0.21853770911693574,
      "num_tokens": 161754540.0,
      "step": 79830
    },
    {
      "entropy": 5.869139337539673,
      "epoch": 6.97675434763611,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001455876186825425,
      "loss": 5.11,
      "mean_token_accuracy": 0.22602218091487886,
      "num_tokens": 161765040.0,
      "step": 79835
    },
    {
      "entropy": 5.8507452487945555,
      "epoch": 6.977191295988814,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014556212900442556,
      "loss": 5.0792,
      "mean_token_accuracy": 0.23248884528875352,
      "num_tokens": 161775020.0,
      "step": 79840
    },
    {
      "entropy": 5.814744234085083,
      "epoch": 6.977628244341519,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001455366418089948,
      "loss": 5.1176,
      "mean_token_accuracy": 0.225098879635334,
      "num_tokens": 161786328.0,
      "step": 79845
    },
    {
      "entropy": 5.758963871002197,
      "epoch": 6.978065192694223,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014551115709673904,
      "loss": 5.0786,
      "mean_token_accuracy": 0.2432709276676178,
      "num_tokens": 161797316.0,
      "step": 79850
    },
    {
      "entropy": 5.90745096206665,
      "epoch": 6.978502141046929,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014548567486814717,
      "loss": 5.137,
      "mean_token_accuracy": 0.2250998631119728,
      "num_tokens": 161807265.0,
      "step": 79855
    },
    {
      "entropy": 5.810847187042237,
      "epoch": 6.978939089399633,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014546019512370792,
      "loss": 5.1087,
      "mean_token_accuracy": 0.22546159625053405,
      "num_tokens": 161817624.0,
      "step": 79860
    },
    {
      "entropy": 5.79423828125,
      "epoch": 6.9793760377523375,
      "grad_norm": 1.25,
      "learning_rate": 0.00014543471786391,
      "loss": 5.0751,
      "mean_token_accuracy": 0.23217845857143402,
      "num_tokens": 161828040.0,
      "step": 79865
    },
    {
      "entropy": 5.806146764755249,
      "epoch": 6.979812986105042,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014540924308924208,
      "loss": 5.0654,
      "mean_token_accuracy": 0.2367757961153984,
      "num_tokens": 161837457.0,
      "step": 79870
    },
    {
      "entropy": 5.803525733947754,
      "epoch": 6.980249934457747,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001453837708001928,
      "loss": 5.1021,
      "mean_token_accuracy": 0.23325297087430955,
      "num_tokens": 161847391.0,
      "step": 79875
    },
    {
      "entropy": 5.864078569412231,
      "epoch": 6.980686882810452,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001453583009972507,
      "loss": 5.1326,
      "mean_token_accuracy": 0.23606744259595872,
      "num_tokens": 161858576.0,
      "step": 79880
    },
    {
      "entropy": 5.859669780731201,
      "epoch": 6.981123831163156,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014533283368090442,
      "loss": 5.1521,
      "mean_token_accuracy": 0.2265734925866127,
      "num_tokens": 161869321.0,
      "step": 79885
    },
    {
      "entropy": 5.900615882873535,
      "epoch": 6.981560779515862,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001453073688516422,
      "loss": 5.1538,
      "mean_token_accuracy": 0.22474009543657303,
      "num_tokens": 161879128.0,
      "step": 79890
    },
    {
      "entropy": 5.895810127258301,
      "epoch": 6.981997727868566,
      "grad_norm": 1.125,
      "learning_rate": 0.00014528190650995267,
      "loss": 5.1975,
      "mean_token_accuracy": 0.2167372465133667,
      "num_tokens": 161889736.0,
      "step": 79895
    },
    {
      "entropy": 5.908981370925903,
      "epoch": 6.982434676221271,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001452564466563241,
      "loss": 5.1879,
      "mean_token_accuracy": 0.22479408085346222,
      "num_tokens": 161899925.0,
      "step": 79900
    },
    {
      "entropy": 5.896118116378784,
      "epoch": 6.982871624573975,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001452309892912449,
      "loss": 5.1944,
      "mean_token_accuracy": 0.22425211519002913,
      "num_tokens": 161910312.0,
      "step": 79905
    },
    {
      "entropy": 5.852931118011474,
      "epoch": 6.98330857292668,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014520553441520328,
      "loss": 5.1231,
      "mean_token_accuracy": 0.23347239792346955,
      "num_tokens": 161921068.0,
      "step": 79910
    },
    {
      "entropy": 5.854577827453613,
      "epoch": 6.983745521279385,
      "grad_norm": 1.25,
      "learning_rate": 0.00014518008202868754,
      "loss": 5.1239,
      "mean_token_accuracy": 0.22039324045181274,
      "num_tokens": 161931120.0,
      "step": 79915
    },
    {
      "entropy": 5.884124040603638,
      "epoch": 6.984182469632089,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014515463213218586,
      "loss": 5.1047,
      "mean_token_accuracy": 0.22989111989736558,
      "num_tokens": 161940422.0,
      "step": 79920
    },
    {
      "entropy": 5.688590621948242,
      "epoch": 6.984619417984794,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014512918472618636,
      "loss": 5.0284,
      "mean_token_accuracy": 0.24835580140352248,
      "num_tokens": 161949972.0,
      "step": 79925
    },
    {
      "entropy": 5.827564525604248,
      "epoch": 6.985056366337499,
      "grad_norm": 1.140625,
      "learning_rate": 0.00014510373981117724,
      "loss": 5.1302,
      "mean_token_accuracy": 0.23033603131771088,
      "num_tokens": 161960665.0,
      "step": 79930
    },
    {
      "entropy": 5.770532464981079,
      "epoch": 6.985493314690204,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001450782973876464,
      "loss": 5.035,
      "mean_token_accuracy": 0.24189864248037338,
      "num_tokens": 161970342.0,
      "step": 79935
    },
    {
      "entropy": 5.745302152633667,
      "epoch": 6.985930263042908,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014505285745608184,
      "loss": 5.0604,
      "mean_token_accuracy": 0.23590618818998338,
      "num_tokens": 161980714.0,
      "step": 79940
    },
    {
      "entropy": 5.755447912216186,
      "epoch": 6.9863672113956135,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001450274200169716,
      "loss": 5.1395,
      "mean_token_accuracy": 0.22840163856744766,
      "num_tokens": 161990799.0,
      "step": 79945
    },
    {
      "entropy": 5.829664802551269,
      "epoch": 6.986804159748318,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00014500198507080354,
      "loss": 5.1002,
      "mean_token_accuracy": 0.22754012644290925,
      "num_tokens": 162000749.0,
      "step": 79950
    },
    {
      "entropy": 5.854704809188843,
      "epoch": 6.987241108101022,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014497655261806551,
      "loss": 5.1073,
      "mean_token_accuracy": 0.22489791214466096,
      "num_tokens": 162009918.0,
      "step": 79955
    },
    {
      "entropy": 5.810322570800781,
      "epoch": 6.987678056453727,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014495112265924538,
      "loss": 5.0849,
      "mean_token_accuracy": 0.23271500915288926,
      "num_tokens": 162019646.0,
      "step": 79960
    },
    {
      "entropy": 5.851027297973633,
      "epoch": 6.988115004806432,
      "grad_norm": 1.25,
      "learning_rate": 0.00014492569519483078,
      "loss": 5.1308,
      "mean_token_accuracy": 0.22865765392780305,
      "num_tokens": 162029486.0,
      "step": 79965
    },
    {
      "entropy": 5.807290363311767,
      "epoch": 6.988551953159137,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014490027022530955,
      "loss": 5.1085,
      "mean_token_accuracy": 0.22780486941337585,
      "num_tokens": 162039408.0,
      "step": 79970
    },
    {
      "entropy": 5.8793645858764645,
      "epoch": 6.988988901511841,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014487484775116934,
      "loss": 5.1903,
      "mean_token_accuracy": 0.2212633326649666,
      "num_tokens": 162049008.0,
      "step": 79975
    },
    {
      "entropy": 5.8799458026885985,
      "epoch": 6.989425849864546,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001448494277728978,
      "loss": 5.1542,
      "mean_token_accuracy": 0.22704489678144454,
      "num_tokens": 162058889.0,
      "step": 79980
    },
    {
      "entropy": 5.865748405456543,
      "epoch": 6.989862798217251,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001448240102909823,
      "loss": 5.1329,
      "mean_token_accuracy": 0.2271329328417778,
      "num_tokens": 162069751.0,
      "step": 79985
    },
    {
      "entropy": 5.85212926864624,
      "epoch": 6.9902997465699555,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014479859530591054,
      "loss": 5.0579,
      "mean_token_accuracy": 0.2360694721341133,
      "num_tokens": 162079315.0,
      "step": 79990
    },
    {
      "entropy": 5.916694641113281,
      "epoch": 6.99073669492266,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014477318281816992,
      "loss": 5.1828,
      "mean_token_accuracy": 0.21714544296264648,
      "num_tokens": 162089992.0,
      "step": 79995
    },
    {
      "entropy": 5.824580287933349,
      "epoch": 6.991173643275365,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001447477728282479,
      "loss": 5.1326,
      "mean_token_accuracy": 0.2287191182374954,
      "num_tokens": 162099802.0,
      "step": 80000
    },
    {
      "entropy": 5.765769958496094,
      "epoch": 6.99161059162807,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00014472236533663192,
      "loss": 5.0168,
      "mean_token_accuracy": 0.23560865223407745,
      "num_tokens": 162110570.0,
      "step": 80005
    },
    {
      "entropy": 5.8739406108856205,
      "epoch": 6.992047539980774,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014469696034380913,
      "loss": 5.202,
      "mean_token_accuracy": 0.2186524137854576,
      "num_tokens": 162121996.0,
      "step": 80010
    },
    {
      "entropy": 5.886889934539795,
      "epoch": 6.992484488333479,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014467155785026698,
      "loss": 5.1152,
      "mean_token_accuracy": 0.22256602197885514,
      "num_tokens": 162132464.0,
      "step": 80015
    },
    {
      "entropy": 5.838918256759643,
      "epoch": 6.992921436686184,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014464615785649262,
      "loss": 5.1606,
      "mean_token_accuracy": 0.21978870332241057,
      "num_tokens": 162142891.0,
      "step": 80020
    },
    {
      "entropy": 5.872475528717041,
      "epoch": 6.9933583850388885,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014462076036297328,
      "loss": 5.1423,
      "mean_token_accuracy": 0.22442663758993148,
      "num_tokens": 162152938.0,
      "step": 80025
    },
    {
      "entropy": 5.870672082901001,
      "epoch": 6.993795333391593,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014459536537019614,
      "loss": 5.163,
      "mean_token_accuracy": 0.2249816507101059,
      "num_tokens": 162164095.0,
      "step": 80030
    },
    {
      "entropy": 5.8594136238098145,
      "epoch": 6.9942322817442975,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001445699728786481,
      "loss": 5.1423,
      "mean_token_accuracy": 0.22452556043863298,
      "num_tokens": 162174677.0,
      "step": 80035
    },
    {
      "entropy": 5.866735553741455,
      "epoch": 6.994669230097003,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001445445828888164,
      "loss": 5.1341,
      "mean_token_accuracy": 0.22405947893857955,
      "num_tokens": 162184662.0,
      "step": 80040
    },
    {
      "entropy": 5.871583461761475,
      "epoch": 6.995106178449707,
      "grad_norm": 1.25,
      "learning_rate": 0.0001445191954011879,
      "loss": 5.1644,
      "mean_token_accuracy": 0.2234962686896324,
      "num_tokens": 162194032.0,
      "step": 80045
    },
    {
      "entropy": 5.871711158752442,
      "epoch": 6.995543126802412,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014449381041624964,
      "loss": 5.1306,
      "mean_token_accuracy": 0.2263210341334343,
      "num_tokens": 162204403.0,
      "step": 80050
    },
    {
      "entropy": 5.840488243103027,
      "epoch": 6.995980075155117,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014446842793448844,
      "loss": 5.1412,
      "mean_token_accuracy": 0.2275191366672516,
      "num_tokens": 162214288.0,
      "step": 80055
    },
    {
      "entropy": 5.806001424789429,
      "epoch": 6.996417023507822,
      "grad_norm": 1.140625,
      "learning_rate": 0.00014444304795639123,
      "loss": 5.0822,
      "mean_token_accuracy": 0.2256384775042534,
      "num_tokens": 162224343.0,
      "step": 80060
    },
    {
      "entropy": 5.855702638626099,
      "epoch": 6.996853971860526,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00014441767048244475,
      "loss": 5.0988,
      "mean_token_accuracy": 0.2285640835762024,
      "num_tokens": 162233911.0,
      "step": 80065
    },
    {
      "entropy": 5.8687643051147464,
      "epoch": 6.9972909202132305,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001443922955131358,
      "loss": 5.1534,
      "mean_token_accuracy": 0.22111861407756805,
      "num_tokens": 162243495.0,
      "step": 80070
    },
    {
      "entropy": 5.8070862770080565,
      "epoch": 6.997727868565936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000144366923048951,
      "loss": 5.1188,
      "mean_token_accuracy": 0.23805126994848252,
      "num_tokens": 162253195.0,
      "step": 80075
    },
    {
      "entropy": 5.852305889129639,
      "epoch": 6.99816481691864,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001443415530903771,
      "loss": 5.1638,
      "mean_token_accuracy": 0.2233758494257927,
      "num_tokens": 162264307.0,
      "step": 80080
    },
    {
      "entropy": 5.826679468154907,
      "epoch": 6.998601765271345,
      "grad_norm": 1.25,
      "learning_rate": 0.00014431618563790071,
      "loss": 5.1937,
      "mean_token_accuracy": 0.2200187012553215,
      "num_tokens": 162273620.0,
      "step": 80085
    },
    {
      "entropy": 5.854904460906982,
      "epoch": 6.999038713624049,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014429082069200832,
      "loss": 5.1331,
      "mean_token_accuracy": 0.22436851710081102,
      "num_tokens": 162283233.0,
      "step": 80090
    },
    {
      "entropy": 5.851778841018676,
      "epoch": 6.999475661976755,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014426545825318648,
      "loss": 5.0818,
      "mean_token_accuracy": 0.23470918089151382,
      "num_tokens": 162294321.0,
      "step": 80095
    },
    {
      "entropy": 5.850057744979859,
      "epoch": 6.999912610329459,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014424009832192166,
      "loss": 5.1058,
      "mean_token_accuracy": 0.2278851717710495,
      "num_tokens": 162304102.0,
      "step": 80100
    },
    {
      "entropy": 5.848377370834351,
      "epoch": 7.000349558682164,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014421474089870029,
      "loss": 5.1099,
      "mean_token_accuracy": 0.22896141111850737,
      "num_tokens": 162313886.0,
      "step": 80105
    },
    {
      "entropy": 5.859063529968262,
      "epoch": 7.000786507034868,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014418938598400872,
      "loss": 5.1575,
      "mean_token_accuracy": 0.22622390687465668,
      "num_tokens": 162324581.0,
      "step": 80110
    },
    {
      "entropy": 5.777341985702515,
      "epoch": 7.001223455387573,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00014416403357833332,
      "loss": 4.9547,
      "mean_token_accuracy": 0.23981003761291503,
      "num_tokens": 162334260.0,
      "step": 80115
    },
    {
      "entropy": 5.796700525283813,
      "epoch": 7.001660403740278,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014413868368216026,
      "loss": 5.045,
      "mean_token_accuracy": 0.23593732863664627,
      "num_tokens": 162344065.0,
      "step": 80120
    },
    {
      "entropy": 5.726021099090576,
      "epoch": 7.002097352092982,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001441133362959758,
      "loss": 5.0256,
      "mean_token_accuracy": 0.24384588450193406,
      "num_tokens": 162355313.0,
      "step": 80125
    },
    {
      "entropy": 5.876555871963501,
      "epoch": 7.002534300445688,
      "grad_norm": 1.125,
      "learning_rate": 0.00014408799142026613,
      "loss": 5.1343,
      "mean_token_accuracy": 0.2290552318096161,
      "num_tokens": 162366678.0,
      "step": 80130
    },
    {
      "entropy": 5.799822473526001,
      "epoch": 7.002971248798392,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014406264905551737,
      "loss": 5.0082,
      "mean_token_accuracy": 0.2379877507686615,
      "num_tokens": 162375851.0,
      "step": 80135
    },
    {
      "entropy": 5.843174266815185,
      "epoch": 7.003408197151097,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014403730920221563,
      "loss": 5.0443,
      "mean_token_accuracy": 0.23409980088472365,
      "num_tokens": 162386472.0,
      "step": 80140
    },
    {
      "entropy": 5.947749233245849,
      "epoch": 7.003845145503801,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00014401197186084694,
      "loss": 5.1888,
      "mean_token_accuracy": 0.2240099087357521,
      "num_tokens": 162396918.0,
      "step": 80145
    },
    {
      "entropy": 5.857141304016113,
      "epoch": 7.0042820938565065,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00014398663703189723,
      "loss": 5.1142,
      "mean_token_accuracy": 0.2320183128118515,
      "num_tokens": 162408024.0,
      "step": 80150
    },
    {
      "entropy": 5.662166929244995,
      "epoch": 7.004719042209211,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001439613047158525,
      "loss": 4.8947,
      "mean_token_accuracy": 0.2556249722838402,
      "num_tokens": 162418603.0,
      "step": 80155
    },
    {
      "entropy": 5.832910966873169,
      "epoch": 7.005155990561915,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001439359749131986,
      "loss": 5.0808,
      "mean_token_accuracy": 0.2313789501786232,
      "num_tokens": 162427728.0,
      "step": 80160
    },
    {
      "entropy": 5.881480979919433,
      "epoch": 7.00559293891462,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00014391064762442145,
      "loss": 5.156,
      "mean_token_accuracy": 0.22813832759857178,
      "num_tokens": 162438715.0,
      "step": 80165
    },
    {
      "entropy": 5.853638505935669,
      "epoch": 7.006029887267325,
      "grad_norm": 1.140625,
      "learning_rate": 0.00014388532285000672,
      "loss": 5.0552,
      "mean_token_accuracy": 0.22456836849451065,
      "num_tokens": 162448627.0,
      "step": 80170
    },
    {
      "entropy": 5.81797194480896,
      "epoch": 7.00646683562003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00014386000059044018,
      "loss": 5.0902,
      "mean_token_accuracy": 0.23177177757024764,
      "num_tokens": 162459419.0,
      "step": 80175
    },
    {
      "entropy": 5.834567308425903,
      "epoch": 7.006903783972734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014383468084620753,
      "loss": 5.088,
      "mean_token_accuracy": 0.23016847521066666,
      "num_tokens": 162469395.0,
      "step": 80180
    },
    {
      "entropy": 5.868826580047608,
      "epoch": 7.0073407323254395,
      "grad_norm": 1.25,
      "learning_rate": 0.00014380936361779446,
      "loss": 5.1783,
      "mean_token_accuracy": 0.21714599877595903,
      "num_tokens": 162478203.0,
      "step": 80185
    },
    {
      "entropy": 5.858123064041138,
      "epoch": 7.007777680678144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001437840489056865,
      "loss": 5.0065,
      "mean_token_accuracy": 0.230560964345932,
      "num_tokens": 162488113.0,
      "step": 80190
    },
    {
      "entropy": 5.783385944366455,
      "epoch": 7.0082146290308485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00014375873671036926,
      "loss": 5.0506,
      "mean_token_accuracy": 0.24375539571046828,
      "num_tokens": 162498098.0,
      "step": 80195
    },
    {
      "entropy": 5.8269023418426515,
      "epoch": 7.008651577383553,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014373342703232823,
      "loss": 5.0699,
      "mean_token_accuracy": 0.23320723026990892,
      "num_tokens": 162508051.0,
      "step": 80200
    },
    {
      "entropy": 5.818337869644165,
      "epoch": 7.009088525736258,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001437081198720488,
      "loss": 5.1011,
      "mean_token_accuracy": 0.2283061996102333,
      "num_tokens": 162517759.0,
      "step": 80205
    },
    {
      "entropy": 5.739656543731689,
      "epoch": 7.009525474088963,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001436828152300165,
      "loss": 5.0201,
      "mean_token_accuracy": 0.24584810733795165,
      "num_tokens": 162527828.0,
      "step": 80210
    },
    {
      "entropy": 5.799351978302002,
      "epoch": 7.009962422441667,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014365751310671664,
      "loss": 5.0026,
      "mean_token_accuracy": 0.23954788148403167,
      "num_tokens": 162538232.0,
      "step": 80215
    },
    {
      "entropy": 5.855374479293824,
      "epoch": 7.010399370794372,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014363221350263445,
      "loss": 5.1504,
      "mean_token_accuracy": 0.2255816087126732,
      "num_tokens": 162548922.0,
      "step": 80220
    },
    {
      "entropy": 5.859834480285644,
      "epoch": 7.010836319147077,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014360691641825525,
      "loss": 5.1123,
      "mean_token_accuracy": 0.22214071005582808,
      "num_tokens": 162559702.0,
      "step": 80225
    },
    {
      "entropy": 5.80768780708313,
      "epoch": 7.0112732674997815,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001435816218540642,
      "loss": 5.0306,
      "mean_token_accuracy": 0.23344553858041764,
      "num_tokens": 162568807.0,
      "step": 80230
    },
    {
      "entropy": 5.783126163482666,
      "epoch": 7.011710215852486,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014355632981054653,
      "loss": 4.9678,
      "mean_token_accuracy": 0.24303219765424727,
      "num_tokens": 162578224.0,
      "step": 80235
    },
    {
      "entropy": 5.798163270950317,
      "epoch": 7.012147164205191,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001435310402881873,
      "loss": 5.0795,
      "mean_token_accuracy": 0.2345705136656761,
      "num_tokens": 162589270.0,
      "step": 80240
    },
    {
      "entropy": 5.801299619674682,
      "epoch": 7.012584112557896,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014350575328747167,
      "loss": 5.072,
      "mean_token_accuracy": 0.23191902041435242,
      "num_tokens": 162599386.0,
      "step": 80245
    },
    {
      "entropy": 5.8738490581512455,
      "epoch": 7.0130210609106,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014348046880888454,
      "loss": 5.1472,
      "mean_token_accuracy": 0.22593959271907807,
      "num_tokens": 162609617.0,
      "step": 80250
    },
    {
      "entropy": 5.821513891220093,
      "epoch": 7.013458009263305,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014345518685291098,
      "loss": 5.008,
      "mean_token_accuracy": 0.227957284450531,
      "num_tokens": 162618922.0,
      "step": 80255
    },
    {
      "entropy": 5.799848031997681,
      "epoch": 7.01389495761601,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014342990742003593,
      "loss": 5.0717,
      "mean_token_accuracy": 0.23358314335346222,
      "num_tokens": 162628873.0,
      "step": 80260
    },
    {
      "entropy": 5.865553998947144,
      "epoch": 7.014331905968715,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014340463051074406,
      "loss": 5.1481,
      "mean_token_accuracy": 0.22470786422491074,
      "num_tokens": 162639274.0,
      "step": 80265
    },
    {
      "entropy": 5.888740062713623,
      "epoch": 7.014768854321419,
      "grad_norm": 1.125,
      "learning_rate": 0.0001433793561255204,
      "loss": 5.139,
      "mean_token_accuracy": 0.23140419721603395,
      "num_tokens": 162649164.0,
      "step": 80270
    },
    {
      "entropy": 5.75354232788086,
      "epoch": 7.0152058026741235,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014335408426484962,
      "loss": 4.9694,
      "mean_token_accuracy": 0.23715795427560807,
      "num_tokens": 162658749.0,
      "step": 80275
    },
    {
      "entropy": 5.780893135070801,
      "epoch": 7.015642751026829,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014332881492921652,
      "loss": 4.966,
      "mean_token_accuracy": 0.24047007411718369,
      "num_tokens": 162669460.0,
      "step": 80280
    },
    {
      "entropy": 5.755866003036499,
      "epoch": 7.016079699379533,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001433035481191057,
      "loss": 5.0502,
      "mean_token_accuracy": 0.2369622439146042,
      "num_tokens": 162679524.0,
      "step": 80285
    },
    {
      "entropy": 5.846691703796386,
      "epoch": 7.016516647732238,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014327828383500185,
      "loss": 5.0569,
      "mean_token_accuracy": 0.23651048839092254,
      "num_tokens": 162689074.0,
      "step": 80290
    },
    {
      "entropy": 5.874466276168823,
      "epoch": 7.016953596084943,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014325302207738954,
      "loss": 5.0827,
      "mean_token_accuracy": 0.22835660725831985,
      "num_tokens": 162699309.0,
      "step": 80295
    },
    {
      "entropy": 5.831011009216309,
      "epoch": 7.017390544437648,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014322776284675333,
      "loss": 5.1157,
      "mean_token_accuracy": 0.22854194194078445,
      "num_tokens": 162709683.0,
      "step": 80300
    },
    {
      "entropy": 5.890192031860352,
      "epoch": 7.017827492790352,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014320250614357764,
      "loss": 5.1989,
      "mean_token_accuracy": 0.2193372830748558,
      "num_tokens": 162720445.0,
      "step": 80305
    },
    {
      "entropy": 5.801942920684814,
      "epoch": 7.018264441143057,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000143177251968347,
      "loss": 5.0211,
      "mean_token_accuracy": 0.24557959288358688,
      "num_tokens": 162729990.0,
      "step": 80310
    },
    {
      "entropy": 5.847371625900268,
      "epoch": 7.018701389495762,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014315200032154573,
      "loss": 5.1039,
      "mean_token_accuracy": 0.23184635639190673,
      "num_tokens": 162739616.0,
      "step": 80315
    },
    {
      "entropy": 5.834288644790649,
      "epoch": 7.019138337848466,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014312675120365815,
      "loss": 5.0759,
      "mean_token_accuracy": 0.22700752317905426,
      "num_tokens": 162749569.0,
      "step": 80320
    },
    {
      "entropy": 5.812982845306396,
      "epoch": 7.019575286201171,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014310150461516858,
      "loss": 5.1024,
      "mean_token_accuracy": 0.2310582160949707,
      "num_tokens": 162760272.0,
      "step": 80325
    },
    {
      "entropy": 5.78361873626709,
      "epoch": 7.020012234553875,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00014307626055656125,
      "loss": 5.0215,
      "mean_token_accuracy": 0.2388245075941086,
      "num_tokens": 162770248.0,
      "step": 80330
    },
    {
      "entropy": 5.83805513381958,
      "epoch": 7.020449182906581,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014305101902832042,
      "loss": 5.1351,
      "mean_token_accuracy": 0.22571347653865814,
      "num_tokens": 162779825.0,
      "step": 80335
    },
    {
      "entropy": 5.783840179443359,
      "epoch": 7.020886131259285,
      "grad_norm": 1.25,
      "learning_rate": 0.0001430257800309301,
      "loss": 5.036,
      "mean_token_accuracy": 0.23664974123239518,
      "num_tokens": 162790541.0,
      "step": 80340
    },
    {
      "entropy": 5.853013753890991,
      "epoch": 7.02132307961199,
      "grad_norm": 1.0625,
      "learning_rate": 0.00014300054356487454,
      "loss": 5.1645,
      "mean_token_accuracy": 0.22745253443717955,
      "num_tokens": 162801666.0,
      "step": 80345
    },
    {
      "entropy": 5.88069715499878,
      "epoch": 7.021760027964695,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014297530963063772,
      "loss": 5.1334,
      "mean_token_accuracy": 0.22692557275295258,
      "num_tokens": 162812019.0,
      "step": 80350
    },
    {
      "entropy": 5.788571834564209,
      "epoch": 7.0221969763173995,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001429500782287036,
      "loss": 5.0786,
      "mean_token_accuracy": 0.2401354730129242,
      "num_tokens": 162822766.0,
      "step": 80355
    },
    {
      "entropy": 5.900389194488525,
      "epoch": 7.022633924670104,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014292484935955627,
      "loss": 5.1599,
      "mean_token_accuracy": 0.2244170844554901,
      "num_tokens": 162832505.0,
      "step": 80360
    },
    {
      "entropy": 5.8466448307037355,
      "epoch": 7.023070873022808,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014289962302367948,
      "loss": 5.0738,
      "mean_token_accuracy": 0.2302054286003113,
      "num_tokens": 162842871.0,
      "step": 80365
    },
    {
      "entropy": 5.786225843429565,
      "epoch": 7.023507821375514,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014287439922155713,
      "loss": 5.0286,
      "mean_token_accuracy": 0.23687347620725632,
      "num_tokens": 162852512.0,
      "step": 80370
    },
    {
      "entropy": 5.809668302536011,
      "epoch": 7.023944769728218,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00014284917795367302,
      "loss": 5.1346,
      "mean_token_accuracy": 0.22694860249757767,
      "num_tokens": 162863222.0,
      "step": 80375
    },
    {
      "entropy": 5.8731162548065186,
      "epoch": 7.024381718080923,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014282395922051092,
      "loss": 5.1325,
      "mean_token_accuracy": 0.22670696675777435,
      "num_tokens": 162873116.0,
      "step": 80380
    },
    {
      "entropy": 5.860728788375854,
      "epoch": 7.024818666433627,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014279874302255453,
      "loss": 5.1033,
      "mean_token_accuracy": 0.2239076778292656,
      "num_tokens": 162883608.0,
      "step": 80385
    },
    {
      "entropy": 5.895155000686645,
      "epoch": 7.0252556147863325,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001427735293602876,
      "loss": 5.0832,
      "mean_token_accuracy": 0.2312702566385269,
      "num_tokens": 162894145.0,
      "step": 80390
    },
    {
      "entropy": 5.791244602203369,
      "epoch": 7.025692563139037,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001427483182341936,
      "loss": 5.0632,
      "mean_token_accuracy": 0.2408476024866104,
      "num_tokens": 162905609.0,
      "step": 80395
    },
    {
      "entropy": 5.811839199066162,
      "epoch": 7.0261295114917415,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00014272310964475615,
      "loss": 5.0395,
      "mean_token_accuracy": 0.2398741975426674,
      "num_tokens": 162916105.0,
      "step": 80400
    },
    {
      "entropy": 5.846279811859131,
      "epoch": 7.026566459844446,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001426979035924588,
      "loss": 5.1204,
      "mean_token_accuracy": 0.23057751804590226,
      "num_tokens": 162926841.0,
      "step": 80405
    },
    {
      "entropy": 5.7883916854858395,
      "epoch": 7.027003408197151,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000142672700077785,
      "loss": 4.984,
      "mean_token_accuracy": 0.24023417830467225,
      "num_tokens": 162936527.0,
      "step": 80410
    },
    {
      "entropy": 5.7678180694580075,
      "epoch": 7.027440356549856,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001426474991012181,
      "loss": 5.0459,
      "mean_token_accuracy": 0.2377466604113579,
      "num_tokens": 162947007.0,
      "step": 80415
    },
    {
      "entropy": 5.885789775848389,
      "epoch": 7.02787730490256,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014262230066324155,
      "loss": 5.2109,
      "mean_token_accuracy": 0.22387794256210328,
      "num_tokens": 162958232.0,
      "step": 80420
    },
    {
      "entropy": 5.8475100040435795,
      "epoch": 7.028314253255266,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014259710476433862,
      "loss": 5.1031,
      "mean_token_accuracy": 0.22849606722593307,
      "num_tokens": 162967751.0,
      "step": 80425
    },
    {
      "entropy": 5.825930881500244,
      "epoch": 7.02875120160797,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001425719114049926,
      "loss": 5.0656,
      "mean_token_accuracy": 0.23117169439792634,
      "num_tokens": 162978405.0,
      "step": 80430
    },
    {
      "entropy": 5.825630521774292,
      "epoch": 7.0291881499606745,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014254672058568662,
      "loss": 5.0762,
      "mean_token_accuracy": 0.2313431367278099,
      "num_tokens": 162987720.0,
      "step": 80435
    },
    {
      "entropy": 5.802772045135498,
      "epoch": 7.029625098313379,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014252153230690407,
      "loss": 5.0746,
      "mean_token_accuracy": 0.2339117631316185,
      "num_tokens": 162997459.0,
      "step": 80440
    },
    {
      "entropy": 5.765105104446411,
      "epoch": 7.030062046666084,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014249634656912795,
      "loss": 5.0971,
      "mean_token_accuracy": 0.23253195434808732,
      "num_tokens": 163008247.0,
      "step": 80445
    },
    {
      "entropy": 5.878671789169312,
      "epoch": 7.030498995018789,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014247116337284132,
      "loss": 5.1065,
      "mean_token_accuracy": 0.22412120103836058,
      "num_tokens": 163018346.0,
      "step": 80450
    },
    {
      "entropy": 5.859527730941773,
      "epoch": 7.030935943371493,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001424459827185272,
      "loss": 5.1143,
      "mean_token_accuracy": 0.2274612158536911,
      "num_tokens": 163028958.0,
      "step": 80455
    },
    {
      "entropy": 5.773471260070801,
      "epoch": 7.031372891724198,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014242080460666857,
      "loss": 5.0446,
      "mean_token_accuracy": 0.23678487688302993,
      "num_tokens": 163039914.0,
      "step": 80460
    },
    {
      "entropy": 5.835818243026734,
      "epoch": 7.031809840076903,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001423956290377484,
      "loss": 5.1615,
      "mean_token_accuracy": 0.22912153601646423,
      "num_tokens": 163050177.0,
      "step": 80465
    },
    {
      "entropy": 5.790298128128052,
      "epoch": 7.032246788429608,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014237045601224953,
      "loss": 4.9611,
      "mean_token_accuracy": 0.2438881665468216,
      "num_tokens": 163060206.0,
      "step": 80470
    },
    {
      "entropy": 5.75718240737915,
      "epoch": 7.032683736782312,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014234528553065483,
      "loss": 5.0249,
      "mean_token_accuracy": 0.23799517303705214,
      "num_tokens": 163069348.0,
      "step": 80475
    },
    {
      "entropy": 5.8289384841918945,
      "epoch": 7.033120685135017,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014232011759344703,
      "loss": 5.1132,
      "mean_token_accuracy": 0.2335625872015953,
      "num_tokens": 163080102.0,
      "step": 80480
    },
    {
      "entropy": 5.786459445953369,
      "epoch": 7.033557633487722,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014229495220110895,
      "loss": 5.0549,
      "mean_token_accuracy": 0.24051871597766877,
      "num_tokens": 163090088.0,
      "step": 80485
    },
    {
      "entropy": 5.846843814849853,
      "epoch": 7.033994581840426,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001422697893541232,
      "loss": 5.0465,
      "mean_token_accuracy": 0.2363039508461952,
      "num_tokens": 163100102.0,
      "step": 80490
    },
    {
      "entropy": 5.832404518127442,
      "epoch": 7.034431530193131,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014224462905297253,
      "loss": 5.053,
      "mean_token_accuracy": 0.2289537027478218,
      "num_tokens": 163109928.0,
      "step": 80495
    },
    {
      "entropy": 5.880283308029175,
      "epoch": 7.034868478545836,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014221947129813934,
      "loss": 5.1546,
      "mean_token_accuracy": 0.2240290716290474,
      "num_tokens": 163119656.0,
      "step": 80500
    },
    {
      "entropy": 5.832079458236694,
      "epoch": 7.035305426898541,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014219431609010624,
      "loss": 5.0956,
      "mean_token_accuracy": 0.2323880597949028,
      "num_tokens": 163130198.0,
      "step": 80505
    },
    {
      "entropy": 5.834211826324463,
      "epoch": 7.035742375251245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014216916342935576,
      "loss": 5.1676,
      "mean_token_accuracy": 0.23127008378505706,
      "num_tokens": 163140374.0,
      "step": 80510
    },
    {
      "entropy": 5.839987421035767,
      "epoch": 7.03617932360395,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014214401331637039,
      "loss": 5.1263,
      "mean_token_accuracy": 0.2309922307729721,
      "num_tokens": 163150134.0,
      "step": 80515
    },
    {
      "entropy": 5.830839967727661,
      "epoch": 7.036616271956655,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001421188657516324,
      "loss": 5.0912,
      "mean_token_accuracy": 0.23455612361431122,
      "num_tokens": 163160162.0,
      "step": 80520
    },
    {
      "entropy": 5.8538905620574955,
      "epoch": 7.037053220309359,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014209372073562421,
      "loss": 5.0456,
      "mean_token_accuracy": 0.23574183285236358,
      "num_tokens": 163170288.0,
      "step": 80525
    },
    {
      "entropy": 5.833933782577515,
      "epoch": 7.037490168662064,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014206857826882808,
      "loss": 5.0897,
      "mean_token_accuracy": 0.23019790649414062,
      "num_tokens": 163180527.0,
      "step": 80530
    },
    {
      "entropy": 5.87375693321228,
      "epoch": 7.037927117014769,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001420434383517263,
      "loss": 5.1485,
      "mean_token_accuracy": 0.225412617623806,
      "num_tokens": 163190446.0,
      "step": 80535
    },
    {
      "entropy": 5.834688329696656,
      "epoch": 7.038364065367474,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014201830098480112,
      "loss": 5.1636,
      "mean_token_accuracy": 0.22777749449014664,
      "num_tokens": 163200339.0,
      "step": 80540
    },
    {
      "entropy": 5.804882621765136,
      "epoch": 7.038801013720178,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00014199316616853453,
      "loss": 5.0564,
      "mean_token_accuracy": 0.22904151529073716,
      "num_tokens": 163211010.0,
      "step": 80545
    },
    {
      "entropy": 5.8139232158660885,
      "epoch": 7.039237962072883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00014196803390340867,
      "loss": 5.0826,
      "mean_token_accuracy": 0.22313954085111617,
      "num_tokens": 163220655.0,
      "step": 80550
    },
    {
      "entropy": 5.838574028015136,
      "epoch": 7.039674910425588,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001419429041899057,
      "loss": 5.1613,
      "mean_token_accuracy": 0.22669770568609238,
      "num_tokens": 163231483.0,
      "step": 80555
    },
    {
      "entropy": 5.819091796875,
      "epoch": 7.0401118587782925,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014191777702850749,
      "loss": 5.0359,
      "mean_token_accuracy": 0.236868816614151,
      "num_tokens": 163241477.0,
      "step": 80560
    },
    {
      "entropy": 5.867238569259643,
      "epoch": 7.040548807130997,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014189265241969603,
      "loss": 5.126,
      "mean_token_accuracy": 0.22856459617614747,
      "num_tokens": 163250589.0,
      "step": 80565
    },
    {
      "entropy": 5.8517759323120115,
      "epoch": 7.040985755483701,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014186753036395328,
      "loss": 5.1821,
      "mean_token_accuracy": 0.21509372740983962,
      "num_tokens": 163260491.0,
      "step": 80570
    },
    {
      "entropy": 5.814687252044678,
      "epoch": 7.041422703836407,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014184241086176105,
      "loss": 5.1005,
      "mean_token_accuracy": 0.22524023652076722,
      "num_tokens": 163271534.0,
      "step": 80575
    },
    {
      "entropy": 5.893459272384644,
      "epoch": 7.041859652189111,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014181729391360114,
      "loss": 5.1954,
      "mean_token_accuracy": 0.22117750942707062,
      "num_tokens": 163282793.0,
      "step": 80580
    },
    {
      "entropy": 5.868964529037475,
      "epoch": 7.042296600541816,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014179217951995537,
      "loss": 5.0935,
      "mean_token_accuracy": 0.2290603771805763,
      "num_tokens": 163292004.0,
      "step": 80585
    },
    {
      "entropy": 5.911292982101441,
      "epoch": 7.042733548894521,
      "grad_norm": 1.25,
      "learning_rate": 0.00014176706768130544,
      "loss": 5.1682,
      "mean_token_accuracy": 0.21920136213302613,
      "num_tokens": 163301079.0,
      "step": 80590
    },
    {
      "entropy": 5.8607481002807615,
      "epoch": 7.0431704972472255,
      "grad_norm": 1.25,
      "learning_rate": 0.00014174195839813285,
      "loss": 5.1303,
      "mean_token_accuracy": 0.23519420623779297,
      "num_tokens": 163312368.0,
      "step": 80595
    },
    {
      "entropy": 5.8304461479187015,
      "epoch": 7.04360744559993,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001417168516709194,
      "loss": 5.1686,
      "mean_token_accuracy": 0.22201151698827742,
      "num_tokens": 163322801.0,
      "step": 80600
    },
    {
      "entropy": 5.803317213058472,
      "epoch": 7.0440443939526345,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014169174750014652,
      "loss": 5.0479,
      "mean_token_accuracy": 0.2405689388513565,
      "num_tokens": 163332694.0,
      "step": 80605
    },
    {
      "entropy": 5.803645086288452,
      "epoch": 7.04448134230534,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001416666458862958,
      "loss": 5.0195,
      "mean_token_accuracy": 0.23353009074926376,
      "num_tokens": 163343022.0,
      "step": 80610
    },
    {
      "entropy": 5.841997575759888,
      "epoch": 7.044918290658044,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014164154682984868,
      "loss": 5.0977,
      "mean_token_accuracy": 0.2304523542523384,
      "num_tokens": 163352316.0,
      "step": 80615
    },
    {
      "entropy": 5.848876047134399,
      "epoch": 7.045355239010749,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014161645033128656,
      "loss": 5.1131,
      "mean_token_accuracy": 0.2304104596376419,
      "num_tokens": 163363993.0,
      "step": 80620
    },
    {
      "entropy": 5.805126428604126,
      "epoch": 7.045792187363453,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001415913563910908,
      "loss": 5.0495,
      "mean_token_accuracy": 0.2372022747993469,
      "num_tokens": 163373561.0,
      "step": 80625
    },
    {
      "entropy": 5.745334148406982,
      "epoch": 7.046229135716159,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014156626500974275,
      "loss": 5.0123,
      "mean_token_accuracy": 0.24666734784841537,
      "num_tokens": 163384341.0,
      "step": 80630
    },
    {
      "entropy": 5.91562852859497,
      "epoch": 7.046666084068863,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014154117618772366,
      "loss": 5.2575,
      "mean_token_accuracy": 0.22463733255863189,
      "num_tokens": 163394956.0,
      "step": 80635
    },
    {
      "entropy": 5.841781377792358,
      "epoch": 7.0471030324215675,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001415160899255148,
      "loss": 5.0791,
      "mean_token_accuracy": 0.2318245843052864,
      "num_tokens": 163405846.0,
      "step": 80640
    },
    {
      "entropy": 5.851828384399414,
      "epoch": 7.047539980774273,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001414910062235972,
      "loss": 5.0915,
      "mean_token_accuracy": 0.2358457401394844,
      "num_tokens": 163415445.0,
      "step": 80645
    },
    {
      "entropy": 5.816616535186768,
      "epoch": 7.047976929126977,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001414659250824521,
      "loss": 5.0584,
      "mean_token_accuracy": 0.2354453220963478,
      "num_tokens": 163424966.0,
      "step": 80650
    },
    {
      "entropy": 5.829914951324463,
      "epoch": 7.048413877479682,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014144084650256049,
      "loss": 5.1676,
      "mean_token_accuracy": 0.22133031338453293,
      "num_tokens": 163434588.0,
      "step": 80655
    },
    {
      "entropy": 5.760605001449585,
      "epoch": 7.048850825832386,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001414157704844034,
      "loss": 5.061,
      "mean_token_accuracy": 0.23388022035360337,
      "num_tokens": 163445140.0,
      "step": 80660
    },
    {
      "entropy": 5.869219017028809,
      "epoch": 7.049287774185092,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014139069702846186,
      "loss": 5.1656,
      "mean_token_accuracy": 0.2304919630289078,
      "num_tokens": 163456027.0,
      "step": 80665
    },
    {
      "entropy": 5.894132947921753,
      "epoch": 7.049724722537796,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001413656261352168,
      "loss": 5.142,
      "mean_token_accuracy": 0.2250139072537422,
      "num_tokens": 163467179.0,
      "step": 80670
    },
    {
      "entropy": 5.849674224853516,
      "epoch": 7.050161670890501,
      "grad_norm": 1.234375,
      "learning_rate": 0.000141340557805149,
      "loss": 5.0513,
      "mean_token_accuracy": 0.23341040462255477,
      "num_tokens": 163476407.0,
      "step": 80675
    },
    {
      "entropy": 5.788924598693848,
      "epoch": 7.050598619243205,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014131549203873935,
      "loss": 5.0586,
      "mean_token_accuracy": 0.24327961802482606,
      "num_tokens": 163486564.0,
      "step": 80680
    },
    {
      "entropy": 5.805481719970703,
      "epoch": 7.05103556759591,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001412904288364687,
      "loss": 5.0894,
      "mean_token_accuracy": 0.2334360510110855,
      "num_tokens": 163497071.0,
      "step": 80685
    },
    {
      "entropy": 5.834565591812134,
      "epoch": 7.051472515948615,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014126536819881764,
      "loss": 5.1427,
      "mean_token_accuracy": 0.22676047086715698,
      "num_tokens": 163506867.0,
      "step": 80690
    },
    {
      "entropy": 5.794651126861572,
      "epoch": 7.051909464301319,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014124031012626684,
      "loss": 5.0232,
      "mean_token_accuracy": 0.23664777874946594,
      "num_tokens": 163517923.0,
      "step": 80695
    },
    {
      "entropy": 5.8682177543640135,
      "epoch": 7.052346412654025,
      "grad_norm": 1.21875,
      "learning_rate": 0.000141215254619297,
      "loss": 5.1289,
      "mean_token_accuracy": 0.2237480565905571,
      "num_tokens": 163529240.0,
      "step": 80700
    },
    {
      "entropy": 5.846333360671997,
      "epoch": 7.052783361006729,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001411902016783887,
      "loss": 5.0678,
      "mean_token_accuracy": 0.23159259408712388,
      "num_tokens": 163539008.0,
      "step": 80705
    },
    {
      "entropy": 5.803796720504761,
      "epoch": 7.053220309359434,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001411651513040224,
      "loss": 5.032,
      "mean_token_accuracy": 0.23295769542455674,
      "num_tokens": 163549062.0,
      "step": 80710
    },
    {
      "entropy": 5.783708286285401,
      "epoch": 7.053657257712138,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014114010349667867,
      "loss": 5.0609,
      "mean_token_accuracy": 0.2379644811153412,
      "num_tokens": 163559116.0,
      "step": 80715
    },
    {
      "entropy": 5.74832010269165,
      "epoch": 7.0540942060648435,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014111505825683786,
      "loss": 5.0682,
      "mean_token_accuracy": 0.237050798535347,
      "num_tokens": 163569366.0,
      "step": 80720
    },
    {
      "entropy": 5.8272850036621096,
      "epoch": 7.054531154417548,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001410900155849804,
      "loss": 5.1451,
      "mean_token_accuracy": 0.22753585875034332,
      "num_tokens": 163578932.0,
      "step": 80725
    },
    {
      "entropy": 5.8588660717010494,
      "epoch": 7.054968102770252,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001410649754815866,
      "loss": 5.1549,
      "mean_token_accuracy": 0.2205544725060463,
      "num_tokens": 163588891.0,
      "step": 80730
    },
    {
      "entropy": 5.860602235794067,
      "epoch": 7.055405051122957,
      "grad_norm": 1.140625,
      "learning_rate": 0.00014103993794713687,
      "loss": 5.0186,
      "mean_token_accuracy": 0.2408956065773964,
      "num_tokens": 163599214.0,
      "step": 80735
    },
    {
      "entropy": 5.8410242080688475,
      "epoch": 7.055841999475662,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014101490298211122,
      "loss": 5.1028,
      "mean_token_accuracy": 0.23435938507318496,
      "num_tokens": 163609208.0,
      "step": 80740
    },
    {
      "entropy": 5.768972253799438,
      "epoch": 7.056278947828367,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014098987058698995,
      "loss": 5.0229,
      "mean_token_accuracy": 0.24090584814548494,
      "num_tokens": 163620457.0,
      "step": 80745
    },
    {
      "entropy": 5.838639879226685,
      "epoch": 7.056715896181071,
      "grad_norm": 1.25,
      "learning_rate": 0.00014096484076225318,
      "loss": 5.0654,
      "mean_token_accuracy": 0.23892679512500764,
      "num_tokens": 163630665.0,
      "step": 80750
    },
    {
      "entropy": 5.8517454147338865,
      "epoch": 7.057152844533776,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014093981350838097,
      "loss": 5.0987,
      "mean_token_accuracy": 0.2271188884973526,
      "num_tokens": 163640259.0,
      "step": 80755
    },
    {
      "entropy": 5.928820848464966,
      "epoch": 7.057589792886481,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001409147888258534,
      "loss": 5.265,
      "mean_token_accuracy": 0.2154049977660179,
      "num_tokens": 163649633.0,
      "step": 80760
    },
    {
      "entropy": 5.886581182479858,
      "epoch": 7.0580267412391855,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014088976671515043,
      "loss": 5.1386,
      "mean_token_accuracy": 0.22785243690013884,
      "num_tokens": 163658611.0,
      "step": 80765
    },
    {
      "entropy": 5.881818675994873,
      "epoch": 7.05846368959189,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014086474717675204,
      "loss": 5.1411,
      "mean_token_accuracy": 0.23161602169275283,
      "num_tokens": 163666869.0,
      "step": 80770
    },
    {
      "entropy": 5.837649965286255,
      "epoch": 7.058900637944595,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014083973021113805,
      "loss": 5.0822,
      "mean_token_accuracy": 0.23257169425487517,
      "num_tokens": 163676768.0,
      "step": 80775
    },
    {
      "entropy": 5.742254304885864,
      "epoch": 7.0593375862973,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014081471581878834,
      "loss": 4.9892,
      "mean_token_accuracy": 0.234380604326725,
      "num_tokens": 163688059.0,
      "step": 80780
    },
    {
      "entropy": 5.884174633026123,
      "epoch": 7.059774534650004,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00014078970400018272,
      "loss": 5.1784,
      "mean_token_accuracy": 0.21981041133403778,
      "num_tokens": 163698760.0,
      "step": 80785
    },
    {
      "entropy": 5.873879146575928,
      "epoch": 7.060211483002709,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014076469475580089,
      "loss": 5.1409,
      "mean_token_accuracy": 0.22747650891542434,
      "num_tokens": 163709069.0,
      "step": 80790
    },
    {
      "entropy": 5.844957780838013,
      "epoch": 7.060648431355414,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014073968808612253,
      "loss": 5.0709,
      "mean_token_accuracy": 0.22285171449184418,
      "num_tokens": 163718734.0,
      "step": 80795
    },
    {
      "entropy": 5.819167184829712,
      "epoch": 7.061085379708119,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014071468399162734,
      "loss": 5.0899,
      "mean_token_accuracy": 0.23324697017669677,
      "num_tokens": 163729416.0,
      "step": 80800
    },
    {
      "entropy": 5.802820634841919,
      "epoch": 7.061522328060823,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014068968247279485,
      "loss": 5.0228,
      "mean_token_accuracy": 0.23675183653831483,
      "num_tokens": 163739387.0,
      "step": 80805
    },
    {
      "entropy": 5.820799779891968,
      "epoch": 7.0619592764135275,
      "grad_norm": 1.25,
      "learning_rate": 0.00014066468353010462,
      "loss": 5.0606,
      "mean_token_accuracy": 0.23346232771873474,
      "num_tokens": 163749732.0,
      "step": 80810
    },
    {
      "entropy": 5.826435565948486,
      "epoch": 7.062396224766233,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014063968716403614,
      "loss": 5.0287,
      "mean_token_accuracy": 0.22853318154811858,
      "num_tokens": 163759495.0,
      "step": 80815
    },
    {
      "entropy": 5.815970325469971,
      "epoch": 7.062833173118937,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014061469337506898,
      "loss": 5.0556,
      "mean_token_accuracy": 0.23442825078964233,
      "num_tokens": 163768158.0,
      "step": 80820
    },
    {
      "entropy": 5.832485628128052,
      "epoch": 7.063270121471642,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014058970216368233,
      "loss": 5.1293,
      "mean_token_accuracy": 0.23375778198242186,
      "num_tokens": 163778254.0,
      "step": 80825
    },
    {
      "entropy": 5.8557531356811525,
      "epoch": 7.063707069824347,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001405647135303556,
      "loss": 5.0709,
      "mean_token_accuracy": 0.23094459474086762,
      "num_tokens": 163787679.0,
      "step": 80830
    },
    {
      "entropy": 5.851583003997803,
      "epoch": 7.064144018177052,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014053972747556814,
      "loss": 5.1241,
      "mean_token_accuracy": 0.22632937282323837,
      "num_tokens": 163798103.0,
      "step": 80835
    },
    {
      "entropy": 5.75381178855896,
      "epoch": 7.064580966529756,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014051474399979914,
      "loss": 5.058,
      "mean_token_accuracy": 0.23743730336427687,
      "num_tokens": 163808623.0,
      "step": 80840
    },
    {
      "entropy": 5.812882041931152,
      "epoch": 7.065017914882461,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014048976310352784,
      "loss": 5.0921,
      "mean_token_accuracy": 0.22806637585163117,
      "num_tokens": 163818262.0,
      "step": 80845
    },
    {
      "entropy": 5.908855152130127,
      "epoch": 7.065454863235166,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014046478478723336,
      "loss": 5.2198,
      "mean_token_accuracy": 0.22039747089147568,
      "num_tokens": 163828140.0,
      "step": 80850
    },
    {
      "entropy": 5.843816757202148,
      "epoch": 7.06589181158787,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014043980905139474,
      "loss": 5.0901,
      "mean_token_accuracy": 0.2318218931555748,
      "num_tokens": 163838187.0,
      "step": 80855
    },
    {
      "entropy": 5.807402467727661,
      "epoch": 7.066328759940575,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014041483589649119,
      "loss": 5.01,
      "mean_token_accuracy": 0.24159313440322877,
      "num_tokens": 163847162.0,
      "step": 80860
    },
    {
      "entropy": 5.8575146198272705,
      "epoch": 7.066765708293279,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014038986532300158,
      "loss": 5.1354,
      "mean_token_accuracy": 0.2244473621249199,
      "num_tokens": 163856897.0,
      "step": 80865
    },
    {
      "entropy": 5.812414979934692,
      "epoch": 7.067202656645985,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014036489733140495,
      "loss": 5.0536,
      "mean_token_accuracy": 0.2302684411406517,
      "num_tokens": 163866910.0,
      "step": 80870
    },
    {
      "entropy": 5.798531866073608,
      "epoch": 7.067639604998689,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014033993192218004,
      "loss": 5.0658,
      "mean_token_accuracy": 0.23625636249780654,
      "num_tokens": 163877218.0,
      "step": 80875
    },
    {
      "entropy": 5.810126256942749,
      "epoch": 7.068076553351394,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014031496909580583,
      "loss": 5.0369,
      "mean_token_accuracy": 0.23468061834573745,
      "num_tokens": 163887290.0,
      "step": 80880
    },
    {
      "entropy": 5.787840700149536,
      "epoch": 7.068513501704099,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014029000885276105,
      "loss": 5.0491,
      "mean_token_accuracy": 0.24083737879991532,
      "num_tokens": 163897471.0,
      "step": 80885
    },
    {
      "entropy": 5.915237474441528,
      "epoch": 7.0689504500568034,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001402650511935245,
      "loss": 5.2363,
      "mean_token_accuracy": 0.21504207849502563,
      "num_tokens": 163909311.0,
      "step": 80890
    },
    {
      "entropy": 5.8320068359375,
      "epoch": 7.069387398409508,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001402400961185749,
      "loss": 5.1004,
      "mean_token_accuracy": 0.22585427314043044,
      "num_tokens": 163919465.0,
      "step": 80895
    },
    {
      "entropy": 5.818203544616699,
      "epoch": 7.069824346762212,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014021514362839085,
      "loss": 5.0656,
      "mean_token_accuracy": 0.23189124017953872,
      "num_tokens": 163929500.0,
      "step": 80900
    },
    {
      "entropy": 5.762989234924317,
      "epoch": 7.070261295114918,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014019019372345096,
      "loss": 5.0172,
      "mean_token_accuracy": 0.23861415982246398,
      "num_tokens": 163939587.0,
      "step": 80905
    },
    {
      "entropy": 5.814257049560547,
      "epoch": 7.070698243467622,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001401652464042338,
      "loss": 5.0856,
      "mean_token_accuracy": 0.22962268739938735,
      "num_tokens": 163948865.0,
      "step": 80910
    },
    {
      "entropy": 5.818520832061767,
      "epoch": 7.071135191820327,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001401403016712179,
      "loss": 5.0191,
      "mean_token_accuracy": 0.23074205964803696,
      "num_tokens": 163958612.0,
      "step": 80915
    },
    {
      "entropy": 5.920399284362793,
      "epoch": 7.071572140173031,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001401153595248817,
      "loss": 5.1899,
      "mean_token_accuracy": 0.21977992951869965,
      "num_tokens": 163969032.0,
      "step": 80920
    },
    {
      "entropy": 5.88128080368042,
      "epoch": 7.0720090885257365,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014009041996570354,
      "loss": 5.1561,
      "mean_token_accuracy": 0.2244659870862961,
      "num_tokens": 163979269.0,
      "step": 80925
    },
    {
      "entropy": 5.834151649475098,
      "epoch": 7.072446036878441,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001400654829941618,
      "loss": 5.0867,
      "mean_token_accuracy": 0.22938863635063172,
      "num_tokens": 163988997.0,
      "step": 80930
    },
    {
      "entropy": 5.819192934036255,
      "epoch": 7.0728829852311454,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014004054861073482,
      "loss": 5.112,
      "mean_token_accuracy": 0.22913422733545302,
      "num_tokens": 164000138.0,
      "step": 80935
    },
    {
      "entropy": 5.866745233535767,
      "epoch": 7.073319933583851,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001400156168159008,
      "loss": 5.1548,
      "mean_token_accuracy": 0.2240639626979828,
      "num_tokens": 164009849.0,
      "step": 80940
    },
    {
      "entropy": 5.8391846179962155,
      "epoch": 7.073756881936555,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013999068761013804,
      "loss": 5.1373,
      "mean_token_accuracy": 0.23096186220645903,
      "num_tokens": 164019121.0,
      "step": 80945
    },
    {
      "entropy": 5.8738565921783445,
      "epoch": 7.07419383028926,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013996576099392457,
      "loss": 5.1478,
      "mean_token_accuracy": 0.22757847905158995,
      "num_tokens": 164029042.0,
      "step": 80950
    },
    {
      "entropy": 5.796914196014404,
      "epoch": 7.074630778641964,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013994083696773862,
      "loss": 5.0275,
      "mean_token_accuracy": 0.24493755400180817,
      "num_tokens": 164040174.0,
      "step": 80955
    },
    {
      "entropy": 5.871514129638672,
      "epoch": 7.07506772699467,
      "grad_norm": 1.21875,
      "learning_rate": 0.00013991591553205816,
      "loss": 5.1003,
      "mean_token_accuracy": 0.23295211046934128,
      "num_tokens": 164049975.0,
      "step": 80960
    },
    {
      "entropy": 5.833293914794922,
      "epoch": 7.075504675347374,
      "grad_norm": 1.25,
      "learning_rate": 0.00013989099668736121,
      "loss": 5.0804,
      "mean_token_accuracy": 0.23140874952077867,
      "num_tokens": 164059997.0,
      "step": 80965
    },
    {
      "entropy": 5.89892840385437,
      "epoch": 7.0759416237000785,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00013986608043412582,
      "loss": 5.1273,
      "mean_token_accuracy": 0.23069861382246018,
      "num_tokens": 164069741.0,
      "step": 80970
    },
    {
      "entropy": 5.855321836471558,
      "epoch": 7.076378572052783,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013984116677282974,
      "loss": 5.1233,
      "mean_token_accuracy": 0.22609489858150483,
      "num_tokens": 164080146.0,
      "step": 80975
    },
    {
      "entropy": 5.767101669311524,
      "epoch": 7.076815520405488,
      "grad_norm": 1.25,
      "learning_rate": 0.0001398162557039509,
      "loss": 4.9924,
      "mean_token_accuracy": 0.2414760023355484,
      "num_tokens": 164089692.0,
      "step": 80980
    },
    {
      "entropy": 5.81832594871521,
      "epoch": 7.077252468758193,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013979134722796717,
      "loss": 5.1091,
      "mean_token_accuracy": 0.23107582777738572,
      "num_tokens": 164098953.0,
      "step": 80985
    },
    {
      "entropy": 5.8178143978118895,
      "epoch": 7.077689417110897,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013976644134535617,
      "loss": 5.1091,
      "mean_token_accuracy": 0.23415422737598418,
      "num_tokens": 164109297.0,
      "step": 80990
    },
    {
      "entropy": 5.818664836883545,
      "epoch": 7.078126365463603,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00013974153805659573,
      "loss": 5.0962,
      "mean_token_accuracy": 0.2305903196334839,
      "num_tokens": 164119452.0,
      "step": 80995
    },
    {
      "entropy": 5.7697914123535154,
      "epoch": 7.078563313816307,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013971663736216342,
      "loss": 5.0154,
      "mean_token_accuracy": 0.23406571596860887,
      "num_tokens": 164129058.0,
      "step": 81000
    },
    {
      "epoch": 7.078563313816307,
      "eval_entropy": 5.568115032592602,
      "eval_loss": 5.2366437911987305,
      "eval_mean_token_accuracy": 0.22898312222823639,
      "eval_num_tokens": 164129058.0,
      "eval_runtime": 27.8964,
      "eval_samples_per_second": 1326.836,
      "eval_steps_per_second": 165.863,
      "step": 81000
    }
  ],
  "logging_steps": 5,
  "max_steps": 114420,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.45644690330368e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}