rus-cyrl-100mb-after-ppt-Dp-100mb-ckpt500_seed3407/checkpoint-24000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0973520929826095,
  "eval_steps": 3000,
  "global_step": 24000,
  "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
    }
  ],
  "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": 7.281382953024e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}