heb-hebr-10mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407/checkpoint-27000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.8892931671036437,
  "eval_steps": 3000,
  "global_step": 27000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.759765005111694,
      "epoch": 0.0005350741077639254,
      "grad_norm": 24.375,
      "learning_rate": 2e-06,
      "loss": 14.199,
      "mean_token_accuracy": 0.004748707497492432,
      "num_tokens": 10038.0,
      "step": 5
    },
    {
      "entropy": 4.763779735565185,
      "epoch": 0.0010701482155278507,
      "grad_norm": 25.125,
      "learning_rate": 4.5e-06,
      "loss": 14.2064,
      "mean_token_accuracy": 0.0057549867779016495,
      "num_tokens": 21916.0,
      "step": 10
    },
    {
      "entropy": 4.7975164413452145,
      "epoch": 0.0016052223232917758,
      "grad_norm": 29.375,
      "learning_rate": 7e-06,
      "loss": 14.1506,
      "mean_token_accuracy": 0.0034938069467898457,
      "num_tokens": 33434.0,
      "step": 15
    },
    {
      "entropy": 4.968814086914063,
      "epoch": 0.0021402964310557014,
      "grad_norm": 38.75,
      "learning_rate": 9.5e-06,
      "loss": 13.7109,
      "mean_token_accuracy": 0.003357043443247676,
      "num_tokens": 44216.0,
      "step": 20
    },
    {
      "entropy": 5.636525678634643,
      "epoch": 0.0026753705388196263,
      "grad_norm": 51.0,
      "learning_rate": 1.2e-05,
      "loss": 12.7453,
      "mean_token_accuracy": 0.0032943477854132652,
      "num_tokens": 54874.0,
      "step": 25
    },
    {
      "entropy": 8.125221157073975,
      "epoch": 0.0032104446465835517,
      "grad_norm": 17.875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 11.4214,
      "mean_token_accuracy": 0.0011823114822618664,
      "num_tokens": 65085.0,
      "step": 30
    },
    {
      "entropy": 10.552507495880127,
      "epoch": 0.003745518754347477,
      "grad_norm": 4.125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.7363,
      "mean_token_accuracy": 0.0020443544257432223,
      "num_tokens": 75601.0,
      "step": 35
    },
    {
      "entropy": 10.720990562438965,
      "epoch": 0.004280592862111403,
      "grad_norm": 3.328125,
      "learning_rate": 1.95e-05,
      "loss": 10.6406,
      "mean_token_accuracy": 0.013470453396439552,
      "num_tokens": 86193.0,
      "step": 40
    },
    {
      "entropy": 10.704570293426514,
      "epoch": 0.004815666969875327,
      "grad_norm": 2.5625,
      "learning_rate": 2.2e-05,
      "loss": 10.4959,
      "mean_token_accuracy": 0.015227390732616187,
      "num_tokens": 96957.0,
      "step": 45
    },
    {
      "entropy": 10.638626861572266,
      "epoch": 0.005350741077639253,
      "grad_norm": 2.25,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.3582,
      "mean_token_accuracy": 0.01454832088202238,
      "num_tokens": 108408.0,
      "step": 50
    },
    {
      "entropy": 10.683897018432617,
      "epoch": 0.005885815185403178,
      "grad_norm": 1.671875,
      "learning_rate": 2.7e-05,
      "loss": 10.2245,
      "mean_token_accuracy": 0.015560841839760541,
      "num_tokens": 119273.0,
      "step": 55
    },
    {
      "entropy": 10.71973533630371,
      "epoch": 0.006420889293167103,
      "grad_norm": 1.859375,
      "learning_rate": 2.95e-05,
      "loss": 10.1245,
      "mean_token_accuracy": 0.03457072209566832,
      "num_tokens": 130817.0,
      "step": 60
    },
    {
      "entropy": 10.70310287475586,
      "epoch": 0.006955963400931029,
      "grad_norm": 1.640625,
      "learning_rate": 3.2e-05,
      "loss": 10.0512,
      "mean_token_accuracy": 0.04912868961691856,
      "num_tokens": 142599.0,
      "step": 65
    },
    {
      "entropy": 10.679985427856446,
      "epoch": 0.007491037508694954,
      "grad_norm": 1.7734375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.9832,
      "mean_token_accuracy": 0.04829649589955807,
      "num_tokens": 153617.0,
      "step": 70
    },
    {
      "entropy": 10.683323860168457,
      "epoch": 0.00802611161645888,
      "grad_norm": 1.5078125,
      "learning_rate": 3.7e-05,
      "loss": 9.9422,
      "mean_token_accuracy": 0.04904289618134498,
      "num_tokens": 165280.0,
      "step": 75
    },
    {
      "entropy": 10.676106262207032,
      "epoch": 0.008561185724222806,
      "grad_norm": 1.3671875,
      "learning_rate": 3.95e-05,
      "loss": 9.8806,
      "mean_token_accuracy": 0.04988635517656803,
      "num_tokens": 176896.0,
      "step": 80
    },
    {
      "entropy": 10.664983177185059,
      "epoch": 0.00909625983198673,
      "grad_norm": 1.5078125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.8275,
      "mean_token_accuracy": 0.05339982435107231,
      "num_tokens": 188000.0,
      "step": 85
    },
    {
      "entropy": 10.663267135620117,
      "epoch": 0.009631333939750655,
      "grad_norm": 1.609375,
      "learning_rate": 4.45e-05,
      "loss": 9.7684,
      "mean_token_accuracy": 0.051240837574005126,
      "num_tokens": 199360.0,
      "step": 90
    },
    {
      "entropy": 10.648833179473877,
      "epoch": 0.01016640804751458,
      "grad_norm": 1.390625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.716,
      "mean_token_accuracy": 0.04947503618896008,
      "num_tokens": 210214.0,
      "step": 95
    },
    {
      "entropy": 10.61328296661377,
      "epoch": 0.010701482155278505,
      "grad_norm": 1.5234375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.5866,
      "mean_token_accuracy": 0.05695506259799003,
      "num_tokens": 220567.0,
      "step": 100
    },
    {
      "entropy": 10.584807205200196,
      "epoch": 0.011236556263042432,
      "grad_norm": 1.4375,
      "learning_rate": 5.2e-05,
      "loss": 9.4837,
      "mean_token_accuracy": 0.05784616619348526,
      "num_tokens": 230585.0,
      "step": 105
    },
    {
      "entropy": 10.56409158706665,
      "epoch": 0.011771630370806356,
      "grad_norm": 1.46875,
      "learning_rate": 5.45e-05,
      "loss": 9.4355,
      "mean_token_accuracy": 0.06093316040933132,
      "num_tokens": 241996.0,
      "step": 110
    },
    {
      "entropy": 10.524872207641602,
      "epoch": 0.012306704478570282,
      "grad_norm": 1.40625,
      "learning_rate": 5.7e-05,
      "loss": 9.3966,
      "mean_token_accuracy": 0.0568605650216341,
      "num_tokens": 253348.0,
      "step": 115
    },
    {
      "entropy": 10.46424674987793,
      "epoch": 0.012841778586334207,
      "grad_norm": 1.4453125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 9.2988,
      "mean_token_accuracy": 0.05472910478711128,
      "num_tokens": 264239.0,
      "step": 120
    },
    {
      "entropy": 10.440286445617676,
      "epoch": 0.013376852694098133,
      "grad_norm": 1.2109375,
      "learning_rate": 6.2e-05,
      "loss": 9.1996,
      "mean_token_accuracy": 0.05646762736141682,
      "num_tokens": 274522.0,
      "step": 125
    },
    {
      "entropy": 10.376360607147216,
      "epoch": 0.013911926801862057,
      "grad_norm": 1.546875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 9.0874,
      "mean_token_accuracy": 0.056457117572426795,
      "num_tokens": 284492.0,
      "step": 130
    },
    {
      "entropy": 10.292502307891846,
      "epoch": 0.014447000909625984,
      "grad_norm": 1.1796875,
      "learning_rate": 6.7e-05,
      "loss": 9.0249,
      "mean_token_accuracy": 0.057459867745637896,
      "num_tokens": 294796.0,
      "step": 135
    },
    {
      "entropy": 10.18876838684082,
      "epoch": 0.014982075017389908,
      "grad_norm": 0.99609375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.9839,
      "mean_token_accuracy": 0.05458516478538513,
      "num_tokens": 305252.0,
      "step": 140
    },
    {
      "entropy": 10.138203525543213,
      "epoch": 0.015517149125153834,
      "grad_norm": 1.171875,
      "learning_rate": 7.2e-05,
      "loss": 8.863,
      "mean_token_accuracy": 0.05204447396099567,
      "num_tokens": 315505.0,
      "step": 145
    },
    {
      "entropy": 10.065136623382568,
      "epoch": 0.01605222323291776,
      "grad_norm": 1.015625,
      "learning_rate": 7.45e-05,
      "loss": 8.8148,
      "mean_token_accuracy": 0.05895447283983231,
      "num_tokens": 326997.0,
      "step": 150
    },
    {
      "entropy": 9.919741535186768,
      "epoch": 0.016587297340681685,
      "grad_norm": 0.96875,
      "learning_rate": 7.7e-05,
      "loss": 8.7167,
      "mean_token_accuracy": 0.05453297197818756,
      "num_tokens": 338089.0,
      "step": 155
    },
    {
      "entropy": 9.827782726287841,
      "epoch": 0.01712237144844561,
      "grad_norm": 0.89453125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.6645,
      "mean_token_accuracy": 0.054577385634183885,
      "num_tokens": 348536.0,
      "step": 160
    },
    {
      "entropy": 9.709816837310791,
      "epoch": 0.017657445556209534,
      "grad_norm": 0.921875,
      "learning_rate": 8.2e-05,
      "loss": 8.6434,
      "mean_token_accuracy": 0.05497662127017975,
      "num_tokens": 359776.0,
      "step": 165
    },
    {
      "entropy": 9.536125087738037,
      "epoch": 0.01819251966397346,
      "grad_norm": 1.03125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 8.5416,
      "mean_token_accuracy": 0.05734114237129688,
      "num_tokens": 370810.0,
      "step": 170
    },
    {
      "entropy": 9.37319803237915,
      "epoch": 0.018727593771737387,
      "grad_norm": 0.76953125,
      "learning_rate": 8.7e-05,
      "loss": 8.512,
      "mean_token_accuracy": 0.05475095845758915,
      "num_tokens": 381788.0,
      "step": 175
    },
    {
      "entropy": 9.286770725250244,
      "epoch": 0.01926266787950131,
      "grad_norm": 0.953125,
      "learning_rate": 8.95e-05,
      "loss": 8.4773,
      "mean_token_accuracy": 0.05117404013872147,
      "num_tokens": 391984.0,
      "step": 180
    },
    {
      "entropy": 9.142317199707032,
      "epoch": 0.019797741987265235,
      "grad_norm": 0.609375,
      "learning_rate": 9.2e-05,
      "loss": 8.462,
      "mean_token_accuracy": 0.05436502285301685,
      "num_tokens": 403430.0,
      "step": 185
    },
    {
      "entropy": 9.018431186676025,
      "epoch": 0.02033281609502916,
      "grad_norm": 0.6484375,
      "learning_rate": 9.45e-05,
      "loss": 8.4383,
      "mean_token_accuracy": 0.052350400015711784,
      "num_tokens": 414467.0,
      "step": 190
    },
    {
      "entropy": 8.938930225372314,
      "epoch": 0.020867890202793088,
      "grad_norm": 0.7265625,
      "learning_rate": 9.7e-05,
      "loss": 8.4015,
      "mean_token_accuracy": 0.05855946056544781,
      "num_tokens": 425752.0,
      "step": 195
    },
    {
      "entropy": 8.85777645111084,
      "epoch": 0.02140296431055701,
      "grad_norm": 0.74609375,
      "learning_rate": 9.95e-05,
      "loss": 8.3693,
      "mean_token_accuracy": 0.05885430946946144,
      "num_tokens": 435908.0,
      "step": 200
    },
    {
      "entropy": 8.804098224639892,
      "epoch": 0.021938038418320937,
      "grad_norm": 0.66796875,
      "learning_rate": 0.000102,
      "loss": 8.4059,
      "mean_token_accuracy": 0.0562417134642601,
      "num_tokens": 446424.0,
      "step": 205
    },
    {
      "entropy": 8.765947341918945,
      "epoch": 0.022473112526084863,
      "grad_norm": 0.765625,
      "learning_rate": 0.00010449999999999999,
      "loss": 8.3715,
      "mean_token_accuracy": 0.06059782654047012,
      "num_tokens": 456854.0,
      "step": 210
    },
    {
      "entropy": 8.768339824676513,
      "epoch": 0.02300818663384879,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000107,
      "loss": 8.3964,
      "mean_token_accuracy": 0.06077548712491989,
      "num_tokens": 467899.0,
      "step": 215
    },
    {
      "entropy": 8.720343017578125,
      "epoch": 0.023543260741612712,
      "grad_norm": 0.859375,
      "learning_rate": 0.0001095,
      "loss": 8.3382,
      "mean_token_accuracy": 0.057309944182634354,
      "num_tokens": 478683.0,
      "step": 220
    },
    {
      "entropy": 8.685015773773193,
      "epoch": 0.02407833484937664,
      "grad_norm": 0.78125,
      "learning_rate": 0.000112,
      "loss": 8.3726,
      "mean_token_accuracy": 0.0574649728834629,
      "num_tokens": 488745.0,
      "step": 225
    },
    {
      "entropy": 8.69738130569458,
      "epoch": 0.024613408957140565,
      "grad_norm": 0.640625,
      "learning_rate": 0.0001145,
      "loss": 8.4486,
      "mean_token_accuracy": 0.05442229248583317,
      "num_tokens": 499448.0,
      "step": 230
    },
    {
      "entropy": 8.748478317260743,
      "epoch": 0.02514848306490449,
      "grad_norm": 0.71875,
      "learning_rate": 0.00011700000000000001,
      "loss": 8.3757,
      "mean_token_accuracy": 0.05678051896393299,
      "num_tokens": 510148.0,
      "step": 235
    },
    {
      "entropy": 8.693877410888671,
      "epoch": 0.025683557172668414,
      "grad_norm": 0.73828125,
      "learning_rate": 0.00011949999999999999,
      "loss": 8.3156,
      "mean_token_accuracy": 0.053839444741606714,
      "num_tokens": 522093.0,
      "step": 240
    },
    {
      "entropy": 8.693435382843017,
      "epoch": 0.02621863128043234,
      "grad_norm": 0.7109375,
      "learning_rate": 0.000122,
      "loss": 8.3365,
      "mean_token_accuracy": 0.060893227159976956,
      "num_tokens": 532598.0,
      "step": 245
    },
    {
      "entropy": 8.625317096710205,
      "epoch": 0.026753705388196266,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0001245,
      "loss": 8.3073,
      "mean_token_accuracy": 0.05914956107735634,
      "num_tokens": 542652.0,
      "step": 250
    },
    {
      "entropy": 8.666687965393066,
      "epoch": 0.027288779495960192,
      "grad_norm": 0.62890625,
      "learning_rate": 0.000127,
      "loss": 8.3343,
      "mean_token_accuracy": 0.05921483002603054,
      "num_tokens": 552526.0,
      "step": 255
    },
    {
      "entropy": 8.59899835586548,
      "epoch": 0.027823853603724115,
      "grad_norm": 0.76171875,
      "learning_rate": 0.0001295,
      "loss": 8.357,
      "mean_token_accuracy": 0.059534988924860956,
      "num_tokens": 563143.0,
      "step": 260
    },
    {
      "entropy": 8.635274982452392,
      "epoch": 0.02835892771148804,
      "grad_norm": 0.7421875,
      "learning_rate": 0.000132,
      "loss": 8.3285,
      "mean_token_accuracy": 0.05812331661581993,
      "num_tokens": 573814.0,
      "step": 265
    },
    {
      "entropy": 8.651222038269044,
      "epoch": 0.028894001819251967,
      "grad_norm": 0.74609375,
      "learning_rate": 0.00013450000000000002,
      "loss": 8.2886,
      "mean_token_accuracy": 0.06336523443460465,
      "num_tokens": 585367.0,
      "step": 270
    },
    {
      "entropy": 8.534408664703369,
      "epoch": 0.02942907592701589,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00013700000000000002,
      "loss": 8.3858,
      "mean_token_accuracy": 0.06117042489349842,
      "num_tokens": 595902.0,
      "step": 275
    },
    {
      "entropy": 8.656291198730468,
      "epoch": 0.029964150034779816,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001395,
      "loss": 8.264,
      "mean_token_accuracy": 0.05527368187904358,
      "num_tokens": 607178.0,
      "step": 280
    },
    {
      "entropy": 8.634803867340088,
      "epoch": 0.030499224142543743,
      "grad_norm": 0.8046875,
      "learning_rate": 0.00014199999999999998,
      "loss": 8.2565,
      "mean_token_accuracy": 0.05927442014217377,
      "num_tokens": 618604.0,
      "step": 285
    },
    {
      "entropy": 8.526193332672118,
      "epoch": 0.03103429825030767,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0001445,
      "loss": 8.3182,
      "mean_token_accuracy": 0.06472528986632824,
      "num_tokens": 628111.0,
      "step": 290
    },
    {
      "entropy": 8.551859188079835,
      "epoch": 0.03156937235807159,
      "grad_norm": 0.62109375,
      "learning_rate": 0.000147,
      "loss": 8.3524,
      "mean_token_accuracy": 0.06350056529045105,
      "num_tokens": 639358.0,
      "step": 295
    },
    {
      "entropy": 8.559032249450684,
      "epoch": 0.03210444646583552,
      "grad_norm": 0.703125,
      "learning_rate": 0.0001495,
      "loss": 8.2369,
      "mean_token_accuracy": 0.061780452728271484,
      "num_tokens": 649504.0,
      "step": 300
    },
    {
      "entropy": 8.562529945373536,
      "epoch": 0.032639520573599444,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000152,
      "loss": 8.2503,
      "mean_token_accuracy": 0.06190255731344223,
      "num_tokens": 659924.0,
      "step": 305
    },
    {
      "entropy": 8.53670949935913,
      "epoch": 0.03317459468136337,
      "grad_norm": 0.73046875,
      "learning_rate": 0.00015450000000000001,
      "loss": 8.2818,
      "mean_token_accuracy": 0.060764011740684507,
      "num_tokens": 671120.0,
      "step": 310
    },
    {
      "entropy": 8.538024520874023,
      "epoch": 0.033709668789127296,
      "grad_norm": 0.84375,
      "learning_rate": 0.000157,
      "loss": 8.3386,
      "mean_token_accuracy": 0.05993119850754738,
      "num_tokens": 681956.0,
      "step": 315
    },
    {
      "entropy": 8.596301746368407,
      "epoch": 0.03424474289689122,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0001595,
      "loss": 8.3111,
      "mean_token_accuracy": 0.060398251935839656,
      "num_tokens": 693226.0,
      "step": 320
    },
    {
      "entropy": 8.568628025054931,
      "epoch": 0.03477981700465514,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000162,
      "loss": 8.3253,
      "mean_token_accuracy": 0.06589160449802875,
      "num_tokens": 704691.0,
      "step": 325
    },
    {
      "entropy": 8.47495059967041,
      "epoch": 0.03531489111241907,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00016450000000000001,
      "loss": 8.2856,
      "mean_token_accuracy": 0.05669932141900062,
      "num_tokens": 716000.0,
      "step": 330
    },
    {
      "entropy": 8.474805164337159,
      "epoch": 0.035849965220182994,
      "grad_norm": 0.77734375,
      "learning_rate": 0.00016700000000000002,
      "loss": 8.1999,
      "mean_token_accuracy": 0.06121533028781414,
      "num_tokens": 726723.0,
      "step": 335
    },
    {
      "entropy": 8.498011589050293,
      "epoch": 0.03638503932794692,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00016950000000000003,
      "loss": 8.218,
      "mean_token_accuracy": 0.06442321352660656,
      "num_tokens": 736740.0,
      "step": 340
    },
    {
      "entropy": 8.463557243347168,
      "epoch": 0.03692011343571085,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00017199999999999998,
      "loss": 8.2888,
      "mean_token_accuracy": 0.06197960674762726,
      "num_tokens": 747567.0,
      "step": 345
    },
    {
      "entropy": 8.562387371063233,
      "epoch": 0.03745518754347477,
      "grad_norm": 0.76953125,
      "learning_rate": 0.00017449999999999999,
      "loss": 8.2184,
      "mean_token_accuracy": 0.06554635129868984,
      "num_tokens": 757937.0,
      "step": 350
    },
    {
      "entropy": 8.381116771697998,
      "epoch": 0.0379902616512387,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000177,
      "loss": 8.2329,
      "mean_token_accuracy": 0.06743667759001255,
      "num_tokens": 768003.0,
      "step": 355
    },
    {
      "entropy": 8.57433042526245,
      "epoch": 0.03852533575900262,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0001795,
      "loss": 8.2918,
      "mean_token_accuracy": 0.060314114391803744,
      "num_tokens": 779098.0,
      "step": 360
    },
    {
      "entropy": 8.524942874908447,
      "epoch": 0.039060409866766545,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000182,
      "loss": 8.2617,
      "mean_token_accuracy": 0.05991893112659454,
      "num_tokens": 790002.0,
      "step": 365
    },
    {
      "entropy": 8.520723247528077,
      "epoch": 0.03959548397453047,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001845,
      "loss": 8.24,
      "mean_token_accuracy": 0.0622980535030365,
      "num_tokens": 800444.0,
      "step": 370
    },
    {
      "entropy": 8.49396734237671,
      "epoch": 0.0401305580822944,
      "grad_norm": 0.80078125,
      "learning_rate": 0.000187,
      "loss": 8.2068,
      "mean_token_accuracy": 0.0646934848278761,
      "num_tokens": 810912.0,
      "step": 375
    },
    {
      "entropy": 8.529602336883546,
      "epoch": 0.04066563219005832,
      "grad_norm": 0.73046875,
      "learning_rate": 0.0001895,
      "loss": 8.2892,
      "mean_token_accuracy": 0.061168758943676946,
      "num_tokens": 822809.0,
      "step": 380
    },
    {
      "entropy": 8.436344432830811,
      "epoch": 0.04120070629782225,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000192,
      "loss": 8.1843,
      "mean_token_accuracy": 0.06499753147363663,
      "num_tokens": 833362.0,
      "step": 385
    },
    {
      "entropy": 8.464655780792237,
      "epoch": 0.041735780405586176,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0001945,
      "loss": 8.2136,
      "mean_token_accuracy": 0.06534497700631618,
      "num_tokens": 844239.0,
      "step": 390
    },
    {
      "entropy": 8.412539768218995,
      "epoch": 0.0422708545133501,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00019700000000000002,
      "loss": 8.21,
      "mean_token_accuracy": 0.05918620526790619,
      "num_tokens": 855178.0,
      "step": 395
    },
    {
      "entropy": 8.480862522125244,
      "epoch": 0.04280592862111402,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00019950000000000002,
      "loss": 8.2122,
      "mean_token_accuracy": 0.060220999643206596,
      "num_tokens": 866120.0,
      "step": 400
    },
    {
      "entropy": 8.434346771240234,
      "epoch": 0.04334100272887795,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000202,
      "loss": 8.1945,
      "mean_token_accuracy": 0.06288457810878753,
      "num_tokens": 876569.0,
      "step": 405
    },
    {
      "entropy": 8.53380298614502,
      "epoch": 0.043876076836641874,
      "grad_norm": 0.8203125,
      "learning_rate": 0.00020449999999999998,
      "loss": 8.2451,
      "mean_token_accuracy": 0.05817388035356998,
      "num_tokens": 887580.0,
      "step": 410
    },
    {
      "entropy": 8.417744541168213,
      "epoch": 0.0444111509444058,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000207,
      "loss": 8.1963,
      "mean_token_accuracy": 0.06088313870131969,
      "num_tokens": 897922.0,
      "step": 415
    },
    {
      "entropy": 8.480956935882569,
      "epoch": 0.044946225052169726,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002095,
      "loss": 8.154,
      "mean_token_accuracy": 0.062450488656759263,
      "num_tokens": 908522.0,
      "step": 420
    },
    {
      "entropy": 8.501266574859619,
      "epoch": 0.04548129915993365,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000212,
      "loss": 8.1712,
      "mean_token_accuracy": 0.06173374205827713,
      "num_tokens": 918253.0,
      "step": 425
    },
    {
      "entropy": 8.375071716308593,
      "epoch": 0.04601637326769758,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002145,
      "loss": 8.1577,
      "mean_token_accuracy": 0.06654028706252575,
      "num_tokens": 929150.0,
      "step": 430
    },
    {
      "entropy": 8.457318782806396,
      "epoch": 0.0465514473754615,
      "grad_norm": 0.8125,
      "learning_rate": 0.00021700000000000002,
      "loss": 8.2037,
      "mean_token_accuracy": 0.06640968844294548,
      "num_tokens": 941041.0,
      "step": 435
    },
    {
      "entropy": 8.420125675201415,
      "epoch": 0.047086521483225424,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002195,
      "loss": 8.1506,
      "mean_token_accuracy": 0.06981723047792912,
      "num_tokens": 952714.0,
      "step": 440
    },
    {
      "entropy": 8.38680305480957,
      "epoch": 0.04762159559098935,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000222,
      "loss": 8.2206,
      "mean_token_accuracy": 0.056760451197624205,
      "num_tokens": 964330.0,
      "step": 445
    },
    {
      "entropy": 8.4091046333313,
      "epoch": 0.04815666969875328,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0002245,
      "loss": 8.1359,
      "mean_token_accuracy": 0.06024715341627598,
      "num_tokens": 975257.0,
      "step": 450
    },
    {
      "entropy": 8.439432144165039,
      "epoch": 0.0486917438065172,
      "grad_norm": 0.71875,
      "learning_rate": 0.00022700000000000002,
      "loss": 8.173,
      "mean_token_accuracy": 0.06317405998706818,
      "num_tokens": 986335.0,
      "step": 455
    },
    {
      "entropy": 8.334716606140137,
      "epoch": 0.04922681791428113,
      "grad_norm": 0.70703125,
      "learning_rate": 0.00022950000000000002,
      "loss": 8.0587,
      "mean_token_accuracy": 0.06826802715659142,
      "num_tokens": 997521.0,
      "step": 460
    },
    {
      "entropy": 8.383316135406494,
      "epoch": 0.049761892022045055,
      "grad_norm": 0.7734375,
      "learning_rate": 0.00023200000000000003,
      "loss": 8.0987,
      "mean_token_accuracy": 0.061755566671490666,
      "num_tokens": 1008383.0,
      "step": 465
    },
    {
      "entropy": 8.42426586151123,
      "epoch": 0.05029696612980898,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023449999999999998,
      "loss": 8.1665,
      "mean_token_accuracy": 0.06531071439385414,
      "num_tokens": 1020552.0,
      "step": 470
    },
    {
      "entropy": 8.319906425476074,
      "epoch": 0.0508320402375729,
      "grad_norm": 1.046875,
      "learning_rate": 0.000237,
      "loss": 8.1349,
      "mean_token_accuracy": 0.06562352441251278,
      "num_tokens": 1031127.0,
      "step": 475
    },
    {
      "entropy": 8.294714164733886,
      "epoch": 0.05136711434533683,
      "grad_norm": 0.72265625,
      "learning_rate": 0.0002395,
      "loss": 8.0537,
      "mean_token_accuracy": 0.06683347783982754,
      "num_tokens": 1043209.0,
      "step": 480
    },
    {
      "entropy": 8.477115249633789,
      "epoch": 0.05190218845310075,
      "grad_norm": 0.77734375,
      "learning_rate": 0.000242,
      "loss": 8.182,
      "mean_token_accuracy": 0.06099085658788681,
      "num_tokens": 1054032.0,
      "step": 485
    },
    {
      "entropy": 8.271202278137206,
      "epoch": 0.05243726256086468,
      "grad_norm": 0.7109375,
      "learning_rate": 0.0002445,
      "loss": 8.0176,
      "mean_token_accuracy": 0.06989072859287263,
      "num_tokens": 1064565.0,
      "step": 490
    },
    {
      "entropy": 8.429040908813477,
      "epoch": 0.052972336668628606,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000247,
      "loss": 8.1709,
      "mean_token_accuracy": 0.06545415259897709,
      "num_tokens": 1076352.0,
      "step": 495
    },
    {
      "entropy": 8.389126300811768,
      "epoch": 0.05350741077639253,
      "grad_norm": 0.921875,
      "learning_rate": 0.0002495,
      "loss": 8.0995,
      "mean_token_accuracy": 0.06924427561461925,
      "num_tokens": 1087697.0,
      "step": 500
    },
    {
      "entropy": 8.324423885345459,
      "epoch": 0.05404248488415646,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000252,
      "loss": 8.0191,
      "mean_token_accuracy": 0.06561592519283295,
      "num_tokens": 1098056.0,
      "step": 505
    },
    {
      "entropy": 8.339075374603272,
      "epoch": 0.054577558991920384,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0002545,
      "loss": 8.0061,
      "mean_token_accuracy": 0.07367064505815506,
      "num_tokens": 1109219.0,
      "step": 510
    },
    {
      "entropy": 8.308711051940918,
      "epoch": 0.055112633099684304,
      "grad_norm": 0.76171875,
      "learning_rate": 0.000257,
      "loss": 8.0719,
      "mean_token_accuracy": 0.06625799126923085,
      "num_tokens": 1120248.0,
      "step": 515
    },
    {
      "entropy": 8.255969905853272,
      "epoch": 0.05564770720744823,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002595,
      "loss": 8.0894,
      "mean_token_accuracy": 0.06586733907461166,
      "num_tokens": 1131278.0,
      "step": 520
    },
    {
      "entropy": 8.364176273345947,
      "epoch": 0.056182781315212156,
      "grad_norm": 0.90625,
      "learning_rate": 0.000262,
      "loss": 8.1062,
      "mean_token_accuracy": 0.05967978574335575,
      "num_tokens": 1143147.0,
      "step": 525
    },
    {
      "entropy": 8.352798652648925,
      "epoch": 0.05671785542297608,
      "grad_norm": 0.7578125,
      "learning_rate": 0.00026450000000000003,
      "loss": 8.0109,
      "mean_token_accuracy": 0.06417375653982163,
      "num_tokens": 1154316.0,
      "step": 530
    },
    {
      "entropy": 8.29659538269043,
      "epoch": 0.05725292953074001,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00026700000000000004,
      "loss": 8.0685,
      "mean_token_accuracy": 0.06354658231139183,
      "num_tokens": 1164696.0,
      "step": 535
    },
    {
      "entropy": 8.285994625091552,
      "epoch": 0.057788003638503935,
      "grad_norm": 0.953125,
      "learning_rate": 0.00026950000000000005,
      "loss": 8.0421,
      "mean_token_accuracy": 0.07412712648510933,
      "num_tokens": 1176129.0,
      "step": 540
    },
    {
      "entropy": 8.28906831741333,
      "epoch": 0.05832307774626786,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00027200000000000005,
      "loss": 8.071,
      "mean_token_accuracy": 0.06631042063236237,
      "num_tokens": 1186499.0,
      "step": 545
    },
    {
      "entropy": 8.289284801483154,
      "epoch": 0.05885815185403178,
      "grad_norm": 0.859375,
      "learning_rate": 0.0002745,
      "loss": 8.0173,
      "mean_token_accuracy": 0.06871062219142914,
      "num_tokens": 1196841.0,
      "step": 550
    },
    {
      "entropy": 8.295639991760254,
      "epoch": 0.059393225961795706,
      "grad_norm": 0.828125,
      "learning_rate": 0.000277,
      "loss": 7.9761,
      "mean_token_accuracy": 0.07033407911658288,
      "num_tokens": 1207647.0,
      "step": 555
    },
    {
      "entropy": 8.243279266357423,
      "epoch": 0.05992830006955963,
      "grad_norm": 0.875,
      "learning_rate": 0.0002795,
      "loss": 7.963,
      "mean_token_accuracy": 0.072792012616992,
      "num_tokens": 1217072.0,
      "step": 560
    },
    {
      "entropy": 8.300912570953368,
      "epoch": 0.06046337417732356,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028199999999999997,
      "loss": 8.073,
      "mean_token_accuracy": 0.06520008333027363,
      "num_tokens": 1228564.0,
      "step": 565
    },
    {
      "entropy": 8.203814697265624,
      "epoch": 0.060998448285087485,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002845,
      "loss": 7.9942,
      "mean_token_accuracy": 0.06911226697266101,
      "num_tokens": 1238755.0,
      "step": 570
    },
    {
      "entropy": 8.239333248138427,
      "epoch": 0.06153352239285141,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000287,
      "loss": 7.9646,
      "mean_token_accuracy": 0.07128063514828682,
      "num_tokens": 1248968.0,
      "step": 575
    },
    {
      "entropy": 8.241281127929687,
      "epoch": 0.06206859650061534,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0002895,
      "loss": 7.9258,
      "mean_token_accuracy": 0.0690683875232935,
      "num_tokens": 1259256.0,
      "step": 580
    },
    {
      "entropy": 8.180652809143066,
      "epoch": 0.06260367060837926,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000292,
      "loss": 7.9653,
      "mean_token_accuracy": 0.07212716266512871,
      "num_tokens": 1270469.0,
      "step": 585
    },
    {
      "entropy": 8.237191200256348,
      "epoch": 0.06313874471614318,
      "grad_norm": 0.78125,
      "learning_rate": 0.0002945,
      "loss": 7.9286,
      "mean_token_accuracy": 0.06839926056563854,
      "num_tokens": 1280558.0,
      "step": 590
    },
    {
      "entropy": 8.18700132369995,
      "epoch": 0.06367381882390712,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000297,
      "loss": 7.9695,
      "mean_token_accuracy": 0.06805149577558041,
      "num_tokens": 1291372.0,
      "step": 595
    },
    {
      "entropy": 8.287422943115235,
      "epoch": 0.06420889293167104,
      "grad_norm": 0.76171875,
      "learning_rate": 0.0002995,
      "loss": 8.0554,
      "mean_token_accuracy": 0.06435842141509056,
      "num_tokens": 1302521.0,
      "step": 600
    },
    {
      "entropy": 8.17969207763672,
      "epoch": 0.06474396703943497,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000302,
      "loss": 7.8954,
      "mean_token_accuracy": 0.07011312432587147,
      "num_tokens": 1313632.0,
      "step": 605
    },
    {
      "entropy": 8.187294721603394,
      "epoch": 0.06527904114719889,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003045,
      "loss": 7.9765,
      "mean_token_accuracy": 0.06986295953392982,
      "num_tokens": 1325247.0,
      "step": 610
    },
    {
      "entropy": 8.145281457901001,
      "epoch": 0.06581411525496281,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000307,
      "loss": 7.9495,
      "mean_token_accuracy": 0.07135364897549153,
      "num_tokens": 1336026.0,
      "step": 615
    },
    {
      "entropy": 8.15605869293213,
      "epoch": 0.06634918936272674,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0003095,
      "loss": 7.8724,
      "mean_token_accuracy": 0.07117584012448788,
      "num_tokens": 1346920.0,
      "step": 620
    },
    {
      "entropy": 8.162275314331055,
      "epoch": 0.06688426347049066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000312,
      "loss": 8.0208,
      "mean_token_accuracy": 0.06915087811648846,
      "num_tokens": 1359752.0,
      "step": 625
    },
    {
      "entropy": 8.272911071777344,
      "epoch": 0.06741933757825459,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003145,
      "loss": 8.0125,
      "mean_token_accuracy": 0.06714854501187802,
      "num_tokens": 1370874.0,
      "step": 630
    },
    {
      "entropy": 8.166405773162841,
      "epoch": 0.06795441168601851,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000317,
      "loss": 7.933,
      "mean_token_accuracy": 0.07350475676357746,
      "num_tokens": 1382316.0,
      "step": 635
    },
    {
      "entropy": 8.12221031188965,
      "epoch": 0.06848948579378245,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0003195,
      "loss": 7.9287,
      "mean_token_accuracy": 0.06466298960149289,
      "num_tokens": 1394656.0,
      "step": 640
    },
    {
      "entropy": 8.132702732086182,
      "epoch": 0.06902455990154636,
      "grad_norm": 0.921875,
      "learning_rate": 0.000322,
      "loss": 7.904,
      "mean_token_accuracy": 0.07535643540322781,
      "num_tokens": 1405061.0,
      "step": 645
    },
    {
      "entropy": 8.1514066696167,
      "epoch": 0.06955963400931028,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.8574,
      "mean_token_accuracy": 0.07722640819847584,
      "num_tokens": 1415573.0,
      "step": 650
    },
    {
      "entropy": 8.076834058761596,
      "epoch": 0.07009470811707422,
      "grad_norm": 0.90625,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.8971,
      "mean_token_accuracy": 0.06901067160069943,
      "num_tokens": 1426103.0,
      "step": 655
    },
    {
      "entropy": 8.042639923095702,
      "epoch": 0.07062978222483814,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.9236,
      "mean_token_accuracy": 0.06627062112092971,
      "num_tokens": 1437256.0,
      "step": 660
    },
    {
      "entropy": 8.271612071990967,
      "epoch": 0.07116485633260207,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.8667,
      "mean_token_accuracy": 0.07184043861925601,
      "num_tokens": 1446975.0,
      "step": 665
    },
    {
      "entropy": 8.110226726531982,
      "epoch": 0.07169993044036599,
      "grad_norm": 0.984375,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.8856,
      "mean_token_accuracy": 0.07188437804579735,
      "num_tokens": 1457751.0,
      "step": 670
    },
    {
      "entropy": 7.960531234741211,
      "epoch": 0.07223500454812992,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000337,
      "loss": 7.8516,
      "mean_token_accuracy": 0.07059676125645638,
      "num_tokens": 1468186.0,
      "step": 675
    },
    {
      "entropy": 8.202023792266846,
      "epoch": 0.07277007865589384,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003395,
      "loss": 7.896,
      "mean_token_accuracy": 0.0678649365901947,
      "num_tokens": 1479780.0,
      "step": 680
    },
    {
      "entropy": 8.120744943618774,
      "epoch": 0.07330515276365776,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000342,
      "loss": 7.841,
      "mean_token_accuracy": 0.07492000721395016,
      "num_tokens": 1490565.0,
      "step": 685
    },
    {
      "entropy": 8.100931549072266,
      "epoch": 0.0738402268714217,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.8612,
      "mean_token_accuracy": 0.07190988808870316,
      "num_tokens": 1501552.0,
      "step": 690
    },
    {
      "entropy": 8.065852403640747,
      "epoch": 0.07437530097918561,
      "grad_norm": 1.140625,
      "learning_rate": 0.000347,
      "loss": 7.8713,
      "mean_token_accuracy": 0.06612264662981034,
      "num_tokens": 1513311.0,
      "step": 695
    },
    {
      "entropy": 8.115069341659545,
      "epoch": 0.07491037508694955,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0003495,
      "loss": 7.895,
      "mean_token_accuracy": 0.07125801891088486,
      "num_tokens": 1524142.0,
      "step": 700
    },
    {
      "entropy": 8.081360721588135,
      "epoch": 0.07544544919471347,
      "grad_norm": 1.0,
      "learning_rate": 0.000352,
      "loss": 7.8645,
      "mean_token_accuracy": 0.07241233922541142,
      "num_tokens": 1535571.0,
      "step": 705
    },
    {
      "entropy": 8.069763135910033,
      "epoch": 0.0759805233024774,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0003545,
      "loss": 7.8535,
      "mean_token_accuracy": 0.07254451140761375,
      "num_tokens": 1546528.0,
      "step": 710
    },
    {
      "entropy": 8.03998236656189,
      "epoch": 0.07651559741024132,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000357,
      "loss": 7.8782,
      "mean_token_accuracy": 0.06598596908152103,
      "num_tokens": 1557264.0,
      "step": 715
    },
    {
      "entropy": 8.095155239105225,
      "epoch": 0.07705067151800524,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003595,
      "loss": 7.8277,
      "mean_token_accuracy": 0.07438212521374225,
      "num_tokens": 1566773.0,
      "step": 720
    },
    {
      "entropy": 7.952378559112549,
      "epoch": 0.07758574562576917,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000362,
      "loss": 7.7892,
      "mean_token_accuracy": 0.07861022278666496,
      "num_tokens": 1578173.0,
      "step": 725
    },
    {
      "entropy": 8.084946393966675,
      "epoch": 0.07812081973353309,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0003645,
      "loss": 7.7521,
      "mean_token_accuracy": 0.07678151875734329,
      "num_tokens": 1588537.0,
      "step": 730
    },
    {
      "entropy": 7.978636407852173,
      "epoch": 0.07865589384129702,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000367,
      "loss": 7.7674,
      "mean_token_accuracy": 0.08002844266593456,
      "num_tokens": 1599159.0,
      "step": 735
    },
    {
      "entropy": 8.004686641693116,
      "epoch": 0.07919096794906094,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003695,
      "loss": 7.8519,
      "mean_token_accuracy": 0.07209457606077194,
      "num_tokens": 1610378.0,
      "step": 740
    },
    {
      "entropy": 8.040560579299926,
      "epoch": 0.07972604205682488,
      "grad_norm": 1.0,
      "learning_rate": 0.000372,
      "loss": 7.861,
      "mean_token_accuracy": 0.07267312295734882,
      "num_tokens": 1621215.0,
      "step": 745
    },
    {
      "entropy": 7.936432123184204,
      "epoch": 0.0802611161645888,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003745,
      "loss": 7.79,
      "mean_token_accuracy": 0.07331596687436104,
      "num_tokens": 1631170.0,
      "step": 750
    },
    {
      "entropy": 8.133660507202148,
      "epoch": 0.08079619027235273,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000377,
      "loss": 7.7989,
      "mean_token_accuracy": 0.07404692210257054,
      "num_tokens": 1641897.0,
      "step": 755
    },
    {
      "entropy": 7.98411865234375,
      "epoch": 0.08133126438011665,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0003795,
      "loss": 7.8606,
      "mean_token_accuracy": 0.07070158869028091,
      "num_tokens": 1653592.0,
      "step": 760
    },
    {
      "entropy": 7.9289227485656735,
      "epoch": 0.08186633848788057,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000382,
      "loss": 7.7832,
      "mean_token_accuracy": 0.06835788786411286,
      "num_tokens": 1665760.0,
      "step": 765
    },
    {
      "entropy": 8.100461387634278,
      "epoch": 0.0824014125956445,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0003845,
      "loss": 7.7963,
      "mean_token_accuracy": 0.07924762107431889,
      "num_tokens": 1676211.0,
      "step": 770
    },
    {
      "entropy": 8.04494333267212,
      "epoch": 0.08293648670340842,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.8154,
      "mean_token_accuracy": 0.07205268889665603,
      "num_tokens": 1686405.0,
      "step": 775
    },
    {
      "entropy": 8.116489696502686,
      "epoch": 0.08347156081117235,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.816,
      "mean_token_accuracy": 0.07210099287331104,
      "num_tokens": 1697340.0,
      "step": 780
    },
    {
      "entropy": 8.015373849868775,
      "epoch": 0.08400663491893627,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.716,
      "mean_token_accuracy": 0.0766339547932148,
      "num_tokens": 1707250.0,
      "step": 785
    },
    {
      "entropy": 7.935082626342774,
      "epoch": 0.0845417090267002,
      "grad_norm": 1.0,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.6759,
      "mean_token_accuracy": 0.07099579833447933,
      "num_tokens": 1718333.0,
      "step": 790
    },
    {
      "entropy": 8.006236696243286,
      "epoch": 0.08507678313446412,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.7727,
      "mean_token_accuracy": 0.07233709581196308,
      "num_tokens": 1730520.0,
      "step": 795
    },
    {
      "entropy": 7.998106575012207,
      "epoch": 0.08561185724222804,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003995,
      "loss": 7.8245,
      "mean_token_accuracy": 0.0690892156213522,
      "num_tokens": 1741898.0,
      "step": 800
    },
    {
      "entropy": 7.937285470962524,
      "epoch": 0.08614693134999198,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000402,
      "loss": 7.7619,
      "mean_token_accuracy": 0.07777528837323189,
      "num_tokens": 1751671.0,
      "step": 805
    },
    {
      "entropy": 7.906507158279419,
      "epoch": 0.0866820054577559,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004045,
      "loss": 7.7796,
      "mean_token_accuracy": 0.07674273923039436,
      "num_tokens": 1762076.0,
      "step": 810
    },
    {
      "entropy": 8.027475690841674,
      "epoch": 0.08721707956551983,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.7598,
      "mean_token_accuracy": 0.07528497837483883,
      "num_tokens": 1772666.0,
      "step": 815
    },
    {
      "entropy": 7.961116218566895,
      "epoch": 0.08775215367328375,
      "grad_norm": 0.875,
      "learning_rate": 0.0004095,
      "loss": 7.7747,
      "mean_token_accuracy": 0.08033046908676625,
      "num_tokens": 1783621.0,
      "step": 820
    },
    {
      "entropy": 7.981131458282471,
      "epoch": 0.08828722778104768,
      "grad_norm": 1.328125,
      "learning_rate": 0.000412,
      "loss": 7.8234,
      "mean_token_accuracy": 0.07162153050303459,
      "num_tokens": 1794654.0,
      "step": 825
    },
    {
      "entropy": 7.881627893447876,
      "epoch": 0.0888223018888116,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004145,
      "loss": 7.7166,
      "mean_token_accuracy": 0.08037983030080795,
      "num_tokens": 1805853.0,
      "step": 830
    },
    {
      "entropy": 7.805557489395142,
      "epoch": 0.08935737599657552,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000417,
      "loss": 7.6985,
      "mean_token_accuracy": 0.07677435502409935,
      "num_tokens": 1817006.0,
      "step": 835
    },
    {
      "entropy": 8.030841255187989,
      "epoch": 0.08989245010433945,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004195,
      "loss": 7.6996,
      "mean_token_accuracy": 0.07248782590031624,
      "num_tokens": 1827166.0,
      "step": 840
    },
    {
      "entropy": 8.012809896469117,
      "epoch": 0.09042752421210337,
      "grad_norm": 0.984375,
      "learning_rate": 0.000422,
      "loss": 7.781,
      "mean_token_accuracy": 0.07629549242556095,
      "num_tokens": 1837182.0,
      "step": 845
    },
    {
      "entropy": 7.861330270767212,
      "epoch": 0.0909625983198673,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004245,
      "loss": 7.7371,
      "mean_token_accuracy": 0.07596477270126342,
      "num_tokens": 1849034.0,
      "step": 850
    },
    {
      "entropy": 7.900086402893066,
      "epoch": 0.09149767242763122,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000427,
      "loss": 7.6039,
      "mean_token_accuracy": 0.08392401710152626,
      "num_tokens": 1859033.0,
      "step": 855
    },
    {
      "entropy": 7.819063472747803,
      "epoch": 0.09203274653539516,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004295,
      "loss": 7.7286,
      "mean_token_accuracy": 0.07808272317051887,
      "num_tokens": 1870274.0,
      "step": 860
    },
    {
      "entropy": 7.888581132888794,
      "epoch": 0.09256782064315908,
      "grad_norm": 1.25,
      "learning_rate": 0.000432,
      "loss": 7.6714,
      "mean_token_accuracy": 0.08056179471313954,
      "num_tokens": 1881306.0,
      "step": 865
    },
    {
      "entropy": 7.833876037597657,
      "epoch": 0.093102894750923,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004345,
      "loss": 7.6631,
      "mean_token_accuracy": 0.07485895380377769,
      "num_tokens": 1892464.0,
      "step": 870
    },
    {
      "entropy": 7.968428325653076,
      "epoch": 0.09363796885868693,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000437,
      "loss": 7.7325,
      "mean_token_accuracy": 0.07241618894040584,
      "num_tokens": 1903743.0,
      "step": 875
    },
    {
      "entropy": 7.845100402832031,
      "epoch": 0.09417304296645085,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004395,
      "loss": 7.6869,
      "mean_token_accuracy": 0.08143132627010345,
      "num_tokens": 1915253.0,
      "step": 880
    },
    {
      "entropy": 7.928981351852417,
      "epoch": 0.09470811707421478,
      "grad_norm": 1.171875,
      "learning_rate": 0.000442,
      "loss": 7.6399,
      "mean_token_accuracy": 0.08035231530666351,
      "num_tokens": 1926113.0,
      "step": 885
    },
    {
      "entropy": 7.823946809768676,
      "epoch": 0.0952431911819787,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004445,
      "loss": 7.735,
      "mean_token_accuracy": 0.07786822356283665,
      "num_tokens": 1937718.0,
      "step": 890
    },
    {
      "entropy": 7.9432416439056395,
      "epoch": 0.09577826528974263,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000447,
      "loss": 7.7198,
      "mean_token_accuracy": 0.07977167144417763,
      "num_tokens": 1948307.0,
      "step": 895
    },
    {
      "entropy": 7.745344495773315,
      "epoch": 0.09631333939750655,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.5286,
      "mean_token_accuracy": 0.08684809878468513,
      "num_tokens": 1957861.0,
      "step": 900
    },
    {
      "entropy": 7.771445035934448,
      "epoch": 0.09684841350527049,
      "grad_norm": 0.81640625,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.5873,
      "mean_token_accuracy": 0.08686093017458915,
      "num_tokens": 1967984.0,
      "step": 905
    },
    {
      "entropy": 7.84833459854126,
      "epoch": 0.0973834876130344,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.5976,
      "mean_token_accuracy": 0.08169375136494636,
      "num_tokens": 1978608.0,
      "step": 910
    },
    {
      "entropy": 7.825530004501343,
      "epoch": 0.09791856172079832,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.6436,
      "mean_token_accuracy": 0.08153785914182662,
      "num_tokens": 1989355.0,
      "step": 915
    },
    {
      "entropy": 7.696963882446289,
      "epoch": 0.09845363582856226,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.6287,
      "mean_token_accuracy": 0.08496365323662758,
      "num_tokens": 1999754.0,
      "step": 920
    },
    {
      "entropy": 7.85645866394043,
      "epoch": 0.09898870993632618,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000462,
      "loss": 7.638,
      "mean_token_accuracy": 0.08035108707845211,
      "num_tokens": 2010877.0,
      "step": 925
    },
    {
      "entropy": 7.902437019348144,
      "epoch": 0.09952378404409011,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004645,
      "loss": 7.7244,
      "mean_token_accuracy": 0.0775139220058918,
      "num_tokens": 2022301.0,
      "step": 930
    },
    {
      "entropy": 7.757477712631226,
      "epoch": 0.10005885815185403,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000467,
      "loss": 7.6194,
      "mean_token_accuracy": 0.08088013455271721,
      "num_tokens": 2033537.0,
      "step": 935
    },
    {
      "entropy": 7.801491928100586,
      "epoch": 0.10059393225961796,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004695,
      "loss": 7.5671,
      "mean_token_accuracy": 0.0867505919188261,
      "num_tokens": 2044360.0,
      "step": 940
    },
    {
      "entropy": 7.825805759429931,
      "epoch": 0.10112900636738188,
      "grad_norm": 0.921875,
      "learning_rate": 0.000472,
      "loss": 7.6443,
      "mean_token_accuracy": 0.08098645284771919,
      "num_tokens": 2055000.0,
      "step": 945
    },
    {
      "entropy": 7.741448211669922,
      "epoch": 0.1016640804751458,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004745,
      "loss": 7.6416,
      "mean_token_accuracy": 0.08422183245420456,
      "num_tokens": 2064807.0,
      "step": 950
    },
    {
      "entropy": 7.827320861816406,
      "epoch": 0.10219915458290973,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000477,
      "loss": 7.6029,
      "mean_token_accuracy": 0.07696961686015129,
      "num_tokens": 2075618.0,
      "step": 955
    },
    {
      "entropy": 7.788719272613525,
      "epoch": 0.10273422869067365,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004795,
      "loss": 7.6011,
      "mean_token_accuracy": 0.0814262680709362,
      "num_tokens": 2086226.0,
      "step": 960
    },
    {
      "entropy": 7.8088513851165775,
      "epoch": 0.10326930279843759,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000482,
      "loss": 7.5913,
      "mean_token_accuracy": 0.078329161927104,
      "num_tokens": 2097222.0,
      "step": 965
    },
    {
      "entropy": 7.7249979972839355,
      "epoch": 0.1038043769062015,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004845,
      "loss": 7.6347,
      "mean_token_accuracy": 0.08402503952383995,
      "num_tokens": 2108919.0,
      "step": 970
    },
    {
      "entropy": 7.705420303344726,
      "epoch": 0.10433945101396544,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000487,
      "loss": 7.5907,
      "mean_token_accuracy": 0.07948351129889489,
      "num_tokens": 2120215.0,
      "step": 975
    },
    {
      "entropy": 7.832537698745727,
      "epoch": 0.10487452512172936,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004895,
      "loss": 7.6102,
      "mean_token_accuracy": 0.07788611873984337,
      "num_tokens": 2131503.0,
      "step": 980
    },
    {
      "entropy": 7.718122386932373,
      "epoch": 0.10540959922949328,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000492,
      "loss": 7.5579,
      "mean_token_accuracy": 0.08647179454565049,
      "num_tokens": 2141551.0,
      "step": 985
    },
    {
      "entropy": 7.673153209686279,
      "epoch": 0.10594467333725721,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004945,
      "loss": 7.5469,
      "mean_token_accuracy": 0.08761123642325401,
      "num_tokens": 2151821.0,
      "step": 990
    },
    {
      "entropy": 7.752789545059204,
      "epoch": 0.10647974744502113,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000497,
      "loss": 7.5979,
      "mean_token_accuracy": 0.0813947595655918,
      "num_tokens": 2162474.0,
      "step": 995
    },
    {
      "entropy": 7.811589527130127,
      "epoch": 0.10701482155278506,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004995,
      "loss": 7.6235,
      "mean_token_accuracy": 0.08300729878246785,
      "num_tokens": 2173985.0,
      "step": 1000
    },
    {
      "entropy": 7.779503297805786,
      "epoch": 0.10754989566054898,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004999999979210101,
      "loss": 7.6631,
      "mean_token_accuracy": 0.07762378752231598,
      "num_tokens": 2186432.0,
      "step": 1005
    },
    {
      "entropy": 7.744147109985351,
      "epoch": 0.10808496976831292,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999999894751138,
      "loss": 7.5197,
      "mean_token_accuracy": 0.0851745329797268,
      "num_tokens": 2197085.0,
      "step": 1010
    },
    {
      "entropy": 7.667597770690918,
      "epoch": 0.10862004387607684,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999999745323744,
      "loss": 7.539,
      "mean_token_accuracy": 0.0865162130445242,
      "num_tokens": 2207162.0,
      "step": 1015
    },
    {
      "entropy": 7.788536834716797,
      "epoch": 0.10915511798384077,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999999530927923,
      "loss": 7.6289,
      "mean_token_accuracy": 0.07713088616728783,
      "num_tokens": 2218714.0,
      "step": 1020
    },
    {
      "entropy": 7.642853784561157,
      "epoch": 0.10969019209160469,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004999999251563682,
      "loss": 7.5508,
      "mean_token_accuracy": 0.08075930066406727,
      "num_tokens": 2229711.0,
      "step": 1025
    },
    {
      "entropy": 7.76542911529541,
      "epoch": 0.11022526619936861,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004999998907231029,
      "loss": 7.559,
      "mean_token_accuracy": 0.08542619198560715,
      "num_tokens": 2239885.0,
      "step": 1030
    },
    {
      "entropy": 7.702492570877075,
      "epoch": 0.11076034030713254,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999998497929974,
      "loss": 7.5658,
      "mean_token_accuracy": 0.08461903110146522,
      "num_tokens": 2250221.0,
      "step": 1035
    },
    {
      "entropy": 7.676487588882447,
      "epoch": 0.11129541441489646,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004999998023660527,
      "loss": 7.6434,
      "mean_token_accuracy": 0.07421790175139904,
      "num_tokens": 2261059.0,
      "step": 1040
    },
    {
      "entropy": 7.728254747390747,
      "epoch": 0.11183048852266039,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004999997484422705,
      "loss": 7.4963,
      "mean_token_accuracy": 0.08282011076807975,
      "num_tokens": 2271174.0,
      "step": 1045
    },
    {
      "entropy": 7.690599012374878,
      "epoch": 0.11236556263042431,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999996880216521,
      "loss": 7.5629,
      "mean_token_accuracy": 0.08049203157424926,
      "num_tokens": 2282431.0,
      "step": 1050
    },
    {
      "entropy": 7.683617353439331,
      "epoch": 0.11290063673818825,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999996211041994,
      "loss": 7.5512,
      "mean_token_accuracy": 0.08223441652953625,
      "num_tokens": 2292891.0,
      "step": 1055
    },
    {
      "entropy": 7.760586452484131,
      "epoch": 0.11343571084595216,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999995476899141,
      "loss": 7.5679,
      "mean_token_accuracy": 0.08535757660865784,
      "num_tokens": 2303764.0,
      "step": 1060
    },
    {
      "entropy": 7.620778036117554,
      "epoch": 0.11397078495371608,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999994677787987,
      "loss": 7.4748,
      "mean_token_accuracy": 0.08901284039020538,
      "num_tokens": 2314172.0,
      "step": 1065
    },
    {
      "entropy": 7.693076801300049,
      "epoch": 0.11450585906148002,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999993813708551,
      "loss": 7.4564,
      "mean_token_accuracy": 0.08755525574088097,
      "num_tokens": 2324848.0,
      "step": 1070
    },
    {
      "entropy": 7.582232046127319,
      "epoch": 0.11504093316924394,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000499999288466086,
      "loss": 7.4952,
      "mean_token_accuracy": 0.08118128105998039,
      "num_tokens": 2336601.0,
      "step": 1075
    },
    {
      "entropy": 7.6848491668701175,
      "epoch": 0.11557600727700787,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004999991890644941,
      "loss": 7.468,
      "mean_token_accuracy": 0.08362995944917202,
      "num_tokens": 2347281.0,
      "step": 1080
    },
    {
      "entropy": 7.642825555801392,
      "epoch": 0.11611108138477179,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999990831660822,
      "loss": 7.5125,
      "mean_token_accuracy": 0.08922351896762848,
      "num_tokens": 2358199.0,
      "step": 1085
    },
    {
      "entropy": 7.686036920547485,
      "epoch": 0.11664615549253572,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999989707708534,
      "loss": 7.5507,
      "mean_token_accuracy": 0.08290127031505108,
      "num_tokens": 2369092.0,
      "step": 1090
    },
    {
      "entropy": 7.613842344284057,
      "epoch": 0.11718122960029964,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999988518788111,
      "loss": 7.5223,
      "mean_token_accuracy": 0.08195193111896515,
      "num_tokens": 2378849.0,
      "step": 1095
    },
    {
      "entropy": 7.688374471664429,
      "epoch": 0.11771630370806356,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004999987264899583,
      "loss": 7.5234,
      "mean_token_accuracy": 0.08619830012321472,
      "num_tokens": 2389587.0,
      "step": 1100
    },
    {
      "entropy": 7.740108346939087,
      "epoch": 0.1182513778158275,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000499998594604299,
      "loss": 7.4909,
      "mean_token_accuracy": 0.08550030142068862,
      "num_tokens": 2400442.0,
      "step": 1105
    },
    {
      "entropy": 7.534623289108277,
      "epoch": 0.11878645192359141,
      "grad_norm": 0.828125,
      "learning_rate": 0.000499998456221837,
      "loss": 7.4883,
      "mean_token_accuracy": 0.08606440126895905,
      "num_tokens": 2410771.0,
      "step": 1110
    },
    {
      "entropy": 7.644052982330322,
      "epoch": 0.11932152603135535,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999983113425762,
      "loss": 7.4918,
      "mean_token_accuracy": 0.09239444956183433,
      "num_tokens": 2421112.0,
      "step": 1115
    },
    {
      "entropy": 7.656321287155151,
      "epoch": 0.11985660013911927,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999981599665207,
      "loss": 7.4718,
      "mean_token_accuracy": 0.08363690003752708,
      "num_tokens": 2431891.0,
      "step": 1120
    },
    {
      "entropy": 7.589166831970215,
      "epoch": 0.1203916742468832,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999980020936748,
      "loss": 7.4192,
      "mean_token_accuracy": 0.09008971899747849,
      "num_tokens": 2442987.0,
      "step": 1125
    },
    {
      "entropy": 7.620885610580444,
      "epoch": 0.12092674835464712,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999978377240434,
      "loss": 7.4522,
      "mean_token_accuracy": 0.08518644496798515,
      "num_tokens": 2452703.0,
      "step": 1130
    },
    {
      "entropy": 7.634619951248169,
      "epoch": 0.12146182246241104,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499997666857631,
      "loss": 7.5143,
      "mean_token_accuracy": 0.08547966852784157,
      "num_tokens": 2462975.0,
      "step": 1135
    },
    {
      "entropy": 7.556209659576416,
      "epoch": 0.12199689657017497,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004999974894944426,
      "loss": 7.465,
      "mean_token_accuracy": 0.08037517294287681,
      "num_tokens": 2473853.0,
      "step": 1140
    },
    {
      "entropy": 7.648143863677978,
      "epoch": 0.12253197067793889,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004999973056344832,
      "loss": 7.4356,
      "mean_token_accuracy": 0.08240287378430367,
      "num_tokens": 2485633.0,
      "step": 1145
    },
    {
      "entropy": 7.544134950637817,
      "epoch": 0.12306704478570282,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004999971152777583,
      "loss": 7.4008,
      "mean_token_accuracy": 0.0854021079838276,
      "num_tokens": 2496922.0,
      "step": 1150
    },
    {
      "entropy": 7.654058313369751,
      "epoch": 0.12360211889346674,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004999969184242733,
      "loss": 7.4612,
      "mean_token_accuracy": 0.08849633410573006,
      "num_tokens": 2507890.0,
      "step": 1155
    },
    {
      "entropy": 7.603252363204956,
      "epoch": 0.12413719300123068,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499996715074034,
      "loss": 7.4782,
      "mean_token_accuracy": 0.08564752526581287,
      "num_tokens": 2519749.0,
      "step": 1160
    },
    {
      "entropy": 7.571427631378174,
      "epoch": 0.1246722671089946,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999965052270461,
      "loss": 7.4656,
      "mean_token_accuracy": 0.08699908182024955,
      "num_tokens": 2530666.0,
      "step": 1165
    },
    {
      "entropy": 7.64540319442749,
      "epoch": 0.12520734121675853,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999962888833157,
      "loss": 7.471,
      "mean_token_accuracy": 0.08744210675358773,
      "num_tokens": 2541594.0,
      "step": 1170
    },
    {
      "entropy": 7.585118579864502,
      "epoch": 0.12574241532452243,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999960660428491,
      "loss": 7.4757,
      "mean_token_accuracy": 0.08734978139400482,
      "num_tokens": 2552330.0,
      "step": 1175
    },
    {
      "entropy": 7.640705585479736,
      "epoch": 0.12627748943228637,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999958367056526,
      "loss": 7.5049,
      "mean_token_accuracy": 0.08555901274085045,
      "num_tokens": 2564601.0,
      "step": 1180
    },
    {
      "entropy": 7.65483775138855,
      "epoch": 0.1268125635400503,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499995600871733,
      "loss": 7.4721,
      "mean_token_accuracy": 0.08904926031827927,
      "num_tokens": 2575539.0,
      "step": 1185
    },
    {
      "entropy": 7.503138256072998,
      "epoch": 0.12734763764781423,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004999953585410971,
      "loss": 7.4135,
      "mean_token_accuracy": 0.08864182159304619,
      "num_tokens": 2586635.0,
      "step": 1190
    },
    {
      "entropy": 7.650179767608643,
      "epoch": 0.12788271175557814,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999951097137518,
      "loss": 7.4342,
      "mean_token_accuracy": 0.09002284705638885,
      "num_tokens": 2596814.0,
      "step": 1195
    },
    {
      "entropy": 7.510055303573608,
      "epoch": 0.12841778586334207,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004999948543897044,
      "loss": 7.4695,
      "mean_token_accuracy": 0.08551743030548095,
      "num_tokens": 2608087.0,
      "step": 1200
    },
    {
      "entropy": 7.754941749572754,
      "epoch": 0.128952859971106,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999945925689621,
      "loss": 7.4764,
      "mean_token_accuracy": 0.09063678905367852,
      "num_tokens": 2619377.0,
      "step": 1205
    },
    {
      "entropy": 7.45523681640625,
      "epoch": 0.12948793407886994,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999943242515325,
      "loss": 7.3686,
      "mean_token_accuracy": 0.08870491459965706,
      "num_tokens": 2629089.0,
      "step": 1210
    },
    {
      "entropy": 7.6135406494140625,
      "epoch": 0.13002300818663384,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999940494374236,
      "loss": 7.4405,
      "mean_token_accuracy": 0.0854348972439766,
      "num_tokens": 2639743.0,
      "step": 1215
    },
    {
      "entropy": 7.64836049079895,
      "epoch": 0.13055808229439778,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000499993768126643,
      "loss": 7.4719,
      "mean_token_accuracy": 0.08272441625595092,
      "num_tokens": 2650631.0,
      "step": 1220
    },
    {
      "entropy": 7.567201042175293,
      "epoch": 0.1310931564021617,
      "grad_norm": 0.875,
      "learning_rate": 0.000499993480319199,
      "loss": 7.4509,
      "mean_token_accuracy": 0.08986086025834084,
      "num_tokens": 2660197.0,
      "step": 1225
    },
    {
      "entropy": 7.526758337020874,
      "epoch": 0.13162823050992561,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999931860150999,
      "loss": 7.4156,
      "mean_token_accuracy": 0.08493227660655975,
      "num_tokens": 2670162.0,
      "step": 1230
    },
    {
      "entropy": 7.630824851989746,
      "epoch": 0.13216330461768955,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004999928852143541,
      "loss": 7.4031,
      "mean_token_accuracy": 0.09029604569077491,
      "num_tokens": 2681233.0,
      "step": 1235
    },
    {
      "entropy": 7.431710243225098,
      "epoch": 0.13269837872545348,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004999925779169703,
      "loss": 7.3671,
      "mean_token_accuracy": 0.09119424894452095,
      "num_tokens": 2692145.0,
      "step": 1240
    },
    {
      "entropy": 7.588590002059936,
      "epoch": 0.13323345283321741,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004999922641229575,
      "loss": 7.4026,
      "mean_token_accuracy": 0.09271374940872193,
      "num_tokens": 2703192.0,
      "step": 1245
    },
    {
      "entropy": 7.56281681060791,
      "epoch": 0.13376852694098132,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004999919438323248,
      "loss": 7.4117,
      "mean_token_accuracy": 0.08688192069530487,
      "num_tokens": 2714804.0,
      "step": 1250
    },
    {
      "entropy": 7.49661545753479,
      "epoch": 0.13430360104874525,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999916170450812,
      "loss": 7.3701,
      "mean_token_accuracy": 0.0922422744333744,
      "num_tokens": 2724340.0,
      "step": 1255
    },
    {
      "entropy": 7.538181924819947,
      "epoch": 0.13483867515650919,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004999912837612364,
      "loss": 7.3767,
      "mean_token_accuracy": 0.09194523319602013,
      "num_tokens": 2735767.0,
      "step": 1260
    },
    {
      "entropy": 7.622563457489013,
      "epoch": 0.1353737492642731,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999909439807997,
      "loss": 7.4463,
      "mean_token_accuracy": 0.08266032785177231,
      "num_tokens": 2747658.0,
      "step": 1265
    },
    {
      "entropy": 7.48611364364624,
      "epoch": 0.13590882337203702,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999905977037813,
      "loss": 7.312,
      "mean_token_accuracy": 0.09179738909006119,
      "num_tokens": 2757901.0,
      "step": 1270
    },
    {
      "entropy": 7.4827721118927,
      "epoch": 0.13644389747980096,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999902449301909,
      "loss": 7.3539,
      "mean_token_accuracy": 0.08355544060468674,
      "num_tokens": 2768945.0,
      "step": 1275
    },
    {
      "entropy": 7.4961072444915775,
      "epoch": 0.1369789715875649,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0004999898856600387,
      "loss": 7.3173,
      "mean_token_accuracy": 0.10180329754948617,
      "num_tokens": 2780276.0,
      "step": 1280
    },
    {
      "entropy": 7.553973293304443,
      "epoch": 0.1375140456953288,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999895198933354,
      "loss": 7.3965,
      "mean_token_accuracy": 0.0919288918375969,
      "num_tokens": 2791183.0,
      "step": 1285
    },
    {
      "entropy": 7.49184365272522,
      "epoch": 0.13804911980309273,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999891476300911,
      "loss": 7.3122,
      "mean_token_accuracy": 0.08812543600797654,
      "num_tokens": 2801928.0,
      "step": 1290
    },
    {
      "entropy": 7.552814531326294,
      "epoch": 0.13858419391085666,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499988768870317,
      "loss": 7.3956,
      "mean_token_accuracy": 0.08465510159730912,
      "num_tokens": 2813415.0,
      "step": 1295
    },
    {
      "entropy": 7.482685708999634,
      "epoch": 0.13911926801862057,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999883836140236,
      "loss": 7.4172,
      "mean_token_accuracy": 0.08735132664442062,
      "num_tokens": 2825288.0,
      "step": 1300
    },
    {
      "entropy": 7.536992979049683,
      "epoch": 0.1396543421263845,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999879918612224,
      "loss": 7.3522,
      "mean_token_accuracy": 0.09331145882606506,
      "num_tokens": 2836298.0,
      "step": 1305
    },
    {
      "entropy": 7.510290431976318,
      "epoch": 0.14018941623414843,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999875936119245,
      "loss": 7.3316,
      "mean_token_accuracy": 0.09056525379419327,
      "num_tokens": 2846367.0,
      "step": 1310
    },
    {
      "entropy": 7.458161878585815,
      "epoch": 0.14072449034191237,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999871888661415,
      "loss": 7.3488,
      "mean_token_accuracy": 0.08874687254428863,
      "num_tokens": 2856832.0,
      "step": 1315
    },
    {
      "entropy": 7.506878137588501,
      "epoch": 0.14125956444967627,
      "grad_norm": 0.84375,
      "learning_rate": 0.000499986777623885,
      "loss": 7.3681,
      "mean_token_accuracy": 0.08353512287139893,
      "num_tokens": 2867954.0,
      "step": 1320
    },
    {
      "entropy": 7.641098499298096,
      "epoch": 0.1417946385574402,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999863598851669,
      "loss": 7.4016,
      "mean_token_accuracy": 0.09097437039017678,
      "num_tokens": 2878714.0,
      "step": 1325
    },
    {
      "entropy": 7.356536340713501,
      "epoch": 0.14232971266520414,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999859356499992,
      "loss": 7.3319,
      "mean_token_accuracy": 0.09417055919766426,
      "num_tokens": 2890697.0,
      "step": 1330
    },
    {
      "entropy": 7.514051151275635,
      "epoch": 0.14286478677296804,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999855049183943,
      "loss": 7.3504,
      "mean_token_accuracy": 0.08786595612764359,
      "num_tokens": 2901468.0,
      "step": 1335
    },
    {
      "entropy": 7.498105907440186,
      "epoch": 0.14339986088073198,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004999850676903646,
      "loss": 7.3144,
      "mean_token_accuracy": 0.09161052480340004,
      "num_tokens": 2913466.0,
      "step": 1340
    },
    {
      "entropy": 7.440569972991943,
      "epoch": 0.1439349349884959,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004999846239659226,
      "loss": 7.2868,
      "mean_token_accuracy": 0.09555015340447426,
      "num_tokens": 2923638.0,
      "step": 1345
    },
    {
      "entropy": 7.524512910842896,
      "epoch": 0.14447000909625984,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999841737450812,
      "loss": 7.3678,
      "mean_token_accuracy": 0.0967521421611309,
      "num_tokens": 2934790.0,
      "step": 1350
    },
    {
      "entropy": 7.4529260158538815,
      "epoch": 0.14500508320402375,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999837170278535,
      "loss": 7.3126,
      "mean_token_accuracy": 0.0890039324760437,
      "num_tokens": 2946320.0,
      "step": 1355
    },
    {
      "entropy": 7.528018951416016,
      "epoch": 0.14554015731178768,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999832538142526,
      "loss": 7.2832,
      "mean_token_accuracy": 0.08959799520671367,
      "num_tokens": 2957077.0,
      "step": 1360
    },
    {
      "entropy": 7.395437908172608,
      "epoch": 0.14607523141955162,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999827841042917,
      "loss": 7.305,
      "mean_token_accuracy": 0.08984916433691978,
      "num_tokens": 2967286.0,
      "step": 1365
    },
    {
      "entropy": 7.475974082946777,
      "epoch": 0.14661030552731552,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999823078979846,
      "loss": 7.342,
      "mean_token_accuracy": 0.0904863603413105,
      "num_tokens": 2979516.0,
      "step": 1370
    },
    {
      "entropy": 7.595507574081421,
      "epoch": 0.14714537963507945,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999818251953449,
      "loss": 7.4628,
      "mean_token_accuracy": 0.08827103711664677,
      "num_tokens": 2991594.0,
      "step": 1375
    },
    {
      "entropy": 7.429067468643188,
      "epoch": 0.1476804537428434,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999813359963867,
      "loss": 7.3428,
      "mean_token_accuracy": 0.08966336324810982,
      "num_tokens": 3002487.0,
      "step": 1380
    },
    {
      "entropy": 7.569780778884888,
      "epoch": 0.14821552785060732,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999808403011241,
      "loss": 7.3603,
      "mean_token_accuracy": 0.09212637171149254,
      "num_tokens": 3013197.0,
      "step": 1385
    },
    {
      "entropy": 7.384968948364258,
      "epoch": 0.14875060195837123,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999803381095712,
      "loss": 7.3808,
      "mean_token_accuracy": 0.08996171355247498,
      "num_tokens": 3024512.0,
      "step": 1390
    },
    {
      "entropy": 7.576485824584961,
      "epoch": 0.14928567606613516,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999798294217428,
      "loss": 7.346,
      "mean_token_accuracy": 0.08968236446380615,
      "num_tokens": 3035621.0,
      "step": 1395
    },
    {
      "entropy": 7.431740236282349,
      "epoch": 0.1498207501738991,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999793142376533,
      "loss": 7.3432,
      "mean_token_accuracy": 0.08854975476861,
      "num_tokens": 3046492.0,
      "step": 1400
    },
    {
      "entropy": 7.440773344039917,
      "epoch": 0.150355824281663,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999787925573177,
      "loss": 7.3546,
      "mean_token_accuracy": 0.09137408435344696,
      "num_tokens": 3057715.0,
      "step": 1405
    },
    {
      "entropy": 7.582657051086426,
      "epoch": 0.15089089838942693,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999782643807513,
      "loss": 7.3135,
      "mean_token_accuracy": 0.08860881105065346,
      "num_tokens": 3068344.0,
      "step": 1410
    },
    {
      "entropy": 7.376219367980957,
      "epoch": 0.15142597249719086,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004999777297079689,
      "loss": 7.2929,
      "mean_token_accuracy": 0.08662425726652145,
      "num_tokens": 3079576.0,
      "step": 1415
    },
    {
      "entropy": 7.48095703125,
      "epoch": 0.1519610466049548,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999771885389863,
      "loss": 7.2626,
      "mean_token_accuracy": 0.09326665848493576,
      "num_tokens": 3090201.0,
      "step": 1420
    },
    {
      "entropy": 7.417229890823364,
      "epoch": 0.1524961207127187,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999766408738189,
      "loss": 7.2349,
      "mean_token_accuracy": 0.09925753399729728,
      "num_tokens": 3099824.0,
      "step": 1425
    },
    {
      "entropy": 7.380178737640381,
      "epoch": 0.15303119482048264,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999760867124827,
      "loss": 7.3052,
      "mean_token_accuracy": 0.09141960963606835,
      "num_tokens": 3109948.0,
      "step": 1430
    },
    {
      "entropy": 7.5375657081604,
      "epoch": 0.15356626892824657,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999755260549937,
      "loss": 7.3051,
      "mean_token_accuracy": 0.09628649353981018,
      "num_tokens": 3121492.0,
      "step": 1435
    },
    {
      "entropy": 7.324652862548828,
      "epoch": 0.15410134303601047,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999749589013677,
      "loss": 7.0668,
      "mean_token_accuracy": 0.11169986054301262,
      "num_tokens": 3131990.0,
      "step": 1440
    },
    {
      "entropy": 7.332792663574219,
      "epoch": 0.1546364171437744,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999743852516217,
      "loss": 7.2535,
      "mean_token_accuracy": 0.09437919408082962,
      "num_tokens": 3142567.0,
      "step": 1445
    },
    {
      "entropy": 7.527145719528198,
      "epoch": 0.15517149125153834,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999738051057717,
      "loss": 7.3012,
      "mean_token_accuracy": 0.09176478385925294,
      "num_tokens": 3153083.0,
      "step": 1450
    },
    {
      "entropy": 7.429166221618653,
      "epoch": 0.15570656535930227,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999732184638347,
      "loss": 7.2866,
      "mean_token_accuracy": 0.09161894842982292,
      "num_tokens": 3163985.0,
      "step": 1455
    },
    {
      "entropy": 7.386832141876221,
      "epoch": 0.15624163946706618,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999726253258278,
      "loss": 7.2769,
      "mean_token_accuracy": 0.09116858392953872,
      "num_tokens": 3175259.0,
      "step": 1460
    },
    {
      "entropy": 7.479492425918579,
      "epoch": 0.1567767135748301,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999720256917678,
      "loss": 7.2395,
      "mean_token_accuracy": 0.09686134830117225,
      "num_tokens": 3185487.0,
      "step": 1465
    },
    {
      "entropy": 7.282340335845947,
      "epoch": 0.15731178768259405,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999714195616723,
      "loss": 7.2391,
      "mean_token_accuracy": 0.09826455116271973,
      "num_tokens": 3196200.0,
      "step": 1470
    },
    {
      "entropy": 7.583243417739868,
      "epoch": 0.15784686179035795,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999708069355586,
      "loss": 7.3415,
      "mean_token_accuracy": 0.08557748645544053,
      "num_tokens": 3207675.0,
      "step": 1475
    },
    {
      "entropy": 7.284139347076416,
      "epoch": 0.15838193589812188,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999701878134445,
      "loss": 7.2096,
      "mean_token_accuracy": 0.09773985520005227,
      "num_tokens": 3218117.0,
      "step": 1480
    },
    {
      "entropy": 7.4924530506134035,
      "epoch": 0.15891701000588582,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999695621953477,
      "loss": 7.2714,
      "mean_token_accuracy": 0.10123906880617142,
      "num_tokens": 3229208.0,
      "step": 1485
    },
    {
      "entropy": 7.381686782836914,
      "epoch": 0.15945208411364975,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999689300812866,
      "loss": 7.1971,
      "mean_token_accuracy": 0.10069452524185181,
      "num_tokens": 3239586.0,
      "step": 1490
    },
    {
      "entropy": 7.443973064422607,
      "epoch": 0.15998715822141366,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004999682914712791,
      "loss": 7.257,
      "mean_token_accuracy": 0.0957643635571003,
      "num_tokens": 3250103.0,
      "step": 1495
    },
    {
      "entropy": 7.41712498664856,
      "epoch": 0.1605222323291776,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004999676463653439,
      "loss": 7.2518,
      "mean_token_accuracy": 0.09973225593566895,
      "num_tokens": 3261844.0,
      "step": 1500
    },
    {
      "entropy": 7.393565559387207,
      "epoch": 0.16105730643694152,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999669947634996,
      "loss": 7.2099,
      "mean_token_accuracy": 0.09588405340909958,
      "num_tokens": 3272073.0,
      "step": 1505
    },
    {
      "entropy": 7.391902828216553,
      "epoch": 0.16159238054470546,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999663366657647,
      "loss": 7.2034,
      "mean_token_accuracy": 0.09728947803378105,
      "num_tokens": 3281725.0,
      "step": 1510
    },
    {
      "entropy": 7.398671960830688,
      "epoch": 0.16212745465246936,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999656720721586,
      "loss": 7.3271,
      "mean_token_accuracy": 0.09106058180332184,
      "num_tokens": 3292708.0,
      "step": 1515
    },
    {
      "entropy": 7.4855554580688475,
      "epoch": 0.1626625287602333,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999650009827004,
      "loss": 7.3079,
      "mean_token_accuracy": 0.094503453373909,
      "num_tokens": 3303548.0,
      "step": 1520
    },
    {
      "entropy": 7.324214076995849,
      "epoch": 0.16319760286799723,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999643233974092,
      "loss": 7.1854,
      "mean_token_accuracy": 0.10153073742985726,
      "num_tokens": 3314261.0,
      "step": 1525
    },
    {
      "entropy": 7.4001932621002195,
      "epoch": 0.16373267697576113,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999636393163049,
      "loss": 7.1979,
      "mean_token_accuracy": 0.09524251073598862,
      "num_tokens": 3324833.0,
      "step": 1530
    },
    {
      "entropy": 7.291304874420166,
      "epoch": 0.16426775108352507,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999629487394071,
      "loss": 7.1051,
      "mean_token_accuracy": 0.10538214668631554,
      "num_tokens": 3334790.0,
      "step": 1535
    },
    {
      "entropy": 7.413803243637085,
      "epoch": 0.164802825191289,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999622516667358,
      "loss": 7.2313,
      "mean_token_accuracy": 0.09709590896964074,
      "num_tokens": 3345115.0,
      "step": 1540
    },
    {
      "entropy": 7.333492660522461,
      "epoch": 0.16533789929905293,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000499961548098311,
      "loss": 7.1207,
      "mean_token_accuracy": 0.10122803598642349,
      "num_tokens": 3356187.0,
      "step": 1545
    },
    {
      "entropy": 7.272494459152222,
      "epoch": 0.16587297340681684,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999608380341533,
      "loss": 7.1359,
      "mean_token_accuracy": 0.10306615009903908,
      "num_tokens": 3367218.0,
      "step": 1550
    },
    {
      "entropy": 7.344357633590699,
      "epoch": 0.16640804751458077,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999601214742829,
      "loss": 7.2837,
      "mean_token_accuracy": 0.09136597141623497,
      "num_tokens": 3377557.0,
      "step": 1555
    },
    {
      "entropy": 7.438312625885009,
      "epoch": 0.1669431216223447,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999593984187206,
      "loss": 7.2729,
      "mean_token_accuracy": 0.09725563302636146,
      "num_tokens": 3387402.0,
      "step": 1560
    },
    {
      "entropy": 7.341435861587525,
      "epoch": 0.1674781957301086,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004999586688674872,
      "loss": 7.1538,
      "mean_token_accuracy": 0.10022934228181839,
      "num_tokens": 3397924.0,
      "step": 1565
    },
    {
      "entropy": 7.39786114692688,
      "epoch": 0.16801326983787254,
      "grad_norm": 0.84375,
      "learning_rate": 0.000499957932820604,
      "loss": 7.2342,
      "mean_token_accuracy": 0.09389038234949112,
      "num_tokens": 3408622.0,
      "step": 1570
    },
    {
      "entropy": 7.419701194763183,
      "epoch": 0.16854834394563648,
      "grad_norm": 0.875,
      "learning_rate": 0.000499957190278092,
      "loss": 7.1859,
      "mean_token_accuracy": 0.09724239408969879,
      "num_tokens": 3419692.0,
      "step": 1575
    },
    {
      "entropy": 7.312865161895752,
      "epoch": 0.1690834180534004,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999564412399728,
      "loss": 7.2472,
      "mean_token_accuracy": 0.093363007158041,
      "num_tokens": 3430986.0,
      "step": 1580
    },
    {
      "entropy": 7.353525066375733,
      "epoch": 0.1696184921611643,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000499955685706268,
      "loss": 7.1695,
      "mean_token_accuracy": 0.096306062489748,
      "num_tokens": 3442211.0,
      "step": 1585
    },
    {
      "entropy": 7.285671186447144,
      "epoch": 0.17015356626892825,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999549236769993,
      "loss": 7.1595,
      "mean_token_accuracy": 0.10178262889385223,
      "num_tokens": 3453585.0,
      "step": 1590
    },
    {
      "entropy": 7.279556846618652,
      "epoch": 0.17068864037669218,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999541551521888,
      "loss": 7.1921,
      "mean_token_accuracy": 0.09866762459278107,
      "num_tokens": 3464156.0,
      "step": 1595
    },
    {
      "entropy": 7.484783601760864,
      "epoch": 0.17122371448445609,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999533801318587,
      "loss": 7.2645,
      "mean_token_accuracy": 0.08954625204205513,
      "num_tokens": 3474853.0,
      "step": 1600
    },
    {
      "entropy": 7.384049606323242,
      "epoch": 0.17175878859222002,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999525986160313,
      "loss": 7.2346,
      "mean_token_accuracy": 0.1001165673136711,
      "num_tokens": 3486841.0,
      "step": 1605
    },
    {
      "entropy": 7.336293649673462,
      "epoch": 0.17229386269998395,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004999518106047293,
      "loss": 7.2373,
      "mean_token_accuracy": 0.09608884081244469,
      "num_tokens": 3498735.0,
      "step": 1610
    },
    {
      "entropy": 7.364135122299194,
      "epoch": 0.17282893680774788,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004999510160979754,
      "loss": 7.2662,
      "mean_token_accuracy": 0.09740655794739723,
      "num_tokens": 3508829.0,
      "step": 1615
    },
    {
      "entropy": 7.306561088562011,
      "epoch": 0.1733640109155118,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999502150957925,
      "loss": 7.189,
      "mean_token_accuracy": 0.09756208583712578,
      "num_tokens": 3519465.0,
      "step": 1620
    },
    {
      "entropy": 7.395194959640503,
      "epoch": 0.17389908502327572,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999494075982037,
      "loss": 7.1279,
      "mean_token_accuracy": 0.10642676651477814,
      "num_tokens": 3530401.0,
      "step": 1625
    },
    {
      "entropy": 7.308092880249023,
      "epoch": 0.17443415913103966,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999485936052324,
      "loss": 7.1994,
      "mean_token_accuracy": 0.09804261550307274,
      "num_tokens": 3541136.0,
      "step": 1630
    },
    {
      "entropy": 7.413141489028931,
      "epoch": 0.17496923323880356,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499947773116902,
      "loss": 7.1996,
      "mean_token_accuracy": 0.09641246423125267,
      "num_tokens": 3552525.0,
      "step": 1635
    },
    {
      "entropy": 7.236075019836425,
      "epoch": 0.1755043073465675,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999469461332365,
      "loss": 7.1501,
      "mean_token_accuracy": 0.10242327377200126,
      "num_tokens": 3563028.0,
      "step": 1640
    },
    {
      "entropy": 7.417467212677002,
      "epoch": 0.17603938145433143,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999461126542592,
      "loss": 7.1158,
      "mean_token_accuracy": 0.10092604756355286,
      "num_tokens": 3573354.0,
      "step": 1645
    },
    {
      "entropy": 7.150946855545044,
      "epoch": 0.17657445556209536,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004999452726799947,
      "loss": 7.104,
      "mean_token_accuracy": 0.09954961612820626,
      "num_tokens": 3585576.0,
      "step": 1650
    },
    {
      "entropy": 7.43422303199768,
      "epoch": 0.17710952966985927,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999444262104671,
      "loss": 7.2255,
      "mean_token_accuracy": 0.09162149503827095,
      "num_tokens": 3596478.0,
      "step": 1655
    },
    {
      "entropy": 7.3217792987823485,
      "epoch": 0.1776446037776232,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999435732457007,
      "loss": 7.1781,
      "mean_token_accuracy": 0.0928413562476635,
      "num_tokens": 3608014.0,
      "step": 1660
    },
    {
      "entropy": 7.3171303272247314,
      "epoch": 0.17817967788538713,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999427137857203,
      "loss": 7.231,
      "mean_token_accuracy": 0.09090021923184395,
      "num_tokens": 3620120.0,
      "step": 1665
    },
    {
      "entropy": 7.364529132843018,
      "epoch": 0.17871475199315104,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999418478305506,
      "loss": 7.1385,
      "mean_token_accuracy": 0.10106266811490058,
      "num_tokens": 3632578.0,
      "step": 1670
    },
    {
      "entropy": 7.233688068389893,
      "epoch": 0.17924982610091497,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999409753802167,
      "loss": 7.1471,
      "mean_token_accuracy": 0.10476156324148178,
      "num_tokens": 3643245.0,
      "step": 1675
    },
    {
      "entropy": 7.329768943786621,
      "epoch": 0.1797849002086789,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004999400964347437,
      "loss": 7.1406,
      "mean_token_accuracy": 0.09999557435512543,
      "num_tokens": 3655832.0,
      "step": 1680
    },
    {
      "entropy": 7.324776554107666,
      "epoch": 0.18031997431644284,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999392109941571,
      "loss": 7.1195,
      "mean_token_accuracy": 0.10120863169431686,
      "num_tokens": 3666566.0,
      "step": 1685
    },
    {
      "entropy": 7.377109861373901,
      "epoch": 0.18085504842420674,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999383190584822,
      "loss": 7.1558,
      "mean_token_accuracy": 0.09975881800055504,
      "num_tokens": 3677572.0,
      "step": 1690
    },
    {
      "entropy": 7.258830785751343,
      "epoch": 0.18139012253197068,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999374206277451,
      "loss": 7.0797,
      "mean_token_accuracy": 0.10004075467586518,
      "num_tokens": 3687116.0,
      "step": 1695
    },
    {
      "entropy": 7.235633325576782,
      "epoch": 0.1819251966397346,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999365157019717,
      "loss": 7.1698,
      "mean_token_accuracy": 0.10185194239020348,
      "num_tokens": 3697961.0,
      "step": 1700
    },
    {
      "entropy": 7.383031129837036,
      "epoch": 0.18246027074749852,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999356042811878,
      "loss": 7.0962,
      "mean_token_accuracy": 0.10583037212491035,
      "num_tokens": 3708834.0,
      "step": 1705
    },
    {
      "entropy": 7.2048969745635985,
      "epoch": 0.18299534485526245,
      "grad_norm": 0.796875,
      "learning_rate": 0.00049993468636542,
      "loss": 7.168,
      "mean_token_accuracy": 0.09880446568131447,
      "num_tokens": 3719835.0,
      "step": 1710
    },
    {
      "entropy": 7.3344207286834715,
      "epoch": 0.18353041896302638,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999337619546948,
      "loss": 7.0614,
      "mean_token_accuracy": 0.10437633171677589,
      "num_tokens": 3730408.0,
      "step": 1715
    },
    {
      "entropy": 7.20396728515625,
      "epoch": 0.18406549307079031,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999328310490387,
      "loss": 7.1371,
      "mean_token_accuracy": 0.10319086164236069,
      "num_tokens": 3741257.0,
      "step": 1720
    },
    {
      "entropy": 7.309871482849121,
      "epoch": 0.18460056717855422,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999318936484789,
      "loss": 7.1195,
      "mean_token_accuracy": 0.10339285656809807,
      "num_tokens": 3752615.0,
      "step": 1725
    },
    {
      "entropy": 7.274545431137085,
      "epoch": 0.18513564128631815,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999309497530422,
      "loss": 7.1536,
      "mean_token_accuracy": 0.09540738239884376,
      "num_tokens": 3763656.0,
      "step": 1730
    },
    {
      "entropy": 7.17682466506958,
      "epoch": 0.1856707153940821,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004999299993627558,
      "loss": 7.1274,
      "mean_token_accuracy": 0.10024873688817024,
      "num_tokens": 3775257.0,
      "step": 1735
    },
    {
      "entropy": 7.275924921035767,
      "epoch": 0.186205789501846,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999290424776475,
      "loss": 7.1085,
      "mean_token_accuracy": 0.10480740368366241,
      "num_tokens": 3786094.0,
      "step": 1740
    },
    {
      "entropy": 7.383003568649292,
      "epoch": 0.18674086360960993,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999280790977445,
      "loss": 7.1905,
      "mean_token_accuracy": 0.09309849962592125,
      "num_tokens": 3796672.0,
      "step": 1745
    },
    {
      "entropy": 7.339628458023071,
      "epoch": 0.18727593771737386,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000499927109223075,
      "loss": 7.1565,
      "mean_token_accuracy": 0.10013005062937737,
      "num_tokens": 3807910.0,
      "step": 1750
    },
    {
      "entropy": 7.22684063911438,
      "epoch": 0.1878110118251378,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004999261328536667,
      "loss": 7.1184,
      "mean_token_accuracy": 0.10545954629778861,
      "num_tokens": 3820792.0,
      "step": 1755
    },
    {
      "entropy": 7.222686910629273,
      "epoch": 0.1883460859329017,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004999251499895479,
      "loss": 7.0348,
      "mean_token_accuracy": 0.10182796642184258,
      "num_tokens": 3831126.0,
      "step": 1760
    },
    {
      "entropy": 7.290640163421631,
      "epoch": 0.18888116004066563,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004999241606307471,
      "loss": 7.1585,
      "mean_token_accuracy": 0.0980623982846737,
      "num_tokens": 3841923.0,
      "step": 1765
    },
    {
      "entropy": 7.286150789260864,
      "epoch": 0.18941623414842956,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999231647772927,
      "loss": 7.0679,
      "mean_token_accuracy": 0.10358999967575074,
      "num_tokens": 3852504.0,
      "step": 1770
    },
    {
      "entropy": 7.19472336769104,
      "epoch": 0.1899513082561935,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999221624292136,
      "loss": 7.0699,
      "mean_token_accuracy": 0.10635367557406425,
      "num_tokens": 3862761.0,
      "step": 1775
    },
    {
      "entropy": 7.289284896850586,
      "epoch": 0.1904863823639574,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999211535865385,
      "loss": 7.0759,
      "mean_token_accuracy": 0.10067537054419518,
      "num_tokens": 3873662.0,
      "step": 1780
    },
    {
      "entropy": 7.168746614456177,
      "epoch": 0.19102145647172133,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999201382492969,
      "loss": 7.0188,
      "mean_token_accuracy": 0.10261877253651619,
      "num_tokens": 3884835.0,
      "step": 1785
    },
    {
      "entropy": 7.28413405418396,
      "epoch": 0.19155653057948527,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999191164175178,
      "loss": 7.1592,
      "mean_token_accuracy": 0.09962226003408432,
      "num_tokens": 3895073.0,
      "step": 1790
    },
    {
      "entropy": 7.337110424041748,
      "epoch": 0.19209160468724917,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999180880912309,
      "loss": 7.1158,
      "mean_token_accuracy": 0.10559302195906639,
      "num_tokens": 3905440.0,
      "step": 1795
    },
    {
      "entropy": 7.21136155128479,
      "epoch": 0.1926266787950131,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999170532704657,
      "loss": 7.1183,
      "mean_token_accuracy": 0.09755382165312768,
      "num_tokens": 3917135.0,
      "step": 1800
    },
    {
      "entropy": 7.304487466812134,
      "epoch": 0.19316175290277704,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0004999160119552523,
      "loss": 7.1225,
      "mean_token_accuracy": 0.10279463827610016,
      "num_tokens": 3928247.0,
      "step": 1805
    },
    {
      "entropy": 7.196545743942261,
      "epoch": 0.19369682701054097,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999149641456206,
      "loss": 7.1084,
      "mean_token_accuracy": 0.10402424335479736,
      "num_tokens": 3938609.0,
      "step": 1810
    },
    {
      "entropy": 7.373271131515503,
      "epoch": 0.19423190111830488,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499913909841601,
      "loss": 7.0897,
      "mean_token_accuracy": 0.10455763563513756,
      "num_tokens": 3949221.0,
      "step": 1815
    },
    {
      "entropy": 7.1703612327575685,
      "epoch": 0.1947669752260688,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999128490432238,
      "loss": 7.0609,
      "mean_token_accuracy": 0.09710687473416328,
      "num_tokens": 3959475.0,
      "step": 1820
    },
    {
      "entropy": 7.2145233154296875,
      "epoch": 0.19530204933383274,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999117817505196,
      "loss": 7.0592,
      "mean_token_accuracy": 0.1138175070285797,
      "num_tokens": 3969638.0,
      "step": 1825
    },
    {
      "entropy": 7.178833293914795,
      "epoch": 0.19583712344159665,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999107079635194,
      "loss": 7.1232,
      "mean_token_accuracy": 0.10179126933217049,
      "num_tokens": 3980543.0,
      "step": 1830
    },
    {
      "entropy": 7.44319224357605,
      "epoch": 0.19637219754936058,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000499909627682254,
      "loss": 7.1773,
      "mean_token_accuracy": 0.08765889480710029,
      "num_tokens": 3993518.0,
      "step": 1835
    },
    {
      "entropy": 7.2637087345123295,
      "epoch": 0.19690727165712452,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999085409067549,
      "loss": 7.1375,
      "mean_token_accuracy": 0.10476782023906708,
      "num_tokens": 4002940.0,
      "step": 1840
    },
    {
      "entropy": 7.169889783859253,
      "epoch": 0.19744234576488845,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999074476370531,
      "loss": 7.0939,
      "mean_token_accuracy": 0.09844619929790496,
      "num_tokens": 4013982.0,
      "step": 1845
    },
    {
      "entropy": 7.24789080619812,
      "epoch": 0.19797741987265235,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004999063478731805,
      "loss": 7.0485,
      "mean_token_accuracy": 0.10713553130626678,
      "num_tokens": 4025607.0,
      "step": 1850
    },
    {
      "entropy": 7.222923517227173,
      "epoch": 0.1985124939804163,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004999052416151687,
      "loss": 7.1148,
      "mean_token_accuracy": 0.10136601626873017,
      "num_tokens": 4037467.0,
      "step": 1855
    },
    {
      "entropy": 7.270085382461548,
      "epoch": 0.19904756808818022,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004999041288630496,
      "loss": 7.1147,
      "mean_token_accuracy": 0.10222559496760368,
      "num_tokens": 4049146.0,
      "step": 1860
    },
    {
      "entropy": 7.300042200088501,
      "epoch": 0.19958264219594413,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004999030096168553,
      "loss": 7.1579,
      "mean_token_accuracy": 0.09921248778700828,
      "num_tokens": 4060500.0,
      "step": 1865
    },
    {
      "entropy": 7.314883184432984,
      "epoch": 0.20011771630370806,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999018838766182,
      "loss": 7.0989,
      "mean_token_accuracy": 0.0998796634376049,
      "num_tokens": 4070553.0,
      "step": 1870
    },
    {
      "entropy": 7.253989124298096,
      "epoch": 0.200652790411472,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499900751642371,
      "loss": 7.0327,
      "mean_token_accuracy": 0.11456628292798995,
      "num_tokens": 4079653.0,
      "step": 1875
    },
    {
      "entropy": 7.165084648132324,
      "epoch": 0.20118786451923593,
      "grad_norm": 0.875,
      "learning_rate": 0.0004998996129141461,
      "loss": 7.1197,
      "mean_token_accuracy": 0.09987151175737381,
      "num_tokens": 4089912.0,
      "step": 1880
    },
    {
      "entropy": 7.321857929229736,
      "epoch": 0.20172293862699983,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998984676919764,
      "loss": 7.0693,
      "mean_token_accuracy": 0.09889922812581062,
      "num_tokens": 4100577.0,
      "step": 1885
    },
    {
      "entropy": 7.251675939559936,
      "epoch": 0.20225801273476376,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004998973159758952,
      "loss": 7.116,
      "mean_token_accuracy": 0.1035578764975071,
      "num_tokens": 4111752.0,
      "step": 1890
    },
    {
      "entropy": 7.180023431777954,
      "epoch": 0.2027930868425277,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998961577659355,
      "loss": 7.0633,
      "mean_token_accuracy": 0.10196540728211403,
      "num_tokens": 4122616.0,
      "step": 1895
    },
    {
      "entropy": 7.318381118774414,
      "epoch": 0.2033281609502916,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004998949930621309,
      "loss": 7.1603,
      "mean_token_accuracy": 0.10213067010045052,
      "num_tokens": 4134275.0,
      "step": 1900
    },
    {
      "entropy": 7.190880680084229,
      "epoch": 0.20386323505805554,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998938218645149,
      "loss": 7.0717,
      "mean_token_accuracy": 0.10233089625835419,
      "num_tokens": 4146412.0,
      "step": 1905
    },
    {
      "entropy": 7.185058450698852,
      "epoch": 0.20439830916581947,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998926441731215,
      "loss": 7.0594,
      "mean_token_accuracy": 0.10094341784715652,
      "num_tokens": 4157592.0,
      "step": 1910
    },
    {
      "entropy": 7.2327782154083256,
      "epoch": 0.2049333832735834,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998914599879845,
      "loss": 7.0388,
      "mean_token_accuracy": 0.10173192545771599,
      "num_tokens": 4168637.0,
      "step": 1915
    },
    {
      "entropy": 7.216303873062134,
      "epoch": 0.2054684573813473,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004998902693091382,
      "loss": 7.073,
      "mean_token_accuracy": 0.10192683935165406,
      "num_tokens": 4179951.0,
      "step": 1920
    },
    {
      "entropy": 7.214565849304199,
      "epoch": 0.20600353148911124,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004998890721366171,
      "loss": 6.9886,
      "mean_token_accuracy": 0.1056469924747944,
      "num_tokens": 4189895.0,
      "step": 1925
    },
    {
      "entropy": 7.218944406509399,
      "epoch": 0.20653860559687517,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998878684704556,
      "loss": 7.0203,
      "mean_token_accuracy": 0.10886224210262299,
      "num_tokens": 4199796.0,
      "step": 1930
    },
    {
      "entropy": 7.145756816864013,
      "epoch": 0.20707367970463908,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998866583106884,
      "loss": 7.0497,
      "mean_token_accuracy": 0.1005377635359764,
      "num_tokens": 4210428.0,
      "step": 1935
    },
    {
      "entropy": 7.31018853187561,
      "epoch": 0.207608753812403,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998854416573506,
      "loss": 7.1162,
      "mean_token_accuracy": 0.10497400760650635,
      "num_tokens": 4221515.0,
      "step": 1940
    },
    {
      "entropy": 7.217876052856445,
      "epoch": 0.20814382792016695,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998842185104772,
      "loss": 7.1294,
      "mean_token_accuracy": 0.10152154713869095,
      "num_tokens": 4232580.0,
      "step": 1945
    },
    {
      "entropy": 7.19381012916565,
      "epoch": 0.20867890202793088,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998829888701036,
      "loss": 7.0433,
      "mean_token_accuracy": 0.10803714469075203,
      "num_tokens": 4242871.0,
      "step": 1950
    },
    {
      "entropy": 7.283778476715088,
      "epoch": 0.20921397613569478,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998817527362654,
      "loss": 7.0692,
      "mean_token_accuracy": 0.09783539399504662,
      "num_tokens": 4253419.0,
      "step": 1955
    },
    {
      "entropy": 7.182916593551636,
      "epoch": 0.20974905024345872,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004998805101089982,
      "loss": 7.0529,
      "mean_token_accuracy": 0.10560425892472267,
      "num_tokens": 4264327.0,
      "step": 1960
    },
    {
      "entropy": 7.240354061126709,
      "epoch": 0.21028412435122265,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004998792609883378,
      "loss": 7.0356,
      "mean_token_accuracy": 0.10485936850309371,
      "num_tokens": 4274568.0,
      "step": 1965
    },
    {
      "entropy": 7.218569707870484,
      "epoch": 0.21081919845898656,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998780053743203,
      "loss": 7.0283,
      "mean_token_accuracy": 0.10580286756157875,
      "num_tokens": 4284895.0,
      "step": 1970
    },
    {
      "entropy": 7.211511325836182,
      "epoch": 0.2113542725667505,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004998767432669822,
      "loss": 7.0579,
      "mean_token_accuracy": 0.10149188414216041,
      "num_tokens": 4297258.0,
      "step": 1975
    },
    {
      "entropy": 7.180872201919556,
      "epoch": 0.21188934667451442,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004998754746663595,
      "loss": 6.9933,
      "mean_token_accuracy": 0.1065586656332016,
      "num_tokens": 4308307.0,
      "step": 1980
    },
    {
      "entropy": 7.1800261497497555,
      "epoch": 0.21242442078227836,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004998741995724892,
      "loss": 7.0546,
      "mean_token_accuracy": 0.10103240609169006,
      "num_tokens": 4318801.0,
      "step": 1985
    },
    {
      "entropy": 7.196444845199585,
      "epoch": 0.21295949489004226,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004998729179854079,
      "loss": 7.0325,
      "mean_token_accuracy": 0.10279674157500267,
      "num_tokens": 4329531.0,
      "step": 1990
    },
    {
      "entropy": 7.1866326332092285,
      "epoch": 0.2134945689978062,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004998716299051527,
      "loss": 7.0133,
      "mean_token_accuracy": 0.10350802689790725,
      "num_tokens": 4341203.0,
      "step": 1995
    },
    {
      "entropy": 7.124855422973633,
      "epoch": 0.21402964310557013,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004998703353317609,
      "loss": 7.0584,
      "mean_token_accuracy": 0.10162648260593414,
      "num_tokens": 4352426.0,
      "step": 2000
    },
    {
      "entropy": 7.259735202789306,
      "epoch": 0.21456471721333403,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004998690342652697,
      "loss": 7.0121,
      "mean_token_accuracy": 0.10889948382973672,
      "num_tokens": 4363413.0,
      "step": 2005
    },
    {
      "entropy": 7.1536116123199465,
      "epoch": 0.21509979132109797,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004998677267057166,
      "loss": 6.9458,
      "mean_token_accuracy": 0.10910924375057221,
      "num_tokens": 4373470.0,
      "step": 2010
    },
    {
      "entropy": 7.114294338226318,
      "epoch": 0.2156348654288619,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998664126531397,
      "loss": 7.0155,
      "mean_token_accuracy": 0.10767255127429962,
      "num_tokens": 4384438.0,
      "step": 2015
    },
    {
      "entropy": 7.260488796234131,
      "epoch": 0.21616993953662583,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998650921075766,
      "loss": 6.9928,
      "mean_token_accuracy": 0.10748248249292373,
      "num_tokens": 4395100.0,
      "step": 2020
    },
    {
      "entropy": 7.089138031005859,
      "epoch": 0.21670501364438974,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004998637650690656,
      "loss": 7.1018,
      "mean_token_accuracy": 0.09796615988016129,
      "num_tokens": 4407131.0,
      "step": 2025
    },
    {
      "entropy": 7.218321037292481,
      "epoch": 0.21724008775215367,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499862431537645,
      "loss": 6.9445,
      "mean_token_accuracy": 0.10602100789546967,
      "num_tokens": 4417701.0,
      "step": 2030
    },
    {
      "entropy": 7.156389379501343,
      "epoch": 0.2177751618599176,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998610915133533,
      "loss": 7.0019,
      "mean_token_accuracy": 0.10421578660607338,
      "num_tokens": 4429309.0,
      "step": 2035
    },
    {
      "entropy": 7.151696300506591,
      "epoch": 0.21831023596768154,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004998597449962292,
      "loss": 7.0604,
      "mean_token_accuracy": 0.1074062243103981,
      "num_tokens": 4440850.0,
      "step": 2040
    },
    {
      "entropy": 7.173864507675171,
      "epoch": 0.21884531007544544,
      "grad_norm": 0.875,
      "learning_rate": 0.0004998583919863115,
      "loss": 7.0455,
      "mean_token_accuracy": 0.10184828117489815,
      "num_tokens": 4451995.0,
      "step": 2045
    },
    {
      "entropy": 7.095217037200928,
      "epoch": 0.21938038418320938,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998570324836393,
      "loss": 6.9646,
      "mean_token_accuracy": 0.10822436287999153,
      "num_tokens": 4461916.0,
      "step": 2050
    },
    {
      "entropy": 7.188471364974975,
      "epoch": 0.2199154582909733,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499855666488252,
      "loss": 7.0417,
      "mean_token_accuracy": 0.10385469049215316,
      "num_tokens": 4471603.0,
      "step": 2055
    },
    {
      "entropy": 7.260391330718994,
      "epoch": 0.22045053239873721,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998542940001889,
      "loss": 7.0733,
      "mean_token_accuracy": 0.10464386790990829,
      "num_tokens": 4483166.0,
      "step": 2060
    },
    {
      "entropy": 7.165126323699951,
      "epoch": 0.22098560650650115,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004998529150194895,
      "loss": 6.984,
      "mean_token_accuracy": 0.10515503585338593,
      "num_tokens": 4493824.0,
      "step": 2065
    },
    {
      "entropy": 7.1778943061828615,
      "epoch": 0.22152068061426508,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004998515295461939,
      "loss": 6.8879,
      "mean_token_accuracy": 0.12077882513403893,
      "num_tokens": 4505674.0,
      "step": 2070
    },
    {
      "entropy": 7.045021867752075,
      "epoch": 0.22205575472202901,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000499850137580342,
      "loss": 6.9944,
      "mean_token_accuracy": 0.11151223108172417,
      "num_tokens": 4516516.0,
      "step": 2075
    },
    {
      "entropy": 7.183830261230469,
      "epoch": 0.22259082882979292,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998487391219739,
      "loss": 6.9922,
      "mean_token_accuracy": 0.10238008573651314,
      "num_tokens": 4527811.0,
      "step": 2080
    },
    {
      "entropy": 7.124439287185669,
      "epoch": 0.22312590293755685,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004998473341711301,
      "loss": 6.9376,
      "mean_token_accuracy": 0.10194609165191651,
      "num_tokens": 4538988.0,
      "step": 2085
    },
    {
      "entropy": 7.129032325744629,
      "epoch": 0.22366097704532079,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998459227278512,
      "loss": 6.964,
      "mean_token_accuracy": 0.10870476886630058,
      "num_tokens": 4550049.0,
      "step": 2090
    },
    {
      "entropy": 7.269771909713745,
      "epoch": 0.2241960511530847,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998445047921777,
      "loss": 7.0811,
      "mean_token_accuracy": 0.1076191745698452,
      "num_tokens": 4559784.0,
      "step": 2095
    },
    {
      "entropy": 7.1074059009552,
      "epoch": 0.22473112526084862,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998430803641509,
      "loss": 6.9183,
      "mean_token_accuracy": 0.10604787841439248,
      "num_tokens": 4569982.0,
      "step": 2100
    },
    {
      "entropy": 7.101476860046387,
      "epoch": 0.22526619936861256,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004998416494438118,
      "loss": 6.9299,
      "mean_token_accuracy": 0.10904891490936279,
      "num_tokens": 4581117.0,
      "step": 2105
    },
    {
      "entropy": 7.105316066741944,
      "epoch": 0.2258012734763765,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998402120312014,
      "loss": 6.9953,
      "mean_token_accuracy": 0.11073005571961403,
      "num_tokens": 4592012.0,
      "step": 2110
    },
    {
      "entropy": 7.180981636047363,
      "epoch": 0.2263363475841404,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004998387681263616,
      "loss": 7.0385,
      "mean_token_accuracy": 0.10344724059104919,
      "num_tokens": 4602591.0,
      "step": 2115
    },
    {
      "entropy": 7.147944641113281,
      "epoch": 0.22687142169190433,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000499837317729334,
      "loss": 6.9648,
      "mean_token_accuracy": 0.10446924120187759,
      "num_tokens": 4613418.0,
      "step": 2120
    },
    {
      "entropy": 7.088868284225464,
      "epoch": 0.22740649579966826,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004998358608401603,
      "loss": 6.8946,
      "mean_token_accuracy": 0.11337370723485947,
      "num_tokens": 4625156.0,
      "step": 2125
    },
    {
      "entropy": 7.165902519226075,
      "epoch": 0.22794156990743217,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000499834397458883,
      "loss": 6.9344,
      "mean_token_accuracy": 0.11145322620868683,
      "num_tokens": 4634523.0,
      "step": 2130
    },
    {
      "entropy": 7.076514863967896,
      "epoch": 0.2284766440151961,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998329275855438,
      "loss": 6.9174,
      "mean_token_accuracy": 0.11302619948983192,
      "num_tokens": 4646292.0,
      "step": 2135
    },
    {
      "entropy": 7.122889423370362,
      "epoch": 0.22901171812296003,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004998314512201855,
      "loss": 6.9004,
      "mean_token_accuracy": 0.10689050704240799,
      "num_tokens": 4656417.0,
      "step": 2140
    },
    {
      "entropy": 7.06037015914917,
      "epoch": 0.22954679223072397,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004998299683628507,
      "loss": 6.912,
      "mean_token_accuracy": 0.11160982474684715,
      "num_tokens": 4666933.0,
      "step": 2145
    },
    {
      "entropy": 7.187684774398804,
      "epoch": 0.23008186633848787,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004998284790135821,
      "loss": 7.1033,
      "mean_token_accuracy": 0.1016832061111927,
      "num_tokens": 4678482.0,
      "step": 2150
    },
    {
      "entropy": 7.139988660812378,
      "epoch": 0.2306169404462518,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998269831724228,
      "loss": 6.9424,
      "mean_token_accuracy": 0.10893533229827881,
      "num_tokens": 4688454.0,
      "step": 2155
    },
    {
      "entropy": 7.114541387557983,
      "epoch": 0.23115201455401574,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998254808394159,
      "loss": 6.9693,
      "mean_token_accuracy": 0.1090779609978199,
      "num_tokens": 4699545.0,
      "step": 2160
    },
    {
      "entropy": 7.173770380020142,
      "epoch": 0.23168708866177964,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0004998239720146048,
      "loss": 6.9818,
      "mean_token_accuracy": 0.10480888038873673,
      "num_tokens": 4711149.0,
      "step": 2165
    },
    {
      "entropy": 7.0449871063232425,
      "epoch": 0.23222216276954358,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004998224566980332,
      "loss": 6.9846,
      "mean_token_accuracy": 0.10382255315780639,
      "num_tokens": 4722651.0,
      "step": 2170
    },
    {
      "entropy": 7.14964804649353,
      "epoch": 0.2327572368773075,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998209348897447,
      "loss": 6.9818,
      "mean_token_accuracy": 0.10008153393864631,
      "num_tokens": 4733883.0,
      "step": 2175
    },
    {
      "entropy": 7.267318153381348,
      "epoch": 0.23329231098507144,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998194065897833,
      "loss": 6.9546,
      "mean_token_accuracy": 0.10759458690881729,
      "num_tokens": 4744517.0,
      "step": 2180
    },
    {
      "entropy": 7.039386034011841,
      "epoch": 0.23382738509283535,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499817871798193,
      "loss": 6.9613,
      "mean_token_accuracy": 0.11343310549855232,
      "num_tokens": 4755481.0,
      "step": 2185
    },
    {
      "entropy": 7.088238620758057,
      "epoch": 0.23436245920059928,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998163305150185,
      "loss": 6.8813,
      "mean_token_accuracy": 0.1131195604801178,
      "num_tokens": 4765535.0,
      "step": 2190
    },
    {
      "entropy": 7.1311625957489015,
      "epoch": 0.23489753330836322,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998147827403038,
      "loss": 7.0102,
      "mean_token_accuracy": 0.1072376735508442,
      "num_tokens": 4777045.0,
      "step": 2195
    },
    {
      "entropy": 7.156279706954956,
      "epoch": 0.23543260741612712,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000499813228474094,
      "loss": 6.8885,
      "mean_token_accuracy": 0.11066517904400826,
      "num_tokens": 4787958.0,
      "step": 2200
    },
    {
      "entropy": 7.07044153213501,
      "epoch": 0.23596768152389105,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004998116677164338,
      "loss": 6.9279,
      "mean_token_accuracy": 0.10897202789783478,
      "num_tokens": 4798502.0,
      "step": 2205
    },
    {
      "entropy": 7.007950210571289,
      "epoch": 0.236502755631655,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998101004673682,
      "loss": 6.9386,
      "mean_token_accuracy": 0.11299166306853295,
      "num_tokens": 4809179.0,
      "step": 2210
    },
    {
      "entropy": 7.225791692733765,
      "epoch": 0.23703782973941892,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998085267269426,
      "loss": 7.0021,
      "mean_token_accuracy": 0.1013013631105423,
      "num_tokens": 4819803.0,
      "step": 2215
    },
    {
      "entropy": 7.156044769287109,
      "epoch": 0.23757290384718283,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004998069464952025,
      "loss": 6.9661,
      "mean_token_accuracy": 0.10483337342739105,
      "num_tokens": 4831226.0,
      "step": 2220
    },
    {
      "entropy": 7.1146551132202145,
      "epoch": 0.23810797795494676,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004998053597721932,
      "loss": 6.9805,
      "mean_token_accuracy": 0.10689718723297119,
      "num_tokens": 4842168.0,
      "step": 2225
    },
    {
      "entropy": 7.112945604324341,
      "epoch": 0.2386430520627107,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004998037665579608,
      "loss": 6.9683,
      "mean_token_accuracy": 0.10687630698084831,
      "num_tokens": 4852655.0,
      "step": 2230
    },
    {
      "entropy": 7.031319570541382,
      "epoch": 0.2391781261704746,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998021668525513,
      "loss": 6.8648,
      "mean_token_accuracy": 0.11213752254843712,
      "num_tokens": 4862164.0,
      "step": 2235
    },
    {
      "entropy": 7.138062334060669,
      "epoch": 0.23971320027823853,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998005606560107,
      "loss": 6.9837,
      "mean_token_accuracy": 0.10962075069546699,
      "num_tokens": 4874233.0,
      "step": 2240
    },
    {
      "entropy": 7.145175266265869,
      "epoch": 0.24024827438600246,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004997989479683856,
      "loss": 6.9847,
      "mean_token_accuracy": 0.1075660951435566,
      "num_tokens": 4885948.0,
      "step": 2245
    },
    {
      "entropy": 7.035645341873169,
      "epoch": 0.2407833484937664,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004997973287897224,
      "loss": 6.9566,
      "mean_token_accuracy": 0.10865945667028427,
      "num_tokens": 4897128.0,
      "step": 2250
    },
    {
      "entropy": 7.135949563980103,
      "epoch": 0.2413184226015303,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004997957031200681,
      "loss": 6.9369,
      "mean_token_accuracy": 0.11560432091355324,
      "num_tokens": 4907699.0,
      "step": 2255
    },
    {
      "entropy": 7.00885157585144,
      "epoch": 0.24185349670929424,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004997940709594692,
      "loss": 6.8547,
      "mean_token_accuracy": 0.11122522950172424,
      "num_tokens": 4918772.0,
      "step": 2260
    },
    {
      "entropy": 7.157874059677124,
      "epoch": 0.24238857081705817,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997924323079733,
      "loss": 6.9092,
      "mean_token_accuracy": 0.10879954397678375,
      "num_tokens": 4930046.0,
      "step": 2265
    },
    {
      "entropy": 7.033572673797607,
      "epoch": 0.24292364492482207,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997907871656273,
      "loss": 6.9361,
      "mean_token_accuracy": 0.11303869113326073,
      "num_tokens": 4940609.0,
      "step": 2270
    },
    {
      "entropy": 7.085981941223144,
      "epoch": 0.243458719032586,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997891355324791,
      "loss": 6.9537,
      "mean_token_accuracy": 0.11050518602132797,
      "num_tokens": 4951339.0,
      "step": 2275
    },
    {
      "entropy": 7.1102917194366455,
      "epoch": 0.24399379314034994,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004997874774085761,
      "loss": 6.8694,
      "mean_token_accuracy": 0.10475466623902321,
      "num_tokens": 4961083.0,
      "step": 2280
    },
    {
      "entropy": 7.02321424484253,
      "epoch": 0.24452886724811387,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997858127939662,
      "loss": 6.8815,
      "mean_token_accuracy": 0.11015179753303528,
      "num_tokens": 4971399.0,
      "step": 2285
    },
    {
      "entropy": 7.133182764053345,
      "epoch": 0.24506394135587778,
      "grad_norm": 0.875,
      "learning_rate": 0.0004997841416886976,
      "loss": 6.9557,
      "mean_token_accuracy": 0.10723848342895508,
      "num_tokens": 4982429.0,
      "step": 2290
    },
    {
      "entropy": 7.134330749511719,
      "epoch": 0.2455990154636417,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997824640928186,
      "loss": 6.9981,
      "mean_token_accuracy": 0.10663792341947556,
      "num_tokens": 4993706.0,
      "step": 2295
    },
    {
      "entropy": 7.106071710586548,
      "epoch": 0.24613408957140565,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004997807800063774,
      "loss": 6.8969,
      "mean_token_accuracy": 0.11038667410612106,
      "num_tokens": 5005411.0,
      "step": 2300
    },
    {
      "entropy": 7.113573980331421,
      "epoch": 0.24666916367916958,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499779089429423,
      "loss": 7.0192,
      "mean_token_accuracy": 0.10511812567710876,
      "num_tokens": 5016011.0,
      "step": 2305
    },
    {
      "entropy": 7.082342147827148,
      "epoch": 0.24720423778693348,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004997773923620037,
      "loss": 6.9766,
      "mean_token_accuracy": 0.1077098935842514,
      "num_tokens": 5027841.0,
      "step": 2310
    },
    {
      "entropy": 7.156000328063965,
      "epoch": 0.24773931189469742,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004997756888041689,
      "loss": 6.9674,
      "mean_token_accuracy": 0.09967386648058892,
      "num_tokens": 5038640.0,
      "step": 2315
    },
    {
      "entropy": 7.031879425048828,
      "epoch": 0.24827438600246135,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004997739787559677,
      "loss": 6.9064,
      "mean_token_accuracy": 0.1101887047290802,
      "num_tokens": 5050423.0,
      "step": 2320
    },
    {
      "entropy": 7.084950113296509,
      "epoch": 0.24880946011022526,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997722622174494,
      "loss": 6.9261,
      "mean_token_accuracy": 0.10869491174817085,
      "num_tokens": 5060654.0,
      "step": 2325
    },
    {
      "entropy": 7.073810243606568,
      "epoch": 0.2493445342179892,
      "grad_norm": 0.875,
      "learning_rate": 0.0004997705391886636,
      "loss": 6.8532,
      "mean_token_accuracy": 0.1153720997273922,
      "num_tokens": 5070165.0,
      "step": 2330
    },
    {
      "entropy": 6.990339374542236,
      "epoch": 0.24987960832575312,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004997688096696601,
      "loss": 6.8889,
      "mean_token_accuracy": 0.10511518865823746,
      "num_tokens": 5080937.0,
      "step": 2335
    },
    {
      "entropy": 7.040760135650634,
      "epoch": 0.25041468243351706,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004997670736604888,
      "loss": 6.8536,
      "mean_token_accuracy": 0.11363203302025796,
      "num_tokens": 5090955.0,
      "step": 2340
    },
    {
      "entropy": 7.100010633468628,
      "epoch": 0.250949756541281,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004997653311611998,
      "loss": 6.9896,
      "mean_token_accuracy": 0.1013368934392929,
      "num_tokens": 5102201.0,
      "step": 2345
    },
    {
      "entropy": 7.11283597946167,
      "epoch": 0.25148483064904487,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997635821718435,
      "loss": 6.912,
      "mean_token_accuracy": 0.11232623681426049,
      "num_tokens": 5112235.0,
      "step": 2350
    },
    {
      "entropy": 7.029507255554199,
      "epoch": 0.2520199047568088,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997618266924704,
      "loss": 6.7832,
      "mean_token_accuracy": 0.10930138230323791,
      "num_tokens": 5122360.0,
      "step": 2355
    },
    {
      "entropy": 7.053276777267456,
      "epoch": 0.25255497886457273,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499760064723131,
      "loss": 6.9599,
      "mean_token_accuracy": 0.10386712178587913,
      "num_tokens": 5133084.0,
      "step": 2360
    },
    {
      "entropy": 7.001993083953858,
      "epoch": 0.25309005297233667,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004997582962638764,
      "loss": 6.9092,
      "mean_token_accuracy": 0.10329946726560593,
      "num_tokens": 5144495.0,
      "step": 2365
    },
    {
      "entropy": 7.14462685585022,
      "epoch": 0.2536251270801006,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004997565213147575,
      "loss": 6.9045,
      "mean_token_accuracy": 0.11057875901460648,
      "num_tokens": 5154824.0,
      "step": 2370
    },
    {
      "entropy": 7.04837121963501,
      "epoch": 0.25416020118786453,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997547398758257,
      "loss": 6.8976,
      "mean_token_accuracy": 0.10639405995607376,
      "num_tokens": 5165061.0,
      "step": 2375
    },
    {
      "entropy": 7.022197437286377,
      "epoch": 0.25469527529562846,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004997529519471323,
      "loss": 6.8471,
      "mean_token_accuracy": 0.11550139263272285,
      "num_tokens": 5174964.0,
      "step": 2380
    },
    {
      "entropy": 7.051410436630249,
      "epoch": 0.25523034940339234,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997511575287291,
      "loss": 6.8892,
      "mean_token_accuracy": 0.11107081025838852,
      "num_tokens": 5185458.0,
      "step": 2385
    },
    {
      "entropy": 7.052774667739868,
      "epoch": 0.2557654235111563,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997493566206677,
      "loss": 6.878,
      "mean_token_accuracy": 0.11274964585900307,
      "num_tokens": 5196193.0,
      "step": 2390
    },
    {
      "entropy": 7.087007999420166,
      "epoch": 0.2563004976189202,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004997475492230003,
      "loss": 7.0016,
      "mean_token_accuracy": 0.10122420638799667,
      "num_tokens": 5208541.0,
      "step": 2395
    },
    {
      "entropy": 7.037209749221802,
      "epoch": 0.25683557172668414,
      "grad_norm": 0.828125,
      "learning_rate": 0.000499745735335779,
      "loss": 6.8283,
      "mean_token_accuracy": 0.11095813065767288,
      "num_tokens": 5218567.0,
      "step": 2400
    },
    {
      "entropy": 7.078734445571899,
      "epoch": 0.2573706458344481,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004997439149590561,
      "loss": 6.9304,
      "mean_token_accuracy": 0.10828692242503166,
      "num_tokens": 5228293.0,
      "step": 2405
    },
    {
      "entropy": 7.043195676803589,
      "epoch": 0.257905719942212,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997420880928843,
      "loss": 6.8715,
      "mean_token_accuracy": 0.11050004214048385,
      "num_tokens": 5238880.0,
      "step": 2410
    },
    {
      "entropy": 7.082090950012207,
      "epoch": 0.25844079404997594,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004997402547373162,
      "loss": 6.9731,
      "mean_token_accuracy": 0.10441731810569763,
      "num_tokens": 5250104.0,
      "step": 2415
    },
    {
      "entropy": 7.139094257354737,
      "epoch": 0.2589758681577399,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004997384148924049,
      "loss": 6.9951,
      "mean_token_accuracy": 0.10500225946307182,
      "num_tokens": 5260355.0,
      "step": 2420
    },
    {
      "entropy": 7.020048427581787,
      "epoch": 0.25951094226550375,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004997365685582035,
      "loss": 6.9181,
      "mean_token_accuracy": 0.10640754848718643,
      "num_tokens": 5272117.0,
      "step": 2425
    },
    {
      "entropy": 7.079926252365112,
      "epoch": 0.2600460163732677,
      "grad_norm": 0.765625,
      "learning_rate": 0.0004997347157347651,
      "loss": 6.9946,
      "mean_token_accuracy": 0.10754604786634445,
      "num_tokens": 5284583.0,
      "step": 2430
    },
    {
      "entropy": 7.123588037490845,
      "epoch": 0.2605810904810316,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004997328564221435,
      "loss": 6.7935,
      "mean_token_accuracy": 0.12066915556788445,
      "num_tokens": 5295846.0,
      "step": 2435
    },
    {
      "entropy": 6.995146322250366,
      "epoch": 0.26111616458879555,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004997309906203922,
      "loss": 6.9324,
      "mean_token_accuracy": 0.10952212065458297,
      "num_tokens": 5307802.0,
      "step": 2440
    },
    {
      "entropy": 7.15656852722168,
      "epoch": 0.2616512386965595,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004997291183295652,
      "loss": 6.8757,
      "mean_token_accuracy": 0.10969269722700119,
      "num_tokens": 5318263.0,
      "step": 2445
    },
    {
      "entropy": 7.028390216827392,
      "epoch": 0.2621863128043234,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004997272395497164,
      "loss": 6.9303,
      "mean_token_accuracy": 0.10968295708298684,
      "num_tokens": 5328555.0,
      "step": 2450
    },
    {
      "entropy": 6.9868518829345705,
      "epoch": 0.26272138691208735,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004997253542809002,
      "loss": 6.8927,
      "mean_token_accuracy": 0.10809791460633278,
      "num_tokens": 5339954.0,
      "step": 2455
    },
    {
      "entropy": 7.193512296676635,
      "epoch": 0.26325646101985123,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004997234625231709,
      "loss": 6.9324,
      "mean_token_accuracy": 0.10953526422381402,
      "num_tokens": 5350919.0,
      "step": 2460
    },
    {
      "entropy": 6.973828077316284,
      "epoch": 0.26379153512761516,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004997215642765832,
      "loss": 6.895,
      "mean_token_accuracy": 0.11250809505581856,
      "num_tokens": 5361844.0,
      "step": 2465
    },
    {
      "entropy": 7.0807849884033205,
      "epoch": 0.2643266092353791,
      "grad_norm": 0.828125,
      "learning_rate": 0.000499719659541192,
      "loss": 6.9167,
      "mean_token_accuracy": 0.10835950300097466,
      "num_tokens": 5372026.0,
      "step": 2470
    },
    {
      "entropy": 7.093844032287597,
      "epoch": 0.26486168334314303,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004997177483170521,
      "loss": 6.9599,
      "mean_token_accuracy": 0.11532194167375565,
      "num_tokens": 5382965.0,
      "step": 2475
    },
    {
      "entropy": 7.004912328720093,
      "epoch": 0.26539675745090696,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004997158306042188,
      "loss": 6.9202,
      "mean_token_accuracy": 0.11223710849881172,
      "num_tokens": 5393738.0,
      "step": 2480
    },
    {
      "entropy": 7.040785694122315,
      "epoch": 0.2659318315586709,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004997139064027476,
      "loss": 6.7822,
      "mean_token_accuracy": 0.1194936640560627,
      "num_tokens": 5404059.0,
      "step": 2485
    },
    {
      "entropy": 6.9767382621765135,
      "epoch": 0.26646690566643483,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004997119757126938,
      "loss": 6.8286,
      "mean_token_accuracy": 0.11358326822519302,
      "num_tokens": 5414152.0,
      "step": 2490
    },
    {
      "entropy": 7.081306219100952,
      "epoch": 0.2670019797741987,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004997100385341133,
      "loss": 7.0047,
      "mean_token_accuracy": 0.11303253397345543,
      "num_tokens": 5425227.0,
      "step": 2495
    },
    {
      "entropy": 7.00633659362793,
      "epoch": 0.26753705388196264,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997080948670621,
      "loss": 6.8416,
      "mean_token_accuracy": 0.11532852947711944,
      "num_tokens": 5435716.0,
      "step": 2500
    },
    {
      "entropy": 7.059450101852417,
      "epoch": 0.26807212798972657,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004997061447115962,
      "loss": 6.8843,
      "mean_token_accuracy": 0.10835974663496017,
      "num_tokens": 5446519.0,
      "step": 2505
    },
    {
      "entropy": 7.0279497623443605,
      "epoch": 0.2686072020974905,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004997041880677719,
      "loss": 6.888,
      "mean_token_accuracy": 0.11044454723596572,
      "num_tokens": 5457496.0,
      "step": 2510
    },
    {
      "entropy": 6.935314702987671,
      "epoch": 0.26914227620525444,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997022249356457,
      "loss": 6.7806,
      "mean_token_accuracy": 0.11524100303649902,
      "num_tokens": 5469059.0,
      "step": 2515
    },
    {
      "entropy": 7.044953393936157,
      "epoch": 0.26967735031301837,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997002553152745,
      "loss": 6.8806,
      "mean_token_accuracy": 0.11304009705781937,
      "num_tokens": 5480324.0,
      "step": 2520
    },
    {
      "entropy": 7.024924993515015,
      "epoch": 0.2702124244207823,
      "grad_norm": 0.875,
      "learning_rate": 0.0004996982792067148,
      "loss": 6.889,
      "mean_token_accuracy": 0.10333864018321037,
      "num_tokens": 5491513.0,
      "step": 2525
    },
    {
      "entropy": 6.982390928268432,
      "epoch": 0.2707474985285462,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000499696296610024,
      "loss": 6.8084,
      "mean_token_accuracy": 0.1128513678908348,
      "num_tokens": 5502414.0,
      "step": 2530
    },
    {
      "entropy": 6.966418361663818,
      "epoch": 0.2712825726363101,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004996943075252592,
      "loss": 6.944,
      "mean_token_accuracy": 0.11141985207796097,
      "num_tokens": 5512962.0,
      "step": 2535
    },
    {
      "entropy": 7.015233278274536,
      "epoch": 0.27181764674407405,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004996923119524777,
      "loss": 6.8026,
      "mean_token_accuracy": 0.11136919856071473,
      "num_tokens": 5524648.0,
      "step": 2540
    },
    {
      "entropy": 6.970895910263062,
      "epoch": 0.272352720851838,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004996903098917375,
      "loss": 6.8846,
      "mean_token_accuracy": 0.10550413206219673,
      "num_tokens": 5536158.0,
      "step": 2545
    },
    {
      "entropy": 6.9682692527771,
      "epoch": 0.2728877949596019,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499688301343096,
      "loss": 6.7946,
      "mean_token_accuracy": 0.11154654324054718,
      "num_tokens": 5547374.0,
      "step": 2550
    },
    {
      "entropy": 7.100108480453491,
      "epoch": 0.27342286906736585,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004996862863066114,
      "loss": 6.8914,
      "mean_token_accuracy": 0.11287438571453094,
      "num_tokens": 5559212.0,
      "step": 2555
    },
    {
      "entropy": 6.999144172668457,
      "epoch": 0.2739579431751298,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996842647823419,
      "loss": 6.8469,
      "mean_token_accuracy": 0.11468030512332916,
      "num_tokens": 5569302.0,
      "step": 2560
    },
    {
      "entropy": 6.963889455795288,
      "epoch": 0.27449301728289366,
      "grad_norm": 0.875,
      "learning_rate": 0.0004996822367703458,
      "loss": 6.8464,
      "mean_token_accuracy": 0.10833697766065598,
      "num_tokens": 5581688.0,
      "step": 2565
    },
    {
      "entropy": 7.065867328643799,
      "epoch": 0.2750280913906576,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996802022706817,
      "loss": 6.7811,
      "mean_token_accuracy": 0.11610291451215744,
      "num_tokens": 5591376.0,
      "step": 2570
    },
    {
      "entropy": 6.911547708511352,
      "epoch": 0.2755631654984215,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996781612834083,
      "loss": 6.8206,
      "mean_token_accuracy": 0.11811210587620735,
      "num_tokens": 5601510.0,
      "step": 2575
    },
    {
      "entropy": 7.062700414657593,
      "epoch": 0.27609823960618546,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996761138085846,
      "loss": 6.9144,
      "mean_token_accuracy": 0.10678515806794167,
      "num_tokens": 5611436.0,
      "step": 2580
    },
    {
      "entropy": 7.025034236907959,
      "epoch": 0.2766333137139494,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996740598462697,
      "loss": 6.9045,
      "mean_token_accuracy": 0.10823121890425683,
      "num_tokens": 5620470.0,
      "step": 2585
    },
    {
      "entropy": 6.927712631225586,
      "epoch": 0.2771683878217133,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996719993965229,
      "loss": 6.803,
      "mean_token_accuracy": 0.1145929142832756,
      "num_tokens": 5630740.0,
      "step": 2590
    },
    {
      "entropy": 6.955496549606323,
      "epoch": 0.27770346192947726,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004996699324594036,
      "loss": 6.8039,
      "mean_token_accuracy": 0.1161543108522892,
      "num_tokens": 5641138.0,
      "step": 2595
    },
    {
      "entropy": 7.075690460205078,
      "epoch": 0.27823853603724114,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004996678590349717,
      "loss": 6.9325,
      "mean_token_accuracy": 0.11235370934009552,
      "num_tokens": 5652474.0,
      "step": 2600
    },
    {
      "entropy": 6.996727037429809,
      "epoch": 0.27877361014500507,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004996657791232868,
      "loss": 6.8411,
      "mean_token_accuracy": 0.11515266224741935,
      "num_tokens": 5663178.0,
      "step": 2605
    },
    {
      "entropy": 7.003625869750977,
      "epoch": 0.279308684252769,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996636927244092,
      "loss": 6.8484,
      "mean_token_accuracy": 0.10914000719785691,
      "num_tokens": 5673565.0,
      "step": 2610
    },
    {
      "entropy": 7.075772523880005,
      "epoch": 0.27984375836053293,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000499661599838399,
      "loss": 6.9003,
      "mean_token_accuracy": 0.10649769231677056,
      "num_tokens": 5684651.0,
      "step": 2615
    },
    {
      "entropy": 6.911233377456665,
      "epoch": 0.28037883246829687,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996595004653167,
      "loss": 6.7904,
      "mean_token_accuracy": 0.11583374589681625,
      "num_tokens": 5696081.0,
      "step": 2620
    },
    {
      "entropy": 6.993151569366455,
      "epoch": 0.2809139065760608,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996573946052229,
      "loss": 6.8294,
      "mean_token_accuracy": 0.1125739298760891,
      "num_tokens": 5706830.0,
      "step": 2625
    },
    {
      "entropy": 6.982648468017578,
      "epoch": 0.28144898068382473,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996552822581783,
      "loss": 6.8391,
      "mean_token_accuracy": 0.1167409248650074,
      "num_tokens": 5717261.0,
      "step": 2630
    },
    {
      "entropy": 7.002830171585083,
      "epoch": 0.2819840547915886,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499653163424244,
      "loss": 6.8305,
      "mean_token_accuracy": 0.11221889033913612,
      "num_tokens": 5727280.0,
      "step": 2635
    },
    {
      "entropy": 6.9843268394470215,
      "epoch": 0.28251912889935255,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004996510381034814,
      "loss": 6.8746,
      "mean_token_accuracy": 0.10521704554557801,
      "num_tokens": 5737485.0,
      "step": 2640
    },
    {
      "entropy": 7.011574792861938,
      "epoch": 0.2830542030071165,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004996489062959515,
      "loss": 6.8336,
      "mean_token_accuracy": 0.11000596359372139,
      "num_tokens": 5747683.0,
      "step": 2645
    },
    {
      "entropy": 6.939910221099853,
      "epoch": 0.2835892771148804,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004996467680017159,
      "loss": 6.9016,
      "mean_token_accuracy": 0.10526101067662239,
      "num_tokens": 5758148.0,
      "step": 2650
    },
    {
      "entropy": 6.925260782241821,
      "epoch": 0.28412435122264434,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004996446232208365,
      "loss": 6.8311,
      "mean_token_accuracy": 0.11652650088071823,
      "num_tokens": 5769109.0,
      "step": 2655
    },
    {
      "entropy": 7.011774253845215,
      "epoch": 0.2846594253304083,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004996424719533753,
      "loss": 6.8467,
      "mean_token_accuracy": 0.11380974426865578,
      "num_tokens": 5780344.0,
      "step": 2660
    },
    {
      "entropy": 6.951174783706665,
      "epoch": 0.2851944994381722,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004996403141993941,
      "loss": 6.6666,
      "mean_token_accuracy": 0.12389900684356689,
      "num_tokens": 5790167.0,
      "step": 2665
    },
    {
      "entropy": 6.946682929992676,
      "epoch": 0.2857295735459361,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004996381499589555,
      "loss": 6.8417,
      "mean_token_accuracy": 0.11240337193012237,
      "num_tokens": 5800076.0,
      "step": 2670
    },
    {
      "entropy": 7.035910606384277,
      "epoch": 0.2862646476537,
      "grad_norm": 0.875,
      "learning_rate": 0.0004996359792321219,
      "loss": 6.7903,
      "mean_token_accuracy": 0.11802778467535972,
      "num_tokens": 5810547.0,
      "step": 2675
    },
    {
      "entropy": 6.917967891693115,
      "epoch": 0.28679972176146395,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499633802018956,
      "loss": 6.8173,
      "mean_token_accuracy": 0.11811300665140152,
      "num_tokens": 5821210.0,
      "step": 2680
    },
    {
      "entropy": 6.960421514511109,
      "epoch": 0.2873347958692279,
      "grad_norm": 0.875,
      "learning_rate": 0.0004996316183195204,
      "loss": 6.8489,
      "mean_token_accuracy": 0.11026589795947075,
      "num_tokens": 5832361.0,
      "step": 2685
    },
    {
      "entropy": 7.026253318786621,
      "epoch": 0.2878698699769918,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004996294281338786,
      "loss": 6.8443,
      "mean_token_accuracy": 0.10899497270584106,
      "num_tokens": 5843242.0,
      "step": 2690
    },
    {
      "entropy": 7.042272329330444,
      "epoch": 0.28840494408475575,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004996272314620936,
      "loss": 6.8897,
      "mean_token_accuracy": 0.10638803094625474,
      "num_tokens": 5855163.0,
      "step": 2695
    },
    {
      "entropy": 6.922289705276489,
      "epoch": 0.2889400181925197,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004996250283042288,
      "loss": 6.8265,
      "mean_token_accuracy": 0.1135579526424408,
      "num_tokens": 5866908.0,
      "step": 2700
    },
    {
      "entropy": 7.018845796585083,
      "epoch": 0.28947509230028357,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499622818660348,
      "loss": 6.8806,
      "mean_token_accuracy": 0.10942182540893555,
      "num_tokens": 5877342.0,
      "step": 2705
    },
    {
      "entropy": 6.949248218536377,
      "epoch": 0.2900101664080475,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004996206025305147,
      "loss": 6.8217,
      "mean_token_accuracy": 0.10982293486595154,
      "num_tokens": 5887960.0,
      "step": 2710
    },
    {
      "entropy": 6.947164630889892,
      "epoch": 0.29054524051581143,
      "grad_norm": 0.875,
      "learning_rate": 0.0004996183799147932,
      "loss": 6.7974,
      "mean_token_accuracy": 0.11249916031956672,
      "num_tokens": 5899887.0,
      "step": 2715
    },
    {
      "entropy": 7.010842323303223,
      "epoch": 0.29108031462357536,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996161508132475,
      "loss": 6.8142,
      "mean_token_accuracy": 0.10648874044418336,
      "num_tokens": 5911022.0,
      "step": 2720
    },
    {
      "entropy": 7.063547801971436,
      "epoch": 0.2916153887313393,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499613915225942,
      "loss": 6.8805,
      "mean_token_accuracy": 0.11401300653815269,
      "num_tokens": 5922780.0,
      "step": 2725
    },
    {
      "entropy": 6.957791137695312,
      "epoch": 0.29215046283910323,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996116731529412,
      "loss": 6.9079,
      "mean_token_accuracy": 0.11033277809619904,
      "num_tokens": 5934112.0,
      "step": 2730
    },
    {
      "entropy": 6.973647975921631,
      "epoch": 0.29268553694686716,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00049960942459431,
      "loss": 6.8414,
      "mean_token_accuracy": 0.10922785475850105,
      "num_tokens": 5944386.0,
      "step": 2735
    },
    {
      "entropy": 6.913372564315796,
      "epoch": 0.29322061105463104,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996071695501132,
      "loss": 6.7672,
      "mean_token_accuracy": 0.11577581316232681,
      "num_tokens": 5955552.0,
      "step": 2740
    },
    {
      "entropy": 6.972257375717163,
      "epoch": 0.293755685162395,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004996049080204159,
      "loss": 6.7788,
      "mean_token_accuracy": 0.11552264243364334,
      "num_tokens": 5966790.0,
      "step": 2745
    },
    {
      "entropy": 6.993252277374268,
      "epoch": 0.2942907592701589,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004996026400052835,
      "loss": 6.7831,
      "mean_token_accuracy": 0.11271130293607712,
      "num_tokens": 5977704.0,
      "step": 2750
    },
    {
      "entropy": 6.870125818252563,
      "epoch": 0.29482583337792284,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004996003655047814,
      "loss": 6.8028,
      "mean_token_accuracy": 0.11466444432735443,
      "num_tokens": 5989221.0,
      "step": 2755
    },
    {
      "entropy": 6.9888458251953125,
      "epoch": 0.2953609074856868,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004995980845189753,
      "loss": 6.834,
      "mean_token_accuracy": 0.1083465002477169,
      "num_tokens": 6000906.0,
      "step": 2760
    },
    {
      "entropy": 6.996977138519287,
      "epoch": 0.2958959815934507,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004995957970479312,
      "loss": 6.8506,
      "mean_token_accuracy": 0.11252145394682884,
      "num_tokens": 6012039.0,
      "step": 2765
    },
    {
      "entropy": 6.935248565673828,
      "epoch": 0.29643105570121464,
      "grad_norm": 0.73828125,
      "learning_rate": 0.0004995935030917148,
      "loss": 6.7372,
      "mean_token_accuracy": 0.11361273676156998,
      "num_tokens": 6024383.0,
      "step": 2770
    },
    {
      "entropy": 7.046666049957276,
      "epoch": 0.2969661298089785,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004995912026503926,
      "loss": 6.8044,
      "mean_token_accuracy": 0.1105383276939392,
      "num_tokens": 6034904.0,
      "step": 2775
    },
    {
      "entropy": 6.906943941116333,
      "epoch": 0.29750120391674245,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499588895724031,
      "loss": 6.8156,
      "mean_token_accuracy": 0.1101146712899208,
      "num_tokens": 6044449.0,
      "step": 2780
    },
    {
      "entropy": 6.984871864318848,
      "epoch": 0.2980362780245064,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004995865823126968,
      "loss": 6.8266,
      "mean_token_accuracy": 0.11390623450279236,
      "num_tokens": 6054465.0,
      "step": 2785
    },
    {
      "entropy": 6.957597541809082,
      "epoch": 0.2985713521322703,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004995842624164564,
      "loss": 6.8403,
      "mean_token_accuracy": 0.11375648602843284,
      "num_tokens": 6065539.0,
      "step": 2790
    },
    {
      "entropy": 6.96545295715332,
      "epoch": 0.29910642624003425,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499581936035377,
      "loss": 6.8012,
      "mean_token_accuracy": 0.11368539929389954,
      "num_tokens": 6076135.0,
      "step": 2795
    },
    {
      "entropy": 6.972835540771484,
      "epoch": 0.2996415003477982,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004995796031695257,
      "loss": 6.8043,
      "mean_token_accuracy": 0.11199271976947785,
      "num_tokens": 6086604.0,
      "step": 2800
    },
    {
      "entropy": 6.936370754241944,
      "epoch": 0.3001765744555621,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00049957726381897,
      "loss": 6.8275,
      "mean_token_accuracy": 0.11501410380005836,
      "num_tokens": 6095672.0,
      "step": 2805
    },
    {
      "entropy": 6.921350908279419,
      "epoch": 0.300711648563326,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004995749179837773,
      "loss": 6.7088,
      "mean_token_accuracy": 0.12310401126742362,
      "num_tokens": 6105652.0,
      "step": 2810
    },
    {
      "entropy": 6.871649312973022,
      "epoch": 0.30124672267108993,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004995725656640155,
      "loss": 6.7453,
      "mean_token_accuracy": 0.11668358147144317,
      "num_tokens": 6116930.0,
      "step": 2815
    },
    {
      "entropy": 6.905279350280762,
      "epoch": 0.30178179677885386,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004995702068597524,
      "loss": 6.7557,
      "mean_token_accuracy": 0.11802282258868217,
      "num_tokens": 6126659.0,
      "step": 2820
    },
    {
      "entropy": 6.936028718948364,
      "epoch": 0.3023168708866178,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004995678415710561,
      "loss": 6.7841,
      "mean_token_accuracy": 0.11382660195231438,
      "num_tokens": 6137227.0,
      "step": 2825
    },
    {
      "entropy": 7.028936529159546,
      "epoch": 0.3028519449943817,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000499565469797995,
      "loss": 6.8442,
      "mean_token_accuracy": 0.11068922504782677,
      "num_tokens": 6147800.0,
      "step": 2830
    },
    {
      "entropy": 6.8801487445831295,
      "epoch": 0.30338701910214566,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004995630915406374,
      "loss": 6.6811,
      "mean_token_accuracy": 0.1217758022248745,
      "num_tokens": 6157671.0,
      "step": 2835
    },
    {
      "entropy": 6.933578777313232,
      "epoch": 0.3039220932099096,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004995607067990521,
      "loss": 6.8252,
      "mean_token_accuracy": 0.11223353296518326,
      "num_tokens": 6168651.0,
      "step": 2840
    },
    {
      "entropy": 6.9156494140625,
      "epoch": 0.30445716731767347,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499558315573308,
      "loss": 6.7999,
      "mean_token_accuracy": 0.1103752687573433,
      "num_tokens": 6179925.0,
      "step": 2845
    },
    {
      "entropy": 6.876695585250855,
      "epoch": 0.3049922414254374,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995559178634741,
      "loss": 6.7933,
      "mean_token_accuracy": 0.1192070983350277,
      "num_tokens": 6190541.0,
      "step": 2850
    },
    {
      "entropy": 7.041800832748413,
      "epoch": 0.30552731553320134,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004995535136696196,
      "loss": 6.838,
      "mean_token_accuracy": 0.10699132680892945,
      "num_tokens": 6201491.0,
      "step": 2855
    },
    {
      "entropy": 6.847650003433228,
      "epoch": 0.30606238964096527,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000499551102991814,
      "loss": 6.6434,
      "mean_token_accuracy": 0.12001704573631286,
      "num_tokens": 6212443.0,
      "step": 2860
    },
    {
      "entropy": 6.952008056640625,
      "epoch": 0.3065974637487292,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004995486858301268,
      "loss": 6.8451,
      "mean_token_accuracy": 0.10546251088380813,
      "num_tokens": 6222623.0,
      "step": 2865
    },
    {
      "entropy": 6.9607843399047855,
      "epoch": 0.30713253785649314,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004995462621846279,
      "loss": 6.7437,
      "mean_token_accuracy": 0.11431733518838882,
      "num_tokens": 6233712.0,
      "step": 2870
    },
    {
      "entropy": 6.921725225448609,
      "epoch": 0.30766761196425707,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995438320553871,
      "loss": 6.8108,
      "mean_token_accuracy": 0.11184044554829597,
      "num_tokens": 6244254.0,
      "step": 2875
    },
    {
      "entropy": 6.933862113952637,
      "epoch": 0.30820268607202095,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004995413954424748,
      "loss": 6.7041,
      "mean_token_accuracy": 0.12413104623556137,
      "num_tokens": 6254202.0,
      "step": 2880
    },
    {
      "entropy": 6.896720361709595,
      "epoch": 0.3087377601797849,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995389523459612,
      "loss": 6.7875,
      "mean_token_accuracy": 0.10941583663225174,
      "num_tokens": 6264496.0,
      "step": 2885
    },
    {
      "entropy": 6.89367413520813,
      "epoch": 0.3092728342875488,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000499536502765917,
      "loss": 6.7968,
      "mean_token_accuracy": 0.11605312377214431,
      "num_tokens": 6275835.0,
      "step": 2890
    },
    {
      "entropy": 6.961062669754028,
      "epoch": 0.30980790839531275,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995340467024128,
      "loss": 6.7421,
      "mean_token_accuracy": 0.11403903961181641,
      "num_tokens": 6286805.0,
      "step": 2895
    },
    {
      "entropy": 6.9369104385375975,
      "epoch": 0.3103429825030767,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004995315841555195,
      "loss": 6.7534,
      "mean_token_accuracy": 0.10973872393369674,
      "num_tokens": 6298444.0,
      "step": 2900
    },
    {
      "entropy": 6.821942901611328,
      "epoch": 0.3108780566108406,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004995291151253083,
      "loss": 6.6866,
      "mean_token_accuracy": 0.11732677444815635,
      "num_tokens": 6308259.0,
      "step": 2905
    },
    {
      "entropy": 6.903902912139893,
      "epoch": 0.31141313071860455,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004995266396118505,
      "loss": 6.762,
      "mean_token_accuracy": 0.10594131126999855,
      "num_tokens": 6320058.0,
      "step": 2910
    },
    {
      "entropy": 6.915698528289795,
      "epoch": 0.3119482048263684,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004995241576152174,
      "loss": 6.7316,
      "mean_token_accuracy": 0.10760388672351837,
      "num_tokens": 6331203.0,
      "step": 2915
    },
    {
      "entropy": 6.955277776718139,
      "epoch": 0.31248327893413236,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499521669135481,
      "loss": 6.8324,
      "mean_token_accuracy": 0.11208058670163154,
      "num_tokens": 6342027.0,
      "step": 2920
    },
    {
      "entropy": 6.928550672531128,
      "epoch": 0.3130183530418963,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995191741727128,
      "loss": 6.7431,
      "mean_token_accuracy": 0.11125858277082443,
      "num_tokens": 6352580.0,
      "step": 2925
    },
    {
      "entropy": 6.969546937942505,
      "epoch": 0.3135534271496602,
      "grad_norm": 0.76171875,
      "learning_rate": 0.000499516672726985,
      "loss": 6.7556,
      "mean_token_accuracy": 0.1169343538582325,
      "num_tokens": 6363997.0,
      "step": 2930
    },
    {
      "entropy": 6.8640326976776125,
      "epoch": 0.31408850125742416,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004995141647983699,
      "loss": 6.8027,
      "mean_token_accuracy": 0.11516904607415199,
      "num_tokens": 6375284.0,
      "step": 2935
    },
    {
      "entropy": 6.883784198760987,
      "epoch": 0.3146235753651881,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004995116503869398,
      "loss": 6.6913,
      "mean_token_accuracy": 0.12319560199975968,
      "num_tokens": 6386698.0,
      "step": 2940
    },
    {
      "entropy": 6.950956916809082,
      "epoch": 0.315158649472952,
      "grad_norm": 0.875,
      "learning_rate": 0.0004995091294927673,
      "loss": 6.7806,
      "mean_token_accuracy": 0.11270102486014366,
      "num_tokens": 6397237.0,
      "step": 2945
    },
    {
      "entropy": 6.972240686416626,
      "epoch": 0.3156937235807159,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995066021159253,
      "loss": 6.8847,
      "mean_token_accuracy": 0.11340105310082435,
      "num_tokens": 6408266.0,
      "step": 2950
    },
    {
      "entropy": 6.8899555683135985,
      "epoch": 0.31622879768847983,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004995040682564867,
      "loss": 6.775,
      "mean_token_accuracy": 0.11872687861323357,
      "num_tokens": 6419563.0,
      "step": 2955
    },
    {
      "entropy": 6.864490795135498,
      "epoch": 0.31676387179624377,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004995015279145247,
      "loss": 6.7399,
      "mean_token_accuracy": 0.11859864667057991,
      "num_tokens": 6430220.0,
      "step": 2960
    },
    {
      "entropy": 6.928368997573853,
      "epoch": 0.3172989459040077,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994989810901126,
      "loss": 6.7558,
      "mean_token_accuracy": 0.11230956315994263,
      "num_tokens": 6440157.0,
      "step": 2965
    },
    {
      "entropy": 6.9050641536712645,
      "epoch": 0.31783402001177163,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994964277833239,
      "loss": 6.7907,
      "mean_token_accuracy": 0.1221705548465252,
      "num_tokens": 6449511.0,
      "step": 2970
    },
    {
      "entropy": 6.942772340774536,
      "epoch": 0.31836909411953557,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004994938679942324,
      "loss": 6.7836,
      "mean_token_accuracy": 0.11035550981760026,
      "num_tokens": 6460608.0,
      "step": 2975
    },
    {
      "entropy": 6.934708499908448,
      "epoch": 0.3189041682272995,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004994913017229121,
      "loss": 6.77,
      "mean_token_accuracy": 0.11715537756681442,
      "num_tokens": 6472737.0,
      "step": 2980
    },
    {
      "entropy": 6.8332374572753904,
      "epoch": 0.31943924233506343,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994887289694371,
      "loss": 6.6674,
      "mean_token_accuracy": 0.11837019473314285,
      "num_tokens": 6482993.0,
      "step": 2985
    },
    {
      "entropy": 7.0178844928741455,
      "epoch": 0.3199743164428273,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004994861497338814,
      "loss": 6.8735,
      "mean_token_accuracy": 0.11568538472056389,
      "num_tokens": 6494530.0,
      "step": 2990
    },
    {
      "entropy": 6.916510963439942,
      "epoch": 0.32050939055059124,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004994835640163198,
      "loss": 6.6973,
      "mean_token_accuracy": 0.1126147136092186,
      "num_tokens": 6504691.0,
      "step": 2995
    },
    {
      "entropy": 6.894684362411499,
      "epoch": 0.3210444646583552,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004994809718168267,
      "loss": 6.7383,
      "mean_token_accuracy": 0.11528538763523102,
      "num_tokens": 6514924.0,
      "step": 3000
    },
    {
      "epoch": 0.3210444646583552,
      "eval_entropy": 6.626076072759705,
      "eval_loss": 6.841334342956543,
      "eval_mean_token_accuracy": 0.11807519927460228,
      "eval_num_tokens": 6514924.0,
      "eval_runtime": 17.6886,
      "eval_samples_per_second": 1677.858,
      "eval_steps_per_second": 209.739,
      "step": 3000
    },
    {
      "entropy": 6.8141203880310055,
      "epoch": 0.3215795387661191,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994783731354771,
      "loss": 6.708,
      "mean_token_accuracy": 0.11827568635344506,
      "num_tokens": 6526097.0,
      "step": 3005
    },
    {
      "entropy": 6.951326608657837,
      "epoch": 0.32211461287388304,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994757679723461,
      "loss": 6.7658,
      "mean_token_accuracy": 0.1161842055618763,
      "num_tokens": 6536384.0,
      "step": 3010
    },
    {
      "entropy": 6.974445486068726,
      "epoch": 0.322649686981647,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004994731563275088,
      "loss": 6.8276,
      "mean_token_accuracy": 0.10699201673269272,
      "num_tokens": 6547161.0,
      "step": 3015
    },
    {
      "entropy": 6.791270780563354,
      "epoch": 0.3231847610894109,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004994705382010406,
      "loss": 6.7301,
      "mean_token_accuracy": 0.11511025950312614,
      "num_tokens": 6556751.0,
      "step": 3020
    },
    {
      "entropy": 6.882776594161987,
      "epoch": 0.3237198351971748,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004994679135930174,
      "loss": 6.7881,
      "mean_token_accuracy": 0.11802244856953621,
      "num_tokens": 6567655.0,
      "step": 3025
    },
    {
      "entropy": 6.978914546966553,
      "epoch": 0.3242549093049387,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994652825035144,
      "loss": 6.8168,
      "mean_token_accuracy": 0.1150478482246399,
      "num_tokens": 6579127.0,
      "step": 3030
    },
    {
      "entropy": 6.948042249679565,
      "epoch": 0.32478998341270265,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994626449326081,
      "loss": 6.7982,
      "mean_token_accuracy": 0.11559739410877228,
      "num_tokens": 6590779.0,
      "step": 3035
    },
    {
      "entropy": 6.8800136089324955,
      "epoch": 0.3253250575204666,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004994600008803745,
      "loss": 6.7028,
      "mean_token_accuracy": 0.12257652655243874,
      "num_tokens": 6600849.0,
      "step": 3040
    },
    {
      "entropy": 6.915249013900757,
      "epoch": 0.3258601316282305,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004994573503468899,
      "loss": 6.6975,
      "mean_token_accuracy": 0.12107903882861137,
      "num_tokens": 6610633.0,
      "step": 3045
    },
    {
      "entropy": 6.825544548034668,
      "epoch": 0.32639520573599445,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004994546933322308,
      "loss": 6.7466,
      "mean_token_accuracy": 0.11225979700684548,
      "num_tokens": 6621327.0,
      "step": 3050
    },
    {
      "entropy": 6.873090410232544,
      "epoch": 0.3269302798437584,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499452029836474,
      "loss": 6.689,
      "mean_token_accuracy": 0.12065511345863342,
      "num_tokens": 6632267.0,
      "step": 3055
    },
    {
      "entropy": 6.888823795318603,
      "epoch": 0.32746535395152226,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004994493598596963,
      "loss": 6.7075,
      "mean_token_accuracy": 0.11505150720477104,
      "num_tokens": 6642714.0,
      "step": 3060
    },
    {
      "entropy": 6.890615558624267,
      "epoch": 0.3280004280592862,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004994466834019749,
      "loss": 6.7309,
      "mean_token_accuracy": 0.12258474081754685,
      "num_tokens": 6652353.0,
      "step": 3065
    },
    {
      "entropy": 6.854852104187012,
      "epoch": 0.32853550216705013,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994440004633871,
      "loss": 6.6994,
      "mean_token_accuracy": 0.11629214882850647,
      "num_tokens": 6663133.0,
      "step": 3070
    },
    {
      "entropy": 6.802253580093383,
      "epoch": 0.32907057627481406,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994413110440103,
      "loss": 6.6619,
      "mean_token_accuracy": 0.12232538983225823,
      "num_tokens": 6673984.0,
      "step": 3075
    },
    {
      "entropy": 6.876217079162598,
      "epoch": 0.329605650382578,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004994386151439222,
      "loss": 6.7476,
      "mean_token_accuracy": 0.11251990124583244,
      "num_tokens": 6686171.0,
      "step": 3080
    },
    {
      "entropy": 6.9876720905303955,
      "epoch": 0.33014072449034193,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004994359127632006,
      "loss": 6.7986,
      "mean_token_accuracy": 0.11241496056318283,
      "num_tokens": 6696609.0,
      "step": 3085
    },
    {
      "entropy": 6.918059539794922,
      "epoch": 0.33067579859810586,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994332039019235,
      "loss": 6.7256,
      "mean_token_accuracy": 0.11809317544102668,
      "num_tokens": 6706188.0,
      "step": 3090
    },
    {
      "entropy": 6.764942979812622,
      "epoch": 0.33121087270586974,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994304885601694,
      "loss": 6.7204,
      "mean_token_accuracy": 0.11531962379813195,
      "num_tokens": 6716276.0,
      "step": 3095
    },
    {
      "entropy": 6.8081374168396,
      "epoch": 0.3317459468136337,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004994277667380162,
      "loss": 6.6702,
      "mean_token_accuracy": 0.12529247477650643,
      "num_tokens": 6727176.0,
      "step": 3100
    },
    {
      "entropy": 6.876701164245605,
      "epoch": 0.3322810209213976,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004994250384355429,
      "loss": 6.761,
      "mean_token_accuracy": 0.11975758746266366,
      "num_tokens": 6737768.0,
      "step": 3105
    },
    {
      "entropy": 6.85945611000061,
      "epoch": 0.33281609502916154,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004994223036528282,
      "loss": 6.6926,
      "mean_token_accuracy": 0.11432832181453705,
      "num_tokens": 6747962.0,
      "step": 3110
    },
    {
      "entropy": 6.838385963439942,
      "epoch": 0.3333511691369255,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499419562389951,
      "loss": 6.6981,
      "mean_token_accuracy": 0.12037034779787063,
      "num_tokens": 6758400.0,
      "step": 3115
    },
    {
      "entropy": 6.864870595932007,
      "epoch": 0.3338862432446894,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994168146469904,
      "loss": 6.6933,
      "mean_token_accuracy": 0.12071826979517937,
      "num_tokens": 6768855.0,
      "step": 3120
    },
    {
      "entropy": 6.902755212783814,
      "epoch": 0.33442131735245334,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004994140604240259,
      "loss": 6.6813,
      "mean_token_accuracy": 0.1156503714621067,
      "num_tokens": 6778656.0,
      "step": 3125
    },
    {
      "entropy": 6.852896070480346,
      "epoch": 0.3349563914602172,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004994112997211369,
      "loss": 6.7448,
      "mean_token_accuracy": 0.11959866657853127,
      "num_tokens": 6789960.0,
      "step": 3130
    },
    {
      "entropy": 6.8003246784210205,
      "epoch": 0.33549146556798115,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994085325384031,
      "loss": 6.7323,
      "mean_token_accuracy": 0.11916140839457512,
      "num_tokens": 6800188.0,
      "step": 3135
    },
    {
      "entropy": 6.8773034572601315,
      "epoch": 0.3360265396757451,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004994057588759045,
      "loss": 6.6903,
      "mean_token_accuracy": 0.12074850872159004,
      "num_tokens": 6811032.0,
      "step": 3140
    },
    {
      "entropy": 6.792485857009888,
      "epoch": 0.336561613783509,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004994029787337212,
      "loss": 6.7104,
      "mean_token_accuracy": 0.11358091682195663,
      "num_tokens": 6823060.0,
      "step": 3145
    },
    {
      "entropy": 6.898675489425659,
      "epoch": 0.33709668789127295,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994001921119333,
      "loss": 6.7803,
      "mean_token_accuracy": 0.117774448543787,
      "num_tokens": 6833559.0,
      "step": 3150
    },
    {
      "entropy": 6.897660970687866,
      "epoch": 0.3376317619990369,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004993973990106215,
      "loss": 6.7276,
      "mean_token_accuracy": 0.11970929205417633,
      "num_tokens": 6843657.0,
      "step": 3155
    },
    {
      "entropy": 6.811191844940185,
      "epoch": 0.3381668361068008,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993945994298662,
      "loss": 6.6928,
      "mean_token_accuracy": 0.11271135210990905,
      "num_tokens": 6853615.0,
      "step": 3160
    },
    {
      "entropy": 6.889709424972534,
      "epoch": 0.3387019102145647,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004993917933697484,
      "loss": 6.7701,
      "mean_token_accuracy": 0.11692628338932991,
      "num_tokens": 6865256.0,
      "step": 3165
    },
    {
      "entropy": 7.066078853607178,
      "epoch": 0.3392369843223286,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004993889808303493,
      "loss": 6.802,
      "mean_token_accuracy": 0.11643452644348144,
      "num_tokens": 6875358.0,
      "step": 3170
    },
    {
      "entropy": 6.784234189987183,
      "epoch": 0.33977205843009256,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004993861618117498,
      "loss": 6.7118,
      "mean_token_accuracy": 0.11509301662445068,
      "num_tokens": 6886584.0,
      "step": 3175
    },
    {
      "entropy": 6.8454875469207765,
      "epoch": 0.3403071325378565,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004993833363140314,
      "loss": 6.6499,
      "mean_token_accuracy": 0.12495435997843743,
      "num_tokens": 6896958.0,
      "step": 3180
    },
    {
      "entropy": 6.899998950958252,
      "epoch": 0.3408422066456204,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004993805043372756,
      "loss": 6.721,
      "mean_token_accuracy": 0.12064734548330307,
      "num_tokens": 6907944.0,
      "step": 3185
    },
    {
      "entropy": 6.884324169158935,
      "epoch": 0.34137728075338436,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004993776658815644,
      "loss": 6.7916,
      "mean_token_accuracy": 0.11133322417736054,
      "num_tokens": 6919486.0,
      "step": 3190
    },
    {
      "entropy": 6.93391604423523,
      "epoch": 0.3419123548611483,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004993748209469796,
      "loss": 6.7743,
      "mean_token_accuracy": 0.12054007947444915,
      "num_tokens": 6930650.0,
      "step": 3195
    },
    {
      "entropy": 6.829514265060425,
      "epoch": 0.34244742896891217,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004993719695336033,
      "loss": 6.7183,
      "mean_token_accuracy": 0.116023488342762,
      "num_tokens": 6941519.0,
      "step": 3200
    },
    {
      "entropy": 6.871858549118042,
      "epoch": 0.3429825030766761,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993691116415179,
      "loss": 6.7584,
      "mean_token_accuracy": 0.11689019277691841,
      "num_tokens": 6952952.0,
      "step": 3205
    },
    {
      "entropy": 6.865701389312744,
      "epoch": 0.34351757718444004,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000499366247270806,
      "loss": 6.7087,
      "mean_token_accuracy": 0.1214831866323948,
      "num_tokens": 6962867.0,
      "step": 3210
    },
    {
      "entropy": 6.8616862297058105,
      "epoch": 0.34405265129220397,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004993633764215502,
      "loss": 6.6533,
      "mean_token_accuracy": 0.12417162358760833,
      "num_tokens": 6973434.0,
      "step": 3215
    },
    {
      "entropy": 6.876732921600341,
      "epoch": 0.3445877253999679,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004993604990938336,
      "loss": 6.6436,
      "mean_token_accuracy": 0.1319417916238308,
      "num_tokens": 6984161.0,
      "step": 3220
    },
    {
      "entropy": 6.887901353836059,
      "epoch": 0.34512279950773184,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993576152877389,
      "loss": 6.6539,
      "mean_token_accuracy": 0.11877508088946342,
      "num_tokens": 6994700.0,
      "step": 3225
    },
    {
      "entropy": 6.865823125839233,
      "epoch": 0.34565787361549577,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004993547250033496,
      "loss": 6.8033,
      "mean_token_accuracy": 0.10758280605077744,
      "num_tokens": 7006497.0,
      "step": 3230
    },
    {
      "entropy": 6.89277138710022,
      "epoch": 0.34619294772325965,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004993518282407493,
      "loss": 6.7089,
      "mean_token_accuracy": 0.1204972542822361,
      "num_tokens": 7016415.0,
      "step": 3235
    },
    {
      "entropy": 6.824367952346802,
      "epoch": 0.3467280218310236,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993489250000213,
      "loss": 6.6683,
      "mean_token_accuracy": 0.11787926405668259,
      "num_tokens": 7026610.0,
      "step": 3240
    },
    {
      "entropy": 6.865869045257568,
      "epoch": 0.3472630959387875,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004993460152812497,
      "loss": 6.6843,
      "mean_token_accuracy": 0.11314239874482154,
      "num_tokens": 7038236.0,
      "step": 3245
    },
    {
      "entropy": 6.821169090270996,
      "epoch": 0.34779817004655145,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993430990845184,
      "loss": 6.7329,
      "mean_token_accuracy": 0.1261778935790062,
      "num_tokens": 7049438.0,
      "step": 3250
    },
    {
      "entropy": 6.8721363067626955,
      "epoch": 0.3483332441543154,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004993401764099118,
      "loss": 6.7005,
      "mean_token_accuracy": 0.11510813310742378,
      "num_tokens": 7061001.0,
      "step": 3255
    },
    {
      "entropy": 6.864164257049561,
      "epoch": 0.3488683182620793,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004993372472575139,
      "loss": 6.6814,
      "mean_token_accuracy": 0.1169338621199131,
      "num_tokens": 7071328.0,
      "step": 3260
    },
    {
      "entropy": 6.914004755020142,
      "epoch": 0.34940339236984325,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004993343116274096,
      "loss": 6.8575,
      "mean_token_accuracy": 0.10851389616727829,
      "num_tokens": 7082597.0,
      "step": 3265
    },
    {
      "entropy": 6.894482135772705,
      "epoch": 0.3499384664776071,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004993313695196836,
      "loss": 6.6804,
      "mean_token_accuracy": 0.1231270395219326,
      "num_tokens": 7093463.0,
      "step": 3270
    },
    {
      "entropy": 6.794412565231323,
      "epoch": 0.35047354058537106,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004993284209344208,
      "loss": 6.7627,
      "mean_token_accuracy": 0.12201037406921386,
      "num_tokens": 7104350.0,
      "step": 3275
    },
    {
      "entropy": 6.926002883911133,
      "epoch": 0.351008614693135,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993254658717065,
      "loss": 6.6938,
      "mean_token_accuracy": 0.12726418748497964,
      "num_tokens": 7114851.0,
      "step": 3280
    },
    {
      "entropy": 6.871751928329468,
      "epoch": 0.3515436888008989,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004993225043316257,
      "loss": 6.7547,
      "mean_token_accuracy": 0.11642403826117516,
      "num_tokens": 7126440.0,
      "step": 3285
    },
    {
      "entropy": 6.894772243499756,
      "epoch": 0.35207876290866286,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993195363142643,
      "loss": 6.775,
      "mean_token_accuracy": 0.11527071371674538,
      "num_tokens": 7137884.0,
      "step": 3290
    },
    {
      "entropy": 6.943993806838989,
      "epoch": 0.3526138370164268,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004993165618197076,
      "loss": 6.7509,
      "mean_token_accuracy": 0.11260337382555008,
      "num_tokens": 7147930.0,
      "step": 3295
    },
    {
      "entropy": 6.814736032485962,
      "epoch": 0.3531489111241907,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004993135808480418,
      "loss": 6.6043,
      "mean_token_accuracy": 0.12858601436018943,
      "num_tokens": 7159200.0,
      "step": 3300
    },
    {
      "entropy": 6.884348058700562,
      "epoch": 0.3536839852319546,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004993105933993529,
      "loss": 6.7094,
      "mean_token_accuracy": 0.12247735410928726,
      "num_tokens": 7169601.0,
      "step": 3305
    },
    {
      "entropy": 6.798114442825318,
      "epoch": 0.35421905933971853,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004993075994737271,
      "loss": 6.6409,
      "mean_token_accuracy": 0.12730678617954255,
      "num_tokens": 7180465.0,
      "step": 3310
    },
    {
      "entropy": 6.855682373046875,
      "epoch": 0.35475413344748247,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004993045990712508,
      "loss": 6.6546,
      "mean_token_accuracy": 0.12138160839676856,
      "num_tokens": 7190603.0,
      "step": 3315
    },
    {
      "entropy": 6.80737338066101,
      "epoch": 0.3552892075552464,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004993015921920109,
      "loss": 6.6164,
      "mean_token_accuracy": 0.12704969346523284,
      "num_tokens": 7201017.0,
      "step": 3320
    },
    {
      "entropy": 6.7541686534881595,
      "epoch": 0.35582428166301033,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499298578836094,
      "loss": 6.6315,
      "mean_token_accuracy": 0.12172743976116181,
      "num_tokens": 7210908.0,
      "step": 3325
    },
    {
      "entropy": 6.896972274780273,
      "epoch": 0.35635935577077427,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499295559003587,
      "loss": 6.6669,
      "mean_token_accuracy": 0.12253023311495781,
      "num_tokens": 7221306.0,
      "step": 3330
    },
    {
      "entropy": 6.821090650558472,
      "epoch": 0.3568944298785382,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004992925326945774,
      "loss": 6.7672,
      "mean_token_accuracy": 0.11355544701218605,
      "num_tokens": 7233076.0,
      "step": 3335
    },
    {
      "entropy": 6.866760158538819,
      "epoch": 0.3574295039863021,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992894999091524,
      "loss": 6.771,
      "mean_token_accuracy": 0.11949312463402748,
      "num_tokens": 7244703.0,
      "step": 3340
    },
    {
      "entropy": 6.897739601135254,
      "epoch": 0.357964578094066,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004992864606473996,
      "loss": 6.7545,
      "mean_token_accuracy": 0.10833368301391602,
      "num_tokens": 7255438.0,
      "step": 3345
    },
    {
      "entropy": 6.838145685195923,
      "epoch": 0.35849965220182994,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004992834149094069,
      "loss": 6.6846,
      "mean_token_accuracy": 0.12186626270413399,
      "num_tokens": 7265720.0,
      "step": 3350
    },
    {
      "entropy": 6.900753784179687,
      "epoch": 0.3590347263095939,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004992803626952619,
      "loss": 6.7495,
      "mean_token_accuracy": 0.11561599299311638,
      "num_tokens": 7277791.0,
      "step": 3355
    },
    {
      "entropy": 6.7818128108978275,
      "epoch": 0.3595698004173578,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499277304005053,
      "loss": 6.6111,
      "mean_token_accuracy": 0.12490361779928208,
      "num_tokens": 7288521.0,
      "step": 3360
    },
    {
      "entropy": 6.859988403320313,
      "epoch": 0.36010487452512174,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004992742388388686,
      "loss": 6.8157,
      "mean_token_accuracy": 0.10504508838057518,
      "num_tokens": 7299298.0,
      "step": 3365
    },
    {
      "entropy": 6.829821062088013,
      "epoch": 0.3606399486328857,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004992711671967968,
      "loss": 6.6284,
      "mean_token_accuracy": 0.11733799502253532,
      "num_tokens": 7310271.0,
      "step": 3370
    },
    {
      "entropy": 6.864026784896851,
      "epoch": 0.36117502274064955,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004992680890789268,
      "loss": 6.6758,
      "mean_token_accuracy": 0.11942022815346717,
      "num_tokens": 7321112.0,
      "step": 3375
    },
    {
      "entropy": 6.750688362121582,
      "epoch": 0.3617100968484135,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004992650044853471,
      "loss": 6.6393,
      "mean_token_accuracy": 0.11746473759412765,
      "num_tokens": 7331976.0,
      "step": 3380
    },
    {
      "entropy": 6.890619802474975,
      "epoch": 0.3622451709561774,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499261913416147,
      "loss": 6.7588,
      "mean_token_accuracy": 0.11621564850211144,
      "num_tokens": 7342800.0,
      "step": 3385
    },
    {
      "entropy": 6.848920202255249,
      "epoch": 0.36278024506394135,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004992588158714157,
      "loss": 6.5817,
      "mean_token_accuracy": 0.11895588338375092,
      "num_tokens": 7354213.0,
      "step": 3390
    },
    {
      "entropy": 6.815144014358521,
      "epoch": 0.3633153191717053,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004992557118512424,
      "loss": 6.6529,
      "mean_token_accuracy": 0.12157573848962784,
      "num_tokens": 7363250.0,
      "step": 3395
    },
    {
      "entropy": 6.803383541107178,
      "epoch": 0.3638503932794692,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992526013557171,
      "loss": 6.6955,
      "mean_token_accuracy": 0.11666846051812171,
      "num_tokens": 7374185.0,
      "step": 3400
    },
    {
      "entropy": 6.794870615005493,
      "epoch": 0.36438546738723315,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004992494843849294,
      "loss": 6.6654,
      "mean_token_accuracy": 0.11655777841806411,
      "num_tokens": 7384579.0,
      "step": 3405
    },
    {
      "entropy": 6.915937185287476,
      "epoch": 0.36492054149499703,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004992463609389694,
      "loss": 6.6681,
      "mean_token_accuracy": 0.11573902368545533,
      "num_tokens": 7395277.0,
      "step": 3410
    },
    {
      "entropy": 6.742115163803101,
      "epoch": 0.36545561560276096,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004992432310179272,
      "loss": 6.6648,
      "mean_token_accuracy": 0.11871817484498023,
      "num_tokens": 7406688.0,
      "step": 3415
    },
    {
      "entropy": 6.779878377914429,
      "epoch": 0.3659906897105249,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004992400946218932,
      "loss": 6.7219,
      "mean_token_accuracy": 0.12013051733374595,
      "num_tokens": 7417378.0,
      "step": 3420
    },
    {
      "entropy": 6.994897317886353,
      "epoch": 0.36652576381828883,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004992369517509581,
      "loss": 6.756,
      "mean_token_accuracy": 0.11834986060857773,
      "num_tokens": 7427675.0,
      "step": 3425
    },
    {
      "entropy": 6.822107934951783,
      "epoch": 0.36706083792605276,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004992338024052123,
      "loss": 6.7854,
      "mean_token_accuracy": 0.11357039287686348,
      "num_tokens": 7438083.0,
      "step": 3430
    },
    {
      "entropy": 6.792652750015259,
      "epoch": 0.3675959120338167,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004992306465847473,
      "loss": 6.6328,
      "mean_token_accuracy": 0.12360611706972122,
      "num_tokens": 7449051.0,
      "step": 3435
    },
    {
      "entropy": 6.7997828960418705,
      "epoch": 0.36813098614158063,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004992274842896536,
      "loss": 6.6609,
      "mean_token_accuracy": 0.12163360193371772,
      "num_tokens": 7459074.0,
      "step": 3440
    },
    {
      "entropy": 6.837110090255737,
      "epoch": 0.3686660602493445,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004992243155200229,
      "loss": 6.7077,
      "mean_token_accuracy": 0.12012316286563873,
      "num_tokens": 7470472.0,
      "step": 3445
    },
    {
      "entropy": 6.830043697357178,
      "epoch": 0.36920113435710844,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004992211402759466,
      "loss": 6.6807,
      "mean_token_accuracy": 0.12608814984560013,
      "num_tokens": 7481477.0,
      "step": 3450
    },
    {
      "entropy": 6.813068199157715,
      "epoch": 0.3697362084648724,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004992179585575164,
      "loss": 6.6334,
      "mean_token_accuracy": 0.12384276241064071,
      "num_tokens": 7491906.0,
      "step": 3455
    },
    {
      "entropy": 6.824427318572998,
      "epoch": 0.3702712825726363,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499214770364824,
      "loss": 6.5898,
      "mean_token_accuracy": 0.11620399430394172,
      "num_tokens": 7502447.0,
      "step": 3460
    },
    {
      "entropy": 6.735115385055542,
      "epoch": 0.37080635668040024,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004992115756979617,
      "loss": 6.5782,
      "mean_token_accuracy": 0.12983837723731995,
      "num_tokens": 7513505.0,
      "step": 3465
    },
    {
      "entropy": 6.775961589813233,
      "epoch": 0.3713414307881642,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004992083745570218,
      "loss": 6.5941,
      "mean_token_accuracy": 0.12342580780386925,
      "num_tokens": 7524884.0,
      "step": 3470
    },
    {
      "entropy": 6.764170551300049,
      "epoch": 0.3718765048959281,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004992051669420964,
      "loss": 6.676,
      "mean_token_accuracy": 0.12419689074158669,
      "num_tokens": 7535942.0,
      "step": 3475
    },
    {
      "entropy": 6.829311561584473,
      "epoch": 0.372411579003692,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004992019528532785,
      "loss": 6.6426,
      "mean_token_accuracy": 0.12233269810676575,
      "num_tokens": 7545506.0,
      "step": 3480
    },
    {
      "entropy": 6.849927377700806,
      "epoch": 0.3729466531114559,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991987322906605,
      "loss": 6.7158,
      "mean_token_accuracy": 0.11780445948243141,
      "num_tokens": 7555328.0,
      "step": 3485
    },
    {
      "entropy": 6.837383031845093,
      "epoch": 0.37348172721921985,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004991955052543357,
      "loss": 6.6114,
      "mean_token_accuracy": 0.12114308625459672,
      "num_tokens": 7565762.0,
      "step": 3490
    },
    {
      "entropy": 6.844478034973145,
      "epoch": 0.3740168013269838,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004991922717443972,
      "loss": 6.6433,
      "mean_token_accuracy": 0.12708517611026765,
      "num_tokens": 7575884.0,
      "step": 3495
    },
    {
      "entropy": 6.724173927307129,
      "epoch": 0.3745518754347477,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004991890317609384,
      "loss": 6.6636,
      "mean_token_accuracy": 0.12050025686621665,
      "num_tokens": 7586100.0,
      "step": 3500
    },
    {
      "entropy": 6.865896463394165,
      "epoch": 0.37508694954251165,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004991857853040528,
      "loss": 6.6853,
      "mean_token_accuracy": 0.11960910484194756,
      "num_tokens": 7597137.0,
      "step": 3505
    },
    {
      "entropy": 6.800037050247193,
      "epoch": 0.3756220236502756,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004991825323738341,
      "loss": 6.7123,
      "mean_token_accuracy": 0.12142123505473137,
      "num_tokens": 7607640.0,
      "step": 3510
    },
    {
      "entropy": 6.873905754089355,
      "epoch": 0.3761570977580395,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004991792729703763,
      "loss": 6.7616,
      "mean_token_accuracy": 0.120759766548872,
      "num_tokens": 7618319.0,
      "step": 3515
    },
    {
      "entropy": 6.834002447128296,
      "epoch": 0.3766921718658034,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004991760070937734,
      "loss": 6.6307,
      "mean_token_accuracy": 0.1232881985604763,
      "num_tokens": 7628350.0,
      "step": 3520
    },
    {
      "entropy": 6.795883989334106,
      "epoch": 0.3772272459735673,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004991727347441199,
      "loss": 6.6589,
      "mean_token_accuracy": 0.11747006997466088,
      "num_tokens": 7637857.0,
      "step": 3525
    },
    {
      "entropy": 6.8453436374664305,
      "epoch": 0.37776232008133126,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004991694559215102,
      "loss": 6.6895,
      "mean_token_accuracy": 0.11790692433714867,
      "num_tokens": 7648242.0,
      "step": 3530
    },
    {
      "entropy": 6.828597497940064,
      "epoch": 0.3782973941890952,
      "grad_norm": 0.7265625,
      "learning_rate": 0.000499166170626039,
      "loss": 6.6843,
      "mean_token_accuracy": 0.12224968299269676,
      "num_tokens": 7660565.0,
      "step": 3535
    },
    {
      "entropy": 6.7761329174041744,
      "epoch": 0.3788324682968591,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499162878857801,
      "loss": 6.7028,
      "mean_token_accuracy": 0.12394043356180191,
      "num_tokens": 7671600.0,
      "step": 3540
    },
    {
      "entropy": 6.851832103729248,
      "epoch": 0.37936754240462306,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004991595806168915,
      "loss": 6.7028,
      "mean_token_accuracy": 0.116616740077734,
      "num_tokens": 7682453.0,
      "step": 3545
    },
    {
      "entropy": 6.792314147949218,
      "epoch": 0.379902616512387,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004991562759034055,
      "loss": 6.6307,
      "mean_token_accuracy": 0.12220644429326058,
      "num_tokens": 7693490.0,
      "step": 3550
    },
    {
      "entropy": 6.855330133438111,
      "epoch": 0.38043769062015087,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004991529647174386,
      "loss": 6.6002,
      "mean_token_accuracy": 0.1216971792280674,
      "num_tokens": 7703478.0,
      "step": 3555
    },
    {
      "entropy": 6.825657939910888,
      "epoch": 0.3809727647279148,
      "grad_norm": 0.875,
      "learning_rate": 0.0004991496470590864,
      "loss": 6.6945,
      "mean_token_accuracy": 0.12218287289142608,
      "num_tokens": 7714647.0,
      "step": 3560
    },
    {
      "entropy": 6.8190515518188475,
      "epoch": 0.38150783883567874,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004991463229284445,
      "loss": 6.6358,
      "mean_token_accuracy": 0.12489367127418519,
      "num_tokens": 7725528.0,
      "step": 3565
    },
    {
      "entropy": 6.758315944671631,
      "epoch": 0.38204291294344267,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004991429923256091,
      "loss": 6.6856,
      "mean_token_accuracy": 0.11895663514733315,
      "num_tokens": 7737708.0,
      "step": 3570
    },
    {
      "entropy": 6.861998987197876,
      "epoch": 0.3825779870512066,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004991396552506762,
      "loss": 6.6692,
      "mean_token_accuracy": 0.11622913479804993,
      "num_tokens": 7749093.0,
      "step": 3575
    },
    {
      "entropy": 6.778561115264893,
      "epoch": 0.38311306115897054,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004991363117037423,
      "loss": 6.6198,
      "mean_token_accuracy": 0.12512530609965325,
      "num_tokens": 7760024.0,
      "step": 3580
    },
    {
      "entropy": 6.771070766448974,
      "epoch": 0.38364813526673447,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004991329616849039,
      "loss": 6.6592,
      "mean_token_accuracy": 0.12613205164670943,
      "num_tokens": 7770801.0,
      "step": 3585
    },
    {
      "entropy": 6.844554233551025,
      "epoch": 0.38418320937449835,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004991296051942578,
      "loss": 6.7097,
      "mean_token_accuracy": 0.12112276777625083,
      "num_tokens": 7782257.0,
      "step": 3590
    },
    {
      "entropy": 6.784887075424194,
      "epoch": 0.3847182834822623,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004991262422319008,
      "loss": 6.6227,
      "mean_token_accuracy": 0.12433370351791381,
      "num_tokens": 7793016.0,
      "step": 3595
    },
    {
      "entropy": 6.7855127334594725,
      "epoch": 0.3852533575900262,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00049912287279793,
      "loss": 6.5883,
      "mean_token_accuracy": 0.12935441955924035,
      "num_tokens": 7802207.0,
      "step": 3600
    },
    {
      "entropy": 6.801633358001709,
      "epoch": 0.38578843169779015,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991194968924429,
      "loss": 6.7064,
      "mean_token_accuracy": 0.11721168383955956,
      "num_tokens": 7813487.0,
      "step": 3605
    },
    {
      "entropy": 6.892347049713135,
      "epoch": 0.3863235058055541,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004991161145155367,
      "loss": 6.6713,
      "mean_token_accuracy": 0.11569483578205109,
      "num_tokens": 7824587.0,
      "step": 3610
    },
    {
      "entropy": 6.836808586120606,
      "epoch": 0.386858579913318,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004991127256673091,
      "loss": 6.6643,
      "mean_token_accuracy": 0.11696894839406013,
      "num_tokens": 7836240.0,
      "step": 3615
    },
    {
      "entropy": 6.777364730834961,
      "epoch": 0.38739365402108195,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004991093303478582,
      "loss": 6.6112,
      "mean_token_accuracy": 0.12282805442810059,
      "num_tokens": 7847014.0,
      "step": 3620
    },
    {
      "entropy": 6.840905952453613,
      "epoch": 0.3879287281288458,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004991059285572818,
      "loss": 6.715,
      "mean_token_accuracy": 0.12360445708036423,
      "num_tokens": 7857899.0,
      "step": 3625
    },
    {
      "entropy": 6.8158944606781,
      "epoch": 0.38846380223660976,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004991025202956781,
      "loss": 6.6287,
      "mean_token_accuracy": 0.1213743507862091,
      "num_tokens": 7868987.0,
      "step": 3630
    },
    {
      "entropy": 6.874086570739746,
      "epoch": 0.3889988763443737,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990991055631457,
      "loss": 6.733,
      "mean_token_accuracy": 0.12092002257704734,
      "num_tokens": 7879547.0,
      "step": 3635
    },
    {
      "entropy": 6.778404331207275,
      "epoch": 0.3895339504521376,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004990956843597832,
      "loss": 6.6001,
      "mean_token_accuracy": 0.12471188604831696,
      "num_tokens": 7889824.0,
      "step": 3640
    },
    {
      "entropy": 6.730760717391968,
      "epoch": 0.39006902455990156,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004990922566856892,
      "loss": 6.5945,
      "mean_token_accuracy": 0.11946317404508591,
      "num_tokens": 7901156.0,
      "step": 3645
    },
    {
      "entropy": 6.825276613235474,
      "epoch": 0.3906040986676655,
      "grad_norm": 1.0,
      "learning_rate": 0.0004990888225409629,
      "loss": 6.6251,
      "mean_token_accuracy": 0.12112335860729218,
      "num_tokens": 7911695.0,
      "step": 3650
    },
    {
      "entropy": 6.731439781188965,
      "epoch": 0.3911391727754294,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004990853819257033,
      "loss": 6.6116,
      "mean_token_accuracy": 0.12454076483845711,
      "num_tokens": 7923559.0,
      "step": 3655
    },
    {
      "entropy": 6.8649694442749025,
      "epoch": 0.3916742468831933,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004990819348400097,
      "loss": 6.6536,
      "mean_token_accuracy": 0.11768093705177307,
      "num_tokens": 7935792.0,
      "step": 3660
    },
    {
      "entropy": 6.825340223312378,
      "epoch": 0.39220932099095723,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004990784812839818,
      "loss": 6.6292,
      "mean_token_accuracy": 0.1266237623989582,
      "num_tokens": 7946987.0,
      "step": 3665
    },
    {
      "entropy": 6.719356107711792,
      "epoch": 0.39274439509872117,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004990750212577192,
      "loss": 6.5651,
      "mean_token_accuracy": 0.1250915713608265,
      "num_tokens": 7957160.0,
      "step": 3670
    },
    {
      "entropy": 6.817027235031128,
      "epoch": 0.3932794692064851,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004990715547613219,
      "loss": 6.6522,
      "mean_token_accuracy": 0.11589304953813553,
      "num_tokens": 7968833.0,
      "step": 3675
    },
    {
      "entropy": 6.832051086425781,
      "epoch": 0.39381454331424903,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004990680817948898,
      "loss": 6.6972,
      "mean_token_accuracy": 0.11313962563872337,
      "num_tokens": 7980661.0,
      "step": 3680
    },
    {
      "entropy": 6.884352779388427,
      "epoch": 0.39434961742201297,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004990646023585234,
      "loss": 6.6231,
      "mean_token_accuracy": 0.12205359265208245,
      "num_tokens": 7992677.0,
      "step": 3685
    },
    {
      "entropy": 6.752724313735962,
      "epoch": 0.3948846915297769,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004990611164523231,
      "loss": 6.6034,
      "mean_token_accuracy": 0.12537218481302262,
      "num_tokens": 8002796.0,
      "step": 3690
    },
    {
      "entropy": 6.684204196929931,
      "epoch": 0.3954197656375408,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004990576240763896,
      "loss": 6.6338,
      "mean_token_accuracy": 0.12649924010038377,
      "num_tokens": 8012779.0,
      "step": 3695
    },
    {
      "entropy": 6.789287137985229,
      "epoch": 0.3959548397453047,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004990541252308237,
      "loss": 6.621,
      "mean_token_accuracy": 0.12149769961833953,
      "num_tokens": 8023442.0,
      "step": 3700
    },
    {
      "entropy": 6.840408802032471,
      "epoch": 0.39648991385306864,
      "grad_norm": 0.875,
      "learning_rate": 0.0004990506199157263,
      "loss": 6.7067,
      "mean_token_accuracy": 0.1183815985918045,
      "num_tokens": 8035163.0,
      "step": 3705
    },
    {
      "entropy": 6.808131217956543,
      "epoch": 0.3970249879608326,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990471081311988,
      "loss": 6.6411,
      "mean_token_accuracy": 0.1226613163948059,
      "num_tokens": 8045330.0,
      "step": 3710
    },
    {
      "entropy": 6.8097728252410885,
      "epoch": 0.3975600620685965,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004990435898773426,
      "loss": 6.66,
      "mean_token_accuracy": 0.12126943394541741,
      "num_tokens": 8056154.0,
      "step": 3715
    },
    {
      "entropy": 6.841643762588501,
      "epoch": 0.39809513617636044,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004990400651542592,
      "loss": 6.6104,
      "mean_token_accuracy": 0.1250067502260208,
      "num_tokens": 8066101.0,
      "step": 3720
    },
    {
      "entropy": 6.837434148788452,
      "epoch": 0.3986302102841244,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004990365339620504,
      "loss": 6.6817,
      "mean_token_accuracy": 0.11575175449252129,
      "num_tokens": 8077249.0,
      "step": 3725
    },
    {
      "entropy": 6.773148679733277,
      "epoch": 0.39916528439188825,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499032996300818,
      "loss": 6.6239,
      "mean_token_accuracy": 0.12317326664924622,
      "num_tokens": 8088532.0,
      "step": 3730
    },
    {
      "entropy": 6.802285242080688,
      "epoch": 0.3997003584996522,
      "grad_norm": 1.0,
      "learning_rate": 0.0004990294521706645,
      "loss": 6.6675,
      "mean_token_accuracy": 0.11330687180161476,
      "num_tokens": 8099541.0,
      "step": 3735
    },
    {
      "entropy": 6.784977293014526,
      "epoch": 0.4002354326074161,
      "grad_norm": 0.828125,
      "learning_rate": 0.000499025901571692,
      "loss": 6.7093,
      "mean_token_accuracy": 0.12464805692434311,
      "num_tokens": 8110532.0,
      "step": 3740
    },
    {
      "entropy": 6.850379228591919,
      "epoch": 0.40077050671518005,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004990223445040031,
      "loss": 6.6445,
      "mean_token_accuracy": 0.1286566063761711,
      "num_tokens": 8120991.0,
      "step": 3745
    },
    {
      "entropy": 6.778837108612061,
      "epoch": 0.401305580822944,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004990187809677004,
      "loss": 6.5695,
      "mean_token_accuracy": 0.12042674124240875,
      "num_tokens": 8130737.0,
      "step": 3750
    },
    {
      "entropy": 6.708615112304687,
      "epoch": 0.4018406549307079,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004990152109628869,
      "loss": 6.5774,
      "mean_token_accuracy": 0.12504880353808404,
      "num_tokens": 8142315.0,
      "step": 3755
    },
    {
      "entropy": 6.7948565006256105,
      "epoch": 0.40237572903847185,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990116344896657,
      "loss": 6.6528,
      "mean_token_accuracy": 0.11660494431853294,
      "num_tokens": 8153618.0,
      "step": 3760
    },
    {
      "entropy": 6.712233686447144,
      "epoch": 0.40291080314623573,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00049900805154814,
      "loss": 6.6012,
      "mean_token_accuracy": 0.1282593384385109,
      "num_tokens": 8164304.0,
      "step": 3765
    },
    {
      "entropy": 6.75973072052002,
      "epoch": 0.40344587725399966,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004990044621384134,
      "loss": 6.5954,
      "mean_token_accuracy": 0.12287844493985176,
      "num_tokens": 8175289.0,
      "step": 3770
    },
    {
      "entropy": 6.851854515075684,
      "epoch": 0.4039809513617636,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004990008662605893,
      "loss": 6.6595,
      "mean_token_accuracy": 0.11499276533722877,
      "num_tokens": 8186274.0,
      "step": 3775
    },
    {
      "entropy": 6.803466510772705,
      "epoch": 0.40451602546952753,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004989972639147717,
      "loss": 6.5814,
      "mean_token_accuracy": 0.12695536240935326,
      "num_tokens": 8196380.0,
      "step": 3780
    },
    {
      "entropy": 6.756079292297363,
      "epoch": 0.40505109957729146,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004989936551010646,
      "loss": 6.6541,
      "mean_token_accuracy": 0.12403199076652527,
      "num_tokens": 8207204.0,
      "step": 3785
    },
    {
      "entropy": 6.709351348876953,
      "epoch": 0.4055861736850554,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004989900398195722,
      "loss": 6.5713,
      "mean_token_accuracy": 0.12644470483064651,
      "num_tokens": 8217745.0,
      "step": 3790
    },
    {
      "entropy": 6.757201862335205,
      "epoch": 0.40612124779281933,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989864180703988,
      "loss": 6.5731,
      "mean_token_accuracy": 0.12548816651105882,
      "num_tokens": 8228171.0,
      "step": 3795
    },
    {
      "entropy": 6.756083679199219,
      "epoch": 0.4066563219005832,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004989827898536491,
      "loss": 6.6065,
      "mean_token_accuracy": 0.11865482330322266,
      "num_tokens": 8239220.0,
      "step": 3800
    },
    {
      "entropy": 6.798965406417847,
      "epoch": 0.40719139600834714,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004989791551694279,
      "loss": 6.6643,
      "mean_token_accuracy": 0.1210280641913414,
      "num_tokens": 8250363.0,
      "step": 3805
    },
    {
      "entropy": 6.774178981781006,
      "epoch": 0.4077264701161111,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004989755140178398,
      "loss": 6.6237,
      "mean_token_accuracy": 0.12307491078972817,
      "num_tokens": 8261207.0,
      "step": 3810
    },
    {
      "entropy": 6.712477159500122,
      "epoch": 0.408261544223875,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004989718663989905,
      "loss": 6.5186,
      "mean_token_accuracy": 0.12528306394815444,
      "num_tokens": 8272343.0,
      "step": 3815
    },
    {
      "entropy": 6.727295970916748,
      "epoch": 0.40879661833163894,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004989682123129849,
      "loss": 6.5847,
      "mean_token_accuracy": 0.12415433675050735,
      "num_tokens": 8282677.0,
      "step": 3820
    },
    {
      "entropy": 6.73093900680542,
      "epoch": 0.40933169243940287,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004989645517599286,
      "loss": 6.596,
      "mean_token_accuracy": 0.1210504062473774,
      "num_tokens": 8292753.0,
      "step": 3825
    },
    {
      "entropy": 6.7588269233703615,
      "epoch": 0.4098667665471668,
      "grad_norm": 0.875,
      "learning_rate": 0.0004989608847399275,
      "loss": 6.5883,
      "mean_token_accuracy": 0.12015826031565666,
      "num_tokens": 8302747.0,
      "step": 3830
    },
    {
      "entropy": 6.7176045894622805,
      "epoch": 0.4104018406549307,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989572112530871,
      "loss": 6.6381,
      "mean_token_accuracy": 0.11648379042744636,
      "num_tokens": 8314766.0,
      "step": 3835
    },
    {
      "entropy": 6.836350011825561,
      "epoch": 0.4109369147626946,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004989535312995139,
      "loss": 6.6636,
      "mean_token_accuracy": 0.12048284634947777,
      "num_tokens": 8325791.0,
      "step": 3840
    },
    {
      "entropy": 6.763233232498169,
      "epoch": 0.41147198887045855,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004989498448793139,
      "loss": 6.6071,
      "mean_token_accuracy": 0.12325042858719826,
      "num_tokens": 8335809.0,
      "step": 3845
    },
    {
      "entropy": 6.73609185218811,
      "epoch": 0.4120070629782225,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004989461519925935,
      "loss": 6.596,
      "mean_token_accuracy": 0.11986378282308578,
      "num_tokens": 8347505.0,
      "step": 3850
    },
    {
      "entropy": 6.789737033843994,
      "epoch": 0.4125421370859864,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989424526394596,
      "loss": 6.602,
      "mean_token_accuracy": 0.12010528445243836,
      "num_tokens": 8358363.0,
      "step": 3855
    },
    {
      "entropy": 6.752479887008667,
      "epoch": 0.41307721119375035,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004989387468200187,
      "loss": 6.631,
      "mean_token_accuracy": 0.12690046951174735,
      "num_tokens": 8369076.0,
      "step": 3860
    },
    {
      "entropy": 6.815656948089599,
      "epoch": 0.4136122853015143,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000498935034534378,
      "loss": 6.6618,
      "mean_token_accuracy": 0.12442757785320283,
      "num_tokens": 8380071.0,
      "step": 3865
    },
    {
      "entropy": 6.747093915939331,
      "epoch": 0.41414735940927816,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004989313157826446,
      "loss": 6.6123,
      "mean_token_accuracy": 0.12264053821563721,
      "num_tokens": 8390574.0,
      "step": 3870
    },
    {
      "entropy": 6.720603561401367,
      "epoch": 0.4146824335170421,
      "grad_norm": 0.796875,
      "learning_rate": 0.000498927590564926,
      "loss": 6.6093,
      "mean_token_accuracy": 0.12397766932845115,
      "num_tokens": 8401154.0,
      "step": 3875
    },
    {
      "entropy": 6.792613697052002,
      "epoch": 0.415217507624806,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004989238588813296,
      "loss": 6.6893,
      "mean_token_accuracy": 0.1173297606408596,
      "num_tokens": 8412113.0,
      "step": 3880
    },
    {
      "entropy": 6.836045598983764,
      "epoch": 0.41575258173256996,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004989201207319634,
      "loss": 6.6742,
      "mean_token_accuracy": 0.12323395013809205,
      "num_tokens": 8422734.0,
      "step": 3885
    },
    {
      "entropy": 6.767079925537109,
      "epoch": 0.4162876558403339,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004989163761169351,
      "loss": 6.6009,
      "mean_token_accuracy": 0.12026888653635978,
      "num_tokens": 8433647.0,
      "step": 3890
    },
    {
      "entropy": 6.77981743812561,
      "epoch": 0.4168227299480978,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989126250363529,
      "loss": 6.6333,
      "mean_token_accuracy": 0.11988705322146416,
      "num_tokens": 8444726.0,
      "step": 3895
    },
    {
      "entropy": 6.7848625659942625,
      "epoch": 0.41735780405586176,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004989088674903251,
      "loss": 6.6331,
      "mean_token_accuracy": 0.12055689990520477,
      "num_tokens": 8455351.0,
      "step": 3900
    },
    {
      "entropy": 6.751009321212768,
      "epoch": 0.41789287816362564,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004989051034789602,
      "loss": 6.6129,
      "mean_token_accuracy": 0.12630177587270736,
      "num_tokens": 8466423.0,
      "step": 3905
    },
    {
      "entropy": 6.756326627731323,
      "epoch": 0.41842795227138957,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000498901333002367,
      "loss": 6.5578,
      "mean_token_accuracy": 0.1217512920498848,
      "num_tokens": 8475841.0,
      "step": 3910
    },
    {
      "entropy": 6.75618543624878,
      "epoch": 0.4189630263791535,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004988975560606541,
      "loss": 6.6614,
      "mean_token_accuracy": 0.11742390394210815,
      "num_tokens": 8487656.0,
      "step": 3915
    },
    {
      "entropy": 6.782296848297119,
      "epoch": 0.41949810048691744,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004988937726539308,
      "loss": 6.67,
      "mean_token_accuracy": 0.11672923117876052,
      "num_tokens": 8497959.0,
      "step": 3920
    },
    {
      "entropy": 6.8015392303466795,
      "epoch": 0.42003317459468137,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004988899827823062,
      "loss": 6.5929,
      "mean_token_accuracy": 0.12176745384931564,
      "num_tokens": 8508228.0,
      "step": 3925
    },
    {
      "entropy": 6.76314640045166,
      "epoch": 0.4205682487024453,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00049888618644589,
      "loss": 6.6357,
      "mean_token_accuracy": 0.11630026996135712,
      "num_tokens": 8518999.0,
      "step": 3930
    },
    {
      "entropy": 6.720941352844238,
      "epoch": 0.42110332281020924,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004988823836447916,
      "loss": 6.6701,
      "mean_token_accuracy": 0.11666706353425979,
      "num_tokens": 8529583.0,
      "step": 3935
    },
    {
      "entropy": 6.880489730834961,
      "epoch": 0.4216383969179731,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004988785743791207,
      "loss": 6.7288,
      "mean_token_accuracy": 0.11347419321537018,
      "num_tokens": 8540069.0,
      "step": 3940
    },
    {
      "entropy": 6.77430419921875,
      "epoch": 0.42217347102573705,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004988747586489876,
      "loss": 6.6066,
      "mean_token_accuracy": 0.12069976776838302,
      "num_tokens": 8550441.0,
      "step": 3945
    },
    {
      "entropy": 6.724952983856201,
      "epoch": 0.422708545133501,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000498870936454502,
      "loss": 6.5822,
      "mean_token_accuracy": 0.1276530534029007,
      "num_tokens": 8560257.0,
      "step": 3950
    },
    {
      "entropy": 6.781778383255005,
      "epoch": 0.4232436192412649,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004988671077957749,
      "loss": 6.6431,
      "mean_token_accuracy": 0.123650910705328,
      "num_tokens": 8570991.0,
      "step": 3955
    },
    {
      "entropy": 6.8194362163543705,
      "epoch": 0.42377869334902885,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004988632726729164,
      "loss": 6.7012,
      "mean_token_accuracy": 0.11642257273197174,
      "num_tokens": 8583668.0,
      "step": 3960
    },
    {
      "entropy": 6.782205390930176,
      "epoch": 0.4243137674567928,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004988594310860374,
      "loss": 6.6329,
      "mean_token_accuracy": 0.12215517908334732,
      "num_tokens": 8593727.0,
      "step": 3965
    },
    {
      "entropy": 6.732775068283081,
      "epoch": 0.4248488415645567,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0004988555830352485,
      "loss": 6.6328,
      "mean_token_accuracy": 0.12335548922419548,
      "num_tokens": 8605470.0,
      "step": 3970
    },
    {
      "entropy": 6.800445938110352,
      "epoch": 0.4253839156723206,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004988517285206614,
      "loss": 6.5566,
      "mean_token_accuracy": 0.12064253017306328,
      "num_tokens": 8617176.0,
      "step": 3975
    },
    {
      "entropy": 6.773797082901001,
      "epoch": 0.4259189897800845,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004988478675423871,
      "loss": 6.6421,
      "mean_token_accuracy": 0.12061286568641663,
      "num_tokens": 8627756.0,
      "step": 3980
    },
    {
      "entropy": 6.770265436172485,
      "epoch": 0.42645406388784846,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004988440001005369,
      "loss": 6.5829,
      "mean_token_accuracy": 0.12523626312613487,
      "num_tokens": 8637766.0,
      "step": 3985
    },
    {
      "entropy": 6.754474687576294,
      "epoch": 0.4269891379956124,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004988401261952227,
      "loss": 6.6195,
      "mean_token_accuracy": 0.11977186426520348,
      "num_tokens": 8648562.0,
      "step": 3990
    },
    {
      "entropy": 6.81419939994812,
      "epoch": 0.4275242121033763,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004988362458265565,
      "loss": 6.6331,
      "mean_token_accuracy": 0.11861854270100594,
      "num_tokens": 8658393.0,
      "step": 3995
    },
    {
      "entropy": 6.745327091217041,
      "epoch": 0.42805928621114026,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004988323589946499,
      "loss": 6.5726,
      "mean_token_accuracy": 0.12602216973900796,
      "num_tokens": 8668365.0,
      "step": 4000
    },
    {
      "entropy": 6.7576432704925535,
      "epoch": 0.4285943603189042,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004988284656996156,
      "loss": 6.5932,
      "mean_token_accuracy": 0.11847178786993026,
      "num_tokens": 8679228.0,
      "step": 4005
    },
    {
      "entropy": 6.70848708152771,
      "epoch": 0.42912943442666807,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004988245659415656,
      "loss": 6.4966,
      "mean_token_accuracy": 0.1268126145005226,
      "num_tokens": 8688940.0,
      "step": 4010
    },
    {
      "entropy": 6.789049434661865,
      "epoch": 0.429664508534432,
      "grad_norm": 0.875,
      "learning_rate": 0.0004988206597206129,
      "loss": 6.6533,
      "mean_token_accuracy": 0.11652215644717216,
      "num_tokens": 8699923.0,
      "step": 4015
    },
    {
      "entropy": 6.688563680648803,
      "epoch": 0.43019958264219593,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00049881674703687,
      "loss": 6.4966,
      "mean_token_accuracy": 0.12585265636444093,
      "num_tokens": 8710445.0,
      "step": 4020
    },
    {
      "entropy": 6.723779106140137,
      "epoch": 0.43073465674995987,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004988128278904499,
      "loss": 6.6101,
      "mean_token_accuracy": 0.12553784623742104,
      "num_tokens": 8720686.0,
      "step": 4025
    },
    {
      "entropy": 6.76382417678833,
      "epoch": 0.4312697308577238,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000498808902281466,
      "loss": 6.6258,
      "mean_token_accuracy": 0.1235751561820507,
      "num_tokens": 8732122.0,
      "step": 4030
    },
    {
      "entropy": 6.71265287399292,
      "epoch": 0.43180480496548773,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004988049702100315,
      "loss": 6.5908,
      "mean_token_accuracy": 0.12264637053012847,
      "num_tokens": 8744122.0,
      "step": 4035
    },
    {
      "entropy": 6.793091678619385,
      "epoch": 0.43233987907325166,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004988010316762598,
      "loss": 6.5945,
      "mean_token_accuracy": 0.12287564054131508,
      "num_tokens": 8754758.0,
      "step": 4040
    },
    {
      "entropy": 6.687934494018554,
      "epoch": 0.4328749531810156,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004987970866802649,
      "loss": 6.6165,
      "mean_token_accuracy": 0.11594383940100669,
      "num_tokens": 8766577.0,
      "step": 4045
    },
    {
      "entropy": 6.767072296142578,
      "epoch": 0.4334100272887795,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004987931352221605,
      "loss": 6.6001,
      "mean_token_accuracy": 0.11936594918370247,
      "num_tokens": 8777698.0,
      "step": 4050
    },
    {
      "entropy": 6.77272047996521,
      "epoch": 0.4339451013965434,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004987891773020606,
      "loss": 6.6255,
      "mean_token_accuracy": 0.12437722161412239,
      "num_tokens": 8788637.0,
      "step": 4055
    },
    {
      "entropy": 6.755249404907227,
      "epoch": 0.43448017550430734,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004987852129200799,
      "loss": 6.6011,
      "mean_token_accuracy": 0.12569011077284814,
      "num_tokens": 8799709.0,
      "step": 4060
    },
    {
      "entropy": 6.790832901000977,
      "epoch": 0.4350152496120713,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004987812420763326,
      "loss": 6.5919,
      "mean_token_accuracy": 0.1250280186533928,
      "num_tokens": 8810469.0,
      "step": 4065
    },
    {
      "entropy": 6.773603105545044,
      "epoch": 0.4355503237198352,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004987772647709333,
      "loss": 6.5687,
      "mean_token_accuracy": 0.12102466896176338,
      "num_tokens": 8821142.0,
      "step": 4070
    },
    {
      "entropy": 6.6965799808502195,
      "epoch": 0.43608539782759914,
      "grad_norm": 0.8125,
      "learning_rate": 0.000498773281003997,
      "loss": 6.5329,
      "mean_token_accuracy": 0.13130529895424842,
      "num_tokens": 8832951.0,
      "step": 4075
    },
    {
      "entropy": 6.716156911849976,
      "epoch": 0.4366204719353631,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004987692907756385,
      "loss": 6.6321,
      "mean_token_accuracy": 0.11649341732263566,
      "num_tokens": 8843579.0,
      "step": 4080
    },
    {
      "entropy": 6.68398380279541,
      "epoch": 0.43715554604312695,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004987652940859733,
      "loss": 6.5737,
      "mean_token_accuracy": 0.119865170866251,
      "num_tokens": 8854612.0,
      "step": 4085
    },
    {
      "entropy": 6.678323125839233,
      "epoch": 0.4376906201508909,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004987612909351167,
      "loss": 6.5225,
      "mean_token_accuracy": 0.12679266855120658,
      "num_tokens": 8865584.0,
      "step": 4090
    },
    {
      "entropy": 6.702922630310058,
      "epoch": 0.4382256942586548,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004987572813231842,
      "loss": 6.5523,
      "mean_token_accuracy": 0.12998123988509178,
      "num_tokens": 8876662.0,
      "step": 4095
    },
    {
      "entropy": 6.730002212524414,
      "epoch": 0.43876076836641875,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004987532652502917,
      "loss": 6.5808,
      "mean_token_accuracy": 0.12212803661823272,
      "num_tokens": 8887249.0,
      "step": 4100
    },
    {
      "entropy": 6.674190044403076,
      "epoch": 0.4392958424741827,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004987492427165551,
      "loss": 6.4124,
      "mean_token_accuracy": 0.14259165078401564,
      "num_tokens": 8898635.0,
      "step": 4105
    },
    {
      "entropy": 6.6693785190582275,
      "epoch": 0.4398309165819466,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004987452137220906,
      "loss": 6.5918,
      "mean_token_accuracy": 0.12297084257006645,
      "num_tokens": 8908794.0,
      "step": 4110
    },
    {
      "entropy": 6.837216567993164,
      "epoch": 0.44036599068971055,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004987411782670144,
      "loss": 6.6168,
      "mean_token_accuracy": 0.11972223967313766,
      "num_tokens": 8919015.0,
      "step": 4115
    },
    {
      "entropy": 6.835398626327515,
      "epoch": 0.44090106479747443,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004987371363514431,
      "loss": 6.6538,
      "mean_token_accuracy": 0.11519176959991455,
      "num_tokens": 8929028.0,
      "step": 4120
    },
    {
      "entropy": 6.677055406570434,
      "epoch": 0.44143613890523836,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004987330879754933,
      "loss": 6.5376,
      "mean_token_accuracy": 0.12245551198720932,
      "num_tokens": 8939789.0,
      "step": 4125
    },
    {
      "entropy": 6.727229261398316,
      "epoch": 0.4419712130130023,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004987290331392823,
      "loss": 6.5154,
      "mean_token_accuracy": 0.13091592267155647,
      "num_tokens": 8950657.0,
      "step": 4130
    },
    {
      "entropy": 6.799406862258911,
      "epoch": 0.44250628712076623,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004987249718429269,
      "loss": 6.6207,
      "mean_token_accuracy": 0.12651147246360778,
      "num_tokens": 8960744.0,
      "step": 4135
    },
    {
      "entropy": 6.71635856628418,
      "epoch": 0.44304136122853016,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004987209040865441,
      "loss": 6.4854,
      "mean_token_accuracy": 0.12618361711502074,
      "num_tokens": 8972411.0,
      "step": 4140
    },
    {
      "entropy": 6.717485332489014,
      "epoch": 0.4435764353362941,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004987168298702519,
      "loss": 6.642,
      "mean_token_accuracy": 0.1232242576777935,
      "num_tokens": 8982818.0,
      "step": 4145
    },
    {
      "entropy": 6.756543302536011,
      "epoch": 0.44411150944405803,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004987127491941675,
      "loss": 6.5719,
      "mean_token_accuracy": 0.12182455062866211,
      "num_tokens": 8993227.0,
      "step": 4150
    },
    {
      "entropy": 6.670086431503296,
      "epoch": 0.4446465835518219,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004987086620584091,
      "loss": 6.5038,
      "mean_token_accuracy": 0.12411652579903602,
      "num_tokens": 9003796.0,
      "step": 4155
    },
    {
      "entropy": 6.707034969329834,
      "epoch": 0.44518165765958584,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004987045684630944,
      "loss": 6.5262,
      "mean_token_accuracy": 0.13413502499461175,
      "num_tokens": 9014931.0,
      "step": 4160
    },
    {
      "entropy": 6.748621940612793,
      "epoch": 0.44571673176734977,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004987004684083417,
      "loss": 6.5949,
      "mean_token_accuracy": 0.12303428947925568,
      "num_tokens": 9025312.0,
      "step": 4165
    },
    {
      "entropy": 6.70564079284668,
      "epoch": 0.4462518058751137,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004986963618942694,
      "loss": 6.5717,
      "mean_token_accuracy": 0.13180141299962997,
      "num_tokens": 9035563.0,
      "step": 4170
    },
    {
      "entropy": 6.804406261444091,
      "epoch": 0.44678687998287764,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004986922489209961,
      "loss": 6.6082,
      "mean_token_accuracy": 0.12511870563030242,
      "num_tokens": 9046065.0,
      "step": 4175
    },
    {
      "entropy": 6.678038454055786,
      "epoch": 0.44732195409064157,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004986881294886406,
      "loss": 6.498,
      "mean_token_accuracy": 0.1308855652809143,
      "num_tokens": 9056696.0,
      "step": 4180
    },
    {
      "entropy": 6.673586320877075,
      "epoch": 0.4478570281984055,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004986840035973218,
      "loss": 6.564,
      "mean_token_accuracy": 0.12332830429077149,
      "num_tokens": 9067035.0,
      "step": 4185
    },
    {
      "entropy": 6.726327848434448,
      "epoch": 0.4483921023061694,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004986798712471588,
      "loss": 6.5423,
      "mean_token_accuracy": 0.12333020940423012,
      "num_tokens": 9078253.0,
      "step": 4190
    },
    {
      "entropy": 6.842236042022705,
      "epoch": 0.4489271764139333,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986757324382709,
      "loss": 6.6082,
      "mean_token_accuracy": 0.11263529285788536,
      "num_tokens": 9088184.0,
      "step": 4195
    },
    {
      "entropy": 6.713875579833984,
      "epoch": 0.44946225052169725,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004986715871707777,
      "loss": 6.671,
      "mean_token_accuracy": 0.11528211683034897,
      "num_tokens": 9099321.0,
      "step": 4200
    },
    {
      "entropy": 6.77054238319397,
      "epoch": 0.4499973246294612,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004986674354447988,
      "loss": 6.6999,
      "mean_token_accuracy": 0.12187818437814713,
      "num_tokens": 9111410.0,
      "step": 4205
    },
    {
      "entropy": 6.772527551651001,
      "epoch": 0.4505323987372251,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004986632772604543,
      "loss": 6.5189,
      "mean_token_accuracy": 0.12885569408535957,
      "num_tokens": 9121932.0,
      "step": 4210
    },
    {
      "entropy": 6.614354991912842,
      "epoch": 0.45106747284498905,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986591126178639,
      "loss": 6.6003,
      "mean_token_accuracy": 0.12762246951460837,
      "num_tokens": 9132686.0,
      "step": 4215
    },
    {
      "entropy": 6.67122106552124,
      "epoch": 0.451602546952753,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004986549415171482,
      "loss": 6.5193,
      "mean_token_accuracy": 0.12586723566055297,
      "num_tokens": 9143582.0,
      "step": 4220
    },
    {
      "entropy": 6.866072940826416,
      "epoch": 0.45213762106051686,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004986507639584274,
      "loss": 6.6057,
      "mean_token_accuracy": 0.11669112965464593,
      "num_tokens": 9156173.0,
      "step": 4225
    },
    {
      "entropy": 6.694578695297241,
      "epoch": 0.4526726951682808,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004986465799418223,
      "loss": 6.6024,
      "mean_token_accuracy": 0.11952880546450614,
      "num_tokens": 9166882.0,
      "step": 4230
    },
    {
      "entropy": 6.651588058471679,
      "epoch": 0.4532077692760447,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004986423894674534,
      "loss": 6.5896,
      "mean_token_accuracy": 0.12054572701454162,
      "num_tokens": 9177525.0,
      "step": 4235
    },
    {
      "entropy": 6.781280660629273,
      "epoch": 0.45374284338380866,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004986381925354422,
      "loss": 6.5576,
      "mean_token_accuracy": 0.12628965452313423,
      "num_tokens": 9188586.0,
      "step": 4240
    },
    {
      "entropy": 6.691897439956665,
      "epoch": 0.4542779174915726,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004986339891459094,
      "loss": 6.4673,
      "mean_token_accuracy": 0.127530487626791,
      "num_tokens": 9199920.0,
      "step": 4245
    },
    {
      "entropy": 6.642374420166016,
      "epoch": 0.4548129915993365,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004986297792989768,
      "loss": 6.4694,
      "mean_token_accuracy": 0.13375552520155906,
      "num_tokens": 9210923.0,
      "step": 4250
    },
    {
      "entropy": 6.779571628570556,
      "epoch": 0.45534806570710046,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986255629947655,
      "loss": 6.6384,
      "mean_token_accuracy": 0.11917833164334297,
      "num_tokens": 9221459.0,
      "step": 4255
    },
    {
      "entropy": 6.746447467803955,
      "epoch": 0.45588313981486434,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004986213402333978,
      "loss": 6.5841,
      "mean_token_accuracy": 0.121009661257267,
      "num_tokens": 9231768.0,
      "step": 4260
    },
    {
      "entropy": 6.689784193038941,
      "epoch": 0.45641821392262827,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004986171110149951,
      "loss": 6.5714,
      "mean_token_accuracy": 0.1208104893565178,
      "num_tokens": 9242230.0,
      "step": 4265
    },
    {
      "entropy": 6.755019187927246,
      "epoch": 0.4569532880303922,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004986128753396798,
      "loss": 6.4709,
      "mean_token_accuracy": 0.13061774894595146,
      "num_tokens": 9252485.0,
      "step": 4270
    },
    {
      "entropy": 6.625811624526977,
      "epoch": 0.45748836213815613,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004986086332075742,
      "loss": 6.5478,
      "mean_token_accuracy": 0.12935234606266022,
      "num_tokens": 9263014.0,
      "step": 4275
    },
    {
      "entropy": 6.703430414199829,
      "epoch": 0.45802343624592007,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004986043846188006,
      "loss": 6.5806,
      "mean_token_accuracy": 0.12819859609007836,
      "num_tokens": 9273777.0,
      "step": 4280
    },
    {
      "entropy": 6.6102972507476805,
      "epoch": 0.458558510353684,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000498600129573482,
      "loss": 6.5214,
      "mean_token_accuracy": 0.11930121630430221,
      "num_tokens": 9284738.0,
      "step": 4285
    },
    {
      "entropy": 6.747187185287475,
      "epoch": 0.45909358446144793,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000498595868071741,
      "loss": 6.5802,
      "mean_token_accuracy": 0.12222979590296745,
      "num_tokens": 9295921.0,
      "step": 4290
    },
    {
      "entropy": 6.741103410720825,
      "epoch": 0.4596286585692118,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004985916001137006,
      "loss": 6.5547,
      "mean_token_accuracy": 0.12327921092510223,
      "num_tokens": 9306568.0,
      "step": 4295
    },
    {
      "entropy": 6.708829498291015,
      "epoch": 0.46016373267697575,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985873256994843,
      "loss": 6.517,
      "mean_token_accuracy": 0.12360680028796196,
      "num_tokens": 9316468.0,
      "step": 4300
    },
    {
      "entropy": 6.688055419921875,
      "epoch": 0.4606988067847397,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004985830448292154,
      "loss": 6.593,
      "mean_token_accuracy": 0.12291383445262909,
      "num_tokens": 9328185.0,
      "step": 4305
    },
    {
      "entropy": 6.687855386734009,
      "epoch": 0.4612338808925036,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004985787575030175,
      "loss": 6.5371,
      "mean_token_accuracy": 0.1250103920698166,
      "num_tokens": 9338591.0,
      "step": 4310
    },
    {
      "entropy": 6.734978055953979,
      "epoch": 0.46176895500026754,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004985744637210144,
      "loss": 6.556,
      "mean_token_accuracy": 0.12288020104169846,
      "num_tokens": 9349286.0,
      "step": 4315
    },
    {
      "entropy": 6.735143327713013,
      "epoch": 0.4623040291080315,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00049857016348333,
      "loss": 6.5502,
      "mean_token_accuracy": 0.13360480219125748,
      "num_tokens": 9359752.0,
      "step": 4320
    },
    {
      "entropy": 6.7408709049224855,
      "epoch": 0.4628391032157954,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004985658567900886,
      "loss": 6.6129,
      "mean_token_accuracy": 0.11857379451394082,
      "num_tokens": 9371823.0,
      "step": 4325
    },
    {
      "entropy": 6.59757809638977,
      "epoch": 0.4633741773235593,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004985615436414145,
      "loss": 6.4569,
      "mean_token_accuracy": 0.12301377132534981,
      "num_tokens": 9383127.0,
      "step": 4330
    },
    {
      "entropy": 6.671346235275268,
      "epoch": 0.4639092514313232,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985572240374321,
      "loss": 6.5678,
      "mean_token_accuracy": 0.12829141393303872,
      "num_tokens": 9394296.0,
      "step": 4335
    },
    {
      "entropy": 6.764807176589966,
      "epoch": 0.46444432553908715,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004985528979782665,
      "loss": 6.5357,
      "mean_token_accuracy": 0.12359946742653846,
      "num_tokens": 9405988.0,
      "step": 4340
    },
    {
      "entropy": 6.7524127006530765,
      "epoch": 0.4649793996468511,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004985485654640423,
      "loss": 6.5599,
      "mean_token_accuracy": 0.12768663391470908,
      "num_tokens": 9417479.0,
      "step": 4345
    },
    {
      "entropy": 6.697694635391235,
      "epoch": 0.465514473754615,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004985442264948847,
      "loss": 6.5056,
      "mean_token_accuracy": 0.13216489255428315,
      "num_tokens": 9428050.0,
      "step": 4350
    },
    {
      "entropy": 6.648604536056519,
      "epoch": 0.46604954786237895,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004985398810709189,
      "loss": 6.4727,
      "mean_token_accuracy": 0.13219011574983597,
      "num_tokens": 9438229.0,
      "step": 4355
    },
    {
      "entropy": 6.6446418285369875,
      "epoch": 0.4665846219701429,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004985355291922704,
      "loss": 6.5434,
      "mean_token_accuracy": 0.12346789464354516,
      "num_tokens": 9449611.0,
      "step": 4360
    },
    {
      "entropy": 6.6712123394012455,
      "epoch": 0.46711969607790677,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000498531170859065,
      "loss": 6.5078,
      "mean_token_accuracy": 0.12737661376595497,
      "num_tokens": 9460234.0,
      "step": 4365
    },
    {
      "entropy": 6.731583738327027,
      "epoch": 0.4676547701856707,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004985268060714285,
      "loss": 6.5717,
      "mean_token_accuracy": 0.12688400596380234,
      "num_tokens": 9472136.0,
      "step": 4370
    },
    {
      "entropy": 6.655231046676636,
      "epoch": 0.46818984429343463,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004985224348294868,
      "loss": 6.519,
      "mean_token_accuracy": 0.1265267513692379,
      "num_tokens": 9482602.0,
      "step": 4375
    },
    {
      "entropy": 6.784295654296875,
      "epoch": 0.46872491840119856,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004985180571333663,
      "loss": 6.5188,
      "mean_token_accuracy": 0.12609197646379472,
      "num_tokens": 9493782.0,
      "step": 4380
    },
    {
      "entropy": 6.6205767631530765,
      "epoch": 0.4692599925089625,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004985136729831932,
      "loss": 6.4878,
      "mean_token_accuracy": 0.1269471064209938,
      "num_tokens": 9504389.0,
      "step": 4385
    },
    {
      "entropy": 6.6520030975341795,
      "epoch": 0.46979506661672643,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004985092823790942,
      "loss": 6.6104,
      "mean_token_accuracy": 0.12734476551413537,
      "num_tokens": 9515166.0,
      "step": 4390
    },
    {
      "entropy": 6.7322594165802006,
      "epoch": 0.47033014072449036,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004985048853211961,
      "loss": 6.5243,
      "mean_token_accuracy": 0.1255890816450119,
      "num_tokens": 9525566.0,
      "step": 4395
    },
    {
      "entropy": 6.735842847824097,
      "epoch": 0.47086521483225424,
      "grad_norm": 0.875,
      "learning_rate": 0.000498500481809626,
      "loss": 6.4401,
      "mean_token_accuracy": 0.13288491889834403,
      "num_tokens": 9536021.0,
      "step": 4400
    },
    {
      "entropy": 6.676405382156372,
      "epoch": 0.4714002889400182,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984960718445107,
      "loss": 6.5484,
      "mean_token_accuracy": 0.12439467236399651,
      "num_tokens": 9546827.0,
      "step": 4405
    },
    {
      "entropy": 6.768996953964233,
      "epoch": 0.4719353630477821,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004984916554259778,
      "loss": 6.6555,
      "mean_token_accuracy": 0.11520670205354691,
      "num_tokens": 9558104.0,
      "step": 4410
    },
    {
      "entropy": 6.817309331893921,
      "epoch": 0.47247043715554604,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004984872325541547,
      "loss": 6.5555,
      "mean_token_accuracy": 0.12726601585745811,
      "num_tokens": 9568487.0,
      "step": 4415
    },
    {
      "entropy": 6.658626842498779,
      "epoch": 0.47300551126331,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004984828032291692,
      "loss": 6.5062,
      "mean_token_accuracy": 0.13354175090789794,
      "num_tokens": 9579509.0,
      "step": 4420
    },
    {
      "entropy": 6.69633994102478,
      "epoch": 0.4735405853710739,
      "grad_norm": 0.875,
      "learning_rate": 0.0004984783674511492,
      "loss": 6.5307,
      "mean_token_accuracy": 0.12417809590697289,
      "num_tokens": 9589931.0,
      "step": 4425
    },
    {
      "entropy": 6.746572542190552,
      "epoch": 0.47407565947883784,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004984739252202227,
      "loss": 6.5656,
      "mean_token_accuracy": 0.12624306231737137,
      "num_tokens": 9599764.0,
      "step": 4430
    },
    {
      "entropy": 6.670381259918213,
      "epoch": 0.4746107335866017,
      "grad_norm": 0.890625,
      "learning_rate": 0.000498469476536518,
      "loss": 6.5282,
      "mean_token_accuracy": 0.12489766702055931,
      "num_tokens": 9611145.0,
      "step": 4435
    },
    {
      "entropy": 6.662897253036499,
      "epoch": 0.47514580769436565,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004984650214001636,
      "loss": 6.5386,
      "mean_token_accuracy": 0.13048497065901757,
      "num_tokens": 9621441.0,
      "step": 4440
    },
    {
      "entropy": 6.6825775623321535,
      "epoch": 0.4756808818021296,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004984605598112881,
      "loss": 6.566,
      "mean_token_accuracy": 0.1295418195426464,
      "num_tokens": 9631874.0,
      "step": 4445
    },
    {
      "entropy": 6.6638665199279785,
      "epoch": 0.4762159559098935,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004984560917700204,
      "loss": 6.544,
      "mean_token_accuracy": 0.11912723928689957,
      "num_tokens": 9643006.0,
      "step": 4450
    },
    {
      "entropy": 6.682493591308594,
      "epoch": 0.47675103001765745,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004984516172764894,
      "loss": 6.4444,
      "mean_token_accuracy": 0.13263275399804114,
      "num_tokens": 9653897.0,
      "step": 4455
    },
    {
      "entropy": 6.6289448738098145,
      "epoch": 0.4772861041254214,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004984471363308243,
      "loss": 6.5144,
      "mean_token_accuracy": 0.12464602962136269,
      "num_tokens": 9664460.0,
      "step": 4460
    },
    {
      "entropy": 6.681101226806641,
      "epoch": 0.4778211782331853,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004984426489331546,
      "loss": 6.5323,
      "mean_token_accuracy": 0.13034801334142684,
      "num_tokens": 9674534.0,
      "step": 4465
    },
    {
      "entropy": 6.785943937301636,
      "epoch": 0.4783562523409492,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004984381550836097,
      "loss": 6.6077,
      "mean_token_accuracy": 0.12375456094741821,
      "num_tokens": 9684828.0,
      "step": 4470
    },
    {
      "entropy": 6.758577537536621,
      "epoch": 0.47889132644871313,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004984336547823196,
      "loss": 6.565,
      "mean_token_accuracy": 0.12997907698154448,
      "num_tokens": 9694536.0,
      "step": 4475
    },
    {
      "entropy": 6.653159952163696,
      "epoch": 0.47942640055647706,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000498429148029414,
      "loss": 6.4319,
      "mean_token_accuracy": 0.13408037573099135,
      "num_tokens": 9705715.0,
      "step": 4480
    },
    {
      "entropy": 6.625295925140381,
      "epoch": 0.479961474664241,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004984246348250233,
      "loss": 6.5029,
      "mean_token_accuracy": 0.13184099048376083,
      "num_tokens": 9717505.0,
      "step": 4485
    },
    {
      "entropy": 6.672426366806031,
      "epoch": 0.4804965487720049,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004984201151692775,
      "loss": 6.4013,
      "mean_token_accuracy": 0.12957659140229225,
      "num_tokens": 9727910.0,
      "step": 4490
    },
    {
      "entropy": 6.6448219299316404,
      "epoch": 0.48103162287976886,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004984155890623074,
      "loss": 6.494,
      "mean_token_accuracy": 0.13356658741831778,
      "num_tokens": 9738528.0,
      "step": 4495
    },
    {
      "entropy": 6.6721742153167725,
      "epoch": 0.4815666969875328,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004984110565042435,
      "loss": 6.5053,
      "mean_token_accuracy": 0.12528721392154693,
      "num_tokens": 9749744.0,
      "step": 4500
    },
    {
      "entropy": 6.722189235687256,
      "epoch": 0.48210177109529667,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004984065174952168,
      "loss": 6.5162,
      "mean_token_accuracy": 0.12471330910921097,
      "num_tokens": 9759968.0,
      "step": 4505
    },
    {
      "entropy": 6.647542953491211,
      "epoch": 0.4826368452030606,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004984019720353582,
      "loss": 6.5304,
      "mean_token_accuracy": 0.12465517371892929,
      "num_tokens": 9769791.0,
      "step": 4510
    },
    {
      "entropy": 6.664258527755737,
      "epoch": 0.48317191931082454,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983974201247991,
      "loss": 6.5143,
      "mean_token_accuracy": 0.13621895313262938,
      "num_tokens": 9780369.0,
      "step": 4515
    },
    {
      "entropy": 6.695696401596069,
      "epoch": 0.48370699341858847,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004983928617636709,
      "loss": 6.5391,
      "mean_token_accuracy": 0.1274557776749134,
      "num_tokens": 9791125.0,
      "step": 4520
    },
    {
      "entropy": 6.72871413230896,
      "epoch": 0.4842420675263524,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004983882969521052,
      "loss": 6.5509,
      "mean_token_accuracy": 0.12490316405892372,
      "num_tokens": 9801697.0,
      "step": 4525
    },
    {
      "entropy": 6.783517837524414,
      "epoch": 0.48477714163411634,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004983837256902338,
      "loss": 6.5859,
      "mean_token_accuracy": 0.13063669353723525,
      "num_tokens": 9812171.0,
      "step": 4530
    },
    {
      "entropy": 6.709694766998291,
      "epoch": 0.48531221574188027,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004983791479781887,
      "loss": 6.4443,
      "mean_token_accuracy": 0.13055962696671486,
      "num_tokens": 9821596.0,
      "step": 4535
    },
    {
      "entropy": 6.698323726654053,
      "epoch": 0.48584728984964415,
      "grad_norm": 0.875,
      "learning_rate": 0.0004983745638161021,
      "loss": 6.5499,
      "mean_token_accuracy": 0.1214986227452755,
      "num_tokens": 9831634.0,
      "step": 4540
    },
    {
      "entropy": 6.669496822357178,
      "epoch": 0.4863823639574081,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004983699732041064,
      "loss": 6.5104,
      "mean_token_accuracy": 0.13144705891609193,
      "num_tokens": 9841662.0,
      "step": 4545
    },
    {
      "entropy": 6.707806730270386,
      "epoch": 0.486917438065172,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000498365376142334,
      "loss": 6.4959,
      "mean_token_accuracy": 0.1266673430800438,
      "num_tokens": 9853797.0,
      "step": 4550
    },
    {
      "entropy": 6.679695320129395,
      "epoch": 0.48745251217293595,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004983607726309179,
      "loss": 6.5456,
      "mean_token_accuracy": 0.13214531913399696,
      "num_tokens": 9865245.0,
      "step": 4555
    },
    {
      "entropy": 6.706293153762817,
      "epoch": 0.4879875862806999,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004983561626699908,
      "loss": 6.6246,
      "mean_token_accuracy": 0.11848078146576882,
      "num_tokens": 9876207.0,
      "step": 4560
    },
    {
      "entropy": 6.682948398590088,
      "epoch": 0.4885226603884638,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004983515462596859,
      "loss": 6.4545,
      "mean_token_accuracy": 0.1299903854727745,
      "num_tokens": 9887210.0,
      "step": 4565
    },
    {
      "entropy": 6.7182141780853275,
      "epoch": 0.48905773449622775,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004983469234001365,
      "loss": 6.6238,
      "mean_token_accuracy": 0.12008717283606529,
      "num_tokens": 9899607.0,
      "step": 4570
    },
    {
      "entropy": 6.620535135269165,
      "epoch": 0.4895928086039917,
      "grad_norm": 1.0,
      "learning_rate": 0.000498342294091476,
      "loss": 6.4071,
      "mean_token_accuracy": 0.1346537686884403,
      "num_tokens": 9909910.0,
      "step": 4575
    },
    {
      "entropy": 6.622541522979736,
      "epoch": 0.49012788271175556,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004983376583338382,
      "loss": 6.5393,
      "mean_token_accuracy": 0.12581005394458772,
      "num_tokens": 9922137.0,
      "step": 4580
    },
    {
      "entropy": 6.662765979766846,
      "epoch": 0.4906629568195195,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004983330161273569,
      "loss": 6.5304,
      "mean_token_accuracy": 0.1304611496627331,
      "num_tokens": 9932514.0,
      "step": 4585
    },
    {
      "entropy": 6.667660045623779,
      "epoch": 0.4911980309272834,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004983283674721662,
      "loss": 6.5536,
      "mean_token_accuracy": 0.12558515146374702,
      "num_tokens": 9943054.0,
      "step": 4590
    },
    {
      "entropy": 6.77188024520874,
      "epoch": 0.49173310503504736,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004983237123684002,
      "loss": 6.5326,
      "mean_token_accuracy": 0.12073590010404586,
      "num_tokens": 9953791.0,
      "step": 4595
    },
    {
      "entropy": 6.65752100944519,
      "epoch": 0.4922681791428113,
      "grad_norm": 0.875,
      "learning_rate": 0.0004983190508161934,
      "loss": 6.4855,
      "mean_token_accuracy": 0.12589664459228517,
      "num_tokens": 9964782.0,
      "step": 4600
    },
    {
      "entropy": 6.7109299659729,
      "epoch": 0.4928032532505752,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004983143828156804,
      "loss": 6.5558,
      "mean_token_accuracy": 0.12979992628097534,
      "num_tokens": 9975843.0,
      "step": 4605
    },
    {
      "entropy": 6.713517475128174,
      "epoch": 0.49333832735833916,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004983097083669959,
      "loss": 6.5394,
      "mean_token_accuracy": 0.12991684526205063,
      "num_tokens": 9987060.0,
      "step": 4610
    },
    {
      "entropy": 6.6631989002227785,
      "epoch": 0.49387340146610303,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004983050274702751,
      "loss": 6.4872,
      "mean_token_accuracy": 0.12487173229455947,
      "num_tokens": 9997380.0,
      "step": 4615
    },
    {
      "entropy": 6.7034773349761965,
      "epoch": 0.49440847557386697,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004983003401256529,
      "loss": 6.5359,
      "mean_token_accuracy": 0.12999945655465125,
      "num_tokens": 10007955.0,
      "step": 4620
    },
    {
      "entropy": 6.818626832962036,
      "epoch": 0.4949435496816309,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004982956463332647,
      "loss": 6.6132,
      "mean_token_accuracy": 0.12420329973101615,
      "num_tokens": 10017645.0,
      "step": 4625
    },
    {
      "entropy": 6.686611795425415,
      "epoch": 0.49547862378939483,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004982909460932463,
      "loss": 6.546,
      "mean_token_accuracy": 0.12951503768563272,
      "num_tokens": 10027619.0,
      "step": 4630
    },
    {
      "entropy": 6.618823432922364,
      "epoch": 0.49601369789715877,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000498286239405733,
      "loss": 6.4206,
      "mean_token_accuracy": 0.1362138234078884,
      "num_tokens": 10037551.0,
      "step": 4635
    },
    {
      "entropy": 6.655293130874634,
      "epoch": 0.4965487720049227,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000498281526270861,
      "loss": 6.4505,
      "mean_token_accuracy": 0.13801832348108292,
      "num_tokens": 10047771.0,
      "step": 4640
    },
    {
      "entropy": 6.7661138534545895,
      "epoch": 0.49708384611268663,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004982768066887663,
      "loss": 6.5863,
      "mean_token_accuracy": 0.11948011070489883,
      "num_tokens": 10058771.0,
      "step": 4645
    },
    {
      "entropy": 6.655697870254516,
      "epoch": 0.4976189202204505,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004982720806595851,
      "loss": 6.4608,
      "mean_token_accuracy": 0.13360660597682,
      "num_tokens": 10070318.0,
      "step": 4650
    },
    {
      "entropy": 6.634393501281738,
      "epoch": 0.49815399432821444,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004982673481834541,
      "loss": 6.4953,
      "mean_token_accuracy": 0.12757233008742333,
      "num_tokens": 10081810.0,
      "step": 4655
    },
    {
      "entropy": 6.659189367294312,
      "epoch": 0.4986890684359784,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004982626092605097,
      "loss": 6.4983,
      "mean_token_accuracy": 0.12614117041230202,
      "num_tokens": 10091724.0,
      "step": 4660
    },
    {
      "entropy": 6.691552066802979,
      "epoch": 0.4992241425437423,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004982578638908888,
      "loss": 6.4887,
      "mean_token_accuracy": 0.12917553260922432,
      "num_tokens": 10102510.0,
      "step": 4665
    },
    {
      "entropy": 6.649389266967773,
      "epoch": 0.49975921665150624,
      "grad_norm": 0.875,
      "learning_rate": 0.0004982531120747286,
      "loss": 6.5663,
      "mean_token_accuracy": 0.12157985270023346,
      "num_tokens": 10112636.0,
      "step": 4670
    },
    {
      "entropy": 6.655121469497681,
      "epoch": 0.5002942907592701,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000498248353812166,
      "loss": 6.5482,
      "mean_token_accuracy": 0.13074629306793212,
      "num_tokens": 10123562.0,
      "step": 4675
    },
    {
      "entropy": 6.707145500183105,
      "epoch": 0.5008293648670341,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004982435891033387,
      "loss": 6.5227,
      "mean_token_accuracy": 0.11998097375035285,
      "num_tokens": 10134731.0,
      "step": 4680
    },
    {
      "entropy": 6.716118907928466,
      "epoch": 0.501364438974798,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000498238817948384,
      "loss": 6.503,
      "mean_token_accuracy": 0.1241519220173359,
      "num_tokens": 10145232.0,
      "step": 4685
    },
    {
      "entropy": 6.6831278800964355,
      "epoch": 0.501899513082562,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004982340403474398,
      "loss": 6.4747,
      "mean_token_accuracy": 0.12642326802015305,
      "num_tokens": 10155857.0,
      "step": 4690
    },
    {
      "entropy": 6.655167865753174,
      "epoch": 0.5024345871903259,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004982292563006441,
      "loss": 6.4852,
      "mean_token_accuracy": 0.13339488878846167,
      "num_tokens": 10166145.0,
      "step": 4695
    },
    {
      "entropy": 6.720065402984619,
      "epoch": 0.5029696612980897,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000498224465808135,
      "loss": 6.6101,
      "mean_token_accuracy": 0.12126021236181259,
      "num_tokens": 10177086.0,
      "step": 4700
    },
    {
      "entropy": 6.721355295181274,
      "epoch": 0.5035047354058537,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004982196688700509,
      "loss": 6.4768,
      "mean_token_accuracy": 0.12901106253266334,
      "num_tokens": 10187120.0,
      "step": 4705
    },
    {
      "entropy": 6.66040358543396,
      "epoch": 0.5040398095136176,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004982148654865302,
      "loss": 6.5379,
      "mean_token_accuracy": 0.12791489511728288,
      "num_tokens": 10197784.0,
      "step": 4710
    },
    {
      "entropy": 6.690552377700806,
      "epoch": 0.5045748836213816,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004982100556577115,
      "loss": 6.394,
      "mean_token_accuracy": 0.13775371462106706,
      "num_tokens": 10209373.0,
      "step": 4715
    },
    {
      "entropy": 6.676543092727661,
      "epoch": 0.5051099577291455,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004982052393837338,
      "loss": 6.5497,
      "mean_token_accuracy": 0.13146620616316795,
      "num_tokens": 10220316.0,
      "step": 4720
    },
    {
      "entropy": 6.618153238296509,
      "epoch": 0.5056450318369095,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004982004166647363,
      "loss": 6.42,
      "mean_token_accuracy": 0.13332393765449524,
      "num_tokens": 10230478.0,
      "step": 4725
    },
    {
      "entropy": 6.578539705276489,
      "epoch": 0.5061801059446733,
      "grad_norm": 0.875,
      "learning_rate": 0.0004981955875008582,
      "loss": 6.5266,
      "mean_token_accuracy": 0.12779373079538345,
      "num_tokens": 10240641.0,
      "step": 4730
    },
    {
      "entropy": 6.663200330734253,
      "epoch": 0.5067151800524372,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004981907518922387,
      "loss": 6.488,
      "mean_token_accuracy": 0.1358850859105587,
      "num_tokens": 10250676.0,
      "step": 4735
    },
    {
      "entropy": 6.759254455566406,
      "epoch": 0.5072502541602012,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004981859098390177,
      "loss": 6.5517,
      "mean_token_accuracy": 0.12901933640241622,
      "num_tokens": 10262016.0,
      "step": 4740
    },
    {
      "entropy": 6.661891937255859,
      "epoch": 0.5077853282679651,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004981810613413348,
      "loss": 6.5024,
      "mean_token_accuracy": 0.1266100212931633,
      "num_tokens": 10271985.0,
      "step": 4745
    },
    {
      "entropy": 6.6667498588562015,
      "epoch": 0.5083204023757291,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981762063993302,
      "loss": 6.4803,
      "mean_token_accuracy": 0.1339687556028366,
      "num_tokens": 10282156.0,
      "step": 4750
    },
    {
      "entropy": 6.594398450851441,
      "epoch": 0.5088554764834929,
      "grad_norm": 0.90625,
      "learning_rate": 0.000498171345013144,
      "loss": 6.4911,
      "mean_token_accuracy": 0.12266959324479103,
      "num_tokens": 10292576.0,
      "step": 4755
    },
    {
      "entropy": 6.635583543777466,
      "epoch": 0.5093905505912569,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004981664771829165,
      "loss": 6.4902,
      "mean_token_accuracy": 0.13049110919237136,
      "num_tokens": 10303648.0,
      "step": 4760
    },
    {
      "entropy": 6.659964656829834,
      "epoch": 0.5099256246990208,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004981616029087884,
      "loss": 6.3731,
      "mean_token_accuracy": 0.14045070260763168,
      "num_tokens": 10314275.0,
      "step": 4765
    },
    {
      "entropy": 6.65841326713562,
      "epoch": 0.5104606988067847,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004981567221909004,
      "loss": 6.5144,
      "mean_token_accuracy": 0.1222583793103695,
      "num_tokens": 10325386.0,
      "step": 4770
    },
    {
      "entropy": 6.664293098449707,
      "epoch": 0.5109957729145487,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004981518350293935,
      "loss": 6.4803,
      "mean_token_accuracy": 0.12706643268465995,
      "num_tokens": 10335041.0,
      "step": 4775
    },
    {
      "entropy": 6.6265764236450195,
      "epoch": 0.5115308470223126,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004981469414244086,
      "loss": 6.4401,
      "mean_token_accuracy": 0.13401475921273232,
      "num_tokens": 10345746.0,
      "step": 4780
    },
    {
      "entropy": 6.611003446578979,
      "epoch": 0.5120659211300765,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000498142041376087,
      "loss": 6.533,
      "mean_token_accuracy": 0.13049467653036118,
      "num_tokens": 10356995.0,
      "step": 4785
    },
    {
      "entropy": 6.653438758850098,
      "epoch": 0.5126009952378404,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004981371348845705,
      "loss": 6.4605,
      "mean_token_accuracy": 0.13274952098727227,
      "num_tokens": 10368445.0,
      "step": 4790
    },
    {
      "entropy": 6.68891224861145,
      "epoch": 0.5131360693456044,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004981322219500006,
      "loss": 6.4091,
      "mean_token_accuracy": 0.13305947482585906,
      "num_tokens": 10378256.0,
      "step": 4795
    },
    {
      "entropy": 6.65896463394165,
      "epoch": 0.5136711434533683,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000498127302572519,
      "loss": 6.5043,
      "mean_token_accuracy": 0.13102471306920052,
      "num_tokens": 10388698.0,
      "step": 4800
    },
    {
      "entropy": 6.588460779190063,
      "epoch": 0.5142062175611322,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981223767522679,
      "loss": 6.4969,
      "mean_token_accuracy": 0.13006809875369071,
      "num_tokens": 10399975.0,
      "step": 4805
    },
    {
      "entropy": 6.686613750457764,
      "epoch": 0.5147412916688962,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004981174444893896,
      "loss": 6.5067,
      "mean_token_accuracy": 0.1265137203037739,
      "num_tokens": 10410480.0,
      "step": 4810
    },
    {
      "entropy": 6.665117025375366,
      "epoch": 0.51527636577666,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004981125057840264,
      "loss": 6.4575,
      "mean_token_accuracy": 0.13104326874017716,
      "num_tokens": 10421023.0,
      "step": 4815
    },
    {
      "entropy": 6.654813051223755,
      "epoch": 0.515811439884424,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000498107560636321,
      "loss": 6.5038,
      "mean_token_accuracy": 0.13196004256606103,
      "num_tokens": 10431152.0,
      "step": 4820
    },
    {
      "entropy": 6.657983207702637,
      "epoch": 0.5163465139921879,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004981026090464161,
      "loss": 6.4142,
      "mean_token_accuracy": 0.1262127973139286,
      "num_tokens": 10441986.0,
      "step": 4825
    },
    {
      "entropy": 6.6557539939880375,
      "epoch": 0.5168815880999519,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004980976510144548,
      "loss": 6.4844,
      "mean_token_accuracy": 0.1264335423707962,
      "num_tokens": 10452932.0,
      "step": 4830
    },
    {
      "entropy": 6.692061710357666,
      "epoch": 0.5174166622077158,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980926865405801,
      "loss": 6.5639,
      "mean_token_accuracy": 0.13087983354926108,
      "num_tokens": 10463411.0,
      "step": 4835
    },
    {
      "entropy": 6.676014041900634,
      "epoch": 0.5179517363154797,
      "grad_norm": 0.875,
      "learning_rate": 0.0004980877156249354,
      "loss": 6.4577,
      "mean_token_accuracy": 0.1286552980542183,
      "num_tokens": 10474820.0,
      "step": 4840
    },
    {
      "entropy": 6.621929407119751,
      "epoch": 0.5184868104232436,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004980827382676643,
      "loss": 6.4737,
      "mean_token_accuracy": 0.1348734863102436,
      "num_tokens": 10485701.0,
      "step": 4845
    },
    {
      "entropy": 6.6411299228668215,
      "epoch": 0.5190218845310075,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004980777544689105,
      "loss": 6.4551,
      "mean_token_accuracy": 0.13278646394610405,
      "num_tokens": 10495825.0,
      "step": 4850
    },
    {
      "entropy": 6.6085044860839846,
      "epoch": 0.5195569586387715,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004980727642288178,
      "loss": 6.4436,
      "mean_token_accuracy": 0.1337565451860428,
      "num_tokens": 10505847.0,
      "step": 4855
    },
    {
      "entropy": 6.689910840988159,
      "epoch": 0.5200920327465354,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004980677675475305,
      "loss": 6.4972,
      "mean_token_accuracy": 0.1257264494895935,
      "num_tokens": 10516596.0,
      "step": 4860
    },
    {
      "entropy": 6.640674066543579,
      "epoch": 0.5206271068542994,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004980627644251926,
      "loss": 6.424,
      "mean_token_accuracy": 0.1361898109316826,
      "num_tokens": 10528175.0,
      "step": 4865
    },
    {
      "entropy": 6.587651538848877,
      "epoch": 0.5211621809620632,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004980577548619489,
      "loss": 6.5086,
      "mean_token_accuracy": 0.1265055663883686,
      "num_tokens": 10540076.0,
      "step": 4870
    },
    {
      "entropy": 6.67489595413208,
      "epoch": 0.5216972550698272,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004980527388579437,
      "loss": 6.5248,
      "mean_token_accuracy": 0.13013955950737,
      "num_tokens": 10552025.0,
      "step": 4875
    },
    {
      "entropy": 6.70230655670166,
      "epoch": 0.5222323291775911,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004980477164133221,
      "loss": 6.522,
      "mean_token_accuracy": 0.12733396738767624,
      "num_tokens": 10562535.0,
      "step": 4880
    },
    {
      "entropy": 6.617531108856201,
      "epoch": 0.522767403285355,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000498042687528229,
      "loss": 6.4483,
      "mean_token_accuracy": 0.132208301872015,
      "num_tokens": 10572989.0,
      "step": 4885
    },
    {
      "entropy": 6.6338588237762455,
      "epoch": 0.523302477393119,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004980376522028098,
      "loss": 6.4665,
      "mean_token_accuracy": 0.13189956918358803,
      "num_tokens": 10583033.0,
      "step": 4890
    },
    {
      "entropy": 6.690638208389283,
      "epoch": 0.5238375515008828,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004980326104372095,
      "loss": 6.5387,
      "mean_token_accuracy": 0.13196553736925126,
      "num_tokens": 10595205.0,
      "step": 4895
    },
    {
      "entropy": 6.6683584690094,
      "epoch": 0.5243726256086468,
      "grad_norm": 0.953125,
      "learning_rate": 0.000498027562231574,
      "loss": 6.3906,
      "mean_token_accuracy": 0.13348505273461342,
      "num_tokens": 10605244.0,
      "step": 4900
    },
    {
      "entropy": 6.6075310707092285,
      "epoch": 0.5249076997164107,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000498022507586049,
      "loss": 6.4814,
      "mean_token_accuracy": 0.12973640710115433,
      "num_tokens": 10615817.0,
      "step": 4905
    },
    {
      "entropy": 6.6691591262817385,
      "epoch": 0.5254427738241747,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004980174465007804,
      "loss": 6.4943,
      "mean_token_accuracy": 0.12603137716650964,
      "num_tokens": 10627160.0,
      "step": 4910
    },
    {
      "entropy": 6.656597852706909,
      "epoch": 0.5259778479319386,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980123789759145,
      "loss": 6.4877,
      "mean_token_accuracy": 0.13532912358641624,
      "num_tokens": 10637552.0,
      "step": 4915
    },
    {
      "entropy": 6.630317211151123,
      "epoch": 0.5265129220397025,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004980073050115973,
      "loss": 6.5171,
      "mean_token_accuracy": 0.13262762352824212,
      "num_tokens": 10648773.0,
      "step": 4920
    },
    {
      "entropy": 6.712649917602539,
      "epoch": 0.5270479961474664,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004980022246079755,
      "loss": 6.5196,
      "mean_token_accuracy": 0.12795575633645057,
      "num_tokens": 10659537.0,
      "step": 4925
    },
    {
      "entropy": 6.715398979187012,
      "epoch": 0.5275830702552303,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000497997137765196,
      "loss": 6.5583,
      "mean_token_accuracy": 0.12573519349098206,
      "num_tokens": 10671390.0,
      "step": 4930
    },
    {
      "entropy": 6.679694366455078,
      "epoch": 0.5281181443629943,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004979920444834054,
      "loss": 6.3888,
      "mean_token_accuracy": 0.14320642352104188,
      "num_tokens": 10682389.0,
      "step": 4935
    },
    {
      "entropy": 6.600399398803711,
      "epoch": 0.5286532184707582,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004979869447627508,
      "loss": 6.4609,
      "mean_token_accuracy": 0.131199461966753,
      "num_tokens": 10692400.0,
      "step": 4940
    },
    {
      "entropy": 6.561653804779053,
      "epoch": 0.5291882925785222,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004979818386033795,
      "loss": 6.3845,
      "mean_token_accuracy": 0.13536884859204293,
      "num_tokens": 10702396.0,
      "step": 4945
    },
    {
      "entropy": 6.637121248245239,
      "epoch": 0.5297233666862861,
      "grad_norm": 0.8046875,
      "learning_rate": 0.000497976726005439,
      "loss": 6.4988,
      "mean_token_accuracy": 0.13272100910544396,
      "num_tokens": 10712863.0,
      "step": 4950
    },
    {
      "entropy": 6.71592607498169,
      "epoch": 0.5302584407940499,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004979716069690768,
      "loss": 6.5057,
      "mean_token_accuracy": 0.13116242736577988,
      "num_tokens": 10724234.0,
      "step": 4955
    },
    {
      "entropy": 6.731963491439819,
      "epoch": 0.5307935149018139,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004979664814944409,
      "loss": 6.5279,
      "mean_token_accuracy": 0.12522164285182952,
      "num_tokens": 10734405.0,
      "step": 4960
    },
    {
      "entropy": 6.664322423934936,
      "epoch": 0.5313285890095778,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004979613495816791,
      "loss": 6.4711,
      "mean_token_accuracy": 0.13456571102142334,
      "num_tokens": 10744573.0,
      "step": 4965
    },
    {
      "entropy": 6.711752843856812,
      "epoch": 0.5318636631173418,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004979562112309397,
      "loss": 6.541,
      "mean_token_accuracy": 0.12778567150235176,
      "num_tokens": 10755555.0,
      "step": 4970
    },
    {
      "entropy": 6.626407051086426,
      "epoch": 0.5323987372251057,
      "grad_norm": 0.8046875,
      "learning_rate": 0.000497951066442371,
      "loss": 6.5437,
      "mean_token_accuracy": 0.12731776833534242,
      "num_tokens": 10766493.0,
      "step": 4975
    },
    {
      "entropy": 6.694499826431274,
      "epoch": 0.5329338113328697,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004979459152161215,
      "loss": 6.5025,
      "mean_token_accuracy": 0.12069962471723557,
      "num_tokens": 10776735.0,
      "step": 4980
    },
    {
      "entropy": 6.6064189910888675,
      "epoch": 0.5334688854406335,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004979407575523402,
      "loss": 6.4902,
      "mean_token_accuracy": 0.12979846075177193,
      "num_tokens": 10787500.0,
      "step": 4985
    },
    {
      "entropy": 6.638271999359131,
      "epoch": 0.5340039595483974,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004979355934511757,
      "loss": 6.5058,
      "mean_token_accuracy": 0.12979728877544403,
      "num_tokens": 10798002.0,
      "step": 4990
    },
    {
      "entropy": 6.662521028518677,
      "epoch": 0.5345390336561614,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004979304229127773,
      "loss": 6.4538,
      "mean_token_accuracy": 0.13369745686650275,
      "num_tokens": 10808982.0,
      "step": 4995
    },
    {
      "entropy": 6.631230354309082,
      "epoch": 0.5350741077639253,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004979252459372944,
      "loss": 6.4146,
      "mean_token_accuracy": 0.1316783122718334,
      "num_tokens": 10821619.0,
      "step": 5000
    },
    {
      "entropy": 6.617094469070435,
      "epoch": 0.5356091818716893,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004979200625248761,
      "loss": 6.4536,
      "mean_token_accuracy": 0.13066781982779502,
      "num_tokens": 10832344.0,
      "step": 5005
    },
    {
      "entropy": 6.660971975326538,
      "epoch": 0.5361442559794531,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004979148726756725,
      "loss": 6.4217,
      "mean_token_accuracy": 0.12439092099666596,
      "num_tokens": 10843149.0,
      "step": 5010
    },
    {
      "entropy": 6.6501552104949955,
      "epoch": 0.5366793300872171,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004979096763898333,
      "loss": 6.4666,
      "mean_token_accuracy": 0.1364528313279152,
      "num_tokens": 10854590.0,
      "step": 5015
    },
    {
      "entropy": 6.684966707229615,
      "epoch": 0.537214404194981,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004979044736675084,
      "loss": 6.4394,
      "mean_token_accuracy": 0.12906123027205468,
      "num_tokens": 10865500.0,
      "step": 5020
    },
    {
      "entropy": 6.601755046844483,
      "epoch": 0.5377494783027449,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004978992645088483,
      "loss": 6.4624,
      "mean_token_accuracy": 0.14228275641798974,
      "num_tokens": 10876866.0,
      "step": 5025
    },
    {
      "entropy": 6.52980055809021,
      "epoch": 0.5382845524105089,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000497894048914003,
      "loss": 6.4285,
      "mean_token_accuracy": 0.12853240743279457,
      "num_tokens": 10888627.0,
      "step": 5030
    },
    {
      "entropy": 6.70817985534668,
      "epoch": 0.5388196265182728,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004978888268831236,
      "loss": 6.5347,
      "mean_token_accuracy": 0.12227588817477227,
      "num_tokens": 10900992.0,
      "step": 5035
    },
    {
      "entropy": 6.764741897583008,
      "epoch": 0.5393547006260367,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004978835984163606,
      "loss": 6.5723,
      "mean_token_accuracy": 0.11759168133139611,
      "num_tokens": 10912323.0,
      "step": 5040
    },
    {
      "entropy": 6.626711463928222,
      "epoch": 0.5398897747338006,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004978783635138649,
      "loss": 6.5012,
      "mean_token_accuracy": 0.1324291668832302,
      "num_tokens": 10923991.0,
      "step": 5045
    },
    {
      "entropy": 6.616898250579834,
      "epoch": 0.5404248488415646,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004978731221757878,
      "loss": 6.3803,
      "mean_token_accuracy": 0.13645441010594367,
      "num_tokens": 10934081.0,
      "step": 5050
    },
    {
      "entropy": 6.646523284912109,
      "epoch": 0.5409599229493285,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004978678744022808,
      "loss": 6.4452,
      "mean_token_accuracy": 0.13217141032218932,
      "num_tokens": 10944729.0,
      "step": 5055
    },
    {
      "entropy": 6.668096733093262,
      "epoch": 0.5414949970570924,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000497862620193495,
      "loss": 6.5113,
      "mean_token_accuracy": 0.12759771272540094,
      "num_tokens": 10955708.0,
      "step": 5060
    },
    {
      "entropy": 6.62319049835205,
      "epoch": 0.5420300711648564,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004978573595495826,
      "loss": 6.4556,
      "mean_token_accuracy": 0.13306488543748857,
      "num_tokens": 10966427.0,
      "step": 5065
    },
    {
      "entropy": 6.63427619934082,
      "epoch": 0.5425651452726202,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000497852092470695,
      "loss": 6.4834,
      "mean_token_accuracy": 0.13202762305736543,
      "num_tokens": 10977741.0,
      "step": 5070
    },
    {
      "entropy": 6.672744941711426,
      "epoch": 0.5431002193803842,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004978468189569847,
      "loss": 6.445,
      "mean_token_accuracy": 0.13317141979932784,
      "num_tokens": 10989127.0,
      "step": 5075
    },
    {
      "entropy": 6.592401599884033,
      "epoch": 0.5436352934881481,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004978415390086038,
      "loss": 6.4634,
      "mean_token_accuracy": 0.12352445498108863,
      "num_tokens": 10999791.0,
      "step": 5080
    },
    {
      "entropy": 6.687466955184936,
      "epoch": 0.5441703675959121,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004978362526257047,
      "loss": 6.5355,
      "mean_token_accuracy": 0.12518023177981377,
      "num_tokens": 11010045.0,
      "step": 5085
    },
    {
      "entropy": 6.659976673126221,
      "epoch": 0.544705441703676,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004978309598084403,
      "loss": 6.4536,
      "mean_token_accuracy": 0.1302490249276161,
      "num_tokens": 11020729.0,
      "step": 5090
    },
    {
      "entropy": 6.654502201080322,
      "epoch": 0.5452405158114398,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004978256605569631,
      "loss": 6.4915,
      "mean_token_accuracy": 0.1297528125345707,
      "num_tokens": 11030641.0,
      "step": 5095
    },
    {
      "entropy": 6.650192928314209,
      "epoch": 0.5457755899192038,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004978203548714262,
      "loss": 6.4833,
      "mean_token_accuracy": 0.12867137864232064,
      "num_tokens": 11041533.0,
      "step": 5100
    },
    {
      "entropy": 6.626247215270996,
      "epoch": 0.5463106640269677,
      "grad_norm": 0.78125,
      "learning_rate": 0.000497815042751983,
      "loss": 6.4235,
      "mean_token_accuracy": 0.12988842129707337,
      "num_tokens": 11052170.0,
      "step": 5105
    },
    {
      "entropy": 6.532457256317139,
      "epoch": 0.5468457381347317,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004978097241987868,
      "loss": 6.4025,
      "mean_token_accuracy": 0.1317355863749981,
      "num_tokens": 11062872.0,
      "step": 5110
    },
    {
      "entropy": 6.630930852890015,
      "epoch": 0.5473808122424956,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000497804399211991,
      "loss": 6.506,
      "mean_token_accuracy": 0.12932709231972694,
      "num_tokens": 11074199.0,
      "step": 5115
    },
    {
      "entropy": 6.651639652252197,
      "epoch": 0.5479158863502596,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004977990677917495,
      "loss": 6.459,
      "mean_token_accuracy": 0.13692924305796622,
      "num_tokens": 11083881.0,
      "step": 5120
    },
    {
      "entropy": 6.591510486602783,
      "epoch": 0.5484509604580234,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004977937299382162,
      "loss": 6.505,
      "mean_token_accuracy": 0.12828393876552582,
      "num_tokens": 11094374.0,
      "step": 5125
    },
    {
      "entropy": 6.615193510055542,
      "epoch": 0.5489860345657873,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004977883856515453,
      "loss": 6.3454,
      "mean_token_accuracy": 0.14302551373839378,
      "num_tokens": 11105174.0,
      "step": 5130
    },
    {
      "entropy": 6.625015830993652,
      "epoch": 0.5495211086735513,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000497783034931891,
      "loss": 6.5183,
      "mean_token_accuracy": 0.13052447959780694,
      "num_tokens": 11117223.0,
      "step": 5135
    },
    {
      "entropy": 6.631676435470581,
      "epoch": 0.5500561827813152,
      "grad_norm": 0.875,
      "learning_rate": 0.000497777677779408,
      "loss": 6.4426,
      "mean_token_accuracy": 0.1306950606405735,
      "num_tokens": 11127240.0,
      "step": 5140
    },
    {
      "entropy": 6.593652677536011,
      "epoch": 0.5505912568890792,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004977723141942509,
      "loss": 6.4036,
      "mean_token_accuracy": 0.1429332561790943,
      "num_tokens": 11137287.0,
      "step": 5145
    },
    {
      "entropy": 6.623790645599366,
      "epoch": 0.551126330996843,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004977669441765743,
      "loss": 6.582,
      "mean_token_accuracy": 0.12360026016831398,
      "num_tokens": 11148988.0,
      "step": 5150
    },
    {
      "entropy": 6.6897046089172365,
      "epoch": 0.551661405104607,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004977615677265336,
      "loss": 6.486,
      "mean_token_accuracy": 0.12912213951349258,
      "num_tokens": 11159874.0,
      "step": 5155
    },
    {
      "entropy": 6.688545560836792,
      "epoch": 0.5521964792123709,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000497756184844284,
      "loss": 6.4355,
      "mean_token_accuracy": 0.13699986264109612,
      "num_tokens": 11170703.0,
      "step": 5160
    },
    {
      "entropy": 6.5400135040283205,
      "epoch": 0.5527315533201348,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004977507955299809,
      "loss": 6.3297,
      "mean_token_accuracy": 0.13553624749183654,
      "num_tokens": 11182230.0,
      "step": 5165
    },
    {
      "entropy": 6.573509454727173,
      "epoch": 0.5532666274278988,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004977453997837797,
      "loss": 6.4224,
      "mean_token_accuracy": 0.12132932245731354,
      "num_tokens": 11192897.0,
      "step": 5170
    },
    {
      "entropy": 6.608253240585327,
      "epoch": 0.5538017015356627,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004977399976058366,
      "loss": 6.495,
      "mean_token_accuracy": 0.12757838517427444,
      "num_tokens": 11203942.0,
      "step": 5175
    },
    {
      "entropy": 6.690961456298828,
      "epoch": 0.5543367756434266,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004977345889963072,
      "loss": 6.548,
      "mean_token_accuracy": 0.12905979976058007,
      "num_tokens": 11214361.0,
      "step": 5180
    },
    {
      "entropy": 6.592653083801269,
      "epoch": 0.5548718497511905,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004977291739553479,
      "loss": 6.4476,
      "mean_token_accuracy": 0.13367174714803695,
      "num_tokens": 11225928.0,
      "step": 5185
    },
    {
      "entropy": 6.604433345794678,
      "epoch": 0.5554069238589545,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004977237524831149,
      "loss": 6.4047,
      "mean_token_accuracy": 0.12101099863648415,
      "num_tokens": 11236660.0,
      "step": 5190
    },
    {
      "entropy": 6.544157266616821,
      "epoch": 0.5559419979667184,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004977183245797649,
      "loss": 6.4277,
      "mean_token_accuracy": 0.13961708843708037,
      "num_tokens": 11247272.0,
      "step": 5195
    },
    {
      "entropy": 6.64446496963501,
      "epoch": 0.5564770720744823,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004977128902454547,
      "loss": 6.529,
      "mean_token_accuracy": 0.12302258089184762,
      "num_tokens": 11258472.0,
      "step": 5200
    },
    {
      "entropy": 6.697921371459961,
      "epoch": 0.5570121461822463,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004977074494803409,
      "loss": 6.4213,
      "mean_token_accuracy": 0.13686134666204453,
      "num_tokens": 11268344.0,
      "step": 5205
    },
    {
      "entropy": 6.626495981216431,
      "epoch": 0.5575472202900101,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004977020022845809,
      "loss": 6.5256,
      "mean_token_accuracy": 0.12177680134773254,
      "num_tokens": 11279931.0,
      "step": 5210
    },
    {
      "entropy": 6.688230848312378,
      "epoch": 0.5580822943977741,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004976965486583318,
      "loss": 6.4646,
      "mean_token_accuracy": 0.1283648930490017,
      "num_tokens": 11290595.0,
      "step": 5215
    },
    {
      "entropy": 6.610369396209717,
      "epoch": 0.558617368505538,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004976910886017513,
      "loss": 6.4855,
      "mean_token_accuracy": 0.1312863752245903,
      "num_tokens": 11300774.0,
      "step": 5220
    },
    {
      "entropy": 6.611403465270996,
      "epoch": 0.559152442613302,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004976856221149969,
      "loss": 6.3636,
      "mean_token_accuracy": 0.14082069247961043,
      "num_tokens": 11310288.0,
      "step": 5225
    },
    {
      "entropy": 6.605205345153808,
      "epoch": 0.5596875167210659,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004976801491982263,
      "loss": 6.4576,
      "mean_token_accuracy": 0.1295248970389366,
      "num_tokens": 11321599.0,
      "step": 5230
    },
    {
      "entropy": 6.677119493484497,
      "epoch": 0.5602225908288297,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004976746698515977,
      "loss": 6.504,
      "mean_token_accuracy": 0.13436227515339852,
      "num_tokens": 11331383.0,
      "step": 5235
    },
    {
      "entropy": 6.601748704910278,
      "epoch": 0.5607576649365937,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004976691840752694,
      "loss": 6.4686,
      "mean_token_accuracy": 0.12666866481304168,
      "num_tokens": 11342085.0,
      "step": 5240
    },
    {
      "entropy": 6.649465560913086,
      "epoch": 0.5612927390443576,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004976636918693996,
      "loss": 6.5105,
      "mean_token_accuracy": 0.12576734274625778,
      "num_tokens": 11354061.0,
      "step": 5245
    },
    {
      "entropy": 6.718101549148559,
      "epoch": 0.5618278131521216,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004976581932341471,
      "loss": 6.4453,
      "mean_token_accuracy": 0.1351204000413418,
      "num_tokens": 11364448.0,
      "step": 5250
    },
    {
      "entropy": 6.71000394821167,
      "epoch": 0.5623628872598855,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004976526881696706,
      "loss": 6.4108,
      "mean_token_accuracy": 0.13188011795282364,
      "num_tokens": 11374838.0,
      "step": 5255
    },
    {
      "entropy": 6.52580189704895,
      "epoch": 0.5628979613676495,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004976471766761288,
      "loss": 6.4683,
      "mean_token_accuracy": 0.13095718398690223,
      "num_tokens": 11385310.0,
      "step": 5260
    },
    {
      "entropy": 6.633668661117554,
      "epoch": 0.5634330354754133,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004976416587536812,
      "loss": 6.4553,
      "mean_token_accuracy": 0.12800781652331353,
      "num_tokens": 11396590.0,
      "step": 5265
    },
    {
      "entropy": 6.64537558555603,
      "epoch": 0.5639681095831772,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004976361344024871,
      "loss": 6.425,
      "mean_token_accuracy": 0.1297707349061966,
      "num_tokens": 11407218.0,
      "step": 5270
    },
    {
      "entropy": 6.509176731109619,
      "epoch": 0.5645031836909412,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004976306036227056,
      "loss": 6.4438,
      "mean_token_accuracy": 0.12941014617681504,
      "num_tokens": 11417716.0,
      "step": 5275
    },
    {
      "entropy": 6.691148519515991,
      "epoch": 0.5650382577987051,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000497625066414497,
      "loss": 6.5224,
      "mean_token_accuracy": 0.12348874360322952,
      "num_tokens": 11428295.0,
      "step": 5280
    },
    {
      "entropy": 6.729833936691284,
      "epoch": 0.5655733319064691,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004976195227780207,
      "loss": 6.4514,
      "mean_token_accuracy": 0.12749890610575676,
      "num_tokens": 11439044.0,
      "step": 5285
    },
    {
      "entropy": 6.631151819229126,
      "epoch": 0.566108406014233,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004976139727134371,
      "loss": 6.4106,
      "mean_token_accuracy": 0.12706983014941214,
      "num_tokens": 11450029.0,
      "step": 5290
    },
    {
      "entropy": 6.603351879119873,
      "epoch": 0.566643480121997,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004976084162209062,
      "loss": 6.5689,
      "mean_token_accuracy": 0.12012592554092408,
      "num_tokens": 11461190.0,
      "step": 5295
    },
    {
      "entropy": 6.639957475662231,
      "epoch": 0.5671785542297608,
      "grad_norm": 0.765625,
      "learning_rate": 0.0004976028533005886,
      "loss": 6.4234,
      "mean_token_accuracy": 0.13233305662870407,
      "num_tokens": 11472143.0,
      "step": 5300
    },
    {
      "entropy": 6.648490476608276,
      "epoch": 0.5677136283375247,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004975972839526449,
      "loss": 6.5461,
      "mean_token_accuracy": 0.12183723077178002,
      "num_tokens": 11483745.0,
      "step": 5305
    },
    {
      "entropy": 6.626348638534546,
      "epoch": 0.5682487024452887,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004975917081772358,
      "loss": 6.4137,
      "mean_token_accuracy": 0.12992365583777427,
      "num_tokens": 11495695.0,
      "step": 5310
    },
    {
      "entropy": 6.653946590423584,
      "epoch": 0.5687837765530526,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004975861259745225,
      "loss": 6.4558,
      "mean_token_accuracy": 0.13149744644761086,
      "num_tokens": 11505576.0,
      "step": 5315
    },
    {
      "entropy": 6.5848089218139645,
      "epoch": 0.5693188506608166,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004975805373446662,
      "loss": 6.31,
      "mean_token_accuracy": 0.13852847293019294,
      "num_tokens": 11516614.0,
      "step": 5320
    },
    {
      "entropy": 6.6240356922149655,
      "epoch": 0.5698539247685804,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000497574942287828,
      "loss": 6.4482,
      "mean_token_accuracy": 0.1252579554915428,
      "num_tokens": 11528032.0,
      "step": 5325
    },
    {
      "entropy": 6.602037477493286,
      "epoch": 0.5703889988763444,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004975693408041697,
      "loss": 6.4417,
      "mean_token_accuracy": 0.127547724545002,
      "num_tokens": 11538822.0,
      "step": 5330
    },
    {
      "entropy": 6.650227212905884,
      "epoch": 0.5709240729841083,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004975637328938529,
      "loss": 6.4659,
      "mean_token_accuracy": 0.12432368323206902,
      "num_tokens": 11548895.0,
      "step": 5335
    },
    {
      "entropy": 6.596523475646973,
      "epoch": 0.5714591470918722,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004975581185570398,
      "loss": 6.3762,
      "mean_token_accuracy": 0.12635455653071404,
      "num_tokens": 11559613.0,
      "step": 5340
    },
    {
      "entropy": 6.602152347564697,
      "epoch": 0.5719942211996362,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004975524977938921,
      "loss": 6.3939,
      "mean_token_accuracy": 0.13863522708415985,
      "num_tokens": 11571092.0,
      "step": 5345
    },
    {
      "entropy": 6.627554702758789,
      "epoch": 0.5725292953074,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004975468706045722,
      "loss": 6.5225,
      "mean_token_accuracy": 0.12761716544628143,
      "num_tokens": 11581448.0,
      "step": 5350
    },
    {
      "entropy": 6.636335706710815,
      "epoch": 0.573064369415164,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004975412369892429,
      "loss": 6.4315,
      "mean_token_accuracy": 0.13777912333607673,
      "num_tokens": 11592273.0,
      "step": 5355
    },
    {
      "entropy": 6.644360065460205,
      "epoch": 0.5735994435229279,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004975355969480665,
      "loss": 6.4013,
      "mean_token_accuracy": 0.13845689967274666,
      "num_tokens": 11602759.0,
      "step": 5360
    },
    {
      "entropy": 6.60584716796875,
      "epoch": 0.5741345176306919,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004975299504812061,
      "loss": 6.4341,
      "mean_token_accuracy": 0.1288062922656536,
      "num_tokens": 11613977.0,
      "step": 5365
    },
    {
      "entropy": 6.668964147567749,
      "epoch": 0.5746695917384558,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004975242975888246,
      "loss": 6.4656,
      "mean_token_accuracy": 0.13144578859210015,
      "num_tokens": 11625425.0,
      "step": 5370
    },
    {
      "entropy": 6.5949547290802,
      "epoch": 0.5752046658462197,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004975186382710854,
      "loss": 6.4424,
      "mean_token_accuracy": 0.1307839035987854,
      "num_tokens": 11636530.0,
      "step": 5375
    },
    {
      "entropy": 6.643009614944458,
      "epoch": 0.5757397399539836,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004975129725281516,
      "loss": 6.4347,
      "mean_token_accuracy": 0.13281472772359848,
      "num_tokens": 11646516.0,
      "step": 5380
    },
    {
      "entropy": 6.616060066223144,
      "epoch": 0.5762748140617475,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004975073003601871,
      "loss": 6.4113,
      "mean_token_accuracy": 0.12877700477838516,
      "num_tokens": 11657424.0,
      "step": 5385
    },
    {
      "entropy": 6.644890213012696,
      "epoch": 0.5768098881695115,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004975016217673556,
      "loss": 6.4461,
      "mean_token_accuracy": 0.12669419050216674,
      "num_tokens": 11669447.0,
      "step": 5390
    },
    {
      "entropy": 6.60390157699585,
      "epoch": 0.5773449622772754,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000497495936749821,
      "loss": 6.4026,
      "mean_token_accuracy": 0.13998956829309464,
      "num_tokens": 11679612.0,
      "step": 5395
    },
    {
      "entropy": 6.594708633422852,
      "epoch": 0.5778800363850394,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004974902453077474,
      "loss": 6.3853,
      "mean_token_accuracy": 0.1350052125751972,
      "num_tokens": 11689527.0,
      "step": 5400
    },
    {
      "entropy": 6.647299957275391,
      "epoch": 0.5784151104928033,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004974845474412993,
      "loss": 6.4248,
      "mean_token_accuracy": 0.13386151045560837,
      "num_tokens": 11700452.0,
      "step": 5405
    },
    {
      "entropy": 6.546576118469238,
      "epoch": 0.5789501846005671,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004974788431506412,
      "loss": 6.4383,
      "mean_token_accuracy": 0.12854185029864312,
      "num_tokens": 11710902.0,
      "step": 5410
    },
    {
      "entropy": 6.623003578186035,
      "epoch": 0.5794852587083311,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004974731324359378,
      "loss": 6.448,
      "mean_token_accuracy": 0.1285107046365738,
      "num_tokens": 11721901.0,
      "step": 5415
    },
    {
      "entropy": 6.633054113388061,
      "epoch": 0.580020332816095,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004974674152973538,
      "loss": 6.4406,
      "mean_token_accuracy": 0.12691360265016555,
      "num_tokens": 11733622.0,
      "step": 5420
    },
    {
      "entropy": 6.659692049026489,
      "epoch": 0.580555406923859,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004974616917350545,
      "loss": 6.4502,
      "mean_token_accuracy": 0.1322809189558029,
      "num_tokens": 11745116.0,
      "step": 5425
    },
    {
      "entropy": 6.534192943572998,
      "epoch": 0.5810904810316229,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004974559617492052,
      "loss": 6.3659,
      "mean_token_accuracy": 0.14103469774127006,
      "num_tokens": 11756203.0,
      "step": 5430
    },
    {
      "entropy": 6.622519063949585,
      "epoch": 0.5816255551393869,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004974502253399712,
      "loss": 6.426,
      "mean_token_accuracy": 0.1334656834602356,
      "num_tokens": 11766797.0,
      "step": 5435
    },
    {
      "entropy": 6.691503620147705,
      "epoch": 0.5821606292471507,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974444825075182,
      "loss": 6.5263,
      "mean_token_accuracy": 0.12716197073459626,
      "num_tokens": 11778110.0,
      "step": 5440
    },
    {
      "entropy": 6.556562280654907,
      "epoch": 0.5826957033549146,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497438733252012,
      "loss": 6.3758,
      "mean_token_accuracy": 0.13210416957736015,
      "num_tokens": 11788396.0,
      "step": 5445
    },
    {
      "entropy": 6.653188276290893,
      "epoch": 0.5832307774626786,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004974329775736188,
      "loss": 6.4567,
      "mean_token_accuracy": 0.13075546100735663,
      "num_tokens": 11798691.0,
      "step": 5450
    },
    {
      "entropy": 6.678783988952636,
      "epoch": 0.5837658515704425,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974272154725044,
      "loss": 6.5191,
      "mean_token_accuracy": 0.12994415536522866,
      "num_tokens": 11808965.0,
      "step": 5455
    },
    {
      "entropy": 6.611812782287598,
      "epoch": 0.5843009256782065,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004974214469488355,
      "loss": 6.3422,
      "mean_token_accuracy": 0.13759486824274064,
      "num_tokens": 11819932.0,
      "step": 5460
    },
    {
      "entropy": 6.634679317474365,
      "epoch": 0.5848359997859703,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004974156720027786,
      "loss": 6.453,
      "mean_token_accuracy": 0.1301378272473812,
      "num_tokens": 11831332.0,
      "step": 5465
    },
    {
      "entropy": 6.622493648529053,
      "epoch": 0.5853710738937343,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004974098906345004,
      "loss": 6.4313,
      "mean_token_accuracy": 0.12657588049769403,
      "num_tokens": 11840947.0,
      "step": 5470
    },
    {
      "entropy": 6.598509502410889,
      "epoch": 0.5859061480014982,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004974041028441679,
      "loss": 6.3798,
      "mean_token_accuracy": 0.12926354855298997,
      "num_tokens": 11851640.0,
      "step": 5475
    },
    {
      "entropy": 6.655171680450439,
      "epoch": 0.5864412221092621,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004973983086319481,
      "loss": 6.493,
      "mean_token_accuracy": 0.12535899356007577,
      "num_tokens": 11863608.0,
      "step": 5480
    },
    {
      "entropy": 6.631993961334229,
      "epoch": 0.5869762962170261,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004973925079980085,
      "loss": 6.3975,
      "mean_token_accuracy": 0.13125982657074928,
      "num_tokens": 11874707.0,
      "step": 5485
    },
    {
      "entropy": 6.593030309677124,
      "epoch": 0.58751137032479,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004973867009425164,
      "loss": 6.4779,
      "mean_token_accuracy": 0.13411123901605607,
      "num_tokens": 11885118.0,
      "step": 5490
    },
    {
      "entropy": 6.689449691772461,
      "epoch": 0.5880464444325539,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004973808874656396,
      "loss": 6.4416,
      "mean_token_accuracy": 0.13338536843657495,
      "num_tokens": 11895859.0,
      "step": 5495
    },
    {
      "entropy": 6.678749036788941,
      "epoch": 0.5885815185403178,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004973750675675458,
      "loss": 6.4658,
      "mean_token_accuracy": 0.12978944033384324,
      "num_tokens": 11907019.0,
      "step": 5500
    },
    {
      "entropy": 6.668684005737305,
      "epoch": 0.5891165926480818,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004973692412484034,
      "loss": 6.4154,
      "mean_token_accuracy": 0.13129291459918022,
      "num_tokens": 11918257.0,
      "step": 5505
    },
    {
      "entropy": 6.582723999023438,
      "epoch": 0.5896516667558457,
      "grad_norm": 0.875,
      "learning_rate": 0.0004973634085083802,
      "loss": 6.411,
      "mean_token_accuracy": 0.14034318551421165,
      "num_tokens": 11929003.0,
      "step": 5510
    },
    {
      "entropy": 6.647145223617554,
      "epoch": 0.5901867408636096,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497357569347645,
      "loss": 6.549,
      "mean_token_accuracy": 0.12624183893203736,
      "num_tokens": 11939786.0,
      "step": 5515
    },
    {
      "entropy": 6.6684633731842045,
      "epoch": 0.5907218149713735,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004973517237663661,
      "loss": 6.4791,
      "mean_token_accuracy": 0.1251661829650402,
      "num_tokens": 11951042.0,
      "step": 5520
    },
    {
      "entropy": 6.658832693099976,
      "epoch": 0.5912568890791374,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004973458717647124,
      "loss": 6.3741,
      "mean_token_accuracy": 0.14139798507094384,
      "num_tokens": 11961982.0,
      "step": 5525
    },
    {
      "entropy": 6.5981471061706545,
      "epoch": 0.5917919631869014,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000497340013342853,
      "loss": 6.4245,
      "mean_token_accuracy": 0.1310732401907444,
      "num_tokens": 11972078.0,
      "step": 5530
    },
    {
      "entropy": 6.588191175460816,
      "epoch": 0.5923270372946653,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000497334148500957,
      "loss": 6.4693,
      "mean_token_accuracy": 0.13160526901483535,
      "num_tokens": 11984037.0,
      "step": 5535
    },
    {
      "entropy": 6.680817985534668,
      "epoch": 0.5928621114024293,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004973282772391936,
      "loss": 6.411,
      "mean_token_accuracy": 0.13250281438231468,
      "num_tokens": 11993688.0,
      "step": 5540
    },
    {
      "entropy": 6.5483081340789795,
      "epoch": 0.5933971855101932,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004973223995577324,
      "loss": 6.3866,
      "mean_token_accuracy": 0.13032643646001815,
      "num_tokens": 12003759.0,
      "step": 5545
    },
    {
      "entropy": 6.577221488952636,
      "epoch": 0.593932259617957,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004973165154567432,
      "loss": 6.4144,
      "mean_token_accuracy": 0.1334957979619503,
      "num_tokens": 12015030.0,
      "step": 5550
    },
    {
      "entropy": 6.6546117782592775,
      "epoch": 0.594467333725721,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004973106249363959,
      "loss": 6.4858,
      "mean_token_accuracy": 0.1276483990252018,
      "num_tokens": 12026181.0,
      "step": 5555
    },
    {
      "entropy": 6.558355188369751,
      "epoch": 0.5950024078334849,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004973047279968606,
      "loss": 6.2504,
      "mean_token_accuracy": 0.14747404307127,
      "num_tokens": 12037336.0,
      "step": 5560
    },
    {
      "entropy": 6.612463665008545,
      "epoch": 0.5955374819412489,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004972988246383074,
      "loss": 6.3157,
      "mean_token_accuracy": 0.13885221257805824,
      "num_tokens": 12048425.0,
      "step": 5565
    },
    {
      "entropy": 6.493126106262207,
      "epoch": 0.5960725560490128,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004972929148609068,
      "loss": 6.2912,
      "mean_token_accuracy": 0.1498216025531292,
      "num_tokens": 12059814.0,
      "step": 5570
    },
    {
      "entropy": 6.545558166503906,
      "epoch": 0.5966076301567768,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004972869986648296,
      "loss": 6.3595,
      "mean_token_accuracy": 0.13906975761055945,
      "num_tokens": 12069377.0,
      "step": 5575
    },
    {
      "entropy": 6.586904287338257,
      "epoch": 0.5971427042645406,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004972810760502467,
      "loss": 6.5038,
      "mean_token_accuracy": 0.1273652084171772,
      "num_tokens": 12080277.0,
      "step": 5580
    },
    {
      "entropy": 6.653348588943482,
      "epoch": 0.5976777783723045,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004972751470173287,
      "loss": 6.3746,
      "mean_token_accuracy": 0.13318905383348464,
      "num_tokens": 12091375.0,
      "step": 5585
    },
    {
      "entropy": 6.579549646377563,
      "epoch": 0.5982128524800685,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004972692115662473,
      "loss": 6.4281,
      "mean_token_accuracy": 0.13375982269644737,
      "num_tokens": 12103110.0,
      "step": 5590
    },
    {
      "entropy": 6.610755443572998,
      "epoch": 0.5987479265878324,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004972632696971735,
      "loss": 6.4737,
      "mean_token_accuracy": 0.12939688861370086,
      "num_tokens": 12112454.0,
      "step": 5595
    },
    {
      "entropy": 6.576534271240234,
      "epoch": 0.5992830006955964,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004972573214102791,
      "loss": 6.3557,
      "mean_token_accuracy": 0.1363022118806839,
      "num_tokens": 12125140.0,
      "step": 5600
    },
    {
      "entropy": 6.645657777786255,
      "epoch": 0.5998180748033602,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004972513667057358,
      "loss": 6.3867,
      "mean_token_accuracy": 0.13274750784039496,
      "num_tokens": 12136061.0,
      "step": 5605
    },
    {
      "entropy": 6.6707840919494625,
      "epoch": 0.6003531489111242,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004972454055837155,
      "loss": 6.4087,
      "mean_token_accuracy": 0.13548065349459648,
      "num_tokens": 12146654.0,
      "step": 5610
    },
    {
      "entropy": 6.639386177062988,
      "epoch": 0.6008882230188881,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004972394380443903,
      "loss": 6.4221,
      "mean_token_accuracy": 0.12807304710149764,
      "num_tokens": 12157881.0,
      "step": 5615
    },
    {
      "entropy": 6.587838172912598,
      "epoch": 0.601423297126652,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004972334640879325,
      "loss": 6.4465,
      "mean_token_accuracy": 0.12886952012777328,
      "num_tokens": 12168534.0,
      "step": 5620
    },
    {
      "entropy": 6.581140470504761,
      "epoch": 0.601958371234416,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004972274837145147,
      "loss": 6.3688,
      "mean_token_accuracy": 0.13126305714249611,
      "num_tokens": 12178773.0,
      "step": 5625
    },
    {
      "entropy": 6.653379631042481,
      "epoch": 0.6024934453421799,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004972214969243096,
      "loss": 6.4028,
      "mean_token_accuracy": 0.13439842611551284,
      "num_tokens": 12188819.0,
      "step": 5630
    },
    {
      "entropy": 6.560933065414429,
      "epoch": 0.6030285194499438,
      "grad_norm": 0.84375,
      "learning_rate": 0.00049721550371749,
      "loss": 6.3844,
      "mean_token_accuracy": 0.14078628718852998,
      "num_tokens": 12199198.0,
      "step": 5635
    },
    {
      "entropy": 6.592670249938965,
      "epoch": 0.6035635935577077,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0004972095040942289,
      "loss": 6.4063,
      "mean_token_accuracy": 0.13120741173624992,
      "num_tokens": 12210765.0,
      "step": 5640
    },
    {
      "entropy": 6.618483829498291,
      "epoch": 0.6040986676654717,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004972034980546996,
      "loss": 6.433,
      "mean_token_accuracy": 0.12770435735583305,
      "num_tokens": 12221067.0,
      "step": 5645
    },
    {
      "entropy": 6.563142490386963,
      "epoch": 0.6046337417732356,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004971974855990755,
      "loss": 6.4267,
      "mean_token_accuracy": 0.13381871506571769,
      "num_tokens": 12232994.0,
      "step": 5650
    },
    {
      "entropy": 6.61626124382019,
      "epoch": 0.6051688158809995,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004971914667275302,
      "loss": 6.3951,
      "mean_token_accuracy": 0.1357675515115261,
      "num_tokens": 12243327.0,
      "step": 5655
    },
    {
      "entropy": 6.537327909469605,
      "epoch": 0.6057038899887635,
      "grad_norm": 0.875,
      "learning_rate": 0.0004971854414402377,
      "loss": 6.3161,
      "mean_token_accuracy": 0.1375333398580551,
      "num_tokens": 12254249.0,
      "step": 5660
    },
    {
      "entropy": 6.5772881507873535,
      "epoch": 0.6062389640965273,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004971794097373716,
      "loss": 6.3624,
      "mean_token_accuracy": 0.1386586993932724,
      "num_tokens": 12264487.0,
      "step": 5665
    },
    {
      "entropy": 6.5285249710083,
      "epoch": 0.6067740382042913,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004971733716191063,
      "loss": 6.3467,
      "mean_token_accuracy": 0.14049620628356935,
      "num_tokens": 12274801.0,
      "step": 5670
    },
    {
      "entropy": 6.616953182220459,
      "epoch": 0.6073091123120552,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004971673270856162,
      "loss": 6.3145,
      "mean_token_accuracy": 0.13489979058504104,
      "num_tokens": 12284647.0,
      "step": 5675
    },
    {
      "entropy": 6.609493827819824,
      "epoch": 0.6078441864198192,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004971612761370756,
      "loss": 6.4419,
      "mean_token_accuracy": 0.1319086618721485,
      "num_tokens": 12295215.0,
      "step": 5680
    },
    {
      "entropy": 6.58382215499878,
      "epoch": 0.6083792605275831,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004971552187736596,
      "loss": 6.3549,
      "mean_token_accuracy": 0.1318963848054409,
      "num_tokens": 12305324.0,
      "step": 5685
    },
    {
      "entropy": 6.569716119766236,
      "epoch": 0.6089143346353469,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004971491549955427,
      "loss": 6.3737,
      "mean_token_accuracy": 0.13600005954504013,
      "num_tokens": 12316399.0,
      "step": 5690
    },
    {
      "entropy": 6.495478010177612,
      "epoch": 0.6094494087431109,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004971430848029002,
      "loss": 6.3536,
      "mean_token_accuracy": 0.13850070238113404,
      "num_tokens": 12328033.0,
      "step": 5695
    },
    {
      "entropy": 6.550614070892334,
      "epoch": 0.6099844828508748,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004971370081959073,
      "loss": 6.3899,
      "mean_token_accuracy": 0.13690576702356339,
      "num_tokens": 12338962.0,
      "step": 5700
    },
    {
      "entropy": 6.6537275314331055,
      "epoch": 0.6105195569586388,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004971309251747396,
      "loss": 6.3776,
      "mean_token_accuracy": 0.13331339582800866,
      "num_tokens": 12350216.0,
      "step": 5705
    },
    {
      "entropy": 6.641787004470825,
      "epoch": 0.6110546310664027,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004971248357395726,
      "loss": 6.4036,
      "mean_token_accuracy": 0.13424265161156654,
      "num_tokens": 12361277.0,
      "step": 5710
    },
    {
      "entropy": 6.506038570404053,
      "epoch": 0.6115897051741667,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004971187398905821,
      "loss": 6.3281,
      "mean_token_accuracy": 0.13117444440722464,
      "num_tokens": 12373308.0,
      "step": 5715
    },
    {
      "entropy": 6.6541353225708,
      "epoch": 0.6121247792819305,
      "grad_norm": 0.875,
      "learning_rate": 0.0004971126376279443,
      "loss": 6.5502,
      "mean_token_accuracy": 0.12353794276714325,
      "num_tokens": 12383943.0,
      "step": 5720
    },
    {
      "entropy": 6.608107709884644,
      "epoch": 0.6126598533896944,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004971065289518354,
      "loss": 6.3202,
      "mean_token_accuracy": 0.13428990244865419,
      "num_tokens": 12394575.0,
      "step": 5725
    },
    {
      "entropy": 6.625621795654297,
      "epoch": 0.6131949274974584,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004971004138624315,
      "loss": 6.4187,
      "mean_token_accuracy": 0.13277268782258034,
      "num_tokens": 12405536.0,
      "step": 5730
    },
    {
      "entropy": 6.525531339645386,
      "epoch": 0.6137300016052223,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004970942923599094,
      "loss": 6.307,
      "mean_token_accuracy": 0.14375876933336257,
      "num_tokens": 12416148.0,
      "step": 5735
    },
    {
      "entropy": 6.575962924957276,
      "epoch": 0.6142650757129863,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004970881644444459,
      "loss": 6.3208,
      "mean_token_accuracy": 0.1355944372713566,
      "num_tokens": 12426182.0,
      "step": 5740
    },
    {
      "entropy": 6.567414903640747,
      "epoch": 0.6148001498207502,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004970820301162178,
      "loss": 6.5025,
      "mean_token_accuracy": 0.12934552878141403,
      "num_tokens": 12437548.0,
      "step": 5745
    },
    {
      "entropy": 6.652767753601074,
      "epoch": 0.6153352239285141,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004970758893754022,
      "loss": 6.4325,
      "mean_token_accuracy": 0.1311009518802166,
      "num_tokens": 12449720.0,
      "step": 5750
    },
    {
      "entropy": 6.704830598831177,
      "epoch": 0.615870298036278,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004970697422221767,
      "loss": 6.4028,
      "mean_token_accuracy": 0.13212565407156945,
      "num_tokens": 12460666.0,
      "step": 5755
    },
    {
      "entropy": 6.573892498016358,
      "epoch": 0.6164053721440419,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004970635886567185,
      "loss": 6.4178,
      "mean_token_accuracy": 0.13249918147921563,
      "num_tokens": 12471472.0,
      "step": 5760
    },
    {
      "entropy": 6.525885772705078,
      "epoch": 0.6169404462518059,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004970574286792054,
      "loss": 6.3997,
      "mean_token_accuracy": 0.1310979500412941,
      "num_tokens": 12482127.0,
      "step": 5765
    },
    {
      "entropy": 6.584674406051636,
      "epoch": 0.6174755203595698,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004970512622898152,
      "loss": 6.3907,
      "mean_token_accuracy": 0.137465713173151,
      "num_tokens": 12493680.0,
      "step": 5770
    },
    {
      "entropy": 6.5630621910095215,
      "epoch": 0.6180105944673338,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497045089488726,
      "loss": 6.3481,
      "mean_token_accuracy": 0.13616480454802513,
      "num_tokens": 12505838.0,
      "step": 5775
    },
    {
      "entropy": 6.611004209518432,
      "epoch": 0.6185456685750976,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004970389102761161,
      "loss": 6.3498,
      "mean_token_accuracy": 0.13807964995503424,
      "num_tokens": 12516334.0,
      "step": 5780
    },
    {
      "entropy": 6.555019569396973,
      "epoch": 0.6190807426828616,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004970327246521638,
      "loss": 6.4176,
      "mean_token_accuracy": 0.13207646533846856,
      "num_tokens": 12526440.0,
      "step": 5785
    },
    {
      "entropy": 6.587235403060913,
      "epoch": 0.6196158167906255,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004970265326170478,
      "loss": 6.3531,
      "mean_token_accuracy": 0.130529922246933,
      "num_tokens": 12538127.0,
      "step": 5790
    },
    {
      "entropy": 6.642105770111084,
      "epoch": 0.6201508908983894,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004970203341709469,
      "loss": 6.4303,
      "mean_token_accuracy": 0.13160809874534607,
      "num_tokens": 12548459.0,
      "step": 5795
    },
    {
      "entropy": 6.528612279891968,
      "epoch": 0.6206859650061534,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00049701412931404,
      "loss": 6.374,
      "mean_token_accuracy": 0.1364937886595726,
      "num_tokens": 12558989.0,
      "step": 5800
    },
    {
      "entropy": 6.578775882720947,
      "epoch": 0.6212210391139172,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004970079180465064,
      "loss": 6.3079,
      "mean_token_accuracy": 0.1425821900367737,
      "num_tokens": 12569449.0,
      "step": 5805
    },
    {
      "entropy": 6.5736206531524655,
      "epoch": 0.6217561132216812,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004970017003685253,
      "loss": 6.3594,
      "mean_token_accuracy": 0.13570875078439712,
      "num_tokens": 12580733.0,
      "step": 5810
    },
    {
      "entropy": 6.564662742614746,
      "epoch": 0.6222911873294451,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004969954762802763,
      "loss": 6.3757,
      "mean_token_accuracy": 0.13068436086177826,
      "num_tokens": 12590869.0,
      "step": 5815
    },
    {
      "entropy": 6.649873065948486,
      "epoch": 0.6228262614372091,
      "grad_norm": 0.984375,
      "learning_rate": 0.000496989245781939,
      "loss": 6.4412,
      "mean_token_accuracy": 0.1330619663000107,
      "num_tokens": 12602109.0,
      "step": 5820
    },
    {
      "entropy": 6.611514139175415,
      "epoch": 0.623361335544973,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004969830088736937,
      "loss": 6.3562,
      "mean_token_accuracy": 0.1308489643037319,
      "num_tokens": 12612234.0,
      "step": 5825
    },
    {
      "entropy": 6.664484024047852,
      "epoch": 0.6238964096527368,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00049697676555572,
      "loss": 6.5161,
      "mean_token_accuracy": 0.12568282708525658,
      "num_tokens": 12623252.0,
      "step": 5830
    },
    {
      "entropy": 6.628607940673828,
      "epoch": 0.6244314837605008,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004969705158281985,
      "loss": 6.2786,
      "mean_token_accuracy": 0.1406656801700592,
      "num_tokens": 12632796.0,
      "step": 5835
    },
    {
      "entropy": 6.589381551742553,
      "epoch": 0.6249665578682647,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004969642596913095,
      "loss": 6.3634,
      "mean_token_accuracy": 0.13309731855988502,
      "num_tokens": 12643278.0,
      "step": 5840
    },
    {
      "entropy": 6.601040267944336,
      "epoch": 0.6255016319760287,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004969579971452337,
      "loss": 6.4447,
      "mean_token_accuracy": 0.1332966536283493,
      "num_tokens": 12654446.0,
      "step": 5845
    },
    {
      "entropy": 6.592562961578369,
      "epoch": 0.6260367060837926,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004969517281901519,
      "loss": 6.4314,
      "mean_token_accuracy": 0.1334735356271267,
      "num_tokens": 12664281.0,
      "step": 5850
    },
    {
      "entropy": 6.592110538482666,
      "epoch": 0.6265717801915566,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004969454528262451,
      "loss": 6.4165,
      "mean_token_accuracy": 0.13084367662668228,
      "num_tokens": 12675252.0,
      "step": 5855
    },
    {
      "entropy": 6.57087025642395,
      "epoch": 0.6271068542993204,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004969391710536946,
      "loss": 6.383,
      "mean_token_accuracy": 0.1349936917424202,
      "num_tokens": 12685764.0,
      "step": 5860
    },
    {
      "entropy": 6.523224925994873,
      "epoch": 0.6276419284070843,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004969328828726817,
      "loss": 6.3027,
      "mean_token_accuracy": 0.13696956783533096,
      "num_tokens": 12696974.0,
      "step": 5865
    },
    {
      "entropy": 6.613940811157226,
      "epoch": 0.6281770025148483,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004969265882833879,
      "loss": 6.4553,
      "mean_token_accuracy": 0.1232716128230095,
      "num_tokens": 12708217.0,
      "step": 5870
    },
    {
      "entropy": 6.612708950042725,
      "epoch": 0.6287120766226122,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004969202872859952,
      "loss": 6.3184,
      "mean_token_accuracy": 0.13500054329633712,
      "num_tokens": 12717840.0,
      "step": 5875
    },
    {
      "entropy": 6.552431583404541,
      "epoch": 0.6292471507303762,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004969139798806854,
      "loss": 6.4025,
      "mean_token_accuracy": 0.13267221301794052,
      "num_tokens": 12728390.0,
      "step": 5880
    },
    {
      "entropy": 6.560757350921631,
      "epoch": 0.6297822248381401,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004969076660676405,
      "loss": 6.4111,
      "mean_token_accuracy": 0.13317102044820786,
      "num_tokens": 12738650.0,
      "step": 5885
    },
    {
      "entropy": 6.619578123092651,
      "epoch": 0.630317298945904,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000496901345847043,
      "loss": 6.4082,
      "mean_token_accuracy": 0.13124213442206384,
      "num_tokens": 12748989.0,
      "step": 5890
    },
    {
      "entropy": 6.637136363983155,
      "epoch": 0.6308523730536679,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004968950192190752,
      "loss": 6.3681,
      "mean_token_accuracy": 0.13284006491303443,
      "num_tokens": 12761132.0,
      "step": 5895
    },
    {
      "entropy": 6.569866418838501,
      "epoch": 0.6313874471614318,
      "grad_norm": 0.859375,
      "learning_rate": 0.00049688868618392,
      "loss": 6.3944,
      "mean_token_accuracy": 0.13606114834547042,
      "num_tokens": 12771478.0,
      "step": 5900
    },
    {
      "entropy": 6.604922771453857,
      "epoch": 0.6319225212691958,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004968823467417602,
      "loss": 6.4543,
      "mean_token_accuracy": 0.1311221204698086,
      "num_tokens": 12782096.0,
      "step": 5905
    },
    {
      "entropy": 6.600997734069824,
      "epoch": 0.6324575953769597,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004968760008927788,
      "loss": 6.3498,
      "mean_token_accuracy": 0.13682234138250352,
      "num_tokens": 12792467.0,
      "step": 5910
    },
    {
      "entropy": 6.644370651245117,
      "epoch": 0.6329926694847237,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000496869648637159,
      "loss": 6.3333,
      "mean_token_accuracy": 0.14047224521636964,
      "num_tokens": 12803248.0,
      "step": 5915
    },
    {
      "entropy": 6.5607692241668705,
      "epoch": 0.6335277435924875,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004968632899750844,
      "loss": 6.3781,
      "mean_token_accuracy": 0.13737768232822417,
      "num_tokens": 12813701.0,
      "step": 5920
    },
    {
      "entropy": 6.618074035644531,
      "epoch": 0.6340628177002515,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004968569249067383,
      "loss": 6.4355,
      "mean_token_accuracy": 0.12714748457074165,
      "num_tokens": 12824945.0,
      "step": 5925
    },
    {
      "entropy": 6.59271092414856,
      "epoch": 0.6345978918080154,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004968505534323047,
      "loss": 6.3896,
      "mean_token_accuracy": 0.12957434356212616,
      "num_tokens": 12836699.0,
      "step": 5930
    },
    {
      "entropy": 6.57825722694397,
      "epoch": 0.6351329659157794,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004968441755519677,
      "loss": 6.3792,
      "mean_token_accuracy": 0.12875413447618483,
      "num_tokens": 12846905.0,
      "step": 5935
    },
    {
      "entropy": 6.626939153671264,
      "epoch": 0.6356680400235433,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004968377912659112,
      "loss": 6.2981,
      "mean_token_accuracy": 0.14117206409573554,
      "num_tokens": 12856919.0,
      "step": 5940
    },
    {
      "entropy": 6.444662714004517,
      "epoch": 0.6362031141313071,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004968314005743196,
      "loss": 6.2765,
      "mean_token_accuracy": 0.14162601307034492,
      "num_tokens": 12867863.0,
      "step": 5945
    },
    {
      "entropy": 6.5123731136322025,
      "epoch": 0.6367381882390711,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004968250034773776,
      "loss": 6.3589,
      "mean_token_accuracy": 0.13333921879529953,
      "num_tokens": 12877877.0,
      "step": 5950
    },
    {
      "entropy": 6.695086145401001,
      "epoch": 0.637273262346835,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004968185999752697,
      "loss": 6.4475,
      "mean_token_accuracy": 0.12420506849884987,
      "num_tokens": 12887954.0,
      "step": 5955
    },
    {
      "entropy": 6.591527891159058,
      "epoch": 0.637808336454599,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004968121900681809,
      "loss": 6.3743,
      "mean_token_accuracy": 0.13082539662718773,
      "num_tokens": 12897744.0,
      "step": 5960
    },
    {
      "entropy": 6.566767930984497,
      "epoch": 0.6383434105623629,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004968057737562964,
      "loss": 6.4402,
      "mean_token_accuracy": 0.13630308136343955,
      "num_tokens": 12909193.0,
      "step": 5965
    },
    {
      "entropy": 6.588399076461792,
      "epoch": 0.6388784846701269,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004967993510398012,
      "loss": 6.3159,
      "mean_token_accuracy": 0.13734279349446296,
      "num_tokens": 12919879.0,
      "step": 5970
    },
    {
      "entropy": 6.559991312026978,
      "epoch": 0.6394135587778907,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004967929219188809,
      "loss": 6.3704,
      "mean_token_accuracy": 0.13482013195753098,
      "num_tokens": 12930297.0,
      "step": 5975
    },
    {
      "entropy": 6.556643629074097,
      "epoch": 0.6399486328856546,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004967864863937212,
      "loss": 6.3235,
      "mean_token_accuracy": 0.14591269195079803,
      "num_tokens": 12940467.0,
      "step": 5980
    },
    {
      "entropy": 6.561069583892822,
      "epoch": 0.6404837069934186,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004967800444645079,
      "loss": 6.3489,
      "mean_token_accuracy": 0.13433404713869096,
      "num_tokens": 12950417.0,
      "step": 5985
    },
    {
      "entropy": 6.556723403930664,
      "epoch": 0.6410187811011825,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004967735961314269,
      "loss": 6.3015,
      "mean_token_accuracy": 0.13700174242258073,
      "num_tokens": 12960957.0,
      "step": 5990
    },
    {
      "entropy": 6.516926383972168,
      "epoch": 0.6415538552089465,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004967671413946646,
      "loss": 6.4051,
      "mean_token_accuracy": 0.1311349555850029,
      "num_tokens": 12971194.0,
      "step": 5995
    },
    {
      "entropy": 6.554597806930542,
      "epoch": 0.6420889293167104,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004967606802544071,
      "loss": 6.3919,
      "mean_token_accuracy": 0.12952575609087943,
      "num_tokens": 12982479.0,
      "step": 6000
    },
    {
      "epoch": 0.6420889293167104,
      "eval_entropy": 6.409764193331777,
      "eval_loss": 6.436302661895752,
      "eval_mean_token_accuracy": 0.13834347560891325,
      "eval_num_tokens": 12982479.0,
      "eval_runtime": 17.6382,
      "eval_samples_per_second": 1682.659,
      "eval_steps_per_second": 210.339,
      "step": 6000
    },
    {
      "entropy": 6.605286121368408,
      "epoch": 0.6426240034244743,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004967542127108412,
      "loss": 6.4464,
      "mean_token_accuracy": 0.1300044409930706,
      "num_tokens": 12994119.0,
      "step": 6005
    },
    {
      "entropy": 6.621917343139648,
      "epoch": 0.6431590775322382,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004967477387641537,
      "loss": 6.3618,
      "mean_token_accuracy": 0.13677581921219825,
      "num_tokens": 13005437.0,
      "step": 6010
    },
    {
      "entropy": 6.528858804702759,
      "epoch": 0.6436941516400021,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004967412584145312,
      "loss": 6.3804,
      "mean_token_accuracy": 0.13444462716579436,
      "num_tokens": 13017278.0,
      "step": 6015
    },
    {
      "entropy": 6.546593952178955,
      "epoch": 0.6442292257477661,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004967347716621611,
      "loss": 6.3433,
      "mean_token_accuracy": 0.1359869047999382,
      "num_tokens": 13028010.0,
      "step": 6020
    },
    {
      "entropy": 6.59840087890625,
      "epoch": 0.64476429985553,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004967282785072306,
      "loss": 6.3459,
      "mean_token_accuracy": 0.13561249598860742,
      "num_tokens": 13038740.0,
      "step": 6025
    },
    {
      "entropy": 6.5341428279876705,
      "epoch": 0.645299373963294,
      "grad_norm": 0.875,
      "learning_rate": 0.0004967217789499272,
      "loss": 6.3663,
      "mean_token_accuracy": 0.13788821995258332,
      "num_tokens": 13049921.0,
      "step": 6030
    },
    {
      "entropy": 6.483286046981812,
      "epoch": 0.6458344480710578,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004967152729904386,
      "loss": 6.2884,
      "mean_token_accuracy": 0.13735996782779694,
      "num_tokens": 13060907.0,
      "step": 6035
    },
    {
      "entropy": 6.555841779708862,
      "epoch": 0.6463695221788218,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004967087606289527,
      "loss": 6.3224,
      "mean_token_accuracy": 0.1412092365324497,
      "num_tokens": 13071720.0,
      "step": 6040
    },
    {
      "entropy": 6.597356939315796,
      "epoch": 0.6469045962865857,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004967022418656575,
      "loss": 6.3672,
      "mean_token_accuracy": 0.13823484629392624,
      "num_tokens": 13082474.0,
      "step": 6045
    },
    {
      "entropy": 6.553841018676758,
      "epoch": 0.6474396703943496,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004966957167007412,
      "loss": 6.2827,
      "mean_token_accuracy": 0.1382516823709011,
      "num_tokens": 13092777.0,
      "step": 6050
    },
    {
      "entropy": 6.499682283401489,
      "epoch": 0.6479747445021136,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004966891851343922,
      "loss": 6.2751,
      "mean_token_accuracy": 0.13983871936798095,
      "num_tokens": 13104019.0,
      "step": 6055
    },
    {
      "entropy": 6.58563814163208,
      "epoch": 0.6485098186098774,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004966826471667992,
      "loss": 6.3197,
      "mean_token_accuracy": 0.13976103737950324,
      "num_tokens": 13115741.0,
      "step": 6060
    },
    {
      "entropy": 6.573620128631592,
      "epoch": 0.6490448927176414,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004966761027981509,
      "loss": 6.3371,
      "mean_token_accuracy": 0.13703611344099045,
      "num_tokens": 13126838.0,
      "step": 6065
    },
    {
      "entropy": 6.50914192199707,
      "epoch": 0.6495799668254053,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004966695520286362,
      "loss": 6.4333,
      "mean_token_accuracy": 0.13525479435920715,
      "num_tokens": 13137485.0,
      "step": 6070
    },
    {
      "entropy": 6.542282056808472,
      "epoch": 0.6501150409331693,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004966629948584444,
      "loss": 6.3128,
      "mean_token_accuracy": 0.14044364169239998,
      "num_tokens": 13147631.0,
      "step": 6075
    },
    {
      "entropy": 6.539883756637574,
      "epoch": 0.6506501150409332,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004966564312877647,
      "loss": 6.3548,
      "mean_token_accuracy": 0.13897788971662522,
      "num_tokens": 13158468.0,
      "step": 6080
    },
    {
      "entropy": 6.591976165771484,
      "epoch": 0.651185189148697,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004966498613167868,
      "loss": 6.3878,
      "mean_token_accuracy": 0.1312116838991642,
      "num_tokens": 13169263.0,
      "step": 6085
    },
    {
      "entropy": 6.565645933151245,
      "epoch": 0.651720263256461,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004966432849457002,
      "loss": 6.3112,
      "mean_token_accuracy": 0.1431890085339546,
      "num_tokens": 13178647.0,
      "step": 6090
    },
    {
      "entropy": 6.518553447723389,
      "epoch": 0.6522553373642249,
      "grad_norm": 0.90625,
      "learning_rate": 0.000496636702174695,
      "loss": 6.3004,
      "mean_token_accuracy": 0.1397631846368313,
      "num_tokens": 13189883.0,
      "step": 6095
    },
    {
      "entropy": 6.526831483840942,
      "epoch": 0.6527904114719889,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004966301130039611,
      "loss": 6.4288,
      "mean_token_accuracy": 0.13575417101383208,
      "num_tokens": 13200920.0,
      "step": 6100
    },
    {
      "entropy": 6.536294841766358,
      "epoch": 0.6533254855797528,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004966235174336889,
      "loss": 6.4238,
      "mean_token_accuracy": 0.12978921085596085,
      "num_tokens": 13211765.0,
      "step": 6105
    },
    {
      "entropy": 6.573797416687012,
      "epoch": 0.6538605596875168,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004966169154640687,
      "loss": 6.3394,
      "mean_token_accuracy": 0.13465530127286912,
      "num_tokens": 13221768.0,
      "step": 6110
    },
    {
      "entropy": 6.514359998703003,
      "epoch": 0.6543956337952807,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004966103070952912,
      "loss": 6.3348,
      "mean_token_accuracy": 0.14100371450185775,
      "num_tokens": 13233545.0,
      "step": 6115
    },
    {
      "entropy": 6.51497449874878,
      "epoch": 0.6549307079030445,
      "grad_norm": 1.0,
      "learning_rate": 0.0004966036923275472,
      "loss": 6.3852,
      "mean_token_accuracy": 0.14398130178451538,
      "num_tokens": 13245018.0,
      "step": 6120
    },
    {
      "entropy": 6.572405004501343,
      "epoch": 0.6554657820108085,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004965970711610278,
      "loss": 6.3031,
      "mean_token_accuracy": 0.13430665656924248,
      "num_tokens": 13255772.0,
      "step": 6125
    },
    {
      "entropy": 6.543505620956421,
      "epoch": 0.6560008561185724,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004965904435959241,
      "loss": 6.3693,
      "mean_token_accuracy": 0.1322019249200821,
      "num_tokens": 13267271.0,
      "step": 6130
    },
    {
      "entropy": 6.534311819076538,
      "epoch": 0.6565359302263364,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004965838096324274,
      "loss": 6.3826,
      "mean_token_accuracy": 0.13832796663045882,
      "num_tokens": 13278467.0,
      "step": 6135
    },
    {
      "entropy": 6.575223445892334,
      "epoch": 0.6570710043341003,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004965771692707294,
      "loss": 6.4417,
      "mean_token_accuracy": 0.12868114784359933,
      "num_tokens": 13288522.0,
      "step": 6140
    },
    {
      "entropy": 6.65201473236084,
      "epoch": 0.6576060784418643,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004965705225110218,
      "loss": 6.4384,
      "mean_token_accuracy": 0.12334849461913108,
      "num_tokens": 13299355.0,
      "step": 6145
    },
    {
      "entropy": 6.629589986801148,
      "epoch": 0.6581411525496281,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004965638693534964,
      "loss": 6.3538,
      "mean_token_accuracy": 0.13837969601154326,
      "num_tokens": 13310334.0,
      "step": 6150
    },
    {
      "entropy": 6.5863532543182375,
      "epoch": 0.658676226657392,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004965572097983455,
      "loss": 6.3851,
      "mean_token_accuracy": 0.1330856777727604,
      "num_tokens": 13320917.0,
      "step": 6155
    },
    {
      "entropy": 6.564312696456909,
      "epoch": 0.659211300765156,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004965505438457613,
      "loss": 6.2986,
      "mean_token_accuracy": 0.1407928004860878,
      "num_tokens": 13331707.0,
      "step": 6160
    },
    {
      "entropy": 6.555730247497559,
      "epoch": 0.6597463748729199,
      "grad_norm": 0.875,
      "learning_rate": 0.0004965438714959363,
      "loss": 6.4441,
      "mean_token_accuracy": 0.129070395976305,
      "num_tokens": 13341784.0,
      "step": 6165
    },
    {
      "entropy": 6.65733118057251,
      "epoch": 0.6602814489806839,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000496537192749063,
      "loss": 6.412,
      "mean_token_accuracy": 0.12962306290864944,
      "num_tokens": 13351633.0,
      "step": 6170
    },
    {
      "entropy": 6.605908918380737,
      "epoch": 0.6608165230884477,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004965305076053345,
      "loss": 6.3106,
      "mean_token_accuracy": 0.1409822680056095,
      "num_tokens": 13362444.0,
      "step": 6175
    },
    {
      "entropy": 6.512597513198853,
      "epoch": 0.6613515971962117,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004965238160649437,
      "loss": 6.4177,
      "mean_token_accuracy": 0.13016311153769494,
      "num_tokens": 13374322.0,
      "step": 6180
    },
    {
      "entropy": 6.5797487735748295,
      "epoch": 0.6618866713039756,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004965171181280838,
      "loss": 6.398,
      "mean_token_accuracy": 0.13184262737631797,
      "num_tokens": 13384188.0,
      "step": 6185
    },
    {
      "entropy": 6.637517690658569,
      "epoch": 0.6624217454117395,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004965104137949483,
      "loss": 6.3927,
      "mean_token_accuracy": 0.12933868244290353,
      "num_tokens": 13395648.0,
      "step": 6190
    },
    {
      "entropy": 6.6057515144348145,
      "epoch": 0.6629568195195035,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004965037030657308,
      "loss": 6.403,
      "mean_token_accuracy": 0.1369745247066021,
      "num_tokens": 13406769.0,
      "step": 6195
    },
    {
      "entropy": 6.596571779251098,
      "epoch": 0.6634918936272673,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004964969859406248,
      "loss": 6.3792,
      "mean_token_accuracy": 0.12673437967896461,
      "num_tokens": 13417509.0,
      "step": 6200
    },
    {
      "entropy": 6.573062944412231,
      "epoch": 0.6640269677350313,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004964902624198246,
      "loss": 6.2996,
      "mean_token_accuracy": 0.13197814524173737,
      "num_tokens": 13429441.0,
      "step": 6205
    },
    {
      "entropy": 6.5056736946105955,
      "epoch": 0.6645620418427952,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004964835325035243,
      "loss": 6.324,
      "mean_token_accuracy": 0.13495202511548995,
      "num_tokens": 13439332.0,
      "step": 6210
    },
    {
      "entropy": 6.519222259521484,
      "epoch": 0.6650971159505592,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004964767961919179,
      "loss": 6.3062,
      "mean_token_accuracy": 0.13584212884306907,
      "num_tokens": 13449400.0,
      "step": 6215
    },
    {
      "entropy": 6.621029233932495,
      "epoch": 0.6656321900583231,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004964700534852003,
      "loss": 6.4161,
      "mean_token_accuracy": 0.13214674815535546,
      "num_tokens": 13460134.0,
      "step": 6220
    },
    {
      "entropy": 6.541699409484863,
      "epoch": 0.666167264166087,
      "grad_norm": 0.875,
      "learning_rate": 0.0004964633043835658,
      "loss": 6.3721,
      "mean_token_accuracy": 0.13698556125164033,
      "num_tokens": 13469954.0,
      "step": 6225
    },
    {
      "entropy": 6.5830151557922365,
      "epoch": 0.666702338273851,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004964565488872097,
      "loss": 6.3515,
      "mean_token_accuracy": 0.1387103796005249,
      "num_tokens": 13479866.0,
      "step": 6230
    },
    {
      "entropy": 6.521968221664428,
      "epoch": 0.6672374123816148,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004964497869963268,
      "loss": 6.307,
      "mean_token_accuracy": 0.13992589563131333,
      "num_tokens": 13490088.0,
      "step": 6235
    },
    {
      "entropy": 6.579476451873779,
      "epoch": 0.6677724864893788,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004964430187111125,
      "loss": 6.3987,
      "mean_token_accuracy": 0.1338411219418049,
      "num_tokens": 13500898.0,
      "step": 6240
    },
    {
      "entropy": 6.549569892883301,
      "epoch": 0.6683075605971427,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004964362440317621,
      "loss": 6.4033,
      "mean_token_accuracy": 0.13586093932390214,
      "num_tokens": 13511537.0,
      "step": 6245
    },
    {
      "entropy": 6.529985809326172,
      "epoch": 0.6688426347049067,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004964294629584712,
      "loss": 6.3411,
      "mean_token_accuracy": 0.1366456300020218,
      "num_tokens": 13522916.0,
      "step": 6250
    },
    {
      "entropy": 6.571378087997436,
      "epoch": 0.6693777088126706,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004964226754914357,
      "loss": 6.3061,
      "mean_token_accuracy": 0.13659246936440467,
      "num_tokens": 13533817.0,
      "step": 6255
    },
    {
      "entropy": 6.5587749004364015,
      "epoch": 0.6699127829204344,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004964158816308517,
      "loss": 6.3934,
      "mean_token_accuracy": 0.13010929077863692,
      "num_tokens": 13543723.0,
      "step": 6260
    },
    {
      "entropy": 6.488784217834473,
      "epoch": 0.6704478570281984,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004964090813769151,
      "loss": 6.2862,
      "mean_token_accuracy": 0.13950706049799919,
      "num_tokens": 13555269.0,
      "step": 6265
    },
    {
      "entropy": 6.559277629852295,
      "epoch": 0.6709829311359623,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004964022747298224,
      "loss": 6.4229,
      "mean_token_accuracy": 0.1281231239438057,
      "num_tokens": 13567465.0,
      "step": 6270
    },
    {
      "entropy": 6.673381423950195,
      "epoch": 0.6715180052437263,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004963954616897701,
      "loss": 6.3595,
      "mean_token_accuracy": 0.13534193560481073,
      "num_tokens": 13578637.0,
      "step": 6275
    },
    {
      "entropy": 6.556958150863648,
      "epoch": 0.6720530793514902,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000496388642256955,
      "loss": 6.3776,
      "mean_token_accuracy": 0.1333651699125767,
      "num_tokens": 13589074.0,
      "step": 6280
    },
    {
      "entropy": 6.444629526138305,
      "epoch": 0.6725881534592542,
      "grad_norm": 0.875,
      "learning_rate": 0.000496381816431574,
      "loss": 6.3303,
      "mean_token_accuracy": 0.13625867664813995,
      "num_tokens": 13599624.0,
      "step": 6285
    },
    {
      "entropy": 6.602360677719116,
      "epoch": 0.673123227567018,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004963749842138241,
      "loss": 6.3538,
      "mean_token_accuracy": 0.1358791708946228,
      "num_tokens": 13610293.0,
      "step": 6290
    },
    {
      "entropy": 6.582174015045166,
      "epoch": 0.6736583016747819,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004963681456039027,
      "loss": 6.3862,
      "mean_token_accuracy": 0.133838801831007,
      "num_tokens": 13621641.0,
      "step": 6295
    },
    {
      "entropy": 6.535614252090454,
      "epoch": 0.6741933757825459,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004963613006020072,
      "loss": 6.3318,
      "mean_token_accuracy": 0.13050647899508477,
      "num_tokens": 13632083.0,
      "step": 6300
    },
    {
      "entropy": 6.512619352340698,
      "epoch": 0.6747284498903098,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004963544492083352,
      "loss": 6.1869,
      "mean_token_accuracy": 0.1617746040225029,
      "num_tokens": 13644084.0,
      "step": 6305
    },
    {
      "entropy": 6.494975948333741,
      "epoch": 0.6752635239980738,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004963475914230845,
      "loss": 6.352,
      "mean_token_accuracy": 0.13421885445713996,
      "num_tokens": 13656231.0,
      "step": 6310
    },
    {
      "entropy": 6.614938879013062,
      "epoch": 0.6757985981058376,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004963407272464533,
      "loss": 6.4083,
      "mean_token_accuracy": 0.13520292639732362,
      "num_tokens": 13667334.0,
      "step": 6315
    },
    {
      "entropy": 6.674988889694214,
      "epoch": 0.6763336722136016,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004963338566786398,
      "loss": 6.3878,
      "mean_token_accuracy": 0.13621892854571344,
      "num_tokens": 13678615.0,
      "step": 6320
    },
    {
      "entropy": 6.569027376174927,
      "epoch": 0.6768687463213655,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004963269797198422,
      "loss": 6.3904,
      "mean_token_accuracy": 0.13982055261731147,
      "num_tokens": 13689890.0,
      "step": 6325
    },
    {
      "entropy": 6.417730903625488,
      "epoch": 0.6774038204291294,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004963200963702592,
      "loss": 6.25,
      "mean_token_accuracy": 0.1447252832353115,
      "num_tokens": 13699120.0,
      "step": 6330
    },
    {
      "entropy": 6.538185358047485,
      "epoch": 0.6779388945368934,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004963132066300894,
      "loss": 6.3607,
      "mean_token_accuracy": 0.13240280523896217,
      "num_tokens": 13710016.0,
      "step": 6335
    },
    {
      "entropy": 6.616557502746582,
      "epoch": 0.6784739686446573,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004963063104995319,
      "loss": 6.4323,
      "mean_token_accuracy": 0.12630163058638572,
      "num_tokens": 13721576.0,
      "step": 6340
    },
    {
      "entropy": 6.619271564483642,
      "epoch": 0.6790090427524212,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004962994079787859,
      "loss": 6.4242,
      "mean_token_accuracy": 0.13799061253666878,
      "num_tokens": 13732568.0,
      "step": 6345
    },
    {
      "entropy": 6.5576300621032715,
      "epoch": 0.6795441168601851,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004962924990680504,
      "loss": 6.2893,
      "mean_token_accuracy": 0.13742328956723213,
      "num_tokens": 13742569.0,
      "step": 6350
    },
    {
      "entropy": 6.517670059204102,
      "epoch": 0.6800791909679491,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004962855837675252,
      "loss": 6.4222,
      "mean_token_accuracy": 0.12688116505742073,
      "num_tokens": 13752689.0,
      "step": 6355
    },
    {
      "entropy": 6.6097806930542,
      "epoch": 0.680614265075713,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004962786620774099,
      "loss": 6.3828,
      "mean_token_accuracy": 0.13684748709201813,
      "num_tokens": 13763709.0,
      "step": 6360
    },
    {
      "entropy": 6.5486103057861325,
      "epoch": 0.6811493391834769,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004962717339979043,
      "loss": 6.2906,
      "mean_token_accuracy": 0.1407897613942623,
      "num_tokens": 13773726.0,
      "step": 6365
    },
    {
      "entropy": 6.491515064239502,
      "epoch": 0.6816844132912409,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0004962647995292084,
      "loss": 6.3401,
      "mean_token_accuracy": 0.13777834102511405,
      "num_tokens": 13785435.0,
      "step": 6370
    },
    {
      "entropy": 6.579515027999878,
      "epoch": 0.6822194873990047,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004962578586715226,
      "loss": 6.3377,
      "mean_token_accuracy": 0.1294751472771168,
      "num_tokens": 13797104.0,
      "step": 6375
    },
    {
      "entropy": 6.6117448806762695,
      "epoch": 0.6827545615067687,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004962509114250471,
      "loss": 6.4275,
      "mean_token_accuracy": 0.13171157985925674,
      "num_tokens": 13807756.0,
      "step": 6380
    },
    {
      "entropy": 6.430712556838989,
      "epoch": 0.6832896356145326,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962439577899828,
      "loss": 6.1153,
      "mean_token_accuracy": 0.15774713307619095,
      "num_tokens": 13818490.0,
      "step": 6385
    },
    {
      "entropy": 6.575515794754028,
      "epoch": 0.6838247097222966,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004962369977665301,
      "loss": 6.3986,
      "mean_token_accuracy": 0.1293700762093067,
      "num_tokens": 13829465.0,
      "step": 6390
    },
    {
      "entropy": 6.609517049789429,
      "epoch": 0.6843597838300605,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004962300313548903,
      "loss": 6.36,
      "mean_token_accuracy": 0.13445046693086624,
      "num_tokens": 13840863.0,
      "step": 6395
    },
    {
      "entropy": 6.590188598632812,
      "epoch": 0.6848948579378243,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004962230585552645,
      "loss": 6.3713,
      "mean_token_accuracy": 0.13608585596084594,
      "num_tokens": 13851711.0,
      "step": 6400
    },
    {
      "entropy": 6.546520328521728,
      "epoch": 0.6854299320455883,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004962160793678539,
      "loss": 6.3934,
      "mean_token_accuracy": 0.13423120304942132,
      "num_tokens": 13864576.0,
      "step": 6405
    },
    {
      "entropy": 6.553374099731445,
      "epoch": 0.6859650061533522,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00049620909379286,
      "loss": 6.3059,
      "mean_token_accuracy": 0.13881907612085342,
      "num_tokens": 13875150.0,
      "step": 6410
    },
    {
      "entropy": 6.558405494689941,
      "epoch": 0.6865000802611162,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004962021018304847,
      "loss": 6.3417,
      "mean_token_accuracy": 0.13712364062666893,
      "num_tokens": 13886070.0,
      "step": 6415
    },
    {
      "entropy": 6.517676591873169,
      "epoch": 0.6870351543688801,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004961951034809298,
      "loss": 6.2697,
      "mean_token_accuracy": 0.13642088323831558,
      "num_tokens": 13897898.0,
      "step": 6420
    },
    {
      "entropy": 6.5993650436401365,
      "epoch": 0.6875702284766441,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004961880987443974,
      "loss": 6.397,
      "mean_token_accuracy": 0.1343232959508896,
      "num_tokens": 13909578.0,
      "step": 6425
    },
    {
      "entropy": 6.423710823059082,
      "epoch": 0.6881053025844079,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004961810876210897,
      "loss": 6.2512,
      "mean_token_accuracy": 0.13721854910254477,
      "num_tokens": 13920537.0,
      "step": 6430
    },
    {
      "entropy": 6.510565853118896,
      "epoch": 0.6886403766921718,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004961740701112091,
      "loss": 6.3016,
      "mean_token_accuracy": 0.14137297347187996,
      "num_tokens": 13931015.0,
      "step": 6435
    },
    {
      "entropy": 6.57059063911438,
      "epoch": 0.6891754507999358,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004961670462149583,
      "loss": 6.3618,
      "mean_token_accuracy": 0.13251473680138587,
      "num_tokens": 13942691.0,
      "step": 6440
    },
    {
      "entropy": 6.646090173721314,
      "epoch": 0.6897105249076997,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004961600159325403,
      "loss": 6.4667,
      "mean_token_accuracy": 0.12274139821529388,
      "num_tokens": 13953945.0,
      "step": 6445
    },
    {
      "entropy": 6.66500506401062,
      "epoch": 0.6902455990154637,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004961529792641577,
      "loss": 6.4344,
      "mean_token_accuracy": 0.12943366393446923,
      "num_tokens": 13965658.0,
      "step": 6450
    },
    {
      "entropy": 6.583506059646607,
      "epoch": 0.6907806731232276,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004961459362100141,
      "loss": 6.4114,
      "mean_token_accuracy": 0.13515917658805848,
      "num_tokens": 13976184.0,
      "step": 6455
    },
    {
      "entropy": 6.515756320953369,
      "epoch": 0.6913157472309915,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004961388867703125,
      "loss": 6.3516,
      "mean_token_accuracy": 0.12608477994799613,
      "num_tokens": 13986243.0,
      "step": 6460
    },
    {
      "entropy": 6.56664662361145,
      "epoch": 0.6918508213387554,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004961318309452566,
      "loss": 6.3156,
      "mean_token_accuracy": 0.13823162913322448,
      "num_tokens": 13995904.0,
      "step": 6465
    },
    {
      "entropy": 6.5749305248260494,
      "epoch": 0.6923858954465193,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004961247687350504,
      "loss": 6.3511,
      "mean_token_accuracy": 0.1384427085518837,
      "num_tokens": 14006711.0,
      "step": 6470
    },
    {
      "entropy": 6.536625957489013,
      "epoch": 0.6929209695542833,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004961177001398974,
      "loss": 6.3534,
      "mean_token_accuracy": 0.13303350359201432,
      "num_tokens": 14018177.0,
      "step": 6475
    },
    {
      "entropy": 6.559611892700195,
      "epoch": 0.6934560436620472,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004961106251600018,
      "loss": 6.2839,
      "mean_token_accuracy": 0.133889614790678,
      "num_tokens": 14028725.0,
      "step": 6480
    },
    {
      "entropy": 6.496445608139038,
      "epoch": 0.6939911177698111,
      "grad_norm": 0.875,
      "learning_rate": 0.0004961035437955681,
      "loss": 6.3436,
      "mean_token_accuracy": 0.146770391613245,
      "num_tokens": 14040448.0,
      "step": 6485
    },
    {
      "entropy": 6.518095779418945,
      "epoch": 0.694526191877575,
      "grad_norm": 0.875,
      "learning_rate": 0.0004960964560468005,
      "loss": 6.3659,
      "mean_token_accuracy": 0.13274956494569778,
      "num_tokens": 14051731.0,
      "step": 6490
    },
    {
      "entropy": 6.610559797286987,
      "epoch": 0.695061265985339,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004960893619139039,
      "loss": 6.3801,
      "mean_token_accuracy": 0.13596992194652557,
      "num_tokens": 14062330.0,
      "step": 6495
    },
    {
      "entropy": 6.557360601425171,
      "epoch": 0.6955963400931029,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004960822613970831,
      "loss": 6.2729,
      "mean_token_accuracy": 0.13554252833127975,
      "num_tokens": 14073991.0,
      "step": 6500
    },
    {
      "entropy": 6.552571201324463,
      "epoch": 0.6961314142008668,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004960751544965431,
      "loss": 6.3204,
      "mean_token_accuracy": 0.1351766750216484,
      "num_tokens": 14085132.0,
      "step": 6505
    },
    {
      "entropy": 6.4797890186309814,
      "epoch": 0.6966664883086308,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004960680412124889,
      "loss": 6.2699,
      "mean_token_accuracy": 0.13892600312829018,
      "num_tokens": 14095371.0,
      "step": 6510
    },
    {
      "entropy": 6.482806444168091,
      "epoch": 0.6972015624163946,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004960609215451261,
      "loss": 6.3285,
      "mean_token_accuracy": 0.13835306018590926,
      "num_tokens": 14106117.0,
      "step": 6515
    },
    {
      "entropy": 6.5573710918426515,
      "epoch": 0.6977366365241586,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004960537954946604,
      "loss": 6.3195,
      "mean_token_accuracy": 0.13604175746440889,
      "num_tokens": 14117112.0,
      "step": 6520
    },
    {
      "entropy": 6.46489372253418,
      "epoch": 0.6982717106319225,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004960466630612974,
      "loss": 6.2667,
      "mean_token_accuracy": 0.145594622194767,
      "num_tokens": 14128037.0,
      "step": 6525
    },
    {
      "entropy": 6.456885957717896,
      "epoch": 0.6988067847396865,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004960395242452429,
      "loss": 6.2725,
      "mean_token_accuracy": 0.14177535697817803,
      "num_tokens": 14140183.0,
      "step": 6530
    },
    {
      "entropy": 6.5688066482543945,
      "epoch": 0.6993418588474504,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004960323790467033,
      "loss": 6.4037,
      "mean_token_accuracy": 0.13741268292069436,
      "num_tokens": 14151635.0,
      "step": 6535
    },
    {
      "entropy": 6.582125186920166,
      "epoch": 0.6998769329552142,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004960252274658849,
      "loss": 6.3653,
      "mean_token_accuracy": 0.13139770030975342,
      "num_tokens": 14162660.0,
      "step": 6540
    },
    {
      "entropy": 6.629947805404663,
      "epoch": 0.7004120070629782,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004960180695029939,
      "loss": 6.3288,
      "mean_token_accuracy": 0.14254391938447952,
      "num_tokens": 14173496.0,
      "step": 6545
    },
    {
      "entropy": 6.50215892791748,
      "epoch": 0.7009470811707421,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004960109051582374,
      "loss": 6.2802,
      "mean_token_accuracy": 0.13397079333662987,
      "num_tokens": 14185479.0,
      "step": 6550
    },
    {
      "entropy": 6.503316020965576,
      "epoch": 0.7014821552785061,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004960037344318221,
      "loss": 6.2599,
      "mean_token_accuracy": 0.13978263661265372,
      "num_tokens": 14197441.0,
      "step": 6555
    },
    {
      "entropy": 6.51508264541626,
      "epoch": 0.70201722938627,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004959965573239548,
      "loss": 6.2287,
      "mean_token_accuracy": 0.13928703740239143,
      "num_tokens": 14207832.0,
      "step": 6560
    },
    {
      "entropy": 6.4884954452514645,
      "epoch": 0.702552303494034,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004959893738348432,
      "loss": 6.2919,
      "mean_token_accuracy": 0.14894637763500213,
      "num_tokens": 14217976.0,
      "step": 6565
    },
    {
      "entropy": 6.609205055236816,
      "epoch": 0.7030873776017978,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004959821839646943,
      "loss": 6.2758,
      "mean_token_accuracy": 0.14147015362977983,
      "num_tokens": 14227557.0,
      "step": 6570
    },
    {
      "entropy": 6.521268224716186,
      "epoch": 0.7036224517095617,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000495974987713716,
      "loss": 6.3663,
      "mean_token_accuracy": 0.13289435282349588,
      "num_tokens": 14238912.0,
      "step": 6575
    },
    {
      "entropy": 6.529506969451904,
      "epoch": 0.7041575258173257,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004959677850821159,
      "loss": 6.3035,
      "mean_token_accuracy": 0.13616457358002662,
      "num_tokens": 14249968.0,
      "step": 6580
    },
    {
      "entropy": 6.5767467498779295,
      "epoch": 0.7046925999250896,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004959605760701021,
      "loss": 6.3797,
      "mean_token_accuracy": 0.1389591544866562,
      "num_tokens": 14260809.0,
      "step": 6585
    },
    {
      "entropy": 6.556084728240966,
      "epoch": 0.7052276740328536,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004959533606778827,
      "loss": 6.3308,
      "mean_token_accuracy": 0.13534696996212006,
      "num_tokens": 14271971.0,
      "step": 6590
    },
    {
      "entropy": 6.595488882064819,
      "epoch": 0.7057627481406175,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004959461389056661,
      "loss": 6.3976,
      "mean_token_accuracy": 0.138909899443388,
      "num_tokens": 14283550.0,
      "step": 6595
    },
    {
      "entropy": 6.502576589584351,
      "epoch": 0.7062978222483814,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004959389107536608,
      "loss": 6.2379,
      "mean_token_accuracy": 0.1382664956152439,
      "num_tokens": 14293769.0,
      "step": 6600
    },
    {
      "entropy": 6.504676055908203,
      "epoch": 0.7068328963561453,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004959316762220754,
      "loss": 6.3281,
      "mean_token_accuracy": 0.13388395681977272,
      "num_tokens": 14304668.0,
      "step": 6605
    },
    {
      "entropy": 6.560839319229126,
      "epoch": 0.7073679704639092,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000495924435311119,
      "loss": 6.3967,
      "mean_token_accuracy": 0.13305563032627105,
      "num_tokens": 14315912.0,
      "step": 6610
    },
    {
      "entropy": 6.529976272583008,
      "epoch": 0.7079030445716732,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004959171880210005,
      "loss": 6.3608,
      "mean_token_accuracy": 0.13332727402448655,
      "num_tokens": 14327386.0,
      "step": 6615
    },
    {
      "entropy": 6.5461677551269535,
      "epoch": 0.7084381186794371,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004959099343519294,
      "loss": 6.3491,
      "mean_token_accuracy": 0.1343730866909027,
      "num_tokens": 14338629.0,
      "step": 6620
    },
    {
      "entropy": 6.601601648330688,
      "epoch": 0.7089731927872011,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004959026743041148,
      "loss": 6.3646,
      "mean_token_accuracy": 0.13949117958545684,
      "num_tokens": 14349195.0,
      "step": 6625
    },
    {
      "entropy": 6.585690832138061,
      "epoch": 0.7095082668949649,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004958954078777665,
      "loss": 6.3903,
      "mean_token_accuracy": 0.13392735421657562,
      "num_tokens": 14359449.0,
      "step": 6630
    },
    {
      "entropy": 6.537525224685669,
      "epoch": 0.7100433410027289,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004958881350730944,
      "loss": 6.2987,
      "mean_token_accuracy": 0.14222483485937118,
      "num_tokens": 14369904.0,
      "step": 6635
    },
    {
      "entropy": 6.52706995010376,
      "epoch": 0.7105784151104928,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004958808558903085,
      "loss": 6.324,
      "mean_token_accuracy": 0.13787742778658868,
      "num_tokens": 14382635.0,
      "step": 6640
    },
    {
      "entropy": 6.534725475311279,
      "epoch": 0.7111134892182567,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958735703296189,
      "loss": 6.2597,
      "mean_token_accuracy": 0.1354966141283512,
      "num_tokens": 14393238.0,
      "step": 6645
    },
    {
      "entropy": 6.5581972122192385,
      "epoch": 0.7116485633260207,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000495866278391236,
      "loss": 6.3119,
      "mean_token_accuracy": 0.13689091429114342,
      "num_tokens": 14403734.0,
      "step": 6650
    },
    {
      "entropy": 6.552613544464111,
      "epoch": 0.7121836374337845,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004958589800753703,
      "loss": 6.2983,
      "mean_token_accuracy": 0.13675035089254378,
      "num_tokens": 14414201.0,
      "step": 6655
    },
    {
      "entropy": 6.533583068847657,
      "epoch": 0.7127187115415485,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004958516753822326,
      "loss": 6.3225,
      "mean_token_accuracy": 0.14106455072760582,
      "num_tokens": 14425153.0,
      "step": 6660
    },
    {
      "entropy": 6.539647817611694,
      "epoch": 0.7132537856493124,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004958443643120338,
      "loss": 6.3111,
      "mean_token_accuracy": 0.14422643184661865,
      "num_tokens": 14435942.0,
      "step": 6665
    },
    {
      "entropy": 6.522118949890137,
      "epoch": 0.7137888597570764,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000495837046864985,
      "loss": 6.3012,
      "mean_token_accuracy": 0.1385548710823059,
      "num_tokens": 14446371.0,
      "step": 6670
    },
    {
      "entropy": 6.533046054840088,
      "epoch": 0.7143239338648403,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004958297230412976,
      "loss": 6.2226,
      "mean_token_accuracy": 0.1426885724067688,
      "num_tokens": 14456598.0,
      "step": 6675
    },
    {
      "entropy": 6.5077948570251465,
      "epoch": 0.7148590079726042,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958223928411829,
      "loss": 6.3241,
      "mean_token_accuracy": 0.13185439109802247,
      "num_tokens": 14466775.0,
      "step": 6680
    },
    {
      "entropy": 6.46589035987854,
      "epoch": 0.7153940820803681,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004958150562648526,
      "loss": 6.2855,
      "mean_token_accuracy": 0.13807128742337227,
      "num_tokens": 14476876.0,
      "step": 6685
    },
    {
      "entropy": 6.4962958812713625,
      "epoch": 0.715929156188132,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004958077133125187,
      "loss": 6.2734,
      "mean_token_accuracy": 0.14248252883553505,
      "num_tokens": 14487698.0,
      "step": 6690
    },
    {
      "entropy": 6.55243148803711,
      "epoch": 0.716464230295896,
      "grad_norm": 0.8046875,
      "learning_rate": 0.000495800363984393,
      "loss": 6.3003,
      "mean_token_accuracy": 0.1408907391130924,
      "num_tokens": 14498572.0,
      "step": 6695
    },
    {
      "entropy": 6.5066344261169435,
      "epoch": 0.7169993044036599,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004957930082806878,
      "loss": 6.3457,
      "mean_token_accuracy": 0.13698889389634133,
      "num_tokens": 14509673.0,
      "step": 6700
    },
    {
      "entropy": 6.630517053604126,
      "epoch": 0.7175343785114239,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004957856462016155,
      "loss": 6.4252,
      "mean_token_accuracy": 0.12491722479462623,
      "num_tokens": 14519974.0,
      "step": 6705
    },
    {
      "entropy": 6.573778200149536,
      "epoch": 0.7180694526191878,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004957782777473888,
      "loss": 6.3398,
      "mean_token_accuracy": 0.13382058069109917,
      "num_tokens": 14531194.0,
      "step": 6710
    },
    {
      "entropy": 6.534556913375854,
      "epoch": 0.7186045267269516,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004957709029182203,
      "loss": 6.2887,
      "mean_token_accuracy": 0.1372295469045639,
      "num_tokens": 14542393.0,
      "step": 6715
    },
    {
      "entropy": 6.506135320663452,
      "epoch": 0.7191396008347156,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004957635217143231,
      "loss": 6.314,
      "mean_token_accuracy": 0.14519082009792328,
      "num_tokens": 14552934.0,
      "step": 6720
    },
    {
      "entropy": 6.524216794967652,
      "epoch": 0.7196746749424795,
      "grad_norm": 0.8125,
      "learning_rate": 0.00049575613413591,
      "loss": 6.3161,
      "mean_token_accuracy": 0.13553231358528137,
      "num_tokens": 14563643.0,
      "step": 6725
    },
    {
      "entropy": 6.551642179489136,
      "epoch": 0.7202097490502435,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004957487401831947,
      "loss": 6.2422,
      "mean_token_accuracy": 0.1390040025115013,
      "num_tokens": 14574422.0,
      "step": 6730
    },
    {
      "entropy": 6.479341602325439,
      "epoch": 0.7207448231580074,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004957413398563906,
      "loss": 6.2293,
      "mean_token_accuracy": 0.13783804848790168,
      "num_tokens": 14585002.0,
      "step": 6735
    },
    {
      "entropy": 6.524044466018677,
      "epoch": 0.7212798972657714,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004957339331557113,
      "loss": 6.2774,
      "mean_token_accuracy": 0.13944358229637147,
      "num_tokens": 14595172.0,
      "step": 6740
    },
    {
      "entropy": 6.52948112487793,
      "epoch": 0.7218149713735352,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004957265200813706,
      "loss": 6.3952,
      "mean_token_accuracy": 0.13071175366640092,
      "num_tokens": 14606316.0,
      "step": 6745
    },
    {
      "entropy": 6.582382488250732,
      "epoch": 0.7223500454812991,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004957191006335827,
      "loss": 6.2874,
      "mean_token_accuracy": 0.1332968533039093,
      "num_tokens": 14616394.0,
      "step": 6750
    },
    {
      "entropy": 6.4622715473175045,
      "epoch": 0.7228851195890631,
      "grad_norm": 0.875,
      "learning_rate": 0.0004957116748125618,
      "loss": 6.3134,
      "mean_token_accuracy": 0.1357070192694664,
      "num_tokens": 14627474.0,
      "step": 6755
    },
    {
      "entropy": 6.491555452346802,
      "epoch": 0.723420193696827,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004957042426185223,
      "loss": 6.2374,
      "mean_token_accuracy": 0.14414020925760268,
      "num_tokens": 14638343.0,
      "step": 6760
    },
    {
      "entropy": 6.517511606216431,
      "epoch": 0.723955267804591,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004956968040516789,
      "loss": 6.3094,
      "mean_token_accuracy": 0.13668713793158532,
      "num_tokens": 14649115.0,
      "step": 6765
    },
    {
      "entropy": 6.515577697753907,
      "epoch": 0.7244903419123548,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004956893591122461,
      "loss": 6.3589,
      "mean_token_accuracy": 0.13698833510279657,
      "num_tokens": 14659714.0,
      "step": 6770
    },
    {
      "entropy": 6.614225006103515,
      "epoch": 0.7250254160201188,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004956819078004392,
      "loss": 6.418,
      "mean_token_accuracy": 0.13083635196089743,
      "num_tokens": 14670481.0,
      "step": 6775
    },
    {
      "entropy": 6.523179149627685,
      "epoch": 0.7255604901278827,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004956744501164732,
      "loss": 6.2267,
      "mean_token_accuracy": 0.14649262502789498,
      "num_tokens": 14680816.0,
      "step": 6780
    },
    {
      "entropy": 6.4830670833587645,
      "epoch": 0.7260955642356466,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004956669860605634,
      "loss": 6.3165,
      "mean_token_accuracy": 0.13714328408241272,
      "num_tokens": 14692723.0,
      "step": 6785
    },
    {
      "entropy": 6.555766201019287,
      "epoch": 0.7266306383434106,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004956595156329254,
      "loss": 6.339,
      "mean_token_accuracy": 0.13901505172252654,
      "num_tokens": 14703824.0,
      "step": 6790
    },
    {
      "entropy": 6.536800241470337,
      "epoch": 0.7271657124511745,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000495652038833775,
      "loss": 6.3199,
      "mean_token_accuracy": 0.14263801351189614,
      "num_tokens": 14715934.0,
      "step": 6795
    },
    {
      "entropy": 6.495520067214966,
      "epoch": 0.7277007865589384,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004956445556633279,
      "loss": 6.3266,
      "mean_token_accuracy": 0.1356288030743599,
      "num_tokens": 14725617.0,
      "step": 6800
    },
    {
      "entropy": 6.548426246643066,
      "epoch": 0.7282358606667023,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004956370661218003,
      "loss": 6.2968,
      "mean_token_accuracy": 0.14091143608093262,
      "num_tokens": 14736430.0,
      "step": 6805
    },
    {
      "entropy": 6.523364543914795,
      "epoch": 0.7287709347744663,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004956295702094084,
      "loss": 6.3017,
      "mean_token_accuracy": 0.14062364101409913,
      "num_tokens": 14746740.0,
      "step": 6810
    },
    {
      "entropy": 6.449228715896607,
      "epoch": 0.7293060088822302,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004956220679263687,
      "loss": 6.2474,
      "mean_token_accuracy": 0.14277183413505554,
      "num_tokens": 14756884.0,
      "step": 6815
    },
    {
      "entropy": 6.5031599521636965,
      "epoch": 0.7298410829899941,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004956145592728976,
      "loss": 6.2846,
      "mean_token_accuracy": 0.14055393263697624,
      "num_tokens": 14766974.0,
      "step": 6820
    },
    {
      "entropy": 6.512714481353759,
      "epoch": 0.730376157097758,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004956070442492123,
      "loss": 6.3963,
      "mean_token_accuracy": 0.1339946910738945,
      "num_tokens": 14778449.0,
      "step": 6825
    },
    {
      "entropy": 6.59020767211914,
      "epoch": 0.7309112312055219,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004955995228555296,
      "loss": 6.3664,
      "mean_token_accuracy": 0.13419365808367728,
      "num_tokens": 14788989.0,
      "step": 6830
    },
    {
      "entropy": 6.5118653774261475,
      "epoch": 0.7314463053132859,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004955919950920666,
      "loss": 6.324,
      "mean_token_accuracy": 0.1472254738211632,
      "num_tokens": 14798454.0,
      "step": 6835
    },
    {
      "entropy": 6.549294137954712,
      "epoch": 0.7319813794210498,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004955844609590408,
      "loss": 6.2685,
      "mean_token_accuracy": 0.13655681014060975,
      "num_tokens": 14808920.0,
      "step": 6840
    },
    {
      "entropy": 6.490383720397949,
      "epoch": 0.7325164535288138,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004955769204566696,
      "loss": 6.2619,
      "mean_token_accuracy": 0.14031370729207993,
      "num_tokens": 14819250.0,
      "step": 6845
    },
    {
      "entropy": 6.547230529785156,
      "epoch": 0.7330515276365777,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004955693735851709,
      "loss": 6.3222,
      "mean_token_accuracy": 0.13652494847774505,
      "num_tokens": 14829789.0,
      "step": 6850
    },
    {
      "entropy": 6.527000188827515,
      "epoch": 0.7335866017443415,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004955618203447625,
      "loss": 6.3076,
      "mean_token_accuracy": 0.1329828307032585,
      "num_tokens": 14840527.0,
      "step": 6855
    },
    {
      "entropy": 6.51408896446228,
      "epoch": 0.7341216758521055,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004955542607356625,
      "loss": 6.3656,
      "mean_token_accuracy": 0.1393475666642189,
      "num_tokens": 14850922.0,
      "step": 6860
    },
    {
      "entropy": 6.557836198806763,
      "epoch": 0.7346567499598694,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004955466947580893,
      "loss": 6.3096,
      "mean_token_accuracy": 0.13479100167751312,
      "num_tokens": 14862226.0,
      "step": 6865
    },
    {
      "entropy": 6.520762014389038,
      "epoch": 0.7351918240676334,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004955391224122611,
      "loss": 6.2171,
      "mean_token_accuracy": 0.15029082745313643,
      "num_tokens": 14872461.0,
      "step": 6870
    },
    {
      "entropy": 6.478526496887207,
      "epoch": 0.7357268981753973,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000495531543698397,
      "loss": 6.3394,
      "mean_token_accuracy": 0.1389941766858101,
      "num_tokens": 14883088.0,
      "step": 6875
    },
    {
      "entropy": 6.548675584793091,
      "epoch": 0.7362619722831613,
      "grad_norm": 0.875,
      "learning_rate": 0.0004955239586167153,
      "loss": 6.3411,
      "mean_token_accuracy": 0.13967593759298325,
      "num_tokens": 14894055.0,
      "step": 6880
    },
    {
      "entropy": 6.5129618644714355,
      "epoch": 0.7367970463909251,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004955163671674354,
      "loss": 6.3237,
      "mean_token_accuracy": 0.1361086040735245,
      "num_tokens": 14904760.0,
      "step": 6885
    },
    {
      "entropy": 6.604186344146728,
      "epoch": 0.737332120498689,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004955087693507764,
      "loss": 6.4348,
      "mean_token_accuracy": 0.1315928190946579,
      "num_tokens": 14915983.0,
      "step": 6890
    },
    {
      "entropy": 6.532724380493164,
      "epoch": 0.737867194606453,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004955011651669577,
      "loss": 6.3269,
      "mean_token_accuracy": 0.1342775173485279,
      "num_tokens": 14925959.0,
      "step": 6895
    },
    {
      "entropy": 6.521912097930908,
      "epoch": 0.7384022687142169,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004954935546161988,
      "loss": 6.2753,
      "mean_token_accuracy": 0.13685757741332055,
      "num_tokens": 14936138.0,
      "step": 6900
    },
    {
      "entropy": 6.575437068939209,
      "epoch": 0.7389373428219809,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004954859376987195,
      "loss": 6.2955,
      "mean_token_accuracy": 0.14261991381645203,
      "num_tokens": 14946979.0,
      "step": 6905
    },
    {
      "entropy": 6.454982900619507,
      "epoch": 0.7394724169297447,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004954783144147397,
      "loss": 6.3498,
      "mean_token_accuracy": 0.1346730634570122,
      "num_tokens": 14958324.0,
      "step": 6910
    },
    {
      "entropy": 6.618794679641724,
      "epoch": 0.7400074910375087,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004954706847644796,
      "loss": 6.3639,
      "mean_token_accuracy": 0.1349130615592003,
      "num_tokens": 14969048.0,
      "step": 6915
    },
    {
      "entropy": 6.579623413085938,
      "epoch": 0.7405425651452726,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004954630487481594,
      "loss": 6.3414,
      "mean_token_accuracy": 0.1382133223116398,
      "num_tokens": 14979825.0,
      "step": 6920
    },
    {
      "entropy": 6.470085096359253,
      "epoch": 0.7410776392530365,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004954554063659998,
      "loss": 6.2848,
      "mean_token_accuracy": 0.13979439586400985,
      "num_tokens": 14990138.0,
      "step": 6925
    },
    {
      "entropy": 6.4993678569793705,
      "epoch": 0.7416127133608005,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004954477576182212,
      "loss": 6.2378,
      "mean_token_accuracy": 0.1421709269285202,
      "num_tokens": 15002276.0,
      "step": 6930
    },
    {
      "entropy": 6.564362239837647,
      "epoch": 0.7421477874685644,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004954401025050445,
      "loss": 6.3841,
      "mean_token_accuracy": 0.13183799609541894,
      "num_tokens": 15013622.0,
      "step": 6935
    },
    {
      "entropy": 6.558071756362915,
      "epoch": 0.7426828615763283,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004954324410266909,
      "loss": 6.3302,
      "mean_token_accuracy": 0.13709006309509278,
      "num_tokens": 15024067.0,
      "step": 6940
    },
    {
      "entropy": 6.538510179519653,
      "epoch": 0.7432179356840922,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004954247731833815,
      "loss": 6.2284,
      "mean_token_accuracy": 0.1457581028342247,
      "num_tokens": 15033646.0,
      "step": 6945
    },
    {
      "entropy": 6.45111870765686,
      "epoch": 0.7437530097918562,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004954170989753377,
      "loss": 6.285,
      "mean_token_accuracy": 0.13467907086014747,
      "num_tokens": 15044655.0,
      "step": 6950
    },
    {
      "entropy": 6.5969995021820065,
      "epoch": 0.7442880838996201,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004954094184027813,
      "loss": 6.3382,
      "mean_token_accuracy": 0.1325977064669132,
      "num_tokens": 15055492.0,
      "step": 6955
    },
    {
      "entropy": 6.575356817245483,
      "epoch": 0.744823158007384,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004954017314659339,
      "loss": 6.3293,
      "mean_token_accuracy": 0.13015183433890343,
      "num_tokens": 15066472.0,
      "step": 6960
    },
    {
      "entropy": 6.547945976257324,
      "epoch": 0.745358232115148,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004953940381650174,
      "loss": 6.2754,
      "mean_token_accuracy": 0.1464118979871273,
      "num_tokens": 15076851.0,
      "step": 6965
    },
    {
      "entropy": 6.414773559570312,
      "epoch": 0.7458933062229118,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004953863385002541,
      "loss": 6.2115,
      "mean_token_accuracy": 0.14125035405158998,
      "num_tokens": 15086997.0,
      "step": 6970
    },
    {
      "entropy": 6.541490793228149,
      "epoch": 0.7464283803306758,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004953786324718661,
      "loss": 6.3837,
      "mean_token_accuracy": 0.13314245343208314,
      "num_tokens": 15098625.0,
      "step": 6975
    },
    {
      "entropy": 6.511341094970703,
      "epoch": 0.7469634544384397,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004953709200800761,
      "loss": 6.2045,
      "mean_token_accuracy": 0.14473722204566003,
      "num_tokens": 15108265.0,
      "step": 6980
    },
    {
      "entropy": 6.541043519973755,
      "epoch": 0.7474985285462037,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004953632013251068,
      "loss": 6.306,
      "mean_token_accuracy": 0.1355377197265625,
      "num_tokens": 15118794.0,
      "step": 6985
    },
    {
      "entropy": 6.528295278549194,
      "epoch": 0.7480336026539676,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004953554762071811,
      "loss": 6.3256,
      "mean_token_accuracy": 0.1343781903386116,
      "num_tokens": 15130084.0,
      "step": 6990
    },
    {
      "entropy": 6.504166746139527,
      "epoch": 0.7485686767617316,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004953477447265218,
      "loss": 6.3191,
      "mean_token_accuracy": 0.13526385575532912,
      "num_tokens": 15142193.0,
      "step": 6995
    },
    {
      "entropy": 6.579152584075928,
      "epoch": 0.7491037508694954,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004953400068833524,
      "loss": 6.4286,
      "mean_token_accuracy": 0.12982157617807388,
      "num_tokens": 15152817.0,
      "step": 7000
    },
    {
      "entropy": 6.481089448928833,
      "epoch": 0.7496388249772593,
      "grad_norm": 1.0,
      "learning_rate": 0.0004953322626778964,
      "loss": 6.1064,
      "mean_token_accuracy": 0.1475074350833893,
      "num_tokens": 15162352.0,
      "step": 7005
    },
    {
      "entropy": 6.530549764633179,
      "epoch": 0.7501738990850233,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004953245121103771,
      "loss": 6.2921,
      "mean_token_accuracy": 0.14548757672309875,
      "num_tokens": 15173091.0,
      "step": 7010
    },
    {
      "entropy": 6.598561382293701,
      "epoch": 0.7507089731927872,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004953167551810185,
      "loss": 6.443,
      "mean_token_accuracy": 0.12762990221381187,
      "num_tokens": 15184422.0,
      "step": 7015
    },
    {
      "entropy": 6.460023689270019,
      "epoch": 0.7512440473005512,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004953089918900447,
      "loss": 6.1932,
      "mean_token_accuracy": 0.14750462397933006,
      "num_tokens": 15194765.0,
      "step": 7020
    },
    {
      "entropy": 6.5149589538574215,
      "epoch": 0.751779121408315,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004953012222376795,
      "loss": 6.3467,
      "mean_token_accuracy": 0.13310825303196908,
      "num_tokens": 15206007.0,
      "step": 7025
    },
    {
      "entropy": 6.551244401931763,
      "epoch": 0.752314195516079,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004952934462241476,
      "loss": 6.2864,
      "mean_token_accuracy": 0.14164330810308456,
      "num_tokens": 15216634.0,
      "step": 7030
    },
    {
      "entropy": 6.539538335800171,
      "epoch": 0.7528492696238429,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004952856638496734,
      "loss": 6.3512,
      "mean_token_accuracy": 0.13117952421307563,
      "num_tokens": 15226931.0,
      "step": 7035
    },
    {
      "entropy": 6.524749994277954,
      "epoch": 0.7533843437316068,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004952778751144817,
      "loss": 6.3248,
      "mean_token_accuracy": 0.13734631463885308,
      "num_tokens": 15239034.0,
      "step": 7040
    },
    {
      "entropy": 6.566304445266724,
      "epoch": 0.7539194178393708,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004952700800187971,
      "loss": 6.3099,
      "mean_token_accuracy": 0.13461821675300598,
      "num_tokens": 15250088.0,
      "step": 7045
    },
    {
      "entropy": 6.544210958480835,
      "epoch": 0.7544544919471347,
      "grad_norm": 1.25,
      "learning_rate": 0.000495262278562845,
      "loss": 6.2501,
      "mean_token_accuracy": 0.1372145541012287,
      "num_tokens": 15261150.0,
      "step": 7050
    },
    {
      "entropy": 6.552422380447387,
      "epoch": 0.7549895660548986,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004952544707468506,
      "loss": 6.3586,
      "mean_token_accuracy": 0.13499465733766555,
      "num_tokens": 15270865.0,
      "step": 7055
    },
    {
      "entropy": 6.519015741348267,
      "epoch": 0.7555246401626625,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004952466565710391,
      "loss": 6.3621,
      "mean_token_accuracy": 0.13283906280994415,
      "num_tokens": 15282809.0,
      "step": 7060
    },
    {
      "entropy": 6.513560247421265,
      "epoch": 0.7560597142704265,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004952388360356366,
      "loss": 6.3135,
      "mean_token_accuracy": 0.14236335307359696,
      "num_tokens": 15293166.0,
      "step": 7065
    },
    {
      "entropy": 6.519959115982056,
      "epoch": 0.7565947883781904,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004952310091408685,
      "loss": 6.2811,
      "mean_token_accuracy": 0.1420826405286789,
      "num_tokens": 15302985.0,
      "step": 7070
    },
    {
      "entropy": 6.504639148712158,
      "epoch": 0.7571298624859543,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004952231758869611,
      "loss": 6.3426,
      "mean_token_accuracy": 0.1433715745806694,
      "num_tokens": 15314737.0,
      "step": 7075
    },
    {
      "entropy": 6.5021086692810055,
      "epoch": 0.7576649365937183,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004952153362741403,
      "loss": 6.2748,
      "mean_token_accuracy": 0.1443006820976734,
      "num_tokens": 15325441.0,
      "step": 7080
    },
    {
      "entropy": 6.490581083297729,
      "epoch": 0.7582000107014821,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004952074903026327,
      "loss": 6.2177,
      "mean_token_accuracy": 0.14506308287382125,
      "num_tokens": 15334939.0,
      "step": 7085
    },
    {
      "entropy": 6.52271990776062,
      "epoch": 0.7587350848092461,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004951996379726647,
      "loss": 6.364,
      "mean_token_accuracy": 0.1351069279015064,
      "num_tokens": 15345655.0,
      "step": 7090
    },
    {
      "entropy": 6.458598327636719,
      "epoch": 0.75927015891701,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004951917792844631,
      "loss": 6.2046,
      "mean_token_accuracy": 0.14718087762594223,
      "num_tokens": 15356035.0,
      "step": 7095
    },
    {
      "entropy": 6.495374298095703,
      "epoch": 0.759805233024774,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004951839142382548,
      "loss": 6.3023,
      "mean_token_accuracy": 0.13995541483163834,
      "num_tokens": 15368509.0,
      "step": 7100
    },
    {
      "entropy": 6.512107515335083,
      "epoch": 0.7603403071325379,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000495176042834267,
      "loss": 6.2935,
      "mean_token_accuracy": 0.14130270704627038,
      "num_tokens": 15379662.0,
      "step": 7105
    },
    {
      "entropy": 6.502401638031006,
      "epoch": 0.7608753812403017,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004951681650727268,
      "loss": 6.2795,
      "mean_token_accuracy": 0.13214101269841194,
      "num_tokens": 15389698.0,
      "step": 7110
    },
    {
      "entropy": 6.577340078353882,
      "epoch": 0.7614104553480657,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004951602809538619,
      "loss": 6.3812,
      "mean_token_accuracy": 0.13417631536722183,
      "num_tokens": 15400724.0,
      "step": 7115
    },
    {
      "entropy": 6.651620864868164,
      "epoch": 0.7619455294558296,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004951523904778997,
      "loss": 6.3795,
      "mean_token_accuracy": 0.13378686904907228,
      "num_tokens": 15410732.0,
      "step": 7120
    },
    {
      "entropy": 6.5081226348876955,
      "epoch": 0.7624806035635936,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004951444936450682,
      "loss": 6.2835,
      "mean_token_accuracy": 0.1417149245738983,
      "num_tokens": 15420185.0,
      "step": 7125
    },
    {
      "entropy": 6.4529128074646,
      "epoch": 0.7630156776713575,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004951365904555954,
      "loss": 6.1971,
      "mean_token_accuracy": 0.14506204500794412,
      "num_tokens": 15431904.0,
      "step": 7130
    },
    {
      "entropy": 6.486575889587402,
      "epoch": 0.7635507517791215,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004951286809097094,
      "loss": 6.3355,
      "mean_token_accuracy": 0.13317205905914306,
      "num_tokens": 15442650.0,
      "step": 7135
    },
    {
      "entropy": 6.51876654624939,
      "epoch": 0.7640858258868853,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004951207650076388,
      "loss": 6.3058,
      "mean_token_accuracy": 0.13968102112412453,
      "num_tokens": 15454273.0,
      "step": 7140
    },
    {
      "entropy": 6.557489824295044,
      "epoch": 0.7646208999946492,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000495112842749612,
      "loss": 6.2649,
      "mean_token_accuracy": 0.14266380667686462,
      "num_tokens": 15465751.0,
      "step": 7145
    },
    {
      "entropy": 6.506114387512207,
      "epoch": 0.7651559741024132,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004951049141358578,
      "loss": 6.3921,
      "mean_token_accuracy": 0.132391407340765,
      "num_tokens": 15475340.0,
      "step": 7150
    },
    {
      "entropy": 6.581349515914917,
      "epoch": 0.7656910482101771,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004950969791666051,
      "loss": 6.3437,
      "mean_token_accuracy": 0.1322751335799694,
      "num_tokens": 15486246.0,
      "step": 7155
    },
    {
      "entropy": 6.5949461460113525,
      "epoch": 0.7662261223179411,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004950890378420831,
      "loss": 6.3449,
      "mean_token_accuracy": 0.13105324506759644,
      "num_tokens": 15497331.0,
      "step": 7160
    },
    {
      "entropy": 6.551503896713257,
      "epoch": 0.766761196425705,
      "grad_norm": 0.84375,
      "learning_rate": 0.000495081090162521,
      "loss": 6.2884,
      "mean_token_accuracy": 0.14032624363899232,
      "num_tokens": 15508036.0,
      "step": 7165
    },
    {
      "entropy": 6.559307289123535,
      "epoch": 0.7672962705334689,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004950731361281483,
      "loss": 6.3013,
      "mean_token_accuracy": 0.1361111432313919,
      "num_tokens": 15518381.0,
      "step": 7170
    },
    {
      "entropy": 6.505682373046875,
      "epoch": 0.7678313446412328,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004950651757391948,
      "loss": 6.2988,
      "mean_token_accuracy": 0.13651533499360086,
      "num_tokens": 15528939.0,
      "step": 7175
    },
    {
      "entropy": 6.531873035430908,
      "epoch": 0.7683664187489967,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004950572089958904,
      "loss": 6.2715,
      "mean_token_accuracy": 0.13551892414689065,
      "num_tokens": 15539718.0,
      "step": 7180
    },
    {
      "entropy": 6.569702625274658,
      "epoch": 0.7689014928567607,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004950492358984648,
      "loss": 6.4458,
      "mean_token_accuracy": 0.12883288115262986,
      "num_tokens": 15550970.0,
      "step": 7185
    },
    {
      "entropy": 6.562612676620484,
      "epoch": 0.7694365669645246,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004950412564471486,
      "loss": 6.2157,
      "mean_token_accuracy": 0.14375220090150834,
      "num_tokens": 15560490.0,
      "step": 7190
    },
    {
      "entropy": 6.476634454727173,
      "epoch": 0.7699716410722885,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000495033270642172,
      "loss": 6.2213,
      "mean_token_accuracy": 0.14110628366470337,
      "num_tokens": 15570188.0,
      "step": 7195
    },
    {
      "entropy": 6.466872978210449,
      "epoch": 0.7705067151800524,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004950252784837655,
      "loss": 6.2564,
      "mean_token_accuracy": 0.1300390236079693,
      "num_tokens": 15581446.0,
      "step": 7200
    },
    {
      "entropy": 6.5604804992675785,
      "epoch": 0.7710417892878164,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004950172799721601,
      "loss": 6.3564,
      "mean_token_accuracy": 0.13786424696445465,
      "num_tokens": 15594362.0,
      "step": 7205
    },
    {
      "entropy": 6.505185461044311,
      "epoch": 0.7715768633955803,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004950092751075866,
      "loss": 6.2248,
      "mean_token_accuracy": 0.14566915854811668,
      "num_tokens": 15605531.0,
      "step": 7210
    },
    {
      "entropy": 6.497638607025147,
      "epoch": 0.7721119375033442,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004950012638902763,
      "loss": 6.2792,
      "mean_token_accuracy": 0.14333579689264297,
      "num_tokens": 15614962.0,
      "step": 7215
    },
    {
      "entropy": 6.473331880569458,
      "epoch": 0.7726470116111082,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004949932463204603,
      "loss": 6.3103,
      "mean_token_accuracy": 0.1413653999567032,
      "num_tokens": 15625475.0,
      "step": 7220
    },
    {
      "entropy": 6.507373666763305,
      "epoch": 0.773182085718872,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004949852223983703,
      "loss": 6.2919,
      "mean_token_accuracy": 0.13588140457868575,
      "num_tokens": 15637359.0,
      "step": 7225
    },
    {
      "entropy": 6.683114147186279,
      "epoch": 0.773717159826636,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004949771921242379,
      "loss": 6.3368,
      "mean_token_accuracy": 0.13007210791110993,
      "num_tokens": 15648368.0,
      "step": 7230
    },
    {
      "entropy": 6.558062982559204,
      "epoch": 0.7742522339343999,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004949691554982951,
      "loss": 6.2277,
      "mean_token_accuracy": 0.14604199230670928,
      "num_tokens": 15659295.0,
      "step": 7235
    },
    {
      "entropy": 6.469854736328125,
      "epoch": 0.7747873080421639,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004949611125207737,
      "loss": 6.2022,
      "mean_token_accuracy": 0.13703280910849572,
      "num_tokens": 15669975.0,
      "step": 7240
    },
    {
      "entropy": 6.430458831787109,
      "epoch": 0.7753223821499278,
      "grad_norm": 0.921875,
      "learning_rate": 0.000494953063191906,
      "loss": 6.2885,
      "mean_token_accuracy": 0.13726247996091842,
      "num_tokens": 15680343.0,
      "step": 7245
    },
    {
      "entropy": 6.5502770900726315,
      "epoch": 0.7758574562576916,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004949450075119247,
      "loss": 6.3639,
      "mean_token_accuracy": 0.13152319192886353,
      "num_tokens": 15692246.0,
      "step": 7250
    },
    {
      "entropy": 6.571636486053467,
      "epoch": 0.7763925303654556,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004949369454810621,
      "loss": 6.2871,
      "mean_token_accuracy": 0.14085712507367135,
      "num_tokens": 15704262.0,
      "step": 7255
    },
    {
      "entropy": 6.47925443649292,
      "epoch": 0.7769276044732195,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004949288770995512,
      "loss": 6.3129,
      "mean_token_accuracy": 0.13671249523758888,
      "num_tokens": 15715078.0,
      "step": 7260
    },
    {
      "entropy": 6.512346458435059,
      "epoch": 0.7774626785809835,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004949208023676249,
      "loss": 6.3437,
      "mean_token_accuracy": 0.13721415549516677,
      "num_tokens": 15726503.0,
      "step": 7265
    },
    {
      "entropy": 6.4813494205474855,
      "epoch": 0.7779977526887474,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004949127212855161,
      "loss": 6.2045,
      "mean_token_accuracy": 0.143607996404171,
      "num_tokens": 15738854.0,
      "step": 7270
    },
    {
      "entropy": 6.522124195098877,
      "epoch": 0.7785328267965114,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004949046338534587,
      "loss": 6.3836,
      "mean_token_accuracy": 0.13659268692135812,
      "num_tokens": 15750836.0,
      "step": 7275
    },
    {
      "entropy": 6.646920108795166,
      "epoch": 0.7790679009042752,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004948965400716857,
      "loss": 6.3841,
      "mean_token_accuracy": 0.1333204463124275,
      "num_tokens": 15762355.0,
      "step": 7280
    },
    {
      "entropy": 6.6230309009552,
      "epoch": 0.7796029750120391,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000494888439940431,
      "loss": 6.3534,
      "mean_token_accuracy": 0.13607660681009293,
      "num_tokens": 15772952.0,
      "step": 7285
    },
    {
      "entropy": 6.444079256057739,
      "epoch": 0.7801380491198031,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004948803334599286,
      "loss": 6.1894,
      "mean_token_accuracy": 0.14478645697236062,
      "num_tokens": 15782997.0,
      "step": 7290
    },
    {
      "entropy": 6.455467271804809,
      "epoch": 0.780673123227567,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004948722206304123,
      "loss": 6.2918,
      "mean_token_accuracy": 0.138990381360054,
      "num_tokens": 15794060.0,
      "step": 7295
    },
    {
      "entropy": 6.578930997848511,
      "epoch": 0.781208197335331,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004948641014521167,
      "loss": 6.3502,
      "mean_token_accuracy": 0.13420951142907142,
      "num_tokens": 15804149.0,
      "step": 7300
    },
    {
      "entropy": 6.543324518203735,
      "epoch": 0.7817432714430949,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000494855975925276,
      "loss": 6.3128,
      "mean_token_accuracy": 0.13801790848374368,
      "num_tokens": 15814643.0,
      "step": 7305
    },
    {
      "entropy": 6.5630086898803714,
      "epoch": 0.7822783455508588,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004948478440501249,
      "loss": 6.2576,
      "mean_token_accuracy": 0.13937826454639435,
      "num_tokens": 15824640.0,
      "step": 7310
    },
    {
      "entropy": 6.443582963943482,
      "epoch": 0.7828134196586227,
      "grad_norm": 1.0,
      "learning_rate": 0.0004948397058268982,
      "loss": 6.2746,
      "mean_token_accuracy": 0.14026655852794648,
      "num_tokens": 15836160.0,
      "step": 7315
    },
    {
      "entropy": 6.480792999267578,
      "epoch": 0.7833484937663866,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004948315612558308,
      "loss": 6.3219,
      "mean_token_accuracy": 0.13636694997549056,
      "num_tokens": 15846886.0,
      "step": 7320
    },
    {
      "entropy": 6.528878974914551,
      "epoch": 0.7838835678741506,
      "grad_norm": 0.90625,
      "learning_rate": 0.000494823410337158,
      "loss": 6.2334,
      "mean_token_accuracy": 0.1459447741508484,
      "num_tokens": 15857309.0,
      "step": 7325
    },
    {
      "entropy": 6.534002637863159,
      "epoch": 0.7844186419819145,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004948152530711153,
      "loss": 6.3793,
      "mean_token_accuracy": 0.13519303053617476,
      "num_tokens": 15869620.0,
      "step": 7330
    },
    {
      "entropy": 6.4702476978302,
      "epoch": 0.7849537160896785,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004948070894579378,
      "loss": 6.2311,
      "mean_token_accuracy": 0.144602470099926,
      "num_tokens": 15880915.0,
      "step": 7335
    },
    {
      "entropy": 6.4955573081970215,
      "epoch": 0.7854887901974423,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004947989194978616,
      "loss": 6.2712,
      "mean_token_accuracy": 0.1428148165345192,
      "num_tokens": 15891150.0,
      "step": 7340
    },
    {
      "entropy": 6.527264595031738,
      "epoch": 0.7860238643052063,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004947907431911225,
      "loss": 6.2905,
      "mean_token_accuracy": 0.14462609365582466,
      "num_tokens": 15902526.0,
      "step": 7345
    },
    {
      "entropy": 6.476318740844727,
      "epoch": 0.7865589384129702,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004947825605379566,
      "loss": 6.2289,
      "mean_token_accuracy": 0.14542306140065192,
      "num_tokens": 15913425.0,
      "step": 7350
    },
    {
      "entropy": 6.498827075958252,
      "epoch": 0.7870940125207341,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004947743715386,
      "loss": 6.2063,
      "mean_token_accuracy": 0.13952323347330092,
      "num_tokens": 15924324.0,
      "step": 7355
    },
    {
      "entropy": 6.465553617477417,
      "epoch": 0.7876290866284981,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004947661761932894,
      "loss": 6.26,
      "mean_token_accuracy": 0.1387181557714939,
      "num_tokens": 15935533.0,
      "step": 7360
    },
    {
      "entropy": 6.505006647109985,
      "epoch": 0.7881641607362619,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004947579745022613,
      "loss": 6.2148,
      "mean_token_accuracy": 0.13753947615623474,
      "num_tokens": 15947007.0,
      "step": 7365
    },
    {
      "entropy": 6.476586437225341,
      "epoch": 0.7886992348440259,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004947497664657527,
      "loss": 6.2928,
      "mean_token_accuracy": 0.13322581648826598,
      "num_tokens": 15957405.0,
      "step": 7370
    },
    {
      "entropy": 6.486299657821656,
      "epoch": 0.7892343089517898,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004947415520840003,
      "loss": 6.2495,
      "mean_token_accuracy": 0.14630692303180695,
      "num_tokens": 15969895.0,
      "step": 7375
    },
    {
      "entropy": 6.485176467895508,
      "epoch": 0.7897693830595538,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004947333313572416,
      "loss": 6.1816,
      "mean_token_accuracy": 0.14981536716222763,
      "num_tokens": 15979854.0,
      "step": 7380
    },
    {
      "entropy": 6.40966272354126,
      "epoch": 0.7903044571673177,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004947251042857137,
      "loss": 6.1905,
      "mean_token_accuracy": 0.14402236491441728,
      "num_tokens": 15991341.0,
      "step": 7385
    },
    {
      "entropy": 6.466861581802368,
      "epoch": 0.7908395312750816,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004947168708696544,
      "loss": 6.2411,
      "mean_token_accuracy": 0.14322528690099717,
      "num_tokens": 16001944.0,
      "step": 7390
    },
    {
      "entropy": 6.463384962081909,
      "epoch": 0.7913746053828455,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004947086311093013,
      "loss": 6.2484,
      "mean_token_accuracy": 0.14022847190499305,
      "num_tokens": 16011771.0,
      "step": 7395
    },
    {
      "entropy": 6.536163091659546,
      "epoch": 0.7919096794906094,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004947003850048924,
      "loss": 6.3837,
      "mean_token_accuracy": 0.1262681730091572,
      "num_tokens": 16023901.0,
      "step": 7400
    },
    {
      "entropy": 6.573896312713623,
      "epoch": 0.7924447535983734,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004946921325566656,
      "loss": 6.3139,
      "mean_token_accuracy": 0.13470809012651444,
      "num_tokens": 16035014.0,
      "step": 7405
    },
    {
      "entropy": 6.566063022613525,
      "epoch": 0.7929798277061373,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004946838737648595,
      "loss": 6.3781,
      "mean_token_accuracy": 0.13269267454743386,
      "num_tokens": 16046927.0,
      "step": 7410
    },
    {
      "entropy": 6.589928483963012,
      "epoch": 0.7935149018139013,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004946756086297124,
      "loss": 6.2954,
      "mean_token_accuracy": 0.14437760785222054,
      "num_tokens": 16057341.0,
      "step": 7415
    },
    {
      "entropy": 6.512710094451904,
      "epoch": 0.7940499759216652,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004946673371514628,
      "loss": 6.2678,
      "mean_token_accuracy": 0.1392816685140133,
      "num_tokens": 16068408.0,
      "step": 7420
    },
    {
      "entropy": 6.451438045501709,
      "epoch": 0.794585050029429,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004946590593303499,
      "loss": 6.2406,
      "mean_token_accuracy": 0.14896183758974074,
      "num_tokens": 16078693.0,
      "step": 7425
    },
    {
      "entropy": 6.485875940322876,
      "epoch": 0.795120124137193,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004946507751666124,
      "loss": 6.2835,
      "mean_token_accuracy": 0.14007917866110803,
      "num_tokens": 16089127.0,
      "step": 7430
    },
    {
      "entropy": 6.452515411376953,
      "epoch": 0.7956551982449569,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004946424846604897,
      "loss": 6.188,
      "mean_token_accuracy": 0.15252198725938798,
      "num_tokens": 16099726.0,
      "step": 7435
    },
    {
      "entropy": 6.505970287322998,
      "epoch": 0.7961902723527209,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004946341878122212,
      "loss": 6.2498,
      "mean_token_accuracy": 0.13865032494068147,
      "num_tokens": 16111017.0,
      "step": 7440
    },
    {
      "entropy": 6.496991491317749,
      "epoch": 0.7967253464604848,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004946258846220462,
      "loss": 6.3182,
      "mean_token_accuracy": 0.14124254807829856,
      "num_tokens": 16121406.0,
      "step": 7445
    },
    {
      "entropy": 6.497278690338135,
      "epoch": 0.7972604205682488,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004946175750902049,
      "loss": 6.2671,
      "mean_token_accuracy": 0.14234999492764472,
      "num_tokens": 16131587.0,
      "step": 7450
    },
    {
      "entropy": 6.464297771453857,
      "epoch": 0.7977954946760126,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004946092592169368,
      "loss": 6.1936,
      "mean_token_accuracy": 0.1488217905163765,
      "num_tokens": 16142578.0,
      "step": 7455
    },
    {
      "entropy": 6.51997389793396,
      "epoch": 0.7983305687837765,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004946009370024822,
      "loss": 6.2168,
      "mean_token_accuracy": 0.14661495462059976,
      "num_tokens": 16153501.0,
      "step": 7460
    },
    {
      "entropy": 6.48798394203186,
      "epoch": 0.7988656428915405,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004945926084470814,
      "loss": 6.302,
      "mean_token_accuracy": 0.13927897214889526,
      "num_tokens": 16164273.0,
      "step": 7465
    },
    {
      "entropy": 6.537490653991699,
      "epoch": 0.7994007169993044,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004945842735509749,
      "loss": 6.2882,
      "mean_token_accuracy": 0.14106250405311585,
      "num_tokens": 16175071.0,
      "step": 7470
    },
    {
      "entropy": 6.559529399871826,
      "epoch": 0.7999357911070684,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004945759323144034,
      "loss": 6.3272,
      "mean_token_accuracy": 0.13412191048264505,
      "num_tokens": 16185404.0,
      "step": 7475
    },
    {
      "entropy": 6.500086450576783,
      "epoch": 0.8004708652148322,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004945675847376077,
      "loss": 6.292,
      "mean_token_accuracy": 0.13708024621009826,
      "num_tokens": 16196881.0,
      "step": 7480
    },
    {
      "entropy": 6.570698213577271,
      "epoch": 0.8010059393225962,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004945592308208288,
      "loss": 6.3272,
      "mean_token_accuracy": 0.13322533369064332,
      "num_tokens": 16207534.0,
      "step": 7485
    },
    {
      "entropy": 6.506867742538452,
      "epoch": 0.8015410134303601,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000494550870564308,
      "loss": 6.1949,
      "mean_token_accuracy": 0.15050898641347885,
      "num_tokens": 16217371.0,
      "step": 7490
    },
    {
      "entropy": 6.510013294219971,
      "epoch": 0.802076087538124,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004945425039682866,
      "loss": 6.2586,
      "mean_token_accuracy": 0.14045843482017517,
      "num_tokens": 16227898.0,
      "step": 7495
    },
    {
      "entropy": 6.509621810913086,
      "epoch": 0.802611161645888,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004945341310330064,
      "loss": 6.2414,
      "mean_token_accuracy": 0.14420711621642113,
      "num_tokens": 16238966.0,
      "step": 7500
    },
    {
      "entropy": 6.464070987701416,
      "epoch": 0.8031462357536518,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004945257517587089,
      "loss": 6.2551,
      "mean_token_accuracy": 0.14895280450582504,
      "num_tokens": 16250129.0,
      "step": 7505
    },
    {
      "entropy": 6.519074869155884,
      "epoch": 0.8036813098614158,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004945173661456361,
      "loss": 6.3103,
      "mean_token_accuracy": 0.13587038367986679,
      "num_tokens": 16260877.0,
      "step": 7510
    },
    {
      "entropy": 6.546694660186768,
      "epoch": 0.8042163839691797,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004945089741940303,
      "loss": 6.2689,
      "mean_token_accuracy": 0.14226654469966887,
      "num_tokens": 16270548.0,
      "step": 7515
    },
    {
      "entropy": 6.504742097854614,
      "epoch": 0.8047514580769437,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004945005759041338,
      "loss": 6.2656,
      "mean_token_accuracy": 0.13865245431661605,
      "num_tokens": 16281014.0,
      "step": 7520
    },
    {
      "entropy": 6.519778442382813,
      "epoch": 0.8052865321847076,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004944921712761889,
      "loss": 6.326,
      "mean_token_accuracy": 0.138186539709568,
      "num_tokens": 16291653.0,
      "step": 7525
    },
    {
      "entropy": 6.530219650268554,
      "epoch": 0.8058216062924715,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004944837603104383,
      "loss": 6.1934,
      "mean_token_accuracy": 0.14021202996373178,
      "num_tokens": 16301703.0,
      "step": 7530
    },
    {
      "entropy": 6.448696517944336,
      "epoch": 0.8063566804002354,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004944753430071252,
      "loss": 6.2575,
      "mean_token_accuracy": 0.14102583453059198,
      "num_tokens": 16312252.0,
      "step": 7535
    },
    {
      "entropy": 6.453697347640992,
      "epoch": 0.8068917545079993,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004944669193664923,
      "loss": 6.2014,
      "mean_token_accuracy": 0.14184984788298607,
      "num_tokens": 16324073.0,
      "step": 7540
    },
    {
      "entropy": 6.533971309661865,
      "epoch": 0.8074268286157633,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004944584893887829,
      "loss": 6.2485,
      "mean_token_accuracy": 0.1400940850377083,
      "num_tokens": 16334637.0,
      "step": 7545
    },
    {
      "entropy": 6.4854789733886715,
      "epoch": 0.8079619027235272,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004944500530742404,
      "loss": 6.3561,
      "mean_token_accuracy": 0.13076531440019606,
      "num_tokens": 16345581.0,
      "step": 7550
    },
    {
      "entropy": 6.54761118888855,
      "epoch": 0.8084969768312912,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004944416104231086,
      "loss": 6.3217,
      "mean_token_accuracy": 0.13845824226737022,
      "num_tokens": 16356995.0,
      "step": 7555
    },
    {
      "entropy": 6.54440860748291,
      "epoch": 0.8090320509390551,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004944331614356311,
      "loss": 6.306,
      "mean_token_accuracy": 0.1379793740808964,
      "num_tokens": 16368841.0,
      "step": 7560
    },
    {
      "entropy": 6.501645946502686,
      "epoch": 0.8095671250468189,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004944247061120519,
      "loss": 6.2125,
      "mean_token_accuracy": 0.13962697982788086,
      "num_tokens": 16379295.0,
      "step": 7565
    },
    {
      "entropy": 6.505613708496094,
      "epoch": 0.8101021991545829,
      "grad_norm": 1.0,
      "learning_rate": 0.0004944162444526151,
      "loss": 6.2443,
      "mean_token_accuracy": 0.14370173290371896,
      "num_tokens": 16389512.0,
      "step": 7570
    },
    {
      "entropy": 6.411747169494629,
      "epoch": 0.8106372732623468,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004944077764575651,
      "loss": 6.1835,
      "mean_token_accuracy": 0.14841234236955642,
      "num_tokens": 16400720.0,
      "step": 7575
    },
    {
      "entropy": 6.469482135772705,
      "epoch": 0.8111723473701108,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004943993021271463,
      "loss": 6.2559,
      "mean_token_accuracy": 0.14133158847689628,
      "num_tokens": 16411278.0,
      "step": 7580
    },
    {
      "entropy": 6.536620950698852,
      "epoch": 0.8117074214778747,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004943908214616035,
      "loss": 6.3042,
      "mean_token_accuracy": 0.1375892236828804,
      "num_tokens": 16421665.0,
      "step": 7585
    },
    {
      "entropy": 6.560115146636963,
      "epoch": 0.8122424955856387,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004943823344611818,
      "loss": 6.3379,
      "mean_token_accuracy": 0.13549405410885812,
      "num_tokens": 16432943.0,
      "step": 7590
    },
    {
      "entropy": 6.537856245040894,
      "epoch": 0.8127775696934025,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004943738411261258,
      "loss": 6.2445,
      "mean_token_accuracy": 0.1387531988322735,
      "num_tokens": 16444506.0,
      "step": 7595
    },
    {
      "entropy": 6.476353454589844,
      "epoch": 0.8133126438011664,
      "grad_norm": 0.875,
      "learning_rate": 0.0004943653414566809,
      "loss": 6.2695,
      "mean_token_accuracy": 0.1386883355677128,
      "num_tokens": 16455403.0,
      "step": 7600
    },
    {
      "entropy": 6.587634515762329,
      "epoch": 0.8138477179089304,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004943568354530927,
      "loss": 6.3723,
      "mean_token_accuracy": 0.1309129886329174,
      "num_tokens": 16465616.0,
      "step": 7605
    },
    {
      "entropy": 6.601719188690185,
      "epoch": 0.8143827920166943,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004943483231156068,
      "loss": 6.3436,
      "mean_token_accuracy": 0.13243747279047965,
      "num_tokens": 16476850.0,
      "step": 7610
    },
    {
      "entropy": 6.530210351943969,
      "epoch": 0.8149178661244583,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004943398044444687,
      "loss": 6.2791,
      "mean_token_accuracy": 0.14042069390416145,
      "num_tokens": 16489118.0,
      "step": 7615
    },
    {
      "entropy": 6.517117214202881,
      "epoch": 0.8154529402322221,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004943312794399246,
      "loss": 6.2352,
      "mean_token_accuracy": 0.14114297330379486,
      "num_tokens": 16499336.0,
      "step": 7620
    },
    {
      "entropy": 6.511140775680542,
      "epoch": 0.8159880143399861,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004943227481022207,
      "loss": 6.3011,
      "mean_token_accuracy": 0.13706371039152146,
      "num_tokens": 16510158.0,
      "step": 7625
    },
    {
      "entropy": 6.4611626148223875,
      "epoch": 0.81652308844775,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004943142104316033,
      "loss": 6.2438,
      "mean_token_accuracy": 0.14061122685670852,
      "num_tokens": 16521080.0,
      "step": 7630
    },
    {
      "entropy": 6.512439632415772,
      "epoch": 0.8170581625555139,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004943056664283189,
      "loss": 6.2567,
      "mean_token_accuracy": 0.14463350176811218,
      "num_tokens": 16531072.0,
      "step": 7635
    },
    {
      "entropy": 6.527822256088257,
      "epoch": 0.8175932366632779,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000494297116092614,
      "loss": 6.354,
      "mean_token_accuracy": 0.13378974944353103,
      "num_tokens": 16542588.0,
      "step": 7640
    },
    {
      "entropy": 6.561722469329834,
      "epoch": 0.8181283107710418,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004942885594247359,
      "loss": 6.2217,
      "mean_token_accuracy": 0.13830168098211287,
      "num_tokens": 16553466.0,
      "step": 7645
    },
    {
      "entropy": 6.495503616333008,
      "epoch": 0.8186633848788057,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004942799964249314,
      "loss": 6.2541,
      "mean_token_accuracy": 0.14222589433193206,
      "num_tokens": 16564084.0,
      "step": 7650
    },
    {
      "entropy": 6.446701526641846,
      "epoch": 0.8191984589865696,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004942714270934479,
      "loss": 6.2261,
      "mean_token_accuracy": 0.1511824294924736,
      "num_tokens": 16575665.0,
      "step": 7655
    },
    {
      "entropy": 6.470211744308472,
      "epoch": 0.8197335330943336,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004942628514305326,
      "loss": 6.2496,
      "mean_token_accuracy": 0.13669308498501778,
      "num_tokens": 16586518.0,
      "step": 7660
    },
    {
      "entropy": 6.476080846786499,
      "epoch": 0.8202686072020975,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004942542694364333,
      "loss": 6.2679,
      "mean_token_accuracy": 0.14210453405976295,
      "num_tokens": 16597345.0,
      "step": 7665
    },
    {
      "entropy": 6.540896797180176,
      "epoch": 0.8208036813098614,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004942456811113978,
      "loss": 6.2887,
      "mean_token_accuracy": 0.14136139005422593,
      "num_tokens": 16609586.0,
      "step": 7670
    },
    {
      "entropy": 6.566203308105469,
      "epoch": 0.8213387554176254,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004942370864556741,
      "loss": 6.1551,
      "mean_token_accuracy": 0.15004199370741844,
      "num_tokens": 16620165.0,
      "step": 7675
    },
    {
      "entropy": 6.405955743789673,
      "epoch": 0.8218738295253892,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004942284854695104,
      "loss": 6.257,
      "mean_token_accuracy": 0.14405664727091788,
      "num_tokens": 16631200.0,
      "step": 7680
    },
    {
      "entropy": 6.488953304290772,
      "epoch": 0.8224089036331532,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004942198781531549,
      "loss": 6.2952,
      "mean_token_accuracy": 0.13799700438976287,
      "num_tokens": 16642553.0,
      "step": 7685
    },
    {
      "entropy": 6.558039665222168,
      "epoch": 0.8229439777409171,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004942112645068561,
      "loss": 6.3167,
      "mean_token_accuracy": 0.1354759894311428,
      "num_tokens": 16653431.0,
      "step": 7690
    },
    {
      "entropy": 6.507707500457764,
      "epoch": 0.8234790518486811,
      "grad_norm": 0.828125,
      "learning_rate": 0.000494202644530863,
      "loss": 6.2518,
      "mean_token_accuracy": 0.13894737660884857,
      "num_tokens": 16664880.0,
      "step": 7695
    },
    {
      "entropy": 6.454554510116577,
      "epoch": 0.824014125956445,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004941940182254244,
      "loss": 6.2644,
      "mean_token_accuracy": 0.14514543265104293,
      "num_tokens": 16676224.0,
      "step": 7700
    },
    {
      "entropy": 6.4975974559783936,
      "epoch": 0.8245492000642088,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004941853855907892,
      "loss": 6.3039,
      "mean_token_accuracy": 0.13956945165991783,
      "num_tokens": 16687732.0,
      "step": 7705
    },
    {
      "entropy": 6.534018516540527,
      "epoch": 0.8250842741719728,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004941767466272068,
      "loss": 6.2422,
      "mean_token_accuracy": 0.14467547386884688,
      "num_tokens": 16698257.0,
      "step": 7710
    },
    {
      "entropy": 6.5065450191497805,
      "epoch": 0.8256193482797367,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004941681013349267,
      "loss": 6.2111,
      "mean_token_accuracy": 0.14334181621670722,
      "num_tokens": 16708860.0,
      "step": 7715
    },
    {
      "entropy": 6.492323589324951,
      "epoch": 0.8261544223875007,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004941594497141985,
      "loss": 6.2213,
      "mean_token_accuracy": 0.14319886043667793,
      "num_tokens": 16719987.0,
      "step": 7720
    },
    {
      "entropy": 6.527457189559937,
      "epoch": 0.8266894964952646,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004941507917652718,
      "loss": 6.3089,
      "mean_token_accuracy": 0.13887255936861037,
      "num_tokens": 16730861.0,
      "step": 7725
    },
    {
      "entropy": 6.4563230037689205,
      "epoch": 0.8272245706030286,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004941421274883969,
      "loss": 6.1677,
      "mean_token_accuracy": 0.1534878171980381,
      "num_tokens": 16740968.0,
      "step": 7730
    },
    {
      "entropy": 6.486476707458496,
      "epoch": 0.8277596447107924,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000494133456883824,
      "loss": 6.2779,
      "mean_token_accuracy": 0.14263985231518744,
      "num_tokens": 16750700.0,
      "step": 7735
    },
    {
      "entropy": 6.537912130355835,
      "epoch": 0.8282947188185563,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004941247799518031,
      "loss": 6.2526,
      "mean_token_accuracy": 0.13385360166430474,
      "num_tokens": 16760719.0,
      "step": 7740
    },
    {
      "entropy": 6.474873352050781,
      "epoch": 0.8288297929263203,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004941160966925851,
      "loss": 6.1429,
      "mean_token_accuracy": 0.1410211443901062,
      "num_tokens": 16771331.0,
      "step": 7745
    },
    {
      "entropy": 6.378145027160644,
      "epoch": 0.8293648670340842,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004941074071064205,
      "loss": 6.2332,
      "mean_token_accuracy": 0.14552246630191804,
      "num_tokens": 16781306.0,
      "step": 7750
    },
    {
      "entropy": 6.519019985198975,
      "epoch": 0.8298999411418482,
      "grad_norm": 0.875,
      "learning_rate": 0.0004940987111935605,
      "loss": 6.266,
      "mean_token_accuracy": 0.14124012738466263,
      "num_tokens": 16793013.0,
      "step": 7755
    },
    {
      "entropy": 6.481528043746948,
      "epoch": 0.830435015249612,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004940900089542558,
      "loss": 6.2726,
      "mean_token_accuracy": 0.14356981664896012,
      "num_tokens": 16803987.0,
      "step": 7760
    },
    {
      "entropy": 6.468053913116455,
      "epoch": 0.830970089357376,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000494081300388758,
      "loss": 6.2754,
      "mean_token_accuracy": 0.1319303944706917,
      "num_tokens": 16815031.0,
      "step": 7765
    },
    {
      "entropy": 6.562944030761718,
      "epoch": 0.8315051634651399,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004940725854973184,
      "loss": 6.3137,
      "mean_token_accuracy": 0.13501427546143532,
      "num_tokens": 16826313.0,
      "step": 7770
    },
    {
      "entropy": 6.538980913162232,
      "epoch": 0.8320402375729038,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004940638642801886,
      "loss": 6.3082,
      "mean_token_accuracy": 0.13758592680096626,
      "num_tokens": 16838470.0,
      "step": 7775
    },
    {
      "entropy": 6.466607522964478,
      "epoch": 0.8325753116806678,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004940551367376205,
      "loss": 6.1996,
      "mean_token_accuracy": 0.14258078560233117,
      "num_tokens": 16849266.0,
      "step": 7780
    },
    {
      "entropy": 6.469249105453491,
      "epoch": 0.8331103857884317,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004940464028698662,
      "loss": 6.3108,
      "mean_token_accuracy": 0.14166692271828651,
      "num_tokens": 16860720.0,
      "step": 7785
    },
    {
      "entropy": 6.491447925567627,
      "epoch": 0.8336454598961957,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004940376626771779,
      "loss": 6.2683,
      "mean_token_accuracy": 0.13823330253362656,
      "num_tokens": 16872004.0,
      "step": 7790
    },
    {
      "entropy": 6.548871755599976,
      "epoch": 0.8341805340039595,
      "grad_norm": 0.875,
      "learning_rate": 0.0004940289161598076,
      "loss": 6.2981,
      "mean_token_accuracy": 0.13453821912407876,
      "num_tokens": 16882898.0,
      "step": 7795
    },
    {
      "entropy": 6.530791330337524,
      "epoch": 0.8347156081117235,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004940201633180084,
      "loss": 6.2278,
      "mean_token_accuracy": 0.1273862227797508,
      "num_tokens": 16894331.0,
      "step": 7800
    },
    {
      "entropy": 6.5691454887390135,
      "epoch": 0.8352506822194874,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004940114041520326,
      "loss": 6.3365,
      "mean_token_accuracy": 0.13795624747872354,
      "num_tokens": 16903656.0,
      "step": 7805
    },
    {
      "entropy": 6.443228816986084,
      "epoch": 0.8357857563272513,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004940026386621333,
      "loss": 6.2962,
      "mean_token_accuracy": 0.14036250188946725,
      "num_tokens": 16915878.0,
      "step": 7810
    },
    {
      "entropy": 6.5711266040802006,
      "epoch": 0.8363208304350153,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004939938668485636,
      "loss": 6.3413,
      "mean_token_accuracy": 0.13679892122745513,
      "num_tokens": 16926577.0,
      "step": 7815
    },
    {
      "entropy": 6.453451776504517,
      "epoch": 0.8368559045427791,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004939850887115768,
      "loss": 6.1853,
      "mean_token_accuracy": 0.14935341253876686,
      "num_tokens": 16938012.0,
      "step": 7820
    },
    {
      "entropy": 6.46830677986145,
      "epoch": 0.8373909786505431,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004939763042514263,
      "loss": 6.1962,
      "mean_token_accuracy": 0.13806031718850137,
      "num_tokens": 16949007.0,
      "step": 7825
    },
    {
      "entropy": 6.37170147895813,
      "epoch": 0.837926052758307,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004939675134683658,
      "loss": 6.1513,
      "mean_token_accuracy": 0.14931921511888505,
      "num_tokens": 16959425.0,
      "step": 7830
    },
    {
      "entropy": 6.509686040878296,
      "epoch": 0.838461126866071,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004939587163626491,
      "loss": 6.2729,
      "mean_token_accuracy": 0.13634693026542663,
      "num_tokens": 16970002.0,
      "step": 7835
    },
    {
      "entropy": 6.523667287826538,
      "epoch": 0.8389962009738349,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004939499129345302,
      "loss": 6.2899,
      "mean_token_accuracy": 0.13898420482873916,
      "num_tokens": 16980739.0,
      "step": 7840
    },
    {
      "entropy": 6.48391580581665,
      "epoch": 0.8395312750815987,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004939411031842633,
      "loss": 6.2508,
      "mean_token_accuracy": 0.14328086525201797,
      "num_tokens": 16991643.0,
      "step": 7845
    },
    {
      "entropy": 6.3353925228118895,
      "epoch": 0.8400663491893627,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000493932287112103,
      "loss": 6.1046,
      "mean_token_accuracy": 0.16147236973047258,
      "num_tokens": 17001975.0,
      "step": 7850
    },
    {
      "entropy": 6.545631694793701,
      "epoch": 0.8406014232971266,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004939234647183035,
      "loss": 6.262,
      "mean_token_accuracy": 0.13767837211489678,
      "num_tokens": 17012658.0,
      "step": 7855
    },
    {
      "entropy": 6.538120460510254,
      "epoch": 0.8411364974048906,
      "grad_norm": 0.875,
      "learning_rate": 0.0004939146360031197,
      "loss": 6.2868,
      "mean_token_accuracy": 0.13649813383817672,
      "num_tokens": 17023196.0,
      "step": 7860
    },
    {
      "entropy": 6.470953941345215,
      "epoch": 0.8416715715126545,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004939058009668067,
      "loss": 6.289,
      "mean_token_accuracy": 0.13703306019306183,
      "num_tokens": 17034589.0,
      "step": 7865
    },
    {
      "entropy": 6.5461596012115475,
      "epoch": 0.8422066456204185,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004938969596096194,
      "loss": 6.2628,
      "mean_token_accuracy": 0.1415180653333664,
      "num_tokens": 17046139.0,
      "step": 7870
    },
    {
      "entropy": 6.485792589187622,
      "epoch": 0.8427417197281823,
      "grad_norm": 0.890625,
      "learning_rate": 0.000493888111931813,
      "loss": 6.2758,
      "mean_token_accuracy": 0.13226577416062354,
      "num_tokens": 17056926.0,
      "step": 7875
    },
    {
      "entropy": 6.390592384338379,
      "epoch": 0.8432767938359462,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004938792579336433,
      "loss": 6.2191,
      "mean_token_accuracy": 0.14395023733377457,
      "num_tokens": 17067339.0,
      "step": 7880
    },
    {
      "entropy": 6.50981068611145,
      "epoch": 0.8438118679437102,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004938703976153657,
      "loss": 6.2371,
      "mean_token_accuracy": 0.13926032856106757,
      "num_tokens": 17078037.0,
      "step": 7885
    },
    {
      "entropy": 6.5204079151153564,
      "epoch": 0.8443469420514741,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004938615309772362,
      "loss": 6.2704,
      "mean_token_accuracy": 0.13414409533143043,
      "num_tokens": 17089069.0,
      "step": 7890
    },
    {
      "entropy": 6.477647829055786,
      "epoch": 0.8448820161592381,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004938526580195107,
      "loss": 6.2387,
      "mean_token_accuracy": 0.1453891061246395,
      "num_tokens": 17100006.0,
      "step": 7895
    },
    {
      "entropy": 6.520467901229859,
      "epoch": 0.845417090267002,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004938437787424454,
      "loss": 6.2639,
      "mean_token_accuracy": 0.14180212914943696,
      "num_tokens": 17111896.0,
      "step": 7900
    },
    {
      "entropy": 6.48339433670044,
      "epoch": 0.845952164374766,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004938348931462969,
      "loss": 6.2511,
      "mean_token_accuracy": 0.14949460998177527,
      "num_tokens": 17123202.0,
      "step": 7905
    },
    {
      "entropy": 6.4982506275177006,
      "epoch": 0.8464872384825298,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004938260012313215,
      "loss": 6.2239,
      "mean_token_accuracy": 0.14612067490816116,
      "num_tokens": 17134013.0,
      "step": 7910
    },
    {
      "entropy": 6.405726146697998,
      "epoch": 0.8470223125902937,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004938171029977761,
      "loss": 6.2254,
      "mean_token_accuracy": 0.14299570694565772,
      "num_tokens": 17144041.0,
      "step": 7915
    },
    {
      "entropy": 6.489075183868408,
      "epoch": 0.8475573866980577,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004938081984459176,
      "loss": 6.2162,
      "mean_token_accuracy": 0.1454182267189026,
      "num_tokens": 17155334.0,
      "step": 7920
    },
    {
      "entropy": 6.491747045516968,
      "epoch": 0.8480924608058216,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000493799287576003,
      "loss": 6.1871,
      "mean_token_accuracy": 0.1432422623038292,
      "num_tokens": 17166243.0,
      "step": 7925
    },
    {
      "entropy": 6.469628572463989,
      "epoch": 0.8486275349135856,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004937903703882898,
      "loss": 6.2404,
      "mean_token_accuracy": 0.13866124749183656,
      "num_tokens": 17177709.0,
      "step": 7930
    },
    {
      "entropy": 6.488557434082031,
      "epoch": 0.8491626090213494,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004937814468830353,
      "loss": 6.2176,
      "mean_token_accuracy": 0.14349672347307205,
      "num_tokens": 17188275.0,
      "step": 7935
    },
    {
      "entropy": 6.531327533721924,
      "epoch": 0.8496976831291134,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004937725170604975,
      "loss": 6.3392,
      "mean_token_accuracy": 0.1405933268368244,
      "num_tokens": 17199527.0,
      "step": 7940
    },
    {
      "entropy": 6.522189092636109,
      "epoch": 0.8502327572368773,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004937635809209339,
      "loss": 6.2527,
      "mean_token_accuracy": 0.13896754756569862,
      "num_tokens": 17209500.0,
      "step": 7945
    },
    {
      "entropy": 6.462730503082275,
      "epoch": 0.8507678313446412,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004937546384646027,
      "loss": 6.2133,
      "mean_token_accuracy": 0.13598527908325195,
      "num_tokens": 17219589.0,
      "step": 7950
    },
    {
      "entropy": 6.540196895599365,
      "epoch": 0.8513029054524052,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004937456896917619,
      "loss": 6.2387,
      "mean_token_accuracy": 0.141167963296175,
      "num_tokens": 17230508.0,
      "step": 7955
    },
    {
      "entropy": 6.443740463256836,
      "epoch": 0.851837979560169,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004937367346026702,
      "loss": 6.2406,
      "mean_token_accuracy": 0.140987516194582,
      "num_tokens": 17240886.0,
      "step": 7960
    },
    {
      "entropy": 6.439098739624024,
      "epoch": 0.852373053667933,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004937277731975858,
      "loss": 6.2671,
      "mean_token_accuracy": 0.14626324623823167,
      "num_tokens": 17252247.0,
      "step": 7965
    },
    {
      "entropy": 6.5244653701782225,
      "epoch": 0.8529081277756969,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004937188054767679,
      "loss": 6.235,
      "mean_token_accuracy": 0.1472860038280487,
      "num_tokens": 17262845.0,
      "step": 7970
    },
    {
      "entropy": 6.489797067642212,
      "epoch": 0.8534432018834609,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004937098314404751,
      "loss": 6.2493,
      "mean_token_accuracy": 0.14316761195659639,
      "num_tokens": 17274247.0,
      "step": 7975
    },
    {
      "entropy": 6.52520899772644,
      "epoch": 0.8539782759912248,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004937008510889668,
      "loss": 6.2469,
      "mean_token_accuracy": 0.14510944336652756,
      "num_tokens": 17285471.0,
      "step": 7980
    },
    {
      "entropy": 6.4983877658844,
      "epoch": 0.8545133500989887,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004936918644225021,
      "loss": 6.2899,
      "mean_token_accuracy": 0.1482240729033947,
      "num_tokens": 17296243.0,
      "step": 7985
    },
    {
      "entropy": 6.4581718921661375,
      "epoch": 0.8550484242067526,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004936828714413405,
      "loss": 6.2332,
      "mean_token_accuracy": 0.14962877556681634,
      "num_tokens": 17307301.0,
      "step": 7990
    },
    {
      "entropy": 6.412732887268066,
      "epoch": 0.8555834983145165,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004936738721457417,
      "loss": 6.1668,
      "mean_token_accuracy": 0.1510644055902958,
      "num_tokens": 17318550.0,
      "step": 7995
    },
    {
      "entropy": 6.552431535720825,
      "epoch": 0.8561185724222805,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004936648665359657,
      "loss": 6.2418,
      "mean_token_accuracy": 0.14080545157194138,
      "num_tokens": 17328577.0,
      "step": 8000
    },
    {
      "entropy": 6.522714710235595,
      "epoch": 0.8566536465300444,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004936558546122723,
      "loss": 6.2965,
      "mean_token_accuracy": 0.13250770717859267,
      "num_tokens": 17341013.0,
      "step": 8005
    },
    {
      "entropy": 6.4302043437957765,
      "epoch": 0.8571887206378084,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004936468363749217,
      "loss": 6.2423,
      "mean_token_accuracy": 0.13588053211569787,
      "num_tokens": 17351956.0,
      "step": 8010
    },
    {
      "entropy": 6.5284051418304445,
      "epoch": 0.8577237947455723,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004936378118241746,
      "loss": 6.2164,
      "mean_token_accuracy": 0.1383969120681286,
      "num_tokens": 17363465.0,
      "step": 8015
    },
    {
      "entropy": 6.487103509902954,
      "epoch": 0.8582588688533361,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004936287809602913,
      "loss": 6.2624,
      "mean_token_accuracy": 0.14479685723781585,
      "num_tokens": 17374458.0,
      "step": 8020
    },
    {
      "entropy": 6.482282447814941,
      "epoch": 0.8587939429611001,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004936197437835329,
      "loss": 6.2722,
      "mean_token_accuracy": 0.1384648010134697,
      "num_tokens": 17385492.0,
      "step": 8025
    },
    {
      "entropy": 6.566893720626831,
      "epoch": 0.859329017068864,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004936107002941599,
      "loss": 6.2234,
      "mean_token_accuracy": 0.14262498915195465,
      "num_tokens": 17395640.0,
      "step": 8030
    },
    {
      "entropy": 6.4909087181091305,
      "epoch": 0.859864091176628,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004936016504924337,
      "loss": 6.255,
      "mean_token_accuracy": 0.13882532864809036,
      "num_tokens": 17405211.0,
      "step": 8035
    },
    {
      "entropy": 6.456467247009277,
      "epoch": 0.8603991652843919,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004935925943786155,
      "loss": 6.2149,
      "mean_token_accuracy": 0.13942714259028435,
      "num_tokens": 17415421.0,
      "step": 8040
    },
    {
      "entropy": 6.501566553115845,
      "epoch": 0.8609342393921559,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004935835319529669,
      "loss": 6.2827,
      "mean_token_accuracy": 0.13601001352071762,
      "num_tokens": 17426040.0,
      "step": 8045
    },
    {
      "entropy": 6.469284915924073,
      "epoch": 0.8614693134999197,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004935744632157496,
      "loss": 6.2635,
      "mean_token_accuracy": 0.14099914506077765,
      "num_tokens": 17436305.0,
      "step": 8050
    },
    {
      "entropy": 6.517049407958984,
      "epoch": 0.8620043876076836,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004935653881672253,
      "loss": 6.3138,
      "mean_token_accuracy": 0.13985636681318284,
      "num_tokens": 17446786.0,
      "step": 8055
    },
    {
      "entropy": 6.504854774475097,
      "epoch": 0.8625394617154476,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004935563068076561,
      "loss": 6.1951,
      "mean_token_accuracy": 0.14708731323480606,
      "num_tokens": 17456613.0,
      "step": 8060
    },
    {
      "entropy": 6.421092748641968,
      "epoch": 0.8630745358232115,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004935472191373044,
      "loss": 6.1528,
      "mean_token_accuracy": 0.14706217721104622,
      "num_tokens": 17467281.0,
      "step": 8065
    },
    {
      "entropy": 6.477051305770874,
      "epoch": 0.8636096099309755,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004935381251564324,
      "loss": 6.2577,
      "mean_token_accuracy": 0.14080820828676224,
      "num_tokens": 17478271.0,
      "step": 8070
    },
    {
      "entropy": 6.490134763717651,
      "epoch": 0.8641446840387393,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004935290248653027,
      "loss": 6.2474,
      "mean_token_accuracy": 0.14129171371459961,
      "num_tokens": 17489519.0,
      "step": 8075
    },
    {
      "entropy": 6.497061729431152,
      "epoch": 0.8646797581465033,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004935199182641781,
      "loss": 6.2219,
      "mean_token_accuracy": 0.14453355669975282,
      "num_tokens": 17500076.0,
      "step": 8080
    },
    {
      "entropy": 6.504789543151856,
      "epoch": 0.8652148322542672,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004935108053533216,
      "loss": 6.2218,
      "mean_token_accuracy": 0.13518178313970566,
      "num_tokens": 17509898.0,
      "step": 8085
    },
    {
      "entropy": 6.4625955581665036,
      "epoch": 0.8657499063620312,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004935016861329964,
      "loss": 6.1768,
      "mean_token_accuracy": 0.1432091474533081,
      "num_tokens": 17519890.0,
      "step": 8090
    },
    {
      "entropy": 6.405915403366089,
      "epoch": 0.8662849804697951,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004934925606034656,
      "loss": 6.2133,
      "mean_token_accuracy": 0.13946884796023368,
      "num_tokens": 17530927.0,
      "step": 8095
    },
    {
      "entropy": 6.526646900177002,
      "epoch": 0.866820054577559,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004934834287649928,
      "loss": 6.2846,
      "mean_token_accuracy": 0.13718777671456336,
      "num_tokens": 17541174.0,
      "step": 8100
    },
    {
      "entropy": 6.503631114959717,
      "epoch": 0.8673551286853229,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004934742906178417,
      "loss": 6.2713,
      "mean_token_accuracy": 0.13760240077972413,
      "num_tokens": 17552460.0,
      "step": 8105
    },
    {
      "entropy": 6.537569236755371,
      "epoch": 0.8678902027930868,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004934651461622762,
      "loss": 6.2636,
      "mean_token_accuracy": 0.14259424656629563,
      "num_tokens": 17563908.0,
      "step": 8110
    },
    {
      "entropy": 6.479424047470093,
      "epoch": 0.8684252769008508,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004934559953985603,
      "loss": 6.2654,
      "mean_token_accuracy": 0.14494671747088433,
      "num_tokens": 17574314.0,
      "step": 8115
    },
    {
      "entropy": 6.485303020477295,
      "epoch": 0.8689603510086147,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004934468383269582,
      "loss": 6.2428,
      "mean_token_accuracy": 0.1445814125239849,
      "num_tokens": 17583719.0,
      "step": 8120
    },
    {
      "entropy": 6.394630765914917,
      "epoch": 0.8694954251163787,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004934376749477344,
      "loss": 6.1925,
      "mean_token_accuracy": 0.1466481104493141,
      "num_tokens": 17594522.0,
      "step": 8125
    },
    {
      "entropy": 6.467332410812378,
      "epoch": 0.8700304992241426,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004934285052611533,
      "loss": 6.1596,
      "mean_token_accuracy": 0.1527232013642788,
      "num_tokens": 17604632.0,
      "step": 8130
    },
    {
      "entropy": 6.518517112731933,
      "epoch": 0.8705655733319064,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00049341932926748,
      "loss": 6.3044,
      "mean_token_accuracy": 0.13584948033094407,
      "num_tokens": 17615459.0,
      "step": 8135
    },
    {
      "entropy": 6.490569067001343,
      "epoch": 0.8711006474396704,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004934101469669791,
      "loss": 6.2553,
      "mean_token_accuracy": 0.1415888287127018,
      "num_tokens": 17625800.0,
      "step": 8140
    },
    {
      "entropy": 6.405483341217041,
      "epoch": 0.8716357215474343,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004934009583599159,
      "loss": 6.0786,
      "mean_token_accuracy": 0.15376727283000946,
      "num_tokens": 17637264.0,
      "step": 8145
    },
    {
      "entropy": 6.439616012573242,
      "epoch": 0.8721707956551983,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004933917634465558,
      "loss": 6.1298,
      "mean_token_accuracy": 0.14518497735261918,
      "num_tokens": 17648256.0,
      "step": 8150
    },
    {
      "entropy": 6.414270544052124,
      "epoch": 0.8727058697629622,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004933825622271642,
      "loss": 6.2603,
      "mean_token_accuracy": 0.13708942756056786,
      "num_tokens": 17659096.0,
      "step": 8155
    },
    {
      "entropy": 6.515870523452759,
      "epoch": 0.8732409438707261,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004933733547020069,
      "loss": 6.2379,
      "mean_token_accuracy": 0.14280832558870316,
      "num_tokens": 17669808.0,
      "step": 8160
    },
    {
      "entropy": 6.526236820220947,
      "epoch": 0.87377601797849,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004933641408713495,
      "loss": 6.2036,
      "mean_token_accuracy": 0.1451032914221287,
      "num_tokens": 17680375.0,
      "step": 8165
    },
    {
      "entropy": 6.3738494396209715,
      "epoch": 0.8743110920862539,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004933549207354582,
      "loss": 6.1588,
      "mean_token_accuracy": 0.1451977863907814,
      "num_tokens": 17690780.0,
      "step": 8170
    },
    {
      "entropy": 6.446805238723755,
      "epoch": 0.8748461661940179,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004933456942945994,
      "loss": 6.2716,
      "mean_token_accuracy": 0.14626665711402892,
      "num_tokens": 17702170.0,
      "step": 8175
    },
    {
      "entropy": 6.44735198020935,
      "epoch": 0.8753812403017818,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004933364615490393,
      "loss": 6.2105,
      "mean_token_accuracy": 0.14683770388364792,
      "num_tokens": 17712631.0,
      "step": 8180
    },
    {
      "entropy": 6.54788179397583,
      "epoch": 0.8759163144095458,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004933272224990445,
      "loss": 6.2949,
      "mean_token_accuracy": 0.1356400080025196,
      "num_tokens": 17723692.0,
      "step": 8185
    },
    {
      "entropy": 6.496940326690674,
      "epoch": 0.8764513885173096,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004933179771448819,
      "loss": 6.1845,
      "mean_token_accuracy": 0.1512119174003601,
      "num_tokens": 17735264.0,
      "step": 8190
    },
    {
      "entropy": 6.366297674179077,
      "epoch": 0.8769864626250736,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004933087254868183,
      "loss": 6.0895,
      "mean_token_accuracy": 0.15264559388160706,
      "num_tokens": 17747060.0,
      "step": 8195
    },
    {
      "entropy": 6.3891722679138185,
      "epoch": 0.8775215367328375,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000493299467525121,
      "loss": 6.2188,
      "mean_token_accuracy": 0.1389477953314781,
      "num_tokens": 17758159.0,
      "step": 8200
    },
    {
      "entropy": 6.500685834884644,
      "epoch": 0.8780566108406014,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004932902032600573,
      "loss": 6.2191,
      "mean_token_accuracy": 0.14177136272192,
      "num_tokens": 17768444.0,
      "step": 8205
    },
    {
      "entropy": 6.461921882629395,
      "epoch": 0.8785916849483654,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004932809326918945,
      "loss": 6.1122,
      "mean_token_accuracy": 0.15678573548793792,
      "num_tokens": 17779303.0,
      "step": 8210
    },
    {
      "entropy": 6.487242603302002,
      "epoch": 0.8791267590561292,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004932716558209005,
      "loss": 6.3052,
      "mean_token_accuracy": 0.1367947705090046,
      "num_tokens": 17789853.0,
      "step": 8215
    },
    {
      "entropy": 6.536150217056274,
      "epoch": 0.8796618331638932,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004932623726473432,
      "loss": 6.1935,
      "mean_token_accuracy": 0.15173025727272033,
      "num_tokens": 17799232.0,
      "step": 8220
    },
    {
      "entropy": 6.3940956592559814,
      "epoch": 0.8801969072716571,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004932530831714904,
      "loss": 6.187,
      "mean_token_accuracy": 0.14445312023162843,
      "num_tokens": 17808870.0,
      "step": 8225
    },
    {
      "entropy": 6.3977783203125,
      "epoch": 0.8807319813794211,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004932437873936107,
      "loss": 6.2626,
      "mean_token_accuracy": 0.13755191415548323,
      "num_tokens": 17819394.0,
      "step": 8230
    },
    {
      "entropy": 6.550758218765258,
      "epoch": 0.881267055487185,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004932344853139721,
      "loss": 6.2453,
      "mean_token_accuracy": 0.13946715593338013,
      "num_tokens": 17830220.0,
      "step": 8235
    },
    {
      "entropy": 6.448937892913818,
      "epoch": 0.8818021295949489,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004932251769328436,
      "loss": 6.1963,
      "mean_token_accuracy": 0.1393178842961788,
      "num_tokens": 17841971.0,
      "step": 8240
    },
    {
      "entropy": 6.448181200027466,
      "epoch": 0.8823372037027128,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004932158622504937,
      "loss": 6.1941,
      "mean_token_accuracy": 0.14694615676999093,
      "num_tokens": 17851890.0,
      "step": 8245
    },
    {
      "entropy": 6.427869701385498,
      "epoch": 0.8828722778104767,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004932065412671915,
      "loss": 6.1862,
      "mean_token_accuracy": 0.15051203742623329,
      "num_tokens": 17862855.0,
      "step": 8250
    },
    {
      "entropy": 6.547014617919922,
      "epoch": 0.8834073519182407,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004931972139832058,
      "loss": 6.1919,
      "mean_token_accuracy": 0.13899358212947846,
      "num_tokens": 17873550.0,
      "step": 8255
    },
    {
      "entropy": 6.462202167510986,
      "epoch": 0.8839424260260046,
      "grad_norm": 0.875,
      "learning_rate": 0.0004931878803988066,
      "loss": 6.221,
      "mean_token_accuracy": 0.14448966532945634,
      "num_tokens": 17884067.0,
      "step": 8260
    },
    {
      "entropy": 6.427721071243286,
      "epoch": 0.8844775001337686,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004931785405142627,
      "loss": 6.2237,
      "mean_token_accuracy": 0.14615930393338203,
      "num_tokens": 17894587.0,
      "step": 8265
    },
    {
      "entropy": 6.475208044052124,
      "epoch": 0.8850125742415325,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004931691943298443,
      "loss": 6.2346,
      "mean_token_accuracy": 0.13537004292011262,
      "num_tokens": 17905297.0,
      "step": 8270
    },
    {
      "entropy": 6.414251136779785,
      "epoch": 0.8855476483492963,
      "grad_norm": 0.84375,
      "learning_rate": 0.000493159841845821,
      "loss": 6.1351,
      "mean_token_accuracy": 0.1493850603699684,
      "num_tokens": 17916225.0,
      "step": 8275
    },
    {
      "entropy": 6.518371868133545,
      "epoch": 0.8860827224570603,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004931504830624629,
      "loss": 6.1896,
      "mean_token_accuracy": 0.1398513376712799,
      "num_tokens": 17925795.0,
      "step": 8280
    },
    {
      "entropy": 6.45190749168396,
      "epoch": 0.8866177965648242,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004931411179800403,
      "loss": 6.1978,
      "mean_token_accuracy": 0.13878842890262605,
      "num_tokens": 17936134.0,
      "step": 8285
    },
    {
      "entropy": 6.476259756088257,
      "epoch": 0.8871528706725882,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004931317465988235,
      "loss": 6.2749,
      "mean_token_accuracy": 0.13695047050714493,
      "num_tokens": 17946393.0,
      "step": 8290
    },
    {
      "entropy": 6.508022928237915,
      "epoch": 0.8876879447803521,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004931223689190832,
      "loss": 6.2731,
      "mean_token_accuracy": 0.1373719498515129,
      "num_tokens": 17957527.0,
      "step": 8295
    },
    {
      "entropy": 6.509348917007446,
      "epoch": 0.8882230188881161,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004931129849410903,
      "loss": 6.2532,
      "mean_token_accuracy": 0.13987435549497604,
      "num_tokens": 17968390.0,
      "step": 8300
    },
    {
      "entropy": 6.516298246383667,
      "epoch": 0.8887580929958799,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004931035946651154,
      "loss": 6.1875,
      "mean_token_accuracy": 0.15575498342514038,
      "num_tokens": 17978667.0,
      "step": 8305
    },
    {
      "entropy": 6.442233037948609,
      "epoch": 0.8892931671036438,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004930941980914299,
      "loss": 6.2262,
      "mean_token_accuracy": 0.13767518401145934,
      "num_tokens": 17989008.0,
      "step": 8310
    },
    {
      "entropy": 6.437238550186157,
      "epoch": 0.8898282412114078,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004930847952203052,
      "loss": 6.2042,
      "mean_token_accuracy": 0.13336346596479415,
      "num_tokens": 18000199.0,
      "step": 8315
    },
    {
      "entropy": 6.496860790252685,
      "epoch": 0.8903633153191717,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004930753860520126,
      "loss": 6.1422,
      "mean_token_accuracy": 0.1425130732357502,
      "num_tokens": 18010941.0,
      "step": 8320
    },
    {
      "entropy": 6.5454202651977536,
      "epoch": 0.8908983894269357,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004930659705868237,
      "loss": 6.2558,
      "mean_token_accuracy": 0.1420603796839714,
      "num_tokens": 18021519.0,
      "step": 8325
    },
    {
      "entropy": 6.47792158126831,
      "epoch": 0.8914334635346995,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004930565488250108,
      "loss": 6.204,
      "mean_token_accuracy": 0.14179358258843422,
      "num_tokens": 18032773.0,
      "step": 8330
    },
    {
      "entropy": 6.393397092819214,
      "epoch": 0.8919685376424635,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004930471207668456,
      "loss": 6.1737,
      "mean_token_accuracy": 0.15024345815181733,
      "num_tokens": 18043250.0,
      "step": 8335
    },
    {
      "entropy": 6.430734014511108,
      "epoch": 0.8925036117502274,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004930376864126005,
      "loss": 6.1775,
      "mean_token_accuracy": 0.14761077091097832,
      "num_tokens": 18055165.0,
      "step": 8340
    },
    {
      "entropy": 6.431783246994018,
      "epoch": 0.8930386858579913,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004930282457625476,
      "loss": 6.1319,
      "mean_token_accuracy": 0.15056303143501282,
      "num_tokens": 18065296.0,
      "step": 8345
    },
    {
      "entropy": 6.412946891784668,
      "epoch": 0.8935737599657553,
      "grad_norm": 0.890625,
      "learning_rate": 0.00049301879881696,
      "loss": 6.1453,
      "mean_token_accuracy": 0.1431378237903118,
      "num_tokens": 18075536.0,
      "step": 8350
    },
    {
      "entropy": 6.492121362686158,
      "epoch": 0.8941088340735192,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004930093455761101,
      "loss": 6.2689,
      "mean_token_accuracy": 0.14712657630443574,
      "num_tokens": 18086253.0,
      "step": 8355
    },
    {
      "entropy": 6.447110652923584,
      "epoch": 0.8946439081812831,
      "grad_norm": 0.890625,
      "learning_rate": 0.000492999886040271,
      "loss": 6.2434,
      "mean_token_accuracy": 0.14176914021372794,
      "num_tokens": 18097335.0,
      "step": 8360
    },
    {
      "entropy": 6.487428617477417,
      "epoch": 0.895178982289047,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004929904202097159,
      "loss": 6.173,
      "mean_token_accuracy": 0.14728611558675767,
      "num_tokens": 18107293.0,
      "step": 8365
    },
    {
      "entropy": 6.43171329498291,
      "epoch": 0.895714056396811,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004929809480847179,
      "loss": 6.3191,
      "mean_token_accuracy": 0.12872769683599472,
      "num_tokens": 18117705.0,
      "step": 8370
    },
    {
      "entropy": 6.522384452819824,
      "epoch": 0.8962491305045749,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004929714696655507,
      "loss": 6.2111,
      "mean_token_accuracy": 0.13840262442827225,
      "num_tokens": 18129003.0,
      "step": 8375
    },
    {
      "entropy": 6.445065498352051,
      "epoch": 0.8967842046123388,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000492961984952488,
      "loss": 6.1513,
      "mean_token_accuracy": 0.15770164132118225,
      "num_tokens": 18140843.0,
      "step": 8380
    },
    {
      "entropy": 6.426892185211182,
      "epoch": 0.8973192787201028,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004929524939458037,
      "loss": 6.2532,
      "mean_token_accuracy": 0.14436969608068467,
      "num_tokens": 18151476.0,
      "step": 8385
    },
    {
      "entropy": 6.453249740600586,
      "epoch": 0.8978543528278666,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004929429966457716,
      "loss": 6.2778,
      "mean_token_accuracy": 0.1357191488146782,
      "num_tokens": 18161304.0,
      "step": 8390
    },
    {
      "entropy": 6.505843734741211,
      "epoch": 0.8983894269356306,
      "grad_norm": 0.875,
      "learning_rate": 0.0004929334930526662,
      "loss": 6.2021,
      "mean_token_accuracy": 0.1475615009665489,
      "num_tokens": 18173696.0,
      "step": 8395
    },
    {
      "entropy": 6.4902702331542965,
      "epoch": 0.8989245010433945,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004929239831667617,
      "loss": 6.2293,
      "mean_token_accuracy": 0.13744482025504112,
      "num_tokens": 18186024.0,
      "step": 8400
    },
    {
      "entropy": 6.477586174011231,
      "epoch": 0.8994595751511585,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004929144669883329,
      "loss": 6.2418,
      "mean_token_accuracy": 0.1429603010416031,
      "num_tokens": 18196976.0,
      "step": 8405
    },
    {
      "entropy": 6.489749479293823,
      "epoch": 0.8999946492589224,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004929049445176545,
      "loss": 6.2311,
      "mean_token_accuracy": 0.13951153978705405,
      "num_tokens": 18208086.0,
      "step": 8410
    },
    {
      "entropy": 6.421720695495606,
      "epoch": 0.9005297233666862,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004928954157550013,
      "loss": 6.1864,
      "mean_token_accuracy": 0.15251387804746627,
      "num_tokens": 18218468.0,
      "step": 8415
    },
    {
      "entropy": 6.506771230697632,
      "epoch": 0.9010647974744502,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004928858807006488,
      "loss": 6.2756,
      "mean_token_accuracy": 0.1449819788336754,
      "num_tokens": 18230489.0,
      "step": 8420
    },
    {
      "entropy": 6.478936433792114,
      "epoch": 0.9015998715822141,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004928763393548721,
      "loss": 6.2273,
      "mean_token_accuracy": 0.14151315689086913,
      "num_tokens": 18239968.0,
      "step": 8425
    },
    {
      "entropy": 6.497046327590942,
      "epoch": 0.9021349456899781,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004928667917179465,
      "loss": 6.248,
      "mean_token_accuracy": 0.14045359268784524,
      "num_tokens": 18250902.0,
      "step": 8430
    },
    {
      "entropy": 6.544751739501953,
      "epoch": 0.902670019797742,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004928572377901481,
      "loss": 6.2649,
      "mean_token_accuracy": 0.14468383342027663,
      "num_tokens": 18262094.0,
      "step": 8435
    },
    {
      "entropy": 6.498642635345459,
      "epoch": 0.903205093905506,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004928476775717524,
      "loss": 6.2446,
      "mean_token_accuracy": 0.14410303086042403,
      "num_tokens": 18272765.0,
      "step": 8440
    },
    {
      "entropy": 6.430245733261108,
      "epoch": 0.9037401680132698,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004928381110630358,
      "loss": 6.1166,
      "mean_token_accuracy": 0.15524550825357436,
      "num_tokens": 18284570.0,
      "step": 8445
    },
    {
      "entropy": 6.579597139358521,
      "epoch": 0.9042752421210337,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004928285382642742,
      "loss": 6.2464,
      "mean_token_accuracy": 0.13878655210137367,
      "num_tokens": 18295170.0,
      "step": 8450
    },
    {
      "entropy": 6.408507251739502,
      "epoch": 0.9048103162287977,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004928189591757443,
      "loss": 6.1957,
      "mean_token_accuracy": 0.14184481278061867,
      "num_tokens": 18306388.0,
      "step": 8455
    },
    {
      "entropy": 6.442728662490845,
      "epoch": 0.9053453903365616,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004928093737977225,
      "loss": 6.1994,
      "mean_token_accuracy": 0.14463388174772263,
      "num_tokens": 18316363.0,
      "step": 8460
    },
    {
      "entropy": 6.353628349304199,
      "epoch": 0.9058804644443256,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004927997821304856,
      "loss": 6.1422,
      "mean_token_accuracy": 0.1478566661477089,
      "num_tokens": 18327397.0,
      "step": 8465
    },
    {
      "entropy": 6.482725620269775,
      "epoch": 0.9064155385520895,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004927901841743107,
      "loss": 6.2601,
      "mean_token_accuracy": 0.13910072967410086,
      "num_tokens": 18338909.0,
      "step": 8470
    },
    {
      "entropy": 6.5338568687438965,
      "epoch": 0.9069506126598534,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004927805799294747,
      "loss": 6.2296,
      "mean_token_accuracy": 0.14250923395156861,
      "num_tokens": 18350110.0,
      "step": 8475
    },
    {
      "entropy": 6.483696174621582,
      "epoch": 0.9074856867676173,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004927709693962551,
      "loss": 6.2818,
      "mean_token_accuracy": 0.13232277110219,
      "num_tokens": 18361590.0,
      "step": 8480
    },
    {
      "entropy": 6.44201135635376,
      "epoch": 0.9080207608753812,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004927613525749295,
      "loss": 6.2238,
      "mean_token_accuracy": 0.14254011660814286,
      "num_tokens": 18372602.0,
      "step": 8485
    },
    {
      "entropy": 6.513248682022095,
      "epoch": 0.9085558349831452,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004927517294657753,
      "loss": 6.2703,
      "mean_token_accuracy": 0.13598200902342797,
      "num_tokens": 18384653.0,
      "step": 8490
    },
    {
      "entropy": 6.564611005783081,
      "epoch": 0.9090909090909091,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004927421000690705,
      "loss": 6.2256,
      "mean_token_accuracy": 0.1415557749569416,
      "num_tokens": 18395379.0,
      "step": 8495
    },
    {
      "entropy": 6.481884431838989,
      "epoch": 0.909625983198673,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004927324643850933,
      "loss": 6.1985,
      "mean_token_accuracy": 0.14312231540679932,
      "num_tokens": 18405915.0,
      "step": 8500
    },
    {
      "entropy": 6.36347222328186,
      "epoch": 0.9101610573064369,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004927228224141217,
      "loss": 6.0887,
      "mean_token_accuracy": 0.1629916936159134,
      "num_tokens": 18416656.0,
      "step": 8505
    },
    {
      "entropy": 6.426639747619629,
      "epoch": 0.9106961314142009,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004927131741564342,
      "loss": 6.228,
      "mean_token_accuracy": 0.13803585097193719,
      "num_tokens": 18427141.0,
      "step": 8510
    },
    {
      "entropy": 6.490043210983276,
      "epoch": 0.9112312055219648,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004927035196123094,
      "loss": 6.2574,
      "mean_token_accuracy": 0.14823196157813073,
      "num_tokens": 18437135.0,
      "step": 8515
    },
    {
      "entropy": 6.606572437286377,
      "epoch": 0.9117662796297287,
      "grad_norm": 0.890625,
      "learning_rate": 0.000492693858782026,
      "loss": 6.3205,
      "mean_token_accuracy": 0.13474073708057405,
      "num_tokens": 18448567.0,
      "step": 8520
    },
    {
      "entropy": 6.561576700210571,
      "epoch": 0.9123013537374927,
      "grad_norm": 1.0,
      "learning_rate": 0.000492684191665863,
      "loss": 6.2943,
      "mean_token_accuracy": 0.13457489982247353,
      "num_tokens": 18459385.0,
      "step": 8525
    },
    {
      "entropy": 6.447676277160644,
      "epoch": 0.9128364278452565,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004926745182640996,
      "loss": 6.2272,
      "mean_token_accuracy": 0.14392890483140947,
      "num_tokens": 18471376.0,
      "step": 8530
    },
    {
      "entropy": 6.511551856994629,
      "epoch": 0.9133715019530205,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004926648385770152,
      "loss": 6.3065,
      "mean_token_accuracy": 0.13559125885367393,
      "num_tokens": 18481950.0,
      "step": 8535
    },
    {
      "entropy": 6.539721441268921,
      "epoch": 0.9139065760607844,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004926551526048891,
      "loss": 6.2217,
      "mean_token_accuracy": 0.14225187674164771,
      "num_tokens": 18492436.0,
      "step": 8540
    },
    {
      "entropy": 6.434160566329956,
      "epoch": 0.9144416501685484,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004926454603480009,
      "loss": 6.0614,
      "mean_token_accuracy": 0.1480466492474079,
      "num_tokens": 18504279.0,
      "step": 8545
    },
    {
      "entropy": 6.473003673553467,
      "epoch": 0.9149767242763123,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004926357618066307,
      "loss": 6.1957,
      "mean_token_accuracy": 0.14428540021181108,
      "num_tokens": 18515262.0,
      "step": 8550
    },
    {
      "entropy": 6.474255132675171,
      "epoch": 0.9155117983840761,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004926260569810586,
      "loss": 6.2633,
      "mean_token_accuracy": 0.14167677313089372,
      "num_tokens": 18525176.0,
      "step": 8555
    },
    {
      "entropy": 6.412099933624267,
      "epoch": 0.9160468724918401,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004926163458715645,
      "loss": 6.2397,
      "mean_token_accuracy": 0.13938599228858947,
      "num_tokens": 18536674.0,
      "step": 8560
    },
    {
      "entropy": 6.487487363815307,
      "epoch": 0.916581946599604,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004926066284784292,
      "loss": 6.2074,
      "mean_token_accuracy": 0.14399363100528717,
      "num_tokens": 18547398.0,
      "step": 8565
    },
    {
      "entropy": 6.50559434890747,
      "epoch": 0.917117020707368,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004925969048019329,
      "loss": 6.25,
      "mean_token_accuracy": 0.14789487645030022,
      "num_tokens": 18558838.0,
      "step": 8570
    },
    {
      "entropy": 6.444234085083008,
      "epoch": 0.9176520948151319,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004925871748423566,
      "loss": 6.1958,
      "mean_token_accuracy": 0.14261153042316438,
      "num_tokens": 18569881.0,
      "step": 8575
    },
    {
      "entropy": 6.445477390289307,
      "epoch": 0.9181871689228959,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004925774385999812,
      "loss": 6.0995,
      "mean_token_accuracy": 0.1520858034491539,
      "num_tokens": 18579556.0,
      "step": 8580
    },
    {
      "entropy": 6.36273717880249,
      "epoch": 0.9187222430306597,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004925676960750878,
      "loss": 6.1261,
      "mean_token_accuracy": 0.1515171892940998,
      "num_tokens": 18590580.0,
      "step": 8585
    },
    {
      "entropy": 6.5159039974212645,
      "epoch": 0.9192573171384236,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004925579472679578,
      "loss": 6.2396,
      "mean_token_accuracy": 0.14403093606233597,
      "num_tokens": 18600813.0,
      "step": 8590
    },
    {
      "entropy": 6.4420863628387455,
      "epoch": 0.9197923912461876,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004925481921788726,
      "loss": 6.1682,
      "mean_token_accuracy": 0.1500229761004448,
      "num_tokens": 18610807.0,
      "step": 8595
    },
    {
      "entropy": 6.371388101577759,
      "epoch": 0.9203274653539515,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000492538430808114,
      "loss": 6.1246,
      "mean_token_accuracy": 0.14546216204762458,
      "num_tokens": 18622108.0,
      "step": 8600
    },
    {
      "entropy": 6.4668769359588625,
      "epoch": 0.9208625394617155,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004925286631559636,
      "loss": 6.2988,
      "mean_token_accuracy": 0.14089180231094361,
      "num_tokens": 18633098.0,
      "step": 8605
    },
    {
      "entropy": 6.491634798049927,
      "epoch": 0.9213976135694794,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004925188892227038,
      "loss": 6.1806,
      "mean_token_accuracy": 0.14678704962134362,
      "num_tokens": 18643603.0,
      "step": 8610
    },
    {
      "entropy": 6.4450047492980955,
      "epoch": 0.9219326876772433,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004925091090086166,
      "loss": 6.2025,
      "mean_token_accuracy": 0.14506881237030028,
      "num_tokens": 18654354.0,
      "step": 8615
    },
    {
      "entropy": 6.532670545578003,
      "epoch": 0.9224677617850072,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004924993225139845,
      "loss": 6.2735,
      "mean_token_accuracy": 0.14608646482229232,
      "num_tokens": 18665349.0,
      "step": 8620
    },
    {
      "entropy": 6.525163078308106,
      "epoch": 0.9230028358927711,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004924895297390899,
      "loss": 6.2269,
      "mean_token_accuracy": 0.13846409171819687,
      "num_tokens": 18676575.0,
      "step": 8625
    },
    {
      "entropy": 6.454398155212402,
      "epoch": 0.9235379100005351,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004924797306842157,
      "loss": 6.1735,
      "mean_token_accuracy": 0.1448497913777828,
      "num_tokens": 18686501.0,
      "step": 8630
    },
    {
      "entropy": 6.4889116287231445,
      "epoch": 0.924072984108299,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004924699253496449,
      "loss": 6.2368,
      "mean_token_accuracy": 0.14226862415671349,
      "num_tokens": 18697246.0,
      "step": 8635
    },
    {
      "entropy": 6.393411493301391,
      "epoch": 0.924608058216063,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004924601137356605,
      "loss": 6.1647,
      "mean_token_accuracy": 0.15329194143414498,
      "num_tokens": 18708177.0,
      "step": 8640
    },
    {
      "entropy": 6.481259727478028,
      "epoch": 0.9251431323238268,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000492450295842546,
      "loss": 6.2126,
      "mean_token_accuracy": 0.13905886486172675,
      "num_tokens": 18719334.0,
      "step": 8645
    },
    {
      "entropy": 6.4964659214019775,
      "epoch": 0.9256782064315908,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004924404716705847,
      "loss": 6.1774,
      "mean_token_accuracy": 0.1499700129032135,
      "num_tokens": 18729174.0,
      "step": 8650
    },
    {
      "entropy": 6.448195075988769,
      "epoch": 0.9262132805393547,
      "grad_norm": 0.875,
      "learning_rate": 0.0004924306412200603,
      "loss": 6.2925,
      "mean_token_accuracy": 0.13642457500100136,
      "num_tokens": 18740277.0,
      "step": 8655
    },
    {
      "entropy": 6.507169485092163,
      "epoch": 0.9267483546471186,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004924208044912567,
      "loss": 6.2314,
      "mean_token_accuracy": 0.14669199436903,
      "num_tokens": 18751864.0,
      "step": 8660
    },
    {
      "entropy": 6.413236808776856,
      "epoch": 0.9272834287548826,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004924109614844579,
      "loss": 6.1443,
      "mean_token_accuracy": 0.1510525941848755,
      "num_tokens": 18762786.0,
      "step": 8665
    },
    {
      "entropy": 6.433039760589599,
      "epoch": 0.9278185028626464,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004924011121999482,
      "loss": 6.203,
      "mean_token_accuracy": 0.14765021950006485,
      "num_tokens": 18772913.0,
      "step": 8670
    },
    {
      "entropy": 6.475260257720947,
      "epoch": 0.9283535769704104,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004923912566380118,
      "loss": 6.1673,
      "mean_token_accuracy": 0.14709473699331282,
      "num_tokens": 18784204.0,
      "step": 8675
    },
    {
      "entropy": 6.388618516921997,
      "epoch": 0.9288886510781743,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004923813947989336,
      "loss": 6.2716,
      "mean_token_accuracy": 0.14042105823755263,
      "num_tokens": 18795425.0,
      "step": 8680
    },
    {
      "entropy": 6.520901441574097,
      "epoch": 0.9294237251859383,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004923715266829981,
      "loss": 6.1898,
      "mean_token_accuracy": 0.14483431875705718,
      "num_tokens": 18806850.0,
      "step": 8685
    },
    {
      "entropy": 6.435678291320801,
      "epoch": 0.9299587992937022,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923616522904903,
      "loss": 6.1387,
      "mean_token_accuracy": 0.15046218186616897,
      "num_tokens": 18817272.0,
      "step": 8690
    },
    {
      "entropy": 6.501014041900635,
      "epoch": 0.930493873401466,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004923517716216953,
      "loss": 6.2869,
      "mean_token_accuracy": 0.13602351620793343,
      "num_tokens": 18827075.0,
      "step": 8695
    },
    {
      "entropy": 6.511406564712525,
      "epoch": 0.93102894750923,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004923418846768985,
      "loss": 6.1981,
      "mean_token_accuracy": 0.1427587501704693,
      "num_tokens": 18837635.0,
      "step": 8700
    },
    {
      "entropy": 6.4627467632293705,
      "epoch": 0.9315640216169939,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004923319914563852,
      "loss": 6.1951,
      "mean_token_accuracy": 0.1493883490562439,
      "num_tokens": 18848559.0,
      "step": 8705
    },
    {
      "entropy": 6.490451526641846,
      "epoch": 0.9320990957247579,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004923220919604413,
      "loss": 6.2163,
      "mean_token_accuracy": 0.14165103286504746,
      "num_tokens": 18859327.0,
      "step": 8710
    },
    {
      "entropy": 6.414099025726318,
      "epoch": 0.9326341698325218,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004923121861893525,
      "loss": 6.1779,
      "mean_token_accuracy": 0.14123260378837585,
      "num_tokens": 18869403.0,
      "step": 8715
    },
    {
      "entropy": 6.4744631290435795,
      "epoch": 0.9331692439402858,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004923022741434049,
      "loss": 6.2033,
      "mean_token_accuracy": 0.140889972448349,
      "num_tokens": 18879890.0,
      "step": 8720
    },
    {
      "entropy": 6.505236625671387,
      "epoch": 0.9337043180480497,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004922923558228846,
      "loss": 6.222,
      "mean_token_accuracy": 0.13713237941265105,
      "num_tokens": 18890361.0,
      "step": 8725
    },
    {
      "entropy": 6.4382130146026615,
      "epoch": 0.9342393921558135,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004922824312280781,
      "loss": 6.1395,
      "mean_token_accuracy": 0.148147551715374,
      "num_tokens": 18902344.0,
      "step": 8730
    },
    {
      "entropy": 6.405410432815552,
      "epoch": 0.9347744662635775,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000492272500359272,
      "loss": 6.1412,
      "mean_token_accuracy": 0.15164890885353088,
      "num_tokens": 18911364.0,
      "step": 8735
    },
    {
      "entropy": 6.434286165237427,
      "epoch": 0.9353095403713414,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004922625632167529,
      "loss": 6.2359,
      "mean_token_accuracy": 0.13671373650431634,
      "num_tokens": 18922254.0,
      "step": 8740
    },
    {
      "entropy": 6.529753828048706,
      "epoch": 0.9358446144791054,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004922526198008078,
      "loss": 6.1991,
      "mean_token_accuracy": 0.14297783002257347,
      "num_tokens": 18932943.0,
      "step": 8745
    },
    {
      "entropy": 6.506849908828736,
      "epoch": 0.9363796885868693,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004922426701117239,
      "loss": 6.3003,
      "mean_token_accuracy": 0.13828289583325387,
      "num_tokens": 18943523.0,
      "step": 8750
    },
    {
      "entropy": 6.537208318710327,
      "epoch": 0.9369147626946333,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004922327141497884,
      "loss": 6.273,
      "mean_token_accuracy": 0.13431756868958472,
      "num_tokens": 18955220.0,
      "step": 8755
    },
    {
      "entropy": 6.558082008361817,
      "epoch": 0.9374498368023971,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004922227519152888,
      "loss": 6.2885,
      "mean_token_accuracy": 0.14038701802492143,
      "num_tokens": 18966714.0,
      "step": 8760
    },
    {
      "entropy": 6.40695595741272,
      "epoch": 0.937984910910161,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004922127834085128,
      "loss": 6.15,
      "mean_token_accuracy": 0.14898365288972854,
      "num_tokens": 18977998.0,
      "step": 8765
    },
    {
      "entropy": 6.435443258285522,
      "epoch": 0.938519985017925,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004922028086297481,
      "loss": 6.1423,
      "mean_token_accuracy": 0.14595036581158638,
      "num_tokens": 18990210.0,
      "step": 8770
    },
    {
      "entropy": 6.5269208431243895,
      "epoch": 0.9390550591256889,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004921928275792828,
      "loss": 6.2425,
      "mean_token_accuracy": 0.14962659180164337,
      "num_tokens": 19000529.0,
      "step": 8775
    },
    {
      "entropy": 6.436434268951416,
      "epoch": 0.9395901332334529,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004921828402574052,
      "loss": 6.1793,
      "mean_token_accuracy": 0.15043628811836243,
      "num_tokens": 19011386.0,
      "step": 8780
    },
    {
      "entropy": 6.437271976470948,
      "epoch": 0.9401252073412167,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004921728466644037,
      "loss": 6.1707,
      "mean_token_accuracy": 0.14790606051683425,
      "num_tokens": 19022196.0,
      "step": 8785
    },
    {
      "entropy": 6.429384708404541,
      "epoch": 0.9406602814489807,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004921628468005666,
      "loss": 6.2038,
      "mean_token_accuracy": 0.1489550918340683,
      "num_tokens": 19033221.0,
      "step": 8790
    },
    {
      "entropy": 6.436385250091552,
      "epoch": 0.9411953555567446,
      "grad_norm": 0.859375,
      "learning_rate": 0.000492152840666183,
      "loss": 6.2469,
      "mean_token_accuracy": 0.13902419656515122,
      "num_tokens": 19044953.0,
      "step": 8795
    },
    {
      "entropy": 6.52896409034729,
      "epoch": 0.9417304296645085,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004921428282615415,
      "loss": 6.2655,
      "mean_token_accuracy": 0.13636692017316818,
      "num_tokens": 19055512.0,
      "step": 8800
    },
    {
      "entropy": 6.52111701965332,
      "epoch": 0.9422655037722725,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004921328095869313,
      "loss": 6.2652,
      "mean_token_accuracy": 0.14065030515193938,
      "num_tokens": 19065659.0,
      "step": 8805
    },
    {
      "entropy": 6.427253198623657,
      "epoch": 0.9428005778800364,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004921227846426418,
      "loss": 6.1655,
      "mean_token_accuracy": 0.1540947064757347,
      "num_tokens": 19076955.0,
      "step": 8810
    },
    {
      "entropy": 6.4817615985870365,
      "epoch": 0.9433356519878003,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004921127534289624,
      "loss": 6.2729,
      "mean_token_accuracy": 0.13906664103269578,
      "num_tokens": 19088305.0,
      "step": 8815
    },
    {
      "entropy": 6.467895793914795,
      "epoch": 0.9438707260955642,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004921027159461829,
      "loss": 6.2452,
      "mean_token_accuracy": 0.140907621383667,
      "num_tokens": 19098046.0,
      "step": 8820
    },
    {
      "entropy": 6.45828161239624,
      "epoch": 0.9444058002033282,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004920926721945927,
      "loss": 6.1678,
      "mean_token_accuracy": 0.14571942687034606,
      "num_tokens": 19109019.0,
      "step": 8825
    },
    {
      "entropy": 6.477725887298584,
      "epoch": 0.9449408743110921,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004920826221744823,
      "loss": 6.186,
      "mean_token_accuracy": 0.1437153622508049,
      "num_tokens": 19119624.0,
      "step": 8830
    },
    {
      "entropy": 6.473362302780151,
      "epoch": 0.945475948418856,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004920725658861417,
      "loss": 6.2235,
      "mean_token_accuracy": 0.13894851878285408,
      "num_tokens": 19130577.0,
      "step": 8835
    },
    {
      "entropy": 6.482684326171875,
      "epoch": 0.94601102252662,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004920625033298611,
      "loss": 6.1426,
      "mean_token_accuracy": 0.14513972103595735,
      "num_tokens": 19140254.0,
      "step": 8840
    },
    {
      "entropy": 6.436223936080933,
      "epoch": 0.9465460966343838,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004920524345059314,
      "loss": 6.169,
      "mean_token_accuracy": 0.14718737304210663,
      "num_tokens": 19151357.0,
      "step": 8845
    },
    {
      "entropy": 6.492574214935303,
      "epoch": 0.9470811707421478,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004920423594146431,
      "loss": 6.182,
      "mean_token_accuracy": 0.14779068231582643,
      "num_tokens": 19162069.0,
      "step": 8850
    },
    {
      "entropy": 6.36601414680481,
      "epoch": 0.9476162448499117,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004920322780562872,
      "loss": 6.0777,
      "mean_token_accuracy": 0.15069809556007385,
      "num_tokens": 19172515.0,
      "step": 8855
    },
    {
      "entropy": 6.360944366455078,
      "epoch": 0.9481513189576757,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004920221904311546,
      "loss": 6.0783,
      "mean_token_accuracy": 0.14836317151784897,
      "num_tokens": 19182450.0,
      "step": 8860
    },
    {
      "entropy": 6.485449266433716,
      "epoch": 0.9486863930654396,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004920120965395369,
      "loss": 6.2865,
      "mean_token_accuracy": 0.14288815706968308,
      "num_tokens": 19193833.0,
      "step": 8865
    },
    {
      "entropy": 6.518713712692261,
      "epoch": 0.9492214671732034,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004920019963817253,
      "loss": 6.2263,
      "mean_token_accuracy": 0.13764797747135163,
      "num_tokens": 19204970.0,
      "step": 8870
    },
    {
      "entropy": 6.458697080612183,
      "epoch": 0.9497565412809674,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004919918899580117,
      "loss": 6.1844,
      "mean_token_accuracy": 0.14529381543397904,
      "num_tokens": 19215795.0,
      "step": 8875
    },
    {
      "entropy": 6.4908928871154785,
      "epoch": 0.9502916153887313,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004919817772686876,
      "loss": 6.22,
      "mean_token_accuracy": 0.13599698767066,
      "num_tokens": 19226375.0,
      "step": 8880
    },
    {
      "entropy": 6.5375683307647705,
      "epoch": 0.9508266894964953,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004919716583140453,
      "loss": 6.2161,
      "mean_token_accuracy": 0.14754707515239715,
      "num_tokens": 19236066.0,
      "step": 8885
    },
    {
      "entropy": 6.399689483642578,
      "epoch": 0.9513617636042592,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004919615330943769,
      "loss": 6.1073,
      "mean_token_accuracy": 0.1497982382774353,
      "num_tokens": 19246757.0,
      "step": 8890
    },
    {
      "entropy": 6.525915575027466,
      "epoch": 0.9518968377120232,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004919514016099746,
      "loss": 6.2522,
      "mean_token_accuracy": 0.141085946559906,
      "num_tokens": 19257875.0,
      "step": 8895
    },
    {
      "entropy": 6.496263742446899,
      "epoch": 0.952431911819787,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000491941263861131,
      "loss": 6.2112,
      "mean_token_accuracy": 0.14586174190044404,
      "num_tokens": 19270236.0,
      "step": 8900
    },
    {
      "entropy": 6.452596187591553,
      "epoch": 0.9529669859275509,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000491931119848139,
      "loss": 6.2059,
      "mean_token_accuracy": 0.1410587064921856,
      "num_tokens": 19280877.0,
      "step": 8905
    },
    {
      "entropy": 6.418635129928589,
      "epoch": 0.9535020600353149,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004919209695712915,
      "loss": 6.085,
      "mean_token_accuracy": 0.15473085418343543,
      "num_tokens": 19292852.0,
      "step": 8910
    },
    {
      "entropy": 6.497902727127075,
      "epoch": 0.9540371341430788,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004919108130308814,
      "loss": 6.3214,
      "mean_token_accuracy": 0.13175884634256363,
      "num_tokens": 19305516.0,
      "step": 8915
    },
    {
      "entropy": 6.450237178802491,
      "epoch": 0.9545722082508428,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004919006502272019,
      "loss": 6.1344,
      "mean_token_accuracy": 0.1516859620809555,
      "num_tokens": 19315436.0,
      "step": 8920
    },
    {
      "entropy": 6.448349285125732,
      "epoch": 0.9551072823586066,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004918904811605468,
      "loss": 6.1739,
      "mean_token_accuracy": 0.14867157340049744,
      "num_tokens": 19325393.0,
      "step": 8925
    },
    {
      "entropy": 6.441112756729126,
      "epoch": 0.9556423564663706,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004918803058312094,
      "loss": 6.1719,
      "mean_token_accuracy": 0.15174808502197265,
      "num_tokens": 19336722.0,
      "step": 8930
    },
    {
      "entropy": 6.553536891937256,
      "epoch": 0.9561774305741345,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004918701242394837,
      "loss": 6.3449,
      "mean_token_accuracy": 0.13422898277640344,
      "num_tokens": 19348406.0,
      "step": 8935
    },
    {
      "entropy": 6.369860744476318,
      "epoch": 0.9567125046818984,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004918599363856637,
      "loss": 6.1443,
      "mean_token_accuracy": 0.15485092550516127,
      "num_tokens": 19359714.0,
      "step": 8940
    },
    {
      "entropy": 6.472765588760376,
      "epoch": 0.9572475787896624,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004918497422700434,
      "loss": 6.1756,
      "mean_token_accuracy": 0.14712189882993698,
      "num_tokens": 19370833.0,
      "step": 8945
    },
    {
      "entropy": 6.502059125900269,
      "epoch": 0.9577826528974263,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004918395418929174,
      "loss": 6.2618,
      "mean_token_accuracy": 0.1383008822798729,
      "num_tokens": 19382098.0,
      "step": 8950
    },
    {
      "entropy": 6.485216426849365,
      "epoch": 0.9583177270051902,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00049182933525458,
      "loss": 6.2229,
      "mean_token_accuracy": 0.14583272486925125,
      "num_tokens": 19393595.0,
      "step": 8955
    },
    {
      "entropy": 6.452768325805664,
      "epoch": 0.9588528011129541,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000491819122355326,
      "loss": 6.18,
      "mean_token_accuracy": 0.14826554507017137,
      "num_tokens": 19405296.0,
      "step": 8960
    },
    {
      "entropy": 6.472521591186523,
      "epoch": 0.9593878752207181,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004918089031954503,
      "loss": 6.1994,
      "mean_token_accuracy": 0.14533568918704987,
      "num_tokens": 19415693.0,
      "step": 8965
    },
    {
      "entropy": 6.451762771606445,
      "epoch": 0.959922949328482,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000491798677775248,
      "loss": 6.2343,
      "mean_token_accuracy": 0.14552236944437028,
      "num_tokens": 19426140.0,
      "step": 8970
    },
    {
      "entropy": 6.5337108135223385,
      "epoch": 0.9604580234362459,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004917884460950144,
      "loss": 6.2858,
      "mean_token_accuracy": 0.13785114735364914,
      "num_tokens": 19437543.0,
      "step": 8975
    },
    {
      "entropy": 6.444455099105835,
      "epoch": 0.9609930975440099,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000491778208155045,
      "loss": 6.1899,
      "mean_token_accuracy": 0.1457691676914692,
      "num_tokens": 19447788.0,
      "step": 8980
    },
    {
      "entropy": 6.517717027664185,
      "epoch": 0.9615281716517737,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004917679639556351,
      "loss": 6.2065,
      "mean_token_accuracy": 0.13845892250537872,
      "num_tokens": 19457587.0,
      "step": 8985
    },
    {
      "entropy": 6.439982891082764,
      "epoch": 0.9620632457595377,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004917577134970808,
      "loss": 6.119,
      "mean_token_accuracy": 0.15125673711299897,
      "num_tokens": 19467422.0,
      "step": 8990
    },
    {
      "entropy": 6.420416069030762,
      "epoch": 0.9625983198673016,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004917474567796778,
      "loss": 6.1781,
      "mean_token_accuracy": 0.1490088813006878,
      "num_tokens": 19479037.0,
      "step": 8995
    },
    {
      "entropy": 6.421410846710205,
      "epoch": 0.9631333939750656,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004917371938037226,
      "loss": 6.1248,
      "mean_token_accuracy": 0.1563099980354309,
      "num_tokens": 19489639.0,
      "step": 9000
    },
    {
      "epoch": 0.9631333939750656,
      "eval_entropy": 6.2821573879519566,
      "eval_loss": 6.2367844581604,
      "eval_mean_token_accuracy": 0.14829185967260816,
      "eval_num_tokens": 19489639.0,
      "eval_runtime": 17.5097,
      "eval_samples_per_second": 1695.006,
      "eval_steps_per_second": 211.883,
      "step": 9000
    },
    {
      "entropy": 6.512316179275513,
      "epoch": 0.9636684680828295,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004917269245695114,
      "loss": 6.2516,
      "mean_token_accuracy": 0.1357911229133606,
      "num_tokens": 19500592.0,
      "step": 9005
    },
    {
      "entropy": 6.500863075256348,
      "epoch": 0.9642035421905933,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004917166490773406,
      "loss": 6.2452,
      "mean_token_accuracy": 0.14373304396867753,
      "num_tokens": 19510818.0,
      "step": 9010
    },
    {
      "entropy": 6.480406856536865,
      "epoch": 0.9647386162983573,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004917063673275071,
      "loss": 6.2731,
      "mean_token_accuracy": 0.1420287720859051,
      "num_tokens": 19521283.0,
      "step": 9015
    },
    {
      "entropy": 6.505241775512696,
      "epoch": 0.9652736904061212,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004916960793203077,
      "loss": 6.1035,
      "mean_token_accuracy": 0.14841227680444719,
      "num_tokens": 19531799.0,
      "step": 9020
    },
    {
      "entropy": 6.449901580810547,
      "epoch": 0.9658087645138852,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004916857850560394,
      "loss": 6.0724,
      "mean_token_accuracy": 0.16073150038719178,
      "num_tokens": 19542377.0,
      "step": 9025
    },
    {
      "entropy": 6.372377395629883,
      "epoch": 0.9663438386216491,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004916754845349996,
      "loss": 6.094,
      "mean_token_accuracy": 0.15285499542951583,
      "num_tokens": 19552654.0,
      "step": 9030
    },
    {
      "entropy": 6.44136610031128,
      "epoch": 0.9668789127294131,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004916651777574856,
      "loss": 6.1786,
      "mean_token_accuracy": 0.14090803936123847,
      "num_tokens": 19563439.0,
      "step": 9035
    },
    {
      "entropy": 6.528454923629761,
      "epoch": 0.9674139868371769,
      "grad_norm": 0.859375,
      "learning_rate": 0.000491654864723795,
      "loss": 6.2028,
      "mean_token_accuracy": 0.1398625247180462,
      "num_tokens": 19574264.0,
      "step": 9040
    },
    {
      "entropy": 6.324220514297485,
      "epoch": 0.9679490609449408,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004916445454342256,
      "loss": 6.0844,
      "mean_token_accuracy": 0.15555037930607796,
      "num_tokens": 19585293.0,
      "step": 9045
    },
    {
      "entropy": 6.443246936798095,
      "epoch": 0.9684841350527048,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004916342198890754,
      "loss": 6.1064,
      "mean_token_accuracy": 0.1432489939033985,
      "num_tokens": 19594384.0,
      "step": 9050
    },
    {
      "entropy": 6.416325330734253,
      "epoch": 0.9690192091604687,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004916238880886426,
      "loss": 6.2339,
      "mean_token_accuracy": 0.1461181789636612,
      "num_tokens": 19605144.0,
      "step": 9055
    },
    {
      "entropy": 6.410384750366211,
      "epoch": 0.9695542832682327,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004916135500332255,
      "loss": 6.1337,
      "mean_token_accuracy": 0.14784870445728301,
      "num_tokens": 19615885.0,
      "step": 9060
    },
    {
      "entropy": 6.442836332321167,
      "epoch": 0.9700893573759966,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004916032057231225,
      "loss": 6.1386,
      "mean_token_accuracy": 0.1452575594186783,
      "num_tokens": 19625749.0,
      "step": 9065
    },
    {
      "entropy": 6.450768375396729,
      "epoch": 0.9706244314837605,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004915928551586324,
      "loss": 6.2368,
      "mean_token_accuracy": 0.13815416768193245,
      "num_tokens": 19638285.0,
      "step": 9070
    },
    {
      "entropy": 6.53087306022644,
      "epoch": 0.9711595055915244,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004915824983400541,
      "loss": 6.2401,
      "mean_token_accuracy": 0.13961710557341575,
      "num_tokens": 19649697.0,
      "step": 9075
    },
    {
      "entropy": 6.5055841445922855,
      "epoch": 0.9716945796992883,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004915721352676864,
      "loss": 6.2645,
      "mean_token_accuracy": 0.1423683024942875,
      "num_tokens": 19659668.0,
      "step": 9080
    },
    {
      "entropy": 6.431279706954956,
      "epoch": 0.9722296538070523,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004915617659418289,
      "loss": 6.1116,
      "mean_token_accuracy": 0.1474567398428917,
      "num_tokens": 19670301.0,
      "step": 9085
    },
    {
      "entropy": 6.460619783401489,
      "epoch": 0.9727647279148162,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004915513903627807,
      "loss": 6.1536,
      "mean_token_accuracy": 0.14713016524910927,
      "num_tokens": 19680926.0,
      "step": 9090
    },
    {
      "entropy": 6.407574987411499,
      "epoch": 0.9732998020225802,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004915410085308417,
      "loss": 6.1432,
      "mean_token_accuracy": 0.15075530707836152,
      "num_tokens": 19691467.0,
      "step": 9095
    },
    {
      "entropy": 6.4866297245025635,
      "epoch": 0.973834876130344,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004915306204463115,
      "loss": 6.1477,
      "mean_token_accuracy": 0.1452942445874214,
      "num_tokens": 19701139.0,
      "step": 9100
    },
    {
      "entropy": 6.434195280075073,
      "epoch": 0.974369950238108,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00049152022610949,
      "loss": 6.1617,
      "mean_token_accuracy": 0.14280082285404205,
      "num_tokens": 19710628.0,
      "step": 9105
    },
    {
      "entropy": 6.468556261062622,
      "epoch": 0.9749050243458719,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004915098255206773,
      "loss": 6.1706,
      "mean_token_accuracy": 0.14397796392440795,
      "num_tokens": 19721824.0,
      "step": 9110
    },
    {
      "entropy": 6.473300123214722,
      "epoch": 0.9754400984536358,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000491499418680174,
      "loss": 6.188,
      "mean_token_accuracy": 0.14112256616353988,
      "num_tokens": 19732557.0,
      "step": 9115
    },
    {
      "entropy": 6.447193050384522,
      "epoch": 0.9759751725613998,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004914890055882804,
      "loss": 6.1703,
      "mean_token_accuracy": 0.14254653677344323,
      "num_tokens": 19743436.0,
      "step": 9120
    },
    {
      "entropy": 6.427279233932495,
      "epoch": 0.9765102466691636,
      "grad_norm": 0.859375,
      "learning_rate": 0.000491478586245297,
      "loss": 6.0804,
      "mean_token_accuracy": 0.14952216893434525,
      "num_tokens": 19755305.0,
      "step": 9125
    },
    {
      "entropy": 6.448919296264648,
      "epoch": 0.9770453207769276,
      "grad_norm": 0.96875,
      "learning_rate": 0.000491468160651525,
      "loss": 6.0853,
      "mean_token_accuracy": 0.15592157691717148,
      "num_tokens": 19766223.0,
      "step": 9130
    },
    {
      "entropy": 6.423963165283203,
      "epoch": 0.9775803948846915,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004914577288072652,
      "loss": 6.205,
      "mean_token_accuracy": 0.1421111471951008,
      "num_tokens": 19777268.0,
      "step": 9135
    },
    {
      "entropy": 6.496765041351319,
      "epoch": 0.9781154689924555,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004914472907128189,
      "loss": 6.2707,
      "mean_token_accuracy": 0.1461112916469574,
      "num_tokens": 19787875.0,
      "step": 9140
    },
    {
      "entropy": 6.483035516738892,
      "epoch": 0.9786505431002194,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004914368463684876,
      "loss": 6.1416,
      "mean_token_accuracy": 0.15096967071294784,
      "num_tokens": 19798919.0,
      "step": 9145
    },
    {
      "entropy": 6.452010536193848,
      "epoch": 0.9791856172079834,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004914263957745726,
      "loss": 6.1404,
      "mean_token_accuracy": 0.15096383094787597,
      "num_tokens": 19809320.0,
      "step": 9150
    },
    {
      "entropy": 6.481369829177856,
      "epoch": 0.9797206913157472,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004914159389313759,
      "loss": 6.2193,
      "mean_token_accuracy": 0.14834315627813338,
      "num_tokens": 19819511.0,
      "step": 9155
    },
    {
      "entropy": 6.523835277557373,
      "epoch": 0.9802557654235111,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004914054758391994,
      "loss": 6.2409,
      "mean_token_accuracy": 0.1416543945670128,
      "num_tokens": 19830381.0,
      "step": 9160
    },
    {
      "entropy": 6.353017997741699,
      "epoch": 0.9807908395312751,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004913950064983452,
      "loss": 6.1001,
      "mean_token_accuracy": 0.143266025185585,
      "num_tokens": 19840806.0,
      "step": 9165
    },
    {
      "entropy": 6.373284578323364,
      "epoch": 0.981325913639039,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004913845309091156,
      "loss": 6.1134,
      "mean_token_accuracy": 0.1534092292189598,
      "num_tokens": 19851095.0,
      "step": 9170
    },
    {
      "entropy": 6.422077417373657,
      "epoch": 0.981860987746803,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004913740490718131,
      "loss": 6.0837,
      "mean_token_accuracy": 0.15422788113355637,
      "num_tokens": 19860235.0,
      "step": 9175
    },
    {
      "entropy": 6.387666273117065,
      "epoch": 0.9823960618545668,
      "grad_norm": 0.875,
      "learning_rate": 0.0004913635609867403,
      "loss": 6.1805,
      "mean_token_accuracy": 0.13871687650680542,
      "num_tokens": 19872264.0,
      "step": 9180
    },
    {
      "entropy": 6.486538076400757,
      "epoch": 0.9829311359623308,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004913530666542001,
      "loss": 6.265,
      "mean_token_accuracy": 0.14309014976024628,
      "num_tokens": 19883628.0,
      "step": 9185
    },
    {
      "entropy": 6.50969443321228,
      "epoch": 0.9834662100700947,
      "grad_norm": 0.875,
      "learning_rate": 0.0004913425660744955,
      "loss": 6.153,
      "mean_token_accuracy": 0.14640728160738944,
      "num_tokens": 19894624.0,
      "step": 9190
    },
    {
      "entropy": 6.479321765899658,
      "epoch": 0.9840012841778586,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004913320592479297,
      "loss": 6.159,
      "mean_token_accuracy": 0.14386670142412186,
      "num_tokens": 19904882.0,
      "step": 9195
    },
    {
      "entropy": 6.37914137840271,
      "epoch": 0.9845363582856226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004913215461748061,
      "loss": 6.1028,
      "mean_token_accuracy": 0.15330933928489685,
      "num_tokens": 19915344.0,
      "step": 9200
    },
    {
      "entropy": 6.410844945907593,
      "epoch": 0.9850714323933865,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004913110268554282,
      "loss": 6.2437,
      "mean_token_accuracy": 0.13824103996157647,
      "num_tokens": 19927527.0,
      "step": 9205
    },
    {
      "entropy": 6.4512537002563475,
      "epoch": 0.9856065065011504,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004913005012900998,
      "loss": 6.1829,
      "mean_token_accuracy": 0.14253388047218324,
      "num_tokens": 19938893.0,
      "step": 9210
    },
    {
      "entropy": 6.453767776489258,
      "epoch": 0.9861415806089143,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004912899694791248,
      "loss": 6.1348,
      "mean_token_accuracy": 0.148092557489872,
      "num_tokens": 19949297.0,
      "step": 9215
    },
    {
      "entropy": 6.403385829925537,
      "epoch": 0.9866766547166783,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004912794314228074,
      "loss": 6.1205,
      "mean_token_accuracy": 0.15008987188339235,
      "num_tokens": 19959194.0,
      "step": 9220
    },
    {
      "entropy": 6.477009010314942,
      "epoch": 0.9872117288244422,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004912688871214518,
      "loss": 6.1784,
      "mean_token_accuracy": 0.14522113054990768,
      "num_tokens": 19969896.0,
      "step": 9225
    },
    {
      "entropy": 6.488388299942017,
      "epoch": 0.9877468029322061,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004912583365753625,
      "loss": 6.1442,
      "mean_token_accuracy": 0.14976364076137544,
      "num_tokens": 19981505.0,
      "step": 9230
    },
    {
      "entropy": 6.431499719619751,
      "epoch": 0.9882818770399701,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004912477797848441,
      "loss": 6.1327,
      "mean_token_accuracy": 0.14974280893802644,
      "num_tokens": 19992770.0,
      "step": 9235
    },
    {
      "entropy": 6.407771348953247,
      "epoch": 0.9888169511477339,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004912372167502014,
      "loss": 6.1259,
      "mean_token_accuracy": 0.14748117029666902,
      "num_tokens": 20003572.0,
      "step": 9240
    },
    {
      "entropy": 6.366019058227539,
      "epoch": 0.9893520252554979,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004912266474717394,
      "loss": 6.173,
      "mean_token_accuracy": 0.1477161854505539,
      "num_tokens": 20014500.0,
      "step": 9245
    },
    {
      "entropy": 6.484816217422486,
      "epoch": 0.9898870993632618,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004912160719497635,
      "loss": 6.1613,
      "mean_token_accuracy": 0.14817373007535933,
      "num_tokens": 20025745.0,
      "step": 9250
    },
    {
      "entropy": 6.434405565261841,
      "epoch": 0.9904221734710258,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004912054901845788,
      "loss": 6.0649,
      "mean_token_accuracy": 0.14782988280057907,
      "num_tokens": 20036346.0,
      "step": 9255
    },
    {
      "entropy": 6.349883651733398,
      "epoch": 0.9909572475787897,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000491194902176491,
      "loss": 6.1925,
      "mean_token_accuracy": 0.14366608560085298,
      "num_tokens": 20048162.0,
      "step": 9260
    },
    {
      "entropy": 6.459431457519531,
      "epoch": 0.9914923216865535,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004911843079258057,
      "loss": 6.2909,
      "mean_token_accuracy": 0.1344110570847988,
      "num_tokens": 20058947.0,
      "step": 9265
    },
    {
      "entropy": 6.550050592422485,
      "epoch": 0.9920273957943175,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004911737074328289,
      "loss": 6.2108,
      "mean_token_accuracy": 0.14493196010589598,
      "num_tokens": 20069282.0,
      "step": 9270
    },
    {
      "entropy": 6.420050811767578,
      "epoch": 0.9925624699020814,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004911631006978667,
      "loss": 6.174,
      "mean_token_accuracy": 0.14369123056530952,
      "num_tokens": 20078347.0,
      "step": 9275
    },
    {
      "entropy": 6.423564386367798,
      "epoch": 0.9930975440098454,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004911524877212254,
      "loss": 6.1882,
      "mean_token_accuracy": 0.1409389466047287,
      "num_tokens": 20089129.0,
      "step": 9280
    },
    {
      "entropy": 6.478132009506226,
      "epoch": 0.9936326181176093,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004911418685032114,
      "loss": 6.1835,
      "mean_token_accuracy": 0.14103031009435654,
      "num_tokens": 20101008.0,
      "step": 9285
    },
    {
      "entropy": 6.395486497879029,
      "epoch": 0.9941676922253733,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004911312430441312,
      "loss": 6.2033,
      "mean_token_accuracy": 0.14719966948032379,
      "num_tokens": 20112272.0,
      "step": 9290
    },
    {
      "entropy": 6.45061469078064,
      "epoch": 0.9947027663331371,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004911206113442918,
      "loss": 6.0842,
      "mean_token_accuracy": 0.15623315870761872,
      "num_tokens": 20122814.0,
      "step": 9295
    },
    {
      "entropy": 6.395213270187378,
      "epoch": 0.995237840440901,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004911099734040001,
      "loss": 6.0479,
      "mean_token_accuracy": 0.1502678319811821,
      "num_tokens": 20133169.0,
      "step": 9300
    },
    {
      "entropy": 6.455768632888794,
      "epoch": 0.995772914548665,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004910993292235634,
      "loss": 6.1887,
      "mean_token_accuracy": 0.13966709449887277,
      "num_tokens": 20144355.0,
      "step": 9305
    },
    {
      "entropy": 6.484028100967407,
      "epoch": 0.9963079886564289,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004910886788032887,
      "loss": 6.1971,
      "mean_token_accuracy": 0.14190244823694229,
      "num_tokens": 20154486.0,
      "step": 9310
    },
    {
      "entropy": 6.434096717834473,
      "epoch": 0.9968430627641929,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004910780221434839,
      "loss": 6.1442,
      "mean_token_accuracy": 0.1508187547326088,
      "num_tokens": 20166359.0,
      "step": 9315
    },
    {
      "entropy": 6.431360101699829,
      "epoch": 0.9973781368719568,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004910673592444566,
      "loss": 6.162,
      "mean_token_accuracy": 0.14606508016586303,
      "num_tokens": 20176469.0,
      "step": 9320
    },
    {
      "entropy": 6.480517768859864,
      "epoch": 0.9979132109797207,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004910566901065147,
      "loss": 6.1951,
      "mean_token_accuracy": 0.14490399807691573,
      "num_tokens": 20188334.0,
      "step": 9325
    },
    {
      "entropy": 6.428728151321411,
      "epoch": 0.9984482850874846,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004910460147299661,
      "loss": 6.0959,
      "mean_token_accuracy": 0.1537097230553627,
      "num_tokens": 20198594.0,
      "step": 9330
    },
    {
      "entropy": 6.407783126831054,
      "epoch": 0.9989833591952485,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004910353331151193,
      "loss": 6.2011,
      "mean_token_accuracy": 0.14391589686274528,
      "num_tokens": 20209456.0,
      "step": 9335
    },
    {
      "entropy": 6.433685207366944,
      "epoch": 0.9995184333030125,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004910246452622825,
      "loss": 6.182,
      "mean_token_accuracy": 0.14907921329140664,
      "num_tokens": 20219835.0,
      "step": 9340
    },
    {
      "entropy": 6.553074094984266,
      "epoch": 1.0,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004910139511717646,
      "loss": 6.2201,
      "mean_token_accuracy": 0.13815473351213667,
      "num_tokens": 20228258.0,
      "step": 9345
    },
    {
      "entropy": 6.429244518280029,
      "epoch": 1.0005350741077639,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004910032508438739,
      "loss": 6.0788,
      "mean_token_accuracy": 0.15253917127847672,
      "num_tokens": 20238875.0,
      "step": 9350
    },
    {
      "entropy": 6.455349540710449,
      "epoch": 1.0010701482155278,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00049099254427892,
      "loss": 6.1126,
      "mean_token_accuracy": 0.13827772065997124,
      "num_tokens": 20249750.0,
      "step": 9355
    },
    {
      "entropy": 6.529637479782105,
      "epoch": 1.0016052223232919,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004909818314772116,
      "loss": 6.1281,
      "mean_token_accuracy": 0.14015104472637177,
      "num_tokens": 20259772.0,
      "step": 9360
    },
    {
      "entropy": 6.407935047149659,
      "epoch": 1.0021402964310557,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004909711124390582,
      "loss": 6.0579,
      "mean_token_accuracy": 0.15395714193582535,
      "num_tokens": 20270230.0,
      "step": 9365
    },
    {
      "entropy": 6.354935789108277,
      "epoch": 1.0026753705388196,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004909603871647692,
      "loss": 5.8953,
      "mean_token_accuracy": 0.16960688978433608,
      "num_tokens": 20280317.0,
      "step": 9370
    },
    {
      "entropy": 6.336078548431397,
      "epoch": 1.0032104446465835,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004909496556546544,
      "loss": 5.9841,
      "mean_token_accuracy": 0.15325066298246384,
      "num_tokens": 20291147.0,
      "step": 9375
    },
    {
      "entropy": 6.407212495803833,
      "epoch": 1.0037455187543476,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004909389179090238,
      "loss": 6.0445,
      "mean_token_accuracy": 0.15600289553403854,
      "num_tokens": 20301095.0,
      "step": 9380
    },
    {
      "entropy": 6.460958099365234,
      "epoch": 1.0042805928621115,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004909281739281871,
      "loss": 6.0674,
      "mean_token_accuracy": 0.15700357407331467,
      "num_tokens": 20311216.0,
      "step": 9385
    },
    {
      "entropy": 6.484631061553955,
      "epoch": 1.0048156669698753,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004909174237124548,
      "loss": 6.1276,
      "mean_token_accuracy": 0.13744072318077089,
      "num_tokens": 20321719.0,
      "step": 9390
    },
    {
      "entropy": 6.4652735710144045,
      "epoch": 1.0053507410776392,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004909066672621372,
      "loss": 6.1226,
      "mean_token_accuracy": 0.1485391989350319,
      "num_tokens": 20331997.0,
      "step": 9395
    },
    {
      "entropy": 6.443803405761718,
      "epoch": 1.005885815185403,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004908959045775449,
      "loss": 6.1446,
      "mean_token_accuracy": 0.1454301282763481,
      "num_tokens": 20342973.0,
      "step": 9400
    },
    {
      "entropy": 6.438637828826904,
      "epoch": 1.0064208892931672,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004908851356589887,
      "loss": 6.0497,
      "mean_token_accuracy": 0.15011420249938964,
      "num_tokens": 20352652.0,
      "step": 9405
    },
    {
      "entropy": 6.444362020492553,
      "epoch": 1.006955963400931,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004908743605067796,
      "loss": 6.1096,
      "mean_token_accuracy": 0.1498057022690773,
      "num_tokens": 20364577.0,
      "step": 9410
    },
    {
      "entropy": 6.434169721603394,
      "epoch": 1.007491037508695,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004908635791212287,
      "loss": 6.1685,
      "mean_token_accuracy": 0.14378399923443794,
      "num_tokens": 20375451.0,
      "step": 9415
    },
    {
      "entropy": 6.437387466430664,
      "epoch": 1.0080261116164588,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004908527915026472,
      "loss": 6.0712,
      "mean_token_accuracy": 0.1470625087618828,
      "num_tokens": 20385174.0,
      "step": 9420
    },
    {
      "entropy": 6.455047845840454,
      "epoch": 1.0085611857242227,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004908419976513467,
      "loss": 6.0143,
      "mean_token_accuracy": 0.15466442108154296,
      "num_tokens": 20395710.0,
      "step": 9425
    },
    {
      "entropy": 6.384525966644287,
      "epoch": 1.0090962598319868,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004908311975676388,
      "loss": 6.0162,
      "mean_token_accuracy": 0.15203627645969392,
      "num_tokens": 20406572.0,
      "step": 9430
    },
    {
      "entropy": 6.3723327159881595,
      "epoch": 1.0096313339397507,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004908203912518355,
      "loss": 6.0075,
      "mean_token_accuracy": 0.15625132620334625,
      "num_tokens": 20417786.0,
      "step": 9435
    },
    {
      "entropy": 6.470874643325805,
      "epoch": 1.0101664080475146,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004908095787042487,
      "loss": 6.054,
      "mean_token_accuracy": 0.1521849125623703,
      "num_tokens": 20427680.0,
      "step": 9440
    },
    {
      "entropy": 6.388336277008056,
      "epoch": 1.0107014821552784,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004907987599251907,
      "loss": 6.0473,
      "mean_token_accuracy": 0.14979023784399031,
      "num_tokens": 20438640.0,
      "step": 9445
    },
    {
      "entropy": 6.42610182762146,
      "epoch": 1.0112365562630425,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004907879349149737,
      "loss": 6.0497,
      "mean_token_accuracy": 0.15402074083685874,
      "num_tokens": 20449939.0,
      "step": 9450
    },
    {
      "entropy": 6.381487131118774,
      "epoch": 1.0117716303708064,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004907771036739105,
      "loss": 5.9872,
      "mean_token_accuracy": 0.15439211055636406,
      "num_tokens": 20460845.0,
      "step": 9455
    },
    {
      "entropy": 6.4228596687316895,
      "epoch": 1.0123067044785703,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004907662662023137,
      "loss": 6.1254,
      "mean_token_accuracy": 0.1472632870078087,
      "num_tokens": 20472407.0,
      "step": 9460
    },
    {
      "entropy": 6.4617791175842285,
      "epoch": 1.0128417785863342,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004907554225004962,
      "loss": 6.1201,
      "mean_token_accuracy": 0.14750466868281364,
      "num_tokens": 20483192.0,
      "step": 9465
    },
    {
      "entropy": 6.489104747772217,
      "epoch": 1.013376852694098,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004907445725687714,
      "loss": 6.1927,
      "mean_token_accuracy": 0.14334388449788094,
      "num_tokens": 20493498.0,
      "step": 9470
    },
    {
      "entropy": 6.456091213226318,
      "epoch": 1.0139119268018622,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004907337164074523,
      "loss": 6.0917,
      "mean_token_accuracy": 0.15335296392440795,
      "num_tokens": 20503781.0,
      "step": 9475
    },
    {
      "entropy": 6.360920429229736,
      "epoch": 1.014447000909626,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004907228540168523,
      "loss": 6.1586,
      "mean_token_accuracy": 0.14568022936582564,
      "num_tokens": 20515142.0,
      "step": 9480
    },
    {
      "entropy": 6.453195190429687,
      "epoch": 1.01498207501739,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004907119853972854,
      "loss": 6.1121,
      "mean_token_accuracy": 0.14408067762851715,
      "num_tokens": 20527568.0,
      "step": 9485
    },
    {
      "entropy": 6.417272615432739,
      "epoch": 1.0155171491251538,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004907011105490651,
      "loss": 6.0252,
      "mean_token_accuracy": 0.15408435761928557,
      "num_tokens": 20538198.0,
      "step": 9490
    },
    {
      "entropy": 6.4382647514343265,
      "epoch": 1.0160522232329177,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004906902294725056,
      "loss": 6.0009,
      "mean_token_accuracy": 0.15535251796245575,
      "num_tokens": 20549355.0,
      "step": 9495
    },
    {
      "entropy": 6.381386375427246,
      "epoch": 1.0165872973406818,
      "grad_norm": 0.984375,
      "learning_rate": 0.000490679342167921,
      "loss": 6.1286,
      "mean_token_accuracy": 0.1497268944978714,
      "num_tokens": 20558984.0,
      "step": 9500
    },
    {
      "entropy": 6.448890161514282,
      "epoch": 1.0171223714484456,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004906684486356258,
      "loss": 6.126,
      "mean_token_accuracy": 0.14346849769353867,
      "num_tokens": 20570143.0,
      "step": 9505
    },
    {
      "entropy": 6.503188991546631,
      "epoch": 1.0176574455562095,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004906575488759343,
      "loss": 6.1294,
      "mean_token_accuracy": 0.14717952758073807,
      "num_tokens": 20580973.0,
      "step": 9510
    },
    {
      "entropy": 6.4686949253082275,
      "epoch": 1.0181925196639734,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004906466428891616,
      "loss": 6.0929,
      "mean_token_accuracy": 0.14252540618181228,
      "num_tokens": 20591473.0,
      "step": 9515
    },
    {
      "entropy": 6.386527824401855,
      "epoch": 1.0187275937717375,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004906357306756222,
      "loss": 6.0498,
      "mean_token_accuracy": 0.147423392534256,
      "num_tokens": 20602347.0,
      "step": 9520
    },
    {
      "entropy": 6.43688850402832,
      "epoch": 1.0192626678795014,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004906248122356315,
      "loss": 6.1312,
      "mean_token_accuracy": 0.1535993129014969,
      "num_tokens": 20613168.0,
      "step": 9525
    },
    {
      "entropy": 6.40423583984375,
      "epoch": 1.0197977419872652,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004906138875695045,
      "loss": 6.0066,
      "mean_token_accuracy": 0.15027929544448854,
      "num_tokens": 20623523.0,
      "step": 9530
    },
    {
      "entropy": 6.347176647186279,
      "epoch": 1.0203328160950291,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004906029566775569,
      "loss": 6.0353,
      "mean_token_accuracy": 0.15138544142246246,
      "num_tokens": 20634085.0,
      "step": 9535
    },
    {
      "entropy": 6.45047607421875,
      "epoch": 1.020867890202793,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004905920195601042,
      "loss": 6.1141,
      "mean_token_accuracy": 0.14865031093358994,
      "num_tokens": 20644747.0,
      "step": 9540
    },
    {
      "entropy": 6.442314529418946,
      "epoch": 1.021402964310557,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004905810762174622,
      "loss": 6.0516,
      "mean_token_accuracy": 0.14463598728179933,
      "num_tokens": 20655704.0,
      "step": 9545
    },
    {
      "entropy": 6.446112442016601,
      "epoch": 1.021938038418321,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004905701266499469,
      "loss": 6.053,
      "mean_token_accuracy": 0.15277446210384368,
      "num_tokens": 20665844.0,
      "step": 9550
    },
    {
      "entropy": 6.441574192047119,
      "epoch": 1.0224731125260849,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004905591708578746,
      "loss": 6.1827,
      "mean_token_accuracy": 0.1400070570409298,
      "num_tokens": 20676921.0,
      "step": 9555
    },
    {
      "entropy": 6.453793907165528,
      "epoch": 1.0230081866338487,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004905482088415615,
      "loss": 6.1055,
      "mean_token_accuracy": 0.14364492073655127,
      "num_tokens": 20688476.0,
      "step": 9560
    },
    {
      "entropy": 6.486232471466065,
      "epoch": 1.0235432607416126,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004905372406013241,
      "loss": 6.1134,
      "mean_token_accuracy": 0.14555611461400986,
      "num_tokens": 20698588.0,
      "step": 9565
    },
    {
      "entropy": 6.445996618270874,
      "epoch": 1.0240783348493767,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004905262661374792,
      "loss": 6.091,
      "mean_token_accuracy": 0.15022807642817498,
      "num_tokens": 20709181.0,
      "step": 9570
    },
    {
      "entropy": 6.381361627578736,
      "epoch": 1.0246134089571406,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004905152854503436,
      "loss": 6.0366,
      "mean_token_accuracy": 0.1591507002711296,
      "num_tokens": 20719724.0,
      "step": 9575
    },
    {
      "entropy": 6.493272113800049,
      "epoch": 1.0251484830649045,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004905042985402345,
      "loss": 6.137,
      "mean_token_accuracy": 0.14701187014579772,
      "num_tokens": 20730548.0,
      "step": 9580
    },
    {
      "entropy": 6.440723085403443,
      "epoch": 1.0256835571726683,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000490493305407469,
      "loss": 6.0983,
      "mean_token_accuracy": 0.14397871047258376,
      "num_tokens": 20742124.0,
      "step": 9585
    },
    {
      "entropy": 6.479833030700684,
      "epoch": 1.0262186312804324,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004904823060523646,
      "loss": 6.176,
      "mean_token_accuracy": 0.14197844713926316,
      "num_tokens": 20753313.0,
      "step": 9590
    },
    {
      "entropy": 6.504648923873901,
      "epoch": 1.0267537053881963,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004904713004752389,
      "loss": 6.1249,
      "mean_token_accuracy": 0.1469336949288845,
      "num_tokens": 20763539.0,
      "step": 9595
    },
    {
      "entropy": 6.442349863052368,
      "epoch": 1.0272887794959602,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004904602886764097,
      "loss": 6.2111,
      "mean_token_accuracy": 0.1398083709180355,
      "num_tokens": 20775849.0,
      "step": 9600
    },
    {
      "entropy": 6.436125612258911,
      "epoch": 1.027823853603724,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004904492706561948,
      "loss": 6.0972,
      "mean_token_accuracy": 0.14340001419186593,
      "num_tokens": 20787460.0,
      "step": 9605
    },
    {
      "entropy": 6.557785224914551,
      "epoch": 1.028358927711488,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004904382464149127,
      "loss": 6.0138,
      "mean_token_accuracy": 0.15302670300006865,
      "num_tokens": 20798478.0,
      "step": 9610
    },
    {
      "entropy": 6.4595612525939945,
      "epoch": 1.028894001819252,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004904272159528814,
      "loss": 6.1096,
      "mean_token_accuracy": 0.15218272805213928,
      "num_tokens": 20809730.0,
      "step": 9615
    },
    {
      "entropy": 6.395774030685425,
      "epoch": 1.029429075927016,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004904161792704195,
      "loss": 6.1008,
      "mean_token_accuracy": 0.15611696019768714,
      "num_tokens": 20821438.0,
      "step": 9620
    },
    {
      "entropy": 6.442331218719483,
      "epoch": 1.0299641500347798,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004904051363678457,
      "loss": 6.0774,
      "mean_token_accuracy": 0.14172109961509705,
      "num_tokens": 20832328.0,
      "step": 9625
    },
    {
      "entropy": 6.3869161128997805,
      "epoch": 1.0304992241425437,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004903940872454789,
      "loss": 6.067,
      "mean_token_accuracy": 0.14965847432613372,
      "num_tokens": 20843956.0,
      "step": 9630
    },
    {
      "entropy": 6.490019607543945,
      "epoch": 1.0310342982503076,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004903830319036381,
      "loss": 6.0943,
      "mean_token_accuracy": 0.15099175721406938,
      "num_tokens": 20854578.0,
      "step": 9635
    },
    {
      "entropy": 6.422026252746582,
      "epoch": 1.0315693723580717,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004903719703426425,
      "loss": 6.0119,
      "mean_token_accuracy": 0.14986705780029297,
      "num_tokens": 20865190.0,
      "step": 9640
    },
    {
      "entropy": 6.438509225845337,
      "epoch": 1.0321044464658355,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004903609025628115,
      "loss": 6.0654,
      "mean_token_accuracy": 0.1490228295326233,
      "num_tokens": 20875733.0,
      "step": 9645
    },
    {
      "entropy": 6.364201927185059,
      "epoch": 1.0326395205735994,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004903498285644647,
      "loss": 6.0807,
      "mean_token_accuracy": 0.1466990239918232,
      "num_tokens": 20886900.0,
      "step": 9650
    },
    {
      "entropy": 6.499811315536499,
      "epoch": 1.0331745946813633,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004903387483479219,
      "loss": 6.136,
      "mean_token_accuracy": 0.14672945886850358,
      "num_tokens": 20898125.0,
      "step": 9655
    },
    {
      "entropy": 6.505327987670898,
      "epoch": 1.0337096687891274,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004903276619135029,
      "loss": 6.1571,
      "mean_token_accuracy": 0.1445252537727356,
      "num_tokens": 20909788.0,
      "step": 9660
    },
    {
      "entropy": 6.4599036693573,
      "epoch": 1.0342447428968913,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000490316569261528,
      "loss": 6.118,
      "mean_token_accuracy": 0.14857554733753203,
      "num_tokens": 20921853.0,
      "step": 9665
    },
    {
      "entropy": 6.438937520980835,
      "epoch": 1.0347798170046552,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004903054703923175,
      "loss": 6.1226,
      "mean_token_accuracy": 0.14140074998140334,
      "num_tokens": 20932574.0,
      "step": 9670
    },
    {
      "entropy": 6.402526473999023,
      "epoch": 1.035314891112419,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004902943653061916,
      "loss": 6.0879,
      "mean_token_accuracy": 0.1465308263897896,
      "num_tokens": 20943458.0,
      "step": 9675
    },
    {
      "entropy": 6.445838212966919,
      "epoch": 1.035849965220183,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004902832540034712,
      "loss": 6.0508,
      "mean_token_accuracy": 0.1533766806125641,
      "num_tokens": 20953213.0,
      "step": 9680
    },
    {
      "entropy": 6.420312738418579,
      "epoch": 1.036385039327947,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004902721364844773,
      "loss": 6.089,
      "mean_token_accuracy": 0.14726726561784745,
      "num_tokens": 20963885.0,
      "step": 9685
    },
    {
      "entropy": 6.380733394622803,
      "epoch": 1.0369201134357109,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004902610127495305,
      "loss": 6.0893,
      "mean_token_accuracy": 0.1529219076037407,
      "num_tokens": 20975529.0,
      "step": 9690
    },
    {
      "entropy": 6.429059028625488,
      "epoch": 1.0374551875434748,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004902498827989523,
      "loss": 6.0297,
      "mean_token_accuracy": 0.15851664990186692,
      "num_tokens": 20985279.0,
      "step": 9695
    },
    {
      "entropy": 6.4679827213287355,
      "epoch": 1.0379902616512386,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004902387466330639,
      "loss": 6.1083,
      "mean_token_accuracy": 0.14936443716287612,
      "num_tokens": 20995818.0,
      "step": 9700
    },
    {
      "entropy": 6.430025053024292,
      "epoch": 1.0385253357590025,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000490227604252187,
      "loss": 6.0207,
      "mean_token_accuracy": 0.14752765446901323,
      "num_tokens": 21006786.0,
      "step": 9705
    },
    {
      "entropy": 6.425715923309326,
      "epoch": 1.0390604098667666,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004902164556566433,
      "loss": 6.0898,
      "mean_token_accuracy": 0.1534474730491638,
      "num_tokens": 21017441.0,
      "step": 9710
    },
    {
      "entropy": 6.305364084243775,
      "epoch": 1.0395954839745305,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004902053008467546,
      "loss": 5.9568,
      "mean_token_accuracy": 0.1562404103577137,
      "num_tokens": 21027812.0,
      "step": 9715
    },
    {
      "entropy": 6.487801361083984,
      "epoch": 1.0401305580822944,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004901941398228431,
      "loss": 6.1407,
      "mean_token_accuracy": 0.1522221088409424,
      "num_tokens": 21038960.0,
      "step": 9720
    },
    {
      "entropy": 6.477895927429199,
      "epoch": 1.0406656321900583,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004901829725852311,
      "loss": 6.0886,
      "mean_token_accuracy": 0.15596613734960557,
      "num_tokens": 21050467.0,
      "step": 9725
    },
    {
      "entropy": 6.345766639709472,
      "epoch": 1.0412007062978224,
      "grad_norm": 0.84375,
      "learning_rate": 0.000490171799134241,
      "loss": 6.0066,
      "mean_token_accuracy": 0.14677918180823327,
      "num_tokens": 21061208.0,
      "step": 9730
    },
    {
      "entropy": 6.4059453964233395,
      "epoch": 1.0417357804055862,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004901606194701954,
      "loss": 6.1144,
      "mean_token_accuracy": 0.15061295330524443,
      "num_tokens": 21072872.0,
      "step": 9735
    },
    {
      "entropy": 6.514674520492553,
      "epoch": 1.04227085451335,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004901494335934172,
      "loss": 6.0688,
      "mean_token_accuracy": 0.15676568150520326,
      "num_tokens": 21084878.0,
      "step": 9740
    },
    {
      "entropy": 6.498925828933716,
      "epoch": 1.042805928621114,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004901382415042293,
      "loss": 6.1317,
      "mean_token_accuracy": 0.1467972993850708,
      "num_tokens": 21095289.0,
      "step": 9745
    },
    {
      "entropy": 6.465736389160156,
      "epoch": 1.0433410027288779,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004901270432029549,
      "loss": 6.0776,
      "mean_token_accuracy": 0.1480264335870743,
      "num_tokens": 21106077.0,
      "step": 9750
    },
    {
      "entropy": 6.356247425079346,
      "epoch": 1.043876076836642,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004901158386899174,
      "loss": 5.9934,
      "mean_token_accuracy": 0.1592910408973694,
      "num_tokens": 21116548.0,
      "step": 9755
    },
    {
      "entropy": 6.395184803009033,
      "epoch": 1.0444111509444058,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004901046279654402,
      "loss": 6.0233,
      "mean_token_accuracy": 0.16085042655467988,
      "num_tokens": 21126163.0,
      "step": 9760
    },
    {
      "entropy": 6.479514789581299,
      "epoch": 1.0449462250521697,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004900934110298471,
      "loss": 6.1383,
      "mean_token_accuracy": 0.14412727057933808,
      "num_tokens": 21136830.0,
      "step": 9765
    },
    {
      "entropy": 6.436767530441284,
      "epoch": 1.0454812991599336,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004900821878834621,
      "loss": 6.0621,
      "mean_token_accuracy": 0.15165750533342362,
      "num_tokens": 21146793.0,
      "step": 9770
    },
    {
      "entropy": 6.42450737953186,
      "epoch": 1.0460163732676975,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000490070958526609,
      "loss": 6.1306,
      "mean_token_accuracy": 0.14566373750567435,
      "num_tokens": 21156799.0,
      "step": 9775
    },
    {
      "entropy": 6.4154270648956295,
      "epoch": 1.0465514473754616,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004900597229596124,
      "loss": 6.1093,
      "mean_token_accuracy": 0.14481442868709565,
      "num_tokens": 21167757.0,
      "step": 9780
    },
    {
      "entropy": 6.4914103031158445,
      "epoch": 1.0470865214832255,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004900484811827963,
      "loss": 6.1016,
      "mean_token_accuracy": 0.15199369937181473,
      "num_tokens": 21178832.0,
      "step": 9785
    },
    {
      "entropy": 6.42375636100769,
      "epoch": 1.0476215955909893,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004900372331964856,
      "loss": 6.0196,
      "mean_token_accuracy": 0.15633775293827057,
      "num_tokens": 21189637.0,
      "step": 9790
    },
    {
      "entropy": 6.361067390441894,
      "epoch": 1.0481566696987532,
      "grad_norm": 0.890625,
      "learning_rate": 0.000490025979001005,
      "loss": 6.065,
      "mean_token_accuracy": 0.14836091548204422,
      "num_tokens": 21200194.0,
      "step": 9795
    },
    {
      "entropy": 6.371837711334228,
      "epoch": 1.0486917438065173,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004900147185966795,
      "loss": 6.0045,
      "mean_token_accuracy": 0.15117005556821822,
      "num_tokens": 21211405.0,
      "step": 9800
    },
    {
      "entropy": 6.438124752044677,
      "epoch": 1.0492268179142812,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004900034519838342,
      "loss": 6.0562,
      "mean_token_accuracy": 0.1450316861271858,
      "num_tokens": 21222722.0,
      "step": 9805
    },
    {
      "entropy": 6.408582401275635,
      "epoch": 1.049761892022045,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004899921791627945,
      "loss": 6.0049,
      "mean_token_accuracy": 0.1568140208721161,
      "num_tokens": 21233510.0,
      "step": 9810
    },
    {
      "entropy": 6.460711717605591,
      "epoch": 1.050296966129809,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004899809001338857,
      "loss": 6.1658,
      "mean_token_accuracy": 0.13992353528738022,
      "num_tokens": 21243235.0,
      "step": 9815
    },
    {
      "entropy": 6.425972032546997,
      "epoch": 1.0508320402375728,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004899696148974338,
      "loss": 6.057,
      "mean_token_accuracy": 0.15077909231185913,
      "num_tokens": 21254680.0,
      "step": 9820
    },
    {
      "entropy": 6.405041742324829,
      "epoch": 1.051367114345337,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004899583234537644,
      "loss": 6.0555,
      "mean_token_accuracy": 0.14797949194908142,
      "num_tokens": 21265116.0,
      "step": 9825
    },
    {
      "entropy": 6.473723125457764,
      "epoch": 1.0519021884531008,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004899470258032034,
      "loss": 6.03,
      "mean_token_accuracy": 0.1500900134444237,
      "num_tokens": 21277066.0,
      "step": 9830
    },
    {
      "entropy": 6.465215349197388,
      "epoch": 1.0524372625608647,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004899357219460775,
      "loss": 6.0982,
      "mean_token_accuracy": 0.14637623280286788,
      "num_tokens": 21287332.0,
      "step": 9835
    },
    {
      "entropy": 6.370449352264404,
      "epoch": 1.0529723366686286,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004899244118827128,
      "loss": 6.0862,
      "mean_token_accuracy": 0.15135086625814437,
      "num_tokens": 21298377.0,
      "step": 9840
    },
    {
      "entropy": 6.443383741378784,
      "epoch": 1.0535074107763924,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004899130956134358,
      "loss": 6.0712,
      "mean_token_accuracy": 0.1481868103146553,
      "num_tokens": 21309966.0,
      "step": 9845
    },
    {
      "entropy": 6.386915254592895,
      "epoch": 1.0540424848841565,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004899017731385735,
      "loss": 5.9877,
      "mean_token_accuracy": 0.15901345014572144,
      "num_tokens": 21320399.0,
      "step": 9850
    },
    {
      "entropy": 6.43333101272583,
      "epoch": 1.0545775589919204,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004898904444584527,
      "loss": 6.0977,
      "mean_token_accuracy": 0.1475136920809746,
      "num_tokens": 21331716.0,
      "step": 9855
    },
    {
      "entropy": 6.496586751937866,
      "epoch": 1.0551126330996843,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004898791095734004,
      "loss": 6.1151,
      "mean_token_accuracy": 0.1421336628496647,
      "num_tokens": 21341664.0,
      "step": 9860
    },
    {
      "entropy": 6.435134410858154,
      "epoch": 1.0556477072074482,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000489867768483744,
      "loss": 6.0717,
      "mean_token_accuracy": 0.14834583997726442,
      "num_tokens": 21352220.0,
      "step": 9865
    },
    {
      "entropy": 6.461541604995728,
      "epoch": 1.0561827813152123,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004898564211898111,
      "loss": 6.1819,
      "mean_token_accuracy": 0.14588122069835663,
      "num_tokens": 21363947.0,
      "step": 9870
    },
    {
      "entropy": 6.487367916107178,
      "epoch": 1.0567178554229761,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004898450676919291,
      "loss": 6.0469,
      "mean_token_accuracy": 0.15410142987966538,
      "num_tokens": 21374011.0,
      "step": 9875
    },
    {
      "entropy": 6.443288278579712,
      "epoch": 1.05725292953074,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000489833707990426,
      "loss": 6.1162,
      "mean_token_accuracy": 0.15177885442972183,
      "num_tokens": 21384232.0,
      "step": 9880
    },
    {
      "entropy": 6.388455009460449,
      "epoch": 1.057788003638504,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004898223420856298,
      "loss": 6.0906,
      "mean_token_accuracy": 0.14735191389918328,
      "num_tokens": 21395928.0,
      "step": 9885
    },
    {
      "entropy": 6.3939306259155275,
      "epoch": 1.0583230777462678,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004898109699778686,
      "loss": 6.0868,
      "mean_token_accuracy": 0.1475905641913414,
      "num_tokens": 21406391.0,
      "step": 9890
    },
    {
      "entropy": 6.423292970657348,
      "epoch": 1.0588581518540319,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000489799591667471,
      "loss": 6.1299,
      "mean_token_accuracy": 0.1456303209066391,
      "num_tokens": 21417859.0,
      "step": 9895
    },
    {
      "entropy": 6.407854080200195,
      "epoch": 1.0593932259617957,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004897882071547651,
      "loss": 5.9815,
      "mean_token_accuracy": 0.15492588877677918,
      "num_tokens": 21428284.0,
      "step": 9900
    },
    {
      "entropy": 6.412936401367188,
      "epoch": 1.0599283000695596,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004897768164400801,
      "loss": 6.133,
      "mean_token_accuracy": 0.14307866841554642,
      "num_tokens": 21439610.0,
      "step": 9905
    },
    {
      "entropy": 6.4674952030181885,
      "epoch": 1.0604633741773235,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004897654195237446,
      "loss": 6.1687,
      "mean_token_accuracy": 0.14330423772335052,
      "num_tokens": 21449319.0,
      "step": 9910
    },
    {
      "entropy": 6.486921739578247,
      "epoch": 1.0609984482850874,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004897540164060879,
      "loss": 6.1349,
      "mean_token_accuracy": 0.15205637887120246,
      "num_tokens": 21459851.0,
      "step": 9915
    },
    {
      "entropy": 6.390908050537109,
      "epoch": 1.0615335223928515,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004897426070874392,
      "loss": 6.0862,
      "mean_token_accuracy": 0.1552753746509552,
      "num_tokens": 21469717.0,
      "step": 9920
    },
    {
      "entropy": 6.456166362762451,
      "epoch": 1.0620685965006154,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004897311915681278,
      "loss": 6.1824,
      "mean_token_accuracy": 0.1430963508784771,
      "num_tokens": 21480815.0,
      "step": 9925
    },
    {
      "entropy": 6.519664478302002,
      "epoch": 1.0626036706083792,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004897197698484834,
      "loss": 6.0867,
      "mean_token_accuracy": 0.14963111132383347,
      "num_tokens": 21492000.0,
      "step": 9930
    },
    {
      "entropy": 6.427057313919067,
      "epoch": 1.0631387447161431,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004897083419288358,
      "loss": 6.0911,
      "mean_token_accuracy": 0.142834185808897,
      "num_tokens": 21502006.0,
      "step": 9935
    },
    {
      "entropy": 6.374684000015259,
      "epoch": 1.0636738188239072,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000489696907809515,
      "loss": 6.1082,
      "mean_token_accuracy": 0.1511937364935875,
      "num_tokens": 21513066.0,
      "step": 9940
    },
    {
      "entropy": 6.4087964534759525,
      "epoch": 1.064208892931671,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004896854674908512,
      "loss": 6.0962,
      "mean_token_accuracy": 0.14399367272853852,
      "num_tokens": 21524853.0,
      "step": 9945
    },
    {
      "entropy": 6.437889862060547,
      "epoch": 1.064743967039435,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004896740209731746,
      "loss": 6.0127,
      "mean_token_accuracy": 0.148801439255476,
      "num_tokens": 21536497.0,
      "step": 9950
    },
    {
      "entropy": 6.399822759628296,
      "epoch": 1.0652790411471988,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004896625682568159,
      "loss": 6.0553,
      "mean_token_accuracy": 0.15451411306858062,
      "num_tokens": 21546213.0,
      "step": 9955
    },
    {
      "entropy": 6.362114286422729,
      "epoch": 1.0658141152549627,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004896511093421057,
      "loss": 6.0244,
      "mean_token_accuracy": 0.1598359927535057,
      "num_tokens": 21556990.0,
      "step": 9960
    },
    {
      "entropy": 6.366931009292602,
      "epoch": 1.0663491893627268,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004896396442293749,
      "loss": 6.0059,
      "mean_token_accuracy": 0.15448964089155198,
      "num_tokens": 21567567.0,
      "step": 9965
    },
    {
      "entropy": 6.48778223991394,
      "epoch": 1.0668842634704907,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004896281729189544,
      "loss": 6.0588,
      "mean_token_accuracy": 0.15108865350484849,
      "num_tokens": 21577457.0,
      "step": 9970
    },
    {
      "entropy": 6.382377624511719,
      "epoch": 1.0674193375782546,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004896166954111758,
      "loss": 6.0482,
      "mean_token_accuracy": 0.15856654793024064,
      "num_tokens": 21588173.0,
      "step": 9975
    },
    {
      "entropy": 6.375876140594483,
      "epoch": 1.0679544116860185,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048960521170637,
      "loss": 6.0862,
      "mean_token_accuracy": 0.14578951001167298,
      "num_tokens": 21599154.0,
      "step": 9980
    },
    {
      "entropy": 6.412415409088135,
      "epoch": 1.0684894857937826,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004895937218048692,
      "loss": 6.1311,
      "mean_token_accuracy": 0.1443185731768608,
      "num_tokens": 21610134.0,
      "step": 9985
    },
    {
      "entropy": 6.413727903366089,
      "epoch": 1.0690245599015464,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004895822257070046,
      "loss": 6.0202,
      "mean_token_accuracy": 0.15281520187854766,
      "num_tokens": 21619989.0,
      "step": 9990
    },
    {
      "entropy": 6.424034118652344,
      "epoch": 1.0695596340093103,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004895707234131084,
      "loss": 6.0658,
      "mean_token_accuracy": 0.15340278148651124,
      "num_tokens": 21632707.0,
      "step": 9995
    },
    {
      "entropy": 6.424519348144531,
      "epoch": 1.0700947081170742,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004895592149235128,
      "loss": 6.0449,
      "mean_token_accuracy": 0.15269885063171387,
      "num_tokens": 21643001.0,
      "step": 10000
    },
    {
      "entropy": 6.446679639816284,
      "epoch": 1.070629782224838,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048954770023855,
      "loss": 6.1167,
      "mean_token_accuracy": 0.14452026709914206,
      "num_tokens": 21653425.0,
      "step": 10005
    },
    {
      "entropy": 6.468544816970825,
      "epoch": 1.0711648563326022,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004895361793585525,
      "loss": 6.1356,
      "mean_token_accuracy": 0.14641227424144745,
      "num_tokens": 21664212.0,
      "step": 10010
    },
    {
      "entropy": 6.419103956222534,
      "epoch": 1.071699930440366,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004895246522838529,
      "loss": 6.1246,
      "mean_token_accuracy": 0.15267803221940995,
      "num_tokens": 21675564.0,
      "step": 10015
    },
    {
      "entropy": 6.404095554351807,
      "epoch": 1.07223500454813,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004895131190147842,
      "loss": 6.1111,
      "mean_token_accuracy": 0.14853642433881759,
      "num_tokens": 21686225.0,
      "step": 10020
    },
    {
      "entropy": 6.431006193161011,
      "epoch": 1.0727700786558938,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004895015795516793,
      "loss": 6.0086,
      "mean_token_accuracy": 0.1532626762986183,
      "num_tokens": 21697479.0,
      "step": 10025
    },
    {
      "entropy": 6.460914754867554,
      "epoch": 1.0733051527636577,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004894900338948714,
      "loss": 6.1548,
      "mean_token_accuracy": 0.14519525989890097,
      "num_tokens": 21708496.0,
      "step": 10030
    },
    {
      "entropy": 6.419927406311035,
      "epoch": 1.0738402268714218,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004894784820446939,
      "loss": 6.0087,
      "mean_token_accuracy": 0.1560596451163292,
      "num_tokens": 21718290.0,
      "step": 10035
    },
    {
      "entropy": 6.354112815856934,
      "epoch": 1.0743753009791857,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004894669240014804,
      "loss": 6.0096,
      "mean_token_accuracy": 0.15154567062854768,
      "num_tokens": 21729185.0,
      "step": 10040
    },
    {
      "entropy": 6.408800506591797,
      "epoch": 1.0749103750869495,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004894553597655646,
      "loss": 6.0823,
      "mean_token_accuracy": 0.14725847393274308,
      "num_tokens": 21740787.0,
      "step": 10045
    },
    {
      "entropy": 6.49581823348999,
      "epoch": 1.0754454491947134,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004894437893372804,
      "loss": 6.1329,
      "mean_token_accuracy": 0.14735653549432753,
      "num_tokens": 21751804.0,
      "step": 10050
    },
    {
      "entropy": 6.492033767700195,
      "epoch": 1.0759805233024773,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000489432212716962,
      "loss": 6.0621,
      "mean_token_accuracy": 0.1503622516989708,
      "num_tokens": 21762730.0,
      "step": 10055
    },
    {
      "entropy": 6.391907262802124,
      "epoch": 1.0765155974102414,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004894206299049436,
      "loss": 6.0812,
      "mean_token_accuracy": 0.1500949591398239,
      "num_tokens": 21773470.0,
      "step": 10060
    },
    {
      "entropy": 6.43964581489563,
      "epoch": 1.0770506715180053,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004894090409015595,
      "loss": 6.0881,
      "mean_token_accuracy": 0.13947335481643677,
      "num_tokens": 21783385.0,
      "step": 10065
    },
    {
      "entropy": 6.3588048934936525,
      "epoch": 1.0775857456257691,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004893974457071445,
      "loss": 6.0945,
      "mean_token_accuracy": 0.14815742075443267,
      "num_tokens": 21794857.0,
      "step": 10070
    },
    {
      "entropy": 6.442140579223633,
      "epoch": 1.078120819733533,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004893858443220335,
      "loss": 6.1031,
      "mean_token_accuracy": 0.14545661434531212,
      "num_tokens": 21804296.0,
      "step": 10075
    },
    {
      "entropy": 6.443773603439331,
      "epoch": 1.0786558938412971,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004893742367465613,
      "loss": 6.0904,
      "mean_token_accuracy": 0.1503361366689205,
      "num_tokens": 21815899.0,
      "step": 10080
    },
    {
      "entropy": 6.4232861518859865,
      "epoch": 1.079190967949061,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004893626229810631,
      "loss": 6.0718,
      "mean_token_accuracy": 0.15055726319551468,
      "num_tokens": 21825964.0,
      "step": 10085
    },
    {
      "entropy": 6.457719850540161,
      "epoch": 1.0797260420568249,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004893510030258743,
      "loss": 6.1641,
      "mean_token_accuracy": 0.14678804129362105,
      "num_tokens": 21836656.0,
      "step": 10090
    },
    {
      "entropy": 6.433513021469116,
      "epoch": 1.0802611161645888,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004893393768813305,
      "loss": 6.1515,
      "mean_token_accuracy": 0.14401523768901825,
      "num_tokens": 21847658.0,
      "step": 10095
    },
    {
      "entropy": 6.4721534729003904,
      "epoch": 1.0807961902723526,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004893277445477673,
      "loss": 6.1345,
      "mean_token_accuracy": 0.14163527637720108,
      "num_tokens": 21858338.0,
      "step": 10100
    },
    {
      "entropy": 6.404046821594238,
      "epoch": 1.0813312643801167,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004893161060255206,
      "loss": 5.9835,
      "mean_token_accuracy": 0.15740612298250198,
      "num_tokens": 21870656.0,
      "step": 10105
    },
    {
      "entropy": 6.417188835144043,
      "epoch": 1.0818663384878806,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004893044613149263,
      "loss": 6.0869,
      "mean_token_accuracy": 0.15112882405519484,
      "num_tokens": 21881806.0,
      "step": 10110
    },
    {
      "entropy": 6.330238628387451,
      "epoch": 1.0824014125956445,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004892928104163209,
      "loss": 6.0246,
      "mean_token_accuracy": 0.14852803498506545,
      "num_tokens": 21892277.0,
      "step": 10115
    },
    {
      "entropy": 6.3847626686096195,
      "epoch": 1.0829364867034084,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004892811533300407,
      "loss": 6.0137,
      "mean_token_accuracy": 0.15253811702132225,
      "num_tokens": 21903509.0,
      "step": 10120
    },
    {
      "entropy": 6.470606470108033,
      "epoch": 1.0834715608111725,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004892694900564223,
      "loss": 6.1331,
      "mean_token_accuracy": 0.14718897119164467,
      "num_tokens": 21914308.0,
      "step": 10125
    },
    {
      "entropy": 6.351409244537353,
      "epoch": 1.0840066349189363,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004892578205958025,
      "loss": 6.0235,
      "mean_token_accuracy": 0.15656714588403703,
      "num_tokens": 21924302.0,
      "step": 10130
    },
    {
      "entropy": 6.44743242263794,
      "epoch": 1.0845417090267002,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004892461449485182,
      "loss": 6.1016,
      "mean_token_accuracy": 0.14411793947219848,
      "num_tokens": 21935066.0,
      "step": 10135
    },
    {
      "entropy": 6.445598173141479,
      "epoch": 1.085076783134464,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004892344631149066,
      "loss": 6.0044,
      "mean_token_accuracy": 0.1540689177811146,
      "num_tokens": 21945798.0,
      "step": 10140
    },
    {
      "entropy": 6.357674694061279,
      "epoch": 1.085611857242228,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000489222775095305,
      "loss": 6.0905,
      "mean_token_accuracy": 0.15099227130413057,
      "num_tokens": 21955667.0,
      "step": 10145
    },
    {
      "entropy": 6.365167808532715,
      "epoch": 1.086146931349992,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004892110808900509,
      "loss": 6.0678,
      "mean_token_accuracy": 0.1491195097565651,
      "num_tokens": 21966107.0,
      "step": 10150
    },
    {
      "entropy": 6.422029495239258,
      "epoch": 1.086682005457756,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004891993804994818,
      "loss": 6.0706,
      "mean_token_accuracy": 0.14995527714490892,
      "num_tokens": 21976408.0,
      "step": 10155
    },
    {
      "entropy": 6.5203235149383545,
      "epoch": 1.0872170795655198,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004891876739239358,
      "loss": 6.0732,
      "mean_token_accuracy": 0.1514231041073799,
      "num_tokens": 21986481.0,
      "step": 10160
    },
    {
      "entropy": 6.466368055343628,
      "epoch": 1.0877521536732837,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004891759611637508,
      "loss": 6.1071,
      "mean_token_accuracy": 0.14710880890488626,
      "num_tokens": 21997554.0,
      "step": 10165
    },
    {
      "entropy": 6.509916305541992,
      "epoch": 1.0882872277810476,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000489164242219265,
      "loss": 6.1564,
      "mean_token_accuracy": 0.143141470849514,
      "num_tokens": 22009048.0,
      "step": 10170
    },
    {
      "entropy": 6.477291011810303,
      "epoch": 1.0888223018888117,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004891525170908169,
      "loss": 6.1669,
      "mean_token_accuracy": 0.14469189792871476,
      "num_tokens": 22020625.0,
      "step": 10175
    },
    {
      "entropy": 6.447618246078491,
      "epoch": 1.0893573759965756,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004891407857787448,
      "loss": 6.0386,
      "mean_token_accuracy": 0.14259916320443153,
      "num_tokens": 22032298.0,
      "step": 10180
    },
    {
      "entropy": 6.395538091659546,
      "epoch": 1.0898924501043394,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004891290482833876,
      "loss": 6.0487,
      "mean_token_accuracy": 0.15133269876241684,
      "num_tokens": 22043693.0,
      "step": 10185
    },
    {
      "entropy": 6.3877575397491455,
      "epoch": 1.0904275242121033,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004891173046050844,
      "loss": 6.1122,
      "mean_token_accuracy": 0.1529858522117138,
      "num_tokens": 22054373.0,
      "step": 10190
    },
    {
      "entropy": 6.412981939315796,
      "epoch": 1.0909625983198672,
      "grad_norm": 1.015625,
      "learning_rate": 0.000489105554744174,
      "loss": 6.0573,
      "mean_token_accuracy": 0.15100773870944978,
      "num_tokens": 22065039.0,
      "step": 10195
    },
    {
      "entropy": 6.464700984954834,
      "epoch": 1.0914976724276313,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004890937987009958,
      "loss": 6.0875,
      "mean_token_accuracy": 0.14925315231084824,
      "num_tokens": 22077003.0,
      "step": 10200
    },
    {
      "entropy": 6.471652984619141,
      "epoch": 1.0920327465353952,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004890820364758892,
      "loss": 6.0759,
      "mean_token_accuracy": 0.14644762724637986,
      "num_tokens": 22088558.0,
      "step": 10205
    },
    {
      "entropy": 6.485191440582275,
      "epoch": 1.092567820643159,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000489070268069194,
      "loss": 6.0541,
      "mean_token_accuracy": 0.14932452589273454,
      "num_tokens": 22097777.0,
      "step": 10210
    },
    {
      "entropy": 6.432035827636719,
      "epoch": 1.093102894750923,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004890584934812498,
      "loss": 5.9858,
      "mean_token_accuracy": 0.15041119903326033,
      "num_tokens": 22107803.0,
      "step": 10215
    },
    {
      "entropy": 6.40109224319458,
      "epoch": 1.093637968858687,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004890467127123967,
      "loss": 6.0063,
      "mean_token_accuracy": 0.15668996125459672,
      "num_tokens": 22117965.0,
      "step": 10220
    },
    {
      "entropy": 6.4437566757202145,
      "epoch": 1.094173042966451,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004890349257629749,
      "loss": 6.0522,
      "mean_token_accuracy": 0.1525336742401123,
      "num_tokens": 22127759.0,
      "step": 10225
    },
    {
      "entropy": 6.387381219863892,
      "epoch": 1.0947081170742148,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004890231326333246,
      "loss": 6.1065,
      "mean_token_accuracy": 0.14486921578645706,
      "num_tokens": 22138477.0,
      "step": 10230
    },
    {
      "entropy": 6.411555576324463,
      "epoch": 1.0952431911819787,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004890113333237865,
      "loss": 6.071,
      "mean_token_accuracy": 0.1488646924495697,
      "num_tokens": 22149568.0,
      "step": 10235
    },
    {
      "entropy": 6.439895296096802,
      "epoch": 1.0957782652897425,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004889995278347012,
      "loss": 6.0563,
      "mean_token_accuracy": 0.15043437480926514,
      "num_tokens": 22159949.0,
      "step": 10240
    },
    {
      "entropy": 6.364970636367798,
      "epoch": 1.0963133393975066,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004889877161664096,
      "loss": 6.0521,
      "mean_token_accuracy": 0.148412124812603,
      "num_tokens": 22170115.0,
      "step": 10245
    },
    {
      "entropy": 6.417066431045532,
      "epoch": 1.0968484135052705,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004889758983192528,
      "loss": 6.0865,
      "mean_token_accuracy": 0.1490737184882164,
      "num_tokens": 22181323.0,
      "step": 10250
    },
    {
      "entropy": 6.410627937316894,
      "epoch": 1.0973834876130344,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004889640742935719,
      "loss": 6.0397,
      "mean_token_accuracy": 0.14545978605747223,
      "num_tokens": 22192078.0,
      "step": 10255
    },
    {
      "entropy": 6.476579809188843,
      "epoch": 1.0979185617207983,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004889522440897085,
      "loss": 6.1118,
      "mean_token_accuracy": 0.14848758429288864,
      "num_tokens": 22203211.0,
      "step": 10260
    },
    {
      "entropy": 6.452811002731323,
      "epoch": 1.0984536358285624,
      "grad_norm": 0.875,
      "learning_rate": 0.0004889404077080041,
      "loss": 6.042,
      "mean_token_accuracy": 0.1454724684357643,
      "num_tokens": 22213487.0,
      "step": 10265
    },
    {
      "entropy": 6.442213916778565,
      "epoch": 1.0989887099363262,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004889285651488005,
      "loss": 6.0716,
      "mean_token_accuracy": 0.14339534267783166,
      "num_tokens": 22223717.0,
      "step": 10270
    },
    {
      "entropy": 6.373847484588623,
      "epoch": 1.0995237840440901,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004889167164124396,
      "loss": 6.0839,
      "mean_token_accuracy": 0.15240889191627502,
      "num_tokens": 22235141.0,
      "step": 10275
    },
    {
      "entropy": 6.4567694664001465,
      "epoch": 1.100058858151854,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004889048614992636,
      "loss": 6.1523,
      "mean_token_accuracy": 0.14074449613690376,
      "num_tokens": 22246601.0,
      "step": 10280
    },
    {
      "entropy": 6.428856897354126,
      "epoch": 1.1005939322596179,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004888930004096148,
      "loss": 6.0853,
      "mean_token_accuracy": 0.14496622681617738,
      "num_tokens": 22257206.0,
      "step": 10285
    },
    {
      "entropy": 6.458989429473877,
      "epoch": 1.101129006367382,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004888811331438356,
      "loss": 6.1451,
      "mean_token_accuracy": 0.13707782328128815,
      "num_tokens": 22267799.0,
      "step": 10290
    },
    {
      "entropy": 6.4701355457305905,
      "epoch": 1.1016640804751459,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004888692597022689,
      "loss": 6.1922,
      "mean_token_accuracy": 0.1472000613808632,
      "num_tokens": 22279368.0,
      "step": 10295
    },
    {
      "entropy": 6.39345350265503,
      "epoch": 1.1021991545829097,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004888573800852572,
      "loss": 5.919,
      "mean_token_accuracy": 0.15674495324492455,
      "num_tokens": 22289998.0,
      "step": 10300
    },
    {
      "entropy": 6.32203311920166,
      "epoch": 1.1027342286906736,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004888454942931438,
      "loss": 6.0482,
      "mean_token_accuracy": 0.16183083951473237,
      "num_tokens": 22300058.0,
      "step": 10305
    },
    {
      "entropy": 6.457044649124145,
      "epoch": 1.1032693027984375,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004888336023262718,
      "loss": 6.1517,
      "mean_token_accuracy": 0.14678664803504943,
      "num_tokens": 22310962.0,
      "step": 10310
    },
    {
      "entropy": 6.436636114120484,
      "epoch": 1.1038043769062016,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004888217041849846,
      "loss": 6.0238,
      "mean_token_accuracy": 0.15562547743320465,
      "num_tokens": 22321526.0,
      "step": 10315
    },
    {
      "entropy": 6.423198986053467,
      "epoch": 1.1043394510139655,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004888097998696256,
      "loss": 6.1058,
      "mean_token_accuracy": 0.14754745438694955,
      "num_tokens": 22332558.0,
      "step": 10320
    },
    {
      "entropy": 6.424881505966186,
      "epoch": 1.1048745251217293,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004887978893805387,
      "loss": 6.064,
      "mean_token_accuracy": 0.15085040926933288,
      "num_tokens": 22342392.0,
      "step": 10325
    },
    {
      "entropy": 6.411435127258301,
      "epoch": 1.1054095992294932,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004887859727180679,
      "loss": 6.1035,
      "mean_token_accuracy": 0.14303725510835646,
      "num_tokens": 22353089.0,
      "step": 10330
    },
    {
      "entropy": 6.399058723449707,
      "epoch": 1.105944673337257,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004887740498825572,
      "loss": 6.0332,
      "mean_token_accuracy": 0.15439766645431519,
      "num_tokens": 22364314.0,
      "step": 10335
    },
    {
      "entropy": 6.3719195365905765,
      "epoch": 1.1064797474450212,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004887621208743507,
      "loss": 6.0441,
      "mean_token_accuracy": 0.15256350636482238,
      "num_tokens": 22374985.0,
      "step": 10340
    },
    {
      "entropy": 6.388993549346924,
      "epoch": 1.107014821552785,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004887501856937932,
      "loss": 5.9572,
      "mean_token_accuracy": 0.1588325172662735,
      "num_tokens": 22385618.0,
      "step": 10345
    },
    {
      "entropy": 6.455599069595337,
      "epoch": 1.107549895660549,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000488738244341229,
      "loss": 6.0636,
      "mean_token_accuracy": 0.15322687849402428,
      "num_tokens": 22395871.0,
      "step": 10350
    },
    {
      "entropy": 6.426832389831543,
      "epoch": 1.1080849697683128,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000488726296817003,
      "loss": 6.0977,
      "mean_token_accuracy": 0.15314423590898513,
      "num_tokens": 22406707.0,
      "step": 10355
    },
    {
      "entropy": 6.438868474960327,
      "epoch": 1.108620043876077,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004887143431214602,
      "loss": 6.1008,
      "mean_token_accuracy": 0.1444241926074028,
      "num_tokens": 22417543.0,
      "step": 10360
    },
    {
      "entropy": 6.359684562683105,
      "epoch": 1.1091551179838408,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004887023832549459,
      "loss": 6.0424,
      "mean_token_accuracy": 0.1540512979030609,
      "num_tokens": 22428283.0,
      "step": 10365
    },
    {
      "entropy": 6.362619781494141,
      "epoch": 1.1096901920916047,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004886904172178052,
      "loss": 6.0205,
      "mean_token_accuracy": 0.15236841067671775,
      "num_tokens": 22439860.0,
      "step": 10370
    },
    {
      "entropy": 6.484773254394531,
      "epoch": 1.1102252661993686,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004886784450103838,
      "loss": 6.0894,
      "mean_token_accuracy": 0.15454527288675307,
      "num_tokens": 22451250.0,
      "step": 10375
    },
    {
      "entropy": 6.375313425064087,
      "epoch": 1.1107603403071324,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004886664666330273,
      "loss": 6.0382,
      "mean_token_accuracy": 0.1503927528858185,
      "num_tokens": 22461112.0,
      "step": 10380
    },
    {
      "entropy": 6.4391029357910154,
      "epoch": 1.1112954144148965,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004886544820860816,
      "loss": 6.0526,
      "mean_token_accuracy": 0.15394495874643327,
      "num_tokens": 22471368.0,
      "step": 10385
    },
    {
      "entropy": 6.4377665519714355,
      "epoch": 1.1118304885226604,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004886424913698927,
      "loss": 6.0507,
      "mean_token_accuracy": 0.14631543010473252,
      "num_tokens": 22482011.0,
      "step": 10390
    },
    {
      "entropy": 6.368098020553589,
      "epoch": 1.1123655626304243,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004886304944848069,
      "loss": 6.0133,
      "mean_token_accuracy": 0.1588986910879612,
      "num_tokens": 22493648.0,
      "step": 10395
    },
    {
      "entropy": 6.457787036895752,
      "epoch": 1.1129006367381882,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004886184914311706,
      "loss": 6.1282,
      "mean_token_accuracy": 0.1451903462409973,
      "num_tokens": 22505577.0,
      "step": 10400
    },
    {
      "entropy": 6.44191575050354,
      "epoch": 1.1134357108459523,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004886064822093305,
      "loss": 6.0945,
      "mean_token_accuracy": 0.1409159407019615,
      "num_tokens": 22515546.0,
      "step": 10405
    },
    {
      "entropy": 6.382582473754883,
      "epoch": 1.1139707849537162,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004885944668196332,
      "loss": 5.9868,
      "mean_token_accuracy": 0.15967881977558135,
      "num_tokens": 22526710.0,
      "step": 10410
    },
    {
      "entropy": 6.468086194992066,
      "epoch": 1.11450585906148,
      "grad_norm": 1.0,
      "learning_rate": 0.0004885824452624254,
      "loss": 6.1175,
      "mean_token_accuracy": 0.14681168496608735,
      "num_tokens": 22537809.0,
      "step": 10415
    },
    {
      "entropy": 6.458578300476074,
      "epoch": 1.115040933169244,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004885704175380548,
      "loss": 6.0995,
      "mean_token_accuracy": 0.14564359188079834,
      "num_tokens": 22549301.0,
      "step": 10420
    },
    {
      "entropy": 6.446738767623901,
      "epoch": 1.1155760072770078,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004885583836468683,
      "loss": 5.9882,
      "mean_token_accuracy": 0.15414031744003295,
      "num_tokens": 22560483.0,
      "step": 10425
    },
    {
      "entropy": 6.396184062957763,
      "epoch": 1.1161110813847719,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004885463435892136,
      "loss": 6.1006,
      "mean_token_accuracy": 0.14650348275899888,
      "num_tokens": 22570524.0,
      "step": 10430
    },
    {
      "entropy": 6.425270080566406,
      "epoch": 1.1166461554925358,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000488534297365438,
      "loss": 6.1184,
      "mean_token_accuracy": 0.14784495830535888,
      "num_tokens": 22581120.0,
      "step": 10435
    },
    {
      "entropy": 6.464936017990112,
      "epoch": 1.1171812296002996,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004885222449758899,
      "loss": 6.0867,
      "mean_token_accuracy": 0.14541475996375083,
      "num_tokens": 22592658.0,
      "step": 10440
    },
    {
      "entropy": 6.46417555809021,
      "epoch": 1.1177163037080635,
      "grad_norm": 1.0,
      "learning_rate": 0.0004885101864209167,
      "loss": 6.0807,
      "mean_token_accuracy": 0.14666630178689957,
      "num_tokens": 22603497.0,
      "step": 10445
    },
    {
      "entropy": 6.388909769058228,
      "epoch": 1.1182513778158274,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000488498121700867,
      "loss": 5.9764,
      "mean_token_accuracy": 0.15548633486032487,
      "num_tokens": 22612916.0,
      "step": 10450
    },
    {
      "entropy": 6.3842089653015135,
      "epoch": 1.1187864519235915,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004884860508160891,
      "loss": 6.073,
      "mean_token_accuracy": 0.1555374503135681,
      "num_tokens": 22625345.0,
      "step": 10455
    },
    {
      "entropy": 6.397184944152832,
      "epoch": 1.1193215260313554,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004884739737669314,
      "loss": 5.9736,
      "mean_token_accuracy": 0.15965810418128967,
      "num_tokens": 22637000.0,
      "step": 10460
    },
    {
      "entropy": 6.405105781555176,
      "epoch": 1.1198566001391193,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004884618905537427,
      "loss": 6.0148,
      "mean_token_accuracy": 0.15448416769504547,
      "num_tokens": 22647025.0,
      "step": 10465
    },
    {
      "entropy": 6.376295137405395,
      "epoch": 1.1203916742468831,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004884498011768719,
      "loss": 5.9948,
      "mean_token_accuracy": 0.1556273430585861,
      "num_tokens": 22657253.0,
      "step": 10470
    },
    {
      "entropy": 6.373455572128296,
      "epoch": 1.120926748354647,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000488437705636668,
      "loss": 6.0346,
      "mean_token_accuracy": 0.15148885846138,
      "num_tokens": 22668432.0,
      "step": 10475
    },
    {
      "entropy": 6.403218126296997,
      "epoch": 1.121461822462411,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004884256039334805,
      "loss": 6.0914,
      "mean_token_accuracy": 0.14390705525875092,
      "num_tokens": 22678538.0,
      "step": 10480
    },
    {
      "entropy": 6.392944288253784,
      "epoch": 1.121996896570175,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004884134960676586,
      "loss": 6.059,
      "mean_token_accuracy": 0.14626166820526124,
      "num_tokens": 22688994.0,
      "step": 10485
    },
    {
      "entropy": 6.433700513839722,
      "epoch": 1.1225319706779389,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000488401382039552,
      "loss": 6.0864,
      "mean_token_accuracy": 0.15349995493888854,
      "num_tokens": 22699584.0,
      "step": 10490
    },
    {
      "entropy": 6.4293218612670895,
      "epoch": 1.1230670447857027,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004883892618495104,
      "loss": 5.9898,
      "mean_token_accuracy": 0.15138272643089296,
      "num_tokens": 22710093.0,
      "step": 10495
    },
    {
      "entropy": 6.431663274765015,
      "epoch": 1.1236021188934668,
      "grad_norm": 0.921875,
      "learning_rate": 0.000488377135497884,
      "loss": 6.0949,
      "mean_token_accuracy": 0.147957893460989,
      "num_tokens": 22720533.0,
      "step": 10500
    },
    {
      "entropy": 6.3571075916290285,
      "epoch": 1.1241371930012307,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004883650029850226,
      "loss": 5.9872,
      "mean_token_accuracy": 0.15093840211629866,
      "num_tokens": 22731010.0,
      "step": 10505
    },
    {
      "entropy": 6.368546438217163,
      "epoch": 1.1246722671089946,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004883528643112769,
      "loss": 6.0575,
      "mean_token_accuracy": 0.1497967891395092,
      "num_tokens": 22743045.0,
      "step": 10510
    },
    {
      "entropy": 6.415064191818237,
      "epoch": 1.1252073412167585,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004883407194769972,
      "loss": 6.0513,
      "mean_token_accuracy": 0.15047580301761626,
      "num_tokens": 22753746.0,
      "step": 10515
    },
    {
      "entropy": 6.439858722686767,
      "epoch": 1.1257424153245223,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004883285684825342,
      "loss": 6.0659,
      "mean_token_accuracy": 0.15383292734622955,
      "num_tokens": 22765264.0,
      "step": 10520
    },
    {
      "entropy": 6.458159303665161,
      "epoch": 1.1262774894322864,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004883164113282386,
      "loss": 6.0528,
      "mean_token_accuracy": 0.15717657506465912,
      "num_tokens": 22775727.0,
      "step": 10525
    },
    {
      "entropy": 6.441124820709229,
      "epoch": 1.1268125635400503,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004883042480144618,
      "loss": 6.0573,
      "mean_token_accuracy": 0.15283898413181304,
      "num_tokens": 22787228.0,
      "step": 10530
    },
    {
      "entropy": 6.470313501358032,
      "epoch": 1.1273476376478142,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004882920785415547,
      "loss": 6.1206,
      "mean_token_accuracy": 0.14563775211572647,
      "num_tokens": 22797467.0,
      "step": 10535
    },
    {
      "entropy": 6.408564233779908,
      "epoch": 1.127882711755578,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004882799029098689,
      "loss": 6.0931,
      "mean_token_accuracy": 0.14806491881608963,
      "num_tokens": 22808281.0,
      "step": 10540
    },
    {
      "entropy": 6.4451995372772215,
      "epoch": 1.1284177858633422,
      "grad_norm": 1.0,
      "learning_rate": 0.00048826772111975583,
      "loss": 6.1324,
      "mean_token_accuracy": 0.15112519264221191,
      "num_tokens": 22819387.0,
      "step": 10545
    },
    {
      "entropy": 6.424782037734985,
      "epoch": 1.128952859971106,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048825553317156717,
      "loss": 6.09,
      "mean_token_accuracy": 0.15004085302352904,
      "num_tokens": 22829649.0,
      "step": 10550
    },
    {
      "entropy": 6.350788974761963,
      "epoch": 1.12948793407887,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004882433390656551,
      "loss": 6.0794,
      "mean_token_accuracy": 0.1528400808572769,
      "num_tokens": 22840667.0,
      "step": 10555
    },
    {
      "entropy": 6.4204323291778564,
      "epoch": 1.1300230081866338,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004882311388023715,
      "loss": 6.035,
      "mean_token_accuracy": 0.15133709609508514,
      "num_tokens": 22852461.0,
      "step": 10560
    },
    {
      "entropy": 6.472927904129028,
      "epoch": 1.1305580822943977,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004882189323820688,
      "loss": 6.1201,
      "mean_token_accuracy": 0.13897055983543397,
      "num_tokens": 22862781.0,
      "step": 10565
    },
    {
      "entropy": 6.485547780990601,
      "epoch": 1.1310931564021618,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048820671980509935,
      "loss": 6.123,
      "mean_token_accuracy": 0.15063353180885314,
      "num_tokens": 22873393.0,
      "step": 10570
    },
    {
      "entropy": 6.483258581161499,
      "epoch": 1.1316282305099257,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048819450107181583,
      "loss": 6.091,
      "mean_token_accuracy": 0.1463221549987793,
      "num_tokens": 22883931.0,
      "step": 10575
    },
    {
      "entropy": 6.396607780456543,
      "epoch": 1.1321633046176895,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004881822761825711,
      "loss": 6.048,
      "mean_token_accuracy": 0.14585273265838622,
      "num_tokens": 22894829.0,
      "step": 10580
    },
    {
      "entropy": 6.385651302337647,
      "epoch": 1.1326983787254534,
      "grad_norm": 0.875,
      "learning_rate": 0.000488170045137718,
      "loss": 6.0762,
      "mean_token_accuracy": 0.14833278059959412,
      "num_tokens": 22906059.0,
      "step": 10585
    },
    {
      "entropy": 6.431635522842408,
      "epoch": 1.1332334528332173,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004881578079376099,
      "loss": 6.0781,
      "mean_token_accuracy": 0.14366692900657654,
      "num_tokens": 22917834.0,
      "step": 10590
    },
    {
      "entropy": 6.409401702880859,
      "epoch": 1.1337685269409814,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048814556458259996,
      "loss": 5.9803,
      "mean_token_accuracy": 0.16076456308364867,
      "num_tokens": 22929157.0,
      "step": 10595
    },
    {
      "entropy": 6.478912734985352,
      "epoch": 1.1343036010487453,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004881333150730419,
      "loss": 6.116,
      "mean_token_accuracy": 0.14745848029851913,
      "num_tokens": 22939396.0,
      "step": 10600
    },
    {
      "entropy": 6.427004337310791,
      "epoch": 1.1348386751565092,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048812105940928917,
      "loss": 6.0744,
      "mean_token_accuracy": 0.15069807916879654,
      "num_tokens": 22950253.0,
      "step": 10605
    },
    {
      "entropy": 6.419635248184204,
      "epoch": 1.135373749264273,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048810879759169593,
      "loss": 6.099,
      "mean_token_accuracy": 0.14996283501386642,
      "num_tokens": 22960417.0,
      "step": 10610
    },
    {
      "entropy": 6.3999100685119625,
      "epoch": 1.135908823372037,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000488096529620616,
      "loss": 6.0661,
      "mean_token_accuracy": 0.1451982378959656,
      "num_tokens": 22971763.0,
      "step": 10615
    },
    {
      "entropy": 6.4945347785949705,
      "epoch": 1.136443897479801,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048808425549640383,
      "loss": 6.0757,
      "mean_token_accuracy": 0.14900879114866256,
      "num_tokens": 22981571.0,
      "step": 10620
    },
    {
      "entropy": 6.428467798233032,
      "epoch": 1.136978971587565,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048807197521941366,
      "loss": 6.0522,
      "mean_token_accuracy": 0.15346492528915406,
      "num_tokens": 22991644.0,
      "step": 10625
    },
    {
      "entropy": 6.344451189041138,
      "epoch": 1.1375140456953288,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004880596887900002,
      "loss": 6.0046,
      "mean_token_accuracy": 0.156754033267498,
      "num_tokens": 23001557.0,
      "step": 10630
    },
    {
      "entropy": 6.427844762802124,
      "epoch": 1.1380491198030926,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048804739620851806,
      "loss": 6.1005,
      "mean_token_accuracy": 0.1445825845003128,
      "num_tokens": 23012715.0,
      "step": 10635
    },
    {
      "entropy": 6.430317831039429,
      "epoch": 1.1385841939108567,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048803509747532235,
      "loss": 6.0879,
      "mean_token_accuracy": 0.14882653206586838,
      "num_tokens": 23023016.0,
      "step": 10640
    },
    {
      "entropy": 6.435520839691162,
      "epoch": 1.1391192680186206,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004880227925907682,
      "loss": 6.0827,
      "mean_token_accuracy": 0.1496237635612488,
      "num_tokens": 23033511.0,
      "step": 10645
    },
    {
      "entropy": 6.418369150161743,
      "epoch": 1.1396543421263845,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004880104815552108,
      "loss": 6.0801,
      "mean_token_accuracy": 0.1548319473862648,
      "num_tokens": 23045149.0,
      "step": 10650
    },
    {
      "entropy": 6.367843151092529,
      "epoch": 1.1401894162341484,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004879981643690056,
      "loss": 6.0415,
      "mean_token_accuracy": 0.15230660885572433,
      "num_tokens": 23054489.0,
      "step": 10655
    },
    {
      "entropy": 6.434671449661255,
      "epoch": 1.1407244903419125,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048798584103250847,
      "loss": 6.1886,
      "mean_token_accuracy": 0.14471326768398285,
      "num_tokens": 23064725.0,
      "step": 10660
    },
    {
      "entropy": 6.451401567459106,
      "epoch": 1.1412595644496764,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000487973511546075,
      "loss": 6.1252,
      "mean_token_accuracy": 0.14657393246889114,
      "num_tokens": 23076473.0,
      "step": 10665
    },
    {
      "entropy": 6.446539640426636,
      "epoch": 1.1417946385574402,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004879611759100614,
      "loss": 6.0472,
      "mean_token_accuracy": 0.15967498272657393,
      "num_tokens": 23087449.0,
      "step": 10670
    },
    {
      "entropy": 6.452880430221557,
      "epoch": 1.1423297126652041,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004879488341248237,
      "loss": 6.0214,
      "mean_token_accuracy": 0.1504603147506714,
      "num_tokens": 23098284.0,
      "step": 10675
    },
    {
      "entropy": 6.3784255027771,
      "epoch": 1.142864786772968,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048793648619071834,
      "loss": 6.0244,
      "mean_token_accuracy": 0.15487734377384185,
      "num_tokens": 23108545.0,
      "step": 10680
    },
    {
      "entropy": 6.412152194976807,
      "epoch": 1.143399860880732,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004879241321081019,
      "loss": 6.0049,
      "mean_token_accuracy": 0.1534179724752903,
      "num_tokens": 23118568.0,
      "step": 10685
    },
    {
      "entropy": 6.43848557472229,
      "epoch": 1.143934934988496,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048791177187733104,
      "loss": 6.1132,
      "mean_token_accuracy": 0.14997347295284272,
      "num_tokens": 23129189.0,
      "step": 10690
    },
    {
      "entropy": 6.454489803314209,
      "epoch": 1.1444700090962598,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004878994054987627,
      "loss": 6.1198,
      "mean_token_accuracy": 0.14506647884845733,
      "num_tokens": 23140238.0,
      "step": 10695
    },
    {
      "entropy": 6.3685537815094,
      "epoch": 1.1450050832040237,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004878870329727539,
      "loss": 6.0011,
      "mean_token_accuracy": 0.1572861760854721,
      "num_tokens": 23149567.0,
      "step": 10700
    },
    {
      "entropy": 6.416757965087891,
      "epoch": 1.1455401573117876,
      "grad_norm": 0.953125,
      "learning_rate": 0.000487874654299662,
      "loss": 6.0041,
      "mean_token_accuracy": 0.1508268818259239,
      "num_tokens": 23159208.0,
      "step": 10705
    },
    {
      "entropy": 6.35107684135437,
      "epoch": 1.1460752314195517,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048786226947984435,
      "loss": 6.0547,
      "mean_token_accuracy": 0.14981927275657653,
      "num_tokens": 23169092.0,
      "step": 10710
    },
    {
      "entropy": 6.402327346801758,
      "epoch": 1.1466103055273156,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004878498785136586,
      "loss": 6.0698,
      "mean_token_accuracy": 0.14950211346149445,
      "num_tokens": 23180385.0,
      "step": 10715
    },
    {
      "entropy": 6.467448329925537,
      "epoch": 1.1471453796350795,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048783748140146245,
      "loss": 6.0548,
      "mean_token_accuracy": 0.14751721769571305,
      "num_tokens": 23191180.0,
      "step": 10720
    },
    {
      "entropy": 6.424088859558106,
      "epoch": 1.1476804537428433,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000487825078143614,
      "loss": 6.0423,
      "mean_token_accuracy": 0.15412194430828094,
      "num_tokens": 23200920.0,
      "step": 10725
    },
    {
      "entropy": 6.37498288154602,
      "epoch": 1.1482155278506072,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004878126687404713,
      "loss": 6.0362,
      "mean_token_accuracy": 0.15680174380540848,
      "num_tokens": 23211140.0,
      "step": 10730
    },
    {
      "entropy": 6.387633609771728,
      "epoch": 1.1487506019583713,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004878002531923927,
      "loss": 6.0763,
      "mean_token_accuracy": 0.14475269690155984,
      "num_tokens": 23222039.0,
      "step": 10735
    },
    {
      "entropy": 6.512231254577637,
      "epoch": 1.1492856760661352,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048778783149973674,
      "loss": 6.1203,
      "mean_token_accuracy": 0.14616891294717788,
      "num_tokens": 23232502.0,
      "step": 10740
    },
    {
      "entropy": 6.468053960800171,
      "epoch": 1.149820750173899,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048777540366286195,
      "loss": 6.0404,
      "mean_token_accuracy": 0.15045567005872726,
      "num_tokens": 23243172.0,
      "step": 10745
    },
    {
      "entropy": 6.435519981384277,
      "epoch": 1.150355824281663,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004877629696821273,
      "loss": 6.1162,
      "mean_token_accuracy": 0.14770963340997695,
      "num_tokens": 23254558.0,
      "step": 10750
    },
    {
      "entropy": 6.455703926086426,
      "epoch": 1.1508908983894268,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048775052955789185,
      "loss": 6.1455,
      "mean_token_accuracy": 0.14684202820062636,
      "num_tokens": 23266283.0,
      "step": 10755
    },
    {
      "entropy": 6.393408870697021,
      "epoch": 1.151425972497191,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004877380832905147,
      "loss": 6.005,
      "mean_token_accuracy": 0.15767347514629365,
      "num_tokens": 23277823.0,
      "step": 10760
    },
    {
      "entropy": 6.392177772521973,
      "epoch": 1.1519610466049548,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048772563088035533,
      "loss": 5.9914,
      "mean_token_accuracy": 0.1517062395811081,
      "num_tokens": 23288159.0,
      "step": 10765
    },
    {
      "entropy": 6.38089337348938,
      "epoch": 1.1524961207127187,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004877131723277732,
      "loss": 6.035,
      "mean_token_accuracy": 0.14856482297182083,
      "num_tokens": 23299379.0,
      "step": 10770
    },
    {
      "entropy": 6.422782230377197,
      "epoch": 1.1530311948204826,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004877007076331282,
      "loss": 6.0547,
      "mean_token_accuracy": 0.14881094843149184,
      "num_tokens": 23310572.0,
      "step": 10775
    },
    {
      "entropy": 6.447106122970581,
      "epoch": 1.1535662689282467,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004876882367967801,
      "loss": 6.049,
      "mean_token_accuracy": 0.15354990661144258,
      "num_tokens": 23320782.0,
      "step": 10780
    },
    {
      "entropy": 6.45343599319458,
      "epoch": 1.1541013430360105,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048767575981908907,
      "loss": 6.1333,
      "mean_token_accuracy": 0.14321690946817398,
      "num_tokens": 23331427.0,
      "step": 10785
    },
    {
      "entropy": 6.449858856201172,
      "epoch": 1.1546364171437744,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048766327670041534,
      "loss": 6.2368,
      "mean_token_accuracy": 0.13702579885721206,
      "num_tokens": 23344250.0,
      "step": 10790
    },
    {
      "entropy": 6.46830005645752,
      "epoch": 1.1551714912515383,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004876507874411194,
      "loss": 6.0842,
      "mean_token_accuracy": 0.15416404157876967,
      "num_tokens": 23354404.0,
      "step": 10795
    },
    {
      "entropy": 6.43124418258667,
      "epoch": 1.1557065653593024,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048763829204156187,
      "loss": 6.1027,
      "mean_token_accuracy": 0.14286879524588586,
      "num_tokens": 23364634.0,
      "step": 10800
    },
    {
      "entropy": 6.33711085319519,
      "epoch": 1.1562416394670663,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048762579050210344,
      "loss": 5.9507,
      "mean_token_accuracy": 0.15245019495487214,
      "num_tokens": 23375816.0,
      "step": 10805
    },
    {
      "entropy": 6.479554891586304,
      "epoch": 1.1567767135748301,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048761328282310534,
      "loss": 6.103,
      "mean_token_accuracy": 0.14906885772943496,
      "num_tokens": 23386583.0,
      "step": 10810
    },
    {
      "entropy": 6.391775465011596,
      "epoch": 1.157311787682594,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048760076900492846,
      "loss": 6.0225,
      "mean_token_accuracy": 0.15515367686748505,
      "num_tokens": 23396826.0,
      "step": 10815
    },
    {
      "entropy": 6.416030597686768,
      "epoch": 1.157846861790358,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004875882490479343,
      "loss": 6.1188,
      "mean_token_accuracy": 0.15226272046566008,
      "num_tokens": 23408661.0,
      "step": 10820
    },
    {
      "entropy": 6.410414028167724,
      "epoch": 1.158381935898122,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048757572295248425,
      "loss": 6.0268,
      "mean_token_accuracy": 0.14845772534608842,
      "num_tokens": 23419308.0,
      "step": 10825
    },
    {
      "entropy": 6.372759437561035,
      "epoch": 1.1589170100058859,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004875631907189402,
      "loss": 6.0172,
      "mean_token_accuracy": 0.15785436779260636,
      "num_tokens": 23430043.0,
      "step": 10830
    },
    {
      "entropy": 6.350531959533692,
      "epoch": 1.1594520841136498,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004875506523476638,
      "loss": 5.9453,
      "mean_token_accuracy": 0.1518328830599785,
      "num_tokens": 23441246.0,
      "step": 10835
    },
    {
      "entropy": 6.444146203994751,
      "epoch": 1.1599871582214136,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048753810783901716,
      "loss": 6.0989,
      "mean_token_accuracy": 0.14499814212322235,
      "num_tokens": 23452250.0,
      "step": 10840
    },
    {
      "entropy": 6.444901609420777,
      "epoch": 1.1605222323291775,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048752555719336257,
      "loss": 6.0786,
      "mean_token_accuracy": 0.14988540560007096,
      "num_tokens": 23464174.0,
      "step": 10845
    },
    {
      "entropy": 6.370390224456787,
      "epoch": 1.1610573064369416,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004875130004110623,
      "loss": 5.9484,
      "mean_token_accuracy": 0.15264058411121367,
      "num_tokens": 23475159.0,
      "step": 10850
    },
    {
      "entropy": 6.325065851211548,
      "epoch": 1.1615923805447055,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048750043749247907,
      "loss": 6.016,
      "mean_token_accuracy": 0.1538507580757141,
      "num_tokens": 23486210.0,
      "step": 10855
    },
    {
      "entropy": 6.379917907714844,
      "epoch": 1.1621274546524694,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004874878684379756,
      "loss": 6.1059,
      "mean_token_accuracy": 0.14704688489437104,
      "num_tokens": 23497122.0,
      "step": 10860
    },
    {
      "entropy": 6.383538341522216,
      "epoch": 1.1626625287602332,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004874752932479148,
      "loss": 5.9659,
      "mean_token_accuracy": 0.15582831650972367,
      "num_tokens": 23507765.0,
      "step": 10865
    },
    {
      "entropy": 6.498394918441773,
      "epoch": 1.1631976028679971,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048746271192265974,
      "loss": 6.0645,
      "mean_token_accuracy": 0.1484560951590538,
      "num_tokens": 23517686.0,
      "step": 10870
    },
    {
      "entropy": 6.4012901306152346,
      "epoch": 1.1637326769757612,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004874501244625737,
      "loss": 6.0295,
      "mean_token_accuracy": 0.14884432405233383,
      "num_tokens": 23528325.0,
      "step": 10875
    },
    {
      "entropy": 6.376734828948974,
      "epoch": 1.164267751083525,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004874375308680202,
      "loss": 6.1055,
      "mean_token_accuracy": 0.1486441746354103,
      "num_tokens": 23540006.0,
      "step": 10880
    },
    {
      "entropy": 6.500550174713135,
      "epoch": 1.164802825191289,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048742493113936274,
      "loss": 6.0118,
      "mean_token_accuracy": 0.14792283177375792,
      "num_tokens": 23550419.0,
      "step": 10885
    },
    {
      "entropy": 6.498251581192017,
      "epoch": 1.1653378992990528,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004874123252769653,
      "loss": 6.12,
      "mean_token_accuracy": 0.1446204587817192,
      "num_tokens": 23561227.0,
      "step": 10890
    },
    {
      "entropy": 6.41230001449585,
      "epoch": 1.1658729734068167,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048739971328119194,
      "loss": 6.0517,
      "mean_token_accuracy": 0.15094673931598662,
      "num_tokens": 23571357.0,
      "step": 10895
    },
    {
      "entropy": 6.440952777862549,
      "epoch": 1.1664080475145808,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004873870951524066,
      "loss": 6.0791,
      "mean_token_accuracy": 0.15124721825122833,
      "num_tokens": 23581875.0,
      "step": 10900
    },
    {
      "entropy": 6.444917011260986,
      "epoch": 1.1669431216223447,
      "grad_norm": 0.875,
      "learning_rate": 0.00048737447089097373,
      "loss": 6.1344,
      "mean_token_accuracy": 0.14199871569871902,
      "num_tokens": 23594251.0,
      "step": 10905
    },
    {
      "entropy": 6.440668821334839,
      "epoch": 1.1674781957301086,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004873618404972579,
      "loss": 6.0765,
      "mean_token_accuracy": 0.14913819432258607,
      "num_tokens": 23605159.0,
      "step": 10910
    },
    {
      "entropy": 6.496748781204223,
      "epoch": 1.1680132698378725,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048734920397162376,
      "loss": 6.0178,
      "mean_token_accuracy": 0.15884051322937012,
      "num_tokens": 23615737.0,
      "step": 10915
    },
    {
      "entropy": 6.414934492111206,
      "epoch": 1.1685483439456366,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004873365613144361,
      "loss": 6.0453,
      "mean_token_accuracy": 0.14735834300518036,
      "num_tokens": 23626207.0,
      "step": 10920
    },
    {
      "entropy": 6.452676105499267,
      "epoch": 1.1690834180534004,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004873239125260602,
      "loss": 6.101,
      "mean_token_accuracy": 0.15018949583172797,
      "num_tokens": 23637900.0,
      "step": 10925
    },
    {
      "entropy": 6.500513505935669,
      "epoch": 1.1696184921611643,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048731125760686106,
      "loss": 5.9856,
      "mean_token_accuracy": 0.15406192615628242,
      "num_tokens": 23647867.0,
      "step": 10930
    },
    {
      "entropy": 6.45203537940979,
      "epoch": 1.1701535662689282,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048729859655720424,
      "loss": 6.0636,
      "mean_token_accuracy": 0.15095616579055787,
      "num_tokens": 23658055.0,
      "step": 10935
    },
    {
      "entropy": 6.3218199729919435,
      "epoch": 1.1706886403766923,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004872859293774553,
      "loss": 6.0392,
      "mean_token_accuracy": 0.1521560087800026,
      "num_tokens": 23669458.0,
      "step": 10940
    },
    {
      "entropy": 6.392317771911621,
      "epoch": 1.1712237144844562,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048727325606797995,
      "loss": 6.0467,
      "mean_token_accuracy": 0.14767611622810364,
      "num_tokens": 23679702.0,
      "step": 10945
    },
    {
      "entropy": 6.436883544921875,
      "epoch": 1.17175878859222,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004872605766291441,
      "loss": 6.043,
      "mean_token_accuracy": 0.15271709114313126,
      "num_tokens": 23690844.0,
      "step": 10950
    },
    {
      "entropy": 6.374674892425537,
      "epoch": 1.172293862699984,
      "grad_norm": 0.9375,
      "learning_rate": 0.000487247891061314,
      "loss": 6.0175,
      "mean_token_accuracy": 0.15097465068101884,
      "num_tokens": 23701484.0,
      "step": 10955
    },
    {
      "entropy": 6.466239404678345,
      "epoch": 1.1728289368077478,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004872351993648559,
      "loss": 6.0758,
      "mean_token_accuracy": 0.14658529683947563,
      "num_tokens": 23711452.0,
      "step": 10960
    },
    {
      "entropy": 6.42755446434021,
      "epoch": 1.173364010915512,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048722250154013613,
      "loss": 6.0141,
      "mean_token_accuracy": 0.1517298236489296,
      "num_tokens": 23723514.0,
      "step": 10965
    },
    {
      "entropy": 6.419243288040161,
      "epoch": 1.1738990850232758,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004872097975875216,
      "loss": 6.0759,
      "mean_token_accuracy": 0.1545051470398903,
      "num_tokens": 23733722.0,
      "step": 10970
    },
    {
      "entropy": 6.47608699798584,
      "epoch": 1.1744341591310397,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004871970875073789,
      "loss": 6.0574,
      "mean_token_accuracy": 0.14802472591400145,
      "num_tokens": 23743403.0,
      "step": 10975
    },
    {
      "entropy": 6.411939477920532,
      "epoch": 1.1749692332388035,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004871843713000751,
      "loss": 6.0187,
      "mean_token_accuracy": 0.15457520484924317,
      "num_tokens": 23754558.0,
      "step": 10980
    },
    {
      "entropy": 6.477199697494507,
      "epoch": 1.1755043073465674,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048717164896597737,
      "loss": 6.082,
      "mean_token_accuracy": 0.1506623700261116,
      "num_tokens": 23766346.0,
      "step": 10985
    },
    {
      "entropy": 6.451038265228272,
      "epoch": 1.1760393814543315,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004871589205054532,
      "loss": 6.0656,
      "mean_token_accuracy": 0.1602258265018463,
      "num_tokens": 23776851.0,
      "step": 10990
    },
    {
      "entropy": 6.395011281967163,
      "epoch": 1.1765744555620954,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048714618591886997,
      "loss": 6.1226,
      "mean_token_accuracy": 0.14618657082319259,
      "num_tokens": 23787350.0,
      "step": 10995
    },
    {
      "entropy": 6.328485059738159,
      "epoch": 1.1771095296698593,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004871334452065954,
      "loss": 5.9943,
      "mean_token_accuracy": 0.15161629319190978,
      "num_tokens": 23797738.0,
      "step": 11000
    },
    {
      "entropy": 6.43249921798706,
      "epoch": 1.1776446037776231,
      "grad_norm": 0.875,
      "learning_rate": 0.0004871206983689974,
      "loss": 5.9723,
      "mean_token_accuracy": 0.15947188436985016,
      "num_tokens": 23809516.0,
      "step": 11005
    },
    {
      "entropy": 6.431351518630981,
      "epoch": 1.178179677885387,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000487107945406444,
      "loss": 6.0466,
      "mean_token_accuracy": 0.15639907345175744,
      "num_tokens": 23819850.0,
      "step": 11010
    },
    {
      "entropy": 6.3724359512329105,
      "epoch": 1.1787147519931511,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004870951863193036,
      "loss": 6.0517,
      "mean_token_accuracy": 0.14713381677865983,
      "num_tokens": 23829891.0,
      "step": 11015
    },
    {
      "entropy": 6.34831976890564,
      "epoch": 1.179249826100915,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004870824211079444,
      "loss": 5.9907,
      "mean_token_accuracy": 0.15624823272228242,
      "num_tokens": 23840816.0,
      "step": 11020
    },
    {
      "entropy": 6.476291608810425,
      "epoch": 1.1797849002086789,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048706964977273515,
      "loss": 6.0853,
      "mean_token_accuracy": 0.15212935209274292,
      "num_tokens": 23852595.0,
      "step": 11025
    },
    {
      "entropy": 6.472467136383057,
      "epoch": 1.1803199743164428,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048705687231404455,
      "loss": 6.0786,
      "mean_token_accuracy": 0.14639779478311538,
      "num_tokens": 23863048.0,
      "step": 11030
    },
    {
      "entropy": 6.415596199035645,
      "epoch": 1.1808550484242066,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048704408873224156,
      "loss": 6.0461,
      "mean_token_accuracy": 0.15126620531082152,
      "num_tokens": 23873499.0,
      "step": 11035
    },
    {
      "entropy": 6.355279350280762,
      "epoch": 1.1813901225319707,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048703129902769526,
      "loss": 5.9984,
      "mean_token_accuracy": 0.1568140096962452,
      "num_tokens": 23885129.0,
      "step": 11040
    },
    {
      "entropy": 6.463844966888428,
      "epoch": 1.1819251966397346,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004870185032007751,
      "loss": 6.0929,
      "mean_token_accuracy": 0.14369981586933137,
      "num_tokens": 23895538.0,
      "step": 11045
    },
    {
      "entropy": 6.448873138427734,
      "epoch": 1.1824602707474985,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004870057012518504,
      "loss": 6.1211,
      "mean_token_accuracy": 0.14315536618232727,
      "num_tokens": 23907341.0,
      "step": 11050
    },
    {
      "entropy": 6.391955423355102,
      "epoch": 1.1829953448552624,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048699289318129087,
      "loss": 5.9636,
      "mean_token_accuracy": 0.15990957915782927,
      "num_tokens": 23918501.0,
      "step": 11055
    },
    {
      "entropy": 6.431356811523438,
      "epoch": 1.1835304189630265,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004869800789894663,
      "loss": 6.0357,
      "mean_token_accuracy": 0.15139036476612092,
      "num_tokens": 23929204.0,
      "step": 11060
    },
    {
      "entropy": 6.39577350616455,
      "epoch": 1.1840654930707903,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004869672586767468,
      "loss": 6.0791,
      "mean_token_accuracy": 0.15489304065704346,
      "num_tokens": 23939855.0,
      "step": 11065
    },
    {
      "entropy": 6.466464900970459,
      "epoch": 1.1846005671785542,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004869544322435024,
      "loss": 6.0116,
      "mean_token_accuracy": 0.1534983828663826,
      "num_tokens": 23949871.0,
      "step": 11070
    },
    {
      "entropy": 6.421636962890625,
      "epoch": 1.185135641286318,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004869415996901036,
      "loss": 6.1261,
      "mean_token_accuracy": 0.14914300367236139,
      "num_tokens": 23961650.0,
      "step": 11075
    },
    {
      "entropy": 6.453940391540527,
      "epoch": 1.1856707153940822,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004869287610169209,
      "loss": 6.1325,
      "mean_token_accuracy": 0.1488512299954891,
      "num_tokens": 23973103.0,
      "step": 11080
    },
    {
      "entropy": 6.4599470615386965,
      "epoch": 1.186205789501846,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000486915916224325,
      "loss": 6.0599,
      "mean_token_accuracy": 0.1506316766142845,
      "num_tokens": 23982972.0,
      "step": 11085
    },
    {
      "entropy": 6.352290344238281,
      "epoch": 1.18674086360961,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004869030653126868,
      "loss": 5.9215,
      "mean_token_accuracy": 0.15819128155708312,
      "num_tokens": 23993044.0,
      "step": 11090
    },
    {
      "entropy": 6.397801113128662,
      "epoch": 1.1872759377173738,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004868902082823774,
      "loss": 6.0122,
      "mean_token_accuracy": 0.15150389820337296,
      "num_tokens": 24003871.0,
      "step": 11095
    },
    {
      "entropy": 6.427644729614258,
      "epoch": 1.1878110118251377,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004868773451337679,
      "loss": 6.0572,
      "mean_token_accuracy": 0.14718835502862931,
      "num_tokens": 24014613.0,
      "step": 11100
    },
    {
      "entropy": 6.4206342697143555,
      "epoch": 1.1883460859329018,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048686447586722995,
      "loss": 6.0725,
      "mean_token_accuracy": 0.15432872474193574,
      "num_tokens": 24024172.0,
      "step": 11105
    },
    {
      "entropy": 6.454071521759033,
      "epoch": 1.1888811600406657,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048685160048313506,
      "loss": 6.1708,
      "mean_token_accuracy": 0.14683951437473297,
      "num_tokens": 24035301.0,
      "step": 11110
    },
    {
      "entropy": 6.479418611526489,
      "epoch": 1.1894162341484296,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048683871898185486,
      "loss": 6.0303,
      "mean_token_accuracy": 0.15984879285097123,
      "num_tokens": 24045661.0,
      "step": 11115
    },
    {
      "entropy": 6.408260250091553,
      "epoch": 1.1899513082561934,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048682583136376145,
      "loss": 6.0426,
      "mean_token_accuracy": 0.15274788737297057,
      "num_tokens": 24057027.0,
      "step": 11120
    },
    {
      "entropy": 6.414361476898193,
      "epoch": 1.1904863823639573,
      "grad_norm": 0.875,
      "learning_rate": 0.0004868129376292269,
      "loss": 6.1632,
      "mean_token_accuracy": 0.14436013549566268,
      "num_tokens": 24068322.0,
      "step": 11125
    },
    {
      "entropy": 6.440590906143188,
      "epoch": 1.1910214564717214,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048680003777862356,
      "loss": 5.9903,
      "mean_token_accuracy": 0.1523751512169838,
      "num_tokens": 24078506.0,
      "step": 11130
    },
    {
      "entropy": 6.424045658111572,
      "epoch": 1.1915565305794853,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048678713181232394,
      "loss": 6.0315,
      "mean_token_accuracy": 0.1511312171816826,
      "num_tokens": 24090296.0,
      "step": 11135
    },
    {
      "entropy": 6.3559572219848635,
      "epoch": 1.1920916046872492,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048677421973070063,
      "loss": 6.0469,
      "mean_token_accuracy": 0.1446344368159771,
      "num_tokens": 24101478.0,
      "step": 11140
    },
    {
      "entropy": 6.4094812870025635,
      "epoch": 1.192626678795013,
      "grad_norm": 0.875,
      "learning_rate": 0.00048676130153412644,
      "loss": 6.0621,
      "mean_token_accuracy": 0.15257982313632965,
      "num_tokens": 24112552.0,
      "step": 11145
    },
    {
      "entropy": 6.326308345794677,
      "epoch": 1.193161752902777,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004867483772229745,
      "loss": 5.9356,
      "mean_token_accuracy": 0.15738652348518373,
      "num_tokens": 24124210.0,
      "step": 11150
    },
    {
      "entropy": 6.466823482513428,
      "epoch": 1.193696827010541,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004867354467976178,
      "loss": 6.0344,
      "mean_token_accuracy": 0.15150919705629348,
      "num_tokens": 24133525.0,
      "step": 11155
    },
    {
      "entropy": 6.445204019546509,
      "epoch": 1.194231901118305,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048672251025842994,
      "loss": 6.0965,
      "mean_token_accuracy": 0.14994918406009675,
      "num_tokens": 24145400.0,
      "step": 11160
    },
    {
      "entropy": 6.395748043060303,
      "epoch": 1.1947669752260688,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004867095676057843,
      "loss": 6.083,
      "mean_token_accuracy": 0.14470206946134567,
      "num_tokens": 24155596.0,
      "step": 11165
    },
    {
      "entropy": 6.469098711013794,
      "epoch": 1.1953020493338327,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048669661884005467,
      "loss": 6.1065,
      "mean_token_accuracy": 0.1415400743484497,
      "num_tokens": 24166683.0,
      "step": 11170
    },
    {
      "entropy": 6.476481103897095,
      "epoch": 1.1958371234415965,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048668366396161503,
      "loss": 6.0103,
      "mean_token_accuracy": 0.15461833626031876,
      "num_tokens": 24177649.0,
      "step": 11175
    },
    {
      "entropy": 6.398001623153687,
      "epoch": 1.1963721975493606,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004866707029708392,
      "loss": 6.059,
      "mean_token_accuracy": 0.15671389251947404,
      "num_tokens": 24188522.0,
      "step": 11180
    },
    {
      "entropy": 6.414281177520752,
      "epoch": 1.1969072716571245,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004866577358681015,
      "loss": 6.0267,
      "mean_token_accuracy": 0.15400335788726807,
      "num_tokens": 24199285.0,
      "step": 11185
    },
    {
      "entropy": 6.378680086135864,
      "epoch": 1.1974423457648884,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004866447626537765,
      "loss": 5.987,
      "mean_token_accuracy": 0.14814073890447615,
      "num_tokens": 24210929.0,
      "step": 11190
    },
    {
      "entropy": 6.402561044692993,
      "epoch": 1.1979774198726523,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004866317833282387,
      "loss": 6.0694,
      "mean_token_accuracy": 0.1501225158572197,
      "num_tokens": 24221394.0,
      "step": 11195
    },
    {
      "entropy": 6.420371580123901,
      "epoch": 1.1985124939804164,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048661879789186295,
      "loss": 6.0316,
      "mean_token_accuracy": 0.15241131484508513,
      "num_tokens": 24230297.0,
      "step": 11200
    },
    {
      "entropy": 6.409456634521485,
      "epoch": 1.1990475680881802,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048660580634502415,
      "loss": 6.1737,
      "mean_token_accuracy": 0.1452004611492157,
      "num_tokens": 24241505.0,
      "step": 11205
    },
    {
      "entropy": 6.404465007781982,
      "epoch": 1.1995826421959441,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048659280868809744,
      "loss": 6.0246,
      "mean_token_accuracy": 0.15384707748889923,
      "num_tokens": 24252638.0,
      "step": 11210
    },
    {
      "entropy": 6.376628589630127,
      "epoch": 1.200117716303708,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000486579804921458,
      "loss": 6.0551,
      "mean_token_accuracy": 0.15416979193687438,
      "num_tokens": 24263156.0,
      "step": 11215
    },
    {
      "entropy": 6.379791259765625,
      "epoch": 1.200652790411472,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048656679504548145,
      "loss": 5.9074,
      "mean_token_accuracy": 0.16190877854824065,
      "num_tokens": 24274384.0,
      "step": 11220
    },
    {
      "entropy": 6.497354412078858,
      "epoch": 1.201187864519236,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004865537790605435,
      "loss": 6.1428,
      "mean_token_accuracy": 0.14229072779417037,
      "num_tokens": 24285159.0,
      "step": 11225
    },
    {
      "entropy": 6.4543530464172365,
      "epoch": 1.2017229386269999,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004865407569670198,
      "loss": 6.0934,
      "mean_token_accuracy": 0.14559346139431,
      "num_tokens": 24296202.0,
      "step": 11230
    },
    {
      "entropy": 6.404468011856079,
      "epoch": 1.2022580127347637,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004865277287652866,
      "loss": 6.0091,
      "mean_token_accuracy": 0.15226322785019875,
      "num_tokens": 24306691.0,
      "step": 11235
    },
    {
      "entropy": 6.395770406723022,
      "epoch": 1.2027930868425276,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004865146944557199,
      "loss": 6.0488,
      "mean_token_accuracy": 0.15193377435207367,
      "num_tokens": 24317499.0,
      "step": 11240
    },
    {
      "entropy": 6.430694341659546,
      "epoch": 1.2033281609502917,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004865016540386961,
      "loss": 6.0302,
      "mean_token_accuracy": 0.14733619689941407,
      "num_tokens": 24328096.0,
      "step": 11245
    },
    {
      "entropy": 6.433403444290161,
      "epoch": 1.2038632350580556,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048648860751459173,
      "loss": 6.0356,
      "mean_token_accuracy": 0.14766712933778764,
      "num_tokens": 24338188.0,
      "step": 11250
    },
    {
      "entropy": 6.431427097320556,
      "epoch": 1.2043983091658195,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048647555488378355,
      "loss": 6.0693,
      "mean_token_accuracy": 0.15442163050174712,
      "num_tokens": 24348578.0,
      "step": 11255
    },
    {
      "entropy": 6.3803808212280275,
      "epoch": 1.2049333832735833,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004864624961466485,
      "loss": 6.0496,
      "mean_token_accuracy": 0.15346357002854347,
      "num_tokens": 24358807.0,
      "step": 11260
    },
    {
      "entropy": 6.432039403915406,
      "epoch": 1.2054684573813472,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004864494313035635,
      "loss": 6.0544,
      "mean_token_accuracy": 0.14514696151018142,
      "num_tokens": 24369074.0,
      "step": 11265
    },
    {
      "entropy": 6.434797477722168,
      "epoch": 1.2060035314891113,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000486436360354906,
      "loss": 6.0279,
      "mean_token_accuracy": 0.15426842272281646,
      "num_tokens": 24380655.0,
      "step": 11270
    },
    {
      "entropy": 6.389752149581909,
      "epoch": 1.2065386055968752,
      "grad_norm": 0.875,
      "learning_rate": 0.00048642328330105323,
      "loss": 6.0068,
      "mean_token_accuracy": 0.15249717831611634,
      "num_tokens": 24391701.0,
      "step": 11275
    },
    {
      "entropy": 6.324358415603638,
      "epoch": 1.207073679704639,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048641020014238286,
      "loss": 6.0965,
      "mean_token_accuracy": 0.1460344523191452,
      "num_tokens": 24403690.0,
      "step": 11280
    },
    {
      "entropy": 6.488384675979614,
      "epoch": 1.207608753812403,
      "grad_norm": 0.875,
      "learning_rate": 0.0004863971108792727,
      "loss": 6.0817,
      "mean_token_accuracy": 0.14792633131146432,
      "num_tokens": 24414670.0,
      "step": 11285
    },
    {
      "entropy": 6.482924175262451,
      "epoch": 1.2081438279201668,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048638401551210063,
      "loss": 6.1303,
      "mean_token_accuracy": 0.14009737074375153,
      "num_tokens": 24425237.0,
      "step": 11290
    },
    {
      "entropy": 6.438837146759033,
      "epoch": 1.208678902027931,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048637091404124484,
      "loss": 6.0802,
      "mean_token_accuracy": 0.15160878971219063,
      "num_tokens": 24437142.0,
      "step": 11295
    },
    {
      "entropy": 6.373235845565796,
      "epoch": 1.2092139761356948,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004863578064670837,
      "loss": 5.9831,
      "mean_token_accuracy": 0.1547194167971611,
      "num_tokens": 24448819.0,
      "step": 11300
    },
    {
      "entropy": 6.363312387466431,
      "epoch": 1.2097490502434587,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048634469278999545,
      "loss": 5.947,
      "mean_token_accuracy": 0.16114952117204667,
      "num_tokens": 24459392.0,
      "step": 11305
    },
    {
      "entropy": 6.428229570388794,
      "epoch": 1.2102841243512226,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048633157301035895,
      "loss": 6.0426,
      "mean_token_accuracy": 0.14748336002230644,
      "num_tokens": 24470298.0,
      "step": 11310
    },
    {
      "entropy": 6.4278522491455075,
      "epoch": 1.2108191984589864,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000486318447128553,
      "loss": 6.0106,
      "mean_token_accuracy": 0.1539096176624298,
      "num_tokens": 24480477.0,
      "step": 11315
    },
    {
      "entropy": 6.446617126464844,
      "epoch": 1.2113542725667505,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004863053151449566,
      "loss": 6.0875,
      "mean_token_accuracy": 0.1393823206424713,
      "num_tokens": 24491890.0,
      "step": 11320
    },
    {
      "entropy": 6.4543599605560305,
      "epoch": 1.2118893466745144,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048629217705994883,
      "loss": 6.0258,
      "mean_token_accuracy": 0.1530631609261036,
      "num_tokens": 24502583.0,
      "step": 11325
    },
    {
      "entropy": 6.437144184112549,
      "epoch": 1.2124244207822783,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004862790328739091,
      "loss": 6.1339,
      "mean_token_accuracy": 0.1446865290403366,
      "num_tokens": 24514491.0,
      "step": 11330
    },
    {
      "entropy": 6.364309978485108,
      "epoch": 1.2129594948900422,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048626588258721717,
      "loss": 5.9662,
      "mean_token_accuracy": 0.1597479909658432,
      "num_tokens": 24524788.0,
      "step": 11335
    },
    {
      "entropy": 6.406537866592407,
      "epoch": 1.2134945689978063,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004862527262002524,
      "loss": 6.0055,
      "mean_token_accuracy": 0.14402141720056533,
      "num_tokens": 24535699.0,
      "step": 11340
    },
    {
      "entropy": 6.387211656570434,
      "epoch": 1.2140296431055702,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048623956371339494,
      "loss": 6.0252,
      "mean_token_accuracy": 0.15207465887069702,
      "num_tokens": 24547339.0,
      "step": 11345
    },
    {
      "entropy": 6.430961227416992,
      "epoch": 1.214564717213334,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048622639512702477,
      "loss": 6.1304,
      "mean_token_accuracy": 0.14383527934551238,
      "num_tokens": 24558778.0,
      "step": 11350
    },
    {
      "entropy": 6.422319412231445,
      "epoch": 1.215099791321098,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048621322044152207,
      "loss": 6.0515,
      "mean_token_accuracy": 0.14978087246417998,
      "num_tokens": 24569032.0,
      "step": 11355
    },
    {
      "entropy": 6.43628888130188,
      "epoch": 1.215634865428862,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048620003965726727,
      "loss": 6.137,
      "mean_token_accuracy": 0.137799970805645,
      "num_tokens": 24581356.0,
      "step": 11360
    },
    {
      "entropy": 6.435864686965942,
      "epoch": 1.2161699395366259,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048618685277464116,
      "loss": 6.0176,
      "mean_token_accuracy": 0.15392738878726958,
      "num_tokens": 24591809.0,
      "step": 11365
    },
    {
      "entropy": 6.331251811981201,
      "epoch": 1.2167050136443898,
      "grad_norm": 0.875,
      "learning_rate": 0.00048617365979402417,
      "loss": 5.9211,
      "mean_token_accuracy": 0.15913875699043273,
      "num_tokens": 24603492.0,
      "step": 11370
    },
    {
      "entropy": 6.405251312255859,
      "epoch": 1.2172400877521536,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004861604607157975,
      "loss": 6.0821,
      "mean_token_accuracy": 0.1485549509525299,
      "num_tokens": 24614837.0,
      "step": 11375
    },
    {
      "entropy": 6.471507835388183,
      "epoch": 1.2177751618599175,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004861472555403423,
      "loss": 6.0668,
      "mean_token_accuracy": 0.1479809433221817,
      "num_tokens": 24624714.0,
      "step": 11380
    },
    {
      "entropy": 6.488359260559082,
      "epoch": 1.2183102359676816,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048613404426803964,
      "loss": 6.1138,
      "mean_token_accuracy": 0.14631285667419433,
      "num_tokens": 24635758.0,
      "step": 11385
    },
    {
      "entropy": 6.464166212081909,
      "epoch": 1.2188453100754455,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048612082689927116,
      "loss": 6.089,
      "mean_token_accuracy": 0.15182169079780578,
      "num_tokens": 24645362.0,
      "step": 11390
    },
    {
      "entropy": 6.41919093132019,
      "epoch": 1.2193803841832094,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004861076034344185,
      "loss": 6.0917,
      "mean_token_accuracy": 0.15264866203069688,
      "num_tokens": 24656898.0,
      "step": 11395
    },
    {
      "entropy": 6.417871522903442,
      "epoch": 1.2199154582909733,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004860943738738634,
      "loss": 6.026,
      "mean_token_accuracy": 0.15403184294700623,
      "num_tokens": 24668191.0,
      "step": 11400
    },
    {
      "entropy": 6.4486246585845945,
      "epoch": 1.2204505323987371,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048608113821798786,
      "loss": 6.0431,
      "mean_token_accuracy": 0.1521573543548584,
      "num_tokens": 24678034.0,
      "step": 11405
    },
    {
      "entropy": 6.425697278976441,
      "epoch": 1.2209856065065012,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004860678964671743,
      "loss": 6.089,
      "mean_token_accuracy": 0.15016046911478043,
      "num_tokens": 24689062.0,
      "step": 11410
    },
    {
      "entropy": 6.478608226776123,
      "epoch": 1.221520680614265,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048605464862180474,
      "loss": 6.0595,
      "mean_token_accuracy": 0.1453390784561634,
      "num_tokens": 24699787.0,
      "step": 11415
    },
    {
      "entropy": 6.439414119720459,
      "epoch": 1.222055754722029,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004860413946822619,
      "loss": 5.9953,
      "mean_token_accuracy": 0.15049341320991516,
      "num_tokens": 24711233.0,
      "step": 11420
    },
    {
      "entropy": 6.3687676906585695,
      "epoch": 1.2225908288297929,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004860281346489284,
      "loss": 6.0303,
      "mean_token_accuracy": 0.14747673124074936,
      "num_tokens": 24722414.0,
      "step": 11425
    },
    {
      "entropy": 6.446717596054077,
      "epoch": 1.2231259029375567,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004860148685221873,
      "loss": 6.0665,
      "mean_token_accuracy": 0.14582476913928985,
      "num_tokens": 24733093.0,
      "step": 11430
    },
    {
      "entropy": 6.414763355255127,
      "epoch": 1.2236609770453208,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048600159630242135,
      "loss": 6.0082,
      "mean_token_accuracy": 0.15332933366298676,
      "num_tokens": 24743195.0,
      "step": 11435
    },
    {
      "entropy": 6.445101308822632,
      "epoch": 1.2241960511530847,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048598831799001406,
      "loss": 6.0262,
      "mean_token_accuracy": 0.1520538315176964,
      "num_tokens": 24753680.0,
      "step": 11440
    },
    {
      "entropy": 6.440192699432373,
      "epoch": 1.2247311252608486,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004859750335853488,
      "loss": 6.0071,
      "mean_token_accuracy": 0.15410007238388063,
      "num_tokens": 24764135.0,
      "step": 11445
    },
    {
      "entropy": 6.308586597442627,
      "epoch": 1.2252661993686125,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000485961743088809,
      "loss": 5.953,
      "mean_token_accuracy": 0.16644514054059983,
      "num_tokens": 24775740.0,
      "step": 11450
    },
    {
      "entropy": 6.433221101760864,
      "epoch": 1.2258012734763766,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004859484465007784,
      "loss": 6.1034,
      "mean_token_accuracy": 0.14898681044578552,
      "num_tokens": 24786169.0,
      "step": 11455
    },
    {
      "entropy": 6.46388463973999,
      "epoch": 1.2263363475841405,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048593514382164123,
      "loss": 5.9788,
      "mean_token_accuracy": 0.15872699171304702,
      "num_tokens": 24796218.0,
      "step": 11460
    },
    {
      "entropy": 6.476220369338989,
      "epoch": 1.2268714216919043,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004859218350517814,
      "loss": 6.1321,
      "mean_token_accuracy": 0.1455128699541092,
      "num_tokens": 24806957.0,
      "step": 11465
    },
    {
      "entropy": 6.473389720916748,
      "epoch": 1.2274064957996682,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004859085201915831,
      "loss": 6.0478,
      "mean_token_accuracy": 0.15187956839799882,
      "num_tokens": 24818761.0,
      "step": 11470
    },
    {
      "entropy": 6.4047716617584225,
      "epoch": 1.227941569907432,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004858951992414311,
      "loss": 5.9812,
      "mean_token_accuracy": 0.16243064850568772,
      "num_tokens": 24829163.0,
      "step": 11475
    },
    {
      "entropy": 6.384244966506958,
      "epoch": 1.2284766440151962,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004858818722017097,
      "loss": 5.9937,
      "mean_token_accuracy": 0.151690986007452,
      "num_tokens": 24839334.0,
      "step": 11480
    },
    {
      "entropy": 6.366472625732422,
      "epoch": 1.22901171812296,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048586853907280395,
      "loss": 6.033,
      "mean_token_accuracy": 0.15324890315532685,
      "num_tokens": 24850312.0,
      "step": 11485
    },
    {
      "entropy": 6.404538917541504,
      "epoch": 1.229546792230724,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004858551998550987,
      "loss": 5.9711,
      "mean_token_accuracy": 0.16004097759723662,
      "num_tokens": 24861835.0,
      "step": 11490
    },
    {
      "entropy": 6.4540643215179445,
      "epoch": 1.2300818663384878,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048584185454897925,
      "loss": 6.0617,
      "mean_token_accuracy": 0.15071349442005158,
      "num_tokens": 24872489.0,
      "step": 11495
    },
    {
      "entropy": 6.430825471878052,
      "epoch": 1.230616940446252,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004858285031548309,
      "loss": 6.0529,
      "mean_token_accuracy": 0.15249309539794922,
      "num_tokens": 24882553.0,
      "step": 11500
    },
    {
      "entropy": 6.464387845993042,
      "epoch": 1.2311520145540158,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004858151456730391,
      "loss": 6.0542,
      "mean_token_accuracy": 0.1572585418820381,
      "num_tokens": 24894386.0,
      "step": 11505
    },
    {
      "entropy": 6.3799848556518555,
      "epoch": 1.2316870886617797,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048580178210398954,
      "loss": 6.0517,
      "mean_token_accuracy": 0.1476918339729309,
      "num_tokens": 24905049.0,
      "step": 11510
    },
    {
      "entropy": 6.346324253082275,
      "epoch": 1.2322221627695435,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048578841244806823,
      "loss": 6.0259,
      "mean_token_accuracy": 0.15691085606813432,
      "num_tokens": 24914831.0,
      "step": 11515
    },
    {
      "entropy": 6.400723600387574,
      "epoch": 1.2327572368773074,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048577503670566116,
      "loss": 6.0728,
      "mean_token_accuracy": 0.15282805263996124,
      "num_tokens": 24925668.0,
      "step": 11520
    },
    {
      "entropy": 6.435077381134033,
      "epoch": 1.2332923109850715,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048576165487715443,
      "loss": 6.0437,
      "mean_token_accuracy": 0.15154671669006348,
      "num_tokens": 24935983.0,
      "step": 11525
    },
    {
      "entropy": 6.393455553054809,
      "epoch": 1.2338273850928354,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048574826696293455,
      "loss": 6.0195,
      "mean_token_accuracy": 0.15619745403528212,
      "num_tokens": 24946673.0,
      "step": 11530
    },
    {
      "entropy": 6.374335908889771,
      "epoch": 1.2343624592005993,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048573487296338814,
      "loss": 5.935,
      "mean_token_accuracy": 0.16131605505943297,
      "num_tokens": 24957075.0,
      "step": 11535
    },
    {
      "entropy": 6.4100484371185305,
      "epoch": 1.2348975333083632,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048572147287890183,
      "loss": 6.0774,
      "mean_token_accuracy": 0.14921035170555114,
      "num_tokens": 24967261.0,
      "step": 11540
    },
    {
      "entropy": 6.3941281795501705,
      "epoch": 1.235432607416127,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00048570806670986263,
      "loss": 6.0151,
      "mean_token_accuracy": 0.15478278696537018,
      "num_tokens": 24978097.0,
      "step": 11545
    },
    {
      "entropy": 6.444482707977295,
      "epoch": 1.2359676815238911,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048569465445665754,
      "loss": 6.0512,
      "mean_token_accuracy": 0.1500304102897644,
      "num_tokens": 24987932.0,
      "step": 11550
    },
    {
      "entropy": 6.436000347137451,
      "epoch": 1.236502755631655,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048568123611967396,
      "loss": 6.0411,
      "mean_token_accuracy": 0.15225322246551515,
      "num_tokens": 24998919.0,
      "step": 11555
    },
    {
      "entropy": 6.352374505996704,
      "epoch": 1.237037829739419,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004856678116992993,
      "loss": 6.0028,
      "mean_token_accuracy": 0.15388862490653993,
      "num_tokens": 25010545.0,
      "step": 11560
    },
    {
      "entropy": 6.371043491363525,
      "epoch": 1.2375729038471828,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004856543811959212,
      "loss": 5.9778,
      "mean_token_accuracy": 0.16465644538402557,
      "num_tokens": 25021133.0,
      "step": 11565
    },
    {
      "entropy": 6.4023792266845705,
      "epoch": 1.2381079779549466,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004856409446099274,
      "loss": 6.0109,
      "mean_token_accuracy": 0.1508641794323921,
      "num_tokens": 25031838.0,
      "step": 11570
    },
    {
      "entropy": 6.434319972991943,
      "epoch": 1.2386430520627107,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048562750194170595,
      "loss": 6.0494,
      "mean_token_accuracy": 0.14732189923524858,
      "num_tokens": 25042737.0,
      "step": 11575
    },
    {
      "entropy": 6.467510795593261,
      "epoch": 1.2391781261704746,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000485614053191645,
      "loss": 6.1329,
      "mean_token_accuracy": 0.15031993091106416,
      "num_tokens": 25054442.0,
      "step": 11580
    },
    {
      "entropy": 6.328031826019287,
      "epoch": 1.2397132002782385,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004856005983601328,
      "loss": 5.8704,
      "mean_token_accuracy": 0.165472574532032,
      "num_tokens": 25065509.0,
      "step": 11585
    },
    {
      "entropy": 6.462453365325928,
      "epoch": 1.2402482743860024,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048558713744755794,
      "loss": 6.1203,
      "mean_token_accuracy": 0.14285247772932053,
      "num_tokens": 25077306.0,
      "step": 11590
    },
    {
      "entropy": 6.450064277648925,
      "epoch": 1.2407833484937665,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004855736704543091,
      "loss": 6.039,
      "mean_token_accuracy": 0.15463847517967225,
      "num_tokens": 25088230.0,
      "step": 11595
    },
    {
      "entropy": 6.504342889785766,
      "epoch": 1.2413184226015304,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004855601973807751,
      "loss": 6.1335,
      "mean_token_accuracy": 0.14597578942775727,
      "num_tokens": 25099658.0,
      "step": 11600
    },
    {
      "entropy": 6.396210193634033,
      "epoch": 1.2418534967092942,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048554671822734495,
      "loss": 6.0188,
      "mean_token_accuracy": 0.15247201025485993,
      "num_tokens": 25110512.0,
      "step": 11605
    },
    {
      "entropy": 6.4116230487823485,
      "epoch": 1.2423885708170581,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048553323299440807,
      "loss": 6.0409,
      "mean_token_accuracy": 0.15291347801685334,
      "num_tokens": 25120193.0,
      "step": 11610
    },
    {
      "entropy": 6.409094333648682,
      "epoch": 1.242923644924822,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048551974168235346,
      "loss": 6.0098,
      "mean_token_accuracy": 0.1537759855389595,
      "num_tokens": 25131752.0,
      "step": 11615
    },
    {
      "entropy": 6.437418508529663,
      "epoch": 1.243458719032586,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000485506244291571,
      "loss": 6.0682,
      "mean_token_accuracy": 0.15034523010253906,
      "num_tokens": 25141541.0,
      "step": 11620
    },
    {
      "entropy": 6.3503830909729,
      "epoch": 1.24399379314035,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004854927408224503,
      "loss": 6.0348,
      "mean_token_accuracy": 0.1480984792113304,
      "num_tokens": 25152219.0,
      "step": 11625
    },
    {
      "entropy": 6.471954870223999,
      "epoch": 1.2445288672481138,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048547923127538126,
      "loss": 6.0643,
      "mean_token_accuracy": 0.1474498689174652,
      "num_tokens": 25163030.0,
      "step": 11630
    },
    {
      "entropy": 6.407954978942871,
      "epoch": 1.2450639413558777,
      "grad_norm": 0.828125,
      "learning_rate": 0.00048546571565075396,
      "loss": 5.927,
      "mean_token_accuracy": 0.16227685213088988,
      "num_tokens": 25174637.0,
      "step": 11635
    },
    {
      "entropy": 6.353334665298462,
      "epoch": 1.2455990154636418,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004854521939489587,
      "loss": 5.9403,
      "mean_token_accuracy": 0.1612449675798416,
      "num_tokens": 25185067.0,
      "step": 11640
    },
    {
      "entropy": 6.370711135864258,
      "epoch": 1.2461340895714057,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048543866617038604,
      "loss": 5.9789,
      "mean_token_accuracy": 0.15409952253103257,
      "num_tokens": 25195046.0,
      "step": 11645
    },
    {
      "entropy": 6.358227109909057,
      "epoch": 1.2466691636791696,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048542513231542635,
      "loss": 6.0608,
      "mean_token_accuracy": 0.14790542423725128,
      "num_tokens": 25206265.0,
      "step": 11650
    },
    {
      "entropy": 6.48202953338623,
      "epoch": 1.2472042377869335,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004854115923844705,
      "loss": 6.0616,
      "mean_token_accuracy": 0.145122529566288,
      "num_tokens": 25217690.0,
      "step": 11655
    },
    {
      "entropy": 6.4813519477844235,
      "epoch": 1.2477393118946973,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004853980463779096,
      "loss": 6.0694,
      "mean_token_accuracy": 0.14623805731534958,
      "num_tokens": 25228229.0,
      "step": 11660
    },
    {
      "entropy": 6.393381929397583,
      "epoch": 1.2482743860024614,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004853844942961346,
      "loss": 5.9987,
      "mean_token_accuracy": 0.15335666686296462,
      "num_tokens": 25240073.0,
      "step": 11665
    },
    {
      "entropy": 6.386109972000122,
      "epoch": 1.2488094601102253,
      "grad_norm": 0.859375,
      "learning_rate": 0.000485370936139537,
      "loss": 6.0454,
      "mean_token_accuracy": 0.154733282327652,
      "num_tokens": 25251915.0,
      "step": 11670
    },
    {
      "entropy": 6.361781263351441,
      "epoch": 1.2493445342179892,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000485357371908508,
      "loss": 6.0321,
      "mean_token_accuracy": 0.15205752104520798,
      "num_tokens": 25262932.0,
      "step": 11675
    },
    {
      "entropy": 6.418450355529785,
      "epoch": 1.249879608325753,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004853438016034396,
      "loss": 6.0866,
      "mean_token_accuracy": 0.15124862492084504,
      "num_tokens": 25274693.0,
      "step": 11680
    },
    {
      "entropy": 6.41953182220459,
      "epoch": 1.250414682433517,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048533022522472344,
      "loss": 5.9723,
      "mean_token_accuracy": 0.15495878607034683,
      "num_tokens": 25285359.0,
      "step": 11685
    },
    {
      "entropy": 6.363399696350098,
      "epoch": 1.250949756541281,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048531664277275154,
      "loss": 6.0005,
      "mean_token_accuracy": 0.15141404271125794,
      "num_tokens": 25296730.0,
      "step": 11690
    },
    {
      "entropy": 6.366191101074219,
      "epoch": 1.251484830649045,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004853030542479162,
      "loss": 6.0453,
      "mean_token_accuracy": 0.15787868499755858,
      "num_tokens": 25306422.0,
      "step": 11695
    },
    {
      "entropy": 6.421235990524292,
      "epoch": 1.2520199047568088,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004852894596506096,
      "loss": 6.0639,
      "mean_token_accuracy": 0.15201850980520248,
      "num_tokens": 25317980.0,
      "step": 11700
    },
    {
      "entropy": 6.41616382598877,
      "epoch": 1.2525549788645727,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004852758589812245,
      "loss": 6.0889,
      "mean_token_accuracy": 0.15790138393640518,
      "num_tokens": 25329905.0,
      "step": 11705
    },
    {
      "entropy": 6.488639974594117,
      "epoch": 1.2530900529723366,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048526225224015355,
      "loss": 6.0589,
      "mean_token_accuracy": 0.14479833841323853,
      "num_tokens": 25340759.0,
      "step": 11710
    },
    {
      "entropy": 6.392478036880493,
      "epoch": 1.2536251270801007,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004852486394277896,
      "loss": 6.08,
      "mean_token_accuracy": 0.1448793701827526,
      "num_tokens": 25351124.0,
      "step": 11715
    },
    {
      "entropy": 6.408259153366089,
      "epoch": 1.2541602011878645,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048523502054452565,
      "loss": 6.0212,
      "mean_token_accuracy": 0.15003727972507477,
      "num_tokens": 25361337.0,
      "step": 11720
    },
    {
      "entropy": 6.4600615978240965,
      "epoch": 1.2546952752956284,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048522139559075507,
      "loss": 6.0509,
      "mean_token_accuracy": 0.1574355334043503,
      "num_tokens": 25371893.0,
      "step": 11725
    },
    {
      "entropy": 6.345823907852173,
      "epoch": 1.2552303494033923,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004852077645668712,
      "loss": 5.939,
      "mean_token_accuracy": 0.15976056605577468,
      "num_tokens": 25383365.0,
      "step": 11730
    },
    {
      "entropy": 6.422120094299316,
      "epoch": 1.2557654235111562,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004851941274732678,
      "loss": 6.0422,
      "mean_token_accuracy": 0.1508241578936577,
      "num_tokens": 25393211.0,
      "step": 11735
    },
    {
      "entropy": 6.323185539245605,
      "epoch": 1.2563004976189203,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004851804843103384,
      "loss": 5.9161,
      "mean_token_accuracy": 0.15132512301206588,
      "num_tokens": 25404828.0,
      "step": 11740
    },
    {
      "entropy": 6.420544910430908,
      "epoch": 1.2568355717266841,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048516683507847705,
      "loss": 6.0381,
      "mean_token_accuracy": 0.16127409189939498,
      "num_tokens": 25416016.0,
      "step": 11745
    },
    {
      "entropy": 6.39495816230774,
      "epoch": 1.257370645834448,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004851531797780779,
      "loss": 5.9922,
      "mean_token_accuracy": 0.15947735011577607,
      "num_tokens": 25426173.0,
      "step": 11750
    },
    {
      "entropy": 6.372569370269775,
      "epoch": 1.2579057199422121,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004851395184095352,
      "loss": 5.9638,
      "mean_token_accuracy": 0.15989621430635453,
      "num_tokens": 25437507.0,
      "step": 11755
    },
    {
      "entropy": 6.415957498550415,
      "epoch": 1.258440794049976,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004851258509732434,
      "loss": 6.0743,
      "mean_token_accuracy": 0.14236538112163544,
      "num_tokens": 25448201.0,
      "step": 11760
    },
    {
      "entropy": 6.442714023590088,
      "epoch": 1.2589758681577399,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048511217746959723,
      "loss": 6.0656,
      "mean_token_accuracy": 0.14910233840346337,
      "num_tokens": 25458640.0,
      "step": 11765
    },
    {
      "entropy": 6.429221773147583,
      "epoch": 1.2595109422655038,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048509849789899146,
      "loss": 6.0535,
      "mean_token_accuracy": 0.1583869442343712,
      "num_tokens": 25469474.0,
      "step": 11770
    },
    {
      "entropy": 6.482155895233154,
      "epoch": 1.2600460163732676,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000485084812261821,
      "loss": 6.1445,
      "mean_token_accuracy": 0.14766237139701843,
      "num_tokens": 25479565.0,
      "step": 11775
    },
    {
      "entropy": 6.450917720794678,
      "epoch": 1.2605810904810317,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004850711205584812,
      "loss": 6.0096,
      "mean_token_accuracy": 0.15592256784439087,
      "num_tokens": 25490142.0,
      "step": 11780
    },
    {
      "entropy": 6.387892770767212,
      "epoch": 1.2611161645887956,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048505742278936726,
      "loss": 6.1071,
      "mean_token_accuracy": 0.151397642493248,
      "num_tokens": 25500824.0,
      "step": 11785
    },
    {
      "entropy": 6.458943605422974,
      "epoch": 1.2616512386965595,
      "grad_norm": 1.0,
      "learning_rate": 0.00048504371895487474,
      "loss": 6.0426,
      "mean_token_accuracy": 0.14935458302497864,
      "num_tokens": 25511251.0,
      "step": 11790
    },
    {
      "entropy": 6.414369535446167,
      "epoch": 1.2621863128043234,
      "grad_norm": 0.84375,
      "learning_rate": 0.00048503000905539945,
      "loss": 6.0294,
      "mean_token_accuracy": 0.14942921698093414,
      "num_tokens": 25521368.0,
      "step": 11795
    },
    {
      "entropy": 6.385678625106811,
      "epoch": 1.2627213869120872,
      "grad_norm": 0.875,
      "learning_rate": 0.00048501629309133707,
      "loss": 6.0228,
      "mean_token_accuracy": 0.1489413097500801,
      "num_tokens": 25532486.0,
      "step": 11800
    },
    {
      "entropy": 6.4228119373321535,
      "epoch": 1.2632564610198513,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004850025710630838,
      "loss": 6.0214,
      "mean_token_accuracy": 0.1576980546116829,
      "num_tokens": 25542867.0,
      "step": 11805
    },
    {
      "entropy": 6.408396863937378,
      "epoch": 1.2637915351276152,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048498884297103575,
      "loss": 5.9895,
      "mean_token_accuracy": 0.15152256190776825,
      "num_tokens": 25553564.0,
      "step": 11810
    },
    {
      "entropy": 6.381280422210693,
      "epoch": 1.264326609235379,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004849751088155894,
      "loss": 5.9917,
      "mean_token_accuracy": 0.15123338401317596,
      "num_tokens": 25564017.0,
      "step": 11815
    },
    {
      "entropy": 6.346936941146851,
      "epoch": 1.264861683343143,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004849613685971413,
      "loss": 5.9438,
      "mean_token_accuracy": 0.1556112661957741,
      "num_tokens": 25575125.0,
      "step": 11820
    },
    {
      "entropy": 6.404648065567017,
      "epoch": 1.2653967574509069,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004849476223160882,
      "loss": 6.0852,
      "mean_token_accuracy": 0.14894888401031495,
      "num_tokens": 25586405.0,
      "step": 11825
    },
    {
      "entropy": 6.436088705062867,
      "epoch": 1.265931831558671,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004849338699728269,
      "loss": 6.0291,
      "mean_token_accuracy": 0.1516052097082138,
      "num_tokens": 25597494.0,
      "step": 11830
    },
    {
      "entropy": 6.39005184173584,
      "epoch": 1.2664669056664348,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048492011156775467,
      "loss": 6.0089,
      "mean_token_accuracy": 0.1586216852068901,
      "num_tokens": 25608425.0,
      "step": 11835
    },
    {
      "entropy": 6.420915508270264,
      "epoch": 1.2670019797741987,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004849063471012688,
      "loss": 6.0345,
      "mean_token_accuracy": 0.1431473083794117,
      "num_tokens": 25619411.0,
      "step": 11840
    },
    {
      "entropy": 6.464434003829956,
      "epoch": 1.2675370538819626,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004848925765737665,
      "loss": 6.0789,
      "mean_token_accuracy": 0.1435423456132412,
      "num_tokens": 25630477.0,
      "step": 11845
    },
    {
      "entropy": 6.434617757797241,
      "epoch": 1.2680721279897265,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048487879998564565,
      "loss": 5.9852,
      "mean_token_accuracy": 0.15037410855293273,
      "num_tokens": 25640129.0,
      "step": 11850
    },
    {
      "entropy": 6.409026432037353,
      "epoch": 1.2686072020974906,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048486501733730386,
      "loss": 6.0641,
      "mean_token_accuracy": 0.15462838411331176,
      "num_tokens": 25650592.0,
      "step": 11855
    },
    {
      "entropy": 6.361433696746826,
      "epoch": 1.2691422762052544,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048485122862913934,
      "loss": 5.9838,
      "mean_token_accuracy": 0.15777004659175872,
      "num_tokens": 25660408.0,
      "step": 11860
    },
    {
      "entropy": 6.3860142707824705,
      "epoch": 1.2696773503130183,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048483743386154994,
      "loss": 6.0677,
      "mean_token_accuracy": 0.15449703186750413,
      "num_tokens": 25671563.0,
      "step": 11865
    },
    {
      "entropy": 6.459661436080933,
      "epoch": 1.2702124244207824,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048482363303493415,
      "loss": 6.0044,
      "mean_token_accuracy": 0.15084402784705161,
      "num_tokens": 25682474.0,
      "step": 11870
    },
    {
      "entropy": 6.485248136520386,
      "epoch": 1.270747498528546,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004848098261496905,
      "loss": 6.0198,
      "mean_token_accuracy": 0.1508568525314331,
      "num_tokens": 25692594.0,
      "step": 11875
    },
    {
      "entropy": 6.271194171905518,
      "epoch": 1.2712825726363102,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004847960132062175,
      "loss": 5.8569,
      "mean_token_accuracy": 0.16723863482475282,
      "num_tokens": 25703724.0,
      "step": 11880
    },
    {
      "entropy": 6.439691829681396,
      "epoch": 1.271817646744074,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004847821942049142,
      "loss": 6.1626,
      "mean_token_accuracy": 0.13662642538547515,
      "num_tokens": 25714911.0,
      "step": 11885
    },
    {
      "entropy": 6.49635877609253,
      "epoch": 1.272352720851838,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048476836914617945,
      "loss": 6.0216,
      "mean_token_accuracy": 0.15288260728120803,
      "num_tokens": 25725098.0,
      "step": 11890
    },
    {
      "entropy": 6.4539141178131105,
      "epoch": 1.272887794959602,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048475453803041254,
      "loss": 6.0099,
      "mean_token_accuracy": 0.1464683398604393,
      "num_tokens": 25735099.0,
      "step": 11895
    },
    {
      "entropy": 6.402464962005615,
      "epoch": 1.273422869067366,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004847407008580128,
      "loss": 6.0006,
      "mean_token_accuracy": 0.1539877399802208,
      "num_tokens": 25746852.0,
      "step": 11900
    },
    {
      "entropy": 6.389354801177978,
      "epoch": 1.2739579431751298,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048472685762937983,
      "loss": 6.0761,
      "mean_token_accuracy": 0.1493126332759857,
      "num_tokens": 25758579.0,
      "step": 11905
    },
    {
      "entropy": 6.361133050918579,
      "epoch": 1.2744930172828937,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00048471300834491325,
      "loss": 6.0525,
      "mean_token_accuracy": 0.1477493703365326,
      "num_tokens": 25769965.0,
      "step": 11910
    },
    {
      "entropy": 6.444669914245606,
      "epoch": 1.2750280913906575,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004846991530050132,
      "loss": 6.0312,
      "mean_token_accuracy": 0.15700097233057023,
      "num_tokens": 25780264.0,
      "step": 11915
    },
    {
      "entropy": 6.4759704113006595,
      "epoch": 1.2755631654984216,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004846852916100794,
      "loss": 5.9803,
      "mean_token_accuracy": 0.15219346284866334,
      "num_tokens": 25791228.0,
      "step": 11920
    },
    {
      "entropy": 6.454332637786865,
      "epoch": 1.2760982396061855,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004846714241605123,
      "loss": 5.9726,
      "mean_token_accuracy": 0.15598224848508835,
      "num_tokens": 25802428.0,
      "step": 11925
    },
    {
      "entropy": 6.410422945022583,
      "epoch": 1.2766333137139494,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048465755065671233,
      "loss": 6.0398,
      "mean_token_accuracy": 0.15440476089715957,
      "num_tokens": 25814220.0,
      "step": 11930
    },
    {
      "entropy": 6.4318338394165036,
      "epoch": 1.2771683878217133,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048464367109908005,
      "loss": 6.041,
      "mean_token_accuracy": 0.15025397688150405,
      "num_tokens": 25824382.0,
      "step": 11935
    },
    {
      "entropy": 6.349942588806153,
      "epoch": 1.2777034619294771,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004846297854880162,
      "loss": 5.9957,
      "mean_token_accuracy": 0.1535354658961296,
      "num_tokens": 25834896.0,
      "step": 11940
    },
    {
      "entropy": 6.428943729400634,
      "epoch": 1.2782385360372412,
      "grad_norm": 0.80859375,
      "learning_rate": 0.00048461589382392176,
      "loss": 6.0979,
      "mean_token_accuracy": 0.14953776374459266,
      "num_tokens": 25847300.0,
      "step": 11945
    },
    {
      "entropy": 6.441252183914185,
      "epoch": 1.2787736101450051,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004846019961071978,
      "loss": 6.0329,
      "mean_token_accuracy": 0.15231782123446463,
      "num_tokens": 25858599.0,
      "step": 11950
    },
    {
      "entropy": 6.423742008209229,
      "epoch": 1.279308684252769,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004845880923382457,
      "loss": 5.9881,
      "mean_token_accuracy": 0.15524668395519256,
      "num_tokens": 25869135.0,
      "step": 11955
    },
    {
      "entropy": 6.438772106170655,
      "epoch": 1.2798437583605329,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048457418251746695,
      "loss": 5.9772,
      "mean_token_accuracy": 0.16028833240270615,
      "num_tokens": 25879234.0,
      "step": 11960
    },
    {
      "entropy": 6.444809532165527,
      "epoch": 1.2803788324682968,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004845602666452629,
      "loss": 6.035,
      "mean_token_accuracy": 0.15932807177305222,
      "num_tokens": 25888230.0,
      "step": 11965
    },
    {
      "entropy": 6.4251810073852536,
      "epoch": 1.2809139065760609,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004845463447220358,
      "loss": 6.0458,
      "mean_token_accuracy": 0.16299064755439757,
      "num_tokens": 25899429.0,
      "step": 11970
    },
    {
      "entropy": 6.352957391738892,
      "epoch": 1.2814489806838247,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004845324167481874,
      "loss": 5.9172,
      "mean_token_accuracy": 0.16644244641065598,
      "num_tokens": 25911042.0,
      "step": 11975
    },
    {
      "entropy": 6.428217601776123,
      "epoch": 1.2819840547915886,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004845184827241199,
      "loss": 5.9783,
      "mean_token_accuracy": 0.1567392498254776,
      "num_tokens": 25921466.0,
      "step": 11980
    },
    {
      "entropy": 6.455148983001709,
      "epoch": 1.2825191288993525,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004845045426502357,
      "loss": 6.1081,
      "mean_token_accuracy": 0.14141944870352746,
      "num_tokens": 25931059.0,
      "step": 11985
    },
    {
      "entropy": 6.466062402725219,
      "epoch": 1.2830542030071164,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048449059652693724,
      "loss": 6.0264,
      "mean_token_accuracy": 0.1564402088522911,
      "num_tokens": 25941669.0,
      "step": 11990
    },
    {
      "entropy": 6.383812761306762,
      "epoch": 1.2835892771148805,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004844766443546272,
      "loss": 6.0431,
      "mean_token_accuracy": 0.1557509034872055,
      "num_tokens": 25952066.0,
      "step": 11995
    },
    {
      "entropy": 6.423954248428345,
      "epoch": 1.2841243512226443,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004844626861337085,
      "loss": 5.9369,
      "mean_token_accuracy": 0.15952450633049012,
      "num_tokens": 25962388.0,
      "step": 12000
    },
    {
      "epoch": 1.2841243512226443,
      "eval_entropy": 6.1652326883331465,
      "eval_loss": 6.136476039886475,
      "eval_mean_token_accuracy": 0.15487081866003471,
      "eval_num_tokens": 25962388.0,
      "eval_runtime": 17.5197,
      "eval_samples_per_second": 1694.033,
      "eval_steps_per_second": 211.761,
      "step": 12000
    },
    {
      "entropy": 6.311177206039429,
      "epoch": 1.2846594253304082,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004844487218645843,
      "loss": 5.9933,
      "mean_token_accuracy": 0.14980660527944564,
      "num_tokens": 25973865.0,
      "step": 12005
    },
    {
      "entropy": 6.367836380004883,
      "epoch": 1.2851944994381723,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004844347515476575,
      "loss": 5.9915,
      "mean_token_accuracy": 0.15764360129833221,
      "num_tokens": 25985113.0,
      "step": 12010
    },
    {
      "entropy": 6.372952079772949,
      "epoch": 1.285729573545936,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048442077518333175,
      "loss": 5.9136,
      "mean_token_accuracy": 0.16656481772661208,
      "num_tokens": 25995980.0,
      "step": 12015
    },
    {
      "entropy": 6.360746812820435,
      "epoch": 1.2862646476537,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048440679277201063,
      "loss": 6.0029,
      "mean_token_accuracy": 0.1548726573586464,
      "num_tokens": 26006362.0,
      "step": 12020
    },
    {
      "entropy": 6.459201335906982,
      "epoch": 1.286799721761464,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048439280431409774,
      "loss": 6.0979,
      "mean_token_accuracy": 0.1495984748005867,
      "num_tokens": 26016623.0,
      "step": 12025
    },
    {
      "entropy": 6.332912826538086,
      "epoch": 1.2873347958692278,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048437880980999694,
      "loss": 5.9027,
      "mean_token_accuracy": 0.1569906145334244,
      "num_tokens": 26027202.0,
      "step": 12030
    },
    {
      "entropy": 6.370944786071777,
      "epoch": 1.287869869976992,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004843648092601125,
      "loss": 6.0085,
      "mean_token_accuracy": 0.1493479236960411,
      "num_tokens": 26037330.0,
      "step": 12035
    },
    {
      "entropy": 6.447657775878906,
      "epoch": 1.2884049440847558,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004843508026648487,
      "loss": 6.0197,
      "mean_token_accuracy": 0.15304963588714598,
      "num_tokens": 26047661.0,
      "step": 12040
    },
    {
      "entropy": 6.463361024856567,
      "epoch": 1.2889400181925197,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004843367900246098,
      "loss": 6.0619,
      "mean_token_accuracy": 0.1487477608025074,
      "num_tokens": 26058582.0,
      "step": 12045
    },
    {
      "entropy": 6.38573431968689,
      "epoch": 1.2894750923002836,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048432277133980055,
      "loss": 6.062,
      "mean_token_accuracy": 0.14511267989873886,
      "num_tokens": 26069117.0,
      "step": 12050
    },
    {
      "entropy": 6.405645275115967,
      "epoch": 1.2900101664080474,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048430874661082565,
      "loss": 6.0289,
      "mean_token_accuracy": 0.15712926238775254,
      "num_tokens": 26079327.0,
      "step": 12055
    },
    {
      "entropy": 6.5040069103240965,
      "epoch": 1.2905452405158115,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048429471583809016,
      "loss": 6.0543,
      "mean_token_accuracy": 0.1460406668484211,
      "num_tokens": 26090709.0,
      "step": 12060
    },
    {
      "entropy": 6.463672876358032,
      "epoch": 1.2910803146235754,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004842806790219991,
      "loss": 6.0428,
      "mean_token_accuracy": 0.1506498008966446,
      "num_tokens": 26101570.0,
      "step": 12065
    },
    {
      "entropy": 6.353507137298584,
      "epoch": 1.2916153887313393,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048426663616295783,
      "loss": 5.9907,
      "mean_token_accuracy": 0.1524586260318756,
      "num_tokens": 26112827.0,
      "step": 12070
    },
    {
      "entropy": 6.421887588500977,
      "epoch": 1.2921504628391032,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048425258726137187,
      "loss": 6.0698,
      "mean_token_accuracy": 0.14892441406846046,
      "num_tokens": 26124141.0,
      "step": 12075
    },
    {
      "entropy": 6.410809469223023,
      "epoch": 1.292685536946867,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048423853231764684,
      "loss": 6.0079,
      "mean_token_accuracy": 0.15069296360015869,
      "num_tokens": 26135372.0,
      "step": 12080
    },
    {
      "entropy": 6.392371797561646,
      "epoch": 1.2932206110546312,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048422447133218867,
      "loss": 5.9907,
      "mean_token_accuracy": 0.1505424976348877,
      "num_tokens": 26145933.0,
      "step": 12085
    },
    {
      "entropy": 6.446551322937012,
      "epoch": 1.293755685162395,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048421040430540323,
      "loss": 6.0564,
      "mean_token_accuracy": 0.14921719431877137,
      "num_tokens": 26155948.0,
      "step": 12090
    },
    {
      "entropy": 6.367982196807861,
      "epoch": 1.294290759270159,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004841963312376967,
      "loss": 5.9675,
      "mean_token_accuracy": 0.16145270466804504,
      "num_tokens": 26166410.0,
      "step": 12095
    },
    {
      "entropy": 6.3258376121521,
      "epoch": 1.2948258333779228,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048418225212947566,
      "loss": 5.9994,
      "mean_token_accuracy": 0.15143728405237197,
      "num_tokens": 26177334.0,
      "step": 12100
    },
    {
      "entropy": 6.486021709442139,
      "epoch": 1.2953609074856867,
      "grad_norm": 1.0,
      "learning_rate": 0.0004841681669811464,
      "loss": 6.0599,
      "mean_token_accuracy": 0.14494219198822975,
      "num_tokens": 26187987.0,
      "step": 12105
    },
    {
      "entropy": 6.459728193283081,
      "epoch": 1.2958959815934508,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004841540757931157,
      "loss": 6.0175,
      "mean_token_accuracy": 0.1463063709437847,
      "num_tokens": 26198600.0,
      "step": 12110
    },
    {
      "entropy": 6.424509143829345,
      "epoch": 1.2964310557012146,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004841399785657904,
      "loss": 6.0496,
      "mean_token_accuracy": 0.14998798072338104,
      "num_tokens": 26210977.0,
      "step": 12115
    },
    {
      "entropy": 6.407670545578003,
      "epoch": 1.2969661298089785,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048412587529957774,
      "loss": 5.9988,
      "mean_token_accuracy": 0.15616723895072937,
      "num_tokens": 26220813.0,
      "step": 12120
    },
    {
      "entropy": 6.415078353881836,
      "epoch": 1.2975012039167424,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004841117659948848,
      "loss": 6.0436,
      "mean_token_accuracy": 0.15693661719560623,
      "num_tokens": 26232042.0,
      "step": 12125
    },
    {
      "entropy": 6.453728914260864,
      "epoch": 1.2980362780245063,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048409765065211884,
      "loss": 6.0487,
      "mean_token_accuracy": 0.14782098680734634,
      "num_tokens": 26242968.0,
      "step": 12130
    },
    {
      "entropy": 6.4647626876831055,
      "epoch": 1.2985713521322704,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048408352927168774,
      "loss": 6.0435,
      "mean_token_accuracy": 0.15253500044345855,
      "num_tokens": 26252751.0,
      "step": 12135
    },
    {
      "entropy": 6.427189874649048,
      "epoch": 1.2991064262400343,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004840694018539991,
      "loss": 6.1087,
      "mean_token_accuracy": 0.14439072981476783,
      "num_tokens": 26263270.0,
      "step": 12140
    },
    {
      "entropy": 6.442751836776734,
      "epoch": 1.2996415003477981,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048405526839946086,
      "loss": 6.1078,
      "mean_token_accuracy": 0.14654383957386016,
      "num_tokens": 26274279.0,
      "step": 12145
    },
    {
      "entropy": 6.428610610961914,
      "epoch": 1.3001765744555622,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048404112890848105,
      "loss": 6.1101,
      "mean_token_accuracy": 0.15012672394514084,
      "num_tokens": 26286020.0,
      "step": 12150
    },
    {
      "entropy": 6.473810005187988,
      "epoch": 1.3007116485633259,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048402698338146803,
      "loss": 6.0218,
      "mean_token_accuracy": 0.15520235747098923,
      "num_tokens": 26296798.0,
      "step": 12155
    },
    {
      "entropy": 6.383613395690918,
      "epoch": 1.30124672267109,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004840128318188303,
      "loss": 6.0184,
      "mean_token_accuracy": 0.14978406801819802,
      "num_tokens": 26306952.0,
      "step": 12160
    },
    {
      "entropy": 6.461830043792725,
      "epoch": 1.3017817967788539,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048399867422097644,
      "loss": 6.0852,
      "mean_token_accuracy": 0.14953851848840713,
      "num_tokens": 26317686.0,
      "step": 12165
    },
    {
      "entropy": 6.4471354484558105,
      "epoch": 1.3023168708866177,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048398451058831514,
      "loss": 6.0651,
      "mean_token_accuracy": 0.15492951571941377,
      "num_tokens": 26327997.0,
      "step": 12170
    },
    {
      "entropy": 6.39400086402893,
      "epoch": 1.3028519449943818,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048397034092125554,
      "loss": 6.0373,
      "mean_token_accuracy": 0.1507429674267769,
      "num_tokens": 26340227.0,
      "step": 12175
    },
    {
      "entropy": 6.553567600250244,
      "epoch": 1.3033870191021457,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048395616522020667,
      "loss": 6.1255,
      "mean_token_accuracy": 0.1529696300625801,
      "num_tokens": 26350549.0,
      "step": 12180
    },
    {
      "entropy": 6.456924200057983,
      "epoch": 1.3039220932099096,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004839419834855779,
      "loss": 6.0411,
      "mean_token_accuracy": 0.15041446387767793,
      "num_tokens": 26361822.0,
      "step": 12185
    },
    {
      "entropy": 6.338099193572998,
      "epoch": 1.3044571673176735,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004839277957177787,
      "loss": 6.0666,
      "mean_token_accuracy": 0.15103471428155898,
      "num_tokens": 26372720.0,
      "step": 12190
    },
    {
      "entropy": 6.374748706817627,
      "epoch": 1.3049922414254373,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004839136019172188,
      "loss": 6.0269,
      "mean_token_accuracy": 0.15443735867738723,
      "num_tokens": 26383092.0,
      "step": 12195
    },
    {
      "entropy": 6.386470365524292,
      "epoch": 1.3055273155332014,
      "grad_norm": 0.9375,
      "learning_rate": 0.000483899402084308,
      "loss": 5.9764,
      "mean_token_accuracy": 0.1609979450702667,
      "num_tokens": 26395188.0,
      "step": 12200
    },
    {
      "entropy": 6.401113033294678,
      "epoch": 1.3060623896409653,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004838851962194563,
      "loss": 6.1064,
      "mean_token_accuracy": 0.14932459145784377,
      "num_tokens": 26406338.0,
      "step": 12205
    },
    {
      "entropy": 6.407237577438354,
      "epoch": 1.3065974637487292,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004838709843230739,
      "loss": 6.0199,
      "mean_token_accuracy": 0.15152190923690795,
      "num_tokens": 26416983.0,
      "step": 12210
    },
    {
      "entropy": 6.456840515136719,
      "epoch": 1.307132537856493,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004838567663955712,
      "loss": 5.9938,
      "mean_token_accuracy": 0.15696968138217926,
      "num_tokens": 26426966.0,
      "step": 12215
    },
    {
      "entropy": 6.391991901397705,
      "epoch": 1.307667611964257,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004838425424373587,
      "loss": 5.949,
      "mean_token_accuracy": 0.1517357423901558,
      "num_tokens": 26437711.0,
      "step": 12220
    },
    {
      "entropy": 6.422004842758179,
      "epoch": 1.308202686072021,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004838283124488472,
      "loss": 6.087,
      "mean_token_accuracy": 0.14809246361255646,
      "num_tokens": 26448657.0,
      "step": 12225
    },
    {
      "entropy": 6.455289745330811,
      "epoch": 1.308737760179785,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048381407643044737,
      "loss": 5.9782,
      "mean_token_accuracy": 0.1566404476761818,
      "num_tokens": 26459363.0,
      "step": 12230
    },
    {
      "entropy": 6.382435846328735,
      "epoch": 1.3092728342875488,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004837998343825705,
      "loss": 6.0123,
      "mean_token_accuracy": 0.1560998558998108,
      "num_tokens": 26470149.0,
      "step": 12235
    },
    {
      "entropy": 6.383670425415039,
      "epoch": 1.3098079083953127,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048378558630562783,
      "loss": 5.98,
      "mean_token_accuracy": 0.16018408834934234,
      "num_tokens": 26480424.0,
      "step": 12240
    },
    {
      "entropy": 6.408652305603027,
      "epoch": 1.3103429825030766,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004837713322000306,
      "loss": 5.9971,
      "mean_token_accuracy": 0.15148263350129126,
      "num_tokens": 26490312.0,
      "step": 12245
    },
    {
      "entropy": 6.339053678512573,
      "epoch": 1.3108780566108407,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004837570720661905,
      "loss": 6.0022,
      "mean_token_accuracy": 0.16391705721616745,
      "num_tokens": 26501119.0,
      "step": 12250
    },
    {
      "entropy": 6.427730941772461,
      "epoch": 1.3114131307186045,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048374280590451934,
      "loss": 6.0572,
      "mean_token_accuracy": 0.15475394576787949,
      "num_tokens": 26512438.0,
      "step": 12255
    },
    {
      "entropy": 6.385207033157348,
      "epoch": 1.3119482048263684,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00048372853371542895,
      "loss": 6.0232,
      "mean_token_accuracy": 0.14870542287826538,
      "num_tokens": 26523930.0,
      "step": 12260
    },
    {
      "entropy": 6.449709272384643,
      "epoch": 1.3124832789341323,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004837142554993315,
      "loss": 6.11,
      "mean_token_accuracy": 0.15153739005327224,
      "num_tokens": 26534937.0,
      "step": 12265
    },
    {
      "entropy": 6.366942501068115,
      "epoch": 1.3130183530418962,
      "grad_norm": 1.0,
      "learning_rate": 0.0004836999712566392,
      "loss": 5.8983,
      "mean_token_accuracy": 0.1573762334883213,
      "num_tokens": 26544555.0,
      "step": 12270
    },
    {
      "entropy": 6.411316823959351,
      "epoch": 1.3135534271496603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048368568098776463,
      "loss": 6.0321,
      "mean_token_accuracy": 0.16070346683263778,
      "num_tokens": 26556144.0,
      "step": 12275
    },
    {
      "entropy": 6.4581047058105465,
      "epoch": 1.3140885012574242,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004836713846931203,
      "loss": 6.0974,
      "mean_token_accuracy": 0.1408153861761093,
      "num_tokens": 26567350.0,
      "step": 12280
    },
    {
      "entropy": 6.384568119049073,
      "epoch": 1.314623575365188,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048365708237311904,
      "loss": 6.0033,
      "mean_token_accuracy": 0.1574316531419754,
      "num_tokens": 26579486.0,
      "step": 12285
    },
    {
      "entropy": 6.415039968490601,
      "epoch": 1.3151586494729521,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004836427740281739,
      "loss": 6.0031,
      "mean_token_accuracy": 0.15133218318223954,
      "num_tokens": 26589665.0,
      "step": 12290
    },
    {
      "entropy": 6.4360028266906735,
      "epoch": 1.3156937235807158,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048362845965869793,
      "loss": 6.0266,
      "mean_token_accuracy": 0.15072643235325814,
      "num_tokens": 26600187.0,
      "step": 12295
    },
    {
      "entropy": 6.410405969619751,
      "epoch": 1.31622879768848,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004836141392651046,
      "loss": 5.9763,
      "mean_token_accuracy": 0.1584290772676468,
      "num_tokens": 26612155.0,
      "step": 12300
    },
    {
      "entropy": 6.412408876419067,
      "epoch": 1.3167638717962438,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048359981284780727,
      "loss": 6.0289,
      "mean_token_accuracy": 0.14702292978763581,
      "num_tokens": 26623077.0,
      "step": 12305
    },
    {
      "entropy": 6.3540013313293455,
      "epoch": 1.3172989459040076,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004835854804072196,
      "loss": 5.9658,
      "mean_token_accuracy": 0.158090803027153,
      "num_tokens": 26632979.0,
      "step": 12310
    },
    {
      "entropy": 6.4409887313842775,
      "epoch": 1.3178340200117717,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048357114194375555,
      "loss": 6.0789,
      "mean_token_accuracy": 0.14963133484125138,
      "num_tokens": 26644377.0,
      "step": 12315
    },
    {
      "entropy": 6.429935026168823,
      "epoch": 1.3183690941195356,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000483556797457829,
      "loss": 6.0478,
      "mean_token_accuracy": 0.1449607327580452,
      "num_tokens": 26655538.0,
      "step": 12320
    },
    {
      "entropy": 6.404887247085571,
      "epoch": 1.3189041682272995,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048354244694985435,
      "loss": 5.9941,
      "mean_token_accuracy": 0.15169127359986306,
      "num_tokens": 26668011.0,
      "step": 12325
    },
    {
      "entropy": 6.399375915527344,
      "epoch": 1.3194392423350634,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004835280904202457,
      "loss": 6.0386,
      "mean_token_accuracy": 0.14957074150443078,
      "num_tokens": 26678996.0,
      "step": 12330
    },
    {
      "entropy": 6.431111478805542,
      "epoch": 1.3199743164428273,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004835137278694178,
      "loss": 5.9893,
      "mean_token_accuracy": 0.15844448655843735,
      "num_tokens": 26689041.0,
      "step": 12335
    },
    {
      "entropy": 6.420051431655883,
      "epoch": 1.3205093905505914,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004834993592977854,
      "loss": 5.9964,
      "mean_token_accuracy": 0.1592840790748596,
      "num_tokens": 26699459.0,
      "step": 12340
    },
    {
      "entropy": 6.422905683517456,
      "epoch": 1.3210444646583552,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004834849847057632,
      "loss": 6.1079,
      "mean_token_accuracy": 0.14611226618289946,
      "num_tokens": 26711663.0,
      "step": 12345
    },
    {
      "entropy": 6.45347957611084,
      "epoch": 1.321579538766119,
      "grad_norm": 0.875,
      "learning_rate": 0.00048347060409376644,
      "loss": 6.0668,
      "mean_token_accuracy": 0.1469884358346462,
      "num_tokens": 26722960.0,
      "step": 12350
    },
    {
      "entropy": 6.3834065914154055,
      "epoch": 1.322114612873883,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048345621746221024,
      "loss": 5.9805,
      "mean_token_accuracy": 0.1598796382546425,
      "num_tokens": 26733274.0,
      "step": 12355
    },
    {
      "entropy": 6.44045877456665,
      "epoch": 1.3226496869816469,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048344182481151003,
      "loss": 6.0471,
      "mean_token_accuracy": 0.14724795073270797,
      "num_tokens": 26743906.0,
      "step": 12360
    },
    {
      "entropy": 6.371076583862305,
      "epoch": 1.323184761089411,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048342742614208145,
      "loss": 6.0386,
      "mean_token_accuracy": 0.1496793136000633,
      "num_tokens": 26755093.0,
      "step": 12365
    },
    {
      "entropy": 6.412216377258301,
      "epoch": 1.3237198351971748,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004834130214543402,
      "loss": 6.0137,
      "mean_token_accuracy": 0.1545899510383606,
      "num_tokens": 26765614.0,
      "step": 12370
    },
    {
      "entropy": 6.382581186294556,
      "epoch": 1.3242549093049387,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048339861074870227,
      "loss": 6.0173,
      "mean_token_accuracy": 0.1531946524977684,
      "num_tokens": 26778078.0,
      "step": 12375
    },
    {
      "entropy": 6.402309226989746,
      "epoch": 1.3247899834127026,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048338419402558375,
      "loss": 6.0283,
      "mean_token_accuracy": 0.15170913487672805,
      "num_tokens": 26788707.0,
      "step": 12380
    },
    {
      "entropy": 6.454970407485962,
      "epoch": 1.3253250575204665,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004833697712854009,
      "loss": 5.9975,
      "mean_token_accuracy": 0.157017220556736,
      "num_tokens": 26798650.0,
      "step": 12385
    },
    {
      "entropy": 6.403705835342407,
      "epoch": 1.3258601316282306,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048335534252857016,
      "loss": 6.0454,
      "mean_token_accuracy": 0.15259954035282136,
      "num_tokens": 26809822.0,
      "step": 12390
    },
    {
      "entropy": 6.390819549560547,
      "epoch": 1.3263952057359945,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004833409077555082,
      "loss": 6.0126,
      "mean_token_accuracy": 0.14680972397327424,
      "num_tokens": 26821061.0,
      "step": 12395
    },
    {
      "entropy": 6.4620825290679935,
      "epoch": 1.3269302798437583,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048332646696663177,
      "loss": 6.0781,
      "mean_token_accuracy": 0.14716461002826692,
      "num_tokens": 26832020.0,
      "step": 12400
    },
    {
      "entropy": 6.439187860488891,
      "epoch": 1.3274653539515222,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004833120201623579,
      "loss": 6.0307,
      "mean_token_accuracy": 0.14897628575563432,
      "num_tokens": 26844287.0,
      "step": 12405
    },
    {
      "entropy": 6.294268655776977,
      "epoch": 1.328000428059286,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048329756734310375,
      "loss": 5.914,
      "mean_token_accuracy": 0.15903573334217072,
      "num_tokens": 26855790.0,
      "step": 12410
    },
    {
      "entropy": 6.413055515289306,
      "epoch": 1.3285355021670502,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048328310850928656,
      "loss": 6.0292,
      "mean_token_accuracy": 0.153592998534441,
      "num_tokens": 26867391.0,
      "step": 12415
    },
    {
      "entropy": 6.453096628189087,
      "epoch": 1.329070576274814,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048326864366132397,
      "loss": 6.0379,
      "mean_token_accuracy": 0.15639247223734856,
      "num_tokens": 26879173.0,
      "step": 12420
    },
    {
      "entropy": 6.430088520050049,
      "epoch": 1.329605650382578,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004832541727996335,
      "loss": 5.9403,
      "mean_token_accuracy": 0.16483087837696075,
      "num_tokens": 26889839.0,
      "step": 12425
    },
    {
      "entropy": 6.332772970199585,
      "epoch": 1.330140724490342,
      "grad_norm": 0.96875,
      "learning_rate": 0.000483239695924633,
      "loss": 6.0185,
      "mean_token_accuracy": 0.15671744495630263,
      "num_tokens": 26900217.0,
      "step": 12430
    },
    {
      "entropy": 6.393820810317993,
      "epoch": 1.330675798598106,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048322521303674064,
      "loss": 6.0134,
      "mean_token_accuracy": 0.1579091504216194,
      "num_tokens": 26911366.0,
      "step": 12435
    },
    {
      "entropy": 6.463123798370361,
      "epoch": 1.3312108727058698,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048321072413637454,
      "loss": 6.0577,
      "mean_token_accuracy": 0.14822871908545493,
      "num_tokens": 26922135.0,
      "step": 12440
    },
    {
      "entropy": 6.394211959838867,
      "epoch": 1.3317459468136337,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004831962292239529,
      "loss": 6.0039,
      "mean_token_accuracy": 0.15831962674856187,
      "num_tokens": 26932502.0,
      "step": 12445
    },
    {
      "entropy": 6.389608240127563,
      "epoch": 1.3322810209213976,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048318172829989447,
      "loss": 6.0292,
      "mean_token_accuracy": 0.1574964389204979,
      "num_tokens": 26944607.0,
      "step": 12450
    },
    {
      "entropy": 6.442240905761719,
      "epoch": 1.3328160950291617,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004831672213646179,
      "loss": 6.0623,
      "mean_token_accuracy": 0.15060238391160966,
      "num_tokens": 26955205.0,
      "step": 12455
    },
    {
      "entropy": 6.384679126739502,
      "epoch": 1.3333511691369255,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004831527084185421,
      "loss": 5.9227,
      "mean_token_accuracy": 0.16366988122463227,
      "num_tokens": 26965558.0,
      "step": 12460
    },
    {
      "entropy": 6.396430397033692,
      "epoch": 1.3338862432446894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000483138189462086,
      "loss": 6.0049,
      "mean_token_accuracy": 0.15095641762018203,
      "num_tokens": 26976176.0,
      "step": 12465
    },
    {
      "entropy": 6.458672046661377,
      "epoch": 1.3344213173524533,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048312366449566895,
      "loss": 6.0661,
      "mean_token_accuracy": 0.15261408537626267,
      "num_tokens": 26987099.0,
      "step": 12470
    },
    {
      "entropy": 6.3919806480407715,
      "epoch": 1.3349563914602172,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004831091335197104,
      "loss": 6.0093,
      "mean_token_accuracy": 0.1528696224093437,
      "num_tokens": 26998122.0,
      "step": 12475
    },
    {
      "entropy": 6.361894512176514,
      "epoch": 1.3354914655679813,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004830945965346298,
      "loss": 5.9665,
      "mean_token_accuracy": 0.1561110347509384,
      "num_tokens": 27009427.0,
      "step": 12480
    },
    {
      "entropy": 6.473658418655395,
      "epoch": 1.3360265396757451,
      "grad_norm": 0.984375,
      "learning_rate": 0.000483080053540847,
      "loss": 6.1111,
      "mean_token_accuracy": 0.14574675038456916,
      "num_tokens": 27020701.0,
      "step": 12485
    },
    {
      "entropy": 6.463243818283081,
      "epoch": 1.336561613783509,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004830655045387818,
      "loss": 6.0648,
      "mean_token_accuracy": 0.14396844804286957,
      "num_tokens": 27033215.0,
      "step": 12490
    },
    {
      "entropy": 6.442213726043701,
      "epoch": 1.337096687891273,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048305094952885453,
      "loss": 6.0463,
      "mean_token_accuracy": 0.14670848101377487,
      "num_tokens": 27043846.0,
      "step": 12495
    },
    {
      "entropy": 6.387128925323486,
      "epoch": 1.3376317619990368,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004830363885114853,
      "loss": 6.027,
      "mean_token_accuracy": 0.1502802163362503,
      "num_tokens": 27054783.0,
      "step": 12500
    },
    {
      "entropy": 6.3706752300262455,
      "epoch": 1.3381668361068009,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004830218214870946,
      "loss": 6.0109,
      "mean_token_accuracy": 0.1545276537537575,
      "num_tokens": 27066051.0,
      "step": 12505
    },
    {
      "entropy": 6.462768125534057,
      "epoch": 1.3387019102145647,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004830072484561029,
      "loss": 6.034,
      "mean_token_accuracy": 0.15640046298503876,
      "num_tokens": 27078210.0,
      "step": 12510
    },
    {
      "entropy": 6.4500679016113285,
      "epoch": 1.3392369843223286,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004829926694189312,
      "loss": 6.0732,
      "mean_token_accuracy": 0.15013144463300704,
      "num_tokens": 27089693.0,
      "step": 12515
    },
    {
      "entropy": 6.41504168510437,
      "epoch": 1.3397720584300925,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004829780843760005,
      "loss": 6.0599,
      "mean_token_accuracy": 0.15103112161159515,
      "num_tokens": 27099856.0,
      "step": 12520
    },
    {
      "entropy": 6.455632448196411,
      "epoch": 1.3403071325378564,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004829634933277317,
      "loss": 6.0499,
      "mean_token_accuracy": 0.1518527165055275,
      "num_tokens": 27110218.0,
      "step": 12525
    },
    {
      "entropy": 6.442966318130493,
      "epoch": 1.3408422066456205,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048294889627454636,
      "loss": 6.01,
      "mean_token_accuracy": 0.15706087350845338,
      "num_tokens": 27120911.0,
      "step": 12530
    },
    {
      "entropy": 6.359386444091797,
      "epoch": 1.3413772807533844,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048293429321686585,
      "loss": 5.9983,
      "mean_token_accuracy": 0.1556496351957321,
      "num_tokens": 27130964.0,
      "step": 12535
    },
    {
      "entropy": 6.415383243560791,
      "epoch": 1.3419123548611482,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048291968415511174,
      "loss": 6.0056,
      "mean_token_accuracy": 0.15176327228546144,
      "num_tokens": 27141641.0,
      "step": 12540
    },
    {
      "entropy": 6.4677286624908445,
      "epoch": 1.3424474289689121,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048290506908970607,
      "loss": 6.0799,
      "mean_token_accuracy": 0.14796556234359742,
      "num_tokens": 27152591.0,
      "step": 12545
    },
    {
      "entropy": 6.486439180374146,
      "epoch": 1.342982503076676,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004828904480210707,
      "loss": 6.1553,
      "mean_token_accuracy": 0.14118096232414246,
      "num_tokens": 27163839.0,
      "step": 12550
    },
    {
      "entropy": 6.437562894821167,
      "epoch": 1.34351757718444,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004828758209496278,
      "loss": 6.0211,
      "mean_token_accuracy": 0.14874017238616943,
      "num_tokens": 27174480.0,
      "step": 12555
    },
    {
      "entropy": 6.343012285232544,
      "epoch": 1.344052651292204,
      "grad_norm": 1.0,
      "learning_rate": 0.00048286118787579983,
      "loss": 6.0136,
      "mean_token_accuracy": 0.14849736616015435,
      "num_tokens": 27184546.0,
      "step": 12560
    },
    {
      "entropy": 6.373420143127442,
      "epoch": 1.3445877253999678,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048284654880000933,
      "loss": 5.8916,
      "mean_token_accuracy": 0.16075090169906617,
      "num_tokens": 27195093.0,
      "step": 12565
    },
    {
      "entropy": 6.425318384170533,
      "epoch": 1.345122799507732,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048283190372267886,
      "loss": 6.0667,
      "mean_token_accuracy": 0.1503993585705757,
      "num_tokens": 27206416.0,
      "step": 12570
    },
    {
      "entropy": 6.4284348487854,
      "epoch": 1.3456578736154958,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004828172526442314,
      "loss": 5.9942,
      "mean_token_accuracy": 0.15560751110315324,
      "num_tokens": 27215533.0,
      "step": 12575
    },
    {
      "entropy": 6.4487693309783936,
      "epoch": 1.3461929477232597,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048280259556509,
      "loss": 6.0438,
      "mean_token_accuracy": 0.15142483860254288,
      "num_tokens": 27225985.0,
      "step": 12580
    },
    {
      "entropy": 6.416169309616089,
      "epoch": 1.3467280218310236,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004827879324856778,
      "loss": 6.007,
      "mean_token_accuracy": 0.15581531673669816,
      "num_tokens": 27235967.0,
      "step": 12585
    },
    {
      "entropy": 6.407985734939575,
      "epoch": 1.3472630959387875,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048277326340641835,
      "loss": 5.9352,
      "mean_token_accuracy": 0.15629177391529084,
      "num_tokens": 27246854.0,
      "step": 12590
    },
    {
      "entropy": 6.387296581268311,
      "epoch": 1.3477981700465516,
      "grad_norm": 1.0,
      "learning_rate": 0.000482758588327735,
      "loss": 6.0679,
      "mean_token_accuracy": 0.1540767177939415,
      "num_tokens": 27256640.0,
      "step": 12595
    },
    {
      "entropy": 6.429937505722046,
      "epoch": 1.3483332441543154,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004827439072500516,
      "loss": 6.0222,
      "mean_token_accuracy": 0.15238182842731476,
      "num_tokens": 27266907.0,
      "step": 12600
    },
    {
      "entropy": 6.438490295410157,
      "epoch": 1.3488683182620793,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00048272922017379217,
      "loss": 6.0149,
      "mean_token_accuracy": 0.14828513860702514,
      "num_tokens": 27277089.0,
      "step": 12605
    },
    {
      "entropy": 6.423541688919068,
      "epoch": 1.3494033923698432,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004827145270993807,
      "loss": 5.981,
      "mean_token_accuracy": 0.1546955168247223,
      "num_tokens": 27286781.0,
      "step": 12610
    },
    {
      "entropy": 6.434300279617309,
      "epoch": 1.349938466477607,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004826998280272414,
      "loss": 6.0056,
      "mean_token_accuracy": 0.15781160444021225,
      "num_tokens": 27298011.0,
      "step": 12615
    },
    {
      "entropy": 6.419676685333252,
      "epoch": 1.3504735405853712,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048268512295779873,
      "loss": 6.0463,
      "mean_token_accuracy": 0.15662085115909577,
      "num_tokens": 27309052.0,
      "step": 12620
    },
    {
      "entropy": 6.341011190414429,
      "epoch": 1.351008614693135,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004826704118914773,
      "loss": 5.9674,
      "mean_token_accuracy": 0.15662161707878114,
      "num_tokens": 27319945.0,
      "step": 12625
    },
    {
      "entropy": 6.411014699935913,
      "epoch": 1.351543688800899,
      "grad_norm": 1.0,
      "learning_rate": 0.000482655694828702,
      "loss": 6.0036,
      "mean_token_accuracy": 0.163016477227211,
      "num_tokens": 27330543.0,
      "step": 12630
    },
    {
      "entropy": 6.443450927734375,
      "epoch": 1.3520787629086628,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004826409717698976,
      "loss": 6.0513,
      "mean_token_accuracy": 0.14720099419355392,
      "num_tokens": 27341957.0,
      "step": 12635
    },
    {
      "entropy": 6.393955421447754,
      "epoch": 1.3526138370164267,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048262624271548936,
      "loss": 5.9897,
      "mean_token_accuracy": 0.1530117064714432,
      "num_tokens": 27351506.0,
      "step": 12640
    },
    {
      "entropy": 6.370419692993164,
      "epoch": 1.3531489111241908,
      "grad_norm": 1.0,
      "learning_rate": 0.00048261150766590257,
      "loss": 5.9702,
      "mean_token_accuracy": 0.15135697722435,
      "num_tokens": 27362208.0,
      "step": 12645
    },
    {
      "entropy": 6.4273213863372805,
      "epoch": 1.3536839852319547,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048259676662156263,
      "loss": 5.9799,
      "mean_token_accuracy": 0.15203302800655366,
      "num_tokens": 27372890.0,
      "step": 12650
    },
    {
      "entropy": 6.3724956035614015,
      "epoch": 1.3542190593397185,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004825820195828952,
      "loss": 5.9483,
      "mean_token_accuracy": 0.15766446590423583,
      "num_tokens": 27382879.0,
      "step": 12655
    },
    {
      "entropy": 6.386530065536499,
      "epoch": 1.3547541334474824,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004825672665503262,
      "loss": 5.9637,
      "mean_token_accuracy": 0.1581568241119385,
      "num_tokens": 27393878.0,
      "step": 12660
    },
    {
      "entropy": 6.400862026214599,
      "epoch": 1.3552892075552463,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004825525075242815,
      "loss": 6.0672,
      "mean_token_accuracy": 0.15068575590848923,
      "num_tokens": 27404928.0,
      "step": 12665
    },
    {
      "entropy": 6.4651120662689205,
      "epoch": 1.3558242816630104,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004825377425051873,
      "loss": 5.9983,
      "mean_token_accuracy": 0.1499782994389534,
      "num_tokens": 27417473.0,
      "step": 12670
    },
    {
      "entropy": 6.4452478885650635,
      "epoch": 1.3563593557707743,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048252297149346997,
      "loss": 6.0692,
      "mean_token_accuracy": 0.14973496198654174,
      "num_tokens": 27428378.0,
      "step": 12675
    },
    {
      "entropy": 6.317526865005493,
      "epoch": 1.3568944298785381,
      "grad_norm": 1.0,
      "learning_rate": 0.00048250819448955595,
      "loss": 6.031,
      "mean_token_accuracy": 0.15386673361063002,
      "num_tokens": 27438815.0,
      "step": 12680
    },
    {
      "entropy": 6.475922918319702,
      "epoch": 1.357429503986302,
      "grad_norm": 0.921875,
      "learning_rate": 0.000482493411493872,
      "loss": 6.0709,
      "mean_token_accuracy": 0.15083241909742356,
      "num_tokens": 27449295.0,
      "step": 12685
    },
    {
      "entropy": 6.389287042617798,
      "epoch": 1.357964578094066,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000482478622506845,
      "loss": 5.9406,
      "mean_token_accuracy": 0.16012440621852875,
      "num_tokens": 27459357.0,
      "step": 12690
    },
    {
      "entropy": 6.406489467620849,
      "epoch": 1.35849965220183,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004824638275289019,
      "loss": 5.9866,
      "mean_token_accuracy": 0.15113565772771836,
      "num_tokens": 27469405.0,
      "step": 12695
    },
    {
      "entropy": 6.394630813598633,
      "epoch": 1.3590347263095939,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004824490265604699,
      "loss": 5.9986,
      "mean_token_accuracy": 0.15261455699801446,
      "num_tokens": 27478818.0,
      "step": 12700
    },
    {
      "entropy": 6.384332942962646,
      "epoch": 1.3595698004173578,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004824342196019764,
      "loss": 5.9511,
      "mean_token_accuracy": 0.16350412666797637,
      "num_tokens": 27488631.0,
      "step": 12705
    },
    {
      "entropy": 6.4433784008026125,
      "epoch": 1.3601048745251219,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048241940665384904,
      "loss": 6.0283,
      "mean_token_accuracy": 0.15179991722106934,
      "num_tokens": 27500895.0,
      "step": 12710
    },
    {
      "entropy": 6.395802974700928,
      "epoch": 1.3606399486328857,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004824045877165154,
      "loss": 6.0523,
      "mean_token_accuracy": 0.14968711733818055,
      "num_tokens": 27512578.0,
      "step": 12715
    },
    {
      "entropy": 6.398001718521118,
      "epoch": 1.3611750227406496,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048238976279040344,
      "loss": 6.0159,
      "mean_token_accuracy": 0.14981312900781632,
      "num_tokens": 27523448.0,
      "step": 12720
    },
    {
      "entropy": 6.411330986022949,
      "epoch": 1.3617100968484135,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004823749318759413,
      "loss": 6.0881,
      "mean_token_accuracy": 0.1469150185585022,
      "num_tokens": 27534919.0,
      "step": 12725
    },
    {
      "entropy": 6.371654033660889,
      "epoch": 1.3622451709561774,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000482360094973557,
      "loss": 5.905,
      "mean_token_accuracy": 0.15699642300605773,
      "num_tokens": 27545603.0,
      "step": 12730
    },
    {
      "entropy": 6.455313444137573,
      "epoch": 1.3627802450639415,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004823452520836792,
      "loss": 6.0261,
      "mean_token_accuracy": 0.15182247757911682,
      "num_tokens": 27556868.0,
      "step": 12735
    },
    {
      "entropy": 6.527916049957275,
      "epoch": 1.3633153191717053,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048233040320673634,
      "loss": 6.1659,
      "mean_token_accuracy": 0.1414632260799408,
      "num_tokens": 27567200.0,
      "step": 12740
    },
    {
      "entropy": 6.367152309417724,
      "epoch": 1.3638503932794692,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048231554834315716,
      "loss": 5.9785,
      "mean_token_accuracy": 0.15740661025047303,
      "num_tokens": 27577759.0,
      "step": 12745
    },
    {
      "entropy": 6.3653933048248295,
      "epoch": 1.364385467387233,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004823006874933708,
      "loss": 6.0696,
      "mean_token_accuracy": 0.15286897271871566,
      "num_tokens": 27588935.0,
      "step": 12750
    },
    {
      "entropy": 6.431991386413574,
      "epoch": 1.364920541494997,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004822858206578061,
      "loss": 6.0114,
      "mean_token_accuracy": 0.1501218095421791,
      "num_tokens": 27600306.0,
      "step": 12755
    },
    {
      "entropy": 6.43401141166687,
      "epoch": 1.365455615602761,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048227094783689244,
      "loss": 6.0516,
      "mean_token_accuracy": 0.15340078622102737,
      "num_tokens": 27612322.0,
      "step": 12760
    },
    {
      "entropy": 6.4369895458221436,
      "epoch": 1.365990689710525,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004822560690310594,
      "loss": 6.0022,
      "mean_token_accuracy": 0.16061661690473555,
      "num_tokens": 27623812.0,
      "step": 12765
    },
    {
      "entropy": 6.369098424911499,
      "epoch": 1.3665257638182888,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048224118424073644,
      "loss": 5.9011,
      "mean_token_accuracy": 0.15581171661615373,
      "num_tokens": 27633848.0,
      "step": 12770
    },
    {
      "entropy": 6.347079658508301,
      "epoch": 1.3670608379260527,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048222629346635336,
      "loss": 6.0097,
      "mean_token_accuracy": 0.1579328656196594,
      "num_tokens": 27644641.0,
      "step": 12775
    },
    {
      "entropy": 6.410189962387085,
      "epoch": 1.3675959120338166,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048221139670834025,
      "loss": 5.9912,
      "mean_token_accuracy": 0.16003549098968506,
      "num_tokens": 27655552.0,
      "step": 12780
    },
    {
      "entropy": 6.3974583625793455,
      "epoch": 1.3681309861415807,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004821964939671272,
      "loss": 6.0175,
      "mean_token_accuracy": 0.1592422679066658,
      "num_tokens": 27666396.0,
      "step": 12785
    },
    {
      "entropy": 6.442167663574219,
      "epoch": 1.3686660602493446,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004821815852431444,
      "loss": 5.9471,
      "mean_token_accuracy": 0.15841109305620193,
      "num_tokens": 27676522.0,
      "step": 12790
    },
    {
      "entropy": 6.367932367324829,
      "epoch": 1.3692011343571084,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048216667053682243,
      "loss": 5.9636,
      "mean_token_accuracy": 0.15813439935445786,
      "num_tokens": 27687088.0,
      "step": 12795
    },
    {
      "entropy": 6.420863962173462,
      "epoch": 1.3697362084648723,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048215174984859207,
      "loss": 5.9996,
      "mean_token_accuracy": 0.1591450899839401,
      "num_tokens": 27697587.0,
      "step": 12800
    },
    {
      "entropy": 6.365783929824829,
      "epoch": 1.3702712825726362,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000482136823178884,
      "loss": 5.9265,
      "mean_token_accuracy": 0.1576545186340809,
      "num_tokens": 27708345.0,
      "step": 12805
    },
    {
      "entropy": 6.442271137237549,
      "epoch": 1.3708063566804003,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004821218905281292,
      "loss": 6.1299,
      "mean_token_accuracy": 0.14670315608382226,
      "num_tokens": 27719592.0,
      "step": 12810
    },
    {
      "entropy": 6.430649662017823,
      "epoch": 1.3713414307881642,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000482106951896759,
      "loss": 5.998,
      "mean_token_accuracy": 0.15390112102031708,
      "num_tokens": 27730071.0,
      "step": 12815
    },
    {
      "entropy": 6.385264778137207,
      "epoch": 1.371876504895928,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048209200728520466,
      "loss": 6.0273,
      "mean_token_accuracy": 0.15088786333799362,
      "num_tokens": 27740514.0,
      "step": 12820
    },
    {
      "entropy": 6.460386562347412,
      "epoch": 1.372411579003692,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048207705669389765,
      "loss": 6.0444,
      "mean_token_accuracy": 0.14829810708761215,
      "num_tokens": 27752088.0,
      "step": 12825
    },
    {
      "entropy": 6.501808118820191,
      "epoch": 1.3729466531114558,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048206210012326976,
      "loss": 6.0557,
      "mean_token_accuracy": 0.14801618158817292,
      "num_tokens": 27763401.0,
      "step": 12830
    },
    {
      "entropy": 6.411707258224487,
      "epoch": 1.37348172721922,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048204713757375283,
      "loss": 5.9854,
      "mean_token_accuracy": 0.1569422721862793,
      "num_tokens": 27773665.0,
      "step": 12835
    },
    {
      "entropy": 6.393361663818359,
      "epoch": 1.3740168013269838,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004820321690457789,
      "loss": 6.008,
      "mean_token_accuracy": 0.15562692135572434,
      "num_tokens": 27785474.0,
      "step": 12840
    },
    {
      "entropy": 6.373070478439331,
      "epoch": 1.3745518754347477,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048201719453978023,
      "loss": 6.0322,
      "mean_token_accuracy": 0.1519990772008896,
      "num_tokens": 27796520.0,
      "step": 12845
    },
    {
      "entropy": 6.493841934204101,
      "epoch": 1.3750869495425118,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004820022140561891,
      "loss": 6.0805,
      "mean_token_accuracy": 0.14758755415678024,
      "num_tokens": 27807559.0,
      "step": 12850
    },
    {
      "entropy": 6.433234071731567,
      "epoch": 1.3756220236502756,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004819872275954381,
      "loss": 6.0302,
      "mean_token_accuracy": 0.15419595390558244,
      "num_tokens": 27818208.0,
      "step": 12855
    },
    {
      "entropy": 6.4251453399658205,
      "epoch": 1.3761570977580395,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048197223515796,
      "loss": 5.9491,
      "mean_token_accuracy": 0.16016671806573868,
      "num_tokens": 27828238.0,
      "step": 12860
    },
    {
      "entropy": 6.4240076541900635,
      "epoch": 1.3766921718658034,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004819572367441877,
      "loss": 6.022,
      "mean_token_accuracy": 0.1624337449669838,
      "num_tokens": 27837885.0,
      "step": 12865
    },
    {
      "entropy": 6.321288776397705,
      "epoch": 1.3772272459735673,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004819422323545543,
      "loss": 5.9404,
      "mean_token_accuracy": 0.16300138533115388,
      "num_tokens": 27847425.0,
      "step": 12870
    },
    {
      "entropy": 6.313952827453614,
      "epoch": 1.3777623200813314,
      "grad_norm": 0.828125,
      "learning_rate": 0.000481927221989493,
      "loss": 5.9732,
      "mean_token_accuracy": 0.1539781793951988,
      "num_tokens": 27859606.0,
      "step": 12875
    },
    {
      "entropy": 6.49549503326416,
      "epoch": 1.3782973941890952,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004819122056494373,
      "loss": 6.0616,
      "mean_token_accuracy": 0.15650817900896072,
      "num_tokens": 27871651.0,
      "step": 12880
    },
    {
      "entropy": 6.457492446899414,
      "epoch": 1.3788324682968591,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048189718333482064,
      "loss": 5.9771,
      "mean_token_accuracy": 0.15548114776611327,
      "num_tokens": 27883138.0,
      "step": 12885
    },
    {
      "entropy": 6.504308891296387,
      "epoch": 1.379367542404623,
      "grad_norm": 1.0,
      "learning_rate": 0.000481882155046077,
      "loss": 6.0483,
      "mean_token_accuracy": 0.1581483416259289,
      "num_tokens": 27893597.0,
      "step": 12890
    },
    {
      "entropy": 6.420593166351319,
      "epoch": 1.3799026165123869,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048186712078364006,
      "loss": 5.9616,
      "mean_token_accuracy": 0.16157536655664445,
      "num_tokens": 27904249.0,
      "step": 12895
    },
    {
      "entropy": 6.372634696960449,
      "epoch": 1.380437690620151,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048185208054794414,
      "loss": 6.0071,
      "mean_token_accuracy": 0.1528368905186653,
      "num_tokens": 27914713.0,
      "step": 12900
    },
    {
      "entropy": 6.345739555358887,
      "epoch": 1.3809727647279149,
      "grad_norm": 0.875,
      "learning_rate": 0.0004818370343394235,
      "loss": 6.0218,
      "mean_token_accuracy": 0.15834377110004424,
      "num_tokens": 27926257.0,
      "step": 12905
    },
    {
      "entropy": 6.411109972000122,
      "epoch": 1.3815078388356787,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004818219821585125,
      "loss": 5.9922,
      "mean_token_accuracy": 0.15973223894834518,
      "num_tokens": 27936839.0,
      "step": 12910
    },
    {
      "entropy": 6.5063948154449465,
      "epoch": 1.3820429129434426,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048180692400564586,
      "loss": 6.1126,
      "mean_token_accuracy": 0.14679630622267723,
      "num_tokens": 27946318.0,
      "step": 12915
    },
    {
      "entropy": 6.475949954986572,
      "epoch": 1.3825779870512065,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048179185988125834,
      "loss": 6.0956,
      "mean_token_accuracy": 0.14862671494483948,
      "num_tokens": 27956990.0,
      "step": 12920
    },
    {
      "entropy": 6.456042289733887,
      "epoch": 1.3831130611589706,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000481776789785785,
      "loss": 6.0312,
      "mean_token_accuracy": 0.14976766258478164,
      "num_tokens": 27968187.0,
      "step": 12925
    },
    {
      "entropy": 6.471331357955933,
      "epoch": 1.3836481352667345,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048176171371966083,
      "loss": 6.0761,
      "mean_token_accuracy": 0.1516728088259697,
      "num_tokens": 27978337.0,
      "step": 12930
    },
    {
      "entropy": 6.403414106369018,
      "epoch": 1.3841832093744983,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004817466316833212,
      "loss": 6.0418,
      "mean_token_accuracy": 0.14986416846513748,
      "num_tokens": 27988681.0,
      "step": 12935
    },
    {
      "entropy": 6.3656025409698485,
      "epoch": 1.3847182834822622,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004817315436772017,
      "loss": 5.9321,
      "mean_token_accuracy": 0.16186131089925765,
      "num_tokens": 27999233.0,
      "step": 12940
    },
    {
      "entropy": 6.46886248588562,
      "epoch": 1.385253357590026,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004817164497017379,
      "loss": 6.0532,
      "mean_token_accuracy": 0.1550150766968727,
      "num_tokens": 28010242.0,
      "step": 12945
    },
    {
      "entropy": 6.400705146789551,
      "epoch": 1.3857884316977902,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004817013497573656,
      "loss": 6.0587,
      "mean_token_accuracy": 0.14842417240142822,
      "num_tokens": 28020424.0,
      "step": 12950
    },
    {
      "entropy": 6.414020633697509,
      "epoch": 1.386323505805554,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000481686243844521,
      "loss": 6.0167,
      "mean_token_accuracy": 0.14889417141675948,
      "num_tokens": 28030287.0,
      "step": 12955
    },
    {
      "entropy": 6.43932204246521,
      "epoch": 1.386858579913318,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048167113196364,
      "loss": 5.9712,
      "mean_token_accuracy": 0.15472842901945114,
      "num_tokens": 28039997.0,
      "step": 12960
    },
    {
      "entropy": 6.395161724090576,
      "epoch": 1.387393654021082,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004816560141151593,
      "loss": 6.0045,
      "mean_token_accuracy": 0.14893352687358857,
      "num_tokens": 28051381.0,
      "step": 12965
    },
    {
      "entropy": 6.385078001022339,
      "epoch": 1.3879287281288457,
      "grad_norm": 1.125,
      "learning_rate": 0.0004816408902995151,
      "loss": 5.9576,
      "mean_token_accuracy": 0.15524173378944398,
      "num_tokens": 28063603.0,
      "step": 12970
    },
    {
      "entropy": 6.405039072036743,
      "epoch": 1.3884638022366098,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004816257605171442,
      "loss": 5.9678,
      "mean_token_accuracy": 0.16037806868553162,
      "num_tokens": 28073942.0,
      "step": 12975
    },
    {
      "entropy": 6.399739122390747,
      "epoch": 1.3889988763443737,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048161062476848363,
      "loss": 6.0043,
      "mean_token_accuracy": 0.15513721555471421,
      "num_tokens": 28083837.0,
      "step": 12980
    },
    {
      "entropy": 6.3457330703735355,
      "epoch": 1.3895339504521376,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004815954830539702,
      "loss": 5.939,
      "mean_token_accuracy": 0.15939247459173203,
      "num_tokens": 28094150.0,
      "step": 12985
    },
    {
      "entropy": 6.385166311264038,
      "epoch": 1.3900690245599017,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048158033537404126,
      "loss": 5.9965,
      "mean_token_accuracy": 0.16018508076667787,
      "num_tokens": 28106345.0,
      "step": 12990
    },
    {
      "entropy": 6.397221183776855,
      "epoch": 1.3906040986676655,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048156518172913417,
      "loss": 5.9234,
      "mean_token_accuracy": 0.15279222130775452,
      "num_tokens": 28116858.0,
      "step": 12995
    },
    {
      "entropy": 6.411376428604126,
      "epoch": 1.3911391727754294,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004815500221196865,
      "loss": 6.0949,
      "mean_token_accuracy": 0.15006548762321473,
      "num_tokens": 28128390.0,
      "step": 13000
    },
    {
      "entropy": 6.348651647567749,
      "epoch": 1.3916742468831933,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048153485654613587,
      "loss": 5.9528,
      "mean_token_accuracy": 0.16057345569133757,
      "num_tokens": 28139421.0,
      "step": 13005
    },
    {
      "entropy": 6.4372154712677006,
      "epoch": 1.3922093209909572,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048151968500892043,
      "loss": 6.0391,
      "mean_token_accuracy": 0.15470600575208665,
      "num_tokens": 28151076.0,
      "step": 13010
    },
    {
      "entropy": 6.485320901870727,
      "epoch": 1.3927443950987213,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048150450750847795,
      "loss": 6.145,
      "mean_token_accuracy": 0.15247728377580644,
      "num_tokens": 28162079.0,
      "step": 13015
    },
    {
      "entropy": 6.431665515899658,
      "epoch": 1.3932794692064852,
      "grad_norm": 0.890625,
      "learning_rate": 0.000481489324045247,
      "loss": 6.0706,
      "mean_token_accuracy": 0.15153184533119202,
      "num_tokens": 28174442.0,
      "step": 13020
    },
    {
      "entropy": 6.420172166824341,
      "epoch": 1.393814543314249,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048147413461966564,
      "loss": 5.9364,
      "mean_token_accuracy": 0.15835790634155272,
      "num_tokens": 28184643.0,
      "step": 13025
    },
    {
      "entropy": 6.32468090057373,
      "epoch": 1.394349617422013,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048145893923217276,
      "loss": 5.9421,
      "mean_token_accuracy": 0.15326581597328187,
      "num_tokens": 28194021.0,
      "step": 13030
    },
    {
      "entropy": 6.432333946228027,
      "epoch": 1.3948846915297768,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004814437378832071,
      "loss": 6.0204,
      "mean_token_accuracy": 0.147834425419569,
      "num_tokens": 28204750.0,
      "step": 13035
    },
    {
      "entropy": 6.4994570255279545,
      "epoch": 1.3954197656375409,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004814285305732074,
      "loss": 6.0646,
      "mean_token_accuracy": 0.15128121376037598,
      "num_tokens": 28216499.0,
      "step": 13040
    },
    {
      "entropy": 6.462815618515014,
      "epoch": 1.3959548397453048,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048141331730261293,
      "loss": 6.0149,
      "mean_token_accuracy": 0.15481323897838592,
      "num_tokens": 28226573.0,
      "step": 13045
    },
    {
      "entropy": 6.382149744033813,
      "epoch": 1.3964899138530686,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000481398098071863,
      "loss": 5.9826,
      "mean_token_accuracy": 0.15709102898836136,
      "num_tokens": 28236688.0,
      "step": 13050
    },
    {
      "entropy": 6.370840406417846,
      "epoch": 1.3970249879608325,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048138287288139686,
      "loss": 6.0408,
      "mean_token_accuracy": 0.15631090402603148,
      "num_tokens": 28247477.0,
      "step": 13055
    },
    {
      "entropy": 6.369821548461914,
      "epoch": 1.3975600620685964,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004813676417316543,
      "loss": 6.0264,
      "mean_token_accuracy": 0.1599254697561264,
      "num_tokens": 28258286.0,
      "step": 13060
    },
    {
      "entropy": 6.523264598846436,
      "epoch": 1.3980951361763605,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004813524046230751,
      "loss": 6.1885,
      "mean_token_accuracy": 0.1423366278409958,
      "num_tokens": 28268587.0,
      "step": 13065
    },
    {
      "entropy": 6.514555597305298,
      "epoch": 1.3986302102841244,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048133716155609926,
      "loss": 6.0554,
      "mean_token_accuracy": 0.14311629682779312,
      "num_tokens": 28279352.0,
      "step": 13070
    },
    {
      "entropy": 6.402999639511108,
      "epoch": 1.3991652843918883,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004813219125311668,
      "loss": 5.9902,
      "mean_token_accuracy": 0.15347336381673812,
      "num_tokens": 28290114.0,
      "step": 13075
    },
    {
      "entropy": 6.397921180725097,
      "epoch": 1.3997003584996521,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048130665754871814,
      "loss": 6.064,
      "mean_token_accuracy": 0.1498659923672676,
      "num_tokens": 28301845.0,
      "step": 13080
    },
    {
      "entropy": 6.484706068038941,
      "epoch": 1.400235432607416,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048129139660919374,
      "loss": 6.0546,
      "mean_token_accuracy": 0.1492530472576618,
      "num_tokens": 28313162.0,
      "step": 13085
    },
    {
      "entropy": 6.432747077941895,
      "epoch": 1.40077050671518,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048127612971303425,
      "loss": 5.9959,
      "mean_token_accuracy": 0.15192358940839767,
      "num_tokens": 28323888.0,
      "step": 13090
    },
    {
      "entropy": 6.412349033355713,
      "epoch": 1.401305580822944,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004812608568606805,
      "loss": 6.0255,
      "mean_token_accuracy": 0.15089134499430656,
      "num_tokens": 28333735.0,
      "step": 13095
    },
    {
      "entropy": 6.446710395812988,
      "epoch": 1.4018406549307079,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004812455780525735,
      "loss": 6.0286,
      "mean_token_accuracy": 0.14879850149154664,
      "num_tokens": 28343947.0,
      "step": 13100
    },
    {
      "entropy": 6.471450138092041,
      "epoch": 1.402375729038472,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004812302932891544,
      "loss": 6.0993,
      "mean_token_accuracy": 0.14374391064047815,
      "num_tokens": 28354635.0,
      "step": 13105
    },
    {
      "entropy": 6.443182373046875,
      "epoch": 1.4029108031462356,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004812150025708646,
      "loss": 6.0186,
      "mean_token_accuracy": 0.14714552238583564,
      "num_tokens": 28365392.0,
      "step": 13110
    },
    {
      "entropy": 6.478414154052734,
      "epoch": 1.4034458772539997,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048119970589814557,
      "loss": 6.0002,
      "mean_token_accuracy": 0.15315165370702744,
      "num_tokens": 28376527.0,
      "step": 13115
    },
    {
      "entropy": 6.4094866752624515,
      "epoch": 1.4039809513617636,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004811844032714389,
      "loss": 6.0265,
      "mean_token_accuracy": 0.15287795811891555,
      "num_tokens": 28387178.0,
      "step": 13120
    },
    {
      "entropy": 6.348618412017823,
      "epoch": 1.4045160254695275,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048116909469118663,
      "loss": 5.9122,
      "mean_token_accuracy": 0.1550012782216072,
      "num_tokens": 28397578.0,
      "step": 13125
    },
    {
      "entropy": 6.375391626358033,
      "epoch": 1.4050510995772916,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004811537801578308,
      "loss": 6.0578,
      "mean_token_accuracy": 0.14940296113491058,
      "num_tokens": 28408933.0,
      "step": 13130
    },
    {
      "entropy": 6.430589818954468,
      "epoch": 1.4055861736850555,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048113845967181346,
      "loss": 6.0384,
      "mean_token_accuracy": 0.14450321421027185,
      "num_tokens": 28421906.0,
      "step": 13135
    },
    {
      "entropy": 6.441525506973266,
      "epoch": 1.4061212477928193,
      "grad_norm": 1.046875,
      "learning_rate": 0.000481123133233577,
      "loss": 5.9741,
      "mean_token_accuracy": 0.15519341230392455,
      "num_tokens": 28432677.0,
      "step": 13140
    },
    {
      "entropy": 6.408242702484131,
      "epoch": 1.4066563219005832,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048110780084356414,
      "loss": 6.0849,
      "mean_token_accuracy": 0.14985138326883315,
      "num_tokens": 28443795.0,
      "step": 13145
    },
    {
      "entropy": 6.4452282905578615,
      "epoch": 1.407191396008347,
      "grad_norm": 0.875,
      "learning_rate": 0.0004810924625022174,
      "loss": 6.0388,
      "mean_token_accuracy": 0.15439170002937316,
      "num_tokens": 28455727.0,
      "step": 13150
    },
    {
      "entropy": 6.483570289611817,
      "epoch": 1.4077264701161112,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004810771182099799,
      "loss": 6.0495,
      "mean_token_accuracy": 0.1467655584216118,
      "num_tokens": 28467554.0,
      "step": 13155
    },
    {
      "entropy": 6.464400148391723,
      "epoch": 1.408261544223875,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004810617679672945,
      "loss": 6.0782,
      "mean_token_accuracy": 0.15515935570001602,
      "num_tokens": 28479268.0,
      "step": 13160
    },
    {
      "entropy": 6.471682214736939,
      "epoch": 1.408796618331639,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048104641177460443,
      "loss": 6.0611,
      "mean_token_accuracy": 0.14725273102521896,
      "num_tokens": 28489887.0,
      "step": 13165
    },
    {
      "entropy": 6.398373317718506,
      "epoch": 1.4093316924394028,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004810310496323533,
      "loss": 5.9508,
      "mean_token_accuracy": 0.16100962907075883,
      "num_tokens": 28500736.0,
      "step": 13170
    },
    {
      "entropy": 6.359110355377197,
      "epoch": 1.4098667665471667,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048101568154098446,
      "loss": 5.9268,
      "mean_token_accuracy": 0.16290731579065323,
      "num_tokens": 28511458.0,
      "step": 13175
    },
    {
      "entropy": 6.476224040985107,
      "epoch": 1.4104018406549308,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048100030750094185,
      "loss": 6.0526,
      "mean_token_accuracy": 0.1446356862783432,
      "num_tokens": 28522354.0,
      "step": 13180
    },
    {
      "entropy": 6.370303440093994,
      "epoch": 1.4109369147626947,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004809849275126693,
      "loss": 5.8837,
      "mean_token_accuracy": 0.16564074754714966,
      "num_tokens": 28533061.0,
      "step": 13185
    },
    {
      "entropy": 6.412190008163452,
      "epoch": 1.4114719888704585,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048096954157661085,
      "loss": 6.0377,
      "mean_token_accuracy": 0.1499859280884266,
      "num_tokens": 28544230.0,
      "step": 13190
    },
    {
      "entropy": 6.436911535263062,
      "epoch": 1.4120070629782224,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048095414969321086,
      "loss": 6.0398,
      "mean_token_accuracy": 0.1566564053297043,
      "num_tokens": 28555461.0,
      "step": 13195
    },
    {
      "entropy": 6.399296236038208,
      "epoch": 1.4125421370859863,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048093875186291376,
      "loss": 5.9515,
      "mean_token_accuracy": 0.15652852654457092,
      "num_tokens": 28565992.0,
      "step": 13200
    },
    {
      "entropy": 6.398607349395752,
      "epoch": 1.4130772111937504,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048092334808616413,
      "loss": 5.9894,
      "mean_token_accuracy": 0.1550082340836525,
      "num_tokens": 28575802.0,
      "step": 13205
    },
    {
      "entropy": 6.404100370407105,
      "epoch": 1.4136122853015143,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004809079383634067,
      "loss": 5.9595,
      "mean_token_accuracy": 0.1601613000035286,
      "num_tokens": 28586982.0,
      "step": 13210
    },
    {
      "entropy": 6.4443018436431885,
      "epoch": 1.4141473594092782,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048089252269508656,
      "loss": 6.056,
      "mean_token_accuracy": 0.152565298974514,
      "num_tokens": 28598267.0,
      "step": 13215
    },
    {
      "entropy": 6.398623895645142,
      "epoch": 1.414682433517042,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048087710108164874,
      "loss": 6.0458,
      "mean_token_accuracy": 0.15094948261976243,
      "num_tokens": 28608213.0,
      "step": 13220
    },
    {
      "entropy": 6.377017545700073,
      "epoch": 1.415217507624806,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004808616735235385,
      "loss": 6.0131,
      "mean_token_accuracy": 0.1473524495959282,
      "num_tokens": 28619674.0,
      "step": 13225
    },
    {
      "entropy": 6.418122625350952,
      "epoch": 1.41575258173257,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004808462400212014,
      "loss": 5.9204,
      "mean_token_accuracy": 0.16133133918046952,
      "num_tokens": 28630824.0,
      "step": 13230
    },
    {
      "entropy": 6.424681186676025,
      "epoch": 1.416287655840334,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048083080057508313,
      "loss": 6.0399,
      "mean_token_accuracy": 0.15107956081628798,
      "num_tokens": 28642270.0,
      "step": 13235
    },
    {
      "entropy": 6.45269889831543,
      "epoch": 1.4168227299480978,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048081535518562933,
      "loss": 6.0601,
      "mean_token_accuracy": 0.15063396841287613,
      "num_tokens": 28653649.0,
      "step": 13240
    },
    {
      "entropy": 6.389961576461792,
      "epoch": 1.4173578040558619,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000480799903853286,
      "loss": 5.9196,
      "mean_token_accuracy": 0.1543866515159607,
      "num_tokens": 28664044.0,
      "step": 13245
    },
    {
      "entropy": 6.392914533615112,
      "epoch": 1.4178928781636255,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004807844465784995,
      "loss": 5.995,
      "mean_token_accuracy": 0.15779202431440353,
      "num_tokens": 28675394.0,
      "step": 13250
    },
    {
      "entropy": 6.3641886711120605,
      "epoch": 1.4184279522713896,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048076898336171596,
      "loss": 5.9731,
      "mean_token_accuracy": 0.16063731610774995,
      "num_tokens": 28686292.0,
      "step": 13255
    },
    {
      "entropy": 6.393302249908447,
      "epoch": 1.4189630263791535,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048075351420338196,
      "loss": 6.0224,
      "mean_token_accuracy": 0.15147142112255096,
      "num_tokens": 28696928.0,
      "step": 13260
    },
    {
      "entropy": 6.3998674869537355,
      "epoch": 1.4194981004869174,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004807380391039441,
      "loss": 6.0237,
      "mean_token_accuracy": 0.1544863164424896,
      "num_tokens": 28707986.0,
      "step": 13265
    },
    {
      "entropy": 6.464109706878662,
      "epoch": 1.4200331745946815,
      "grad_norm": 1.0,
      "learning_rate": 0.00048072255806384936,
      "loss": 6.0437,
      "mean_token_accuracy": 0.15485527515411376,
      "num_tokens": 28719296.0,
      "step": 13270
    },
    {
      "entropy": 6.382425212860108,
      "epoch": 1.4205682487024454,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004807070710835447,
      "loss": 5.9076,
      "mean_token_accuracy": 0.16298083811998368,
      "num_tokens": 28729523.0,
      "step": 13275
    },
    {
      "entropy": 6.43955545425415,
      "epoch": 1.4211033228102092,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004806915781634771,
      "loss": 6.0363,
      "mean_token_accuracy": 0.14845321774482728,
      "num_tokens": 28739026.0,
      "step": 13280
    },
    {
      "entropy": 6.444441843032837,
      "epoch": 1.4216383969179731,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004806760793040942,
      "loss": 6.0971,
      "mean_token_accuracy": 0.15063622444868088,
      "num_tokens": 28749513.0,
      "step": 13285
    },
    {
      "entropy": 6.477699327468872,
      "epoch": 1.422173471025737,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048066057450584334,
      "loss": 5.9876,
      "mean_token_accuracy": 0.15694974809885026,
      "num_tokens": 28760722.0,
      "step": 13290
    },
    {
      "entropy": 6.409042406082153,
      "epoch": 1.422708545133501,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004806450637691723,
      "loss": 5.9473,
      "mean_token_accuracy": 0.16600931733846663,
      "num_tokens": 28771235.0,
      "step": 13295
    },
    {
      "entropy": 6.358709383010864,
      "epoch": 1.423243619241265,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048062954709452907,
      "loss": 5.9556,
      "mean_token_accuracy": 0.15550183802843093,
      "num_tokens": 28782642.0,
      "step": 13300
    },
    {
      "entropy": 6.4370581150054935,
      "epoch": 1.4237786933490288,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004806140244823615,
      "loss": 5.9946,
      "mean_token_accuracy": 0.15230996757745743,
      "num_tokens": 28793628.0,
      "step": 13305
    },
    {
      "entropy": 6.479377412796021,
      "epoch": 1.4243137674567927,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048059849593311776,
      "loss": 6.0402,
      "mean_token_accuracy": 0.15544017106294633,
      "num_tokens": 28804034.0,
      "step": 13310
    },
    {
      "entropy": 6.441398334503174,
      "epoch": 1.4248488415645566,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004805829614472464,
      "loss": 5.9995,
      "mean_token_accuracy": 0.15001312643289566,
      "num_tokens": 28814679.0,
      "step": 13315
    },
    {
      "entropy": 6.404942989349365,
      "epoch": 1.4253839156723207,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048056742102519593,
      "loss": 6.0123,
      "mean_token_accuracy": 0.15564872771501542,
      "num_tokens": 28825130.0,
      "step": 13320
    },
    {
      "entropy": 6.396114730834961,
      "epoch": 1.4259189897800846,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048055187466741506,
      "loss": 6.0397,
      "mean_token_accuracy": 0.14774275273084642,
      "num_tokens": 28835500.0,
      "step": 13325
    },
    {
      "entropy": 6.392610263824463,
      "epoch": 1.4264540638878485,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004805363223743527,
      "loss": 5.9845,
      "mean_token_accuracy": 0.15305309295654296,
      "num_tokens": 28845794.0,
      "step": 13330
    },
    {
      "entropy": 6.4702568531036375,
      "epoch": 1.4269891379956123,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004805207641464579,
      "loss": 6.0006,
      "mean_token_accuracy": 0.1523097947239876,
      "num_tokens": 28856326.0,
      "step": 13335
    },
    {
      "entropy": 6.342141723632812,
      "epoch": 1.4275242121033762,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004805051999841799,
      "loss": 5.9759,
      "mean_token_accuracy": 0.15488241463899613,
      "num_tokens": 28867592.0,
      "step": 13340
    },
    {
      "entropy": 6.487649059295654,
      "epoch": 1.4280592862111403,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004804896298879682,
      "loss": 6.1233,
      "mean_token_accuracy": 0.1472506955265999,
      "num_tokens": 28880352.0,
      "step": 13345
    },
    {
      "entropy": 6.485336780548096,
      "epoch": 1.4285943603189042,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004804740538582723,
      "loss": 5.9819,
      "mean_token_accuracy": 0.15547581985592843,
      "num_tokens": 28892012.0,
      "step": 13350
    },
    {
      "entropy": 6.431128787994385,
      "epoch": 1.429129434426668,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004804584718955419,
      "loss": 5.9543,
      "mean_token_accuracy": 0.1571507066488266,
      "num_tokens": 28902247.0,
      "step": 13355
    },
    {
      "entropy": 6.391356611251831,
      "epoch": 1.429664508534432,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000480442884000227,
      "loss": 6.0919,
      "mean_token_accuracy": 0.14986072108149529,
      "num_tokens": 28913222.0,
      "step": 13360
    },
    {
      "entropy": 6.461181545257569,
      "epoch": 1.4301995826421958,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004804272901727778,
      "loss": 6.0038,
      "mean_token_accuracy": 0.15332399904727936,
      "num_tokens": 28924038.0,
      "step": 13365
    },
    {
      "entropy": 6.4158540725708,
      "epoch": 1.43073465674996,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048041169041364437,
      "loss": 6.0301,
      "mean_token_accuracy": 0.1550060033798218,
      "num_tokens": 28934448.0,
      "step": 13370
    },
    {
      "entropy": 6.3995373249053955,
      "epoch": 1.4312697308577238,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048039608472327724,
      "loss": 5.9824,
      "mean_token_accuracy": 0.15337499976158142,
      "num_tokens": 28945175.0,
      "step": 13375
    },
    {
      "entropy": 6.403515386581421,
      "epoch": 1.4318048049654877,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048038047310212706,
      "loss": 5.9583,
      "mean_token_accuracy": 0.15990348756313325,
      "num_tokens": 28955116.0,
      "step": 13380
    },
    {
      "entropy": 6.459186029434204,
      "epoch": 1.4323398790732518,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048036485555064454,
      "loss": 6.0337,
      "mean_token_accuracy": 0.14969629868865014,
      "num_tokens": 28966999.0,
      "step": 13385
    },
    {
      "entropy": 6.376125478744507,
      "epoch": 1.4328749531810157,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048034923206928065,
      "loss": 6.0273,
      "mean_token_accuracy": 0.1474001705646515,
      "num_tokens": 28977388.0,
      "step": 13390
    },
    {
      "entropy": 6.3951828956604,
      "epoch": 1.4334100272887795,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048033360265848664,
      "loss": 5.9132,
      "mean_token_accuracy": 0.15966878160834314,
      "num_tokens": 28987792.0,
      "step": 13395
    },
    {
      "entropy": 6.498046636581421,
      "epoch": 1.4339451013965434,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048031796731871364,
      "loss": 6.0576,
      "mean_token_accuracy": 0.150737763941288,
      "num_tokens": 28998319.0,
      "step": 13400
    },
    {
      "entropy": 6.386184644699097,
      "epoch": 1.4344801755043073,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048030232605041323,
      "loss": 6.052,
      "mean_token_accuracy": 0.14849743992090225,
      "num_tokens": 29007942.0,
      "step": 13405
    },
    {
      "entropy": 6.432157897949219,
      "epoch": 1.4350152496120714,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004802866788540369,
      "loss": 5.9889,
      "mean_token_accuracy": 0.15755589604377745,
      "num_tokens": 29018217.0,
      "step": 13410
    },
    {
      "entropy": 6.41338415145874,
      "epoch": 1.4355503237198353,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004802710257300367,
      "loss": 6.0044,
      "mean_token_accuracy": 0.1552721731364727,
      "num_tokens": 29028472.0,
      "step": 13415
    },
    {
      "entropy": 6.420782661437988,
      "epoch": 1.4360853978275991,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004802553666788644,
      "loss": 5.9917,
      "mean_token_accuracy": 0.15654241293668747,
      "num_tokens": 29040553.0,
      "step": 13420
    },
    {
      "entropy": 6.456770181655884,
      "epoch": 1.436620471935363,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048023970170097226,
      "loss": 6.0289,
      "mean_token_accuracy": 0.1433689758181572,
      "num_tokens": 29050142.0,
      "step": 13425
    },
    {
      "entropy": 6.412277936935425,
      "epoch": 1.437155546043127,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048022403079681254,
      "loss": 5.9544,
      "mean_token_accuracy": 0.1506963700056076,
      "num_tokens": 29061762.0,
      "step": 13430
    },
    {
      "entropy": 6.450308132171631,
      "epoch": 1.437690620150891,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048020835396683775,
      "loss": 6.0261,
      "mean_token_accuracy": 0.15470581501722336,
      "num_tokens": 29072095.0,
      "step": 13435
    },
    {
      "entropy": 6.442930221557617,
      "epoch": 1.4382256942586549,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048019267121150066,
      "loss": 6.035,
      "mean_token_accuracy": 0.15525248795747756,
      "num_tokens": 29082561.0,
      "step": 13440
    },
    {
      "entropy": 6.426079511642456,
      "epoch": 1.4387607683664188,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048017698253125397,
      "loss": 5.9938,
      "mean_token_accuracy": 0.15459677577018738,
      "num_tokens": 29093813.0,
      "step": 13445
    },
    {
      "entropy": 6.418557167053223,
      "epoch": 1.4392958424741826,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004801612879265507,
      "loss": 6.0014,
      "mean_token_accuracy": 0.1529178000986576,
      "num_tokens": 29104064.0,
      "step": 13450
    },
    {
      "entropy": 6.414542102813721,
      "epoch": 1.4398309165819465,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004801455873978442,
      "loss": 6.0014,
      "mean_token_accuracy": 0.1561982348561287,
      "num_tokens": 29115086.0,
      "step": 13455
    },
    {
      "entropy": 6.394630002975464,
      "epoch": 1.4403659906897106,
      "grad_norm": 1.0,
      "learning_rate": 0.0004801298809455876,
      "loss": 5.9804,
      "mean_token_accuracy": 0.1622392788529396,
      "num_tokens": 29125536.0,
      "step": 13460
    },
    {
      "entropy": 6.466509103775024,
      "epoch": 1.4409010647974745,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004801141685702345,
      "loss": 6.0484,
      "mean_token_accuracy": 0.1571807473897934,
      "num_tokens": 29137007.0,
      "step": 13465
    },
    {
      "entropy": 6.415415287017822,
      "epoch": 1.4414361389052384,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048009845027223864,
      "loss": 6.041,
      "mean_token_accuracy": 0.14565799608826638,
      "num_tokens": 29147072.0,
      "step": 13470
    },
    {
      "entropy": 6.391107702255249,
      "epoch": 1.4419712130130022,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048008272605205377,
      "loss": 6.0098,
      "mean_token_accuracy": 0.15148337334394454,
      "num_tokens": 29157803.0,
      "step": 13475
    },
    {
      "entropy": 6.4483705997467045,
      "epoch": 1.4425062871207661,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048006699591013417,
      "loss": 5.9789,
      "mean_token_accuracy": 0.15339577943086624,
      "num_tokens": 29168173.0,
      "step": 13480
    },
    {
      "entropy": 6.494020557403564,
      "epoch": 1.4430413612285302,
      "grad_norm": 1.0,
      "learning_rate": 0.0004800512598469337,
      "loss": 6.0356,
      "mean_token_accuracy": 0.14880047142505645,
      "num_tokens": 29179377.0,
      "step": 13485
    },
    {
      "entropy": 6.422605705261231,
      "epoch": 1.443576435336294,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000480035517862907,
      "loss": 6.0587,
      "mean_token_accuracy": 0.14996186047792434,
      "num_tokens": 29190909.0,
      "step": 13490
    },
    {
      "entropy": 6.398040914535523,
      "epoch": 1.444111509444058,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048001976995850856,
      "loss": 5.967,
      "mean_token_accuracy": 0.15803209394216539,
      "num_tokens": 29200977.0,
      "step": 13495
    },
    {
      "entropy": 6.4279557228088375,
      "epoch": 1.4446465835518219,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048000401613419296,
      "loss": 5.9976,
      "mean_token_accuracy": 0.15082981884479524,
      "num_tokens": 29210702.0,
      "step": 13500
    },
    {
      "entropy": 6.463345336914062,
      "epoch": 1.4451816576595857,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00047998825639041534,
      "loss": 6.0932,
      "mean_token_accuracy": 0.14795197695493698,
      "num_tokens": 29221338.0,
      "step": 13505
    },
    {
      "entropy": 6.435945749282837,
      "epoch": 1.4457167317673498,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004799724907276306,
      "loss": 5.9391,
      "mean_token_accuracy": 0.16381806433200835,
      "num_tokens": 29232091.0,
      "step": 13510
    },
    {
      "entropy": 6.355152368545532,
      "epoch": 1.4462518058751137,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004799567191462939,
      "loss": 5.9534,
      "mean_token_accuracy": 0.1590733915567398,
      "num_tokens": 29241903.0,
      "step": 13515
    },
    {
      "entropy": 6.40584979057312,
      "epoch": 1.4467868799828776,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047994094164686074,
      "loss": 6.06,
      "mean_token_accuracy": 0.15764431059360504,
      "num_tokens": 29253414.0,
      "step": 13520
    },
    {
      "entropy": 6.441178798675537,
      "epoch": 1.4473219540906417,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004799251582297868,
      "loss": 5.9391,
      "mean_token_accuracy": 0.16360065788030625,
      "num_tokens": 29264066.0,
      "step": 13525
    },
    {
      "entropy": 6.457836580276489,
      "epoch": 1.4478570281984056,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047990936889552767,
      "loss": 6.0028,
      "mean_token_accuracy": 0.1533188506960869,
      "num_tokens": 29274915.0,
      "step": 13530
    },
    {
      "entropy": 6.417357444763184,
      "epoch": 1.4483921023061694,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004798935736445392,
      "loss": 5.9937,
      "mean_token_accuracy": 0.15109521597623826,
      "num_tokens": 29287345.0,
      "step": 13535
    },
    {
      "entropy": 6.3747169971466064,
      "epoch": 1.4489271764139333,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004798777724772777,
      "loss": 5.9715,
      "mean_token_accuracy": 0.16671939343214034,
      "num_tokens": 29297432.0,
      "step": 13540
    },
    {
      "entropy": 6.4038022518157955,
      "epoch": 1.4494622505216972,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047986196539419936,
      "loss": 5.9031,
      "mean_token_accuracy": 0.16230253726243973,
      "num_tokens": 29307534.0,
      "step": 13545
    },
    {
      "entropy": 6.389487218856812,
      "epoch": 1.4499973246294613,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004798461523957604,
      "loss": 5.9635,
      "mean_token_accuracy": 0.1593337371945381,
      "num_tokens": 29319412.0,
      "step": 13550
    },
    {
      "entropy": 6.327165603637695,
      "epoch": 1.4505323987372252,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047983033348241767,
      "loss": 5.9643,
      "mean_token_accuracy": 0.16500141471624374,
      "num_tokens": 29330477.0,
      "step": 13555
    },
    {
      "entropy": 6.362191438674927,
      "epoch": 1.451067472844989,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004798145086546278,
      "loss": 5.9893,
      "mean_token_accuracy": 0.1563024863600731,
      "num_tokens": 29341481.0,
      "step": 13560
    },
    {
      "entropy": 6.423067235946656,
      "epoch": 1.451602546952753,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004797986779128478,
      "loss": 5.9803,
      "mean_token_accuracy": 0.15672291964292526,
      "num_tokens": 29353273.0,
      "step": 13565
    },
    {
      "entropy": 6.421946859359741,
      "epoch": 1.4521376210605168,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004797828412575348,
      "loss": 5.9919,
      "mean_token_accuracy": 0.15289032757282256,
      "num_tokens": 29364108.0,
      "step": 13570
    },
    {
      "entropy": 6.3255541801452635,
      "epoch": 1.452672695168281,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047976699868914594,
      "loss": 5.9582,
      "mean_token_accuracy": 0.16502645015716552,
      "num_tokens": 29375272.0,
      "step": 13575
    },
    {
      "entropy": 6.396491813659668,
      "epoch": 1.4532077692760448,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004797511502081388,
      "loss": 5.9866,
      "mean_token_accuracy": 0.1540304221212864,
      "num_tokens": 29385749.0,
      "step": 13580
    },
    {
      "entropy": 6.342773962020874,
      "epoch": 1.4537428433838087,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000479735295814971,
      "loss": 5.9292,
      "mean_token_accuracy": 0.1604256421327591,
      "num_tokens": 29396430.0,
      "step": 13585
    },
    {
      "entropy": 6.430709600448608,
      "epoch": 1.4542779174915725,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047971943551010035,
      "loss": 5.9905,
      "mean_token_accuracy": 0.15311016589403154,
      "num_tokens": 29407020.0,
      "step": 13590
    },
    {
      "entropy": 6.4546572208404545,
      "epoch": 1.4548129915993364,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047970356929398467,
      "loss": 6.002,
      "mean_token_accuracy": 0.1531175598502159,
      "num_tokens": 29418428.0,
      "step": 13595
    },
    {
      "entropy": 6.444525671005249,
      "epoch": 1.4553480657071005,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004796876971670822,
      "loss": 6.073,
      "mean_token_accuracy": 0.15434489026665688,
      "num_tokens": 29430611.0,
      "step": 13600
    },
    {
      "entropy": 6.40570855140686,
      "epoch": 1.4558831398148644,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004796718191298512,
      "loss": 5.9848,
      "mean_token_accuracy": 0.1489190176129341,
      "num_tokens": 29441545.0,
      "step": 13605
    },
    {
      "entropy": 6.459817934036255,
      "epoch": 1.4564182139226283,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004796559351827503,
      "loss": 6.0621,
      "mean_token_accuracy": 0.15318683385849,
      "num_tokens": 29452529.0,
      "step": 13610
    },
    {
      "entropy": 6.3890868663787845,
      "epoch": 1.4569532880303921,
      "grad_norm": 0.890625,
      "learning_rate": 0.000479640045326238,
      "loss": 5.9781,
      "mean_token_accuracy": 0.16061681210994722,
      "num_tokens": 29463860.0,
      "step": 13615
    },
    {
      "entropy": 6.484077978134155,
      "epoch": 1.457488362138156,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004796241495607731,
      "loss": 6.0446,
      "mean_token_accuracy": 0.1536493867635727,
      "num_tokens": 29475004.0,
      "step": 13620
    },
    {
      "entropy": 6.446160078048706,
      "epoch": 1.4580234362459201,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004796082478868146,
      "loss": 6.1092,
      "mean_token_accuracy": 0.14367302432656287,
      "num_tokens": 29485921.0,
      "step": 13625
    },
    {
      "entropy": 6.452400207519531,
      "epoch": 1.458558510353684,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047959234030482185,
      "loss": 5.9688,
      "mean_token_accuracy": 0.1667572572827339,
      "num_tokens": 29495729.0,
      "step": 13630
    },
    {
      "entropy": 6.481155586242676,
      "epoch": 1.4590935844614479,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004795764268152538,
      "loss": 6.0406,
      "mean_token_accuracy": 0.149743016064167,
      "num_tokens": 29506051.0,
      "step": 13635
    },
    {
      "entropy": 6.389807653427124,
      "epoch": 1.4596286585692118,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004795605074185704,
      "loss": 5.9998,
      "mean_token_accuracy": 0.15389131456613542,
      "num_tokens": 29516234.0,
      "step": 13640
    },
    {
      "entropy": 6.449402809143066,
      "epoch": 1.4601637326769756,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047954458211523095,
      "loss": 5.9792,
      "mean_token_accuracy": 0.15142796486616134,
      "num_tokens": 29527649.0,
      "step": 13645
    },
    {
      "entropy": 6.465574789047241,
      "epoch": 1.4606988067847397,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047952865090569545,
      "loss": 6.0487,
      "mean_token_accuracy": 0.14836390540003777,
      "num_tokens": 29539150.0,
      "step": 13650
    },
    {
      "entropy": 6.3607728481292725,
      "epoch": 1.4612338808925036,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047951271379042393,
      "loss": 5.9349,
      "mean_token_accuracy": 0.16002043038606645,
      "num_tokens": 29550058.0,
      "step": 13655
    },
    {
      "entropy": 6.338524675369262,
      "epoch": 1.4617689550002675,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004794967707698765,
      "loss": 5.9262,
      "mean_token_accuracy": 0.1668383464217186,
      "num_tokens": 29560377.0,
      "step": 13660
    },
    {
      "entropy": 6.359252977371216,
      "epoch": 1.4623040291080316,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004794808218445136,
      "loss": 5.8692,
      "mean_token_accuracy": 0.1667836606502533,
      "num_tokens": 29571378.0,
      "step": 13665
    },
    {
      "entropy": 6.458996915817261,
      "epoch": 1.4628391032157955,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047946486701479576,
      "loss": 6.0339,
      "mean_token_accuracy": 0.15407999008893966,
      "num_tokens": 29582187.0,
      "step": 13670
    },
    {
      "entropy": 6.394173097610474,
      "epoch": 1.4633741773235593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004794489062811835,
      "loss": 6.0111,
      "mean_token_accuracy": 0.15108111426234244,
      "num_tokens": 29592982.0,
      "step": 13675
    },
    {
      "entropy": 6.439515876770019,
      "epoch": 1.4639092514313232,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004794329396441378,
      "loss": 5.9866,
      "mean_token_accuracy": 0.15276691168546677,
      "num_tokens": 29603639.0,
      "step": 13680
    },
    {
      "entropy": 6.3766645908355715,
      "epoch": 1.464444325539087,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047941696710411975,
      "loss": 5.9279,
      "mean_token_accuracy": 0.16186307817697526,
      "num_tokens": 29615103.0,
      "step": 13685
    },
    {
      "entropy": 6.455867195129395,
      "epoch": 1.4649793996468512,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004794009886615904,
      "loss": 5.999,
      "mean_token_accuracy": 0.15398305505514145,
      "num_tokens": 29626184.0,
      "step": 13690
    },
    {
      "entropy": 6.438571166992188,
      "epoch": 1.465514473754615,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047938500431701135,
      "loss": 5.9664,
      "mean_token_accuracy": 0.1660277619957924,
      "num_tokens": 29637615.0,
      "step": 13695
    },
    {
      "entropy": 6.45487871170044,
      "epoch": 1.466049547862379,
      "grad_norm": 0.953125,
      "learning_rate": 0.000479369014070844,
      "loss": 6.0356,
      "mean_token_accuracy": 0.15271387696266175,
      "num_tokens": 29648648.0,
      "step": 13700
    },
    {
      "entropy": 6.433786821365357,
      "epoch": 1.4665846219701428,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047935301792355003,
      "loss": 5.9765,
      "mean_token_accuracy": 0.16103079915046692,
      "num_tokens": 29660542.0,
      "step": 13705
    },
    {
      "entropy": 6.449999094009399,
      "epoch": 1.4671196960779067,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004793370158755914,
      "loss": 6.0169,
      "mean_token_accuracy": 0.1522742435336113,
      "num_tokens": 29670342.0,
      "step": 13710
    },
    {
      "entropy": 6.410958194732666,
      "epoch": 1.4676547701856708,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047932100792743014,
      "loss": 5.9871,
      "mean_token_accuracy": 0.16180376559495926,
      "num_tokens": 29681191.0,
      "step": 13715
    },
    {
      "entropy": 6.4375732421875,
      "epoch": 1.4681898442934347,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047930499407952855,
      "loss": 5.9252,
      "mean_token_accuracy": 0.1655012309551239,
      "num_tokens": 29691299.0,
      "step": 13720
    },
    {
      "entropy": 6.404075574874878,
      "epoch": 1.4687249184011986,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047928897433234893,
      "loss": 5.9871,
      "mean_token_accuracy": 0.15798449963331224,
      "num_tokens": 29702125.0,
      "step": 13725
    },
    {
      "entropy": 6.421619510650634,
      "epoch": 1.4692599925089624,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004792729486863539,
      "loss": 5.9993,
      "mean_token_accuracy": 0.15197849720716478,
      "num_tokens": 29712679.0,
      "step": 13730
    },
    {
      "entropy": 6.410307121276856,
      "epoch": 1.4697950666167263,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004792569171420061,
      "loss": 5.9458,
      "mean_token_accuracy": 0.16102083921432495,
      "num_tokens": 29723833.0,
      "step": 13735
    },
    {
      "entropy": 6.479249477386475,
      "epoch": 1.4703301407244904,
      "grad_norm": 1.0,
      "learning_rate": 0.0004792408796997687,
      "loss": 6.0474,
      "mean_token_accuracy": 0.15478676408529282,
      "num_tokens": 29734646.0,
      "step": 13740
    },
    {
      "entropy": 6.433805131912232,
      "epoch": 1.4708652148322543,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004792248363601044,
      "loss": 5.9915,
      "mean_token_accuracy": 0.1579342007637024,
      "num_tokens": 29745129.0,
      "step": 13745
    },
    {
      "entropy": 6.442555904388428,
      "epoch": 1.4714002889400182,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004792087871234767,
      "loss": 6.0151,
      "mean_token_accuracy": 0.15217642039060592,
      "num_tokens": 29756308.0,
      "step": 13750
    },
    {
      "entropy": 6.405927991867065,
      "epoch": 1.471935363047782,
      "grad_norm": 1.0,
      "learning_rate": 0.000479192731990349,
      "loss": 6.0071,
      "mean_token_accuracy": 0.15515146553516387,
      "num_tokens": 29767835.0,
      "step": 13755
    },
    {
      "entropy": 6.431542634963989,
      "epoch": 1.472470437155546,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004791766709611849,
      "loss": 6.0456,
      "mean_token_accuracy": 0.15362401604652404,
      "num_tokens": 29778776.0,
      "step": 13760
    },
    {
      "entropy": 6.498543167114258,
      "epoch": 1.47300551126331,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000479160604036448,
      "loss": 6.0428,
      "mean_token_accuracy": 0.14693716466426848,
      "num_tokens": 29788800.0,
      "step": 13765
    },
    {
      "entropy": 6.415443181991577,
      "epoch": 1.473540585371074,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004791445312166025,
      "loss": 5.9529,
      "mean_token_accuracy": 0.16358682960271836,
      "num_tokens": 29799375.0,
      "step": 13770
    },
    {
      "entropy": 6.479481744766235,
      "epoch": 1.4740756594788378,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004791284525021122,
      "loss": 6.0833,
      "mean_token_accuracy": 0.14749659076333047,
      "num_tokens": 29811267.0,
      "step": 13775
    },
    {
      "entropy": 6.425268697738647,
      "epoch": 1.4746107335866017,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004791123678934416,
      "loss": 6.0148,
      "mean_token_accuracy": 0.15197368189692498,
      "num_tokens": 29821616.0,
      "step": 13780
    },
    {
      "entropy": 6.40714201927185,
      "epoch": 1.4751458076943655,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004790962773910551,
      "loss": 5.9941,
      "mean_token_accuracy": 0.1568580076098442,
      "num_tokens": 29832233.0,
      "step": 13785
    },
    {
      "entropy": 6.4556028842926025,
      "epoch": 1.4756808818021296,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047908018099541717,
      "loss": 5.9637,
      "mean_token_accuracy": 0.15423155725002288,
      "num_tokens": 29843465.0,
      "step": 13790
    },
    {
      "entropy": 6.390689182281494,
      "epoch": 1.4762159559098935,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004790640787069927,
      "loss": 5.8413,
      "mean_token_accuracy": 0.1641766294836998,
      "num_tokens": 29854123.0,
      "step": 13795
    },
    {
      "entropy": 6.460968351364135,
      "epoch": 1.4767510300176574,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004790479705262467,
      "loss": 6.0746,
      "mean_token_accuracy": 0.1459658406674862,
      "num_tokens": 29864831.0,
      "step": 13800
    },
    {
      "entropy": 6.390366697311402,
      "epoch": 1.4772861041254215,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004790318564536441,
      "loss": 5.9562,
      "mean_token_accuracy": 0.15159278064966203,
      "num_tokens": 29876273.0,
      "step": 13805
    },
    {
      "entropy": 6.42622594833374,
      "epoch": 1.4778211782331854,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047901573648965047,
      "loss": 5.9604,
      "mean_token_accuracy": 0.15375650823116302,
      "num_tokens": 29886560.0,
      "step": 13810
    },
    {
      "entropy": 6.396084356307983,
      "epoch": 1.4783562523409493,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000478999610634731,
      "loss": 5.9582,
      "mean_token_accuracy": 0.15774335116147994,
      "num_tokens": 29896916.0,
      "step": 13815
    },
    {
      "entropy": 6.396129846572876,
      "epoch": 1.4788913264487131,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004789834788893515,
      "loss": 6.0286,
      "mean_token_accuracy": 0.14876941069960595,
      "num_tokens": 29907288.0,
      "step": 13820
    },
    {
      "entropy": 6.374079895019531,
      "epoch": 1.479426400556477,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004789673412539777,
      "loss": 5.9399,
      "mean_token_accuracy": 0.15566962957382202,
      "num_tokens": 29918194.0,
      "step": 13825
    },
    {
      "entropy": 6.428502607345581,
      "epoch": 1.479961474664241,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047895119772907566,
      "loss": 5.9873,
      "mean_token_accuracy": 0.155208583176136,
      "num_tokens": 29928521.0,
      "step": 13830
    },
    {
      "entropy": 6.386731338500977,
      "epoch": 1.480496548772005,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004789350483151113,
      "loss": 5.9516,
      "mean_token_accuracy": 0.15276003628969193,
      "num_tokens": 29938885.0,
      "step": 13835
    },
    {
      "entropy": 6.425902318954468,
      "epoch": 1.4810316228797689,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004789188930125512,
      "loss": 6.0059,
      "mean_token_accuracy": 0.16351278722286225,
      "num_tokens": 29948921.0,
      "step": 13840
    },
    {
      "entropy": 6.390540742874146,
      "epoch": 1.4815666969875327,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004789027318218616,
      "loss": 5.9269,
      "mean_token_accuracy": 0.16290507912635804,
      "num_tokens": 29959633.0,
      "step": 13845
    },
    {
      "entropy": 6.376818704605102,
      "epoch": 1.4821017710952966,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004788865647435093,
      "loss": 5.966,
      "mean_token_accuracy": 0.15059732943773269,
      "num_tokens": 29970639.0,
      "step": 13850
    },
    {
      "entropy": 6.4237480640411375,
      "epoch": 1.4826368452030607,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004788703917779612,
      "loss": 6.0587,
      "mean_token_accuracy": 0.153861166536808,
      "num_tokens": 29981752.0,
      "step": 13855
    },
    {
      "entropy": 6.453693389892578,
      "epoch": 1.4831719193108246,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004788542129256841,
      "loss": 5.9154,
      "mean_token_accuracy": 0.16017114669084548,
      "num_tokens": 29991037.0,
      "step": 13860
    },
    {
      "entropy": 6.448595428466797,
      "epoch": 1.4837069934185885,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047883802818714526,
      "loss": 6.0188,
      "mean_token_accuracy": 0.14515757337212562,
      "num_tokens": 30001980.0,
      "step": 13865
    },
    {
      "entropy": 6.420425510406494,
      "epoch": 1.4842420675263523,
      "grad_norm": 1.0,
      "learning_rate": 0.000478821837562812,
      "loss": 6.013,
      "mean_token_accuracy": 0.15398439168930053,
      "num_tokens": 30013377.0,
      "step": 13870
    },
    {
      "entropy": 6.423045015335083,
      "epoch": 1.4847771416341162,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004788056410531518,
      "loss": 6.0673,
      "mean_token_accuracy": 0.15447486639022828,
      "num_tokens": 30023909.0,
      "step": 13875
    },
    {
      "entropy": 6.366085624694824,
      "epoch": 1.4853122157418803,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004787894386586324,
      "loss": 5.9285,
      "mean_token_accuracy": 0.1586606577038765,
      "num_tokens": 30034069.0,
      "step": 13880
    },
    {
      "entropy": 6.363991165161133,
      "epoch": 1.4858472898496442,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047877323037972153,
      "loss": 5.934,
      "mean_token_accuracy": 0.15956230461597443,
      "num_tokens": 30043940.0,
      "step": 13885
    },
    {
      "entropy": 6.350185298919678,
      "epoch": 1.486382363957408,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004787570162168874,
      "loss": 5.9157,
      "mean_token_accuracy": 0.16537891179323197,
      "num_tokens": 30054985.0,
      "step": 13890
    },
    {
      "entropy": 6.4401092529296875,
      "epoch": 1.486917438065172,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047874079617059783,
      "loss": 5.9613,
      "mean_token_accuracy": 0.15726698338985443,
      "num_tokens": 30065675.0,
      "step": 13895
    },
    {
      "entropy": 6.395420503616333,
      "epoch": 1.4874525121729358,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004787245702413216,
      "loss": 5.9187,
      "mean_token_accuracy": 0.16010231226682664,
      "num_tokens": 30075305.0,
      "step": 13900
    },
    {
      "entropy": 6.429739093780517,
      "epoch": 1.4879875862807,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047870833842952697,
      "loss": 6.0117,
      "mean_token_accuracy": 0.1562752529978752,
      "num_tokens": 30087309.0,
      "step": 13905
    },
    {
      "entropy": 6.414200687408448,
      "epoch": 1.4885226603884638,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004786921007356827,
      "loss": 5.9296,
      "mean_token_accuracy": 0.15422847419977187,
      "num_tokens": 30097262.0,
      "step": 13910
    },
    {
      "entropy": 6.4299500465393065,
      "epoch": 1.4890577344962277,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047867585716025767,
      "loss": 6.0302,
      "mean_token_accuracy": 0.15622605681419371,
      "num_tokens": 30107067.0,
      "step": 13915
    },
    {
      "entropy": 6.443173551559449,
      "epoch": 1.4895928086039918,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004786596077037209,
      "loss": 6.037,
      "mean_token_accuracy": 0.15037016719579696,
      "num_tokens": 30117115.0,
      "step": 13920
    },
    {
      "entropy": 6.421395683288575,
      "epoch": 1.4901278827117554,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047864335236654144,
      "loss": 5.9514,
      "mean_token_accuracy": 0.15858543366193772,
      "num_tokens": 30127202.0,
      "step": 13925
    },
    {
      "entropy": 6.343588352203369,
      "epoch": 1.4906629568195195,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004786270911491889,
      "loss": 5.9646,
      "mean_token_accuracy": 0.15753924548625947,
      "num_tokens": 30138608.0,
      "step": 13930
    },
    {
      "entropy": 6.465522432327271,
      "epoch": 1.4911980309272834,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004786108240521327,
      "loss": 6.0031,
      "mean_token_accuracy": 0.15762083530426024,
      "num_tokens": 30148500.0,
      "step": 13935
    },
    {
      "entropy": 6.344076108932495,
      "epoch": 1.4917331050350473,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047859455107584256,
      "loss": 5.9456,
      "mean_token_accuracy": 0.1641395717859268,
      "num_tokens": 30159062.0,
      "step": 13940
    },
    {
      "entropy": 6.372643089294433,
      "epoch": 1.4922681791428114,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047857827222078835,
      "loss": 5.9841,
      "mean_token_accuracy": 0.15420911312103272,
      "num_tokens": 30170468.0,
      "step": 13945
    },
    {
      "entropy": 6.360788822174072,
      "epoch": 1.4928032532505753,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004785619874874402,
      "loss": 5.8835,
      "mean_token_accuracy": 0.1622602790594101,
      "num_tokens": 30180944.0,
      "step": 13950
    },
    {
      "entropy": 6.341245937347412,
      "epoch": 1.4933383273583392,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004785456968762682,
      "loss": 5.8921,
      "mean_token_accuracy": 0.16010242700576782,
      "num_tokens": 30191614.0,
      "step": 13955
    },
    {
      "entropy": 6.349891424179077,
      "epoch": 1.493873401466103,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047852940038774276,
      "loss": 5.9128,
      "mean_token_accuracy": 0.1636913612484932,
      "num_tokens": 30202035.0,
      "step": 13960
    },
    {
      "entropy": 6.3552042007446286,
      "epoch": 1.494408475573867,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047851309802233457,
      "loss": 5.9985,
      "mean_token_accuracy": 0.15794553905725478,
      "num_tokens": 30213205.0,
      "step": 13965
    },
    {
      "entropy": 6.467336177825928,
      "epoch": 1.494943549681631,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004784967897805143,
      "loss": 6.0057,
      "mean_token_accuracy": 0.15847853794693947,
      "num_tokens": 30223971.0,
      "step": 13970
    },
    {
      "entropy": 6.426643323898316,
      "epoch": 1.495478623789395,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004784804756627528,
      "loss": 6.0098,
      "mean_token_accuracy": 0.151532331854105,
      "num_tokens": 30235346.0,
      "step": 13975
    },
    {
      "entropy": 6.3789012908935545,
      "epoch": 1.4960136978971588,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047846415566952106,
      "loss": 5.9568,
      "mean_token_accuracy": 0.15841448605060576,
      "num_tokens": 30246033.0,
      "step": 13980
    },
    {
      "entropy": 6.379653358459473,
      "epoch": 1.4965487720049226,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047844782980129057,
      "loss": 6.0085,
      "mean_token_accuracy": 0.15052564144134523,
      "num_tokens": 30258208.0,
      "step": 13985
    },
    {
      "entropy": 6.400771284103394,
      "epoch": 1.4970838461126865,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047843149805853254,
      "loss": 6.0231,
      "mean_token_accuracy": 0.15609513819217682,
      "num_tokens": 30268824.0,
      "step": 13990
    },
    {
      "entropy": 6.454229354858398,
      "epoch": 1.4976189202204506,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004784151604417185,
      "loss": 6.0406,
      "mean_token_accuracy": 0.14706123173236846,
      "num_tokens": 30280078.0,
      "step": 13995
    },
    {
      "entropy": 6.394439029693603,
      "epoch": 1.4981539943282145,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047839881695132034,
      "loss": 5.9483,
      "mean_token_accuracy": 0.16360280960798262,
      "num_tokens": 30290939.0,
      "step": 14000
    },
    {
      "entropy": 6.402804327011109,
      "epoch": 1.4986890684359784,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047838246758780996,
      "loss": 5.9766,
      "mean_token_accuracy": 0.1586063176393509,
      "num_tokens": 30301198.0,
      "step": 14005
    },
    {
      "entropy": 6.385310363769531,
      "epoch": 1.4992241425437423,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004783661123516593,
      "loss": 5.9415,
      "mean_token_accuracy": 0.1622556135058403,
      "num_tokens": 30312021.0,
      "step": 14010
    },
    {
      "entropy": 6.431384801864624,
      "epoch": 1.4997592166515061,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004783497512433408,
      "loss": 5.9565,
      "mean_token_accuracy": 0.16146570444107056,
      "num_tokens": 30323335.0,
      "step": 14015
    },
    {
      "entropy": 6.40609917640686,
      "epoch": 1.5002942907592702,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004783333842633268,
      "loss": 5.989,
      "mean_token_accuracy": 0.15663225650787355,
      "num_tokens": 30334389.0,
      "step": 14020
    },
    {
      "entropy": 6.357702922821045,
      "epoch": 1.500829364867034,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047831701141208994,
      "loss": 5.9555,
      "mean_token_accuracy": 0.1489044263958931,
      "num_tokens": 30345407.0,
      "step": 14025
    },
    {
      "entropy": 6.4298515796661375,
      "epoch": 1.501364438974798,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047830063269010286,
      "loss": 5.9692,
      "mean_token_accuracy": 0.15171137154102327,
      "num_tokens": 30356718.0,
      "step": 14030
    },
    {
      "entropy": 6.381672048568726,
      "epoch": 1.501899513082562,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047828424809783865,
      "loss": 5.9211,
      "mean_token_accuracy": 0.1572643518447876,
      "num_tokens": 30367617.0,
      "step": 14035
    },
    {
      "entropy": 6.393596839904785,
      "epoch": 1.5024345871903257,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004782678576357703,
      "loss": 5.9891,
      "mean_token_accuracy": 0.15565940290689467,
      "num_tokens": 30377923.0,
      "step": 14040
    },
    {
      "entropy": 6.400784730911255,
      "epoch": 1.5029696612980898,
      "grad_norm": 1.0,
      "learning_rate": 0.0004782514613043711,
      "loss": 5.9639,
      "mean_token_accuracy": 0.16317369490861894,
      "num_tokens": 30388719.0,
      "step": 14045
    },
    {
      "entropy": 6.391443109512329,
      "epoch": 1.5035047354058537,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047823505910411465,
      "loss": 5.911,
      "mean_token_accuracy": 0.15768683850765228,
      "num_tokens": 30400191.0,
      "step": 14050
    },
    {
      "entropy": 6.403419923782349,
      "epoch": 1.5040398095136176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047821865103547437,
      "loss": 5.993,
      "mean_token_accuracy": 0.15274875462055207,
      "num_tokens": 30410267.0,
      "step": 14055
    },
    {
      "entropy": 6.4074605941772464,
      "epoch": 1.5045748836213817,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000478202237098924,
      "loss": 6.0015,
      "mean_token_accuracy": 0.15294764637947084,
      "num_tokens": 30420852.0,
      "step": 14060
    },
    {
      "entropy": 6.439791631698609,
      "epoch": 1.5051099577291454,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004781858172949376,
      "loss": 6.0356,
      "mean_token_accuracy": 0.15536686331033706,
      "num_tokens": 30432214.0,
      "step": 14065
    },
    {
      "entropy": 6.4741294384002686,
      "epoch": 1.5056450318369095,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004781693916239894,
      "loss": 6.057,
      "mean_token_accuracy": 0.14787335991859435,
      "num_tokens": 30443919.0,
      "step": 14070
    },
    {
      "entropy": 6.449910306930542,
      "epoch": 1.5061801059446733,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047815296008655347,
      "loss": 6.1114,
      "mean_token_accuracy": 0.1467147499322891,
      "num_tokens": 30453930.0,
      "step": 14075
    },
    {
      "entropy": 6.432980585098266,
      "epoch": 1.5067151800524372,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00047813652268310436,
      "loss": 6.0129,
      "mean_token_accuracy": 0.14897721707820893,
      "num_tokens": 30464612.0,
      "step": 14080
    },
    {
      "entropy": 6.409858369827271,
      "epoch": 1.5072502541602013,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047812007941411673,
      "loss": 5.94,
      "mean_token_accuracy": 0.15463443398475646,
      "num_tokens": 30474819.0,
      "step": 14085
    },
    {
      "entropy": 6.3661237239837645,
      "epoch": 1.507785328267965,
      "grad_norm": 0.859375,
      "learning_rate": 0.00047810363028006535,
      "loss": 5.9384,
      "mean_token_accuracy": 0.16861377209424971,
      "num_tokens": 30485481.0,
      "step": 14090
    },
    {
      "entropy": 6.355788612365723,
      "epoch": 1.508320402375729,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004780871752814252,
      "loss": 5.9726,
      "mean_token_accuracy": 0.16148585826158524,
      "num_tokens": 30496321.0,
      "step": 14095
    },
    {
      "entropy": 6.399319171905518,
      "epoch": 1.508855476483493,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004780707144186714,
      "loss": 6.0701,
      "mean_token_accuracy": 0.14837516248226165,
      "num_tokens": 30507578.0,
      "step": 14100
    },
    {
      "entropy": 6.459050798416138,
      "epoch": 1.5093905505912568,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047805424769227923,
      "loss": 5.9978,
      "mean_token_accuracy": 0.15501926392316817,
      "num_tokens": 30518328.0,
      "step": 14105
    },
    {
      "entropy": 6.3831109523773195,
      "epoch": 1.509925624699021,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047803777510272425,
      "loss": 5.9547,
      "mean_token_accuracy": 0.15705757439136506,
      "num_tokens": 30528555.0,
      "step": 14110
    },
    {
      "entropy": 6.353852224349976,
      "epoch": 1.5104606988067846,
      "grad_norm": 1.03125,
      "learning_rate": 0.000478021296650482,
      "loss": 5.8829,
      "mean_token_accuracy": 0.16317364424467087,
      "num_tokens": 30540217.0,
      "step": 14115
    },
    {
      "entropy": 6.458622407913208,
      "epoch": 1.5109957729145487,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004780048123360283,
      "loss": 6.0221,
      "mean_token_accuracy": 0.1585195854306221,
      "num_tokens": 30550937.0,
      "step": 14120
    },
    {
      "entropy": 6.468946552276611,
      "epoch": 1.5115308470223126,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047798832215983916,
      "loss": 6.0118,
      "mean_token_accuracy": 0.15595668628811837,
      "num_tokens": 30562248.0,
      "step": 14125
    },
    {
      "entropy": 6.453252935409546,
      "epoch": 1.5120659211300764,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004779718261223908,
      "loss": 6.0016,
      "mean_token_accuracy": 0.15973061174154282,
      "num_tokens": 30573362.0,
      "step": 14130
    },
    {
      "entropy": 6.4029356956481935,
      "epoch": 1.5126009952378405,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047795532422415947,
      "loss": 5.9952,
      "mean_token_accuracy": 0.15406877398490906,
      "num_tokens": 30584052.0,
      "step": 14135
    },
    {
      "entropy": 6.429622650146484,
      "epoch": 1.5131360693456044,
      "grad_norm": 1.0,
      "learning_rate": 0.0004779388164656217,
      "loss": 6.0235,
      "mean_token_accuracy": 0.1528918921947479,
      "num_tokens": 30594630.0,
      "step": 14140
    },
    {
      "entropy": 6.401149654388428,
      "epoch": 1.5136711434533683,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047792230284725415,
      "loss": 5.9811,
      "mean_token_accuracy": 0.15375052392482758,
      "num_tokens": 30605625.0,
      "step": 14145
    },
    {
      "entropy": 6.409199523925781,
      "epoch": 1.5142062175611322,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047790578336953356,
      "loss": 5.9643,
      "mean_token_accuracy": 0.16028316020965577,
      "num_tokens": 30616520.0,
      "step": 14150
    },
    {
      "entropy": 6.40304274559021,
      "epoch": 1.514741291668896,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000477889258032937,
      "loss": 5.9366,
      "mean_token_accuracy": 0.16202332377433776,
      "num_tokens": 30627372.0,
      "step": 14155
    },
    {
      "entropy": 6.421124601364136,
      "epoch": 1.5152763657766601,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047787272683794157,
      "loss": 5.9345,
      "mean_token_accuracy": 0.16611140221357346,
      "num_tokens": 30636974.0,
      "step": 14160
    },
    {
      "entropy": 6.421116828918457,
      "epoch": 1.515811439884424,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004778561897850247,
      "loss": 5.9881,
      "mean_token_accuracy": 0.16020333766937256,
      "num_tokens": 30648202.0,
      "step": 14165
    },
    {
      "entropy": 6.411683416366577,
      "epoch": 1.516346513992188,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047783964687466384,
      "loss": 6.0315,
      "mean_token_accuracy": 0.1523759439587593,
      "num_tokens": 30657981.0,
      "step": 14170
    },
    {
      "entropy": 6.428750658035279,
      "epoch": 1.516881588099952,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047782309810733676,
      "loss": 5.9983,
      "mean_token_accuracy": 0.1508225053548813,
      "num_tokens": 30668496.0,
      "step": 14175
    },
    {
      "entropy": 6.374180030822754,
      "epoch": 1.5174166622077156,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047780654348352115,
      "loss": 5.8976,
      "mean_token_accuracy": 0.1611080601811409,
      "num_tokens": 30678868.0,
      "step": 14180
    },
    {
      "entropy": 6.416233348846435,
      "epoch": 1.5179517363154797,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047778998300369516,
      "loss": 6.0394,
      "mean_token_accuracy": 0.15413734912872315,
      "num_tokens": 30690066.0,
      "step": 14185
    },
    {
      "entropy": 6.397771310806275,
      "epoch": 1.5184868104232436,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004777734166683368,
      "loss": 5.9262,
      "mean_token_accuracy": 0.15959364473819732,
      "num_tokens": 30700624.0,
      "step": 14190
    },
    {
      "entropy": 6.371306371688843,
      "epoch": 1.5190218845310075,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047775684447792457,
      "loss": 5.9359,
      "mean_token_accuracy": 0.16291745007038116,
      "num_tokens": 30711183.0,
      "step": 14195
    },
    {
      "entropy": 6.3782402038574215,
      "epoch": 1.5195569586387716,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000477740266432937,
      "loss": 5.962,
      "mean_token_accuracy": 0.16599517166614533,
      "num_tokens": 30721446.0,
      "step": 14200
    },
    {
      "entropy": 6.453416252136231,
      "epoch": 1.5200920327465353,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047772368253385267,
      "loss": 6.0122,
      "mean_token_accuracy": 0.14951234310865402,
      "num_tokens": 30731901.0,
      "step": 14205
    },
    {
      "entropy": 6.434371662139893,
      "epoch": 1.5206271068542994,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004777070927811505,
      "loss": 5.9936,
      "mean_token_accuracy": 0.15339210480451584,
      "num_tokens": 30741701.0,
      "step": 14210
    },
    {
      "entropy": 6.405278158187866,
      "epoch": 1.5211621809620632,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047769049717530956,
      "loss": 5.9804,
      "mean_token_accuracy": 0.15703128576278685,
      "num_tokens": 30751594.0,
      "step": 14215
    },
    {
      "entropy": 6.384680318832397,
      "epoch": 1.5216972550698271,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00047767389571680896,
      "loss": 6.0502,
      "mean_token_accuracy": 0.14325668290257454,
      "num_tokens": 30764418.0,
      "step": 14220
    },
    {
      "entropy": 6.460424566268921,
      "epoch": 1.5222323291775912,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004776572884061281,
      "loss": 5.9725,
      "mean_token_accuracy": 0.16051809638738632,
      "num_tokens": 30773976.0,
      "step": 14225
    },
    {
      "entropy": 6.438876342773438,
      "epoch": 1.5227674032853549,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004776406752437465,
      "loss": 6.0061,
      "mean_token_accuracy": 0.15622590631246566,
      "num_tokens": 30784210.0,
      "step": 14230
    },
    {
      "entropy": 6.41880054473877,
      "epoch": 1.523302477393119,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004776240562301438,
      "loss": 6.064,
      "mean_token_accuracy": 0.1507749505341053,
      "num_tokens": 30794322.0,
      "step": 14235
    },
    {
      "entropy": 6.450243330001831,
      "epoch": 1.5238375515008828,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004776074313658001,
      "loss": 5.9878,
      "mean_token_accuracy": 0.16299980729818345,
      "num_tokens": 30805264.0,
      "step": 14240
    },
    {
      "entropy": 6.344057321548462,
      "epoch": 1.5243726256086467,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004775908006511952,
      "loss": 5.9755,
      "mean_token_accuracy": 0.15285835266113282,
      "num_tokens": 30815909.0,
      "step": 14245
    },
    {
      "entropy": 6.449803972244263,
      "epoch": 1.5249076997164108,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004775741640868095,
      "loss": 5.983,
      "mean_token_accuracy": 0.15893309339880943,
      "num_tokens": 30826867.0,
      "step": 14250
    },
    {
      "entropy": 6.45894193649292,
      "epoch": 1.5254427738241747,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047755752167312314,
      "loss": 5.9138,
      "mean_token_accuracy": 0.15791400521993637,
      "num_tokens": 30837696.0,
      "step": 14255
    },
    {
      "entropy": 6.364669561386108,
      "epoch": 1.5259778479319386,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004775408734106169,
      "loss": 5.9904,
      "mean_token_accuracy": 0.15583638995885848,
      "num_tokens": 30848155.0,
      "step": 14260
    },
    {
      "entropy": 6.351717138290406,
      "epoch": 1.5265129220397025,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047752421929977143,
      "loss": 5.8747,
      "mean_token_accuracy": 0.1629194051027298,
      "num_tokens": 30857804.0,
      "step": 14265
    },
    {
      "entropy": 6.440001201629639,
      "epoch": 1.5270479961474663,
      "grad_norm": 1.0,
      "learning_rate": 0.0004775075593410675,
      "loss": 6.0109,
      "mean_token_accuracy": 0.1550878942012787,
      "num_tokens": 30867290.0,
      "step": 14270
    },
    {
      "entropy": 6.426958179473877,
      "epoch": 1.5275830702552304,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047749089353498625,
      "loss": 6.0748,
      "mean_token_accuracy": 0.15661276280879974,
      "num_tokens": 30877634.0,
      "step": 14275
    },
    {
      "entropy": 6.429671001434326,
      "epoch": 1.5281181443629943,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000477474221882009,
      "loss": 6.0396,
      "mean_token_accuracy": 0.1554855450987816,
      "num_tokens": 30889176.0,
      "step": 14280
    },
    {
      "entropy": 6.402042293548584,
      "epoch": 1.5286532184707582,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004774575443826169,
      "loss": 5.9371,
      "mean_token_accuracy": 0.16026026457548143,
      "num_tokens": 30899516.0,
      "step": 14285
    },
    {
      "entropy": 6.416169786453247,
      "epoch": 1.5291882925785223,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004774408610372918,
      "loss": 6.0196,
      "mean_token_accuracy": 0.15382248908281326,
      "num_tokens": 30910155.0,
      "step": 14290
    },
    {
      "entropy": 6.448124790191651,
      "epoch": 1.529723366686286,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004774241718465152,
      "loss": 6.0472,
      "mean_token_accuracy": 0.15000636577606202,
      "num_tokens": 30921486.0,
      "step": 14295
    },
    {
      "entropy": 6.440995168685913,
      "epoch": 1.53025844079405,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004774074768107691,
      "loss": 5.9746,
      "mean_token_accuracy": 0.15477679818868637,
      "num_tokens": 30932361.0,
      "step": 14300
    },
    {
      "entropy": 6.381921815872192,
      "epoch": 1.530793514901814,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047739077593053554,
      "loss": 5.9645,
      "mean_token_accuracy": 0.15766846388578415,
      "num_tokens": 30943033.0,
      "step": 14305
    },
    {
      "entropy": 6.399259948730469,
      "epoch": 1.5313285890095778,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047737406920629675,
      "loss": 5.9649,
      "mean_token_accuracy": 0.15719179660081864,
      "num_tokens": 30953094.0,
      "step": 14310
    },
    {
      "entropy": 6.424063968658447,
      "epoch": 1.531863663117342,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047735735663853514,
      "loss": 5.9764,
      "mean_token_accuracy": 0.1572314038872719,
      "num_tokens": 30964604.0,
      "step": 14315
    },
    {
      "entropy": 6.416403293609619,
      "epoch": 1.5323987372251056,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004773406382277334,
      "loss": 6.0627,
      "mean_token_accuracy": 0.15206461399793625,
      "num_tokens": 30975681.0,
      "step": 14320
    },
    {
      "entropy": 6.473670482635498,
      "epoch": 1.5329338113328697,
      "grad_norm": 1.0,
      "learning_rate": 0.0004773239139743741,
      "loss": 6.0155,
      "mean_token_accuracy": 0.15775269716978074,
      "num_tokens": 30986319.0,
      "step": 14325
    },
    {
      "entropy": 6.442071533203125,
      "epoch": 1.5334688854406335,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004773071838789402,
      "loss": 5.9881,
      "mean_token_accuracy": 0.15276835709810258,
      "num_tokens": 30997077.0,
      "step": 14330
    },
    {
      "entropy": 6.3459302425384525,
      "epoch": 1.5340039595483974,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004772904479419148,
      "loss": 5.8969,
      "mean_token_accuracy": 0.1599753737449646,
      "num_tokens": 31007485.0,
      "step": 14335
    },
    {
      "entropy": 6.377594614028931,
      "epoch": 1.5345390336561615,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004772737061637811,
      "loss": 5.9762,
      "mean_token_accuracy": 0.16264894902706145,
      "num_tokens": 31017827.0,
      "step": 14340
    },
    {
      "entropy": 6.4241735458374025,
      "epoch": 1.5350741077639252,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047725695854502267,
      "loss": 6.0248,
      "mean_token_accuracy": 0.158640319108963,
      "num_tokens": 31028167.0,
      "step": 14345
    },
    {
      "entropy": 6.398904180526733,
      "epoch": 1.5356091818716893,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047724020508612287,
      "loss": 5.9661,
      "mean_token_accuracy": 0.1588180586695671,
      "num_tokens": 31038941.0,
      "step": 14350
    },
    {
      "entropy": 6.38801326751709,
      "epoch": 1.5361442559794531,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047722344578756564,
      "loss": 5.9615,
      "mean_token_accuracy": 0.1547321930527687,
      "num_tokens": 31049949.0,
      "step": 14355
    },
    {
      "entropy": 6.416993856430054,
      "epoch": 1.536679330087217,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047720668064983476,
      "loss": 5.9593,
      "mean_token_accuracy": 0.1580192670226097,
      "num_tokens": 31059299.0,
      "step": 14360
    },
    {
      "entropy": 6.414626359939575,
      "epoch": 1.5372144041949811,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004771899096734145,
      "loss": 6.0737,
      "mean_token_accuracy": 0.14753491282463074,
      "num_tokens": 31070508.0,
      "step": 14365
    },
    {
      "entropy": 6.35029354095459,
      "epoch": 1.5377494783027448,
      "grad_norm": 0.875,
      "learning_rate": 0.00047717313285878886,
      "loss": 5.9439,
      "mean_token_accuracy": 0.16161288619041442,
      "num_tokens": 31082354.0,
      "step": 14370
    },
    {
      "entropy": 6.574608182907104,
      "epoch": 1.5382845524105089,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047715635020644247,
      "loss": 6.077,
      "mean_token_accuracy": 0.14998216778039933,
      "num_tokens": 31093496.0,
      "step": 14375
    },
    {
      "entropy": 6.42903504371643,
      "epoch": 1.5388196265182728,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047713956171685994,
      "loss": 5.9509,
      "mean_token_accuracy": 0.1600353680551052,
      "num_tokens": 31104289.0,
      "step": 14380
    },
    {
      "entropy": 6.279636859893799,
      "epoch": 1.5393547006260366,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000477122767390526,
      "loss": 5.891,
      "mean_token_accuracy": 0.16902408972382546,
      "num_tokens": 31114998.0,
      "step": 14385
    },
    {
      "entropy": 6.389548063278198,
      "epoch": 1.5398897747338007,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047710596722792543,
      "loss": 5.9506,
      "mean_token_accuracy": 0.15374597087502478,
      "num_tokens": 31127393.0,
      "step": 14390
    },
    {
      "entropy": 6.352050924301148,
      "epoch": 1.5404248488415646,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004770891612295435,
      "loss": 5.7922,
      "mean_token_accuracy": 0.17488379180431365,
      "num_tokens": 31137484.0,
      "step": 14395
    },
    {
      "entropy": 6.413282251358032,
      "epoch": 1.5409599229493285,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004770723493958654,
      "loss": 6.0363,
      "mean_token_accuracy": 0.15503147393465042,
      "num_tokens": 31147863.0,
      "step": 14400
    },
    {
      "entropy": 6.36036319732666,
      "epoch": 1.5414949970570924,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004770555317273767,
      "loss": 5.905,
      "mean_token_accuracy": 0.16022974401712417,
      "num_tokens": 31159018.0,
      "step": 14405
    },
    {
      "entropy": 6.409140729904175,
      "epoch": 1.5420300711648562,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047703870822456287,
      "loss": 5.9216,
      "mean_token_accuracy": 0.15990501493215561,
      "num_tokens": 31170673.0,
      "step": 14410
    },
    {
      "entropy": 6.2663576126098635,
      "epoch": 1.5425651452726203,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004770218788879097,
      "loss": 5.8073,
      "mean_token_accuracy": 0.16898298114538193,
      "num_tokens": 31181165.0,
      "step": 14415
    },
    {
      "entropy": 6.290889120101928,
      "epoch": 1.5431002193803842,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004770050437179033,
      "loss": 5.9278,
      "mean_token_accuracy": 0.16292359605431556,
      "num_tokens": 31192976.0,
      "step": 14420
    },
    {
      "entropy": 6.491928815841675,
      "epoch": 1.543635293488148,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004769882027150295,
      "loss": 6.0369,
      "mean_token_accuracy": 0.15594052374362946,
      "num_tokens": 31203979.0,
      "step": 14425
    },
    {
      "entropy": 6.459931516647339,
      "epoch": 1.5441703675959122,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047697135587977473,
      "loss": 5.8989,
      "mean_token_accuracy": 0.16245424896478652,
      "num_tokens": 31214972.0,
      "step": 14430
    },
    {
      "entropy": 6.454630422592163,
      "epoch": 1.5447054417036759,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047695450321262546,
      "loss": 6.0312,
      "mean_token_accuracy": 0.15493249744176865,
      "num_tokens": 31224726.0,
      "step": 14435
    },
    {
      "entropy": 6.395449256896972,
      "epoch": 1.54524051581144,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004769376447140683,
      "loss": 5.9108,
      "mean_token_accuracy": 0.16531994491815566,
      "num_tokens": 31235761.0,
      "step": 14440
    },
    {
      "entropy": 6.392197227478027,
      "epoch": 1.5457755899192038,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047692078038459015,
      "loss": 6.0076,
      "mean_token_accuracy": 0.1632586345076561,
      "num_tokens": 31246221.0,
      "step": 14445
    },
    {
      "entropy": 6.425009870529175,
      "epoch": 1.5463106640269677,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004769039102246777,
      "loss": 5.9724,
      "mean_token_accuracy": 0.15540244430303574,
      "num_tokens": 31257364.0,
      "step": 14450
    },
    {
      "entropy": 6.436997127532959,
      "epoch": 1.5468457381347318,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047688703423481827,
      "loss": 5.9892,
      "mean_token_accuracy": 0.1529560998082161,
      "num_tokens": 31268343.0,
      "step": 14455
    },
    {
      "entropy": 6.420995759963989,
      "epoch": 1.5473808122424955,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047687015241549896,
      "loss": 6.0115,
      "mean_token_accuracy": 0.15487167835235596,
      "num_tokens": 31278335.0,
      "step": 14460
    },
    {
      "entropy": 6.418414354324341,
      "epoch": 1.5479158863502596,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004768532647672075,
      "loss": 5.9392,
      "mean_token_accuracy": 0.15308126360177993,
      "num_tokens": 31288537.0,
      "step": 14465
    },
    {
      "entropy": 6.355852508544922,
      "epoch": 1.5484509604580234,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004768363712904314,
      "loss": 5.9596,
      "mean_token_accuracy": 0.15877459049224854,
      "num_tokens": 31299353.0,
      "step": 14470
    },
    {
      "entropy": 6.468207263946534,
      "epoch": 1.5489860345657873,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004768194719856584,
      "loss": 5.9863,
      "mean_token_accuracy": 0.15388087332248687,
      "num_tokens": 31309655.0,
      "step": 14475
    },
    {
      "entropy": 6.404808759689331,
      "epoch": 1.5495211086735514,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004768025668533765,
      "loss": 6.0156,
      "mean_token_accuracy": 0.15866301357746124,
      "num_tokens": 31321008.0,
      "step": 14480
    },
    {
      "entropy": 6.354135322570801,
      "epoch": 1.550056182781315,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047678565589407383,
      "loss": 5.8986,
      "mean_token_accuracy": 0.15900446325540543,
      "num_tokens": 31331978.0,
      "step": 14485
    },
    {
      "entropy": 6.416317272186279,
      "epoch": 1.5505912568890792,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004767687391082387,
      "loss": 5.9605,
      "mean_token_accuracy": 0.16246418952941893,
      "num_tokens": 31341703.0,
      "step": 14490
    },
    {
      "entropy": 6.394450855255127,
      "epoch": 1.551126330996843,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047675181649635965,
      "loss": 5.9977,
      "mean_token_accuracy": 0.1645517796278,
      "num_tokens": 31352787.0,
      "step": 14495
    },
    {
      "entropy": 6.373354101181031,
      "epoch": 1.551661405104607,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047673488805892515,
      "loss": 5.9617,
      "mean_token_accuracy": 0.15728401988744736,
      "num_tokens": 31363702.0,
      "step": 14500
    },
    {
      "entropy": 6.4237652778625485,
      "epoch": 1.552196479212371,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047671795379642415,
      "loss": 5.9479,
      "mean_token_accuracy": 0.1494924709200859,
      "num_tokens": 31374719.0,
      "step": 14505
    },
    {
      "entropy": 6.504313659667969,
      "epoch": 1.5527315533201347,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004767010137093456,
      "loss": 6.0727,
      "mean_token_accuracy": 0.1476470947265625,
      "num_tokens": 31386289.0,
      "step": 14510
    },
    {
      "entropy": 6.42588210105896,
      "epoch": 1.5532666274278988,
      "grad_norm": 1.0,
      "learning_rate": 0.00047668406779817864,
      "loss": 5.8823,
      "mean_token_accuracy": 0.1613766834139824,
      "num_tokens": 31397103.0,
      "step": 14515
    },
    {
      "entropy": 6.397244119644165,
      "epoch": 1.5538017015356627,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004766671160634125,
      "loss": 6.0049,
      "mean_token_accuracy": 0.1609807014465332,
      "num_tokens": 31408715.0,
      "step": 14520
    },
    {
      "entropy": 6.495431709289551,
      "epoch": 1.5543367756434265,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004766501585055367,
      "loss": 6.0228,
      "mean_token_accuracy": 0.1490361288189888,
      "num_tokens": 31420365.0,
      "step": 14525
    },
    {
      "entropy": 6.399900817871094,
      "epoch": 1.5548718497511906,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000476633195125041,
      "loss": 5.9169,
      "mean_token_accuracy": 0.15983837097883224,
      "num_tokens": 31431635.0,
      "step": 14530
    },
    {
      "entropy": 6.414658737182617,
      "epoch": 1.5554069238589545,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047661622592241507,
      "loss": 5.9584,
      "mean_token_accuracy": 0.1577182337641716,
      "num_tokens": 31442513.0,
      "step": 14535
    },
    {
      "entropy": 6.40077018737793,
      "epoch": 1.5559419979667184,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004765992508981489,
      "loss": 5.9944,
      "mean_token_accuracy": 0.14715637713670732,
      "num_tokens": 31453726.0,
      "step": 14540
    },
    {
      "entropy": 6.4203564643859865,
      "epoch": 1.5564770720744823,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004765822700527328,
      "loss": 5.9053,
      "mean_token_accuracy": 0.16186023950576783,
      "num_tokens": 31463980.0,
      "step": 14545
    },
    {
      "entropy": 6.436416959762573,
      "epoch": 1.5570121461822461,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004765652833866569,
      "loss": 6.0605,
      "mean_token_accuracy": 0.14987971037626266,
      "num_tokens": 31474615.0,
      "step": 14550
    },
    {
      "entropy": 6.352265357971191,
      "epoch": 1.5575472202900102,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004765482909004118,
      "loss": 5.9082,
      "mean_token_accuracy": 0.16493077874183654,
      "num_tokens": 31485283.0,
      "step": 14555
    },
    {
      "entropy": 6.400459337234497,
      "epoch": 1.5580822943977741,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004765312925944881,
      "loss": 5.9628,
      "mean_token_accuracy": 0.15394781529903412,
      "num_tokens": 31496997.0,
      "step": 14560
    },
    {
      "entropy": 6.435737991333008,
      "epoch": 1.558617368505538,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047651428846937673,
      "loss": 5.9738,
      "mean_token_accuracy": 0.16104926764965058,
      "num_tokens": 31507585.0,
      "step": 14565
    },
    {
      "entropy": 6.423663520812989,
      "epoch": 1.559152442613302,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047649727852556854,
      "loss": 5.9069,
      "mean_token_accuracy": 0.16171808242797853,
      "num_tokens": 31517078.0,
      "step": 14570
    },
    {
      "entropy": 6.337673711776733,
      "epoch": 1.5596875167210658,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004764802627635547,
      "loss": 5.8966,
      "mean_token_accuracy": 0.16957565099000932,
      "num_tokens": 31528576.0,
      "step": 14575
    },
    {
      "entropy": 6.483899354934692,
      "epoch": 1.5602225908288299,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004764632411838266,
      "loss": 6.0101,
      "mean_token_accuracy": 0.15115008801221846,
      "num_tokens": 31540095.0,
      "step": 14580
    },
    {
      "entropy": 6.407795238494873,
      "epoch": 1.5607576649365937,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047644621378687573,
      "loss": 5.9071,
      "mean_token_accuracy": 0.15755292624235154,
      "num_tokens": 31550980.0,
      "step": 14585
    },
    {
      "entropy": 6.245433521270752,
      "epoch": 1.5612927390443576,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047642918057319377,
      "loss": 5.7135,
      "mean_token_accuracy": 0.1860009104013443,
      "num_tokens": 31561842.0,
      "step": 14590
    },
    {
      "entropy": 6.408733892440796,
      "epoch": 1.5618278131521217,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004764121415432725,
      "loss": 6.0639,
      "mean_token_accuracy": 0.14809912890195848,
      "num_tokens": 31572798.0,
      "step": 14595
    },
    {
      "entropy": 6.408991050720215,
      "epoch": 1.5623628872598854,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004763950966976039,
      "loss": 6.0136,
      "mean_token_accuracy": 0.15565649718046187,
      "num_tokens": 31584774.0,
      "step": 14600
    },
    {
      "entropy": 6.417719173431396,
      "epoch": 1.5628979613676495,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004763780460366803,
      "loss": 5.9488,
      "mean_token_accuracy": 0.15416090339422225,
      "num_tokens": 31596197.0,
      "step": 14605
    },
    {
      "entropy": 6.4419619083404545,
      "epoch": 1.5634330354754133,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004763609895609939,
      "loss": 5.9677,
      "mean_token_accuracy": 0.15382450371980666,
      "num_tokens": 31607455.0,
      "step": 14610
    },
    {
      "entropy": 6.430599117279053,
      "epoch": 1.5639681095831772,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047634392727103714,
      "loss": 6.0029,
      "mean_token_accuracy": 0.15627100169658661,
      "num_tokens": 31618995.0,
      "step": 14615
    },
    {
      "entropy": 6.3388307094573975,
      "epoch": 1.5645031836909413,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004763268591673028,
      "loss": 5.8675,
      "mean_token_accuracy": 0.1619246020913124,
      "num_tokens": 31629984.0,
      "step": 14620
    },
    {
      "entropy": 6.410207366943359,
      "epoch": 1.565038257798705,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004763097852502837,
      "loss": 5.8978,
      "mean_token_accuracy": 0.15848351269960403,
      "num_tokens": 31641362.0,
      "step": 14625
    },
    {
      "entropy": 6.445670127868652,
      "epoch": 1.565573331906469,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047629270552047284,
      "loss": 6.0807,
      "mean_token_accuracy": 0.1460634544491768,
      "num_tokens": 31652650.0,
      "step": 14630
    },
    {
      "entropy": 6.398959684371948,
      "epoch": 1.566108406014233,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047627561997836337,
      "loss": 5.9878,
      "mean_token_accuracy": 0.1489569813013077,
      "num_tokens": 31664201.0,
      "step": 14635
    },
    {
      "entropy": 6.450074911117554,
      "epoch": 1.5666434801219968,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004762585286244487,
      "loss": 5.9373,
      "mean_token_accuracy": 0.15585556775331497,
      "num_tokens": 31674480.0,
      "step": 14640
    },
    {
      "entropy": 6.425080728530884,
      "epoch": 1.567178554229761,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047624143145922227,
      "loss": 5.9577,
      "mean_token_accuracy": 0.1560557872056961,
      "num_tokens": 31685377.0,
      "step": 14645
    },
    {
      "entropy": 6.426592922210693,
      "epoch": 1.5677136283375246,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047622432848317775,
      "loss": 5.9472,
      "mean_token_accuracy": 0.15826667845249176,
      "num_tokens": 31695199.0,
      "step": 14650
    },
    {
      "entropy": 6.3653017520904545,
      "epoch": 1.5682487024452887,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004762072196968091,
      "loss": 5.9789,
      "mean_token_accuracy": 0.15650330930948259,
      "num_tokens": 31705528.0,
      "step": 14655
    },
    {
      "entropy": 6.292976570129395,
      "epoch": 1.5687837765530526,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004761901051006102,
      "loss": 5.8973,
      "mean_token_accuracy": 0.163994500041008,
      "num_tokens": 31718156.0,
      "step": 14660
    },
    {
      "entropy": 6.45362491607666,
      "epoch": 1.5693188506608164,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047617298469507534,
      "loss": 5.9461,
      "mean_token_accuracy": 0.15857689827680588,
      "num_tokens": 31729783.0,
      "step": 14665
    },
    {
      "entropy": 6.4490966796875,
      "epoch": 1.5698539247685805,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004761558584806988,
      "loss": 5.9284,
      "mean_token_accuracy": 0.15827034860849382,
      "num_tokens": 31740796.0,
      "step": 14670
    },
    {
      "entropy": 6.456551837921142,
      "epoch": 1.5703889988763444,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004761387264579751,
      "loss": 6.0161,
      "mean_token_accuracy": 0.15322857797145845,
      "num_tokens": 31750559.0,
      "step": 14675
    },
    {
      "entropy": 6.382494783401489,
      "epoch": 1.5709240729841083,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004761215886273989,
      "loss": 5.9099,
      "mean_token_accuracy": 0.15894279778003692,
      "num_tokens": 31761559.0,
      "step": 14680
    },
    {
      "entropy": 6.376495218276977,
      "epoch": 1.5714591470918722,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047610444498946507,
      "loss": 5.8805,
      "mean_token_accuracy": 0.16104390770196914,
      "num_tokens": 31771677.0,
      "step": 14685
    },
    {
      "entropy": 6.424437856674194,
      "epoch": 1.571994221199636,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004760872955446688,
      "loss": 6.0295,
      "mean_token_accuracy": 0.15021618008613585,
      "num_tokens": 31783028.0,
      "step": 14690
    },
    {
      "entropy": 6.430985355377198,
      "epoch": 1.5725292953074002,
      "grad_norm": 1.125,
      "learning_rate": 0.00047607014029350504,
      "loss": 5.9888,
      "mean_token_accuracy": 0.15902926921844482,
      "num_tokens": 31793110.0,
      "step": 14695
    },
    {
      "entropy": 6.355496644973755,
      "epoch": 1.573064369415164,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047605297923646923,
      "loss": 5.9434,
      "mean_token_accuracy": 0.15966074615716935,
      "num_tokens": 31804193.0,
      "step": 14700
    },
    {
      "entropy": 6.3916956901550295,
      "epoch": 1.573599443522928,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047603581237405684,
      "loss": 5.9601,
      "mean_token_accuracy": 0.15990700125694274,
      "num_tokens": 31815752.0,
      "step": 14705
    },
    {
      "entropy": 6.44994158744812,
      "epoch": 1.574134517630692,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047601863970676367,
      "loss": 6.0235,
      "mean_token_accuracy": 0.15400762110948563,
      "num_tokens": 31825936.0,
      "step": 14710
    },
    {
      "entropy": 6.403060579299927,
      "epoch": 1.5746695917384557,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047600146123508554,
      "loss": 5.9302,
      "mean_token_accuracy": 0.15498108863830568,
      "num_tokens": 31836601.0,
      "step": 14715
    },
    {
      "entropy": 6.2967394351959225,
      "epoch": 1.5752046658462198,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047598427695951846,
      "loss": 5.7873,
      "mean_token_accuracy": 0.17311608493328096,
      "num_tokens": 31846476.0,
      "step": 14720
    },
    {
      "entropy": 6.324089193344117,
      "epoch": 1.5757397399539836,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047596708688055857,
      "loss": 5.8173,
      "mean_token_accuracy": 0.16601997166872023,
      "num_tokens": 31856151.0,
      "step": 14725
    },
    {
      "entropy": 6.420996236801147,
      "epoch": 1.5762748140617475,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004759498909987023,
      "loss": 6.0164,
      "mean_token_accuracy": 0.1533183604478836,
      "num_tokens": 31867419.0,
      "step": 14730
    },
    {
      "entropy": 6.432789373397827,
      "epoch": 1.5768098881695116,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004759326893144462,
      "loss": 6.0388,
      "mean_token_accuracy": 0.15360198616981507,
      "num_tokens": 31878036.0,
      "step": 14735
    },
    {
      "entropy": 6.411780071258545,
      "epoch": 1.5773449622772753,
      "grad_norm": 0.96875,
      "learning_rate": 0.000475915481828287,
      "loss": 5.9716,
      "mean_token_accuracy": 0.15401259809732437,
      "num_tokens": 31887919.0,
      "step": 14740
    },
    {
      "entropy": 6.358624029159546,
      "epoch": 1.5778800363850394,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047589826854072133,
      "loss": 5.9731,
      "mean_token_accuracy": 0.15420302003622055,
      "num_tokens": 31898962.0,
      "step": 14745
    },
    {
      "entropy": 6.403563022613525,
      "epoch": 1.5784151104928033,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047588104945224654,
      "loss": 6.0819,
      "mean_token_accuracy": 0.1497616410255432,
      "num_tokens": 31911027.0,
      "step": 14750
    },
    {
      "entropy": 6.486897134780884,
      "epoch": 1.5789501846005671,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004758638245633595,
      "loss": 6.0031,
      "mean_token_accuracy": 0.15857743322849274,
      "num_tokens": 31921226.0,
      "step": 14755
    },
    {
      "entropy": 6.374368143081665,
      "epoch": 1.5794852587083312,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004758465938745579,
      "loss": 5.9417,
      "mean_token_accuracy": 0.15719680786132811,
      "num_tokens": 31932848.0,
      "step": 14760
    },
    {
      "entropy": 6.324529647827148,
      "epoch": 1.5800203328160949,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047582935738633915,
      "loss": 5.8508,
      "mean_token_accuracy": 0.1599285915493965,
      "num_tokens": 31943757.0,
      "step": 14765
    },
    {
      "entropy": 6.432357311248779,
      "epoch": 1.580555406923859,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004758121150992009,
      "loss": 6.0071,
      "mean_token_accuracy": 0.15109248757362365,
      "num_tokens": 31954391.0,
      "step": 14770
    },
    {
      "entropy": 6.438722276687622,
      "epoch": 1.5810904810316229,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047579486701364093,
      "loss": 6.0096,
      "mean_token_accuracy": 0.15839426070451737,
      "num_tokens": 31964497.0,
      "step": 14775
    },
    {
      "entropy": 6.484709548950195,
      "epoch": 1.5816255551393867,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004757776131301575,
      "loss": 6.0087,
      "mean_token_accuracy": 0.1484030395746231,
      "num_tokens": 31974466.0,
      "step": 14780
    },
    {
      "entropy": 6.460316944122314,
      "epoch": 1.5821606292471508,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004757603534492487,
      "loss": 6.0236,
      "mean_token_accuracy": 0.15612092912197112,
      "num_tokens": 31985890.0,
      "step": 14785
    },
    {
      "entropy": 6.409747171401977,
      "epoch": 1.5826957033549145,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004757430879714128,
      "loss": 6.0726,
      "mean_token_accuracy": 0.15832225233316422,
      "num_tokens": 31996865.0,
      "step": 14790
    },
    {
      "entropy": 6.3112400531768795,
      "epoch": 1.5832307774626786,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004757258166971485,
      "loss": 5.8459,
      "mean_token_accuracy": 0.16624507904052735,
      "num_tokens": 32008121.0,
      "step": 14795
    },
    {
      "entropy": 6.421030712127686,
      "epoch": 1.5837658515704425,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004757085396269545,
      "loss": 6.0475,
      "mean_token_accuracy": 0.1541065789759159,
      "num_tokens": 32020552.0,
      "step": 14800
    },
    {
      "entropy": 6.4080511093139645,
      "epoch": 1.5843009256782064,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047569125676132963,
      "loss": 5.9966,
      "mean_token_accuracy": 0.15653155148029327,
      "num_tokens": 32032361.0,
      "step": 14805
    },
    {
      "entropy": 6.501546382904053,
      "epoch": 1.5848359997859705,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047567396810077286,
      "loss": 5.9864,
      "mean_token_accuracy": 0.1513087585568428,
      "num_tokens": 32043233.0,
      "step": 14810
    },
    {
      "entropy": 6.437207889556885,
      "epoch": 1.5853710738937343,
      "grad_norm": 1.0,
      "learning_rate": 0.00047565667364578344,
      "loss": 5.955,
      "mean_token_accuracy": 0.16178649440407752,
      "num_tokens": 32054273.0,
      "step": 14815
    },
    {
      "entropy": 6.401141786575318,
      "epoch": 1.5859061480014982,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004756393733968608,
      "loss": 5.9396,
      "mean_token_accuracy": 0.16384750008583068,
      "num_tokens": 32064200.0,
      "step": 14820
    },
    {
      "entropy": 6.341894340515137,
      "epoch": 1.586441222109262,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004756220673545045,
      "loss": 5.934,
      "mean_token_accuracy": 0.15380320996046065,
      "num_tokens": 32074823.0,
      "step": 14825
    },
    {
      "entropy": 6.412326002120972,
      "epoch": 1.586976296217026,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047560475551921416,
      "loss": 6.0218,
      "mean_token_accuracy": 0.15705028027296067,
      "num_tokens": 32085333.0,
      "step": 14830
    },
    {
      "entropy": 6.439935874938965,
      "epoch": 1.58751137032479,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004755874378914897,
      "loss": 6.0033,
      "mean_token_accuracy": 0.14712767228484153,
      "num_tokens": 32095827.0,
      "step": 14835
    },
    {
      "entropy": 6.418816423416137,
      "epoch": 1.588046444432554,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004755701144718311,
      "loss": 6.0013,
      "mean_token_accuracy": 0.15488713681697847,
      "num_tokens": 32106949.0,
      "step": 14840
    },
    {
      "entropy": 6.406856298446655,
      "epoch": 1.5885815185403178,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047555278526073867,
      "loss": 5.9411,
      "mean_token_accuracy": 0.15591731071472167,
      "num_tokens": 32118269.0,
      "step": 14845
    },
    {
      "entropy": 6.390598917007447,
      "epoch": 1.589116592648082,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004755354502587127,
      "loss": 5.9725,
      "mean_token_accuracy": 0.1553051769733429,
      "num_tokens": 32128406.0,
      "step": 14850
    },
    {
      "entropy": 6.346534585952758,
      "epoch": 1.5896516667558456,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047551810946625385,
      "loss": 5.971,
      "mean_token_accuracy": 0.1595791608095169,
      "num_tokens": 32140112.0,
      "step": 14855
    },
    {
      "entropy": 6.436548137664795,
      "epoch": 1.5901867408636097,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047550076288386276,
      "loss": 5.9099,
      "mean_token_accuracy": 0.16902150362730026,
      "num_tokens": 32150010.0,
      "step": 14860
    },
    {
      "entropy": 6.433078575134277,
      "epoch": 1.5907218149713735,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004754834105120403,
      "loss": 6.0212,
      "mean_token_accuracy": 0.15099380165338516,
      "num_tokens": 32160694.0,
      "step": 14865
    },
    {
      "entropy": 6.403913497924805,
      "epoch": 1.5912568890791374,
      "grad_norm": 1.0,
      "learning_rate": 0.00047546605235128755,
      "loss": 6.0307,
      "mean_token_accuracy": 0.14985934719443322,
      "num_tokens": 32171947.0,
      "step": 14870
    },
    {
      "entropy": 6.4543421268463135,
      "epoch": 1.5917919631869015,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004754486884021057,
      "loss": 5.9903,
      "mean_token_accuracy": 0.15334563627839087,
      "num_tokens": 32182849.0,
      "step": 14875
    },
    {
      "entropy": 6.427846908569336,
      "epoch": 1.5923270372946652,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004754313186649961,
      "loss": 6.008,
      "mean_token_accuracy": 0.15399690419435502,
      "num_tokens": 32192871.0,
      "step": 14880
    },
    {
      "entropy": 6.453654623031616,
      "epoch": 1.5928621114024293,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047541394314046037,
      "loss": 6.0418,
      "mean_token_accuracy": 0.1468898192048073,
      "num_tokens": 32203289.0,
      "step": 14885
    },
    {
      "entropy": 6.4335126876831055,
      "epoch": 1.5933971855101932,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004753965618290002,
      "loss": 5.9904,
      "mean_token_accuracy": 0.1564868539571762,
      "num_tokens": 32213238.0,
      "step": 14890
    },
    {
      "entropy": 6.360141801834106,
      "epoch": 1.593932259617957,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004753791747311175,
      "loss": 5.8773,
      "mean_token_accuracy": 0.16788973808288574,
      "num_tokens": 32223384.0,
      "step": 14895
    },
    {
      "entropy": 6.416360712051391,
      "epoch": 1.5944673337257211,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047536178184731426,
      "loss": 5.9828,
      "mean_token_accuracy": 0.15664443224668503,
      "num_tokens": 32235358.0,
      "step": 14900
    },
    {
      "entropy": 6.550871086120606,
      "epoch": 1.5950024078334848,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004753443831780927,
      "loss": 6.0343,
      "mean_token_accuracy": 0.15315500348806382,
      "num_tokens": 32246459.0,
      "step": 14905
    },
    {
      "entropy": 6.412514543533325,
      "epoch": 1.595537481941249,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004753269787239552,
      "loss": 5.9226,
      "mean_token_accuracy": 0.16310400068759917,
      "num_tokens": 32257798.0,
      "step": 14910
    },
    {
      "entropy": 6.402127599716186,
      "epoch": 1.5960725560490128,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047530956848540435,
      "loss": 5.9612,
      "mean_token_accuracy": 0.15815509706735612,
      "num_tokens": 32268118.0,
      "step": 14915
    },
    {
      "entropy": 6.344193792343139,
      "epoch": 1.5966076301567766,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004752921524629429,
      "loss": 5.8796,
      "mean_token_accuracy": 0.15868421494960785,
      "num_tokens": 32279676.0,
      "step": 14920
    },
    {
      "entropy": 6.4014301776885985,
      "epoch": 1.5971427042645407,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004752747306570737,
      "loss": 5.943,
      "mean_token_accuracy": 0.15817199051380157,
      "num_tokens": 32290613.0,
      "step": 14925
    },
    {
      "entropy": 6.42701153755188,
      "epoch": 1.5976777783723044,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047525730306829977,
      "loss": 5.9788,
      "mean_token_accuracy": 0.1573183059692383,
      "num_tokens": 32300873.0,
      "step": 14930
    },
    {
      "entropy": 6.408425283432007,
      "epoch": 1.5982128524800685,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004752398696971244,
      "loss": 5.9904,
      "mean_token_accuracy": 0.15795271918177606,
      "num_tokens": 32312326.0,
      "step": 14935
    },
    {
      "entropy": 6.45890965461731,
      "epoch": 1.5987479265878324,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047522243054405085,
      "loss": 5.9609,
      "mean_token_accuracy": 0.1466933473944664,
      "num_tokens": 32324322.0,
      "step": 14940
    },
    {
      "entropy": 6.409703588485717,
      "epoch": 1.5992830006955963,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004752049856095828,
      "loss": 5.9507,
      "mean_token_accuracy": 0.15573473125696183,
      "num_tokens": 32334397.0,
      "step": 14945
    },
    {
      "entropy": 6.397755241394043,
      "epoch": 1.5998180748033604,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047518753489422393,
      "loss": 5.9337,
      "mean_token_accuracy": 0.1609349951148033,
      "num_tokens": 32345131.0,
      "step": 14950
    },
    {
      "entropy": 6.360000228881836,
      "epoch": 1.6003531489111242,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047517007839847816,
      "loss": 5.9081,
      "mean_token_accuracy": 0.1572778731584549,
      "num_tokens": 32355520.0,
      "step": 14955
    },
    {
      "entropy": 6.411602306365967,
      "epoch": 1.6008882230188881,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004751526161228494,
      "loss": 6.013,
      "mean_token_accuracy": 0.16260802149772643,
      "num_tokens": 32366695.0,
      "step": 14960
    },
    {
      "entropy": 6.353814029693604,
      "epoch": 1.601423297126652,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000475135148067842,
      "loss": 5.969,
      "mean_token_accuracy": 0.15295666158199311,
      "num_tokens": 32377467.0,
      "step": 14965
    },
    {
      "entropy": 6.4740540981292725,
      "epoch": 1.6019583712344159,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047511767423396033,
      "loss": 5.9616,
      "mean_token_accuracy": 0.15341430380940438,
      "num_tokens": 32388021.0,
      "step": 14970
    },
    {
      "entropy": 6.4092442989349365,
      "epoch": 1.60249344534218,
      "grad_norm": 1.015625,
      "learning_rate": 0.000475100194621709,
      "loss": 5.9252,
      "mean_token_accuracy": 0.1597691372036934,
      "num_tokens": 32398105.0,
      "step": 14975
    },
    {
      "entropy": 6.38935923576355,
      "epoch": 1.6030285194499438,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004750827092315926,
      "loss": 6.066,
      "mean_token_accuracy": 0.15135482251644133,
      "num_tokens": 32409876.0,
      "step": 14980
    },
    {
      "entropy": 6.427843475341797,
      "epoch": 1.6035635935577077,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047506521806411613,
      "loss": 5.9285,
      "mean_token_accuracy": 0.17237737029790878,
      "num_tokens": 32420283.0,
      "step": 14985
    },
    {
      "entropy": 6.43850302696228,
      "epoch": 1.6040986676654718,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047504772111978463,
      "loss": 6.0523,
      "mean_token_accuracy": 0.14464129954576493,
      "num_tokens": 32431164.0,
      "step": 14990
    },
    {
      "entropy": 6.4384643077850345,
      "epoch": 1.6046337417732355,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004750302183991032,
      "loss": 5.9441,
      "mean_token_accuracy": 0.1565098874270916,
      "num_tokens": 32442253.0,
      "step": 14995
    },
    {
      "entropy": 6.443842172622681,
      "epoch": 1.6051688158809996,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004750127099025774,
      "loss": 6.038,
      "mean_token_accuracy": 0.1550551787018776,
      "num_tokens": 32452800.0,
      "step": 15000
    },
    {
      "epoch": 1.6051688158809996,
      "eval_entropy": 6.23317083631243,
      "eval_loss": 6.05530309677124,
      "eval_mean_token_accuracy": 0.1603909620575888,
      "eval_num_tokens": 32452800.0,
      "eval_runtime": 17.6505,
      "eval_samples_per_second": 1681.48,
      "eval_steps_per_second": 210.192,
      "step": 15000
    },
    {
      "entropy": 6.450182819366455,
      "epoch": 1.6057038899887635,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004749951956307126,
      "loss": 5.992,
      "mean_token_accuracy": 0.15406063199043274,
      "num_tokens": 32462890.0,
      "step": 15005
    },
    {
      "entropy": 6.37549147605896,
      "epoch": 1.6062389640965273,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004749776755840148,
      "loss": 5.9402,
      "mean_token_accuracy": 0.16463536769151688,
      "num_tokens": 32473422.0,
      "step": 15010
    },
    {
      "entropy": 6.383621788024902,
      "epoch": 1.6067740382042914,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047496014976298957,
      "loss": 5.9868,
      "mean_token_accuracy": 0.15814928263425826,
      "num_tokens": 32483980.0,
      "step": 15015
    },
    {
      "entropy": 6.493856382369995,
      "epoch": 1.607309112312055,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004749426181681432,
      "loss": 6.004,
      "mean_token_accuracy": 0.15338917821645737,
      "num_tokens": 32494905.0,
      "step": 15020
    },
    {
      "entropy": 6.5152153968811035,
      "epoch": 1.6078441864198192,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047492508079998173,
      "loss": 6.0386,
      "mean_token_accuracy": 0.1521016925573349,
      "num_tokens": 32505799.0,
      "step": 15025
    },
    {
      "entropy": 6.445469045639038,
      "epoch": 1.608379260527583,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004749075376590117,
      "loss": 6.0276,
      "mean_token_accuracy": 0.15248295217752456,
      "num_tokens": 32516328.0,
      "step": 15030
    },
    {
      "entropy": 6.435227775573731,
      "epoch": 1.608914334635347,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004748899887457396,
      "loss": 5.9952,
      "mean_token_accuracy": 0.16168775856494905,
      "num_tokens": 32526335.0,
      "step": 15035
    },
    {
      "entropy": 6.4173565864562985,
      "epoch": 1.609449408743111,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047487243406067227,
      "loss": 5.9517,
      "mean_token_accuracy": 0.16079121381044387,
      "num_tokens": 32536896.0,
      "step": 15040
    },
    {
      "entropy": 6.362458276748657,
      "epoch": 1.6099844828508747,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004748548736043164,
      "loss": 5.8706,
      "mean_token_accuracy": 0.16273298189044,
      "num_tokens": 32547604.0,
      "step": 15045
    },
    {
      "entropy": 6.46412672996521,
      "epoch": 1.6105195569586388,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047483730737717914,
      "loss": 6.0076,
      "mean_token_accuracy": 0.15045315027236938,
      "num_tokens": 32558539.0,
      "step": 15050
    },
    {
      "entropy": 6.479256057739258,
      "epoch": 1.6110546310664027,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004748197353797677,
      "loss": 6.073,
      "mean_token_accuracy": 0.15133537948131562,
      "num_tokens": 32569787.0,
      "step": 15055
    },
    {
      "entropy": 6.420907115936279,
      "epoch": 1.6115897051741666,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004748021576125895,
      "loss": 6.0095,
      "mean_token_accuracy": 0.1551932781934738,
      "num_tokens": 32580146.0,
      "step": 15060
    },
    {
      "entropy": 6.398712491989135,
      "epoch": 1.6121247792819307,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004747845740761521,
      "loss": 5.978,
      "mean_token_accuracy": 0.15450613871216773,
      "num_tokens": 32591243.0,
      "step": 15065
    },
    {
      "entropy": 6.447966194152832,
      "epoch": 1.6126598533896943,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004747669847709631,
      "loss": 5.9974,
      "mean_token_accuracy": 0.16027649641036987,
      "num_tokens": 32602424.0,
      "step": 15070
    },
    {
      "entropy": 6.474065923690796,
      "epoch": 1.6131949274974584,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004747493896975306,
      "loss": 6.083,
      "mean_token_accuracy": 0.14921532571315765,
      "num_tokens": 32613703.0,
      "step": 15075
    },
    {
      "entropy": 6.431403160095215,
      "epoch": 1.6137300016052223,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004747317888563625,
      "loss": 6.0269,
      "mean_token_accuracy": 0.1481316603720188,
      "num_tokens": 32624371.0,
      "step": 15080
    },
    {
      "entropy": 6.459831809997558,
      "epoch": 1.6142650757129862,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047471418224796704,
      "loss": 5.9824,
      "mean_token_accuracy": 0.15581246912479402,
      "num_tokens": 32634039.0,
      "step": 15085
    },
    {
      "entropy": 6.419746541976929,
      "epoch": 1.6148001498207503,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004746965698728526,
      "loss": 5.9349,
      "mean_token_accuracy": 0.16012608110904694,
      "num_tokens": 32644644.0,
      "step": 15090
    },
    {
      "entropy": 6.459805202484131,
      "epoch": 1.6153352239285141,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047467895173152787,
      "loss": 5.9924,
      "mean_token_accuracy": 0.16298181414604188,
      "num_tokens": 32655961.0,
      "step": 15095
    },
    {
      "entropy": 6.386929512023926,
      "epoch": 1.615870298036278,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047466132782450146,
      "loss": 5.905,
      "mean_token_accuracy": 0.16620202511548995,
      "num_tokens": 32665506.0,
      "step": 15100
    },
    {
      "entropy": 6.348428726196289,
      "epoch": 1.616405372144042,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047464369815228215,
      "loss": 5.9264,
      "mean_token_accuracy": 0.15948545634746553,
      "num_tokens": 32676278.0,
      "step": 15105
    },
    {
      "entropy": 6.412627172470093,
      "epoch": 1.6169404462518058,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004746260627153792,
      "loss": 5.9169,
      "mean_token_accuracy": 0.15350343883037568,
      "num_tokens": 32686835.0,
      "step": 15110
    },
    {
      "entropy": 6.488045310974121,
      "epoch": 1.6174755203595699,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047460842151430166,
      "loss": 6.0747,
      "mean_token_accuracy": 0.14917765259742738,
      "num_tokens": 32696856.0,
      "step": 15115
    },
    {
      "entropy": 6.456389284133911,
      "epoch": 1.6180105944673338,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047459077454955914,
      "loss": 5.9642,
      "mean_token_accuracy": 0.154352068901062,
      "num_tokens": 32707514.0,
      "step": 15120
    },
    {
      "entropy": 6.43286280632019,
      "epoch": 1.6185456685750976,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047457312182166094,
      "loss": 6.0651,
      "mean_token_accuracy": 0.15182463750243186,
      "num_tokens": 32718934.0,
      "step": 15125
    },
    {
      "entropy": 6.4588745594024655,
      "epoch": 1.6190807426828617,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004745554633311169,
      "loss": 5.9422,
      "mean_token_accuracy": 0.15749026238918304,
      "num_tokens": 32729249.0,
      "step": 15130
    },
    {
      "entropy": 6.466430377960205,
      "epoch": 1.6196158167906254,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047453779907843686,
      "loss": 6.0294,
      "mean_token_accuracy": 0.1522265799343586,
      "num_tokens": 32739948.0,
      "step": 15135
    },
    {
      "entropy": 6.415965747833252,
      "epoch": 1.6201508908983895,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047452012906413105,
      "loss": 6.0149,
      "mean_token_accuracy": 0.15517785847187043,
      "num_tokens": 32751482.0,
      "step": 15140
    },
    {
      "entropy": 6.423072433471679,
      "epoch": 1.6206859650061534,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047450245328870943,
      "loss": 5.8517,
      "mean_token_accuracy": 0.1672578603029251,
      "num_tokens": 32761260.0,
      "step": 15145
    },
    {
      "entropy": 6.477628469467163,
      "epoch": 1.6212210391139172,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047448477175268255,
      "loss": 5.9874,
      "mean_token_accuracy": 0.15562281310558318,
      "num_tokens": 32772684.0,
      "step": 15150
    },
    {
      "entropy": 6.277187919616699,
      "epoch": 1.6217561132216813,
      "grad_norm": 1.0,
      "learning_rate": 0.00047446708445656095,
      "loss": 5.7913,
      "mean_token_accuracy": 0.16923835426568984,
      "num_tokens": 32784022.0,
      "step": 15155
    },
    {
      "entropy": 6.374023389816284,
      "epoch": 1.622291187329445,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004744493914008552,
      "loss": 5.9232,
      "mean_token_accuracy": 0.15668537020683287,
      "num_tokens": 32794104.0,
      "step": 15160
    },
    {
      "entropy": 6.446038484573364,
      "epoch": 1.622826261437209,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004744316925860764,
      "loss": 5.9273,
      "mean_token_accuracy": 0.1630878508090973,
      "num_tokens": 32804038.0,
      "step": 15165
    },
    {
      "entropy": 6.410863542556763,
      "epoch": 1.623361335544973,
      "grad_norm": 1.0,
      "learning_rate": 0.0004744139880127355,
      "loss": 5.9857,
      "mean_token_accuracy": 0.15929034948349,
      "num_tokens": 32815620.0,
      "step": 15170
    },
    {
      "entropy": 6.395233917236328,
      "epoch": 1.6238964096527368,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047439627768134366,
      "loss": 5.8832,
      "mean_token_accuracy": 0.16801934987306594,
      "num_tokens": 32826353.0,
      "step": 15175
    },
    {
      "entropy": 6.4252564907073975,
      "epoch": 1.624431483760501,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004743785615924124,
      "loss": 6.0346,
      "mean_token_accuracy": 0.14915986657142638,
      "num_tokens": 32836582.0,
      "step": 15180
    },
    {
      "entropy": 6.374610090255738,
      "epoch": 1.6249665578682646,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004743608397464531,
      "loss": 5.9758,
      "mean_token_accuracy": 0.1584508091211319,
      "num_tokens": 32848073.0,
      "step": 15185
    },
    {
      "entropy": 6.418408441543579,
      "epoch": 1.6255016319760287,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047434311214397765,
      "loss": 6.0118,
      "mean_token_accuracy": 0.1524984508752823,
      "num_tokens": 32857839.0,
      "step": 15190
    },
    {
      "entropy": 6.434469604492188,
      "epoch": 1.6260367060837926,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047432537878549786,
      "loss": 5.9026,
      "mean_token_accuracy": 0.1755142629146576,
      "num_tokens": 32868087.0,
      "step": 15195
    },
    {
      "entropy": 6.438942098617554,
      "epoch": 1.6265717801915565,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047430763967152564,
      "loss": 5.9843,
      "mean_token_accuracy": 0.15411747395992278,
      "num_tokens": 32880622.0,
      "step": 15200
    },
    {
      "entropy": 6.478228378295898,
      "epoch": 1.6271068542993206,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047428989480257337,
      "loss": 5.9812,
      "mean_token_accuracy": 0.15548082441091537,
      "num_tokens": 32891405.0,
      "step": 15205
    },
    {
      "entropy": 6.417403030395508,
      "epoch": 1.6276419284070842,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004742721441791535,
      "loss": 5.9272,
      "mean_token_accuracy": 0.15932783335447312,
      "num_tokens": 32900487.0,
      "step": 15210
    },
    {
      "entropy": 6.376625490188599,
      "epoch": 1.6281770025148483,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004742543878017784,
      "loss": 5.8801,
      "mean_token_accuracy": 0.16860706359148026,
      "num_tokens": 32912139.0,
      "step": 15215
    },
    {
      "entropy": 6.43062801361084,
      "epoch": 1.6287120766226122,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047423662567096085,
      "loss": 5.9667,
      "mean_token_accuracy": 0.15549841225147248,
      "num_tokens": 32922972.0,
      "step": 15220
    },
    {
      "entropy": 6.415035581588745,
      "epoch": 1.629247150730376,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047421885778721376,
      "loss": 5.9145,
      "mean_token_accuracy": 0.16154803484678268,
      "num_tokens": 32932609.0,
      "step": 15225
    },
    {
      "entropy": 6.352713251113892,
      "epoch": 1.6297822248381402,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047420108415105006,
      "loss": 5.9405,
      "mean_token_accuracy": 0.16452530771493912,
      "num_tokens": 32942706.0,
      "step": 15230
    },
    {
      "entropy": 6.352208375930786,
      "epoch": 1.630317298945904,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047418330476298317,
      "loss": 5.9358,
      "mean_token_accuracy": 0.16049565225839615,
      "num_tokens": 32952240.0,
      "step": 15235
    },
    {
      "entropy": 6.40999698638916,
      "epoch": 1.630852373053668,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004741655196235262,
      "loss": 6.0119,
      "mean_token_accuracy": 0.15121943578124047,
      "num_tokens": 32963249.0,
      "step": 15240
    },
    {
      "entropy": 6.461125707626342,
      "epoch": 1.6313874471614318,
      "grad_norm": 1.0,
      "learning_rate": 0.00047414772873319294,
      "loss": 5.972,
      "mean_token_accuracy": 0.1617278650403023,
      "num_tokens": 32974004.0,
      "step": 15245
    },
    {
      "entropy": 6.450625848770142,
      "epoch": 1.6319225212691957,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000474129932092497,
      "loss": 5.9579,
      "mean_token_accuracy": 0.15511960610747338,
      "num_tokens": 32985236.0,
      "step": 15250
    },
    {
      "entropy": 6.4713146686553955,
      "epoch": 1.6324575953769598,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004741121297019522,
      "loss": 6.0155,
      "mean_token_accuracy": 0.15602864027023317,
      "num_tokens": 32995155.0,
      "step": 15255
    },
    {
      "entropy": 6.440524911880493,
      "epoch": 1.6329926694847237,
      "grad_norm": 0.859375,
      "learning_rate": 0.00047409432156207264,
      "loss": 5.9744,
      "mean_token_accuracy": 0.1658593624830246,
      "num_tokens": 33007562.0,
      "step": 15260
    },
    {
      "entropy": 6.4152562618255615,
      "epoch": 1.6335277435924875,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047407650767337256,
      "loss": 5.9647,
      "mean_token_accuracy": 0.15219125300645828,
      "num_tokens": 33018653.0,
      "step": 15265
    },
    {
      "entropy": 6.495700263977051,
      "epoch": 1.6340628177002516,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004740586880363663,
      "loss": 6.0558,
      "mean_token_accuracy": 0.14779894500970842,
      "num_tokens": 33029470.0,
      "step": 15270
    },
    {
      "entropy": 6.430942535400391,
      "epoch": 1.6345978918080153,
      "grad_norm": 1.0,
      "learning_rate": 0.00047404086265156835,
      "loss": 5.9425,
      "mean_token_accuracy": 0.16005455255508422,
      "num_tokens": 33039597.0,
      "step": 15275
    },
    {
      "entropy": 6.395409107208252,
      "epoch": 1.6351329659157794,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004740230315194935,
      "loss": 5.9533,
      "mean_token_accuracy": 0.15904719531536102,
      "num_tokens": 33050984.0,
      "step": 15280
    },
    {
      "entropy": 6.3798521041870115,
      "epoch": 1.6356680400235433,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047400519464065656,
      "loss": 6.0159,
      "mean_token_accuracy": 0.15770427584648133,
      "num_tokens": 33062250.0,
      "step": 15285
    },
    {
      "entropy": 6.4674512386322025,
      "epoch": 1.6362031141313071,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004739873520155726,
      "loss": 6.0552,
      "mean_token_accuracy": 0.14853655993938447,
      "num_tokens": 33072129.0,
      "step": 15290
    },
    {
      "entropy": 6.455130195617675,
      "epoch": 1.6367381882390712,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004739695036447568,
      "loss": 5.9125,
      "mean_token_accuracy": 0.15661536306142806,
      "num_tokens": 33082475.0,
      "step": 15295
    },
    {
      "entropy": 6.374242305755615,
      "epoch": 1.637273262346835,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047395164952872455,
      "loss": 5.9329,
      "mean_token_accuracy": 0.16938816159963607,
      "num_tokens": 33093791.0,
      "step": 15300
    },
    {
      "entropy": 6.419148683547974,
      "epoch": 1.637808336454599,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004739337896679914,
      "loss": 5.9305,
      "mean_token_accuracy": 0.16107006072998048,
      "num_tokens": 33104231.0,
      "step": 15305
    },
    {
      "entropy": 6.378033113479614,
      "epoch": 1.6383434105623629,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047391592406307296,
      "loss": 5.9383,
      "mean_token_accuracy": 0.15781633406877518,
      "num_tokens": 33115118.0,
      "step": 15310
    },
    {
      "entropy": 6.417173624038696,
      "epoch": 1.6388784846701268,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004738980527144852,
      "loss": 5.9175,
      "mean_token_accuracy": 0.16207022666931153,
      "num_tokens": 33126154.0,
      "step": 15315
    },
    {
      "entropy": 6.428819274902343,
      "epoch": 1.6394135587778909,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004738801756227441,
      "loss": 6.0055,
      "mean_token_accuracy": 0.15580477714538574,
      "num_tokens": 33137342.0,
      "step": 15320
    },
    {
      "entropy": 6.375725364685058,
      "epoch": 1.6399486328856545,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047386229278836595,
      "loss": 5.9603,
      "mean_token_accuracy": 0.16194958612322807,
      "num_tokens": 33147661.0,
      "step": 15325
    },
    {
      "entropy": 6.413193321228027,
      "epoch": 1.6404837069934186,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047384440421186695,
      "loss": 5.9969,
      "mean_token_accuracy": 0.15656197667121888,
      "num_tokens": 33157106.0,
      "step": 15330
    },
    {
      "entropy": 6.437152290344239,
      "epoch": 1.6410187811011825,
      "grad_norm": 1.0,
      "learning_rate": 0.0004738265098937638,
      "loss": 5.9824,
      "mean_token_accuracy": 0.15636766105890273,
      "num_tokens": 33169451.0,
      "step": 15335
    },
    {
      "entropy": 6.476279163360596,
      "epoch": 1.6415538552089464,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000473808609834573,
      "loss": 5.9894,
      "mean_token_accuracy": 0.15725223124027252,
      "num_tokens": 33180799.0,
      "step": 15340
    },
    {
      "entropy": 6.49586238861084,
      "epoch": 1.6420889293167105,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047379070403481165,
      "loss": 6.0559,
      "mean_token_accuracy": 0.15022989585995675,
      "num_tokens": 33191948.0,
      "step": 15345
    },
    {
      "entropy": 6.434716415405274,
      "epoch": 1.6426240034244743,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047377279249499656,
      "loss": 5.9736,
      "mean_token_accuracy": 0.15477521419525148,
      "num_tokens": 33203536.0,
      "step": 15350
    },
    {
      "entropy": 6.410928392410279,
      "epoch": 1.6431590775322382,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004737548752156451,
      "loss": 5.9659,
      "mean_token_accuracy": 0.15715570896863937,
      "num_tokens": 33214526.0,
      "step": 15355
    },
    {
      "entropy": 6.429980039596558,
      "epoch": 1.643694151640002,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004737369521972745,
      "loss": 5.9395,
      "mean_token_accuracy": 0.16168728321790696,
      "num_tokens": 33226560.0,
      "step": 15360
    },
    {
      "entropy": 6.383514499664306,
      "epoch": 1.644229225747766,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047371902344040234,
      "loss": 5.9515,
      "mean_token_accuracy": 0.15770779848098754,
      "num_tokens": 33237441.0,
      "step": 15365
    },
    {
      "entropy": 6.40056300163269,
      "epoch": 1.64476429985553,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047370108894554633,
      "loss": 5.9237,
      "mean_token_accuracy": 0.15634870678186416,
      "num_tokens": 33248229.0,
      "step": 15370
    },
    {
      "entropy": 6.408377695083618,
      "epoch": 1.645299373963294,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004736831487132243,
      "loss": 5.877,
      "mean_token_accuracy": 0.16904453486204146,
      "num_tokens": 33258992.0,
      "step": 15375
    },
    {
      "entropy": 6.3732161045074465,
      "epoch": 1.6458344480710578,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004736652027439542,
      "loss": 5.8787,
      "mean_token_accuracy": 0.16121681034564972,
      "num_tokens": 33269583.0,
      "step": 15380
    },
    {
      "entropy": 6.463597154617309,
      "epoch": 1.646369522178822,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047364725103825437,
      "loss": 6.0376,
      "mean_token_accuracy": 0.15236882269382476,
      "num_tokens": 33280236.0,
      "step": 15385
    },
    {
      "entropy": 6.38706865310669,
      "epoch": 1.6469045962865856,
      "grad_norm": 0.953125,
      "learning_rate": 0.000473629293596643,
      "loss": 5.9406,
      "mean_token_accuracy": 0.15549387782812119,
      "num_tokens": 33291327.0,
      "step": 15390
    },
    {
      "entropy": 6.433351039886475,
      "epoch": 1.6474396703943497,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047361133041963873,
      "loss": 6.0263,
      "mean_token_accuracy": 0.15215435922145842,
      "num_tokens": 33302125.0,
      "step": 15395
    },
    {
      "entropy": 6.439089107513428,
      "epoch": 1.6479747445021136,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004735933615077602,
      "loss": 5.9754,
      "mean_token_accuracy": 0.15693295747041702,
      "num_tokens": 33312576.0,
      "step": 15400
    },
    {
      "entropy": 6.446839141845703,
      "epoch": 1.6485098186098774,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047357538686152627,
      "loss": 5.9509,
      "mean_token_accuracy": 0.15447341799736022,
      "num_tokens": 33322821.0,
      "step": 15405
    },
    {
      "entropy": 6.464156579971314,
      "epoch": 1.6490448927176415,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047355740648145596,
      "loss": 5.9845,
      "mean_token_accuracy": 0.15157102644443513,
      "num_tokens": 33333951.0,
      "step": 15410
    },
    {
      "entropy": 6.432973003387451,
      "epoch": 1.6495799668254052,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004735394203680684,
      "loss": 5.9448,
      "mean_token_accuracy": 0.15865984112024306,
      "num_tokens": 33344739.0,
      "step": 15415
    },
    {
      "entropy": 6.4591991901397705,
      "epoch": 1.6501150409331693,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047352142852188304,
      "loss": 5.8987,
      "mean_token_accuracy": 0.15979873538017272,
      "num_tokens": 33354956.0,
      "step": 15420
    },
    {
      "entropy": 6.376287364959717,
      "epoch": 1.6506501150409332,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004735034309434193,
      "loss": 5.8836,
      "mean_token_accuracy": 0.16407324969768525,
      "num_tokens": 33364809.0,
      "step": 15425
    },
    {
      "entropy": 6.465846681594849,
      "epoch": 1.651185189148697,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004734854276331968,
      "loss": 6.002,
      "mean_token_accuracy": 0.1632540062069893,
      "num_tokens": 33375324.0,
      "step": 15430
    },
    {
      "entropy": 6.476783609390258,
      "epoch": 1.6517202632564612,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047346741859173564,
      "loss": 6.0655,
      "mean_token_accuracy": 0.1516342356801033,
      "num_tokens": 33385963.0,
      "step": 15435
    },
    {
      "entropy": 6.480970096588135,
      "epoch": 1.6522553373642248,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004734494038195555,
      "loss": 6.0392,
      "mean_token_accuracy": 0.15840765386819838,
      "num_tokens": 33395927.0,
      "step": 15440
    },
    {
      "entropy": 6.452879858016968,
      "epoch": 1.652790411471989,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004734313833171768,
      "loss": 5.9551,
      "mean_token_accuracy": 0.16107676327228546,
      "num_tokens": 33406294.0,
      "step": 15445
    },
    {
      "entropy": 6.437997817993164,
      "epoch": 1.6533254855797528,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004734133570851198,
      "loss": 5.9159,
      "mean_token_accuracy": 0.16598698049783706,
      "num_tokens": 33416995.0,
      "step": 15450
    },
    {
      "entropy": 6.324536466598511,
      "epoch": 1.6538605596875167,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004733953251239049,
      "loss": 5.8828,
      "mean_token_accuracy": 0.1558977484703064,
      "num_tokens": 33428087.0,
      "step": 15455
    },
    {
      "entropy": 6.445274877548218,
      "epoch": 1.6543956337952808,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004733772874340529,
      "loss": 6.0585,
      "mean_token_accuracy": 0.1508308008313179,
      "num_tokens": 33439151.0,
      "step": 15460
    },
    {
      "entropy": 6.4477485656738285,
      "epoch": 1.6549307079030444,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004733592440160846,
      "loss": 5.947,
      "mean_token_accuracy": 0.1630619913339615,
      "num_tokens": 33449257.0,
      "step": 15465
    },
    {
      "entropy": 6.393534469604492,
      "epoch": 1.6554657820108085,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047334119487052104,
      "loss": 5.9216,
      "mean_token_accuracy": 0.15876551866531372,
      "num_tokens": 33460472.0,
      "step": 15470
    },
    {
      "entropy": 6.438268184661865,
      "epoch": 1.6560008561185724,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047332313999788334,
      "loss": 5.9219,
      "mean_token_accuracy": 0.15083212852478028,
      "num_tokens": 33470193.0,
      "step": 15475
    },
    {
      "entropy": 6.326861429214477,
      "epoch": 1.6565359302263363,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047330507939869284,
      "loss": 5.8654,
      "mean_token_accuracy": 0.1626855455338955,
      "num_tokens": 33481098.0,
      "step": 15480
    },
    {
      "entropy": 6.450273942947388,
      "epoch": 1.6570710043341004,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000473287013073471,
      "loss": 5.9489,
      "mean_token_accuracy": 0.15904858559370041,
      "num_tokens": 33492514.0,
      "step": 15485
    },
    {
      "entropy": 6.430668783187866,
      "epoch": 1.6576060784418643,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004732689410227396,
      "loss": 5.9412,
      "mean_token_accuracy": 0.15924587845802307,
      "num_tokens": 33504254.0,
      "step": 15490
    },
    {
      "entropy": 6.452382516860962,
      "epoch": 1.6581411525496281,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004732508632470203,
      "loss": 5.9621,
      "mean_token_accuracy": 0.15595217198133468,
      "num_tokens": 33515101.0,
      "step": 15495
    },
    {
      "entropy": 6.436896753311157,
      "epoch": 1.658676226657392,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004732327797468352,
      "loss": 5.9686,
      "mean_token_accuracy": 0.16074855476617814,
      "num_tokens": 33525052.0,
      "step": 15500
    },
    {
      "entropy": 6.46158618927002,
      "epoch": 1.6592113007651559,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004732146905227065,
      "loss": 6.1132,
      "mean_token_accuracy": 0.14265801310539244,
      "num_tokens": 33536841.0,
      "step": 15505
    },
    {
      "entropy": 6.461019372940063,
      "epoch": 1.65974637487292,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004731965955751564,
      "loss": 5.9845,
      "mean_token_accuracy": 0.15497457534074782,
      "num_tokens": 33547292.0,
      "step": 15510
    },
    {
      "entropy": 6.446432065963745,
      "epoch": 1.6602814489806839,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004731784949047075,
      "loss": 5.9225,
      "mean_token_accuracy": 0.15660507529973983,
      "num_tokens": 33557674.0,
      "step": 15515
    },
    {
      "entropy": 6.438103723526001,
      "epoch": 1.6608165230884477,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004731603885118824,
      "loss": 5.9912,
      "mean_token_accuracy": 0.15227999687194824,
      "num_tokens": 33568396.0,
      "step": 15520
    },
    {
      "entropy": 6.426874351501465,
      "epoch": 1.6613515971962118,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004731422763972039,
      "loss": 6.0386,
      "mean_token_accuracy": 0.15762048587203026,
      "num_tokens": 33579688.0,
      "step": 15525
    },
    {
      "entropy": 6.460370588302612,
      "epoch": 1.6618866713039755,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047312415856119505,
      "loss": 6.0399,
      "mean_token_accuracy": 0.1559126392006874,
      "num_tokens": 33591550.0,
      "step": 15530
    },
    {
      "entropy": 6.4916259765625,
      "epoch": 1.6624217454117396,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047310603500437895,
      "loss": 5.988,
      "mean_token_accuracy": 0.16130850315093995,
      "num_tokens": 33600792.0,
      "step": 15535
    },
    {
      "entropy": 6.3687997341156,
      "epoch": 1.6629568195195035,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047308790572727895,
      "loss": 5.9249,
      "mean_token_accuracy": 0.15879255831241607,
      "num_tokens": 33611873.0,
      "step": 15540
    },
    {
      "entropy": 6.381060218811035,
      "epoch": 1.6634918936272673,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004730697707304184,
      "loss": 5.9681,
      "mean_token_accuracy": 0.15335929095745088,
      "num_tokens": 33622960.0,
      "step": 15545
    },
    {
      "entropy": 6.47402720451355,
      "epoch": 1.6640269677350314,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004730516300143211,
      "loss": 6.0412,
      "mean_token_accuracy": 0.1538678824901581,
      "num_tokens": 33635391.0,
      "step": 15550
    },
    {
      "entropy": 6.465884351730347,
      "epoch": 1.664562041842795,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004730334835795109,
      "loss": 5.9923,
      "mean_token_accuracy": 0.15214723646640776,
      "num_tokens": 33647942.0,
      "step": 15555
    },
    {
      "entropy": 6.477712678909302,
      "epoch": 1.6650971159505592,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004730153314265116,
      "loss": 5.9704,
      "mean_token_accuracy": 0.15489294081926347,
      "num_tokens": 33658425.0,
      "step": 15560
    },
    {
      "entropy": 6.385942029953003,
      "epoch": 1.665632190058323,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004729971735558475,
      "loss": 5.9251,
      "mean_token_accuracy": 0.16535238176584244,
      "num_tokens": 33669645.0,
      "step": 15565
    },
    {
      "entropy": 6.442303276062011,
      "epoch": 1.666167264166087,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004729790099680428,
      "loss": 6.0165,
      "mean_token_accuracy": 0.15558583810925483,
      "num_tokens": 33680415.0,
      "step": 15570
    },
    {
      "entropy": 6.407535552978516,
      "epoch": 1.666702338273851,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000472960840663622,
      "loss": 6.0353,
      "mean_token_accuracy": 0.15089727491140364,
      "num_tokens": 33691425.0,
      "step": 15575
    },
    {
      "entropy": 6.410505485534668,
      "epoch": 1.6672374123816147,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004729426656431097,
      "loss": 5.9712,
      "mean_token_accuracy": 0.16122428178787232,
      "num_tokens": 33703845.0,
      "step": 15580
    },
    {
      "entropy": 6.492136096954345,
      "epoch": 1.6677724864893788,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004729244849070308,
      "loss": 5.9532,
      "mean_token_accuracy": 0.15973744988441468,
      "num_tokens": 33714737.0,
      "step": 15585
    },
    {
      "entropy": 6.424393367767334,
      "epoch": 1.6683075605971427,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047290629845591026,
      "loss": 5.9076,
      "mean_token_accuracy": 0.15455199852585794,
      "num_tokens": 33725042.0,
      "step": 15590
    },
    {
      "entropy": 6.376716089248657,
      "epoch": 1.6688426347049066,
      "grad_norm": 1.03125,
      "learning_rate": 0.000472888106290273,
      "loss": 5.9504,
      "mean_token_accuracy": 0.1547821819782257,
      "num_tokens": 33736306.0,
      "step": 15595
    },
    {
      "entropy": 6.50233039855957,
      "epoch": 1.6693777088126707,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047286990841064467,
      "loss": 6.0131,
      "mean_token_accuracy": 0.14870304316282273,
      "num_tokens": 33747327.0,
      "step": 15600
    },
    {
      "entropy": 6.443639278411865,
      "epoch": 1.6699127829204343,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004728517048175504,
      "loss": 5.9564,
      "mean_token_accuracy": 0.15668172538280487,
      "num_tokens": 33757723.0,
      "step": 15605
    },
    {
      "entropy": 6.334149551391602,
      "epoch": 1.6704478570281984,
      "grad_norm": 0.953125,
      "learning_rate": 0.000472833495511516,
      "loss": 5.9148,
      "mean_token_accuracy": 0.16143670231103896,
      "num_tokens": 33769958.0,
      "step": 15610
    },
    {
      "entropy": 6.441114997863769,
      "epoch": 1.6709829311359623,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047281528049306727,
      "loss": 5.9483,
      "mean_token_accuracy": 0.15884028971195222,
      "num_tokens": 33780107.0,
      "step": 15615
    },
    {
      "entropy": 6.397679424285888,
      "epoch": 1.6715180052437262,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004727970597627301,
      "loss": 5.962,
      "mean_token_accuracy": 0.15535331517457962,
      "num_tokens": 33791023.0,
      "step": 15620
    },
    {
      "entropy": 6.360067653656006,
      "epoch": 1.6720530793514903,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004727788333210305,
      "loss": 5.9721,
      "mean_token_accuracy": 0.16189103275537492,
      "num_tokens": 33801195.0,
      "step": 15625
    },
    {
      "entropy": 6.434132289886475,
      "epoch": 1.6725881534592542,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000472760601168495,
      "loss": 5.9181,
      "mean_token_accuracy": 0.1669492170214653,
      "num_tokens": 33812957.0,
      "step": 15630
    },
    {
      "entropy": 6.443881511688232,
      "epoch": 1.673123227567018,
      "grad_norm": 1.0,
      "learning_rate": 0.00047274236330564995,
      "loss": 5.9934,
      "mean_token_accuracy": 0.1576312616467476,
      "num_tokens": 33823874.0,
      "step": 15635
    },
    {
      "entropy": 6.421902990341186,
      "epoch": 1.673658301674782,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004727241197330219,
      "loss": 5.9682,
      "mean_token_accuracy": 0.16214772760868074,
      "num_tokens": 33834284.0,
      "step": 15640
    },
    {
      "entropy": 6.451789426803589,
      "epoch": 1.6741933757825458,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004727058704511376,
      "loss": 5.9986,
      "mean_token_accuracy": 0.15734495371580123,
      "num_tokens": 33845544.0,
      "step": 15645
    },
    {
      "entropy": 6.4993171215057375,
      "epoch": 1.67472844989031,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004726876154605242,
      "loss": 5.967,
      "mean_token_accuracy": 0.16032592654228212,
      "num_tokens": 33855420.0,
      "step": 15650
    },
    {
      "entropy": 6.4031023502349855,
      "epoch": 1.6752635239980738,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004726693547617086,
      "loss": 5.8595,
      "mean_token_accuracy": 0.16592632234096527,
      "num_tokens": 33864904.0,
      "step": 15655
    },
    {
      "entropy": 6.3813042640686035,
      "epoch": 1.6757985981058376,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047265108835521825,
      "loss": 5.9248,
      "mean_token_accuracy": 0.1565372332930565,
      "num_tokens": 33875139.0,
      "step": 15660
    },
    {
      "entropy": 6.399698686599732,
      "epoch": 1.6763336722136017,
      "grad_norm": 1.0,
      "learning_rate": 0.0004726328162415804,
      "loss": 5.9317,
      "mean_token_accuracy": 0.16567895710468292,
      "num_tokens": 33886401.0,
      "step": 15665
    },
    {
      "entropy": 6.427884769439697,
      "epoch": 1.6768687463213654,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047261453842132277,
      "loss": 5.9821,
      "mean_token_accuracy": 0.16256581097841263,
      "num_tokens": 33896382.0,
      "step": 15670
    },
    {
      "entropy": 6.440084981918335,
      "epoch": 1.6774038204291295,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047259625489497317,
      "loss": 5.9621,
      "mean_token_accuracy": 0.16072721183300018,
      "num_tokens": 33906403.0,
      "step": 15675
    },
    {
      "entropy": 6.334707641601563,
      "epoch": 1.6779388945368934,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047257796566305946,
      "loss": 5.8522,
      "mean_token_accuracy": 0.15992193073034286,
      "num_tokens": 33916992.0,
      "step": 15680
    },
    {
      "entropy": 6.430480337142944,
      "epoch": 1.6784739686446573,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004725596707261097,
      "loss": 5.9678,
      "mean_token_accuracy": 0.15988713800907134,
      "num_tokens": 33927599.0,
      "step": 15685
    },
    {
      "entropy": 6.4616804122924805,
      "epoch": 1.6790090427524214,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004725413700846522,
      "loss": 6.0694,
      "mean_token_accuracy": 0.14797115325927734,
      "num_tokens": 33938957.0,
      "step": 15690
    },
    {
      "entropy": 6.458152818679809,
      "epoch": 1.679544116860185,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00047252306373921543,
      "loss": 5.961,
      "mean_token_accuracy": 0.16493951231241227,
      "num_tokens": 33950169.0,
      "step": 15695
    },
    {
      "entropy": 6.4391848087310795,
      "epoch": 1.680079190967949,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047250475169032804,
      "loss": 5.9664,
      "mean_token_accuracy": 0.15679520815610887,
      "num_tokens": 33960800.0,
      "step": 15700
    },
    {
      "entropy": 6.432637691497803,
      "epoch": 1.680614265075713,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004724864339385185,
      "loss": 6.0337,
      "mean_token_accuracy": 0.14718221127986908,
      "num_tokens": 33973060.0,
      "step": 15705
    },
    {
      "entropy": 6.468172407150268,
      "epoch": 1.6811493391834769,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004724681104843161,
      "loss": 6.0472,
      "mean_token_accuracy": 0.1544102042913437,
      "num_tokens": 33983185.0,
      "step": 15710
    },
    {
      "entropy": 6.461758852005005,
      "epoch": 1.681684413291241,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004724497813282497,
      "loss": 5.9817,
      "mean_token_accuracy": 0.1602817088365555,
      "num_tokens": 33993373.0,
      "step": 15715
    },
    {
      "entropy": 6.343892240524292,
      "epoch": 1.6822194873990046,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004724314464708486,
      "loss": 5.8578,
      "mean_token_accuracy": 0.17205298244953154,
      "num_tokens": 34005793.0,
      "step": 15720
    },
    {
      "entropy": 6.422130870819092,
      "epoch": 1.6827545615067687,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004724131059126423,
      "loss": 6.0373,
      "mean_token_accuracy": 0.14594856202602385,
      "num_tokens": 34016415.0,
      "step": 15725
    },
    {
      "entropy": 6.395816898345947,
      "epoch": 1.6832896356145326,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047239475965416016,
      "loss": 5.9128,
      "mean_token_accuracy": 0.1585589349269867,
      "num_tokens": 34026859.0,
      "step": 15730
    },
    {
      "entropy": 6.421891546249389,
      "epoch": 1.6838247097222965,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047237640769593224,
      "loss": 5.8066,
      "mean_token_accuracy": 0.17537156194448472,
      "num_tokens": 34037942.0,
      "step": 15735
    },
    {
      "entropy": 6.411176061630249,
      "epoch": 1.6843597838300606,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004723580500384882,
      "loss": 5.9045,
      "mean_token_accuracy": 0.16626847237348558,
      "num_tokens": 34048659.0,
      "step": 15740
    },
    {
      "entropy": 6.380535221099853,
      "epoch": 1.6848948579378242,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047233968668235817,
      "loss": 5.9316,
      "mean_token_accuracy": 0.1588290050625801,
      "num_tokens": 34059961.0,
      "step": 15745
    },
    {
      "entropy": 6.410470628738404,
      "epoch": 1.6854299320455883,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047232131762807244,
      "loss": 5.9681,
      "mean_token_accuracy": 0.1631573975086212,
      "num_tokens": 34069978.0,
      "step": 15750
    },
    {
      "entropy": 6.4301896572113035,
      "epoch": 1.6859650061533522,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004723029428761614,
      "loss": 5.971,
      "mean_token_accuracy": 0.15752568542957307,
      "num_tokens": 34080301.0,
      "step": 15755
    },
    {
      "entropy": 6.405263614654541,
      "epoch": 1.686500080261116,
      "grad_norm": 1.375,
      "learning_rate": 0.00047228456242715564,
      "loss": 5.9651,
      "mean_token_accuracy": 0.1617084637284279,
      "num_tokens": 34092376.0,
      "step": 15760
    },
    {
      "entropy": 6.437398433685303,
      "epoch": 1.6870351543688802,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004722661762815858,
      "loss": 5.9429,
      "mean_token_accuracy": 0.16088494211435317,
      "num_tokens": 34102761.0,
      "step": 15765
    },
    {
      "entropy": 6.4151012897491455,
      "epoch": 1.687570228476644,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004722477844399829,
      "loss": 5.9527,
      "mean_token_accuracy": 0.1504283793270588,
      "num_tokens": 34113370.0,
      "step": 15770
    },
    {
      "entropy": 6.437972927093506,
      "epoch": 1.688105302584408,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004722293869028779,
      "loss": 5.9544,
      "mean_token_accuracy": 0.16395780444145203,
      "num_tokens": 34125038.0,
      "step": 15775
    },
    {
      "entropy": 6.483677053451538,
      "epoch": 1.6886403766921718,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004722109836708021,
      "loss": 6.0005,
      "mean_token_accuracy": 0.15217196196317673,
      "num_tokens": 34135595.0,
      "step": 15780
    },
    {
      "entropy": 6.455944538116455,
      "epoch": 1.6891754507999357,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047219257474428685,
      "loss": 5.9831,
      "mean_token_accuracy": 0.15272802412509917,
      "num_tokens": 34146755.0,
      "step": 15785
    },
    {
      "entropy": 6.3476344585418705,
      "epoch": 1.6897105249076998,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047217416012386367,
      "loss": 5.828,
      "mean_token_accuracy": 0.17567530572414397,
      "num_tokens": 34158465.0,
      "step": 15790
    },
    {
      "entropy": 6.3985601425170895,
      "epoch": 1.6902455990154637,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004721557398100644,
      "loss": 5.9295,
      "mean_token_accuracy": 0.15895601809024812,
      "num_tokens": 34170007.0,
      "step": 15795
    },
    {
      "entropy": 6.470225524902344,
      "epoch": 1.6907806731232276,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047213731380342077,
      "loss": 6.0111,
      "mean_token_accuracy": 0.15572379827499389,
      "num_tokens": 34181595.0,
      "step": 15800
    },
    {
      "entropy": 6.43712854385376,
      "epoch": 1.6913157472309917,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000472118882104465,
      "loss": 5.9719,
      "mean_token_accuracy": 0.15918021351099015,
      "num_tokens": 34191791.0,
      "step": 15805
    },
    {
      "entropy": 6.446450424194336,
      "epoch": 1.6918508213387553,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047210044471372915,
      "loss": 5.8889,
      "mean_token_accuracy": 0.16053694039583205,
      "num_tokens": 34202636.0,
      "step": 15810
    },
    {
      "entropy": 6.500013399124145,
      "epoch": 1.6923858954465194,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004720820016317456,
      "loss": 6.0286,
      "mean_token_accuracy": 0.15069264322519302,
      "num_tokens": 34213333.0,
      "step": 15815
    },
    {
      "entropy": 6.45832052230835,
      "epoch": 1.6929209695542833,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000472063552859047,
      "loss": 5.9635,
      "mean_token_accuracy": 0.15968910306692125,
      "num_tokens": 34223679.0,
      "step": 15820
    },
    {
      "entropy": 6.452071332931519,
      "epoch": 1.6934560436620472,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047204509839616604,
      "loss": 6.0878,
      "mean_token_accuracy": 0.14816973730921745,
      "num_tokens": 34234528.0,
      "step": 15825
    },
    {
      "entropy": 6.437414646148682,
      "epoch": 1.6939911177698113,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004720266382436355,
      "loss": 5.9131,
      "mean_token_accuracy": 0.160316102206707,
      "num_tokens": 34246203.0,
      "step": 15830
    },
    {
      "entropy": 6.450861263275146,
      "epoch": 1.694526191877575,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047200817240198846,
      "loss": 5.9223,
      "mean_token_accuracy": 0.15651085376739501,
      "num_tokens": 34257231.0,
      "step": 15835
    },
    {
      "entropy": 6.4111381530761715,
      "epoch": 1.695061265985339,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047198970087175817,
      "loss": 5.9331,
      "mean_token_accuracy": 0.1610191211104393,
      "num_tokens": 34268900.0,
      "step": 15840
    },
    {
      "entropy": 6.44896674156189,
      "epoch": 1.695596340093103,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004719712236534779,
      "loss": 6.0212,
      "mean_token_accuracy": 0.15441965013742448,
      "num_tokens": 34279343.0,
      "step": 15845
    },
    {
      "entropy": 6.435469388961792,
      "epoch": 1.6961314142008668,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004719527407476812,
      "loss": 5.9526,
      "mean_token_accuracy": 0.16289241760969161,
      "num_tokens": 34291952.0,
      "step": 15850
    },
    {
      "entropy": 6.441124439239502,
      "epoch": 1.6966664883086309,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047193425215490186,
      "loss": 5.9734,
      "mean_token_accuracy": 0.14896577149629592,
      "num_tokens": 34302516.0,
      "step": 15855
    },
    {
      "entropy": 6.435949373245239,
      "epoch": 1.6972015624163945,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047191575787567373,
      "loss": 5.8997,
      "mean_token_accuracy": 0.16195925176143647,
      "num_tokens": 34313094.0,
      "step": 15860
    },
    {
      "entropy": 6.482620334625244,
      "epoch": 1.6977366365241586,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004718972579105306,
      "loss": 6.0282,
      "mean_token_accuracy": 0.15430918186903,
      "num_tokens": 34324803.0,
      "step": 15865
    },
    {
      "entropy": 6.44781174659729,
      "epoch": 1.6982717106319225,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004718787522600069,
      "loss": 5.9774,
      "mean_token_accuracy": 0.15356167256832123,
      "num_tokens": 34336386.0,
      "step": 15870
    },
    {
      "entropy": 6.4255204677581785,
      "epoch": 1.6988067847396864,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004718602409246369,
      "loss": 5.9969,
      "mean_token_accuracy": 0.15500651597976683,
      "num_tokens": 34347085.0,
      "step": 15875
    },
    {
      "entropy": 6.413449478149414,
      "epoch": 1.6993418588474505,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047184172390495505,
      "loss": 5.9316,
      "mean_token_accuracy": 0.16569200307130813,
      "num_tokens": 34358146.0,
      "step": 15880
    },
    {
      "entropy": 6.433770418167114,
      "epoch": 1.6998769329552141,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004718232012014961,
      "loss": 5.9639,
      "mean_token_accuracy": 0.15669197142124175,
      "num_tokens": 34370403.0,
      "step": 15885
    },
    {
      "entropy": 6.446292972564697,
      "epoch": 1.7004120070629782,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047180467281479496,
      "loss": 5.9647,
      "mean_token_accuracy": 0.1609950289130211,
      "num_tokens": 34381130.0,
      "step": 15890
    },
    {
      "entropy": 6.4022864818573,
      "epoch": 1.7009470811707421,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004717861387453865,
      "loss": 5.9906,
      "mean_token_accuracy": 0.15622309297323228,
      "num_tokens": 34391610.0,
      "step": 15895
    },
    {
      "entropy": 6.460147523880005,
      "epoch": 1.701482155278506,
      "grad_norm": 1.0,
      "learning_rate": 0.0004717675989938059,
      "loss": 5.9729,
      "mean_token_accuracy": 0.15942391604185105,
      "num_tokens": 34402262.0,
      "step": 15900
    },
    {
      "entropy": 6.441456079483032,
      "epoch": 1.70201722938627,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047174905356058857,
      "loss": 5.9651,
      "mean_token_accuracy": 0.15343908220529556,
      "num_tokens": 34412707.0,
      "step": 15905
    },
    {
      "entropy": 6.369121408462524,
      "epoch": 1.702552303494034,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047173050244626993,
      "loss": 5.9047,
      "mean_token_accuracy": 0.1619341641664505,
      "num_tokens": 34423203.0,
      "step": 15910
    },
    {
      "entropy": 6.410768985748291,
      "epoch": 1.7030873776017978,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004717119456513857,
      "loss": 5.9919,
      "mean_token_accuracy": 0.15724618136882781,
      "num_tokens": 34433279.0,
      "step": 15915
    },
    {
      "entropy": 6.441387081146241,
      "epoch": 1.7036224517095617,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047169338317647164,
      "loss": 6.0238,
      "mean_token_accuracy": 0.15807254910469054,
      "num_tokens": 34444601.0,
      "step": 15920
    },
    {
      "entropy": 6.473701667785645,
      "epoch": 1.7041575258173256,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047167481502206383,
      "loss": 5.9822,
      "mean_token_accuracy": 0.15675383657217026,
      "num_tokens": 34455103.0,
      "step": 15925
    },
    {
      "entropy": 6.446041822433472,
      "epoch": 1.7046925999250897,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047165624118869834,
      "loss": 6.0027,
      "mean_token_accuracy": 0.15278667211532593,
      "num_tokens": 34465835.0,
      "step": 15930
    },
    {
      "entropy": 6.4933130741119385,
      "epoch": 1.7052276740328536,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004716376616769115,
      "loss": 6.0335,
      "mean_token_accuracy": 0.15085156708955766,
      "num_tokens": 34476602.0,
      "step": 15935
    },
    {
      "entropy": 6.466088819503784,
      "epoch": 1.7057627481406175,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004716190764872399,
      "loss": 5.9219,
      "mean_token_accuracy": 0.15774450525641442,
      "num_tokens": 34488026.0,
      "step": 15940
    },
    {
      "entropy": 6.436269617080688,
      "epoch": 1.7062978222483816,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047160048562022006,
      "loss": 6.0105,
      "mean_token_accuracy": 0.16350340098142624,
      "num_tokens": 34497837.0,
      "step": 15945
    },
    {
      "entropy": 6.380255842208863,
      "epoch": 1.7068328963561452,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004715818890763888,
      "loss": 5.9482,
      "mean_token_accuracy": 0.161816044151783,
      "num_tokens": 34507727.0,
      "step": 15950
    },
    {
      "entropy": 6.4403749942779545,
      "epoch": 1.7073679704639093,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047156328685628307,
      "loss": 5.9649,
      "mean_token_accuracy": 0.15586953163146972,
      "num_tokens": 34518595.0,
      "step": 15955
    },
    {
      "entropy": 6.406744432449341,
      "epoch": 1.7079030445716732,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004715446789604401,
      "loss": 5.8927,
      "mean_token_accuracy": 0.16284205615520478,
      "num_tokens": 34529415.0,
      "step": 15960
    },
    {
      "entropy": 6.419968366622925,
      "epoch": 1.708438118679437,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004715260653893972,
      "loss": 5.9396,
      "mean_token_accuracy": 0.16475329101085662,
      "num_tokens": 34539593.0,
      "step": 15965
    },
    {
      "entropy": 6.447049188613891,
      "epoch": 1.7089731927872012,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004715074461436917,
      "loss": 5.9365,
      "mean_token_accuracy": 0.1539831943809986,
      "num_tokens": 34549441.0,
      "step": 15970
    },
    {
      "entropy": 6.433720636367798,
      "epoch": 1.7095082668949648,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047148882122386126,
      "loss": 5.8963,
      "mean_token_accuracy": 0.1639431819319725,
      "num_tokens": 34560211.0,
      "step": 15975
    },
    {
      "entropy": 6.41886944770813,
      "epoch": 1.710043341002729,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047147019063044387,
      "loss": 5.9952,
      "mean_token_accuracy": 0.15636261701583862,
      "num_tokens": 34570479.0,
      "step": 15980
    },
    {
      "entropy": 6.446100807189941,
      "epoch": 1.7105784151104928,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047145155436397726,
      "loss": 5.9907,
      "mean_token_accuracy": 0.15785836577415466,
      "num_tokens": 34581382.0,
      "step": 15985
    },
    {
      "entropy": 6.435035562515258,
      "epoch": 1.7111134892182567,
      "grad_norm": 1.0,
      "learning_rate": 0.0004714329124249996,
      "loss": 5.9119,
      "mean_token_accuracy": 0.1645201861858368,
      "num_tokens": 34592131.0,
      "step": 15990
    },
    {
      "entropy": 6.409049701690674,
      "epoch": 1.7116485633260208,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004714142648140492,
      "loss": 5.994,
      "mean_token_accuracy": 0.15563116520643233,
      "num_tokens": 34604407.0,
      "step": 15995
    },
    {
      "entropy": 6.378651332855225,
      "epoch": 1.7121836374337844,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004713956115316645,
      "loss": 5.8863,
      "mean_token_accuracy": 0.1574580505490303,
      "num_tokens": 34614842.0,
      "step": 16000
    },
    {
      "entropy": 6.330024242401123,
      "epoch": 1.7127187115415485,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004713769525783841,
      "loss": 5.839,
      "mean_token_accuracy": 0.16385052502155303,
      "num_tokens": 34625991.0,
      "step": 16005
    },
    {
      "entropy": 6.404508018493653,
      "epoch": 1.7132537856493124,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004713582879547469,
      "loss": 6.0018,
      "mean_token_accuracy": 0.1588558793067932,
      "num_tokens": 34637317.0,
      "step": 16010
    },
    {
      "entropy": 6.468359565734863,
      "epoch": 1.7137888597570763,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004713396176612916,
      "loss": 5.9964,
      "mean_token_accuracy": 0.1544583722949028,
      "num_tokens": 34648282.0,
      "step": 16015
    },
    {
      "entropy": 6.43749771118164,
      "epoch": 1.7143239338648404,
      "grad_norm": 1.0,
      "learning_rate": 0.0004713209416985575,
      "loss": 5.9513,
      "mean_token_accuracy": 0.15715266317129134,
      "num_tokens": 34658638.0,
      "step": 16020
    },
    {
      "entropy": 6.409981203079224,
      "epoch": 1.714859007972604,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047130226006708375,
      "loss": 5.9892,
      "mean_token_accuracy": 0.15603242367506026,
      "num_tokens": 34669308.0,
      "step": 16025
    },
    {
      "entropy": 6.381965780258179,
      "epoch": 1.7153940820803681,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047128357276740986,
      "loss": 5.9522,
      "mean_token_accuracy": 0.155669729411602,
      "num_tokens": 34679513.0,
      "step": 16030
    },
    {
      "entropy": 6.431146717071533,
      "epoch": 1.715929156188132,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004712648798000754,
      "loss": 5.9662,
      "mean_token_accuracy": 0.15777787566184998,
      "num_tokens": 34690736.0,
      "step": 16035
    },
    {
      "entropy": 6.422944593429565,
      "epoch": 1.716464230295896,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047124618116562003,
      "loss": 5.9445,
      "mean_token_accuracy": 0.1538891613483429,
      "num_tokens": 34701272.0,
      "step": 16040
    },
    {
      "entropy": 6.36087474822998,
      "epoch": 1.71699930440366,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004712274768645839,
      "loss": 5.8551,
      "mean_token_accuracy": 0.16429322063922883,
      "num_tokens": 34712486.0,
      "step": 16045
    },
    {
      "entropy": 6.480742883682251,
      "epoch": 1.7175343785114239,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047120876689750685,
      "loss": 5.9498,
      "mean_token_accuracy": 0.1573661133646965,
      "num_tokens": 34721979.0,
      "step": 16050
    },
    {
      "entropy": 6.397354650497436,
      "epoch": 1.7180694526191878,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004711900512649292,
      "loss": 5.9325,
      "mean_token_accuracy": 0.16165987253189087,
      "num_tokens": 34733496.0,
      "step": 16055
    },
    {
      "entropy": 6.431666278839112,
      "epoch": 1.7186045267269516,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004711713299673914,
      "loss": 5.9487,
      "mean_token_accuracy": 0.16020318120718002,
      "num_tokens": 34744223.0,
      "step": 16060
    },
    {
      "entropy": 6.421742296218872,
      "epoch": 1.7191396008347155,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047115260300543407,
      "loss": 5.9906,
      "mean_token_accuracy": 0.15847385227680205,
      "num_tokens": 34754931.0,
      "step": 16065
    },
    {
      "entropy": 6.370933294296265,
      "epoch": 1.7196746749424796,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004711338703795978,
      "loss": 5.8488,
      "mean_token_accuracy": 0.1637336105108261,
      "num_tokens": 34766285.0,
      "step": 16070
    },
    {
      "entropy": 6.409761810302735,
      "epoch": 1.7202097490502435,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004711151320904237,
      "loss": 5.9552,
      "mean_token_accuracy": 0.1639688044786453,
      "num_tokens": 34776979.0,
      "step": 16075
    },
    {
      "entropy": 6.490874195098877,
      "epoch": 1.7207448231580074,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004710963881384527,
      "loss": 6.0362,
      "mean_token_accuracy": 0.15185218751430513,
      "num_tokens": 34787466.0,
      "step": 16080
    },
    {
      "entropy": 6.5128826141357425,
      "epoch": 1.7212798972657715,
      "grad_norm": 1.046875,
      "learning_rate": 0.000471077638524226,
      "loss": 5.9394,
      "mean_token_accuracy": 0.16182179152965545,
      "num_tokens": 34797664.0,
      "step": 16085
    },
    {
      "entropy": 6.368173599243164,
      "epoch": 1.7218149713735351,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047105888324828506,
      "loss": 5.9257,
      "mean_token_accuracy": 0.17095199525356292,
      "num_tokens": 34809082.0,
      "step": 16090
    },
    {
      "entropy": 6.388035774230957,
      "epoch": 1.7223500454812992,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047104012231117143,
      "loss": 5.947,
      "mean_token_accuracy": 0.1561230719089508,
      "num_tokens": 34820223.0,
      "step": 16095
    },
    {
      "entropy": 6.336936521530151,
      "epoch": 1.722885119589063,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047102135571342683,
      "loss": 5.8626,
      "mean_token_accuracy": 0.15993381291627884,
      "num_tokens": 34830428.0,
      "step": 16100
    },
    {
      "entropy": 6.3960143566131595,
      "epoch": 1.723420193696827,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047100258345559307,
      "loss": 5.9773,
      "mean_token_accuracy": 0.1547223523259163,
      "num_tokens": 34842787.0,
      "step": 16105
    },
    {
      "entropy": 6.4165990352630615,
      "epoch": 1.723955267804591,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004709838055382123,
      "loss": 5.9253,
      "mean_token_accuracy": 0.15452154576778412,
      "num_tokens": 34854323.0,
      "step": 16110
    },
    {
      "entropy": 6.437535142898559,
      "epoch": 1.7244903419123547,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004709650219618266,
      "loss": 5.9044,
      "mean_token_accuracy": 0.16143249869346618,
      "num_tokens": 34865279.0,
      "step": 16115
    },
    {
      "entropy": 6.4142493724823,
      "epoch": 1.7250254160201188,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004709462327269785,
      "loss": 5.9269,
      "mean_token_accuracy": 0.15828490108251572,
      "num_tokens": 34875951.0,
      "step": 16120
    },
    {
      "entropy": 6.346537256240845,
      "epoch": 1.7255604901278827,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004709274378342105,
      "loss": 5.9673,
      "mean_token_accuracy": 0.15984275192022324,
      "num_tokens": 34885815.0,
      "step": 16125
    },
    {
      "entropy": 6.418897485733032,
      "epoch": 1.7260955642356466,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047090863728406517,
      "loss": 5.9838,
      "mean_token_accuracy": 0.15564633160829544,
      "num_tokens": 34897158.0,
      "step": 16130
    },
    {
      "entropy": 6.486556768417358,
      "epoch": 1.7266306383434107,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004708898310770856,
      "loss": 5.9952,
      "mean_token_accuracy": 0.15636667013168334,
      "num_tokens": 34907815.0,
      "step": 16135
    },
    {
      "entropy": 6.412474632263184,
      "epoch": 1.7271657124511743,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004708710192138146,
      "loss": 5.9306,
      "mean_token_accuracy": 0.16338492184877396,
      "num_tokens": 34918820.0,
      "step": 16140
    },
    {
      "entropy": 6.47990984916687,
      "epoch": 1.7277007865589384,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004708522016947954,
      "loss": 6.0056,
      "mean_token_accuracy": 0.1570424482226372,
      "num_tokens": 34929998.0,
      "step": 16145
    },
    {
      "entropy": 6.41492338180542,
      "epoch": 1.7282358606667023,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004708333785205715,
      "loss": 5.8989,
      "mean_token_accuracy": 0.17244210243225097,
      "num_tokens": 34940989.0,
      "step": 16150
    },
    {
      "entropy": 6.456949996948242,
      "epoch": 1.7287709347744662,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004708145496916862,
      "loss": 5.9365,
      "mean_token_accuracy": 0.15786129534244536,
      "num_tokens": 34952433.0,
      "step": 16155
    },
    {
      "entropy": 6.45723443031311,
      "epoch": 1.7293060088822303,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047079571520868344,
      "loss": 5.996,
      "mean_token_accuracy": 0.15286411046981813,
      "num_tokens": 34963074.0,
      "step": 16160
    },
    {
      "entropy": 6.382172155380249,
      "epoch": 1.729841082989994,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047077687507210674,
      "loss": 5.9243,
      "mean_token_accuracy": 0.1649669200181961,
      "num_tokens": 34974315.0,
      "step": 16165
    },
    {
      "entropy": 6.374082040786743,
      "epoch": 1.730376157097758,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004707580292825004,
      "loss": 5.8681,
      "mean_token_accuracy": 0.17248545736074447,
      "num_tokens": 34985059.0,
      "step": 16170
    },
    {
      "entropy": 6.464826154708862,
      "epoch": 1.730911231205522,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004707391778404085,
      "loss": 6.0503,
      "mean_token_accuracy": 0.1486973397433758,
      "num_tokens": 34996156.0,
      "step": 16175
    },
    {
      "entropy": 6.4432188987731935,
      "epoch": 1.7314463053132858,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004707203207463752,
      "loss": 5.9527,
      "mean_token_accuracy": 0.15894730538129806,
      "num_tokens": 35007410.0,
      "step": 16180
    },
    {
      "entropy": 6.442901659011841,
      "epoch": 1.73198137942105,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004707014580009453,
      "loss": 5.9288,
      "mean_token_accuracy": 0.16117143929004668,
      "num_tokens": 35018492.0,
      "step": 16185
    },
    {
      "entropy": 6.443019151687622,
      "epoch": 1.7325164535288138,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004706825896046632,
      "loss": 5.988,
      "mean_token_accuracy": 0.15979912877082825,
      "num_tokens": 35029923.0,
      "step": 16190
    },
    {
      "entropy": 6.400047779083252,
      "epoch": 1.7330515276365777,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047066371555807386,
      "loss": 5.9245,
      "mean_token_accuracy": 0.15788981467485427,
      "num_tokens": 35040370.0,
      "step": 16195
    },
    {
      "entropy": 6.452683448791504,
      "epoch": 1.7335866017443415,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004706448358617222,
      "loss": 5.9836,
      "mean_token_accuracy": 0.1532462626695633,
      "num_tokens": 35051414.0,
      "step": 16200
    },
    {
      "entropy": 6.477778625488281,
      "epoch": 1.7341216758521054,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047062595051615343,
      "loss": 5.9165,
      "mean_token_accuracy": 0.16308738440275192,
      "num_tokens": 35061015.0,
      "step": 16205
    },
    {
      "entropy": 6.424928426742554,
      "epoch": 1.7346567499598695,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004706070595219128,
      "loss": 6.0203,
      "mean_token_accuracy": 0.15545460656285287,
      "num_tokens": 35071757.0,
      "step": 16210
    },
    {
      "entropy": 6.458475637435913,
      "epoch": 1.7351918240676334,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047058816287954587,
      "loss": 5.997,
      "mean_token_accuracy": 0.15527721345424653,
      "num_tokens": 35082912.0,
      "step": 16215
    },
    {
      "entropy": 6.453078985214233,
      "epoch": 1.7357268981753973,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047056926058959813,
      "loss": 5.9215,
      "mean_token_accuracy": 0.16881765872240068,
      "num_tokens": 35094103.0,
      "step": 16220
    },
    {
      "entropy": 6.397449350357055,
      "epoch": 1.7362619722831614,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004705503526526155,
      "loss": 5.8706,
      "mean_token_accuracy": 0.167387056350708,
      "num_tokens": 35105960.0,
      "step": 16225
    },
    {
      "entropy": 6.327276086807251,
      "epoch": 1.736797046390925,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047053143906914405,
      "loss": 5.8899,
      "mean_token_accuracy": 0.16938157379627228,
      "num_tokens": 35116482.0,
      "step": 16230
    },
    {
      "entropy": 6.425247287750244,
      "epoch": 1.7373321204986891,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047051251983972966,
      "loss": 5.9733,
      "mean_token_accuracy": 0.1594018116593361,
      "num_tokens": 35127976.0,
      "step": 16235
    },
    {
      "entropy": 6.456327152252197,
      "epoch": 1.737867194606453,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004704935949649187,
      "loss": 5.9903,
      "mean_token_accuracy": 0.15916103720664979,
      "num_tokens": 35137712.0,
      "step": 16240
    },
    {
      "entropy": 6.388078403472901,
      "epoch": 1.7384022687142169,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004704746644452577,
      "loss": 5.8978,
      "mean_token_accuracy": 0.16431068629026413,
      "num_tokens": 35148524.0,
      "step": 16245
    },
    {
      "entropy": 6.440047025680542,
      "epoch": 1.738937342821981,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047045572828129323,
      "loss": 5.969,
      "mean_token_accuracy": 0.1584414303302765,
      "num_tokens": 35159080.0,
      "step": 16250
    },
    {
      "entropy": 6.316601037979126,
      "epoch": 1.7394724169297446,
      "grad_norm": 1.046875,
      "learning_rate": 0.000470436786473572,
      "loss": 5.876,
      "mean_token_accuracy": 0.16394800543785096,
      "num_tokens": 35169993.0,
      "step": 16255
    },
    {
      "entropy": 6.386191368103027,
      "epoch": 1.7400074910375087,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004704178390226411,
      "loss": 5.8982,
      "mean_token_accuracy": 0.16015563607215882,
      "num_tokens": 35181413.0,
      "step": 16260
    },
    {
      "entropy": 6.403861618041992,
      "epoch": 1.7405425651452726,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047039888592904753,
      "loss": 5.9139,
      "mean_token_accuracy": 0.16403191089630126,
      "num_tokens": 35191600.0,
      "step": 16265
    },
    {
      "entropy": 6.446494388580322,
      "epoch": 1.7410776392530365,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047037992719333864,
      "loss": 5.9236,
      "mean_token_accuracy": 0.15600271970033647,
      "num_tokens": 35202535.0,
      "step": 16270
    },
    {
      "entropy": 6.405620861053467,
      "epoch": 1.7416127133608006,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047036096281606176,
      "loss": 5.9205,
      "mean_token_accuracy": 0.15477754175662994,
      "num_tokens": 35214798.0,
      "step": 16275
    },
    {
      "entropy": 6.372258996963501,
      "epoch": 1.7421477874685642,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047034199279776456,
      "loss": 5.904,
      "mean_token_accuracy": 0.16286729723215104,
      "num_tokens": 35224870.0,
      "step": 16280
    },
    {
      "entropy": 6.386773538589478,
      "epoch": 1.7426828615763283,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004703230171389948,
      "loss": 5.9237,
      "mean_token_accuracy": 0.1544351652264595,
      "num_tokens": 35235408.0,
      "step": 16285
    },
    {
      "entropy": 6.3953063011169435,
      "epoch": 1.7432179356840922,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047030403584030024,
      "loss": 5.9399,
      "mean_token_accuracy": 0.16197500973939896,
      "num_tokens": 35246310.0,
      "step": 16290
    },
    {
      "entropy": 6.419058418273925,
      "epoch": 1.743753009791856,
      "grad_norm": 1.0,
      "learning_rate": 0.0004702850489022292,
      "loss": 5.9527,
      "mean_token_accuracy": 0.15810076296329498,
      "num_tokens": 35257027.0,
      "step": 16295
    },
    {
      "entropy": 6.465060997009277,
      "epoch": 1.7442880838996202,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047026605632532976,
      "loss": 5.9206,
      "mean_token_accuracy": 0.16240539997816086,
      "num_tokens": 35267397.0,
      "step": 16300
    },
    {
      "entropy": 6.3458967208862305,
      "epoch": 1.7448231580073839,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047024705811015045,
      "loss": 5.8725,
      "mean_token_accuracy": 0.16006311923265457,
      "num_tokens": 35278775.0,
      "step": 16305
    },
    {
      "entropy": 6.370694780349732,
      "epoch": 1.745358232115148,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004702280542572397,
      "loss": 5.828,
      "mean_token_accuracy": 0.16718346029520034,
      "num_tokens": 35288818.0,
      "step": 16310
    },
    {
      "entropy": 6.448410940170288,
      "epoch": 1.7458933062229118,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004702090447671464,
      "loss": 6.0299,
      "mean_token_accuracy": 0.14842126220464708,
      "num_tokens": 35299385.0,
      "step": 16315
    },
    {
      "entropy": 6.426753234863281,
      "epoch": 1.7464283803306757,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004701900296404193,
      "loss": 5.9037,
      "mean_token_accuracy": 0.16668517291545867,
      "num_tokens": 35310626.0,
      "step": 16320
    },
    {
      "entropy": 6.452135801315308,
      "epoch": 1.7469634544384398,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004701710088776075,
      "loss": 5.9784,
      "mean_token_accuracy": 0.1582433059811592,
      "num_tokens": 35321615.0,
      "step": 16325
    },
    {
      "entropy": 6.47968487739563,
      "epoch": 1.7474985285462037,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047015198247926025,
      "loss": 6.0024,
      "mean_token_accuracy": 0.15670483708381652,
      "num_tokens": 35332012.0,
      "step": 16330
    },
    {
      "entropy": 6.4822084426879885,
      "epoch": 1.7480336026539676,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000470132950445927,
      "loss": 6.0201,
      "mean_token_accuracy": 0.14689060151576996,
      "num_tokens": 35342153.0,
      "step": 16335
    },
    {
      "entropy": 6.464757061004638,
      "epoch": 1.7485686767617317,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004701139127781571,
      "loss": 5.9536,
      "mean_token_accuracy": 0.1580946996808052,
      "num_tokens": 35352858.0,
      "step": 16340
    },
    {
      "entropy": 6.414715719223023,
      "epoch": 1.7491037508694953,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004700948694765005,
      "loss": 5.879,
      "mean_token_accuracy": 0.16521502137184144,
      "num_tokens": 35363489.0,
      "step": 16345
    },
    {
      "entropy": 6.459300518035889,
      "epoch": 1.7496388249772594,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004700758205415069,
      "loss": 6.0404,
      "mean_token_accuracy": 0.15430581271648408,
      "num_tokens": 35375816.0,
      "step": 16350
    },
    {
      "entropy": 6.402688884735108,
      "epoch": 1.7501738990850233,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047005676597372645,
      "loss": 5.9545,
      "mean_token_accuracy": 0.15438281893730163,
      "num_tokens": 35386708.0,
      "step": 16355
    },
    {
      "entropy": 6.443779039382934,
      "epoch": 1.7507089731927872,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047003770577370925,
      "loss": 5.9978,
      "mean_token_accuracy": 0.1553271472454071,
      "num_tokens": 35397734.0,
      "step": 16360
    },
    {
      "entropy": 6.529666328430176,
      "epoch": 1.7512440473005513,
      "grad_norm": 1.0,
      "learning_rate": 0.00047001863994200574,
      "loss": 5.9354,
      "mean_token_accuracy": 0.16243588477373122,
      "num_tokens": 35408870.0,
      "step": 16365
    },
    {
      "entropy": 6.3906721591949465,
      "epoch": 1.751779121408315,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004699995684791664,
      "loss": 5.8962,
      "mean_token_accuracy": 0.16072052270174025,
      "num_tokens": 35419498.0,
      "step": 16370
    },
    {
      "entropy": 6.352797031402588,
      "epoch": 1.752314195516079,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046998049138574177,
      "loss": 5.9234,
      "mean_token_accuracy": 0.16501273959875107,
      "num_tokens": 35430690.0,
      "step": 16375
    },
    {
      "entropy": 6.3840748310089115,
      "epoch": 1.752849269623843,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000469961408662283,
      "loss": 5.8836,
      "mean_token_accuracy": 0.16383645981550216,
      "num_tokens": 35441007.0,
      "step": 16380
    },
    {
      "entropy": 6.389566278457641,
      "epoch": 1.7533843437316068,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000469942320309341,
      "loss": 5.9251,
      "mean_token_accuracy": 0.16795536279678344,
      "num_tokens": 35453202.0,
      "step": 16385
    },
    {
      "entropy": 6.387524175643921,
      "epoch": 1.7539194178393709,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046992322632746677,
      "loss": 5.9484,
      "mean_token_accuracy": 0.1652113303542137,
      "num_tokens": 35463303.0,
      "step": 16390
    },
    {
      "entropy": 6.444651365280151,
      "epoch": 1.7544544919471345,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046990412671721186,
      "loss": 5.9489,
      "mean_token_accuracy": 0.16309733539819718,
      "num_tokens": 35474104.0,
      "step": 16395
    },
    {
      "entropy": 6.431228351593018,
      "epoch": 1.7549895660548986,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004698850214791276,
      "loss": 5.9412,
      "mean_token_accuracy": 0.1625020131468773,
      "num_tokens": 35486412.0,
      "step": 16400
    },
    {
      "entropy": 6.3739354610443115,
      "epoch": 1.7555246401626625,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004698659106137658,
      "loss": 5.9236,
      "mean_token_accuracy": 0.1591517984867096,
      "num_tokens": 35497505.0,
      "step": 16405
    },
    {
      "entropy": 6.451943016052246,
      "epoch": 1.7560597142704264,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004698467941216782,
      "loss": 5.9708,
      "mean_token_accuracy": 0.16370343416929245,
      "num_tokens": 35508397.0,
      "step": 16410
    },
    {
      "entropy": 6.462685203552246,
      "epoch": 1.7565947883781905,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046982767200341675,
      "loss": 6.0105,
      "mean_token_accuracy": 0.15388662964105607,
      "num_tokens": 35520261.0,
      "step": 16415
    },
    {
      "entropy": 6.444361448287964,
      "epoch": 1.7571298624859542,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046980854425953365,
      "loss": 5.9136,
      "mean_token_accuracy": 0.1549459218978882,
      "num_tokens": 35531816.0,
      "step": 16420
    },
    {
      "entropy": 6.511369276046753,
      "epoch": 1.7576649365937183,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004697894108905813,
      "loss": 5.9969,
      "mean_token_accuracy": 0.15505112260580062,
      "num_tokens": 35542233.0,
      "step": 16425
    },
    {
      "entropy": 6.419626665115357,
      "epoch": 1.7582000107014821,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046977027189711197,
      "loss": 5.9305,
      "mean_token_accuracy": 0.16072979867458342,
      "num_tokens": 35552881.0,
      "step": 16430
    },
    {
      "entropy": 6.412137985229492,
      "epoch": 1.758735084809246,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004697511272796785,
      "loss": 5.8919,
      "mean_token_accuracy": 0.16424526274204254,
      "num_tokens": 35562392.0,
      "step": 16435
    },
    {
      "entropy": 6.444373941421508,
      "epoch": 1.75927015891701,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004697319770388335,
      "loss": 5.9595,
      "mean_token_accuracy": 0.15557699650526047,
      "num_tokens": 35574587.0,
      "step": 16440
    },
    {
      "entropy": 6.399566745758056,
      "epoch": 1.759805233024774,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004697128211751301,
      "loss": 5.8382,
      "mean_token_accuracy": 0.16117219179868697,
      "num_tokens": 35585178.0,
      "step": 16445
    },
    {
      "entropy": 6.38471565246582,
      "epoch": 1.7603403071325379,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004696936596891213,
      "loss": 5.8887,
      "mean_token_accuracy": 0.1594669535756111,
      "num_tokens": 35595501.0,
      "step": 16450
    },
    {
      "entropy": 6.390929412841797,
      "epoch": 1.7608753812403017,
      "grad_norm": 1.0,
      "learning_rate": 0.0004696744925813605,
      "loss": 5.9309,
      "mean_token_accuracy": 0.1628015473484993,
      "num_tokens": 35606794.0,
      "step": 16455
    },
    {
      "entropy": 6.3977405548095705,
      "epoch": 1.7614104553480656,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004696553198524011,
      "loss": 5.9379,
      "mean_token_accuracy": 0.16080324053764344,
      "num_tokens": 35618732.0,
      "step": 16460
    },
    {
      "entropy": 6.394991302490235,
      "epoch": 1.7619455294558297,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004696361415027966,
      "loss": 5.8739,
      "mean_token_accuracy": 0.16381841599941255,
      "num_tokens": 35629802.0,
      "step": 16465
    },
    {
      "entropy": 6.431935119628906,
      "epoch": 1.7624806035635936,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004696169575331009,
      "loss": 5.9954,
      "mean_token_accuracy": 0.15382932126522064,
      "num_tokens": 35640826.0,
      "step": 16470
    },
    {
      "entropy": 6.4772542953491214,
      "epoch": 1.7630156776713575,
      "grad_norm": 1.046875,
      "learning_rate": 0.000469597767943868,
      "loss": 5.9815,
      "mean_token_accuracy": 0.15719096958637238,
      "num_tokens": 35651613.0,
      "step": 16475
    },
    {
      "entropy": 6.396968269348145,
      "epoch": 1.7635507517791216,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004695785727356517,
      "loss": 5.8301,
      "mean_token_accuracy": 0.16165972203016282,
      "num_tokens": 35662334.0,
      "step": 16480
    },
    {
      "entropy": 6.434882688522339,
      "epoch": 1.7640858258868852,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004695593719090066,
      "loss": 5.9998,
      "mean_token_accuracy": 0.15829534828662872,
      "num_tokens": 35673466.0,
      "step": 16485
    },
    {
      "entropy": 6.387656831741333,
      "epoch": 1.7646208999946493,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004695401654644869,
      "loss": 5.9486,
      "mean_token_accuracy": 0.1606345981359482,
      "num_tokens": 35684708.0,
      "step": 16490
    },
    {
      "entropy": 6.39096302986145,
      "epoch": 1.7651559741024132,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046952095340264727,
      "loss": 5.8218,
      "mean_token_accuracy": 0.16419054120779036,
      "num_tokens": 35694888.0,
      "step": 16495
    },
    {
      "entropy": 6.420748996734619,
      "epoch": 1.765691048210177,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046950173572404255,
      "loss": 5.8912,
      "mean_token_accuracy": 0.15797501057386398,
      "num_tokens": 35704844.0,
      "step": 16500
    },
    {
      "entropy": 6.411362981796264,
      "epoch": 1.7662261223179412,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004694825124292275,
      "loss": 5.9456,
      "mean_token_accuracy": 0.16276140064001082,
      "num_tokens": 35715468.0,
      "step": 16505
    },
    {
      "entropy": 6.469284582138061,
      "epoch": 1.7667611964257048,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004694632835187572,
      "loss": 6.0148,
      "mean_token_accuracy": 0.15289920419454575,
      "num_tokens": 35725542.0,
      "step": 16510
    },
    {
      "entropy": 6.484205961227417,
      "epoch": 1.767296270533469,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046944404899318685,
      "loss": 5.9195,
      "mean_token_accuracy": 0.16155464053153992,
      "num_tokens": 35737032.0,
      "step": 16515
    },
    {
      "entropy": 6.434043264389038,
      "epoch": 1.7678313446412328,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000469424808853072,
      "loss": 5.9231,
      "mean_token_accuracy": 0.1591553807258606,
      "num_tokens": 35747742.0,
      "step": 16520
    },
    {
      "entropy": 6.368545246124268,
      "epoch": 1.7683664187489967,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004694055630989681,
      "loss": 5.9017,
      "mean_token_accuracy": 0.15973269790410996,
      "num_tokens": 35757795.0,
      "step": 16525
    },
    {
      "entropy": 6.381338500976563,
      "epoch": 1.7689014928567608,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004693863117314308,
      "loss": 5.8846,
      "mean_token_accuracy": 0.16769442111253738,
      "num_tokens": 35767962.0,
      "step": 16530
    },
    {
      "entropy": 6.42633638381958,
      "epoch": 1.7694365669645244,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004693670547510161,
      "loss": 5.99,
      "mean_token_accuracy": 0.15432004928588866,
      "num_tokens": 35778564.0,
      "step": 16535
    },
    {
      "entropy": 6.4411835193634035,
      "epoch": 1.7699716410722885,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046934779215828006,
      "loss": 5.9175,
      "mean_token_accuracy": 0.1674918830394745,
      "num_tokens": 35788417.0,
      "step": 16540
    },
    {
      "entropy": 6.4104005813598635,
      "epoch": 1.7705067151800524,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004693285239537788,
      "loss": 5.9641,
      "mean_token_accuracy": 0.15438820123672486,
      "num_tokens": 35799724.0,
      "step": 16545
    },
    {
      "entropy": 6.445405960083008,
      "epoch": 1.7710417892878163,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046930925013806873,
      "loss": 5.8848,
      "mean_token_accuracy": 0.16504203230142594,
      "num_tokens": 35810306.0,
      "step": 16550
    },
    {
      "entropy": 6.484496355056763,
      "epoch": 1.7715768633955804,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004692899707117064,
      "loss": 5.9201,
      "mean_token_accuracy": 0.16128137409687043,
      "num_tokens": 35821103.0,
      "step": 16555
    },
    {
      "entropy": 6.388918542861939,
      "epoch": 1.772111937503344,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004692706856752484,
      "loss": 5.9564,
      "mean_token_accuracy": 0.16151179894804954,
      "num_tokens": 35832271.0,
      "step": 16560
    },
    {
      "entropy": 6.410341072082519,
      "epoch": 1.7726470116111082,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046925139502925167,
      "loss": 5.9681,
      "mean_token_accuracy": 0.154161474108696,
      "num_tokens": 35843875.0,
      "step": 16565
    },
    {
      "entropy": 6.457885885238648,
      "epoch": 1.773182085718872,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004692320987742732,
      "loss": 5.9629,
      "mean_token_accuracy": 0.15896851122379302,
      "num_tokens": 35854085.0,
      "step": 16570
    },
    {
      "entropy": 6.479429435729981,
      "epoch": 1.773717159826636,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004692127969108702,
      "loss": 5.9821,
      "mean_token_accuracy": 0.15932624489068986,
      "num_tokens": 35864777.0,
      "step": 16575
    },
    {
      "entropy": 6.443502616882324,
      "epoch": 1.7742522339344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004691934894396,
      "loss": 6.0037,
      "mean_token_accuracy": 0.1522153675556183,
      "num_tokens": 35876822.0,
      "step": 16580
    },
    {
      "entropy": 6.500642871856689,
      "epoch": 1.774787308042164,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046917417636102,
      "loss": 5.9166,
      "mean_token_accuracy": 0.16281876862049102,
      "num_tokens": 35886997.0,
      "step": 16585
    },
    {
      "entropy": 6.455738925933838,
      "epoch": 1.7753223821499278,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000469154857675688,
      "loss": 5.9755,
      "mean_token_accuracy": 0.15964241325855255,
      "num_tokens": 35897871.0,
      "step": 16590
    },
    {
      "entropy": 6.356819772720337,
      "epoch": 1.7758574562576916,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004691355333841617,
      "loss": 5.9643,
      "mean_token_accuracy": 0.16855514496564866,
      "num_tokens": 35908791.0,
      "step": 16595
    },
    {
      "entropy": 6.466270589828492,
      "epoch": 1.7763925303654555,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004691162034869993,
      "loss": 5.992,
      "mean_token_accuracy": 0.15536285638809205,
      "num_tokens": 35919436.0,
      "step": 16600
    },
    {
      "entropy": 6.481278991699218,
      "epoch": 1.7769276044732196,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004690968679847586,
      "loss": 5.8743,
      "mean_token_accuracy": 0.1599051460623741,
      "num_tokens": 35929851.0,
      "step": 16605
    },
    {
      "entropy": 6.422090721130371,
      "epoch": 1.7774626785809835,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004690775268779983,
      "loss": 5.9115,
      "mean_token_accuracy": 0.1601647436618805,
      "num_tokens": 35940271.0,
      "step": 16610
    },
    {
      "entropy": 6.372260999679566,
      "epoch": 1.7779977526887474,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004690581801672765,
      "loss": 6.022,
      "mean_token_accuracy": 0.1523737922310829,
      "num_tokens": 35951460.0,
      "step": 16615
    },
    {
      "entropy": 6.393130683898926,
      "epoch": 1.7785328267965115,
      "grad_norm": 1.0,
      "learning_rate": 0.00046903882785315215,
      "loss": 5.9081,
      "mean_token_accuracy": 0.15956609100103378,
      "num_tokens": 35961530.0,
      "step": 16620
    },
    {
      "entropy": 6.400392007827759,
      "epoch": 1.7790679009042751,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004690194699361839,
      "loss": 5.8879,
      "mean_token_accuracy": 0.16717612892389297,
      "num_tokens": 35971085.0,
      "step": 16625
    },
    {
      "entropy": 6.375500774383545,
      "epoch": 1.7796029750120392,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046900010641693073,
      "loss": 5.9304,
      "mean_token_accuracy": 0.15317088589072228,
      "num_tokens": 35981393.0,
      "step": 16630
    },
    {
      "entropy": 6.394368505477905,
      "epoch": 1.7801380491198031,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004689807372959517,
      "loss": 5.8954,
      "mean_token_accuracy": 0.16572047173976898,
      "num_tokens": 35992847.0,
      "step": 16635
    },
    {
      "entropy": 6.39155125617981,
      "epoch": 1.780673123227567,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046896136257380615,
      "loss": 5.9502,
      "mean_token_accuracy": 0.15711943656206132,
      "num_tokens": 36002991.0,
      "step": 16640
    },
    {
      "entropy": 6.407492971420288,
      "epoch": 1.781208197335331,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00046894198225105356,
      "loss": 5.9432,
      "mean_token_accuracy": 0.16548591554164888,
      "num_tokens": 36015140.0,
      "step": 16645
    },
    {
      "entropy": 6.396450042724609,
      "epoch": 1.7817432714430947,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004689225963282534,
      "loss": 5.8566,
      "mean_token_accuracy": 0.16230641752481462,
      "num_tokens": 36026014.0,
      "step": 16650
    },
    {
      "entropy": 6.519558143615723,
      "epoch": 1.7822783455508588,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004689032048059656,
      "loss": 6.0242,
      "mean_token_accuracy": 0.15558198243379592,
      "num_tokens": 36035926.0,
      "step": 16655
    },
    {
      "entropy": 6.430045032501221,
      "epoch": 1.7828134196586227,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046888380768475,
      "loss": 5.9347,
      "mean_token_accuracy": 0.159661927819252,
      "num_tokens": 36046406.0,
      "step": 16660
    },
    {
      "entropy": 6.426459884643554,
      "epoch": 1.7833484937663866,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004688644049651667,
      "loss": 5.8816,
      "mean_token_accuracy": 0.16179632842540742,
      "num_tokens": 36056440.0,
      "step": 16665
    },
    {
      "entropy": 6.377339124679565,
      "epoch": 1.7838835678741507,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046884499664777586,
      "loss": 5.966,
      "mean_token_accuracy": 0.1547910153865814,
      "num_tokens": 36067244.0,
      "step": 16670
    },
    {
      "entropy": 6.399812889099121,
      "epoch": 1.7844186419819144,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004688255827331381,
      "loss": 5.8391,
      "mean_token_accuracy": 0.16652311533689498,
      "num_tokens": 36078110.0,
      "step": 16675
    },
    {
      "entropy": 6.437540006637573,
      "epoch": 1.7849537160896785,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004688061632218137,
      "loss": 5.9256,
      "mean_token_accuracy": 0.16004159152507783,
      "num_tokens": 36088805.0,
      "step": 16680
    },
    {
      "entropy": 6.392685461044311,
      "epoch": 1.7854887901974423,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046878673811436375,
      "loss": 5.93,
      "mean_token_accuracy": 0.164516843855381,
      "num_tokens": 36098619.0,
      "step": 16685
    },
    {
      "entropy": 6.453076696395874,
      "epoch": 1.7860238643052062,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004687673074113489,
      "loss": 5.9862,
      "mean_token_accuracy": 0.15923232436180115,
      "num_tokens": 36110604.0,
      "step": 16690
    },
    {
      "entropy": 6.496310377120972,
      "epoch": 1.7865589384129703,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004687478711133303,
      "loss": 5.9845,
      "mean_token_accuracy": 0.15685592740774154,
      "num_tokens": 36121669.0,
      "step": 16695
    },
    {
      "entropy": 6.306751728057861,
      "epoch": 1.787094012520734,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004687284292208691,
      "loss": 5.767,
      "mean_token_accuracy": 0.17262455224990844,
      "num_tokens": 36132165.0,
      "step": 16700
    },
    {
      "entropy": 6.380014133453369,
      "epoch": 1.787629086628498,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046870898173452676,
      "loss": 5.9159,
      "mean_token_accuracy": 0.16204007118940353,
      "num_tokens": 36142830.0,
      "step": 16705
    },
    {
      "entropy": 6.3993144035339355,
      "epoch": 1.788164160736262,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004686895286548648,
      "loss": 5.8916,
      "mean_token_accuracy": 0.16442764848470687,
      "num_tokens": 36153812.0,
      "step": 16710
    },
    {
      "entropy": 6.374171447753906,
      "epoch": 1.7886992348440258,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046867006998244494,
      "loss": 5.933,
      "mean_token_accuracy": 0.16392334401607514,
      "num_tokens": 36163599.0,
      "step": 16715
    },
    {
      "entropy": 6.419242572784424,
      "epoch": 1.78923430895179,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046865060571782896,
      "loss": 5.9675,
      "mean_token_accuracy": 0.15254207253456115,
      "num_tokens": 36173815.0,
      "step": 16720
    },
    {
      "entropy": 6.528357887268067,
      "epoch": 1.7897693830595538,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046863113586157894,
      "loss": 6.032,
      "mean_token_accuracy": 0.15590094923973083,
      "num_tokens": 36186131.0,
      "step": 16725
    },
    {
      "entropy": 6.449932193756103,
      "epoch": 1.7903044571673177,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004686116604142572,
      "loss": 5.9007,
      "mean_token_accuracy": 0.16529717147350312,
      "num_tokens": 36196084.0,
      "step": 16730
    },
    {
      "entropy": 6.340091896057129,
      "epoch": 1.7908395312750816,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004685921793764259,
      "loss": 5.8563,
      "mean_token_accuracy": 0.16949861496686935,
      "num_tokens": 36206948.0,
      "step": 16735
    },
    {
      "entropy": 6.452722072601318,
      "epoch": 1.7913746053828454,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004685726927486476,
      "loss": 5.9997,
      "mean_token_accuracy": 0.15411876440048217,
      "num_tokens": 36217202.0,
      "step": 16740
    },
    {
      "entropy": 6.519041872024536,
      "epoch": 1.7919096794906095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046855320053148505,
      "loss": 6.037,
      "mean_token_accuracy": 0.15328154265880584,
      "num_tokens": 36227914.0,
      "step": 16745
    },
    {
      "entropy": 6.484017372131348,
      "epoch": 1.7924447535983734,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000468533702725501,
      "loss": 5.9779,
      "mean_token_accuracy": 0.15647002607583999,
      "num_tokens": 36239344.0,
      "step": 16750
    },
    {
      "entropy": 6.429100227355957,
      "epoch": 1.7929798277061373,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046851419933125843,
      "loss": 6.0028,
      "mean_token_accuracy": 0.155867899954319,
      "num_tokens": 36250438.0,
      "step": 16755
    },
    {
      "entropy": 6.423089504241943,
      "epoch": 1.7935149018139014,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004684946903493206,
      "loss": 5.9525,
      "mean_token_accuracy": 0.1552292972803116,
      "num_tokens": 36260545.0,
      "step": 16760
    },
    {
      "entropy": 6.347021675109863,
      "epoch": 1.794049975921665,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004684751757802508,
      "loss": 5.865,
      "mean_token_accuracy": 0.16056947857141496,
      "num_tokens": 36272577.0,
      "step": 16765
    },
    {
      "entropy": 6.438727903366089,
      "epoch": 1.7945850500294291,
      "grad_norm": 1.0,
      "learning_rate": 0.00046845565562461243,
      "loss": 5.9009,
      "mean_token_accuracy": 0.1647948741912842,
      "num_tokens": 36284161.0,
      "step": 16770
    },
    {
      "entropy": 6.436196565628052,
      "epoch": 1.795120124137193,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046843612988296926,
      "loss": 5.9262,
      "mean_token_accuracy": 0.15516040325164795,
      "num_tokens": 36294910.0,
      "step": 16775
    },
    {
      "entropy": 6.409628677368164,
      "epoch": 1.795655198244957,
      "grad_norm": 1.125,
      "learning_rate": 0.0004684165985558849,
      "loss": 5.8907,
      "mean_token_accuracy": 0.16259705573320388,
      "num_tokens": 36305388.0,
      "step": 16780
    },
    {
      "entropy": 6.395497226715088,
      "epoch": 1.796190272352721,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004683970616439236,
      "loss": 5.9353,
      "mean_token_accuracy": 0.16593302488327027,
      "num_tokens": 36316696.0,
      "step": 16785
    },
    {
      "entropy": 6.381643390655517,
      "epoch": 1.7967253464604847,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004683775191476492,
      "loss": 5.8763,
      "mean_token_accuracy": 0.17079982459545134,
      "num_tokens": 36327246.0,
      "step": 16790
    },
    {
      "entropy": 6.435485219955444,
      "epoch": 1.7972604205682488,
      "grad_norm": 1.0,
      "learning_rate": 0.0004683579710676262,
      "loss": 6.0039,
      "mean_token_accuracy": 0.15580273568630218,
      "num_tokens": 36338089.0,
      "step": 16795
    },
    {
      "entropy": 6.426194095611573,
      "epoch": 1.7977954946760126,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046833841740441886,
      "loss": 5.9613,
      "mean_token_accuracy": 0.16057096868753434,
      "num_tokens": 36350285.0,
      "step": 16800
    },
    {
      "entropy": 6.4751416683197025,
      "epoch": 1.7983305687837765,
      "grad_norm": 1.0,
      "learning_rate": 0.0004683188581585919,
      "loss": 5.9429,
      "mean_token_accuracy": 0.1562327727675438,
      "num_tokens": 36360431.0,
      "step": 16805
    },
    {
      "entropy": 6.41259799003601,
      "epoch": 1.7988656428915406,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004682992933307101,
      "loss": 5.8991,
      "mean_token_accuracy": 0.16734291464090348,
      "num_tokens": 36371257.0,
      "step": 16810
    },
    {
      "entropy": 6.352460718154907,
      "epoch": 1.7994007169993043,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046827972292133835,
      "loss": 5.8642,
      "mean_token_accuracy": 0.1639757364988327,
      "num_tokens": 36382613.0,
      "step": 16815
    },
    {
      "entropy": 6.4879059314727785,
      "epoch": 1.7999357911070684,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004682601469310418,
      "loss": 5.9497,
      "mean_token_accuracy": 0.1578674241900444,
      "num_tokens": 36393799.0,
      "step": 16820
    },
    {
      "entropy": 6.452397346496582,
      "epoch": 1.8004708652148322,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046824056536038565,
      "loss": 5.9415,
      "mean_token_accuracy": 0.15785589963197708,
      "num_tokens": 36403914.0,
      "step": 16825
    },
    {
      "entropy": 6.345038604736328,
      "epoch": 1.8010059393225961,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004682209782099353,
      "loss": 5.9079,
      "mean_token_accuracy": 0.16415497064590454,
      "num_tokens": 36415022.0,
      "step": 16830
    },
    {
      "entropy": 6.414320230484009,
      "epoch": 1.8015410134303602,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046820138548025634,
      "loss": 5.8673,
      "mean_token_accuracy": 0.16815999299287795,
      "num_tokens": 36426155.0,
      "step": 16835
    },
    {
      "entropy": 6.409469127655029,
      "epoch": 1.8020760875381239,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046818178717191464,
      "loss": 5.9276,
      "mean_token_accuracy": 0.1586792454123497,
      "num_tokens": 36437345.0,
      "step": 16840
    },
    {
      "entropy": 6.437321996688842,
      "epoch": 1.802611161645888,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004681621832854759,
      "loss": 5.9563,
      "mean_token_accuracy": 0.16967403143644333,
      "num_tokens": 36448634.0,
      "step": 16845
    },
    {
      "entropy": 6.424160385131836,
      "epoch": 1.8031462357536518,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046814257382150627,
      "loss": 5.8914,
      "mean_token_accuracy": 0.17238007634878158,
      "num_tokens": 36458870.0,
      "step": 16850
    },
    {
      "entropy": 6.433578538894653,
      "epoch": 1.8036813098614157,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046812295878057193,
      "loss": 6.0431,
      "mean_token_accuracy": 0.1555313527584076,
      "num_tokens": 36469865.0,
      "step": 16855
    },
    {
      "entropy": 6.440585565567017,
      "epoch": 1.8042163839691798,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004681033381632393,
      "loss": 5.9321,
      "mean_token_accuracy": 0.15892516672611237,
      "num_tokens": 36480502.0,
      "step": 16860
    },
    {
      "entropy": 6.469013595581055,
      "epoch": 1.8047514580769437,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004680837119700749,
      "loss": 5.9494,
      "mean_token_accuracy": 0.16083444952964782,
      "num_tokens": 36492356.0,
      "step": 16865
    },
    {
      "entropy": 6.318448925018311,
      "epoch": 1.8052865321847076,
      "grad_norm": 0.890625,
      "learning_rate": 0.00046806408020164553,
      "loss": 5.9179,
      "mean_token_accuracy": 0.16209062486886977,
      "num_tokens": 36504659.0,
      "step": 16870
    },
    {
      "entropy": 6.391596031188965,
      "epoch": 1.8058216062924715,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046804444285851796,
      "loss": 5.9255,
      "mean_token_accuracy": 0.1637495830655098,
      "num_tokens": 36516042.0,
      "step": 16875
    },
    {
      "entropy": 6.475486707687378,
      "epoch": 1.8063566804002353,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046802479994125916,
      "loss": 6.017,
      "mean_token_accuracy": 0.15332066714763642,
      "num_tokens": 36527417.0,
      "step": 16880
    },
    {
      "entropy": 6.416657209396362,
      "epoch": 1.8068917545079994,
      "grad_norm": 1.0,
      "learning_rate": 0.0004680051514504364,
      "loss": 5.9272,
      "mean_token_accuracy": 0.16534601151943207,
      "num_tokens": 36538182.0,
      "step": 16885
    },
    {
      "entropy": 6.393430614471436,
      "epoch": 1.8074268286157633,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004679854973866171,
      "loss": 5.9756,
      "mean_token_accuracy": 0.16015183925628662,
      "num_tokens": 36549835.0,
      "step": 16890
    },
    {
      "entropy": 6.3928971767425535,
      "epoch": 1.8079619027235272,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046796583775036857,
      "loss": 5.9117,
      "mean_token_accuracy": 0.1678117632865906,
      "num_tokens": 36560238.0,
      "step": 16895
    },
    {
      "entropy": 6.437711763381958,
      "epoch": 1.8084969768312913,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046794617254225865,
      "loss": 5.9399,
      "mean_token_accuracy": 0.1601879268884659,
      "num_tokens": 36571267.0,
      "step": 16900
    },
    {
      "entropy": 6.43780632019043,
      "epoch": 1.809032050939055,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004679265017628551,
      "loss": 5.9993,
      "mean_token_accuracy": 0.1619711771607399,
      "num_tokens": 36582174.0,
      "step": 16905
    },
    {
      "entropy": 6.488168001174927,
      "epoch": 1.809567125046819,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004679068254127259,
      "loss": 5.9471,
      "mean_token_accuracy": 0.16253016889095306,
      "num_tokens": 36593054.0,
      "step": 16910
    },
    {
      "entropy": 6.442882013320923,
      "epoch": 1.810102199154583,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046788714349243923,
      "loss": 5.9509,
      "mean_token_accuracy": 0.16032088100910186,
      "num_tokens": 36603962.0,
      "step": 16915
    },
    {
      "entropy": 6.435347986221314,
      "epoch": 1.8106372732623468,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004678674560025634,
      "loss": 5.9318,
      "mean_token_accuracy": 0.16086623221635818,
      "num_tokens": 36616003.0,
      "step": 16920
    },
    {
      "entropy": 6.497943305969239,
      "epoch": 1.811172347370111,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004678477629436669,
      "loss": 6.0311,
      "mean_token_accuracy": 0.14607261568307878,
      "num_tokens": 36627189.0,
      "step": 16925
    },
    {
      "entropy": 6.476930284500122,
      "epoch": 1.8117074214778746,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004678280643163184,
      "loss": 5.9761,
      "mean_token_accuracy": 0.15763429701328277,
      "num_tokens": 36637756.0,
      "step": 16930
    },
    {
      "entropy": 6.40727686882019,
      "epoch": 1.8122424955856387,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004678083601210865,
      "loss": 5.9576,
      "mean_token_accuracy": 0.16215287446975707,
      "num_tokens": 36648946.0,
      "step": 16935
    },
    {
      "entropy": 6.49388747215271,
      "epoch": 1.8127775696934025,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004677886503585404,
      "loss": 5.9892,
      "mean_token_accuracy": 0.16390926539897918,
      "num_tokens": 36660184.0,
      "step": 16940
    },
    {
      "entropy": 6.518471670150757,
      "epoch": 1.8133126438011664,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046776893502924904,
      "loss": 6.0181,
      "mean_token_accuracy": 0.15671016573905944,
      "num_tokens": 36671346.0,
      "step": 16945
    },
    {
      "entropy": 6.424312162399292,
      "epoch": 1.8138477179089305,
      "grad_norm": 1.0,
      "learning_rate": 0.0004677492141337818,
      "loss": 5.9396,
      "mean_token_accuracy": 0.15987934917211533,
      "num_tokens": 36682917.0,
      "step": 16950
    },
    {
      "entropy": 6.415992593765258,
      "epoch": 1.8143827920166942,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004677294876727081,
      "loss": 6.0076,
      "mean_token_accuracy": 0.1501097284257412,
      "num_tokens": 36693577.0,
      "step": 16955
    },
    {
      "entropy": 6.527229595184326,
      "epoch": 1.8149178661244583,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004677097556465975,
      "loss": 5.9541,
      "mean_token_accuracy": 0.15965773910284042,
      "num_tokens": 36704915.0,
      "step": 16960
    },
    {
      "entropy": 6.540899705886841,
      "epoch": 1.8154529402322221,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00046769001805601983,
      "loss": 6.0637,
      "mean_token_accuracy": 0.15031189918518068,
      "num_tokens": 36716086.0,
      "step": 16965
    },
    {
      "entropy": 6.376270723342896,
      "epoch": 1.815988014339986,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046767027490154494,
      "loss": 5.9199,
      "mean_token_accuracy": 0.15519306436181068,
      "num_tokens": 36726110.0,
      "step": 16970
    },
    {
      "entropy": 6.4361354351043705,
      "epoch": 1.8165230884477501,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046765052618374296,
      "loss": 5.8777,
      "mean_token_accuracy": 0.16603032052516936,
      "num_tokens": 36735846.0,
      "step": 16975
    },
    {
      "entropy": 6.440337705612182,
      "epoch": 1.8170581625555138,
      "grad_norm": 0.96875,
      "learning_rate": 0.000467630771903184,
      "loss": 5.9656,
      "mean_token_accuracy": 0.16021115332841873,
      "num_tokens": 36746851.0,
      "step": 16980
    },
    {
      "entropy": 6.45834527015686,
      "epoch": 1.8175932366632779,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004676110120604387,
      "loss": 5.9089,
      "mean_token_accuracy": 0.1626890629529953,
      "num_tokens": 36758032.0,
      "step": 16985
    },
    {
      "entropy": 6.454544639587402,
      "epoch": 1.8181283107710418,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004675912466560775,
      "loss": 5.9675,
      "mean_token_accuracy": 0.1603601887822151,
      "num_tokens": 36768939.0,
      "step": 16990
    },
    {
      "entropy": 6.46444787979126,
      "epoch": 1.8186633848788056,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046757147569067106,
      "loss": 6.0103,
      "mean_token_accuracy": 0.1547171950340271,
      "num_tokens": 36780338.0,
      "step": 16995
    },
    {
      "entropy": 6.448343801498413,
      "epoch": 1.8191984589865697,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004675516991647903,
      "loss": 5.88,
      "mean_token_accuracy": 0.152728820592165,
      "num_tokens": 36791384.0,
      "step": 17000
    },
    {
      "entropy": 6.43792576789856,
      "epoch": 1.8197335330943336,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046753191707900636,
      "loss": 5.9229,
      "mean_token_accuracy": 0.15291493088006974,
      "num_tokens": 36802460.0,
      "step": 17005
    },
    {
      "entropy": 6.468645668029785,
      "epoch": 1.8202686072020975,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046751212943389026,
      "loss": 5.9952,
      "mean_token_accuracy": 0.15336545258760453,
      "num_tokens": 36813599.0,
      "step": 17010
    },
    {
      "entropy": 6.451990938186645,
      "epoch": 1.8208036813098614,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004674923362300135,
      "loss": 5.8764,
      "mean_token_accuracy": 0.1619970306754112,
      "num_tokens": 36823100.0,
      "step": 17015
    },
    {
      "entropy": 6.471307849884033,
      "epoch": 1.8213387554176252,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046747253746794767,
      "loss": 5.9915,
      "mean_token_accuracy": 0.1510108396410942,
      "num_tokens": 36834245.0,
      "step": 17020
    },
    {
      "entropy": 6.4341333389282225,
      "epoch": 1.8218738295253893,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046745273314826425,
      "loss": 5.8847,
      "mean_token_accuracy": 0.16606824547052385,
      "num_tokens": 36844126.0,
      "step": 17025
    },
    {
      "entropy": 6.421468734741211,
      "epoch": 1.8224089036331532,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004674329232715353,
      "loss": 5.8711,
      "mean_token_accuracy": 0.17317622750997544,
      "num_tokens": 36855451.0,
      "step": 17030
    },
    {
      "entropy": 6.425603008270263,
      "epoch": 1.822943977740917,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004674131078383327,
      "loss": 5.9562,
      "mean_token_accuracy": 0.16040031760931014,
      "num_tokens": 36864994.0,
      "step": 17035
    },
    {
      "entropy": 6.409979391098022,
      "epoch": 1.8234790518486812,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004673932868492286,
      "loss": 5.9064,
      "mean_token_accuracy": 0.15954847633838654,
      "num_tokens": 36875131.0,
      "step": 17040
    },
    {
      "entropy": 6.483668279647827,
      "epoch": 1.8240141259564449,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046737346030479546,
      "loss": 5.9822,
      "mean_token_accuracy": 0.15937425047159196,
      "num_tokens": 36884789.0,
      "step": 17045
    },
    {
      "entropy": 6.396101713180542,
      "epoch": 1.824549200064209,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046735362820560564,
      "loss": 5.8551,
      "mean_token_accuracy": 0.15868693441152573,
      "num_tokens": 36895199.0,
      "step": 17050
    },
    {
      "entropy": 6.426017904281617,
      "epoch": 1.8250842741719728,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004673337905522318,
      "loss": 5.9795,
      "mean_token_accuracy": 0.1534508153796196,
      "num_tokens": 36906495.0,
      "step": 17055
    },
    {
      "entropy": 6.424411487579346,
      "epoch": 1.8256193482797367,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046731394734524686,
      "loss": 5.9428,
      "mean_token_accuracy": 0.15226158499717712,
      "num_tokens": 36916907.0,
      "step": 17060
    },
    {
      "entropy": 6.4135347366333,
      "epoch": 1.8261544223875008,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046729409858522363,
      "loss": 5.9132,
      "mean_token_accuracy": 0.16037887334823608,
      "num_tokens": 36927872.0,
      "step": 17065
    },
    {
      "entropy": 6.440005779266357,
      "epoch": 1.8266894964952645,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004672742442727354,
      "loss": 5.9564,
      "mean_token_accuracy": 0.16131239086389543,
      "num_tokens": 36938575.0,
      "step": 17070
    },
    {
      "entropy": 6.399770736694336,
      "epoch": 1.8272245706030286,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046725438440835536,
      "loss": 5.8333,
      "mean_token_accuracy": 0.17475120425224305,
      "num_tokens": 36949179.0,
      "step": 17075
    },
    {
      "entropy": 6.3860071182250975,
      "epoch": 1.8277596447107924,
      "grad_norm": 1.046875,
      "learning_rate": 0.000467234518992657,
      "loss": 5.8819,
      "mean_token_accuracy": 0.16672852039337158,
      "num_tokens": 36959164.0,
      "step": 17080
    },
    {
      "entropy": 6.503335905075073,
      "epoch": 1.8282947188185563,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046721464802621383,
      "loss": 6.0139,
      "mean_token_accuracy": 0.16069333106279374,
      "num_tokens": 36969091.0,
      "step": 17085
    },
    {
      "entropy": 6.476371431350708,
      "epoch": 1.8288297929263204,
      "grad_norm": 0.890625,
      "learning_rate": 0.00046719477150959984,
      "loss": 5.9471,
      "mean_token_accuracy": 0.15733551383018493,
      "num_tokens": 36980481.0,
      "step": 17090
    },
    {
      "entropy": 6.433249616622925,
      "epoch": 1.829364867034084,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046717488944338873,
      "loss": 5.9195,
      "mean_token_accuracy": 0.1608563929796219,
      "num_tokens": 36991349.0,
      "step": 17095
    },
    {
      "entropy": 6.394046878814697,
      "epoch": 1.8298999411418482,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046715500182815473,
      "loss": 6.0047,
      "mean_token_accuracy": 0.15723074674606324,
      "num_tokens": 37002803.0,
      "step": 17100
    },
    {
      "entropy": 6.48158221244812,
      "epoch": 1.830435015249612,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046713510866447204,
      "loss": 5.9592,
      "mean_token_accuracy": 0.1639849826693535,
      "num_tokens": 37013578.0,
      "step": 17105
    },
    {
      "entropy": 6.451555824279785,
      "epoch": 1.830970089357376,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004671152099529152,
      "loss": 5.8702,
      "mean_token_accuracy": 0.16197686046361923,
      "num_tokens": 37024012.0,
      "step": 17110
    },
    {
      "entropy": 6.420357656478882,
      "epoch": 1.83150516346514,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046709530569405855,
      "loss": 5.8819,
      "mean_token_accuracy": 0.16420998871326448,
      "num_tokens": 37034149.0,
      "step": 17115
    },
    {
      "entropy": 6.357254219055176,
      "epoch": 1.8320402375729037,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004670753958884769,
      "loss": 5.8575,
      "mean_token_accuracy": 0.1658901035785675,
      "num_tokens": 37045669.0,
      "step": 17120
    },
    {
      "entropy": 6.401954460144043,
      "epoch": 1.8325753116806678,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00046705548053674524,
      "loss": 5.953,
      "mean_token_accuracy": 0.1608389899134636,
      "num_tokens": 37056611.0,
      "step": 17125
    },
    {
      "entropy": 6.388093090057373,
      "epoch": 1.8331103857884317,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046703555963943855,
      "loss": 5.9234,
      "mean_token_accuracy": 0.15868833065032958,
      "num_tokens": 37067387.0,
      "step": 17130
    },
    {
      "entropy": 6.532819080352783,
      "epoch": 1.8336454598961955,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046701563319713214,
      "loss": 5.9989,
      "mean_token_accuracy": 0.15195754468441008,
      "num_tokens": 37078151.0,
      "step": 17135
    },
    {
      "entropy": 6.480370235443115,
      "epoch": 1.8341805340039596,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046699570121040126,
      "loss": 5.9737,
      "mean_token_accuracy": 0.16074947267770767,
      "num_tokens": 37088219.0,
      "step": 17140
    },
    {
      "entropy": 6.385111093521118,
      "epoch": 1.8347156081117235,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004669757636798215,
      "loss": 5.8654,
      "mean_token_accuracy": 0.15907205790281295,
      "num_tokens": 37098769.0,
      "step": 17145
    },
    {
      "entropy": 6.464435815811157,
      "epoch": 1.8352506822194874,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004669558206059684,
      "loss": 6.0028,
      "mean_token_accuracy": 0.14793941751122475,
      "num_tokens": 37110716.0,
      "step": 17150
    },
    {
      "entropy": 6.442920351028443,
      "epoch": 1.8357857563272513,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004669358719894181,
      "loss": 5.8986,
      "mean_token_accuracy": 0.1567833125591278,
      "num_tokens": 37122785.0,
      "step": 17155
    },
    {
      "entropy": 6.3875213146209715,
      "epoch": 1.8363208304350152,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004669159178307465,
      "loss": 5.9177,
      "mean_token_accuracy": 0.16035102754831315,
      "num_tokens": 37133308.0,
      "step": 17160
    },
    {
      "entropy": 6.335878801345825,
      "epoch": 1.8368559045427792,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004668959581305296,
      "loss": 5.8831,
      "mean_token_accuracy": 0.16129842400550842,
      "num_tokens": 37143916.0,
      "step": 17165
    },
    {
      "entropy": 6.413056707382202,
      "epoch": 1.8373909786505431,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046687599288934394,
      "loss": 5.9023,
      "mean_token_accuracy": 0.15948037058115005,
      "num_tokens": 37154335.0,
      "step": 17170
    },
    {
      "entropy": 6.43286395072937,
      "epoch": 1.837926052758307,
      "grad_norm": 1.0625,
      "learning_rate": 0.000466856022107766,
      "loss": 5.9419,
      "mean_token_accuracy": 0.1580684944987297,
      "num_tokens": 37164116.0,
      "step": 17175
    },
    {
      "entropy": 6.433859872817993,
      "epoch": 1.838461126866071,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046683604578637235,
      "loss": 6.0058,
      "mean_token_accuracy": 0.16007013767957687,
      "num_tokens": 37175407.0,
      "step": 17180
    },
    {
      "entropy": 6.4126873970031735,
      "epoch": 1.8389962009738348,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046681606392573975,
      "loss": 5.896,
      "mean_token_accuracy": 0.1595083713531494,
      "num_tokens": 37185339.0,
      "step": 17185
    },
    {
      "entropy": 6.459780025482178,
      "epoch": 1.8395312750815989,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004667960765264454,
      "loss": 5.9816,
      "mean_token_accuracy": 0.15751735121011734,
      "num_tokens": 37195145.0,
      "step": 17190
    },
    {
      "entropy": 6.4546839714050295,
      "epoch": 1.8400663491893627,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004667760835890662,
      "loss": 6.0053,
      "mean_token_accuracy": 0.15362785756587982,
      "num_tokens": 37206454.0,
      "step": 17195
    },
    {
      "entropy": 6.436693716049194,
      "epoch": 1.8406014232971266,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046675608511417957,
      "loss": 5.9083,
      "mean_token_accuracy": 0.1607656165957451,
      "num_tokens": 37216646.0,
      "step": 17200
    },
    {
      "entropy": 6.327355861663818,
      "epoch": 1.8411364974048907,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046673608110236293,
      "loss": 5.8563,
      "mean_token_accuracy": 0.16786355525255203,
      "num_tokens": 37227368.0,
      "step": 17205
    },
    {
      "entropy": 6.46039605140686,
      "epoch": 1.8416715715126544,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004667160715541939,
      "loss": 5.9801,
      "mean_token_accuracy": 0.1626427322626114,
      "num_tokens": 37238150.0,
      "step": 17210
    },
    {
      "entropy": 6.447904920578003,
      "epoch": 1.8422066456204185,
      "grad_norm": 1.125,
      "learning_rate": 0.00046669605647025014,
      "loss": 5.9087,
      "mean_token_accuracy": 0.15959220379590988,
      "num_tokens": 37249534.0,
      "step": 17215
    },
    {
      "entropy": 6.510118722915649,
      "epoch": 1.8427417197281823,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004666760358511098,
      "loss": 6.0157,
      "mean_token_accuracy": 0.1596991777420044,
      "num_tokens": 37260633.0,
      "step": 17220
    },
    {
      "entropy": 6.458601665496826,
      "epoch": 1.8432767938359462,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046665600969735086,
      "loss": 5.9576,
      "mean_token_accuracy": 0.15845999419689177,
      "num_tokens": 37271978.0,
      "step": 17225
    },
    {
      "entropy": 6.359670686721802,
      "epoch": 1.8438118679437103,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004666359780095515,
      "loss": 5.8323,
      "mean_token_accuracy": 0.16788816452026367,
      "num_tokens": 37284130.0,
      "step": 17230
    },
    {
      "entropy": 6.436881113052368,
      "epoch": 1.844346942051474,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004666159407882902,
      "loss": 6.0163,
      "mean_token_accuracy": 0.15542217642068862,
      "num_tokens": 37295648.0,
      "step": 17235
    },
    {
      "entropy": 6.459294176101684,
      "epoch": 1.844882016159238,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004665958980341456,
      "loss": 5.9735,
      "mean_token_accuracy": 0.16088982969522475,
      "num_tokens": 37306715.0,
      "step": 17240
    },
    {
      "entropy": 6.451521396636963,
      "epoch": 1.845417090267002,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046657584974769636,
      "loss": 5.9841,
      "mean_token_accuracy": 0.1615831062197685,
      "num_tokens": 37317605.0,
      "step": 17245
    },
    {
      "entropy": 6.405914783477783,
      "epoch": 1.8459521643747658,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004665557959295213,
      "loss": 5.9142,
      "mean_token_accuracy": 0.1616660937666893,
      "num_tokens": 37328898.0,
      "step": 17250
    },
    {
      "entropy": 6.41571569442749,
      "epoch": 1.84648723848253,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046653573658019963,
      "loss": 5.8794,
      "mean_token_accuracy": 0.16181340515613557,
      "num_tokens": 37340108.0,
      "step": 17255
    },
    {
      "entropy": 6.4372334480285645,
      "epoch": 1.8470223125902936,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046651567170031053,
      "loss": 5.9996,
      "mean_token_accuracy": 0.16056283712387084,
      "num_tokens": 37351773.0,
      "step": 17260
    },
    {
      "entropy": 6.407811403274536,
      "epoch": 1.8475573866980577,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004664956012904332,
      "loss": 5.9179,
      "mean_token_accuracy": 0.1612090840935707,
      "num_tokens": 37362748.0,
      "step": 17265
    },
    {
      "entropy": 6.424991226196289,
      "epoch": 1.8480924608058216,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004664755253511474,
      "loss": 6.0053,
      "mean_token_accuracy": 0.15509541481733322,
      "num_tokens": 37373487.0,
      "step": 17270
    },
    {
      "entropy": 6.4147059440612795,
      "epoch": 1.8486275349135854,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004664554438830326,
      "loss": 5.8508,
      "mean_token_accuracy": 0.1616704523563385,
      "num_tokens": 37383302.0,
      "step": 17275
    },
    {
      "entropy": 6.430067920684815,
      "epoch": 1.8491626090213495,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004664353568866689,
      "loss": 5.8955,
      "mean_token_accuracy": 0.1582562506198883,
      "num_tokens": 37393085.0,
      "step": 17280
    },
    {
      "entropy": 6.4770910263061525,
      "epoch": 1.8496976831291134,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000466415264362636,
      "loss": 5.9835,
      "mean_token_accuracy": 0.15887303799390792,
      "num_tokens": 37403774.0,
      "step": 17285
    },
    {
      "entropy": 6.440846395492554,
      "epoch": 1.8502327572368773,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046639516631151435,
      "loss": 5.9251,
      "mean_token_accuracy": 0.16732970476150513,
      "num_tokens": 37414356.0,
      "step": 17290
    },
    {
      "entropy": 6.426045942306518,
      "epoch": 1.8507678313446412,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046637506273388416,
      "loss": 5.9673,
      "mean_token_accuracy": 0.15677843987941742,
      "num_tokens": 37426381.0,
      "step": 17295
    },
    {
      "entropy": 6.44749116897583,
      "epoch": 1.851302905452405,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046635495363032595,
      "loss": 5.936,
      "mean_token_accuracy": 0.15884182304143907,
      "num_tokens": 37436259.0,
      "step": 17300
    },
    {
      "entropy": 6.4382908821105955,
      "epoch": 1.8518379795601692,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00046633483900142026,
      "loss": 5.9044,
      "mean_token_accuracy": 0.1668746665120125,
      "num_tokens": 37447498.0,
      "step": 17305
    },
    {
      "entropy": 6.408259487152099,
      "epoch": 1.852373053667933,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000466314718847748,
      "loss": 5.9115,
      "mean_token_accuracy": 0.16106233596801758,
      "num_tokens": 37457323.0,
      "step": 17310
    },
    {
      "entropy": 6.413173341751099,
      "epoch": 1.852908127775697,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004662945931698901,
      "loss": 5.9276,
      "mean_token_accuracy": 0.15842394530773163,
      "num_tokens": 37468121.0,
      "step": 17315
    },
    {
      "entropy": 6.486238813400268,
      "epoch": 1.853443201883461,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046627446196842784,
      "loss": 5.9803,
      "mean_token_accuracy": 0.1549091726541519,
      "num_tokens": 37479205.0,
      "step": 17320
    },
    {
      "entropy": 6.501701068878174,
      "epoch": 1.8539782759912247,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004662543252439422,
      "loss": 6.0506,
      "mean_token_accuracy": 0.154263374209404,
      "num_tokens": 37491015.0,
      "step": 17325
    },
    {
      "entropy": 6.507241725921631,
      "epoch": 1.8545133500989888,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046623418299701494,
      "loss": 5.9919,
      "mean_token_accuracy": 0.15429921746253966,
      "num_tokens": 37503078.0,
      "step": 17330
    },
    {
      "entropy": 6.470237112045288,
      "epoch": 1.8550484242067526,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004662140352282274,
      "loss": 5.9485,
      "mean_token_accuracy": 0.16935261934995652,
      "num_tokens": 37514623.0,
      "step": 17335
    },
    {
      "entropy": 6.405385541915893,
      "epoch": 1.8555834983145165,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046619388193816155,
      "loss": 5.8974,
      "mean_token_accuracy": 0.15966669768095015,
      "num_tokens": 37525618.0,
      "step": 17340
    },
    {
      "entropy": 6.456606197357178,
      "epoch": 1.8561185724222806,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046617372312739917,
      "loss": 5.9726,
      "mean_token_accuracy": 0.16194069981575013,
      "num_tokens": 37536023.0,
      "step": 17345
    },
    {
      "entropy": 6.499948596954345,
      "epoch": 1.8566536465300443,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004661535587965224,
      "loss": 5.9842,
      "mean_token_accuracy": 0.15191233456134795,
      "num_tokens": 37545886.0,
      "step": 17350
    },
    {
      "entropy": 6.420600557327271,
      "epoch": 1.8571887206378084,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046613338894611347,
      "loss": 5.9092,
      "mean_token_accuracy": 0.1579471319913864,
      "num_tokens": 37557587.0,
      "step": 17355
    },
    {
      "entropy": 6.351703453063965,
      "epoch": 1.8577237947455723,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004661132135767548,
      "loss": 5.8152,
      "mean_token_accuracy": 0.1742329105734825,
      "num_tokens": 37569379.0,
      "step": 17360
    },
    {
      "entropy": 6.495197391510009,
      "epoch": 1.8582588688533361,
      "grad_norm": 0.953125,
      "learning_rate": 0.000466093032689029,
      "loss": 6.0097,
      "mean_token_accuracy": 0.15547834932804108,
      "num_tokens": 37580478.0,
      "step": 17365
    },
    {
      "entropy": 6.38857069015503,
      "epoch": 1.8587939429611002,
      "grad_norm": 1.0,
      "learning_rate": 0.0004660728462835187,
      "loss": 5.9313,
      "mean_token_accuracy": 0.16494786292314528,
      "num_tokens": 37591506.0,
      "step": 17370
    },
    {
      "entropy": 6.3723536968231205,
      "epoch": 1.8593290170688639,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004660526543608068,
      "loss": 5.7749,
      "mean_token_accuracy": 0.17164039313793183,
      "num_tokens": 37602074.0,
      "step": 17375
    },
    {
      "entropy": 6.4484227180480955,
      "epoch": 1.859864091176628,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004660324569214763,
      "loss": 5.9606,
      "mean_token_accuracy": 0.15657860040664673,
      "num_tokens": 37612180.0,
      "step": 17380
    },
    {
      "entropy": 6.430492305755616,
      "epoch": 1.8603991652843919,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004660122539661106,
      "loss": 5.941,
      "mean_token_accuracy": 0.161201548576355,
      "num_tokens": 37622588.0,
      "step": 17385
    },
    {
      "entropy": 6.465597534179688,
      "epoch": 1.8609342393921557,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004659920454952928,
      "loss": 5.9437,
      "mean_token_accuracy": 0.15979873090982438,
      "num_tokens": 37634323.0,
      "step": 17390
    },
    {
      "entropy": 6.461930274963379,
      "epoch": 1.8614693134999198,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004659718315096065,
      "loss": 5.958,
      "mean_token_accuracy": 0.1588006466627121,
      "num_tokens": 37645816.0,
      "step": 17395
    },
    {
      "entropy": 6.481200695037842,
      "epoch": 1.8620043876076835,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046595161200963545,
      "loss": 5.9908,
      "mean_token_accuracy": 0.15920123159885408,
      "num_tokens": 37656143.0,
      "step": 17400
    },
    {
      "entropy": 6.439121866226197,
      "epoch": 1.8625394617154476,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046593138699596343,
      "loss": 5.8833,
      "mean_token_accuracy": 0.15994448065757752,
      "num_tokens": 37665959.0,
      "step": 17405
    },
    {
      "entropy": 6.412542486190796,
      "epoch": 1.8630745358232115,
      "grad_norm": 1.765625,
      "learning_rate": 0.00046591115646917443,
      "loss": 5.8999,
      "mean_token_accuracy": 0.15889732986688615,
      "num_tokens": 37677176.0,
      "step": 17410
    },
    {
      "entropy": 6.4394136428833,
      "epoch": 1.8636096099309754,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004658909204298526,
      "loss": 5.9398,
      "mean_token_accuracy": 0.16338930130004883,
      "num_tokens": 37687740.0,
      "step": 17415
    },
    {
      "entropy": 6.371346282958984,
      "epoch": 1.8641446840387395,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004658706788785823,
      "loss": 5.8821,
      "mean_token_accuracy": 0.16818764358758925,
      "num_tokens": 37698536.0,
      "step": 17420
    },
    {
      "entropy": 6.463601875305176,
      "epoch": 1.8646797581465033,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046585043181594793,
      "loss": 5.9254,
      "mean_token_accuracy": 0.15881510078907013,
      "num_tokens": 37708795.0,
      "step": 17425
    },
    {
      "entropy": 6.458835077285767,
      "epoch": 1.8652148322542672,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046583017924253426,
      "loss": 5.9263,
      "mean_token_accuracy": 0.16641103327274323,
      "num_tokens": 37719136.0,
      "step": 17430
    },
    {
      "entropy": 6.555261898040771,
      "epoch": 1.8657499063620313,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004658099211589259,
      "loss": 6.0298,
      "mean_token_accuracy": 0.15872676968574523,
      "num_tokens": 37730513.0,
      "step": 17435
    },
    {
      "entropy": 6.417062759399414,
      "epoch": 1.866284980469795,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046578965756570787,
      "loss": 5.9335,
      "mean_token_accuracy": 0.1665346145629883,
      "num_tokens": 37741811.0,
      "step": 17440
    },
    {
      "entropy": 6.3664796352386475,
      "epoch": 1.866820054577559,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046576938846346527,
      "loss": 5.7858,
      "mean_token_accuracy": 0.17328321933746338,
      "num_tokens": 37752213.0,
      "step": 17445
    },
    {
      "entropy": 6.3313905715942385,
      "epoch": 1.867355128685323,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00046574911385278343,
      "loss": 5.7828,
      "mean_token_accuracy": 0.17409576922655107,
      "num_tokens": 37763253.0,
      "step": 17450
    },
    {
      "entropy": 6.429786300659179,
      "epoch": 1.8678902027930868,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046572883373424776,
      "loss": 5.9849,
      "mean_token_accuracy": 0.158577361702919,
      "num_tokens": 37775087.0,
      "step": 17455
    },
    {
      "entropy": 6.455647945404053,
      "epoch": 1.868425276900851,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046570854810844373,
      "loss": 5.9743,
      "mean_token_accuracy": 0.15177940875291823,
      "num_tokens": 37786119.0,
      "step": 17460
    },
    {
      "entropy": 6.503290700912475,
      "epoch": 1.8689603510086146,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046568825697595724,
      "loss": 5.9857,
      "mean_token_accuracy": 0.15619521588087082,
      "num_tokens": 37797482.0,
      "step": 17465
    },
    {
      "entropy": 6.431302213668824,
      "epoch": 1.8694954251163787,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004656679603373741,
      "loss": 5.911,
      "mean_token_accuracy": 0.15908150672912597,
      "num_tokens": 37808067.0,
      "step": 17470
    },
    {
      "entropy": 6.4500470638275145,
      "epoch": 1.8700304992241426,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004656476581932804,
      "loss": 6.0553,
      "mean_token_accuracy": 0.15313214361667632,
      "num_tokens": 37819762.0,
      "step": 17475
    },
    {
      "entropy": 6.421302318572998,
      "epoch": 1.8705655733319064,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004656273505442624,
      "loss": 5.9257,
      "mean_token_accuracy": 0.16384139508008957,
      "num_tokens": 37829510.0,
      "step": 17480
    },
    {
      "entropy": 6.429576587677002,
      "epoch": 1.8711006474396705,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046560703739090633,
      "loss": 6.0077,
      "mean_token_accuracy": 0.15038282424211502,
      "num_tokens": 37839431.0,
      "step": 17485
    },
    {
      "entropy": 6.50909538269043,
      "epoch": 1.8716357215474342,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046558671873379886,
      "loss": 5.9987,
      "mean_token_accuracy": 0.14964407086372375,
      "num_tokens": 37850531.0,
      "step": 17490
    },
    {
      "entropy": 6.528608798980713,
      "epoch": 1.8721707956551983,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004655663945735268,
      "loss": 6.0141,
      "mean_token_accuracy": 0.15310536623001098,
      "num_tokens": 37861964.0,
      "step": 17495
    },
    {
      "entropy": 6.338653326034546,
      "epoch": 1.8727058697629622,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004655460649106768,
      "loss": 5.729,
      "mean_token_accuracy": 0.17541573792696,
      "num_tokens": 37873598.0,
      "step": 17500
    },
    {
      "entropy": 6.4520101070404055,
      "epoch": 1.873240943870726,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046552572974583597,
      "loss": 6.0006,
      "mean_token_accuracy": 0.14675063118338585,
      "num_tokens": 37884643.0,
      "step": 17505
    },
    {
      "entropy": 6.409591102600098,
      "epoch": 1.8737760179784901,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004655053890795914,
      "loss": 5.9802,
      "mean_token_accuracy": 0.1599051386117935,
      "num_tokens": 37896566.0,
      "step": 17510
    },
    {
      "entropy": 6.4921318054199215,
      "epoch": 1.8743110920862538,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046548504291253054,
      "loss": 6.013,
      "mean_token_accuracy": 0.15558438003063202,
      "num_tokens": 37907424.0,
      "step": 17515
    },
    {
      "entropy": 6.484284400939941,
      "epoch": 1.874846166194018,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004654646912452408,
      "loss": 5.9568,
      "mean_token_accuracy": 0.16489457637071608,
      "num_tokens": 37917318.0,
      "step": 17520
    },
    {
      "entropy": 6.417637729644776,
      "epoch": 1.8753812403017818,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004654443340783099,
      "loss": 5.9071,
      "mean_token_accuracy": 0.16302186101675034,
      "num_tokens": 37928521.0,
      "step": 17525
    },
    {
      "entropy": 6.439956140518189,
      "epoch": 1.8759163144095456,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046542397141232554,
      "loss": 5.9847,
      "mean_token_accuracy": 0.16198912560939788,
      "num_tokens": 37939930.0,
      "step": 17530
    },
    {
      "entropy": 6.430138731002808,
      "epoch": 1.8764513885173097,
      "grad_norm": 1.0,
      "learning_rate": 0.0004654036032478759,
      "loss": 5.89,
      "mean_token_accuracy": 0.1618873655796051,
      "num_tokens": 37949690.0,
      "step": 17535
    },
    {
      "entropy": 6.454503011703491,
      "epoch": 1.8769864626250736,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046538322958554885,
      "loss": 5.8989,
      "mean_token_accuracy": 0.16442804485559465,
      "num_tokens": 37962162.0,
      "step": 17540
    },
    {
      "entropy": 6.524032974243164,
      "epoch": 1.8775215367328375,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004653628504259329,
      "loss": 5.9688,
      "mean_token_accuracy": 0.15331066250801087,
      "num_tokens": 37972707.0,
      "step": 17545
    },
    {
      "entropy": 6.391521310806274,
      "epoch": 1.8780566108406014,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046534246576961633,
      "loss": 5.8998,
      "mean_token_accuracy": 0.160857155919075,
      "num_tokens": 37983101.0,
      "step": 17550
    },
    {
      "entropy": 6.313796854019165,
      "epoch": 1.8785916849483653,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046532207561718775,
      "loss": 5.834,
      "mean_token_accuracy": 0.16787783652544022,
      "num_tokens": 37994025.0,
      "step": 17555
    },
    {
      "entropy": 6.411320781707763,
      "epoch": 1.8791267590561294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046530167996923613,
      "loss": 5.9491,
      "mean_token_accuracy": 0.1531756490468979,
      "num_tokens": 38004761.0,
      "step": 17560
    },
    {
      "entropy": 6.492489910125732,
      "epoch": 1.8796618331638932,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004652812788263501,
      "loss": 5.9287,
      "mean_token_accuracy": 0.16238341033458709,
      "num_tokens": 38015129.0,
      "step": 17565
    },
    {
      "entropy": 6.4708099365234375,
      "epoch": 1.8801969072716571,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046526087218911894,
      "loss": 5.9573,
      "mean_token_accuracy": 0.16719554960727692,
      "num_tokens": 38025501.0,
      "step": 17570
    },
    {
      "entropy": 6.454316282272339,
      "epoch": 1.8807319813794212,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046524046005813185,
      "loss": 5.9266,
      "mean_token_accuracy": 0.157777638733387,
      "num_tokens": 38034912.0,
      "step": 17575
    },
    {
      "entropy": 6.393937587738037,
      "epoch": 1.8812670554871849,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004652200424339782,
      "loss": 5.8818,
      "mean_token_accuracy": 0.16032685041427613,
      "num_tokens": 38045417.0,
      "step": 17580
    },
    {
      "entropy": 6.455037546157837,
      "epoch": 1.881802129594949,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004651996193172475,
      "loss": 6.0016,
      "mean_token_accuracy": 0.1510241448879242,
      "num_tokens": 38057033.0,
      "step": 17585
    },
    {
      "entropy": 6.510967779159546,
      "epoch": 1.8823372037027128,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004651791907085296,
      "loss": 5.9332,
      "mean_token_accuracy": 0.16016403660178186,
      "num_tokens": 38067804.0,
      "step": 17590
    },
    {
      "entropy": 6.444961833953857,
      "epoch": 1.8828722778104767,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004651587566084143,
      "loss": 5.9131,
      "mean_token_accuracy": 0.1674808979034424,
      "num_tokens": 38079246.0,
      "step": 17595
    },
    {
      "entropy": 6.432911539077759,
      "epoch": 1.8834073519182408,
      "grad_norm": 0.9375,
      "learning_rate": 0.00046513831701749155,
      "loss": 6.0623,
      "mean_token_accuracy": 0.1542695939540863,
      "num_tokens": 38090628.0,
      "step": 17600
    },
    {
      "entropy": 6.435601091384887,
      "epoch": 1.8839424260260045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046511787193635165,
      "loss": 5.8664,
      "mean_token_accuracy": 0.16674543172121048,
      "num_tokens": 38100848.0,
      "step": 17605
    },
    {
      "entropy": 6.5163774490356445,
      "epoch": 1.8844775001337686,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046509742136558493,
      "loss": 5.916,
      "mean_token_accuracy": 0.16077620387077332,
      "num_tokens": 38110732.0,
      "step": 17610
    },
    {
      "entropy": 6.379998683929443,
      "epoch": 1.8850125742415325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046507696530578185,
      "loss": 5.8985,
      "mean_token_accuracy": 0.17174023985862732,
      "num_tokens": 38121002.0,
      "step": 17615
    },
    {
      "entropy": 6.414114284515381,
      "epoch": 1.8855476483492963,
      "grad_norm": 1.0,
      "learning_rate": 0.0004650565037575331,
      "loss": 5.9616,
      "mean_token_accuracy": 0.16290385350584985,
      "num_tokens": 38132294.0,
      "step": 17620
    },
    {
      "entropy": 6.423145723342896,
      "epoch": 1.8860827224570604,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046503603672142955,
      "loss": 5.8637,
      "mean_token_accuracy": 0.16349444389343262,
      "num_tokens": 38142962.0,
      "step": 17625
    },
    {
      "entropy": 6.516419363021851,
      "epoch": 1.886617796564824,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004650155641980621,
      "loss": 5.9528,
      "mean_token_accuracy": 0.15955508649349212,
      "num_tokens": 38153737.0,
      "step": 17630
    },
    {
      "entropy": 6.479144763946533,
      "epoch": 1.8871528706725882,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046499508618802195,
      "loss": 5.9541,
      "mean_token_accuracy": 0.15738388448953627,
      "num_tokens": 38163897.0,
      "step": 17635
    },
    {
      "entropy": 6.365886402130127,
      "epoch": 1.887687944780352,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004649746026919004,
      "loss": 5.9065,
      "mean_token_accuracy": 0.16960834711790085,
      "num_tokens": 38175123.0,
      "step": 17640
    },
    {
      "entropy": 6.363177013397217,
      "epoch": 1.888223018888116,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046495411371028886,
      "loss": 5.9387,
      "mean_token_accuracy": 0.16893213540315627,
      "num_tokens": 38186100.0,
      "step": 17645
    },
    {
      "entropy": 6.455349111557007,
      "epoch": 1.88875809299588,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046493361924377903,
      "loss": 5.9506,
      "mean_token_accuracy": 0.16183405071496965,
      "num_tokens": 38196385.0,
      "step": 17650
    },
    {
      "entropy": 6.389872407913208,
      "epoch": 1.8892931671036437,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004649131192929626,
      "loss": 5.8982,
      "mean_token_accuracy": 0.15838708132505416,
      "num_tokens": 38207260.0,
      "step": 17655
    },
    {
      "entropy": 6.3908679485321045,
      "epoch": 1.8898282412114078,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004648926138584315,
      "loss": 5.8121,
      "mean_token_accuracy": 0.17005134522914886,
      "num_tokens": 38217537.0,
      "step": 17660
    },
    {
      "entropy": 6.477864789962768,
      "epoch": 1.8903633153191717,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046487210294077794,
      "loss": 5.9154,
      "mean_token_accuracy": 0.1615554451942444,
      "num_tokens": 38228956.0,
      "step": 17665
    },
    {
      "entropy": 6.335651493072509,
      "epoch": 1.8908983894269356,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046485158654059405,
      "loss": 5.8411,
      "mean_token_accuracy": 0.16752301007509232,
      "num_tokens": 38240236.0,
      "step": 17670
    },
    {
      "entropy": 6.436702537536621,
      "epoch": 1.8914334635346997,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004648310646584723,
      "loss": 5.9811,
      "mean_token_accuracy": 0.15702261924743652,
      "num_tokens": 38251739.0,
      "step": 17675
    },
    {
      "entropy": 6.452213430404663,
      "epoch": 1.8919685376424635,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046481053729500516,
      "loss": 5.9254,
      "mean_token_accuracy": 0.15175360888242723,
      "num_tokens": 38262090.0,
      "step": 17680
    },
    {
      "entropy": 6.451668548583984,
      "epoch": 1.8925036117502274,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004647900044507855,
      "loss": 5.9318,
      "mean_token_accuracy": 0.16032156795263292,
      "num_tokens": 38272770.0,
      "step": 17685
    },
    {
      "entropy": 6.419056844711304,
      "epoch": 1.8930386858579913,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004647694661264061,
      "loss": 5.895,
      "mean_token_accuracy": 0.16049622595310212,
      "num_tokens": 38283449.0,
      "step": 17690
    },
    {
      "entropy": 6.424036073684692,
      "epoch": 1.8935737599657552,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046474892232246,
      "loss": 5.9362,
      "mean_token_accuracy": 0.16442353278398514,
      "num_tokens": 38293831.0,
      "step": 17695
    },
    {
      "entropy": 6.449624729156494,
      "epoch": 1.8941088340735193,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046472837303954053,
      "loss": 5.8948,
      "mean_token_accuracy": 0.15501539632678032,
      "num_tokens": 38304782.0,
      "step": 17700
    },
    {
      "entropy": 6.40440731048584,
      "epoch": 1.8946439081812831,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046470781827824087,
      "loss": 5.9017,
      "mean_token_accuracy": 0.1716387927532196,
      "num_tokens": 38316547.0,
      "step": 17705
    },
    {
      "entropy": 6.496830606460572,
      "epoch": 1.895178982289047,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046468725803915464,
      "loss": 5.9508,
      "mean_token_accuracy": 0.1650058001279831,
      "num_tokens": 38329488.0,
      "step": 17710
    },
    {
      "entropy": 6.324218654632569,
      "epoch": 1.8957140563968111,
      "grad_norm": 0.9375,
      "learning_rate": 0.00046466669232287543,
      "loss": 5.8155,
      "mean_token_accuracy": 0.16862730085849761,
      "num_tokens": 38341290.0,
      "step": 17715
    },
    {
      "entropy": 6.410222864151001,
      "epoch": 1.8962491305045748,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004646461211299973,
      "loss": 5.9498,
      "mean_token_accuracy": 0.1581570714712143,
      "num_tokens": 38352445.0,
      "step": 17720
    },
    {
      "entropy": 6.42872462272644,
      "epoch": 1.8967842046123389,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004646255444611139,
      "loss": 5.9507,
      "mean_token_accuracy": 0.16218846440315246,
      "num_tokens": 38363825.0,
      "step": 17725
    },
    {
      "entropy": 6.430000829696655,
      "epoch": 1.8973192787201028,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00046460496231681957,
      "loss": 5.9988,
      "mean_token_accuracy": 0.16433898359537125,
      "num_tokens": 38375454.0,
      "step": 17730
    },
    {
      "entropy": 6.413379907608032,
      "epoch": 1.8978543528278666,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046458437469770866,
      "loss": 5.9758,
      "mean_token_accuracy": 0.1612471505999565,
      "num_tokens": 38385938.0,
      "step": 17735
    },
    {
      "entropy": 6.4481244564056395,
      "epoch": 1.8983894269356307,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046456378160437555,
      "loss": 5.8782,
      "mean_token_accuracy": 0.16309349983930588,
      "num_tokens": 38396143.0,
      "step": 17740
    },
    {
      "entropy": 6.389001512527466,
      "epoch": 1.8989245010433944,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004645431830374149,
      "loss": 5.9818,
      "mean_token_accuracy": 0.14898470640182496,
      "num_tokens": 38407445.0,
      "step": 17745
    },
    {
      "entropy": 6.480663776397705,
      "epoch": 1.8994595751511585,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046452257899742144,
      "loss": 6.0061,
      "mean_token_accuracy": 0.15055964291095733,
      "num_tokens": 38418438.0,
      "step": 17750
    },
    {
      "entropy": 6.427168846130371,
      "epoch": 1.8999946492589224,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046450196948499015,
      "loss": 5.8928,
      "mean_token_accuracy": 0.15883404463529588,
      "num_tokens": 38428821.0,
      "step": 17755
    },
    {
      "entropy": 6.417350101470947,
      "epoch": 1.9005297233666862,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004644813545007161,
      "loss": 5.8486,
      "mean_token_accuracy": 0.1658239707350731,
      "num_tokens": 38439560.0,
      "step": 17760
    },
    {
      "entropy": 6.460046243667603,
      "epoch": 1.9010647974744503,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004644607340451946,
      "loss": 5.9434,
      "mean_token_accuracy": 0.16246250718832017,
      "num_tokens": 38449724.0,
      "step": 17765
    },
    {
      "entropy": 6.454975128173828,
      "epoch": 1.901599871582214,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046444010811902105,
      "loss": 5.8959,
      "mean_token_accuracy": 0.1618331953883171,
      "num_tokens": 38459833.0,
      "step": 17770
    },
    {
      "entropy": 6.367010974884034,
      "epoch": 1.902134945689978,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046441947672279095,
      "loss": 5.9121,
      "mean_token_accuracy": 0.1700920879840851,
      "num_tokens": 38470682.0,
      "step": 17775
    },
    {
      "entropy": 6.437992668151855,
      "epoch": 1.902670019797742,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004643988398571001,
      "loss": 5.9447,
      "mean_token_accuracy": 0.15996199995279312,
      "num_tokens": 38481362.0,
      "step": 17780
    },
    {
      "entropy": 6.436229419708252,
      "epoch": 1.9032050939055059,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004643781975225444,
      "loss": 5.9038,
      "mean_token_accuracy": 0.16439609229564667,
      "num_tokens": 38492941.0,
      "step": 17785
    },
    {
      "entropy": 6.506550931930542,
      "epoch": 1.90374016801327,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046435754971971976,
      "loss": 5.9466,
      "mean_token_accuracy": 0.15639031678438187,
      "num_tokens": 38503823.0,
      "step": 17790
    },
    {
      "entropy": 6.421533012390137,
      "epoch": 1.9042752421210336,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046433689644922256,
      "loss": 5.8767,
      "mean_token_accuracy": 0.16825971901416778,
      "num_tokens": 38514482.0,
      "step": 17795
    },
    {
      "entropy": 6.425713682174683,
      "epoch": 1.9048103162287977,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046431623771164897,
      "loss": 5.9421,
      "mean_token_accuracy": 0.16251826882362366,
      "num_tokens": 38525580.0,
      "step": 17800
    },
    {
      "entropy": 6.369844627380371,
      "epoch": 1.9053453903365616,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004642955735075958,
      "loss": 5.9346,
      "mean_token_accuracy": 0.16432115584611892,
      "num_tokens": 38537025.0,
      "step": 17805
    },
    {
      "entropy": 6.439187812805176,
      "epoch": 1.9058804644443255,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046427490383765943,
      "loss": 5.9022,
      "mean_token_accuracy": 0.16268587112426758,
      "num_tokens": 38548974.0,
      "step": 17810
    },
    {
      "entropy": 6.39447603225708,
      "epoch": 1.9064155385520896,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004642542287024368,
      "loss": 5.7899,
      "mean_token_accuracy": 0.17938645780086518,
      "num_tokens": 38559971.0,
      "step": 17815
    },
    {
      "entropy": 6.420039272308349,
      "epoch": 1.9069506126598534,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000464233548102525,
      "loss": 5.8806,
      "mean_token_accuracy": 0.16273204535245894,
      "num_tokens": 38570326.0,
      "step": 17820
    },
    {
      "entropy": 6.399229288101196,
      "epoch": 1.9074856867676173,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000464212862038521,
      "loss": 5.9006,
      "mean_token_accuracy": 0.16652079224586486,
      "num_tokens": 38580799.0,
      "step": 17825
    },
    {
      "entropy": 6.373348712921143,
      "epoch": 1.9080207608753812,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046419217051102223,
      "loss": 6.0038,
      "mean_token_accuracy": 0.1531835302710533,
      "num_tokens": 38591136.0,
      "step": 17830
    },
    {
      "entropy": 6.437010717391968,
      "epoch": 1.908555834983145,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004641714735206261,
      "loss": 5.8749,
      "mean_token_accuracy": 0.1662692219018936,
      "num_tokens": 38602600.0,
      "step": 17835
    },
    {
      "entropy": 6.419898319244385,
      "epoch": 1.9090909090909092,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046415077106793024,
      "loss": 5.9091,
      "mean_token_accuracy": 0.16158178299665452,
      "num_tokens": 38612641.0,
      "step": 17840
    },
    {
      "entropy": 6.408333015441895,
      "epoch": 1.909625983198673,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004641300631535325,
      "loss": 5.9146,
      "mean_token_accuracy": 0.15945911407470703,
      "num_tokens": 38622873.0,
      "step": 17845
    },
    {
      "entropy": 6.4041060447692875,
      "epoch": 1.910161057306437,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004641093497780307,
      "loss": 5.9093,
      "mean_token_accuracy": 0.16685594767332076,
      "num_tokens": 38633523.0,
      "step": 17850
    },
    {
      "entropy": 6.409204483032227,
      "epoch": 1.910696131414201,
      "grad_norm": 1.0,
      "learning_rate": 0.0004640886309420231,
      "loss": 5.9234,
      "mean_token_accuracy": 0.1561405435204506,
      "num_tokens": 38645635.0,
      "step": 17855
    },
    {
      "entropy": 6.431383466720581,
      "epoch": 1.9112312055219647,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004640679066461078,
      "loss": 5.8832,
      "mean_token_accuracy": 0.16442970633506776,
      "num_tokens": 38656293.0,
      "step": 17860
    },
    {
      "entropy": 6.451780796051025,
      "epoch": 1.9117662796297288,
      "grad_norm": 1.0,
      "learning_rate": 0.0004640471768908832,
      "loss": 6.0092,
      "mean_token_accuracy": 0.15386151969432832,
      "num_tokens": 38667307.0,
      "step": 17865
    },
    {
      "entropy": 6.475836229324341,
      "epoch": 1.9123013537374927,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004640264416769481,
      "loss": 5.8831,
      "mean_token_accuracy": 0.1608160436153412,
      "num_tokens": 38677775.0,
      "step": 17870
    },
    {
      "entropy": 6.399198150634765,
      "epoch": 1.9128364278452565,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046400570100490086,
      "loss": 5.9791,
      "mean_token_accuracy": 0.15859391167759895,
      "num_tokens": 38688652.0,
      "step": 17875
    },
    {
      "entropy": 6.45950984954834,
      "epoch": 1.9133715019530206,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00046398495487534066,
      "loss": 5.9554,
      "mean_token_accuracy": 0.15493216067552568,
      "num_tokens": 38698694.0,
      "step": 17880
    },
    {
      "entropy": 6.501062870025635,
      "epoch": 1.9139065760607843,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004639642032888664,
      "loss": 5.9552,
      "mean_token_accuracy": 0.16079901307821273,
      "num_tokens": 38709864.0,
      "step": 17885
    },
    {
      "entropy": 6.46007227897644,
      "epoch": 1.9144416501685484,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004639434462460774,
      "loss": 5.9558,
      "mean_token_accuracy": 0.15473742932081222,
      "num_tokens": 38721034.0,
      "step": 17890
    },
    {
      "entropy": 6.430035781860352,
      "epoch": 1.9149767242763123,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046392268374757285,
      "loss": 5.9458,
      "mean_token_accuracy": 0.15513525456190108,
      "num_tokens": 38731874.0,
      "step": 17895
    },
    {
      "entropy": 6.4010009765625,
      "epoch": 1.9155117983840761,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004639019157939524,
      "loss": 5.9092,
      "mean_token_accuracy": 0.16017600893974304,
      "num_tokens": 38742447.0,
      "step": 17900
    },
    {
      "entropy": 6.41221604347229,
      "epoch": 1.9160468724918402,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004638811423858157,
      "loss": 5.8302,
      "mean_token_accuracy": 0.16523364782333375,
      "num_tokens": 38754117.0,
      "step": 17905
    },
    {
      "entropy": 6.4230735301971436,
      "epoch": 1.916581946599604,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046386036352376255,
      "loss": 5.9398,
      "mean_token_accuracy": 0.1557144492864609,
      "num_tokens": 38765057.0,
      "step": 17910
    },
    {
      "entropy": 6.420705032348633,
      "epoch": 1.917117020707368,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046383957920839285,
      "loss": 5.886,
      "mean_token_accuracy": 0.16277136504650117,
      "num_tokens": 38775938.0,
      "step": 17915
    },
    {
      "entropy": 6.345596075057983,
      "epoch": 1.9176520948151319,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046381878944030697,
      "loss": 5.8099,
      "mean_token_accuracy": 0.16462095379829406,
      "num_tokens": 38787013.0,
      "step": 17920
    },
    {
      "entropy": 6.425973653793335,
      "epoch": 1.9181871689228958,
      "grad_norm": 1.0,
      "learning_rate": 0.000463797994220105,
      "loss": 5.8482,
      "mean_token_accuracy": 0.16974557638168336,
      "num_tokens": 38798412.0,
      "step": 17925
    },
    {
      "entropy": 6.450766229629517,
      "epoch": 1.9187222430306599,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046377719354838756,
      "loss": 5.9405,
      "mean_token_accuracy": 0.15929118245840074,
      "num_tokens": 38809983.0,
      "step": 17930
    },
    {
      "entropy": 6.439217615127563,
      "epoch": 1.9192573171384235,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046375638742575506,
      "loss": 5.9321,
      "mean_token_accuracy": 0.1615123689174652,
      "num_tokens": 38820418.0,
      "step": 17935
    },
    {
      "entropy": 6.48678297996521,
      "epoch": 1.9197923912461876,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004637355758528085,
      "loss": 5.9459,
      "mean_token_accuracy": 0.16076479256153106,
      "num_tokens": 38831482.0,
      "step": 17940
    },
    {
      "entropy": 6.513797473907471,
      "epoch": 1.9203274653539515,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004637147588301486,
      "loss": 5.9643,
      "mean_token_accuracy": 0.156097112596035,
      "num_tokens": 38842110.0,
      "step": 17945
    },
    {
      "entropy": 6.392921447753906,
      "epoch": 1.9208625394617154,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046369393635837657,
      "loss": 5.8352,
      "mean_token_accuracy": 0.16181326061487197,
      "num_tokens": 38852861.0,
      "step": 17950
    },
    {
      "entropy": 6.422309160232544,
      "epoch": 1.9213976135694795,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046367310843809374,
      "loss": 5.9675,
      "mean_token_accuracy": 0.15539708063006402,
      "num_tokens": 38864067.0,
      "step": 17955
    },
    {
      "entropy": 6.434354639053344,
      "epoch": 1.9219326876772433,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046365227506990137,
      "loss": 5.9342,
      "mean_token_accuracy": 0.1581297591328621,
      "num_tokens": 38875363.0,
      "step": 17960
    },
    {
      "entropy": 6.485410976409912,
      "epoch": 1.9224677617850072,
      "grad_norm": 1.0,
      "learning_rate": 0.0004636314362544011,
      "loss": 5.9473,
      "mean_token_accuracy": 0.16451057344675063,
      "num_tokens": 38886414.0,
      "step": 17965
    },
    {
      "entropy": 6.503412199020386,
      "epoch": 1.923002835892771,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046361059199219457,
      "loss": 6.014,
      "mean_token_accuracy": 0.15617304295301437,
      "num_tokens": 38897708.0,
      "step": 17970
    },
    {
      "entropy": 6.371181297302246,
      "epoch": 1.923537910000535,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004635897422838837,
      "loss": 5.8893,
      "mean_token_accuracy": 0.1663953557610512,
      "num_tokens": 38908683.0,
      "step": 17975
    },
    {
      "entropy": 6.367205190658569,
      "epoch": 1.924072984108299,
      "grad_norm": 0.890625,
      "learning_rate": 0.00046356888713007055,
      "loss": 5.868,
      "mean_token_accuracy": 0.1667026996612549,
      "num_tokens": 38919279.0,
      "step": 17980
    },
    {
      "entropy": 6.490095376968384,
      "epoch": 1.924608058216063,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046354802653135723,
      "loss": 5.9629,
      "mean_token_accuracy": 0.15217874050140381,
      "num_tokens": 38929353.0,
      "step": 17985
    },
    {
      "entropy": 6.428515386581421,
      "epoch": 1.9251431323238268,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004635271604883462,
      "loss": 5.8963,
      "mean_token_accuracy": 0.16209135204553604,
      "num_tokens": 38939296.0,
      "step": 17990
    },
    {
      "entropy": 6.365480995178222,
      "epoch": 1.925678206431591,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046350628900163987,
      "loss": 5.8989,
      "mean_token_accuracy": 0.16429791301488877,
      "num_tokens": 38950149.0,
      "step": 17995
    },
    {
      "entropy": 6.466815853118897,
      "epoch": 1.9262132805393546,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000463485412071841,
      "loss": 5.9669,
      "mean_token_accuracy": 0.15798321217298508,
      "num_tokens": 38961052.0,
      "step": 18000
    },
    {
      "epoch": 1.9262132805393546,
      "eval_entropy": 6.243131867637532,
      "eval_loss": 5.9925689697265625,
      "eval_mean_token_accuracy": 0.16536695748025357,
      "eval_num_tokens": 38961052.0,
      "eval_runtime": 17.719,
      "eval_samples_per_second": 1674.981,
      "eval_steps_per_second": 209.38,
      "step": 18000
    },
    {
      "entropy": 6.468381738662719,
      "epoch": 1.9267483546471187,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046346452969955214,
      "loss": 5.955,
      "mean_token_accuracy": 0.16015141308307648,
      "num_tokens": 38972344.0,
      "step": 18005
    },
    {
      "entropy": 6.448053598403931,
      "epoch": 1.9272834287548826,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004634436418853767,
      "loss": 5.9294,
      "mean_token_accuracy": 0.1534495994448662,
      "num_tokens": 38982877.0,
      "step": 18010
    },
    {
      "entropy": 6.434416246414185,
      "epoch": 1.9278185028626464,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046342274862991745,
      "loss": 5.9216,
      "mean_token_accuracy": 0.16064331829547882,
      "num_tokens": 38994322.0,
      "step": 18015
    },
    {
      "entropy": 6.420562267303467,
      "epoch": 1.9283535769704105,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046340184993377793,
      "loss": 5.8672,
      "mean_token_accuracy": 0.16912854164838792,
      "num_tokens": 39003447.0,
      "step": 18020
    },
    {
      "entropy": 6.421012020111084,
      "epoch": 1.9288886510781742,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046338094579756136,
      "loss": 5.9383,
      "mean_token_accuracy": 0.1553300991654396,
      "num_tokens": 39013760.0,
      "step": 18025
    },
    {
      "entropy": 6.316494369506836,
      "epoch": 1.9294237251859383,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046336003622187143,
      "loss": 5.7472,
      "mean_token_accuracy": 0.1724447190761566,
      "num_tokens": 39025559.0,
      "step": 18030
    },
    {
      "entropy": 6.405377531051636,
      "epoch": 1.9299587992937022,
      "grad_norm": 0.96875,
      "learning_rate": 0.000463339121207312,
      "loss": 5.9161,
      "mean_token_accuracy": 0.155572010576725,
      "num_tokens": 39036004.0,
      "step": 18035
    },
    {
      "entropy": 6.440870904922486,
      "epoch": 1.930493873401466,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004633182007544868,
      "loss": 5.942,
      "mean_token_accuracy": 0.15852470099925994,
      "num_tokens": 39047251.0,
      "step": 18040
    },
    {
      "entropy": 6.447377014160156,
      "epoch": 1.9310289475092302,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046329727486400015,
      "loss": 5.9765,
      "mean_token_accuracy": 0.15784023553133011,
      "num_tokens": 39057255.0,
      "step": 18045
    },
    {
      "entropy": 6.481149005889892,
      "epoch": 1.9315640216169938,
      "grad_norm": 1.0,
      "learning_rate": 0.0004632763435364561,
      "loss": 5.9785,
      "mean_token_accuracy": 0.15898712426424028,
      "num_tokens": 39068186.0,
      "step": 18050
    },
    {
      "entropy": 6.366122579574585,
      "epoch": 1.932099095724758,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046325540677245907,
      "loss": 5.8842,
      "mean_token_accuracy": 0.1620979979634285,
      "num_tokens": 39079173.0,
      "step": 18055
    },
    {
      "entropy": 6.4245435237884525,
      "epoch": 1.9326341698325218,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046323446457261365,
      "loss": 5.9515,
      "mean_token_accuracy": 0.16234587132930756,
      "num_tokens": 39089939.0,
      "step": 18060
    },
    {
      "entropy": 6.403949737548828,
      "epoch": 1.9331692439402857,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004632135169375245,
      "loss": 5.9458,
      "mean_token_accuracy": 0.1642809510231018,
      "num_tokens": 39100464.0,
      "step": 18065
    },
    {
      "entropy": 6.47834324836731,
      "epoch": 1.9337043180480498,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004631925638677965,
      "loss": 5.8973,
      "mean_token_accuracy": 0.1621638223528862,
      "num_tokens": 39111161.0,
      "step": 18070
    },
    {
      "entropy": 6.457088661193848,
      "epoch": 1.9342393921558134,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046317160536403463,
      "loss": 5.9092,
      "mean_token_accuracy": 0.1670028418302536,
      "num_tokens": 39122469.0,
      "step": 18075
    },
    {
      "entropy": 6.3874267578125,
      "epoch": 1.9347744662635775,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046315064142684416,
      "loss": 5.8798,
      "mean_token_accuracy": 0.16602083444595336,
      "num_tokens": 39131871.0,
      "step": 18080
    },
    {
      "entropy": 6.400062417984008,
      "epoch": 1.9353095403713414,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004631296720568304,
      "loss": 5.9304,
      "mean_token_accuracy": 0.15639459788799287,
      "num_tokens": 39143061.0,
      "step": 18085
    },
    {
      "entropy": 6.393586111068726,
      "epoch": 1.9358446144791053,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004631086972545987,
      "loss": 5.9445,
      "mean_token_accuracy": 0.16730112284421922,
      "num_tokens": 39152452.0,
      "step": 18090
    },
    {
      "entropy": 6.361050701141357,
      "epoch": 1.9363796885868694,
      "grad_norm": 1.0,
      "learning_rate": 0.0004630877170207548,
      "loss": 5.8897,
      "mean_token_accuracy": 0.16431441009044648,
      "num_tokens": 39163196.0,
      "step": 18095
    },
    {
      "entropy": 6.341789484024048,
      "epoch": 1.9369147626946333,
      "grad_norm": 1.0,
      "learning_rate": 0.0004630667313559045,
      "loss": 5.8839,
      "mean_token_accuracy": 0.1596880689263344,
      "num_tokens": 39173958.0,
      "step": 18100
    },
    {
      "entropy": 6.435337066650391,
      "epoch": 1.9374498368023971,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046304574026065376,
      "loss": 5.911,
      "mean_token_accuracy": 0.16723097637295722,
      "num_tokens": 39184483.0,
      "step": 18105
    },
    {
      "entropy": 6.434160661697388,
      "epoch": 1.937984910910161,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004630247437356087,
      "loss": 5.9471,
      "mean_token_accuracy": 0.158493971824646,
      "num_tokens": 39196500.0,
      "step": 18110
    },
    {
      "entropy": 6.478815364837646,
      "epoch": 1.9385199850179249,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046300374178137566,
      "loss": 5.9936,
      "mean_token_accuracy": 0.15307266861200333,
      "num_tokens": 39207492.0,
      "step": 18115
    },
    {
      "entropy": 6.425041437149048,
      "epoch": 1.939055059125689,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004629827343985609,
      "loss": 5.9213,
      "mean_token_accuracy": 0.16120508313179016,
      "num_tokens": 39218204.0,
      "step": 18120
    },
    {
      "entropy": 6.453165435791016,
      "epoch": 1.9395901332334529,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046296172158777116,
      "loss": 5.9672,
      "mean_token_accuracy": 0.15995916426181794,
      "num_tokens": 39229646.0,
      "step": 18125
    },
    {
      "entropy": 6.41486496925354,
      "epoch": 1.9401252073412167,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004629407033496131,
      "loss": 5.9229,
      "mean_token_accuracy": 0.16224279552698134,
      "num_tokens": 39240151.0,
      "step": 18130
    },
    {
      "entropy": 6.476422262191773,
      "epoch": 1.9406602814489808,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004629196796846936,
      "loss": 5.9795,
      "mean_token_accuracy": 0.16061938405036927,
      "num_tokens": 39251571.0,
      "step": 18135
    },
    {
      "entropy": 6.437278699874878,
      "epoch": 1.9411953555567445,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046289865059361983,
      "loss": 5.9076,
      "mean_token_accuracy": 0.1621767744421959,
      "num_tokens": 39262937.0,
      "step": 18140
    },
    {
      "entropy": 6.439812707901001,
      "epoch": 1.9417304296645086,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046287761607699887,
      "loss": 5.9923,
      "mean_token_accuracy": 0.1553113877773285,
      "num_tokens": 39274354.0,
      "step": 18145
    },
    {
      "entropy": 6.436888933181763,
      "epoch": 1.9422655037722725,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004628565761354382,
      "loss": 5.8731,
      "mean_token_accuracy": 0.16670443415641784,
      "num_tokens": 39284314.0,
      "step": 18150
    },
    {
      "entropy": 6.399976444244385,
      "epoch": 1.9428005778800364,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046283553076954527,
      "loss": 5.9006,
      "mean_token_accuracy": 0.1616524264216423,
      "num_tokens": 39295516.0,
      "step": 18155
    },
    {
      "entropy": 6.436187028884888,
      "epoch": 1.9433356519878004,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004628144799799277,
      "loss": 5.9088,
      "mean_token_accuracy": 0.1612467110157013,
      "num_tokens": 39307221.0,
      "step": 18160
    },
    {
      "entropy": 6.402547979354859,
      "epoch": 1.943870726095564,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004627934237671935,
      "loss": 5.9426,
      "mean_token_accuracy": 0.15756385177373886,
      "num_tokens": 39318448.0,
      "step": 18165
    },
    {
      "entropy": 6.36914029121399,
      "epoch": 1.9444058002033282,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046277236213195064,
      "loss": 5.8344,
      "mean_token_accuracy": 0.166275230050087,
      "num_tokens": 39329384.0,
      "step": 18170
    },
    {
      "entropy": 6.427836084365845,
      "epoch": 1.944940874311092,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046275129507480705,
      "loss": 5.9733,
      "mean_token_accuracy": 0.15605104118585586,
      "num_tokens": 39340342.0,
      "step": 18175
    },
    {
      "entropy": 6.328458309173584,
      "epoch": 1.945475948418856,
      "grad_norm": 1.125,
      "learning_rate": 0.0004627302225963713,
      "loss": 5.8286,
      "mean_token_accuracy": 0.1692350059747696,
      "num_tokens": 39350825.0,
      "step": 18180
    },
    {
      "entropy": 6.508645391464233,
      "epoch": 1.94601102252662,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046270914469725174,
      "loss": 5.9643,
      "mean_token_accuracy": 0.15831917226314546,
      "num_tokens": 39361986.0,
      "step": 18185
    },
    {
      "entropy": 6.412636041641235,
      "epoch": 1.9465460966343837,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046268806137805704,
      "loss": 5.8599,
      "mean_token_accuracy": 0.16053353995084763,
      "num_tokens": 39372440.0,
      "step": 18190
    },
    {
      "entropy": 6.370061349868775,
      "epoch": 1.9470811707421478,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004626669726393958,
      "loss": 5.8976,
      "mean_token_accuracy": 0.1725822299718857,
      "num_tokens": 39383455.0,
      "step": 18195
    },
    {
      "entropy": 6.384465074539184,
      "epoch": 1.9476162448499117,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004626458784818772,
      "loss": 5.9061,
      "mean_token_accuracy": 0.15797876119613646,
      "num_tokens": 39393511.0,
      "step": 18200
    },
    {
      "entropy": 6.399247550964356,
      "epoch": 1.9481513189576756,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046262477890611026,
      "loss": 5.9119,
      "mean_token_accuracy": 0.16168417334556578,
      "num_tokens": 39404049.0,
      "step": 18205
    },
    {
      "entropy": 6.4510139465332035,
      "epoch": 1.9486863930654397,
      "grad_norm": 1.0,
      "learning_rate": 0.00046260367391270414,
      "loss": 5.8948,
      "mean_token_accuracy": 0.1647850379347801,
      "num_tokens": 39414771.0,
      "step": 18210
    },
    {
      "entropy": 6.377371788024902,
      "epoch": 1.9492214671732033,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046258256350226835,
      "loss": 5.8929,
      "mean_token_accuracy": 0.16534430533647537,
      "num_tokens": 39424869.0,
      "step": 18215
    },
    {
      "entropy": 6.417412185668946,
      "epoch": 1.9497565412809674,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004625614476754123,
      "loss": 5.963,
      "mean_token_accuracy": 0.15685632973909377,
      "num_tokens": 39435442.0,
      "step": 18220
    },
    {
      "entropy": 6.439911222457885,
      "epoch": 1.9502916153887313,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046254032643274596,
      "loss": 5.8934,
      "mean_token_accuracy": 0.16704519242048263,
      "num_tokens": 39446704.0,
      "step": 18225
    },
    {
      "entropy": 6.391701507568359,
      "epoch": 1.9508266894964952,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00046251919977487887,
      "loss": 5.9094,
      "mean_token_accuracy": 0.16541609168052673,
      "num_tokens": 39458079.0,
      "step": 18230
    },
    {
      "entropy": 6.415241050720215,
      "epoch": 1.9513617636042593,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046249806770242135,
      "loss": 5.8937,
      "mean_token_accuracy": 0.16578327417373656,
      "num_tokens": 39469867.0,
      "step": 18235
    },
    {
      "entropy": 6.412796068191528,
      "epoch": 1.9518968377120232,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004624769302159834,
      "loss": 5.9121,
      "mean_token_accuracy": 0.15946312099695206,
      "num_tokens": 39481138.0,
      "step": 18240
    },
    {
      "entropy": 6.4097757816314695,
      "epoch": 1.952431911819787,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004624557873161755,
      "loss": 5.8655,
      "mean_token_accuracy": 0.16274818778038025,
      "num_tokens": 39492066.0,
      "step": 18245
    },
    {
      "entropy": 6.477092742919922,
      "epoch": 1.952966985927551,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004624346390036081,
      "loss": 5.9337,
      "mean_token_accuracy": 0.1626860648393631,
      "num_tokens": 39502576.0,
      "step": 18250
    },
    {
      "entropy": 6.381869220733643,
      "epoch": 1.9535020600353148,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004624134852788918,
      "loss": 5.8856,
      "mean_token_accuracy": 0.16135708540678023,
      "num_tokens": 39513373.0,
      "step": 18255
    },
    {
      "entropy": 6.446647453308105,
      "epoch": 1.954037134143079,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004623923261426374,
      "loss": 5.9527,
      "mean_token_accuracy": 0.1599346488714218,
      "num_tokens": 39523970.0,
      "step": 18260
    },
    {
      "entropy": 6.429325103759766,
      "epoch": 1.9545722082508428,
      "grad_norm": 1.015625,
      "learning_rate": 0.000462371161595456,
      "loss": 5.9345,
      "mean_token_accuracy": 0.1644115775823593,
      "num_tokens": 39533893.0,
      "step": 18265
    },
    {
      "entropy": 6.384197473526001,
      "epoch": 1.9551072823586066,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004623499916379586,
      "loss": 5.8999,
      "mean_token_accuracy": 0.15987617075443267,
      "num_tokens": 39544276.0,
      "step": 18270
    },
    {
      "entropy": 6.41569333076477,
      "epoch": 1.9556423564663707,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004623288162707565,
      "loss": 5.8812,
      "mean_token_accuracy": 0.16483618766069413,
      "num_tokens": 39554443.0,
      "step": 18275
    },
    {
      "entropy": 6.431766939163208,
      "epoch": 1.9561774305741344,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046230763549446116,
      "loss": 5.8557,
      "mean_token_accuracy": 0.1647702068090439,
      "num_tokens": 39564508.0,
      "step": 18280
    },
    {
      "entropy": 6.4850257396697994,
      "epoch": 1.9567125046818985,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004622864493096842,
      "loss": 5.9369,
      "mean_token_accuracy": 0.15621879696846008,
      "num_tokens": 39574757.0,
      "step": 18285
    },
    {
      "entropy": 6.454825305938721,
      "epoch": 1.9572475787896624,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046226525771703733,
      "loss": 6.0012,
      "mean_token_accuracy": 0.15614123940467833,
      "num_tokens": 39585911.0,
      "step": 18290
    },
    {
      "entropy": 6.463834238052368,
      "epoch": 1.9577826528974263,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004622440607171324,
      "loss": 5.9034,
      "mean_token_accuracy": 0.16386361569166183,
      "num_tokens": 39597443.0,
      "step": 18295
    },
    {
      "entropy": 6.458410215377808,
      "epoch": 1.9583177270051904,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046222285831058155,
      "loss": 5.9691,
      "mean_token_accuracy": 0.15887991189956666,
      "num_tokens": 39608693.0,
      "step": 18300
    },
    {
      "entropy": 6.425221729278564,
      "epoch": 1.958852801112954,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000462201650497997,
      "loss": 5.9573,
      "mean_token_accuracy": 0.16053914725780488,
      "num_tokens": 39620002.0,
      "step": 18305
    },
    {
      "entropy": 6.420577335357666,
      "epoch": 1.9593878752207181,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046218043727999105,
      "loss": 5.9103,
      "mean_token_accuracy": 0.16786695271730423,
      "num_tokens": 39629856.0,
      "step": 18310
    },
    {
      "entropy": 6.348625516891479,
      "epoch": 1.959922949328482,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046215921865717637,
      "loss": 5.8863,
      "mean_token_accuracy": 0.16581773310899733,
      "num_tokens": 39640418.0,
      "step": 18315
    },
    {
      "entropy": 6.480826663970947,
      "epoch": 1.9604580234362459,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004621379946301655,
      "loss": 5.9864,
      "mean_token_accuracy": 0.15514304488897324,
      "num_tokens": 39652871.0,
      "step": 18320
    },
    {
      "entropy": 6.436758327484131,
      "epoch": 1.96099309754401,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004621167651995713,
      "loss": 5.894,
      "mean_token_accuracy": 0.16388921737670897,
      "num_tokens": 39663864.0,
      "step": 18325
    },
    {
      "entropy": 6.360592937469482,
      "epoch": 1.9615281716517736,
      "grad_norm": 1.0,
      "learning_rate": 0.0004620955303660068,
      "loss": 5.8912,
      "mean_token_accuracy": 0.169964899122715,
      "num_tokens": 39675835.0,
      "step": 18330
    },
    {
      "entropy": 6.484938144683838,
      "epoch": 1.9620632457595377,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046207429013008525,
      "loss": 5.9622,
      "mean_token_accuracy": 0.15777293145656585,
      "num_tokens": 39686799.0,
      "step": 18335
    },
    {
      "entropy": 6.466398429870606,
      "epoch": 1.9625983198673016,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046205304449241976,
      "loss": 5.9142,
      "mean_token_accuracy": 0.16187693178653717,
      "num_tokens": 39697564.0,
      "step": 18340
    },
    {
      "entropy": 6.4050215721130375,
      "epoch": 1.9631333939750655,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004620317934536239,
      "loss": 5.9221,
      "mean_token_accuracy": 0.1610904037952423,
      "num_tokens": 39708064.0,
      "step": 18345
    },
    {
      "entropy": 6.359973001480102,
      "epoch": 1.9636684680828296,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004620105370143112,
      "loss": 5.8066,
      "mean_token_accuracy": 0.17131211012601852,
      "num_tokens": 39718076.0,
      "step": 18350
    },
    {
      "entropy": 6.451185178756714,
      "epoch": 1.9642035421905932,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004619892751750957,
      "loss": 5.9039,
      "mean_token_accuracy": 0.16007694900035857,
      "num_tokens": 39729532.0,
      "step": 18355
    },
    {
      "entropy": 6.491407966613769,
      "epoch": 1.9647386162983573,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046196800793659116,
      "loss": 5.9031,
      "mean_token_accuracy": 0.1640763595700264,
      "num_tokens": 39740372.0,
      "step": 18360
    },
    {
      "entropy": 6.3663853168487545,
      "epoch": 1.9652736904061212,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004619467352994115,
      "loss": 5.7757,
      "mean_token_accuracy": 0.1707020252943039,
      "num_tokens": 39751283.0,
      "step": 18365
    },
    {
      "entropy": 6.393907976150513,
      "epoch": 1.965808764513885,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004619254572641712,
      "loss": 6.0038,
      "mean_token_accuracy": 0.15818903595209122,
      "num_tokens": 39763295.0,
      "step": 18370
    },
    {
      "entropy": 6.414424133300781,
      "epoch": 1.9663438386216492,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046190417383148465,
      "loss": 5.9185,
      "mean_token_accuracy": 0.15921248495578766,
      "num_tokens": 39773849.0,
      "step": 18375
    },
    {
      "entropy": 6.356018447875977,
      "epoch": 1.966878912729413,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004618828850019663,
      "loss": 5.7462,
      "mean_token_accuracy": 0.17030975520610808,
      "num_tokens": 39784366.0,
      "step": 18380
    },
    {
      "entropy": 6.400374555587769,
      "epoch": 1.967413986837177,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046186159077623093,
      "loss": 5.8657,
      "mean_token_accuracy": 0.16042062044143676,
      "num_tokens": 39794226.0,
      "step": 18385
    },
    {
      "entropy": 6.359219694137574,
      "epoch": 1.9679490609449408,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004618402911548933,
      "loss": 5.885,
      "mean_token_accuracy": 0.16302887871861457,
      "num_tokens": 39805557.0,
      "step": 18390
    },
    {
      "entropy": 6.412019348144531,
      "epoch": 1.9684841350527047,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004618189861385686,
      "loss": 5.8529,
      "mean_token_accuracy": 0.1684012606739998,
      "num_tokens": 39816090.0,
      "step": 18395
    },
    {
      "entropy": 6.472478771209717,
      "epoch": 1.9690192091604688,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004617976757278719,
      "loss": 5.9742,
      "mean_token_accuracy": 0.15239567160606385,
      "num_tokens": 39826662.0,
      "step": 18400
    },
    {
      "entropy": 6.464862489700318,
      "epoch": 1.9695542832682327,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046177635992341855,
      "loss": 5.9342,
      "mean_token_accuracy": 0.15821207463741302,
      "num_tokens": 39837805.0,
      "step": 18405
    },
    {
      "entropy": 6.4902912139892575,
      "epoch": 1.9700893573759966,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004617550387258241,
      "loss": 5.9412,
      "mean_token_accuracy": 0.15482579916715622,
      "num_tokens": 39848052.0,
      "step": 18410
    },
    {
      "entropy": 6.459542512893677,
      "epoch": 1.9706244314837607,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004617337121357041,
      "loss": 5.9386,
      "mean_token_accuracy": 0.1576685905456543,
      "num_tokens": 39859448.0,
      "step": 18415
    },
    {
      "entropy": 6.435079956054688,
      "epoch": 1.9711595055915243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004617123801536744,
      "loss": 5.8929,
      "mean_token_accuracy": 0.16609100848436356,
      "num_tokens": 39870053.0,
      "step": 18420
    },
    {
      "entropy": 6.470058107376099,
      "epoch": 1.9716945796992884,
      "grad_norm": 1.125,
      "learning_rate": 0.000461691042780351,
      "loss": 6.0046,
      "mean_token_accuracy": 0.1557839535176754,
      "num_tokens": 39881547.0,
      "step": 18425
    },
    {
      "entropy": 6.456443405151367,
      "epoch": 1.9722296538070523,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046166970001635,
      "loss": 5.984,
      "mean_token_accuracy": 0.16228879243135452,
      "num_tokens": 39892974.0,
      "step": 18430
    },
    {
      "entropy": 6.449027967453003,
      "epoch": 1.9727647279148162,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046164835186228757,
      "loss": 5.9467,
      "mean_token_accuracy": 0.16461389213800431,
      "num_tokens": 39904210.0,
      "step": 18435
    },
    {
      "entropy": 6.400702333450317,
      "epoch": 1.9732998020225803,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00046162699831878026,
      "loss": 5.9054,
      "mean_token_accuracy": 0.16400906443595886,
      "num_tokens": 39915464.0,
      "step": 18440
    },
    {
      "entropy": 6.387275552749633,
      "epoch": 1.973834876130344,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004616056393864446,
      "loss": 5.9331,
      "mean_token_accuracy": 0.16192064583301544,
      "num_tokens": 39927117.0,
      "step": 18445
    },
    {
      "entropy": 6.474756336212158,
      "epoch": 1.974369950238108,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004615842750658972,
      "loss": 5.9349,
      "mean_token_accuracy": 0.16186013519763948,
      "num_tokens": 39937363.0,
      "step": 18450
    },
    {
      "entropy": 6.369938135147095,
      "epoch": 1.974905024345872,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004615629053577552,
      "loss": 5.8217,
      "mean_token_accuracy": 0.17046794444322586,
      "num_tokens": 39948934.0,
      "step": 18455
    },
    {
      "entropy": 6.387050199508667,
      "epoch": 1.9754400984536358,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004615415302626355,
      "loss": 5.88,
      "mean_token_accuracy": 0.1656728431582451,
      "num_tokens": 39959435.0,
      "step": 18460
    },
    {
      "entropy": 6.401864051818848,
      "epoch": 1.9759751725613999,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046152014978115535,
      "loss": 5.9335,
      "mean_token_accuracy": 0.16058844178915024,
      "num_tokens": 39969006.0,
      "step": 18465
    },
    {
      "entropy": 6.436132860183716,
      "epoch": 1.9765102466691635,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000461498763913932,
      "loss": 5.8787,
      "mean_token_accuracy": 0.15883197635412216,
      "num_tokens": 39979264.0,
      "step": 18470
    },
    {
      "entropy": 6.502075862884522,
      "epoch": 1.9770453207769276,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046147737266158314,
      "loss": 6.017,
      "mean_token_accuracy": 0.15477853119373322,
      "num_tokens": 39989791.0,
      "step": 18475
    },
    {
      "entropy": 6.444992399215698,
      "epoch": 1.9775803948846915,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004614559760247263,
      "loss": 5.8894,
      "mean_token_accuracy": 0.15949317812919617,
      "num_tokens": 39999591.0,
      "step": 18480
    },
    {
      "entropy": 6.396151304244995,
      "epoch": 1.9781154689924554,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004614345740039794,
      "loss": 5.997,
      "mean_token_accuracy": 0.15930678248405455,
      "num_tokens": 40010811.0,
      "step": 18485
    },
    {
      "entropy": 6.435017490386963,
      "epoch": 1.9786505431002195,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046141316659996037,
      "loss": 5.9675,
      "mean_token_accuracy": 0.15814695209264756,
      "num_tokens": 40022190.0,
      "step": 18490
    },
    {
      "entropy": 6.431499052047729,
      "epoch": 1.9791856172079834,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046139175381328725,
      "loss": 5.9124,
      "mean_token_accuracy": 0.15724365264177323,
      "num_tokens": 40033905.0,
      "step": 18495
    },
    {
      "entropy": 6.458973217010498,
      "epoch": 1.9797206913157472,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004613703356445786,
      "loss": 5.9812,
      "mean_token_accuracy": 0.15946464091539383,
      "num_tokens": 40044923.0,
      "step": 18500
    },
    {
      "entropy": 6.440467786788941,
      "epoch": 1.9802557654235111,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004613489120944526,
      "loss": 5.8757,
      "mean_token_accuracy": 0.16418402940034865,
      "num_tokens": 40054988.0,
      "step": 18505
    },
    {
      "entropy": 6.479482173919678,
      "epoch": 1.980790839531275,
      "grad_norm": 1.0,
      "learning_rate": 0.0004613274831635279,
      "loss": 5.9893,
      "mean_token_accuracy": 0.1530245989561081,
      "num_tokens": 40065565.0,
      "step": 18510
    },
    {
      "entropy": 6.427524089813232,
      "epoch": 1.981325913639039,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046130604885242333,
      "loss": 5.9626,
      "mean_token_accuracy": 0.15988202393054962,
      "num_tokens": 40076282.0,
      "step": 18515
    },
    {
      "entropy": 6.406351566314697,
      "epoch": 1.981860987746803,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004612846091617578,
      "loss": 5.8938,
      "mean_token_accuracy": 0.16580274999141692,
      "num_tokens": 40087546.0,
      "step": 18520
    },
    {
      "entropy": 6.448495388031006,
      "epoch": 1.9823960618545668,
      "grad_norm": 1.0,
      "learning_rate": 0.0004612631640921504,
      "loss": 5.9655,
      "mean_token_accuracy": 0.16052975058555602,
      "num_tokens": 40097937.0,
      "step": 18525
    },
    {
      "entropy": 6.440968656539917,
      "epoch": 1.982931135962331,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046124171364422025,
      "loss": 5.9134,
      "mean_token_accuracy": 0.16766863614320754,
      "num_tokens": 40108639.0,
      "step": 18530
    },
    {
      "entropy": 6.395347023010254,
      "epoch": 1.9834662100700946,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004612202578185868,
      "loss": 5.9108,
      "mean_token_accuracy": 0.16512931734323502,
      "num_tokens": 40119533.0,
      "step": 18535
    },
    {
      "entropy": 6.451768445968628,
      "epoch": 1.9840012841778587,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004611987966158695,
      "loss": 5.9338,
      "mean_token_accuracy": 0.1607542634010315,
      "num_tokens": 40131176.0,
      "step": 18540
    },
    {
      "entropy": 6.4383711338043215,
      "epoch": 1.9845363582856226,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046117733003668825,
      "loss": 5.9065,
      "mean_token_accuracy": 0.16346924006938934,
      "num_tokens": 40143293.0,
      "step": 18545
    },
    {
      "entropy": 6.443876886367798,
      "epoch": 1.9850714323933865,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004611558580816628,
      "loss": 5.8746,
      "mean_token_accuracy": 0.15765966922044755,
      "num_tokens": 40155312.0,
      "step": 18550
    },
    {
      "entropy": 6.445737409591675,
      "epoch": 1.9856065065011506,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000461134380751413,
      "loss": 5.918,
      "mean_token_accuracy": 0.16174044013023375,
      "num_tokens": 40166261.0,
      "step": 18555
    },
    {
      "entropy": 6.3754926204681395,
      "epoch": 1.9861415806089142,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046111289804655906,
      "loss": 5.8742,
      "mean_token_accuracy": 0.1706941619515419,
      "num_tokens": 40176641.0,
      "step": 18560
    },
    {
      "entropy": 6.43683648109436,
      "epoch": 1.9866766547166783,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004610914099677214,
      "loss": 5.9557,
      "mean_token_accuracy": 0.15996893644332885,
      "num_tokens": 40187784.0,
      "step": 18565
    },
    {
      "entropy": 6.415470552444458,
      "epoch": 1.9872117288244422,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004610699165155204,
      "loss": 5.9458,
      "mean_token_accuracy": 0.1585106372833252,
      "num_tokens": 40198219.0,
      "step": 18570
    },
    {
      "entropy": 6.406395816802979,
      "epoch": 1.987746802932206,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046104841769057677,
      "loss": 5.9152,
      "mean_token_accuracy": 0.15870713740587233,
      "num_tokens": 40208589.0,
      "step": 18575
    },
    {
      "entropy": 6.457934904098511,
      "epoch": 1.9882818770399702,
      "grad_norm": 1.125,
      "learning_rate": 0.0004610269134935112,
      "loss": 5.8816,
      "mean_token_accuracy": 0.1650537759065628,
      "num_tokens": 40219587.0,
      "step": 18580
    },
    {
      "entropy": 6.448610067367554,
      "epoch": 1.9888169511477338,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004610054039249446,
      "loss": 5.9142,
      "mean_token_accuracy": 0.16151822358369827,
      "num_tokens": 40230681.0,
      "step": 18585
    },
    {
      "entropy": 6.3264251232147215,
      "epoch": 1.989352025255498,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046098388898549816,
      "loss": 5.8628,
      "mean_token_accuracy": 0.16871621161699296,
      "num_tokens": 40241356.0,
      "step": 18590
    },
    {
      "entropy": 6.42724723815918,
      "epoch": 1.9898870993632618,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046096236867579306,
      "loss": 5.9018,
      "mean_token_accuracy": 0.16487225592136384,
      "num_tokens": 40251652.0,
      "step": 18595
    },
    {
      "entropy": 6.412282419204712,
      "epoch": 1.9904221734710257,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046094084299645065,
      "loss": 5.8783,
      "mean_token_accuracy": 0.1667523443698883,
      "num_tokens": 40262819.0,
      "step": 18600
    },
    {
      "entropy": 6.459940004348755,
      "epoch": 1.9909572475787898,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046091931194809253,
      "loss": 5.9872,
      "mean_token_accuracy": 0.15552957653999328,
      "num_tokens": 40274314.0,
      "step": 18605
    },
    {
      "entropy": 6.481340646743774,
      "epoch": 1.9914923216865534,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004608977755313404,
      "loss": 5.9213,
      "mean_token_accuracy": 0.1616710126399994,
      "num_tokens": 40284733.0,
      "step": 18610
    },
    {
      "entropy": 6.417234802246094,
      "epoch": 1.9920273957943175,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004608762337468162,
      "loss": 5.9061,
      "mean_token_accuracy": 0.1637854278087616,
      "num_tokens": 40295484.0,
      "step": 18615
    },
    {
      "entropy": 6.456670188903809,
      "epoch": 1.9925624699020814,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004608546865951418,
      "loss": 5.8798,
      "mean_token_accuracy": 0.16493650525808334,
      "num_tokens": 40305470.0,
      "step": 18620
    },
    {
      "entropy": 6.481755208969116,
      "epoch": 1.9930975440098453,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046083313407693945,
      "loss": 6.0226,
      "mean_token_accuracy": 0.15443455278873444,
      "num_tokens": 40316055.0,
      "step": 18625
    },
    {
      "entropy": 6.453444099426269,
      "epoch": 1.9936326181176094,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004608115761928315,
      "loss": 5.9239,
      "mean_token_accuracy": 0.16142624616622925,
      "num_tokens": 40326689.0,
      "step": 18630
    },
    {
      "entropy": 6.417324209213257,
      "epoch": 1.9941676922253733,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046079001294344036,
      "loss": 5.9148,
      "mean_token_accuracy": 0.1604055419564247,
      "num_tokens": 40338511.0,
      "step": 18635
    },
    {
      "entropy": 6.408674049377441,
      "epoch": 1.9947027663331371,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004607684443293887,
      "loss": 5.904,
      "mean_token_accuracy": 0.15971864014863968,
      "num_tokens": 40350395.0,
      "step": 18640
    },
    {
      "entropy": 6.44920597076416,
      "epoch": 1.995237840440901,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004607468703512993,
      "loss": 5.9018,
      "mean_token_accuracy": 0.1688707321882248,
      "num_tokens": 40360166.0,
      "step": 18645
    },
    {
      "entropy": 6.380733633041382,
      "epoch": 1.995772914548665,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046072529100979514,
      "loss": 5.916,
      "mean_token_accuracy": 0.15737992525100708,
      "num_tokens": 40370655.0,
      "step": 18650
    },
    {
      "entropy": 6.429322338104248,
      "epoch": 1.996307988656429,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004607037063054993,
      "loss": 5.8874,
      "mean_token_accuracy": 0.16440557986497878,
      "num_tokens": 40381502.0,
      "step": 18655
    },
    {
      "entropy": 6.535812473297119,
      "epoch": 1.9968430627641929,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000460682116239035,
      "loss": 5.974,
      "mean_token_accuracy": 0.1560654327273369,
      "num_tokens": 40391815.0,
      "step": 18660
    },
    {
      "entropy": 6.4074688911437985,
      "epoch": 1.9973781368719568,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046066052081102576,
      "loss": 5.9428,
      "mean_token_accuracy": 0.1637543708086014,
      "num_tokens": 40403306.0,
      "step": 18665
    },
    {
      "entropy": 6.381930828094482,
      "epoch": 1.9979132109797209,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046063892002209505,
      "loss": 5.8857,
      "mean_token_accuracy": 0.1701988011598587,
      "num_tokens": 40414210.0,
      "step": 18670
    },
    {
      "entropy": 6.295237874984741,
      "epoch": 1.9984482850874845,
      "grad_norm": 1.0,
      "learning_rate": 0.0004606173138728666,
      "loss": 5.8468,
      "mean_token_accuracy": 0.17813299596309662,
      "num_tokens": 40424368.0,
      "step": 18675
    },
    {
      "entropy": 6.416673278808593,
      "epoch": 1.9989833591952486,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004605957023639643,
      "loss": 5.9337,
      "mean_token_accuracy": 0.15823312997817993,
      "num_tokens": 40436252.0,
      "step": 18680
    },
    {
      "entropy": 6.470952749252319,
      "epoch": 1.9995184333030125,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046057408549601223,
      "loss": 5.9415,
      "mean_token_accuracy": 0.1642047330737114,
      "num_tokens": 40447256.0,
      "step": 18685
    },
    {
      "entropy": 6.449306858910455,
      "epoch": 2.0,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004605524632696344,
      "loss": 5.8908,
      "mean_token_accuracy": 0.16257203453116947,
      "num_tokens": 40456516.0,
      "step": 18690
    },
    {
      "entropy": 6.432154417037964,
      "epoch": 2.000535074107764,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004605308356854553,
      "loss": 5.8029,
      "mean_token_accuracy": 0.16607019305229187,
      "num_tokens": 40467563.0,
      "step": 18695
    },
    {
      "entropy": 6.405585765838623,
      "epoch": 2.0010701482155278,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004605092027440994,
      "loss": 5.873,
      "mean_token_accuracy": 0.16990351378917695,
      "num_tokens": 40478202.0,
      "step": 18700
    },
    {
      "entropy": 6.441783094406128,
      "epoch": 2.001605222323292,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004604875644461914,
      "loss": 5.8027,
      "mean_token_accuracy": 0.17338819652795792,
      "num_tokens": 40489677.0,
      "step": 18705
    },
    {
      "entropy": 6.463828659057617,
      "epoch": 2.0021402964310555,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046046592079235595,
      "loss": 5.857,
      "mean_token_accuracy": 0.16318298727273942,
      "num_tokens": 40500694.0,
      "step": 18710
    },
    {
      "entropy": 6.410423135757446,
      "epoch": 2.0026753705388196,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004604442717832181,
      "loss": 5.8244,
      "mean_token_accuracy": 0.16354424208402635,
      "num_tokens": 40511539.0,
      "step": 18715
    },
    {
      "entropy": 6.374074840545655,
      "epoch": 2.0032104446465837,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000460422617419403,
      "loss": 5.7822,
      "mean_token_accuracy": 0.16821132302284242,
      "num_tokens": 40521906.0,
      "step": 18720
    },
    {
      "entropy": 6.4704255104064945,
      "epoch": 2.0037455187543474,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046040095770153594,
      "loss": 5.8483,
      "mean_token_accuracy": 0.1592197984457016,
      "num_tokens": 40531454.0,
      "step": 18725
    },
    {
      "entropy": 6.489177703857422,
      "epoch": 2.0042805928621115,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004603792926302422,
      "loss": 5.8932,
      "mean_token_accuracy": 0.16078903079032897,
      "num_tokens": 40541497.0,
      "step": 18730
    },
    {
      "entropy": 6.444369649887085,
      "epoch": 2.004815666969875,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004603576222061475,
      "loss": 5.8883,
      "mean_token_accuracy": 0.16434308439493178,
      "num_tokens": 40552576.0,
      "step": 18735
    },
    {
      "entropy": 6.405531215667724,
      "epoch": 2.005350741077639,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004603359464298774,
      "loss": 5.8416,
      "mean_token_accuracy": 0.16577156335115434,
      "num_tokens": 40562906.0,
      "step": 18740
    },
    {
      "entropy": 6.375948667526245,
      "epoch": 2.0058858151854033,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046031426530205796,
      "loss": 5.7548,
      "mean_token_accuracy": 0.1727544352412224,
      "num_tokens": 40574510.0,
      "step": 18745
    },
    {
      "entropy": 6.354838705062866,
      "epoch": 2.006420889293167,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004602925788233152,
      "loss": 5.8112,
      "mean_token_accuracy": 0.17822517901659013,
      "num_tokens": 40586170.0,
      "step": 18750
    },
    {
      "entropy": 6.458474349975586,
      "epoch": 2.006955963400931,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046027088699427516,
      "loss": 5.8447,
      "mean_token_accuracy": 0.16062600314617156,
      "num_tokens": 40597613.0,
      "step": 18755
    },
    {
      "entropy": 6.495531272888184,
      "epoch": 2.007491037508695,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046024918981556434,
      "loss": 5.904,
      "mean_token_accuracy": 0.16428085118532182,
      "num_tokens": 40608366.0,
      "step": 18760
    },
    {
      "entropy": 6.412400960922241,
      "epoch": 2.008026111616459,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004602274872878092,
      "loss": 5.8232,
      "mean_token_accuracy": 0.17287563234567643,
      "num_tokens": 40618616.0,
      "step": 18765
    },
    {
      "entropy": 6.317733192443848,
      "epoch": 2.008561185724223,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046020577941163635,
      "loss": 5.7053,
      "mean_token_accuracy": 0.17611220479011536,
      "num_tokens": 40628342.0,
      "step": 18770
    },
    {
      "entropy": 6.368038558959961,
      "epoch": 2.0090962598319866,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046018406618767273,
      "loss": 5.824,
      "mean_token_accuracy": 0.1612042546272278,
      "num_tokens": 40638644.0,
      "step": 18775
    },
    {
      "entropy": 6.431771945953369,
      "epoch": 2.0096313339397507,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004601623476165451,
      "loss": 5.8205,
      "mean_token_accuracy": 0.1686380088329315,
      "num_tokens": 40649945.0,
      "step": 18780
    },
    {
      "entropy": 6.522971963882446,
      "epoch": 2.010166408047515,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046014062369888084,
      "loss": 5.9422,
      "mean_token_accuracy": 0.15927932858467103,
      "num_tokens": 40660896.0,
      "step": 18785
    },
    {
      "entropy": 6.446690702438355,
      "epoch": 2.0107014821552784,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000460118894435307,
      "loss": 5.897,
      "mean_token_accuracy": 0.16289696395397185,
      "num_tokens": 40672072.0,
      "step": 18790
    },
    {
      "entropy": 6.458041524887085,
      "epoch": 2.0112365562630425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004600971598264511,
      "loss": 5.8925,
      "mean_token_accuracy": 0.16279026567935945,
      "num_tokens": 40682622.0,
      "step": 18795
    },
    {
      "entropy": 6.381924629211426,
      "epoch": 2.011771630370806,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004600754198729407,
      "loss": 5.8121,
      "mean_token_accuracy": 0.1670759454369545,
      "num_tokens": 40693127.0,
      "step": 18800
    },
    {
      "entropy": 6.426330423355102,
      "epoch": 2.0123067044785703,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046005367457540366,
      "loss": 5.8159,
      "mean_token_accuracy": 0.16736077815294265,
      "num_tokens": 40705009.0,
      "step": 18805
    },
    {
      "entropy": 6.472621870040894,
      "epoch": 2.0128417785863344,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046003192393446763,
      "loss": 5.8679,
      "mean_token_accuracy": 0.16330384314060212,
      "num_tokens": 40716498.0,
      "step": 18810
    },
    {
      "entropy": 6.459681463241577,
      "epoch": 2.013376852694098,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004600101679507609,
      "loss": 5.8592,
      "mean_token_accuracy": 0.17091048806905745,
      "num_tokens": 40727802.0,
      "step": 18815
    },
    {
      "entropy": 6.482123184204101,
      "epoch": 2.013911926801862,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004599884066249115,
      "loss": 5.8677,
      "mean_token_accuracy": 0.16809843629598617,
      "num_tokens": 40738866.0,
      "step": 18820
    },
    {
      "entropy": 6.380206537246704,
      "epoch": 2.014447000909626,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004599666399575479,
      "loss": 5.8283,
      "mean_token_accuracy": 0.16551585793495177,
      "num_tokens": 40750214.0,
      "step": 18825
    },
    {
      "entropy": 6.430353784561158,
      "epoch": 2.01498207501739,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004599448679492986,
      "loss": 5.8427,
      "mean_token_accuracy": 0.16924750953912734,
      "num_tokens": 40759830.0,
      "step": 18830
    },
    {
      "entropy": 6.468142318725586,
      "epoch": 2.015517149125154,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004599230906007921,
      "loss": 5.8994,
      "mean_token_accuracy": 0.16013766676187516,
      "num_tokens": 40770007.0,
      "step": 18835
    },
    {
      "entropy": 6.464301204681396,
      "epoch": 2.0160522232329177,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004599013079126575,
      "loss": 5.7881,
      "mean_token_accuracy": 0.16244660913944245,
      "num_tokens": 40781495.0,
      "step": 18840
    },
    {
      "entropy": 6.447126054763794,
      "epoch": 2.0165872973406818,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004598795198855235,
      "loss": 5.8016,
      "mean_token_accuracy": 0.16637408435344697,
      "num_tokens": 40792875.0,
      "step": 18845
    },
    {
      "entropy": 6.437228441238403,
      "epoch": 2.0171223714484454,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004598577265200193,
      "loss": 5.8946,
      "mean_token_accuracy": 0.15787415355443954,
      "num_tokens": 40804484.0,
      "step": 18850
    },
    {
      "entropy": 6.394803285598755,
      "epoch": 2.0176574455562095,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004598359278167744,
      "loss": 5.8223,
      "mean_token_accuracy": 0.16159922778606414,
      "num_tokens": 40814692.0,
      "step": 18855
    },
    {
      "entropy": 6.427354860305786,
      "epoch": 2.0181925196639736,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045981412377641795,
      "loss": 5.7892,
      "mean_token_accuracy": 0.17356277853250504,
      "num_tokens": 40825443.0,
      "step": 18860
    },
    {
      "entropy": 6.437495946884155,
      "epoch": 2.0187275937717373,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045979231439957966,
      "loss": 5.8514,
      "mean_token_accuracy": 0.16634950637817383,
      "num_tokens": 40836699.0,
      "step": 18865
    },
    {
      "entropy": 6.4581891059875485,
      "epoch": 2.0192626678795014,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004597704996868893,
      "loss": 5.8334,
      "mean_token_accuracy": 0.16337066441774367,
      "num_tokens": 40846736.0,
      "step": 18870
    },
    {
      "entropy": 6.47485876083374,
      "epoch": 2.019797741987265,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004597486796389766,
      "loss": 5.9179,
      "mean_token_accuracy": 0.1683817058801651,
      "num_tokens": 40858362.0,
      "step": 18875
    },
    {
      "entropy": 6.39816575050354,
      "epoch": 2.020332816095029,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045972685425647186,
      "loss": 5.8382,
      "mean_token_accuracy": 0.16412338465452195,
      "num_tokens": 40869346.0,
      "step": 18880
    },
    {
      "entropy": 6.450319528579712,
      "epoch": 2.0208678902027932,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004597050235400051,
      "loss": 5.9083,
      "mean_token_accuracy": 0.16675252318382264,
      "num_tokens": 40880003.0,
      "step": 18885
    },
    {
      "entropy": 6.490572786331176,
      "epoch": 2.021402964310557,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00045968318749020675,
      "loss": 5.8692,
      "mean_token_accuracy": 0.16374387890100478,
      "num_tokens": 40891515.0,
      "step": 18890
    },
    {
      "entropy": 6.452636241912842,
      "epoch": 2.021938038418321,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004596613461077073,
      "loss": 5.7991,
      "mean_token_accuracy": 0.1686956912279129,
      "num_tokens": 40902024.0,
      "step": 18895
    },
    {
      "entropy": 6.408902835845947,
      "epoch": 2.022473112526085,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045963949939313754,
      "loss": 5.8166,
      "mean_token_accuracy": 0.1717332422733307,
      "num_tokens": 40912160.0,
      "step": 18900
    },
    {
      "entropy": 6.44699273109436,
      "epoch": 2.0230081866338487,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000459617647347128,
      "loss": 5.9007,
      "mean_token_accuracy": 0.15373365432024003,
      "num_tokens": 40922516.0,
      "step": 18905
    },
    {
      "entropy": 6.436704206466675,
      "epoch": 2.023543260741613,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045959578997031,
      "loss": 5.8665,
      "mean_token_accuracy": 0.16302557587623595,
      "num_tokens": 40934148.0,
      "step": 18910
    },
    {
      "entropy": 6.476804542541504,
      "epoch": 2.0240783348493765,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045957392726331443,
      "loss": 5.8068,
      "mean_token_accuracy": 0.17280222624540328,
      "num_tokens": 40944409.0,
      "step": 18915
    },
    {
      "entropy": 6.4406524181365965,
      "epoch": 2.0246134089571406,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045955205922677264,
      "loss": 5.8664,
      "mean_token_accuracy": 0.16473795026540755,
      "num_tokens": 40955133.0,
      "step": 18920
    },
    {
      "entropy": 6.414217138290406,
      "epoch": 2.0251484830649047,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004595301858613161,
      "loss": 5.9321,
      "mean_token_accuracy": 0.16428882628679276,
      "num_tokens": 40965508.0,
      "step": 18925
    },
    {
      "entropy": 6.539606809616089,
      "epoch": 2.0256835571726683,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004595083071675763,
      "loss": 5.9204,
      "mean_token_accuracy": 0.17141903340816497,
      "num_tokens": 40975931.0,
      "step": 18930
    },
    {
      "entropy": 6.484955501556397,
      "epoch": 2.0262186312804324,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045948642314618514,
      "loss": 5.8681,
      "mean_token_accuracy": 0.16216625422239303,
      "num_tokens": 40985797.0,
      "step": 18935
    },
    {
      "entropy": 6.427921915054322,
      "epoch": 2.026753705388196,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004594645337977744,
      "loss": 5.848,
      "mean_token_accuracy": 0.1661387026309967,
      "num_tokens": 40996501.0,
      "step": 18940
    },
    {
      "entropy": 6.418433475494385,
      "epoch": 2.02728877949596,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004594426391229762,
      "loss": 5.8315,
      "mean_token_accuracy": 0.1672544911503792,
      "num_tokens": 41005970.0,
      "step": 18945
    },
    {
      "entropy": 6.411648607254028,
      "epoch": 2.0278238536037243,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004594207391224227,
      "loss": 5.8163,
      "mean_token_accuracy": 0.16006234735250474,
      "num_tokens": 41017208.0,
      "step": 18950
    },
    {
      "entropy": 6.5205397605896,
      "epoch": 2.028358927711488,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045939883379674624,
      "loss": 5.99,
      "mean_token_accuracy": 0.15978730767965316,
      "num_tokens": 41028435.0,
      "step": 18955
    },
    {
      "entropy": 6.42169041633606,
      "epoch": 2.028894001819252,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045937692314657936,
      "loss": 5.8363,
      "mean_token_accuracy": 0.17131348103284835,
      "num_tokens": 41039256.0,
      "step": 18960
    },
    {
      "entropy": 6.360827112197876,
      "epoch": 2.0294290759270157,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004593550071725548,
      "loss": 5.8401,
      "mean_token_accuracy": 0.1646688237786293,
      "num_tokens": 41050527.0,
      "step": 18965
    },
    {
      "entropy": 6.453940057754517,
      "epoch": 2.02996415003478,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045933308587530527,
      "loss": 5.8736,
      "mean_token_accuracy": 0.1621284767985344,
      "num_tokens": 41062259.0,
      "step": 18970
    },
    {
      "entropy": 6.462147998809814,
      "epoch": 2.030499224142544,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004593111592554638,
      "loss": 5.9022,
      "mean_token_accuracy": 0.16435067504644393,
      "num_tokens": 41073011.0,
      "step": 18975
    },
    {
      "entropy": 6.42039122581482,
      "epoch": 2.0310342982503076,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004592892273136635,
      "loss": 5.7503,
      "mean_token_accuracy": 0.17424234747886658,
      "num_tokens": 41083482.0,
      "step": 18980
    },
    {
      "entropy": 6.393030452728271,
      "epoch": 2.0315693723580717,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045926729005053764,
      "loss": 5.8225,
      "mean_token_accuracy": 0.1715554878115654,
      "num_tokens": 41094028.0,
      "step": 18985
    },
    {
      "entropy": 6.374712228775024,
      "epoch": 2.0321044464658353,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004592453474667197,
      "loss": 5.8059,
      "mean_token_accuracy": 0.16624326556921004,
      "num_tokens": 41106346.0,
      "step": 18990
    },
    {
      "entropy": 6.47103328704834,
      "epoch": 2.0326395205735994,
      "grad_norm": 1.125,
      "learning_rate": 0.0004592233995628432,
      "loss": 5.7773,
      "mean_token_accuracy": 0.16701408177614213,
      "num_tokens": 41116130.0,
      "step": 18995
    },
    {
      "entropy": 6.420724010467529,
      "epoch": 2.0331745946813635,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000459201446339542,
      "loss": 5.7799,
      "mean_token_accuracy": 0.1676821932196617,
      "num_tokens": 41127103.0,
      "step": 19000
    },
    {
      "entropy": 6.36404767036438,
      "epoch": 2.033709668789127,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004591794877974499,
      "loss": 5.8108,
      "mean_token_accuracy": 0.16331264078617097,
      "num_tokens": 41138258.0,
      "step": 19005
    },
    {
      "entropy": 6.466695070266724,
      "epoch": 2.0342447428968913,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045915752393720094,
      "loss": 5.8657,
      "mean_token_accuracy": 0.16688111275434495,
      "num_tokens": 41148859.0,
      "step": 19010
    },
    {
      "entropy": 6.432927417755127,
      "epoch": 2.034779817004655,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045913555475942937,
      "loss": 5.7724,
      "mean_token_accuracy": 0.1649308294057846,
      "num_tokens": 41158614.0,
      "step": 19015
    },
    {
      "entropy": 6.340429973602295,
      "epoch": 2.035314891112419,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004591135802647695,
      "loss": 5.8095,
      "mean_token_accuracy": 0.1767549678683281,
      "num_tokens": 41170215.0,
      "step": 19020
    },
    {
      "entropy": 6.423883724212646,
      "epoch": 2.035849965220183,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004590916004538559,
      "loss": 5.8353,
      "mean_token_accuracy": 0.16390147805213928,
      "num_tokens": 41180784.0,
      "step": 19025
    },
    {
      "entropy": 6.472837448120117,
      "epoch": 2.036385039327947,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045906961532732316,
      "loss": 5.8702,
      "mean_token_accuracy": 0.163063944876194,
      "num_tokens": 41191305.0,
      "step": 19030
    },
    {
      "entropy": 6.431164407730103,
      "epoch": 2.036920113435711,
      "grad_norm": 1.0,
      "learning_rate": 0.0004590476248858062,
      "loss": 5.8809,
      "mean_token_accuracy": 0.160915544629097,
      "num_tokens": 41202388.0,
      "step": 19035
    },
    {
      "entropy": 6.3996514797210695,
      "epoch": 2.037455187543475,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004590256291299399,
      "loss": 5.8088,
      "mean_token_accuracy": 0.16877888143062592,
      "num_tokens": 41212136.0,
      "step": 19040
    },
    {
      "entropy": 6.405546712875366,
      "epoch": 2.0379902616512386,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045900362806035945,
      "loss": 5.859,
      "mean_token_accuracy": 0.1653733640909195,
      "num_tokens": 41223390.0,
      "step": 19045
    },
    {
      "entropy": 6.391432189941407,
      "epoch": 2.0385253357590027,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045898162167770006,
      "loss": 5.8875,
      "mean_token_accuracy": 0.16809014678001405,
      "num_tokens": 41234775.0,
      "step": 19050
    },
    {
      "entropy": 6.448359727859497,
      "epoch": 2.0390604098667664,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004589596099825973,
      "loss": 5.808,
      "mean_token_accuracy": 0.16744986176490784,
      "num_tokens": 41245627.0,
      "step": 19055
    },
    {
      "entropy": 6.408150911331177,
      "epoch": 2.0395954839745305,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045893759297568654,
      "loss": 5.8679,
      "mean_token_accuracy": 0.16578415930271148,
      "num_tokens": 41255441.0,
      "step": 19060
    },
    {
      "entropy": 6.370366334915161,
      "epoch": 2.0401305580822946,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004589155706576036,
      "loss": 5.7786,
      "mean_token_accuracy": 0.16955500841140747,
      "num_tokens": 41268686.0,
      "step": 19065
    },
    {
      "entropy": 6.499679374694824,
      "epoch": 2.0406656321900583,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045889354302898445,
      "loss": 5.9221,
      "mean_token_accuracy": 0.15905057340860368,
      "num_tokens": 41279384.0,
      "step": 19070
    },
    {
      "entropy": 6.500144624710083,
      "epoch": 2.0412007062978224,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004588715100904651,
      "loss": 5.9058,
      "mean_token_accuracy": 0.16238564103841782,
      "num_tokens": 41289923.0,
      "step": 19075
    },
    {
      "entropy": 6.401065969467163,
      "epoch": 2.041735780405586,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004588494718426818,
      "loss": 5.8419,
      "mean_token_accuracy": 0.16651138812303543,
      "num_tokens": 41300486.0,
      "step": 19080
    },
    {
      "entropy": 6.403222179412841,
      "epoch": 2.04227085451335,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004588274282862708,
      "loss": 5.831,
      "mean_token_accuracy": 0.16521429792046546,
      "num_tokens": 41310507.0,
      "step": 19085
    },
    {
      "entropy": 6.436733818054199,
      "epoch": 2.042805928621114,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045880537942186855,
      "loss": 5.8611,
      "mean_token_accuracy": 0.17306173890829085,
      "num_tokens": 41321284.0,
      "step": 19090
    },
    {
      "entropy": 6.397989654541016,
      "epoch": 2.043341002728878,
      "grad_norm": 1.125,
      "learning_rate": 0.0004587833252501118,
      "loss": 5.7814,
      "mean_token_accuracy": 0.1747141271829605,
      "num_tokens": 41332037.0,
      "step": 19095
    },
    {
      "entropy": 6.331292057037354,
      "epoch": 2.043876076836642,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004587612657716374,
      "loss": 5.8424,
      "mean_token_accuracy": 0.1679975226521492,
      "num_tokens": 41343480.0,
      "step": 19100
    },
    {
      "entropy": 6.48652982711792,
      "epoch": 2.0444111509444056,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004587392009870824,
      "loss": 5.8985,
      "mean_token_accuracy": 0.1544149935245514,
      "num_tokens": 41353656.0,
      "step": 19105
    },
    {
      "entropy": 6.462438869476318,
      "epoch": 2.0449462250521697,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004587171308970836,
      "loss": 5.8375,
      "mean_token_accuracy": 0.16339320987462996,
      "num_tokens": 41363640.0,
      "step": 19110
    },
    {
      "entropy": 6.425931549072265,
      "epoch": 2.045481299159934,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045869505550227857,
      "loss": 5.8673,
      "mean_token_accuracy": 0.15748696625232697,
      "num_tokens": 41374194.0,
      "step": 19115
    },
    {
      "entropy": 6.428192472457885,
      "epoch": 2.0460163732676975,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004586729748033046,
      "loss": 5.8311,
      "mean_token_accuracy": 0.1696629449725151,
      "num_tokens": 41384968.0,
      "step": 19120
    },
    {
      "entropy": 6.387308025360108,
      "epoch": 2.0465514473754616,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045865088880079924,
      "loss": 5.7411,
      "mean_token_accuracy": 0.1766682580113411,
      "num_tokens": 41395117.0,
      "step": 19125
    },
    {
      "entropy": 6.436519002914428,
      "epoch": 2.0470865214832252,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004586287974954004,
      "loss": 5.8579,
      "mean_token_accuracy": 0.1664932206273079,
      "num_tokens": 41406366.0,
      "step": 19130
    },
    {
      "entropy": 6.479413223266602,
      "epoch": 2.0476215955909893,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004586067008877457,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16447473168373108,
      "num_tokens": 41417388.0,
      "step": 19135
    },
    {
      "entropy": 6.395864391326905,
      "epoch": 2.0481566696987534,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004585845989784734,
      "loss": 5.8044,
      "mean_token_accuracy": 0.16751595139503478,
      "num_tokens": 41428085.0,
      "step": 19140
    },
    {
      "entropy": 6.478150224685669,
      "epoch": 2.048691743806517,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045856249176822155,
      "loss": 5.9089,
      "mean_token_accuracy": 0.1594173327088356,
      "num_tokens": 41439117.0,
      "step": 19145
    },
    {
      "entropy": 6.4349699974060055,
      "epoch": 2.049226817914281,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045854037925762857,
      "loss": 5.9,
      "mean_token_accuracy": 0.15863346308469772,
      "num_tokens": 41450042.0,
      "step": 19150
    },
    {
      "entropy": 6.41898136138916,
      "epoch": 2.049761892022045,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045851826144733296,
      "loss": 5.7917,
      "mean_token_accuracy": 0.1716499701142311,
      "num_tokens": 41460554.0,
      "step": 19155
    },
    {
      "entropy": 6.466725635528564,
      "epoch": 2.050296966129809,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045849613833797326,
      "loss": 5.8855,
      "mean_token_accuracy": 0.16671936064958573,
      "num_tokens": 41471403.0,
      "step": 19160
    },
    {
      "entropy": 6.433062124252319,
      "epoch": 2.050832040237573,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045847400993018835,
      "loss": 5.8523,
      "mean_token_accuracy": 0.16403007805347442,
      "num_tokens": 41482098.0,
      "step": 19165
    },
    {
      "entropy": 6.408970022201538,
      "epoch": 2.0513671143453367,
      "grad_norm": 1.125,
      "learning_rate": 0.00045845187622461724,
      "loss": 5.8171,
      "mean_token_accuracy": 0.17036739438772203,
      "num_tokens": 41493126.0,
      "step": 19170
    },
    {
      "entropy": 6.48432674407959,
      "epoch": 2.051902188453101,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000458429737221899,
      "loss": 5.923,
      "mean_token_accuracy": 0.1629864305257797,
      "num_tokens": 41504242.0,
      "step": 19175
    },
    {
      "entropy": 6.410525226593018,
      "epoch": 2.052437262560865,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004584075929226728,
      "loss": 5.8279,
      "mean_token_accuracy": 0.16594446152448655,
      "num_tokens": 41515368.0,
      "step": 19180
    },
    {
      "entropy": 6.461184120178222,
      "epoch": 2.0529723366686286,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004583854433275782,
      "loss": 5.8714,
      "mean_token_accuracy": 0.1596730425953865,
      "num_tokens": 41525939.0,
      "step": 19185
    },
    {
      "entropy": 6.402989435195923,
      "epoch": 2.0535074107763926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045836328843725465,
      "loss": 5.897,
      "mean_token_accuracy": 0.16326470971107482,
      "num_tokens": 41536505.0,
      "step": 19190
    },
    {
      "entropy": 6.433830118179321,
      "epoch": 2.0540424848841563,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045834112825234193,
      "loss": 5.7988,
      "mean_token_accuracy": 0.16663869321346284,
      "num_tokens": 41546293.0,
      "step": 19195
    },
    {
      "entropy": 6.455318403244019,
      "epoch": 2.0545775589919204,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045831896277347985,
      "loss": 5.826,
      "mean_token_accuracy": 0.16914356648921966,
      "num_tokens": 41556596.0,
      "step": 19200
    },
    {
      "entropy": 6.422694826126099,
      "epoch": 2.0551126330996845,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004582967920013085,
      "loss": 5.8595,
      "mean_token_accuracy": 0.166616852581501,
      "num_tokens": 41566732.0,
      "step": 19205
    },
    {
      "entropy": 6.431009721755982,
      "epoch": 2.055647707207448,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045827461593646806,
      "loss": 5.8808,
      "mean_token_accuracy": 0.16329149156808853,
      "num_tokens": 41577777.0,
      "step": 19210
    },
    {
      "entropy": 6.427589845657349,
      "epoch": 2.0561827813152123,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004582524345795989,
      "loss": 5.8615,
      "mean_token_accuracy": 0.16921883225440978,
      "num_tokens": 41587978.0,
      "step": 19215
    },
    {
      "entropy": 6.416416978836059,
      "epoch": 2.056717855422976,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004582302479313413,
      "loss": 5.8035,
      "mean_token_accuracy": 0.16821190416812898,
      "num_tokens": 41598977.0,
      "step": 19220
    },
    {
      "entropy": 6.44900221824646,
      "epoch": 2.05725292953074,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004582080559923362,
      "loss": 5.8592,
      "mean_token_accuracy": 0.16027268022298813,
      "num_tokens": 41609664.0,
      "step": 19225
    },
    {
      "entropy": 6.464775848388672,
      "epoch": 2.057788003638504,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004581858587632241,
      "loss": 5.8503,
      "mean_token_accuracy": 0.1669194996356964,
      "num_tokens": 41619427.0,
      "step": 19230
    },
    {
      "entropy": 6.41856050491333,
      "epoch": 2.0583230777462678,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004581636562446461,
      "loss": 5.8427,
      "mean_token_accuracy": 0.16197112202644348,
      "num_tokens": 41629551.0,
      "step": 19235
    },
    {
      "entropy": 6.3747453689575195,
      "epoch": 2.058858151854032,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004581414484372433,
      "loss": 5.8438,
      "mean_token_accuracy": 0.16379487216472627,
      "num_tokens": 41640183.0,
      "step": 19240
    },
    {
      "entropy": 6.419865083694458,
      "epoch": 2.0593932259617955,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004581192353416569,
      "loss": 5.8419,
      "mean_token_accuracy": 0.16793860793113707,
      "num_tokens": 41651934.0,
      "step": 19245
    },
    {
      "entropy": 6.362654542922973,
      "epoch": 2.0599283000695596,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004580970169585283,
      "loss": 5.7321,
      "mean_token_accuracy": 0.184805990755558,
      "num_tokens": 41662418.0,
      "step": 19250
    },
    {
      "entropy": 6.3563093662261965,
      "epoch": 2.0604633741773237,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045807479328849906,
      "loss": 5.7598,
      "mean_token_accuracy": 0.17166978120803833,
      "num_tokens": 41673419.0,
      "step": 19255
    },
    {
      "entropy": 6.376647186279297,
      "epoch": 2.0609984482850874,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045805256433221087,
      "loss": 5.7939,
      "mean_token_accuracy": 0.17786844372749328,
      "num_tokens": 41684132.0,
      "step": 19260
    },
    {
      "entropy": 6.457601070404053,
      "epoch": 2.0615335223928515,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004580303300903056,
      "loss": 5.8646,
      "mean_token_accuracy": 0.16545597165822984,
      "num_tokens": 41695051.0,
      "step": 19265
    },
    {
      "entropy": 6.443119382858276,
      "epoch": 2.062068596500615,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045800809056342534,
      "loss": 5.8833,
      "mean_token_accuracy": 0.16630606055259706,
      "num_tokens": 41705986.0,
      "step": 19270
    },
    {
      "entropy": 6.428728866577148,
      "epoch": 2.0626036706083792,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045798584575221213,
      "loss": 5.8719,
      "mean_token_accuracy": 0.16321321278810502,
      "num_tokens": 41717082.0,
      "step": 19275
    },
    {
      "entropy": 6.43544373512268,
      "epoch": 2.0631387447161433,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004579635956573084,
      "loss": 5.8798,
      "mean_token_accuracy": 0.16473229676485063,
      "num_tokens": 41728783.0,
      "step": 19280
    },
    {
      "entropy": 6.429499530792237,
      "epoch": 2.063673818823907,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045794134027935646,
      "loss": 5.8653,
      "mean_token_accuracy": 0.16325037330389022,
      "num_tokens": 41740452.0,
      "step": 19285
    },
    {
      "entropy": 6.428386306762695,
      "epoch": 2.064208892931671,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004579190796189991,
      "loss": 5.8135,
      "mean_token_accuracy": 0.16666142493486405,
      "num_tokens": 41750966.0,
      "step": 19290
    },
    {
      "entropy": 6.47124662399292,
      "epoch": 2.064743967039435,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000457896813676879,
      "loss": 5.7944,
      "mean_token_accuracy": 0.17130086421966553,
      "num_tokens": 41762174.0,
      "step": 19295
    },
    {
      "entropy": 6.42952790260315,
      "epoch": 2.065279041147199,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045787454245363907,
      "loss": 5.8963,
      "mean_token_accuracy": 0.16256206184625627,
      "num_tokens": 41773396.0,
      "step": 19300
    },
    {
      "entropy": 6.457822608947754,
      "epoch": 2.065814115254963,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004578522659499225,
      "loss": 5.8584,
      "mean_token_accuracy": 0.16979958713054658,
      "num_tokens": 41783833.0,
      "step": 19305
    },
    {
      "entropy": 6.39550552368164,
      "epoch": 2.0663491893627266,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045782998416637235,
      "loss": 5.8175,
      "mean_token_accuracy": 0.1746004641056061,
      "num_tokens": 41795363.0,
      "step": 19310
    },
    {
      "entropy": 6.427926015853882,
      "epoch": 2.0668842634704907,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004578076971036321,
      "loss": 5.8784,
      "mean_token_accuracy": 0.16432900130748748,
      "num_tokens": 41807575.0,
      "step": 19315
    },
    {
      "entropy": 6.38566484451294,
      "epoch": 2.067419337578255,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004577854047623454,
      "loss": 5.7951,
      "mean_token_accuracy": 0.16903940588235855,
      "num_tokens": 41818496.0,
      "step": 19320
    },
    {
      "entropy": 6.491126871109008,
      "epoch": 2.0679544116860185,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004577631071431557,
      "loss": 5.8489,
      "mean_token_accuracy": 0.16823755502700805,
      "num_tokens": 41829106.0,
      "step": 19325
    },
    {
      "entropy": 6.385643815994262,
      "epoch": 2.0684894857937826,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045774080424670704,
      "loss": 5.8269,
      "mean_token_accuracy": 0.16632779240608214,
      "num_tokens": 41838095.0,
      "step": 19330
    },
    {
      "entropy": 6.339726734161377,
      "epoch": 2.069024559901546,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004577184960736433,
      "loss": 5.838,
      "mean_token_accuracy": 0.15676534920930862,
      "num_tokens": 41847980.0,
      "step": 19335
    },
    {
      "entropy": 6.457884407043457,
      "epoch": 2.0695596340093103,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045769618262460864,
      "loss": 5.8539,
      "mean_token_accuracy": 0.1625257283449173,
      "num_tokens": 41858625.0,
      "step": 19340
    },
    {
      "entropy": 6.498803520202637,
      "epoch": 2.0700947081170744,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004576738639002475,
      "loss": 5.869,
      "mean_token_accuracy": 0.16258393377065658,
      "num_tokens": 41870939.0,
      "step": 19345
    },
    {
      "entropy": 6.485565328598023,
      "epoch": 2.070629782224838,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004576515399012041,
      "loss": 5.8936,
      "mean_token_accuracy": 0.1642105534672737,
      "num_tokens": 41882356.0,
      "step": 19350
    },
    {
      "entropy": 6.422587299346924,
      "epoch": 2.071164856332602,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004576292106281231,
      "loss": 5.8775,
      "mean_token_accuracy": 0.16385020911693574,
      "num_tokens": 41892306.0,
      "step": 19355
    },
    {
      "entropy": 6.4395404815673825,
      "epoch": 2.071699930440366,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045760687608164933,
      "loss": 5.9155,
      "mean_token_accuracy": 0.16501900255680085,
      "num_tokens": 41902450.0,
      "step": 19360
    },
    {
      "entropy": 6.421039342880249,
      "epoch": 2.07223500454813,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004575845362624278,
      "loss": 5.8222,
      "mean_token_accuracy": 0.17102037668228148,
      "num_tokens": 41913319.0,
      "step": 19365
    },
    {
      "entropy": 6.435607290267944,
      "epoch": 2.072770078655894,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045756219117110337,
      "loss": 5.88,
      "mean_token_accuracy": 0.16113779395818711,
      "num_tokens": 41923598.0,
      "step": 19370
    },
    {
      "entropy": 6.455588436126709,
      "epoch": 2.0733051527636577,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045753984080832127,
      "loss": 5.8366,
      "mean_token_accuracy": 0.1645994707942009,
      "num_tokens": 41934482.0,
      "step": 19375
    },
    {
      "entropy": 6.443126058578491,
      "epoch": 2.0738402268714218,
      "grad_norm": 0.984375,
      "learning_rate": 0.000457517485174727,
      "loss": 5.8234,
      "mean_token_accuracy": 0.16701181679964067,
      "num_tokens": 41945994.0,
      "step": 19380
    },
    {
      "entropy": 6.436656045913696,
      "epoch": 2.0743753009791854,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045749512427096595,
      "loss": 5.8591,
      "mean_token_accuracy": 0.16500892341136933,
      "num_tokens": 41956564.0,
      "step": 19385
    },
    {
      "entropy": 6.4072184562683105,
      "epoch": 2.0749103750869495,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045747275809768386,
      "loss": 5.8436,
      "mean_token_accuracy": 0.16753415763378143,
      "num_tokens": 41966847.0,
      "step": 19390
    },
    {
      "entropy": 6.4771934986114506,
      "epoch": 2.0754454491947136,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004574503866555265,
      "loss": 5.9288,
      "mean_token_accuracy": 0.15067253708839418,
      "num_tokens": 41977752.0,
      "step": 19395
    },
    {
      "entropy": 6.497445440292358,
      "epoch": 2.0759805233024773,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045742800994513983,
      "loss": 5.8661,
      "mean_token_accuracy": 0.16400457918643951,
      "num_tokens": 41987837.0,
      "step": 19400
    },
    {
      "entropy": 6.472946882247925,
      "epoch": 2.0765155974102414,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045740562796717005,
      "loss": 5.8567,
      "mean_token_accuracy": 0.16134626567363738,
      "num_tokens": 41997988.0,
      "step": 19405
    },
    {
      "entropy": 6.325835514068603,
      "epoch": 2.077050671518005,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004573832407222634,
      "loss": 5.7672,
      "mean_token_accuracy": 0.16991885453462602,
      "num_tokens": 42009643.0,
      "step": 19410
    },
    {
      "entropy": 6.4249334812164305,
      "epoch": 2.077585745625769,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004573608482110663,
      "loss": 5.8405,
      "mean_token_accuracy": 0.1642468422651291,
      "num_tokens": 42019972.0,
      "step": 19415
    },
    {
      "entropy": 6.476372146606446,
      "epoch": 2.0781208197335332,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045733845043422535,
      "loss": 5.8184,
      "mean_token_accuracy": 0.1694321811199188,
      "num_tokens": 42030536.0,
      "step": 19420
    },
    {
      "entropy": 6.44429087638855,
      "epoch": 2.078655893841297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004573160473923872,
      "loss": 5.8371,
      "mean_token_accuracy": 0.16232705861330032,
      "num_tokens": 42041217.0,
      "step": 19425
    },
    {
      "entropy": 6.294672203063965,
      "epoch": 2.079190967949061,
      "grad_norm": 1.125,
      "learning_rate": 0.0004572936390861988,
      "loss": 5.7427,
      "mean_token_accuracy": 0.16894438713788987,
      "num_tokens": 42051346.0,
      "step": 19430
    },
    {
      "entropy": 6.522471714019775,
      "epoch": 2.0797260420568247,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004572712255163072,
      "loss": 5.9091,
      "mean_token_accuracy": 0.1654258370399475,
      "num_tokens": 42062718.0,
      "step": 19435
    },
    {
      "entropy": 6.4600245475769045,
      "epoch": 2.0802611161645888,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004572488066833596,
      "loss": 5.8355,
      "mean_token_accuracy": 0.1634783536195755,
      "num_tokens": 42073587.0,
      "step": 19440
    },
    {
      "entropy": 6.42154450416565,
      "epoch": 2.080796190272353,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045722638258800335,
      "loss": 5.8708,
      "mean_token_accuracy": 0.17019712179899216,
      "num_tokens": 42083763.0,
      "step": 19445
    },
    {
      "entropy": 6.346534633636475,
      "epoch": 2.0813312643801165,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004572039532308858,
      "loss": 5.797,
      "mean_token_accuracy": 0.16481893211603166,
      "num_tokens": 42094062.0,
      "step": 19450
    },
    {
      "entropy": 6.438830184936523,
      "epoch": 2.0818663384878806,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004571815186126549,
      "loss": 5.8596,
      "mean_token_accuracy": 0.16061777770519256,
      "num_tokens": 42106136.0,
      "step": 19455
    },
    {
      "entropy": 6.448560190200806,
      "epoch": 2.0824014125956447,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045715907873395805,
      "loss": 5.869,
      "mean_token_accuracy": 0.1603648766875267,
      "num_tokens": 42116309.0,
      "step": 19460
    },
    {
      "entropy": 6.393404150009156,
      "epoch": 2.0829364867034084,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004571366335954435,
      "loss": 5.8348,
      "mean_token_accuracy": 0.17005988210439682,
      "num_tokens": 42127747.0,
      "step": 19465
    },
    {
      "entropy": 6.349496984481812,
      "epoch": 2.0834715608111725,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004571141831977592,
      "loss": 5.7782,
      "mean_token_accuracy": 0.17349997013807297,
      "num_tokens": 42137992.0,
      "step": 19470
    },
    {
      "entropy": 6.4399731159210205,
      "epoch": 2.084006634918936,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004570917275415535,
      "loss": 5.8001,
      "mean_token_accuracy": 0.16876267045736312,
      "num_tokens": 42148833.0,
      "step": 19475
    },
    {
      "entropy": 6.3964794158935545,
      "epoch": 2.0845417090267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004570692666274747,
      "loss": 5.857,
      "mean_token_accuracy": 0.16541005969047545,
      "num_tokens": 42159276.0,
      "step": 19480
    },
    {
      "entropy": 6.483205938339234,
      "epoch": 2.0850767831344643,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004570468004561715,
      "loss": 5.9182,
      "mean_token_accuracy": 0.16767588555812835,
      "num_tokens": 42171541.0,
      "step": 19485
    },
    {
      "entropy": 6.445080184936524,
      "epoch": 2.085611857242228,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045702432902829245,
      "loss": 5.8611,
      "mean_token_accuracy": 0.16777804791927337,
      "num_tokens": 42182888.0,
      "step": 19490
    },
    {
      "entropy": 6.418055582046509,
      "epoch": 2.086146931349992,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004570018523444865,
      "loss": 5.8002,
      "mean_token_accuracy": 0.17425960451364517,
      "num_tokens": 42193405.0,
      "step": 19495
    },
    {
      "entropy": 6.431553220748901,
      "epoch": 2.0866820054577557,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045697937040540263,
      "loss": 5.8093,
      "mean_token_accuracy": 0.1678914725780487,
      "num_tokens": 42203355.0,
      "step": 19500
    },
    {
      "entropy": 6.403125333786011,
      "epoch": 2.08721707956552,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045695688321169,
      "loss": 5.8358,
      "mean_token_accuracy": 0.15813987255096434,
      "num_tokens": 42213191.0,
      "step": 19505
    },
    {
      "entropy": 6.441274929046631,
      "epoch": 2.087752153673284,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045693439076399797,
      "loss": 5.8984,
      "mean_token_accuracy": 0.15743620991706847,
      "num_tokens": 42223357.0,
      "step": 19510
    },
    {
      "entropy": 6.46363205909729,
      "epoch": 2.0882872277810476,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045691189306297595,
      "loss": 5.8237,
      "mean_token_accuracy": 0.16976215541362763,
      "num_tokens": 42234917.0,
      "step": 19515
    },
    {
      "entropy": 6.503038311004639,
      "epoch": 2.0888223018888117,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045688939010927353,
      "loss": 5.8638,
      "mean_token_accuracy": 0.16359584480524064,
      "num_tokens": 42245977.0,
      "step": 19520
    },
    {
      "entropy": 6.41424298286438,
      "epoch": 2.0893573759965753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004568668819035406,
      "loss": 5.8359,
      "mean_token_accuracy": 0.16314694434404373,
      "num_tokens": 42257523.0,
      "step": 19525
    },
    {
      "entropy": 6.4496622562408445,
      "epoch": 2.0898924501043394,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045684436844642694,
      "loss": 5.8479,
      "mean_token_accuracy": 0.16868808269500732,
      "num_tokens": 42267727.0,
      "step": 19530
    },
    {
      "entropy": 6.470313835144043,
      "epoch": 2.0904275242121035,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004568218497385828,
      "loss": 5.8206,
      "mean_token_accuracy": 0.17099614441394806,
      "num_tokens": 42278201.0,
      "step": 19535
    },
    {
      "entropy": 6.457871055603027,
      "epoch": 2.090962598319867,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004567993257806582,
      "loss": 5.9364,
      "mean_token_accuracy": 0.16013086587190628,
      "num_tokens": 42288695.0,
      "step": 19540
    },
    {
      "entropy": 6.44616003036499,
      "epoch": 2.0914976724276313,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004567767965733036,
      "loss": 5.9027,
      "mean_token_accuracy": 0.17182280570268632,
      "num_tokens": 42299418.0,
      "step": 19545
    },
    {
      "entropy": 6.38820972442627,
      "epoch": 2.092032746535395,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045675426211716963,
      "loss": 5.7475,
      "mean_token_accuracy": 0.17313225865364074,
      "num_tokens": 42310901.0,
      "step": 19550
    },
    {
      "entropy": 6.466144943237305,
      "epoch": 2.092567820643159,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004567317224129069,
      "loss": 5.8164,
      "mean_token_accuracy": 0.17221021205186843,
      "num_tokens": 42320055.0,
      "step": 19555
    },
    {
      "entropy": 6.503711128234864,
      "epoch": 2.093102894750923,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004567091774611661,
      "loss": 5.9235,
      "mean_token_accuracy": 0.1646993950009346,
      "num_tokens": 42330393.0,
      "step": 19560
    },
    {
      "entropy": 6.379061508178711,
      "epoch": 2.093637968858687,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045668662726259846,
      "loss": 5.8381,
      "mean_token_accuracy": 0.1641651064157486,
      "num_tokens": 42341384.0,
      "step": 19565
    },
    {
      "entropy": 6.379297733306885,
      "epoch": 2.094173042966451,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045666407181785496,
      "loss": 5.7787,
      "mean_token_accuracy": 0.17544497847557067,
      "num_tokens": 42352351.0,
      "step": 19570
    },
    {
      "entropy": 6.351498985290528,
      "epoch": 2.094708117074215,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004566415111275869,
      "loss": 5.7725,
      "mean_token_accuracy": 0.17138027995824814,
      "num_tokens": 42363912.0,
      "step": 19575
    },
    {
      "entropy": 6.430650234222412,
      "epoch": 2.0952431911819787,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004566189451924457,
      "loss": 5.8772,
      "mean_token_accuracy": 0.1662553519010544,
      "num_tokens": 42374638.0,
      "step": 19580
    },
    {
      "entropy": 6.3737787246704105,
      "epoch": 2.0957782652897428,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045659637401308296,
      "loss": 5.8011,
      "mean_token_accuracy": 0.17567936480045318,
      "num_tokens": 42385976.0,
      "step": 19585
    },
    {
      "entropy": 6.46123628616333,
      "epoch": 2.0963133393975064,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004565737975901505,
      "loss": 5.8597,
      "mean_token_accuracy": 0.15881111174821855,
      "num_tokens": 42395714.0,
      "step": 19590
    },
    {
      "entropy": 6.417952919006348,
      "epoch": 2.0968484135052705,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004565512159243001,
      "loss": 5.7651,
      "mean_token_accuracy": 0.16989606022834777,
      "num_tokens": 42406027.0,
      "step": 19595
    },
    {
      "entropy": 6.500837564468384,
      "epoch": 2.0973834876130346,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004565286290161839,
      "loss": 5.9137,
      "mean_token_accuracy": 0.1602156013250351,
      "num_tokens": 42417756.0,
      "step": 19600
    },
    {
      "entropy": 6.458520126342774,
      "epoch": 2.0979185617207983,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045650603686645403,
      "loss": 5.9095,
      "mean_token_accuracy": 0.16059592962265015,
      "num_tokens": 42429321.0,
      "step": 19605
    },
    {
      "entropy": 6.35654878616333,
      "epoch": 2.0984536358285624,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045648343947576284,
      "loss": 5.7366,
      "mean_token_accuracy": 0.17664788365364076,
      "num_tokens": 42440003.0,
      "step": 19610
    },
    {
      "entropy": 6.438579368591308,
      "epoch": 2.098988709936326,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004564608368447628,
      "loss": 5.8361,
      "mean_token_accuracy": 0.16752046048641206,
      "num_tokens": 42450743.0,
      "step": 19615
    },
    {
      "entropy": 6.429192399978637,
      "epoch": 2.09952378404409,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004564382289741067,
      "loss": 5.8132,
      "mean_token_accuracy": 0.16579100042581557,
      "num_tokens": 42460885.0,
      "step": 19620
    },
    {
      "entropy": 6.398460054397583,
      "epoch": 2.1000588581518542,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045641561586444706,
      "loss": 5.8381,
      "mean_token_accuracy": 0.16966400295495987,
      "num_tokens": 42470854.0,
      "step": 19625
    },
    {
      "entropy": 6.3822503089904785,
      "epoch": 2.100593932259618,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045639299751643714,
      "loss": 5.902,
      "mean_token_accuracy": 0.16290476322174072,
      "num_tokens": 42482392.0,
      "step": 19630
    },
    {
      "entropy": 6.437476634979248,
      "epoch": 2.101129006367382,
      "grad_norm": 1.0,
      "learning_rate": 0.00045637037393072996,
      "loss": 5.876,
      "mean_token_accuracy": 0.15513724088668823,
      "num_tokens": 42494156.0,
      "step": 19635
    },
    {
      "entropy": 6.441825866699219,
      "epoch": 2.1016640804751456,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004563477451079786,
      "loss": 5.8632,
      "mean_token_accuracy": 0.1629658818244934,
      "num_tokens": 42505185.0,
      "step": 19640
    },
    {
      "entropy": 6.398845052719116,
      "epoch": 2.1021991545829097,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045632511104883657,
      "loss": 5.7748,
      "mean_token_accuracy": 0.16904022842645644,
      "num_tokens": 42516415.0,
      "step": 19645
    },
    {
      "entropy": 6.436602640151977,
      "epoch": 2.102734228690674,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004563024717539575,
      "loss": 5.8548,
      "mean_token_accuracy": 0.17213092893362045,
      "num_tokens": 42526945.0,
      "step": 19650
    },
    {
      "entropy": 6.500004291534424,
      "epoch": 2.1032693027984375,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045627982722399506,
      "loss": 5.9158,
      "mean_token_accuracy": 0.16360370367765426,
      "num_tokens": 42537629.0,
      "step": 19655
    },
    {
      "entropy": 6.364980268478393,
      "epoch": 2.1038043769062016,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045625717745960306,
      "loss": 5.8506,
      "mean_token_accuracy": 0.1688908278942108,
      "num_tokens": 42547952.0,
      "step": 19660
    },
    {
      "entropy": 6.397638559341431,
      "epoch": 2.1043394510139652,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045623452246143547,
      "loss": 5.8335,
      "mean_token_accuracy": 0.16678587049245835,
      "num_tokens": 42559026.0,
      "step": 19665
    },
    {
      "entropy": 6.509052991867065,
      "epoch": 2.1048745251217293,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004562118622301466,
      "loss": 5.9138,
      "mean_token_accuracy": 0.16348755061626435,
      "num_tokens": 42570046.0,
      "step": 19670
    },
    {
      "entropy": 6.494258165359497,
      "epoch": 2.1054095992294934,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004561891967663906,
      "loss": 5.8879,
      "mean_token_accuracy": 0.16034430414438247,
      "num_tokens": 42580440.0,
      "step": 19675
    },
    {
      "entropy": 6.440071392059326,
      "epoch": 2.105944673337257,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045616652607082203,
      "loss": 5.8371,
      "mean_token_accuracy": 0.1618095889687538,
      "num_tokens": 42591872.0,
      "step": 19680
    },
    {
      "entropy": 6.440607643127441,
      "epoch": 2.106479747445021,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004561438501440955,
      "loss": 5.7797,
      "mean_token_accuracy": 0.16314346343278885,
      "num_tokens": 42602843.0,
      "step": 19685
    },
    {
      "entropy": 6.389559555053711,
      "epoch": 2.107014821552785,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004561211689868657,
      "loss": 5.8642,
      "mean_token_accuracy": 0.17499472051858903,
      "num_tokens": 42613476.0,
      "step": 19690
    },
    {
      "entropy": 6.334925603866577,
      "epoch": 2.107549895660549,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045609848259978766,
      "loss": 5.744,
      "mean_token_accuracy": 0.18339178264141082,
      "num_tokens": 42624059.0,
      "step": 19695
    },
    {
      "entropy": 6.474901103973389,
      "epoch": 2.108084969768313,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004560757909835164,
      "loss": 5.9227,
      "mean_token_accuracy": 0.15808606296777725,
      "num_tokens": 42633980.0,
      "step": 19700
    },
    {
      "entropy": 6.487460517883301,
      "epoch": 2.1086200438760767,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045605309413870697,
      "loss": 5.9227,
      "mean_token_accuracy": 0.15779781341552734,
      "num_tokens": 42645504.0,
      "step": 19705
    },
    {
      "entropy": 6.480719709396363,
      "epoch": 2.109155117983841,
      "grad_norm": 1.125,
      "learning_rate": 0.0004560303920660151,
      "loss": 5.8678,
      "mean_token_accuracy": 0.16409000158309936,
      "num_tokens": 42655403.0,
      "step": 19710
    },
    {
      "entropy": 6.464799070358277,
      "epoch": 2.1096901920916045,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045600768476609596,
      "loss": 5.8983,
      "mean_token_accuracy": 0.16897065490484237,
      "num_tokens": 42666236.0,
      "step": 19715
    },
    {
      "entropy": 6.466550207138061,
      "epoch": 2.1102252661993686,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045598497223960533,
      "loss": 5.9679,
      "mean_token_accuracy": 0.15529453605413437,
      "num_tokens": 42677365.0,
      "step": 19720
    },
    {
      "entropy": 6.456188535690307,
      "epoch": 2.1107603403071327,
      "grad_norm": 1.125,
      "learning_rate": 0.00045596225448719913,
      "loss": 5.8407,
      "mean_token_accuracy": 0.16145552545785904,
      "num_tokens": 42688518.0,
      "step": 19725
    },
    {
      "entropy": 6.453122663497925,
      "epoch": 2.1112954144148963,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045593953150953326,
      "loss": 5.8514,
      "mean_token_accuracy": 0.16865471452474595,
      "num_tokens": 42699877.0,
      "step": 19730
    },
    {
      "entropy": 6.438025188446045,
      "epoch": 2.1118304885226604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004559168033072638,
      "loss": 5.8474,
      "mean_token_accuracy": 0.16999187171459199,
      "num_tokens": 42710191.0,
      "step": 19735
    },
    {
      "entropy": 6.465034818649292,
      "epoch": 2.1123655626304245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045589406988104717,
      "loss": 5.9153,
      "mean_token_accuracy": 0.16506102532148362,
      "num_tokens": 42720874.0,
      "step": 19740
    },
    {
      "entropy": 6.436861276626587,
      "epoch": 2.112900636738188,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004558713312315396,
      "loss": 5.8783,
      "mean_token_accuracy": 0.16609982699155806,
      "num_tokens": 42731783.0,
      "step": 19745
    },
    {
      "entropy": 6.4107780933380125,
      "epoch": 2.1134357108459523,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004558485873593978,
      "loss": 5.84,
      "mean_token_accuracy": 0.16419390439987183,
      "num_tokens": 42741602.0,
      "step": 19750
    },
    {
      "entropy": 6.503105545043946,
      "epoch": 2.113970784953716,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045582583826527835,
      "loss": 5.8036,
      "mean_token_accuracy": 0.17330844849348068,
      "num_tokens": 42751249.0,
      "step": 19755
    },
    {
      "entropy": 6.421742725372314,
      "epoch": 2.11450585906148,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004558030839498383,
      "loss": 5.8196,
      "mean_token_accuracy": 0.1729217976331711,
      "num_tokens": 42762083.0,
      "step": 19760
    },
    {
      "entropy": 6.312354612350464,
      "epoch": 2.115040933169244,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004557803244137347,
      "loss": 5.7048,
      "mean_token_accuracy": 0.18285813331604003,
      "num_tokens": 42773443.0,
      "step": 19765
    },
    {
      "entropy": 6.425844383239746,
      "epoch": 2.115576007277008,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004557575596576245,
      "loss": 5.8789,
      "mean_token_accuracy": 0.15897115021944047,
      "num_tokens": 42784126.0,
      "step": 19770
    },
    {
      "entropy": 6.421220827102661,
      "epoch": 2.116111081384772,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045573478968216526,
      "loss": 5.7715,
      "mean_token_accuracy": 0.17537297010421754,
      "num_tokens": 42795472.0,
      "step": 19775
    },
    {
      "entropy": 6.4308172225952145,
      "epoch": 2.1166461554925355,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045571201448801424,
      "loss": 5.8272,
      "mean_token_accuracy": 0.17260844111442566,
      "num_tokens": 42805279.0,
      "step": 19780
    },
    {
      "entropy": 6.429988861083984,
      "epoch": 2.1171812296002996,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004556892340758293,
      "loss": 5.8787,
      "mean_token_accuracy": 0.16584087610244752,
      "num_tokens": 42815680.0,
      "step": 19785
    },
    {
      "entropy": 6.443085145950318,
      "epoch": 2.1177163037080637,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004556664484462681,
      "loss": 5.9369,
      "mean_token_accuracy": 0.15768889337778091,
      "num_tokens": 42826724.0,
      "step": 19790
    },
    {
      "entropy": 6.446796321868897,
      "epoch": 2.1182513778158274,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045564365759998856,
      "loss": 5.8577,
      "mean_token_accuracy": 0.163572196662426,
      "num_tokens": 42837875.0,
      "step": 19795
    },
    {
      "entropy": 6.425797176361084,
      "epoch": 2.1187864519235915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004556208615376488,
      "loss": 5.8481,
      "mean_token_accuracy": 0.16009988337755204,
      "num_tokens": 42848714.0,
      "step": 19800
    },
    {
      "entropy": 6.431131410598755,
      "epoch": 2.119321526031355,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004555980602599071,
      "loss": 5.9185,
      "mean_token_accuracy": 0.16831838637590407,
      "num_tokens": 42859764.0,
      "step": 19805
    },
    {
      "entropy": 6.376730012893677,
      "epoch": 2.1198566001391193,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004555752537674217,
      "loss": 5.7447,
      "mean_token_accuracy": 0.16924744099378586,
      "num_tokens": 42870346.0,
      "step": 19810
    },
    {
      "entropy": 6.432367372512817,
      "epoch": 2.1203916742468834,
      "grad_norm": 1.0,
      "learning_rate": 0.0004555524420608513,
      "loss": 5.8505,
      "mean_token_accuracy": 0.16876229345798494,
      "num_tokens": 42882706.0,
      "step": 19815
    },
    {
      "entropy": 6.407188987731933,
      "epoch": 2.120926748354647,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045552962514085443,
      "loss": 5.8444,
      "mean_token_accuracy": 0.17595774680376053,
      "num_tokens": 42893611.0,
      "step": 19820
    },
    {
      "entropy": 6.448733234405518,
      "epoch": 2.121461822462411,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045550680300809003,
      "loss": 5.8229,
      "mean_token_accuracy": 0.16195975691080094,
      "num_tokens": 42904034.0,
      "step": 19825
    },
    {
      "entropy": 6.452269792556763,
      "epoch": 2.1219968965701748,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045548397566321706,
      "loss": 5.8243,
      "mean_token_accuracy": 0.16534460484981536,
      "num_tokens": 42915530.0,
      "step": 19830
    },
    {
      "entropy": 6.416247034072876,
      "epoch": 2.122531970677939,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004554611431068947,
      "loss": 5.8177,
      "mean_token_accuracy": 0.16752826571464538,
      "num_tokens": 42926696.0,
      "step": 19835
    },
    {
      "entropy": 6.440952205657959,
      "epoch": 2.123067044785703,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004554383053397822,
      "loss": 5.8726,
      "mean_token_accuracy": 0.17040227353572845,
      "num_tokens": 42936985.0,
      "step": 19840
    },
    {
      "entropy": 6.428706932067871,
      "epoch": 2.1236021188934666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045541546236253896,
      "loss": 5.8012,
      "mean_token_accuracy": 0.1696324199438095,
      "num_tokens": 42947996.0,
      "step": 19845
    },
    {
      "entropy": 6.412336444854736,
      "epoch": 2.1241371930012307,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045539261417582456,
      "loss": 5.7447,
      "mean_token_accuracy": 0.17253318578004836,
      "num_tokens": 42959669.0,
      "step": 19850
    },
    {
      "entropy": 6.4095934391021725,
      "epoch": 2.124672267108995,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045536976078029874,
      "loss": 5.7309,
      "mean_token_accuracy": 0.17257265001535416,
      "num_tokens": 42969687.0,
      "step": 19855
    },
    {
      "entropy": 6.461356353759766,
      "epoch": 2.1252073412167585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004553469021766215,
      "loss": 5.8995,
      "mean_token_accuracy": 0.16130252033472062,
      "num_tokens": 42980752.0,
      "step": 19860
    },
    {
      "entropy": 6.480481815338135,
      "epoch": 2.1257424153245226,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045532403836545286,
      "loss": 5.9272,
      "mean_token_accuracy": 0.16294864416122437,
      "num_tokens": 42991949.0,
      "step": 19865
    },
    {
      "entropy": 6.376125621795654,
      "epoch": 2.1262774894322862,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004553011693474528,
      "loss": 5.8023,
      "mean_token_accuracy": 0.17984508872032165,
      "num_tokens": 43002689.0,
      "step": 19870
    },
    {
      "entropy": 6.472100305557251,
      "epoch": 2.1268125635400503,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004552782951232819,
      "loss": 5.8521,
      "mean_token_accuracy": 0.1648621827363968,
      "num_tokens": 43012939.0,
      "step": 19875
    },
    {
      "entropy": 6.45720329284668,
      "epoch": 2.1273476376478144,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045525541569360053,
      "loss": 5.8101,
      "mean_token_accuracy": 0.17297794818878173,
      "num_tokens": 43023887.0,
      "step": 19880
    },
    {
      "entropy": 6.370115470886231,
      "epoch": 2.127882711755578,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004552325310590694,
      "loss": 5.7527,
      "mean_token_accuracy": 0.1725412517786026,
      "num_tokens": 43034554.0,
      "step": 19885
    },
    {
      "entropy": 6.449231815338135,
      "epoch": 2.128417785863342,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004552096412203492,
      "loss": 5.9123,
      "mean_token_accuracy": 0.16310523748397826,
      "num_tokens": 43046322.0,
      "step": 19890
    },
    {
      "entropy": 6.470745801925659,
      "epoch": 2.128952859971106,
      "grad_norm": 1.125,
      "learning_rate": 0.000455186746178101,
      "loss": 5.7791,
      "mean_token_accuracy": 0.173251973092556,
      "num_tokens": 43055981.0,
      "step": 19895
    },
    {
      "entropy": 6.477341365814209,
      "epoch": 2.12948793407887,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045516384593298584,
      "loss": 5.8473,
      "mean_token_accuracy": 0.16170074939727783,
      "num_tokens": 43067217.0,
      "step": 19900
    },
    {
      "entropy": 6.469232797622681,
      "epoch": 2.130023008186634,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045514094048566486,
      "loss": 5.8571,
      "mean_token_accuracy": 0.1650273934006691,
      "num_tokens": 43078312.0,
      "step": 19905
    },
    {
      "entropy": 6.271515798568726,
      "epoch": 2.1305580822943977,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045511802983679955,
      "loss": 5.6522,
      "mean_token_accuracy": 0.18910305798053742,
      "num_tokens": 43090210.0,
      "step": 19910
    },
    {
      "entropy": 6.388755559921265,
      "epoch": 2.131093156402162,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004550951139870515,
      "loss": 5.7879,
      "mean_token_accuracy": 0.17058865427970887,
      "num_tokens": 43101282.0,
      "step": 19915
    },
    {
      "entropy": 6.470105981826782,
      "epoch": 2.1316282305099254,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045507219293708227,
      "loss": 5.9382,
      "mean_token_accuracy": 0.16146279126405716,
      "num_tokens": 43112856.0,
      "step": 19920
    },
    {
      "entropy": 6.449237251281739,
      "epoch": 2.1321633046176895,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045504926668755384,
      "loss": 5.7806,
      "mean_token_accuracy": 0.17230825871229172,
      "num_tokens": 43123898.0,
      "step": 19925
    },
    {
      "entropy": 6.447227239608765,
      "epoch": 2.1326983787254536,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004550263352391281,
      "loss": 5.8438,
      "mean_token_accuracy": 0.1702183559536934,
      "num_tokens": 43134651.0,
      "step": 19930
    },
    {
      "entropy": 6.452787208557129,
      "epoch": 2.1332334528332173,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004550033985924672,
      "loss": 5.8642,
      "mean_token_accuracy": 0.1691132292151451,
      "num_tokens": 43145385.0,
      "step": 19935
    },
    {
      "entropy": 6.4669641017913815,
      "epoch": 2.1337685269409814,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004549804567482335,
      "loss": 5.8384,
      "mean_token_accuracy": 0.16461127251386642,
      "num_tokens": 43157356.0,
      "step": 19940
    },
    {
      "entropy": 6.45161623954773,
      "epoch": 2.134303601048745,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045495750970708944,
      "loss": 5.8674,
      "mean_token_accuracy": 0.17037493586540223,
      "num_tokens": 43168289.0,
      "step": 19945
    },
    {
      "entropy": 6.347868013381958,
      "epoch": 2.134838675156509,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045493455746969746,
      "loss": 5.8082,
      "mean_token_accuracy": 0.17548425495624542,
      "num_tokens": 43177796.0,
      "step": 19950
    },
    {
      "entropy": 6.371859550476074,
      "epoch": 2.1353737492642733,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004549116000367205,
      "loss": 5.8259,
      "mean_token_accuracy": 0.16878054291009903,
      "num_tokens": 43188846.0,
      "step": 19955
    },
    {
      "entropy": 6.494574022293091,
      "epoch": 2.135908823372037,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045488863740882133,
      "loss": 5.8919,
      "mean_token_accuracy": 0.1697712242603302,
      "num_tokens": 43199852.0,
      "step": 19960
    },
    {
      "entropy": 6.4777617931365965,
      "epoch": 2.136443897479801,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045486566958666307,
      "loss": 5.8016,
      "mean_token_accuracy": 0.1669518917798996,
      "num_tokens": 43210591.0,
      "step": 19965
    },
    {
      "entropy": 6.457054138183594,
      "epoch": 2.136978971587565,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045484269657090887,
      "loss": 5.8107,
      "mean_token_accuracy": 0.16093191355466843,
      "num_tokens": 43221760.0,
      "step": 19970
    },
    {
      "entropy": 6.44816403388977,
      "epoch": 2.1375140456953288,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045481971836222205,
      "loss": 5.9022,
      "mean_token_accuracy": 0.15732348412275315,
      "num_tokens": 43233693.0,
      "step": 19975
    },
    {
      "entropy": 6.461922645568848,
      "epoch": 2.138049119803093,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045479673496126615,
      "loss": 5.8791,
      "mean_token_accuracy": 0.16669392585754395,
      "num_tokens": 43244565.0,
      "step": 19980
    },
    {
      "entropy": 6.388365650177002,
      "epoch": 2.1385841939108565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045477374636870485,
      "loss": 5.7901,
      "mean_token_accuracy": 0.16869880110025406,
      "num_tokens": 43254913.0,
      "step": 19985
    },
    {
      "entropy": 6.4834935665130615,
      "epoch": 2.1391192680186206,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045475075258520175,
      "loss": 5.8599,
      "mean_token_accuracy": 0.16778527200222015,
      "num_tokens": 43265525.0,
      "step": 19990
    },
    {
      "entropy": 6.4844005584716795,
      "epoch": 2.1396543421263843,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004547277536114211,
      "loss": 5.9355,
      "mean_token_accuracy": 0.15683637857437133,
      "num_tokens": 43276954.0,
      "step": 19995
    },
    {
      "entropy": 6.4616292953491214,
      "epoch": 2.1401894162341484,
      "grad_norm": 1.125,
      "learning_rate": 0.0004547047494480267,
      "loss": 5.8459,
      "mean_token_accuracy": 0.16908542364835738,
      "num_tokens": 43287491.0,
      "step": 20000
    },
    {
      "entropy": 6.428159475326538,
      "epoch": 2.1407244903419125,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004546817400956829,
      "loss": 5.8881,
      "mean_token_accuracy": 0.16143570989370346,
      "num_tokens": 43298341.0,
      "step": 20005
    },
    {
      "entropy": 6.392088413238525,
      "epoch": 2.141259564449676,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045465872555505417,
      "loss": 5.8075,
      "mean_token_accuracy": 0.17105790823698044,
      "num_tokens": 43309450.0,
      "step": 20010
    },
    {
      "entropy": 6.456686401367188,
      "epoch": 2.1417946385574402,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000454635705826805,
      "loss": 5.8323,
      "mean_token_accuracy": 0.16263357251882554,
      "num_tokens": 43320126.0,
      "step": 20015
    },
    {
      "entropy": 6.466338491439819,
      "epoch": 2.1423297126652043,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004546126809116,
      "loss": 5.9253,
      "mean_token_accuracy": 0.1573258936405182,
      "num_tokens": 43331347.0,
      "step": 20020
    },
    {
      "entropy": 6.465284109115601,
      "epoch": 2.142864786772968,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004545896508101041,
      "loss": 5.8226,
      "mean_token_accuracy": 0.16958659887313843,
      "num_tokens": 43341351.0,
      "step": 20025
    },
    {
      "entropy": 6.466542673110962,
      "epoch": 2.143399860880732,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004545666155229823,
      "loss": 5.9064,
      "mean_token_accuracy": 0.1621461033821106,
      "num_tokens": 43352471.0,
      "step": 20030
    },
    {
      "entropy": 6.489681720733643,
      "epoch": 2.1439349349884957,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045454357505089965,
      "loss": 5.7907,
      "mean_token_accuracy": 0.16481289863586426,
      "num_tokens": 43363041.0,
      "step": 20035
    },
    {
      "entropy": 6.43073582649231,
      "epoch": 2.14447000909626,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004545205293945215,
      "loss": 5.8166,
      "mean_token_accuracy": 0.16622084379196167,
      "num_tokens": 43375034.0,
      "step": 20040
    },
    {
      "entropy": 6.4339653015136715,
      "epoch": 2.145005083204024,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004544974785545134,
      "loss": 5.9432,
      "mean_token_accuracy": 0.1552848517894745,
      "num_tokens": 43386110.0,
      "step": 20045
    },
    {
      "entropy": 6.432501888275146,
      "epoch": 2.1455401573117876,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004544744225315407,
      "loss": 5.8476,
      "mean_token_accuracy": 0.16745607256889344,
      "num_tokens": 43396150.0,
      "step": 20050
    },
    {
      "entropy": 6.481930732727051,
      "epoch": 2.1460752314195517,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004544513613262693,
      "loss": 5.8658,
      "mean_token_accuracy": 0.1645260512828827,
      "num_tokens": 43407237.0,
      "step": 20055
    },
    {
      "entropy": 6.397608375549316,
      "epoch": 2.1466103055273154,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004544282949393652,
      "loss": 5.7524,
      "mean_token_accuracy": 0.1735348954796791,
      "num_tokens": 43417626.0,
      "step": 20060
    },
    {
      "entropy": 6.400396537780762,
      "epoch": 2.1471453796350795,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004544052233714941,
      "loss": 5.8377,
      "mean_token_accuracy": 0.16382003724575042,
      "num_tokens": 43427368.0,
      "step": 20065
    },
    {
      "entropy": 6.391905927658081,
      "epoch": 2.1476804537428436,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045438214662332245,
      "loss": 5.8034,
      "mean_token_accuracy": 0.16685877591371537,
      "num_tokens": 43438159.0,
      "step": 20070
    },
    {
      "entropy": 6.4730797290802,
      "epoch": 2.148215527850607,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045435906469551657,
      "loss": 5.8672,
      "mean_token_accuracy": 0.16880974918603897,
      "num_tokens": 43447403.0,
      "step": 20075
    },
    {
      "entropy": 6.3886829853057865,
      "epoch": 2.1487506019583713,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045433597758874286,
      "loss": 5.7476,
      "mean_token_accuracy": 0.1798299804329872,
      "num_tokens": 43457713.0,
      "step": 20080
    },
    {
      "entropy": 6.441000175476074,
      "epoch": 2.149285676066135,
      "grad_norm": 1.140625,
      "learning_rate": 0.000454312885303668,
      "loss": 5.8829,
      "mean_token_accuracy": 0.16633155196905136,
      "num_tokens": 43468520.0,
      "step": 20085
    },
    {
      "entropy": 6.494701957702636,
      "epoch": 2.149820750173899,
      "grad_norm": 1.125,
      "learning_rate": 0.00045428978784095873,
      "loss": 5.8915,
      "mean_token_accuracy": 0.16226353049278258,
      "num_tokens": 43480501.0,
      "step": 20090
    },
    {
      "entropy": 6.4419779777526855,
      "epoch": 2.150355824281663,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045426668520128204,
      "loss": 5.8468,
      "mean_token_accuracy": 0.1678907558321953,
      "num_tokens": 43491297.0,
      "step": 20095
    },
    {
      "entropy": 6.46983380317688,
      "epoch": 2.150890898389427,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004542435773853051,
      "loss": 5.8725,
      "mean_token_accuracy": 0.16529197692871095,
      "num_tokens": 43501212.0,
      "step": 20100
    },
    {
      "entropy": 6.450108242034912,
      "epoch": 2.151425972497191,
      "grad_norm": 1.125,
      "learning_rate": 0.000454220464393695,
      "loss": 5.9125,
      "mean_token_accuracy": 0.15975917875766754,
      "num_tokens": 43511136.0,
      "step": 20105
    },
    {
      "entropy": 6.516258192062378,
      "epoch": 2.1519610466049546,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004541973462271192,
      "loss": 5.9133,
      "mean_token_accuracy": 0.16334874778985978,
      "num_tokens": 43522006.0,
      "step": 20110
    },
    {
      "entropy": 6.452516078948975,
      "epoch": 2.1524961207127187,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004541742228862451,
      "loss": 5.8538,
      "mean_token_accuracy": 0.16222350150346757,
      "num_tokens": 43532746.0,
      "step": 20115
    },
    {
      "entropy": 6.344159078598023,
      "epoch": 2.1530311948204828,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045415109437174064,
      "loss": 5.8333,
      "mean_token_accuracy": 0.17120255082845687,
      "num_tokens": 43543707.0,
      "step": 20120
    },
    {
      "entropy": 6.4369463443756105,
      "epoch": 2.1535662689282464,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045412796068427337,
      "loss": 5.8298,
      "mean_token_accuracy": 0.16158076673746108,
      "num_tokens": 43555029.0,
      "step": 20125
    },
    {
      "entropy": 6.504056453704834,
      "epoch": 2.1541013430360105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045410482182451153,
      "loss": 5.8731,
      "mean_token_accuracy": 0.1642980009317398,
      "num_tokens": 43565639.0,
      "step": 20130
    },
    {
      "entropy": 6.385388898849487,
      "epoch": 2.1546364171437746,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045408167779312306,
      "loss": 5.7806,
      "mean_token_accuracy": 0.1735271230340004,
      "num_tokens": 43576333.0,
      "step": 20135
    },
    {
      "entropy": 6.435035037994385,
      "epoch": 2.1551714912515383,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045405852859077626,
      "loss": 5.8018,
      "mean_token_accuracy": 0.1672034427523613,
      "num_tokens": 43587012.0,
      "step": 20140
    },
    {
      "entropy": 6.421123600006103,
      "epoch": 2.1557065653593024,
      "grad_norm": 1.125,
      "learning_rate": 0.00045403537421813975,
      "loss": 5.7712,
      "mean_token_accuracy": 0.17514100968837737,
      "num_tokens": 43597779.0,
      "step": 20145
    },
    {
      "entropy": 6.434720754623413,
      "epoch": 2.156241639467066,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045401221467588184,
      "loss": 5.8302,
      "mean_token_accuracy": 0.16701741516590118,
      "num_tokens": 43608631.0,
      "step": 20150
    },
    {
      "entropy": 6.463970899581909,
      "epoch": 2.15677671357483,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004539890499646714,
      "loss": 5.7919,
      "mean_token_accuracy": 0.1753564238548279,
      "num_tokens": 43620254.0,
      "step": 20155
    },
    {
      "entropy": 6.437707233428955,
      "epoch": 2.1573117876825942,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004539658800851773,
      "loss": 5.7712,
      "mean_token_accuracy": 0.16801337599754335,
      "num_tokens": 43630986.0,
      "step": 20160
    },
    {
      "entropy": 6.37644100189209,
      "epoch": 2.157846861790358,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004539427050380686,
      "loss": 5.787,
      "mean_token_accuracy": 0.16474586427211763,
      "num_tokens": 43641260.0,
      "step": 20165
    },
    {
      "entropy": 6.437306451797485,
      "epoch": 2.158381935898122,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045391952482401444,
      "loss": 5.943,
      "mean_token_accuracy": 0.16102536767721176,
      "num_tokens": 43652093.0,
      "step": 20170
    },
    {
      "entropy": 6.456766510009766,
      "epoch": 2.1589170100058857,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045389633944368406,
      "loss": 5.858,
      "mean_token_accuracy": 0.16350767761468887,
      "num_tokens": 43662799.0,
      "step": 20175
    },
    {
      "entropy": 6.427477169036865,
      "epoch": 2.1594520841136498,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004538731488977471,
      "loss": 5.7892,
      "mean_token_accuracy": 0.17015786468982697,
      "num_tokens": 43674198.0,
      "step": 20180
    },
    {
      "entropy": 6.444305467605591,
      "epoch": 2.159987158221414,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004538499531868731,
      "loss": 5.8386,
      "mean_token_accuracy": 0.16580623239278794,
      "num_tokens": 43684841.0,
      "step": 20185
    },
    {
      "entropy": 6.478142213821411,
      "epoch": 2.1605222323291775,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045382675231173174,
      "loss": 5.9305,
      "mean_token_accuracy": 0.15923755913972854,
      "num_tokens": 43695903.0,
      "step": 20190
    },
    {
      "entropy": 6.485306930541992,
      "epoch": 2.1610573064369416,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004538035462729931,
      "loss": 5.8366,
      "mean_token_accuracy": 0.16566680371761322,
      "num_tokens": 43706740.0,
      "step": 20195
    },
    {
      "entropy": 6.424800539016724,
      "epoch": 2.1615923805447053,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004537803350713273,
      "loss": 5.7829,
      "mean_token_accuracy": 0.16781352013349532,
      "num_tokens": 43717124.0,
      "step": 20200
    },
    {
      "entropy": 6.362027168273926,
      "epoch": 2.1621274546524694,
      "grad_norm": 1.0,
      "learning_rate": 0.0004537571187074043,
      "loss": 5.8035,
      "mean_token_accuracy": 0.17307131439447404,
      "num_tokens": 43727346.0,
      "step": 20205
    },
    {
      "entropy": 6.490067005157471,
      "epoch": 2.1626625287602335,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045373389718189464,
      "loss": 5.9625,
      "mean_token_accuracy": 0.1524321973323822,
      "num_tokens": 43738297.0,
      "step": 20210
    },
    {
      "entropy": 6.4593791484832765,
      "epoch": 2.163197602867997,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004537106704954689,
      "loss": 5.8471,
      "mean_token_accuracy": 0.16342633664608003,
      "num_tokens": 43749543.0,
      "step": 20215
    },
    {
      "entropy": 6.478094387054443,
      "epoch": 2.163732676975761,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045368743864879755,
      "loss": 5.821,
      "mean_token_accuracy": 0.1666335567831993,
      "num_tokens": 43760324.0,
      "step": 20220
    },
    {
      "entropy": 6.465610599517822,
      "epoch": 2.164267751083525,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045366420164255156,
      "loss": 5.901,
      "mean_token_accuracy": 0.1629411980509758,
      "num_tokens": 43770391.0,
      "step": 20225
    },
    {
      "entropy": 6.366758823394775,
      "epoch": 2.164802825191289,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004536409594774019,
      "loss": 5.8053,
      "mean_token_accuracy": 0.16362592875957488,
      "num_tokens": 43781516.0,
      "step": 20230
    },
    {
      "entropy": 6.382858610153198,
      "epoch": 2.165337899299053,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045361771215401953,
      "loss": 5.8409,
      "mean_token_accuracy": 0.17172675132751464,
      "num_tokens": 43794152.0,
      "step": 20235
    },
    {
      "entropy": 6.500121545791626,
      "epoch": 2.1658729734068167,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045359445967307595,
      "loss": 5.83,
      "mean_token_accuracy": 0.16793027371168137,
      "num_tokens": 43805712.0,
      "step": 20240
    },
    {
      "entropy": 6.506805419921875,
      "epoch": 2.166408047514581,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004535712020352423,
      "loss": 5.8946,
      "mean_token_accuracy": 0.16577986031770706,
      "num_tokens": 43816985.0,
      "step": 20245
    },
    {
      "entropy": 6.424125671386719,
      "epoch": 2.166943121622345,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004535479392411904,
      "loss": 5.8064,
      "mean_token_accuracy": 0.16764383316040038,
      "num_tokens": 43828048.0,
      "step": 20250
    },
    {
      "entropy": 6.441467571258545,
      "epoch": 2.1674781957301086,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045352467129159183,
      "loss": 5.8426,
      "mean_token_accuracy": 0.1664919599890709,
      "num_tokens": 43838507.0,
      "step": 20255
    },
    {
      "entropy": 6.390451288223266,
      "epoch": 2.1680132698378727,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045350139818711846,
      "loss": 5.7614,
      "mean_token_accuracy": 0.17237259596586227,
      "num_tokens": 43849328.0,
      "step": 20260
    },
    {
      "entropy": 6.426527500152588,
      "epoch": 2.1685483439456363,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004534781199284423,
      "loss": 5.8586,
      "mean_token_accuracy": 0.16145385950803756,
      "num_tokens": 43860311.0,
      "step": 20265
    },
    {
      "entropy": 6.49206051826477,
      "epoch": 2.1690834180534004,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004534548365162354,
      "loss": 5.9271,
      "mean_token_accuracy": 0.16517846286296844,
      "num_tokens": 43871167.0,
      "step": 20270
    },
    {
      "entropy": 6.411978816986084,
      "epoch": 2.169618492161164,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045343154795117024,
      "loss": 5.7879,
      "mean_token_accuracy": 0.17606168240308762,
      "num_tokens": 43882332.0,
      "step": 20275
    },
    {
      "entropy": 6.460506868362427,
      "epoch": 2.170153566268928,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004534082542339192,
      "loss": 5.7495,
      "mean_token_accuracy": 0.169540236890316,
      "num_tokens": 43892971.0,
      "step": 20280
    },
    {
      "entropy": 6.419000768661499,
      "epoch": 2.1706886403766923,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045338495536515497,
      "loss": 5.8235,
      "mean_token_accuracy": 0.17327871322631835,
      "num_tokens": 43904305.0,
      "step": 20285
    },
    {
      "entropy": 6.312623357772827,
      "epoch": 2.171223714484456,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004533616513455502,
      "loss": 5.8022,
      "mean_token_accuracy": 0.167039255797863,
      "num_tokens": 43914431.0,
      "step": 20290
    },
    {
      "entropy": 6.402218198776245,
      "epoch": 2.17175878859222,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045333834217577775,
      "loss": 5.8319,
      "mean_token_accuracy": 0.1741215318441391,
      "num_tokens": 43925350.0,
      "step": 20295
    },
    {
      "entropy": 6.396366691589355,
      "epoch": 2.172293862699984,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045331502785651075,
      "loss": 5.8735,
      "mean_token_accuracy": 0.1637540727853775,
      "num_tokens": 43936194.0,
      "step": 20300
    },
    {
      "entropy": 6.462814378738403,
      "epoch": 2.172828936807748,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004532917083884224,
      "loss": 5.8476,
      "mean_token_accuracy": 0.16281537264585494,
      "num_tokens": 43946692.0,
      "step": 20305
    },
    {
      "entropy": 6.425861787796021,
      "epoch": 2.173364010915512,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000453268383772186,
      "loss": 5.8012,
      "mean_token_accuracy": 0.16659508943557738,
      "num_tokens": 43957493.0,
      "step": 20310
    },
    {
      "entropy": 6.468068361282349,
      "epoch": 2.1738990850232756,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045324505400847506,
      "loss": 5.9174,
      "mean_token_accuracy": 0.16395393162965774,
      "num_tokens": 43970203.0,
      "step": 20315
    },
    {
      "entropy": 6.448598718643188,
      "epoch": 2.1744341591310397,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045322171909796327,
      "loss": 5.8195,
      "mean_token_accuracy": 0.16278603821992874,
      "num_tokens": 43980778.0,
      "step": 20320
    },
    {
      "entropy": 6.470990562438965,
      "epoch": 2.1749692332388038,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045319837904132435,
      "loss": 5.802,
      "mean_token_accuracy": 0.1748654916882515,
      "num_tokens": 43992239.0,
      "step": 20325
    },
    {
      "entropy": 6.477198839187622,
      "epoch": 2.1755043073465674,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004531750338392323,
      "loss": 5.8648,
      "mean_token_accuracy": 0.16977407932281494,
      "num_tokens": 44001882.0,
      "step": 20330
    },
    {
      "entropy": 6.409822940826416,
      "epoch": 2.1760393814543315,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004531516834923611,
      "loss": 5.8394,
      "mean_token_accuracy": 0.16394179314374924,
      "num_tokens": 44013192.0,
      "step": 20335
    },
    {
      "entropy": 6.509022331237793,
      "epoch": 2.176574455562095,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004531283280013852,
      "loss": 5.8894,
      "mean_token_accuracy": 0.15883799642324448,
      "num_tokens": 44023072.0,
      "step": 20340
    },
    {
      "entropy": 6.466934490203857,
      "epoch": 2.1771095296698593,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045310496736697875,
      "loss": 5.8636,
      "mean_token_accuracy": 0.16605733036994935,
      "num_tokens": 44034238.0,
      "step": 20345
    },
    {
      "entropy": 6.474547719955444,
      "epoch": 2.1776446037776234,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045308160158981646,
      "loss": 5.8451,
      "mean_token_accuracy": 0.16260211020708085,
      "num_tokens": 44044392.0,
      "step": 20350
    },
    {
      "entropy": 6.510711240768432,
      "epoch": 2.178179677885387,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000453058230670573,
      "loss": 5.882,
      "mean_token_accuracy": 0.16680019199848176,
      "num_tokens": 44056129.0,
      "step": 20355
    },
    {
      "entropy": 6.395761585235595,
      "epoch": 2.178714751993151,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045303485460992303,
      "loss": 5.8186,
      "mean_token_accuracy": 0.17645716667175293,
      "num_tokens": 44067600.0,
      "step": 20360
    },
    {
      "entropy": 6.434391021728516,
      "epoch": 2.1792498261009148,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045301147340854177,
      "loss": 5.7332,
      "mean_token_accuracy": 0.1751248300075531,
      "num_tokens": 44077373.0,
      "step": 20365
    },
    {
      "entropy": 6.47512788772583,
      "epoch": 2.179784900208679,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045298808706710416,
      "loss": 5.9311,
      "mean_token_accuracy": 0.16311613321304322,
      "num_tokens": 44087422.0,
      "step": 20370
    },
    {
      "entropy": 6.470017147064209,
      "epoch": 2.180319974316443,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045296469558628555,
      "loss": 5.8821,
      "mean_token_accuracy": 0.16400287449359893,
      "num_tokens": 44098524.0,
      "step": 20375
    },
    {
      "entropy": 6.410015630722046,
      "epoch": 2.1808550484242066,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045294129896676134,
      "loss": 5.8026,
      "mean_token_accuracy": 0.17204558849334717,
      "num_tokens": 44108340.0,
      "step": 20380
    },
    {
      "entropy": 6.439930152893067,
      "epoch": 2.1813901225319707,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004529178972092071,
      "loss": 5.8387,
      "mean_token_accuracy": 0.15966206341981887,
      "num_tokens": 44118985.0,
      "step": 20385
    },
    {
      "entropy": 6.481768178939819,
      "epoch": 2.1819251966397344,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045289449031429873,
      "loss": 5.84,
      "mean_token_accuracy": 0.17172563970088958,
      "num_tokens": 44129126.0,
      "step": 20390
    },
    {
      "entropy": 6.461998224258423,
      "epoch": 2.1824602707474985,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045287107828271184,
      "loss": 5.8807,
      "mean_token_accuracy": 0.16899194270372392,
      "num_tokens": 44139731.0,
      "step": 20395
    },
    {
      "entropy": 6.43447699546814,
      "epoch": 2.1829953448552626,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004528476611151226,
      "loss": 5.819,
      "mean_token_accuracy": 0.16424878388643266,
      "num_tokens": 44149924.0,
      "step": 20400
    },
    {
      "entropy": 6.432070207595825,
      "epoch": 2.1835304189630262,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004528242388122071,
      "loss": 5.8176,
      "mean_token_accuracy": 0.16355322897434235,
      "num_tokens": 44161155.0,
      "step": 20405
    },
    {
      "entropy": 6.426794624328613,
      "epoch": 2.1840654930707903,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004528008113746417,
      "loss": 5.8859,
      "mean_token_accuracy": 0.1637388989329338,
      "num_tokens": 44171563.0,
      "step": 20410
    },
    {
      "entropy": 6.4031189441680905,
      "epoch": 2.1846005671785544,
      "grad_norm": 1.0,
      "learning_rate": 0.00045277737880310284,
      "loss": 5.7835,
      "mean_token_accuracy": 0.16842422783374786,
      "num_tokens": 44182719.0,
      "step": 20415
    },
    {
      "entropy": 6.432194471359253,
      "epoch": 2.185135641286318,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045275394109826715,
      "loss": 5.8078,
      "mean_token_accuracy": 0.17268998473882674,
      "num_tokens": 44192214.0,
      "step": 20420
    },
    {
      "entropy": 6.438079833984375,
      "epoch": 2.185670715394082,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045273049826081145,
      "loss": 5.8355,
      "mean_token_accuracy": 0.163115057349205,
      "num_tokens": 44204074.0,
      "step": 20425
    },
    {
      "entropy": 6.515549755096435,
      "epoch": 2.186205789501846,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004527070502914125,
      "loss": 5.9597,
      "mean_token_accuracy": 0.15994883626699447,
      "num_tokens": 44214402.0,
      "step": 20430
    },
    {
      "entropy": 6.467284393310547,
      "epoch": 2.18674086360961,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004526835971907475,
      "loss": 5.857,
      "mean_token_accuracy": 0.1631206527352333,
      "num_tokens": 44224479.0,
      "step": 20435
    },
    {
      "entropy": 6.466757488250733,
      "epoch": 2.187275937717374,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045266013895949356,
      "loss": 5.8746,
      "mean_token_accuracy": 0.1612037554383278,
      "num_tokens": 44234069.0,
      "step": 20440
    },
    {
      "entropy": 6.44365758895874,
      "epoch": 2.1878110118251377,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004526366755983281,
      "loss": 5.85,
      "mean_token_accuracy": 0.1616448849439621,
      "num_tokens": 44244978.0,
      "step": 20445
    },
    {
      "entropy": 6.42487359046936,
      "epoch": 2.188346085932902,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004526132071079286,
      "loss": 5.8184,
      "mean_token_accuracy": 0.1731092393398285,
      "num_tokens": 44255657.0,
      "step": 20450
    },
    {
      "entropy": 6.373812961578369,
      "epoch": 2.1888811600406655,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045258973348897267,
      "loss": 5.8191,
      "mean_token_accuracy": 0.1688300684094429,
      "num_tokens": 44267370.0,
      "step": 20455
    },
    {
      "entropy": 6.4339526176452635,
      "epoch": 2.1894162341484296,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004525662547421382,
      "loss": 5.8419,
      "mean_token_accuracy": 0.16625480502843856,
      "num_tokens": 44277855.0,
      "step": 20460
    },
    {
      "entropy": 6.471925830841064,
      "epoch": 2.1899513082561937,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000452542770868103,
      "loss": 5.8145,
      "mean_token_accuracy": 0.1695030227303505,
      "num_tokens": 44288233.0,
      "step": 20465
    },
    {
      "entropy": 6.437334680557251,
      "epoch": 2.1904863823639573,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045251928186754533,
      "loss": 5.8688,
      "mean_token_accuracy": 0.1709280863404274,
      "num_tokens": 44298054.0,
      "step": 20470
    },
    {
      "entropy": 6.432727670669555,
      "epoch": 2.1910214564717214,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004524957877411432,
      "loss": 5.82,
      "mean_token_accuracy": 0.16887895464897157,
      "num_tokens": 44309435.0,
      "step": 20475
    },
    {
      "entropy": 6.465598821640015,
      "epoch": 2.191556530579485,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004524722884895753,
      "loss": 5.9038,
      "mean_token_accuracy": 0.16243264973163604,
      "num_tokens": 44320502.0,
      "step": 20480
    },
    {
      "entropy": 6.388786792755127,
      "epoch": 2.192091604687249,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045244878411352,
      "loss": 5.7497,
      "mean_token_accuracy": 0.17645825892686845,
      "num_tokens": 44330911.0,
      "step": 20485
    },
    {
      "entropy": 6.418095827102661,
      "epoch": 2.1926266787950133,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004524252746136559,
      "loss": 5.8616,
      "mean_token_accuracy": 0.1577995538711548,
      "num_tokens": 44342159.0,
      "step": 20490
    },
    {
      "entropy": 6.432256078720092,
      "epoch": 2.193161752902777,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045240175999066194,
      "loss": 5.815,
      "mean_token_accuracy": 0.17039343267679213,
      "num_tokens": 44352293.0,
      "step": 20495
    },
    {
      "entropy": 6.42401089668274,
      "epoch": 2.193696827010541,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045237824024521714,
      "loss": 5.9707,
      "mean_token_accuracy": 0.16243116706609725,
      "num_tokens": 44362321.0,
      "step": 20500
    },
    {
      "entropy": 6.429323959350586,
      "epoch": 2.1942319011183047,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004523547153780005,
      "loss": 5.8459,
      "mean_token_accuracy": 0.1614599734544754,
      "num_tokens": 44372682.0,
      "step": 20505
    },
    {
      "entropy": 6.475095939636231,
      "epoch": 2.194766975226069,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004523311853896915,
      "loss": 5.8522,
      "mean_token_accuracy": 0.16537991315126419,
      "num_tokens": 44383482.0,
      "step": 20510
    },
    {
      "entropy": 6.308330440521241,
      "epoch": 2.195302049333833,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045230765028096925,
      "loss": 5.6518,
      "mean_token_accuracy": 0.1928430527448654,
      "num_tokens": 44395333.0,
      "step": 20515
    },
    {
      "entropy": 6.440374851226807,
      "epoch": 2.1958371234415965,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045228411005251366,
      "loss": 5.9108,
      "mean_token_accuracy": 0.1580565169453621,
      "num_tokens": 44406210.0,
      "step": 20520
    },
    {
      "entropy": 6.466749763488769,
      "epoch": 2.1963721975493606,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004522605647050043,
      "loss": 5.8715,
      "mean_token_accuracy": 0.1591958612203598,
      "num_tokens": 44415834.0,
      "step": 20525
    },
    {
      "entropy": 6.56388578414917,
      "epoch": 2.1969072716571247,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045223701423912103,
      "loss": 5.9039,
      "mean_token_accuracy": 0.1607590451836586,
      "num_tokens": 44427902.0,
      "step": 20530
    },
    {
      "entropy": 6.452849960327148,
      "epoch": 2.1974423457648884,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045221345865554386,
      "loss": 5.7785,
      "mean_token_accuracy": 0.16791199445724486,
      "num_tokens": 44437406.0,
      "step": 20535
    },
    {
      "entropy": 6.398679971694946,
      "epoch": 2.1979774198726525,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045218989795495294,
      "loss": 5.7951,
      "mean_token_accuracy": 0.170055028796196,
      "num_tokens": 44446889.0,
      "step": 20540
    },
    {
      "entropy": 6.45182580947876,
      "epoch": 2.198512493980416,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045216633213802866,
      "loss": 5.8818,
      "mean_token_accuracy": 0.16411300450563432,
      "num_tokens": 44457132.0,
      "step": 20545
    },
    {
      "entropy": 6.424293088912964,
      "epoch": 2.1990475680881802,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045214276120545137,
      "loss": 5.8766,
      "mean_token_accuracy": 0.16349293440580367,
      "num_tokens": 44468006.0,
      "step": 20550
    },
    {
      "entropy": 6.423907232284546,
      "epoch": 2.199582642195944,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004521191851579019,
      "loss": 5.7661,
      "mean_token_accuracy": 0.1700239434838295,
      "num_tokens": 44478650.0,
      "step": 20555
    },
    {
      "entropy": 6.411964750289917,
      "epoch": 2.200117716303708,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004520956039960607,
      "loss": 5.8755,
      "mean_token_accuracy": 0.16337748020887374,
      "num_tokens": 44489842.0,
      "step": 20560
    },
    {
      "entropy": 6.404282903671264,
      "epoch": 2.200652790411472,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045207201772060887,
      "loss": 5.8123,
      "mean_token_accuracy": 0.172593554854393,
      "num_tokens": 44500444.0,
      "step": 20565
    },
    {
      "entropy": 6.3722145557403564,
      "epoch": 2.2011878645192358,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004520484263322274,
      "loss": 5.7589,
      "mean_token_accuracy": 0.1719468802213669,
      "num_tokens": 44510711.0,
      "step": 20570
    },
    {
      "entropy": 6.398149585723877,
      "epoch": 2.201722938627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004520248298315975,
      "loss": 5.8139,
      "mean_token_accuracy": 0.1685259684920311,
      "num_tokens": 44522343.0,
      "step": 20575
    },
    {
      "entropy": 6.405088567733765,
      "epoch": 2.202258012734764,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045200122821940057,
      "loss": 5.8653,
      "mean_token_accuracy": 0.1656480312347412,
      "num_tokens": 44533218.0,
      "step": 20580
    },
    {
      "entropy": 6.46086049079895,
      "epoch": 2.2027930868425276,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045197762149631795,
      "loss": 5.9003,
      "mean_token_accuracy": 0.16250620037317276,
      "num_tokens": 44543406.0,
      "step": 20585
    },
    {
      "entropy": 6.501491832733154,
      "epoch": 2.2033281609502917,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045195400966303146,
      "loss": 5.9046,
      "mean_token_accuracy": 0.17000547796487808,
      "num_tokens": 44555329.0,
      "step": 20590
    },
    {
      "entropy": 6.486339569091797,
      "epoch": 2.2038632350580554,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004519303927202228,
      "loss": 5.9318,
      "mean_token_accuracy": 0.16505258679389953,
      "num_tokens": 44565191.0,
      "step": 20595
    },
    {
      "entropy": 6.416538286209106,
      "epoch": 2.2043983091658195,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045190677066857384,
      "loss": 5.7981,
      "mean_token_accuracy": 0.17578189820051193,
      "num_tokens": 44576374.0,
      "step": 20600
    },
    {
      "entropy": 6.389271688461304,
      "epoch": 2.2049333832735836,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004518831435087668,
      "loss": 5.7936,
      "mean_token_accuracy": 0.1706012785434723,
      "num_tokens": 44586533.0,
      "step": 20605
    },
    {
      "entropy": 6.349596261978149,
      "epoch": 2.2054684573813472,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004518595112414838,
      "loss": 5.7014,
      "mean_token_accuracy": 0.17708251476287842,
      "num_tokens": 44596753.0,
      "step": 20610
    },
    {
      "entropy": 6.49101243019104,
      "epoch": 2.2060035314891113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004518358738674073,
      "loss": 5.9515,
      "mean_token_accuracy": 0.15542939752340318,
      "num_tokens": 44607632.0,
      "step": 20615
    },
    {
      "entropy": 6.4842266082763675,
      "epoch": 2.206538605596875,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004518122313872198,
      "loss": 5.847,
      "mean_token_accuracy": 0.16618276089429856,
      "num_tokens": 44618786.0,
      "step": 20620
    },
    {
      "entropy": 6.405200386047364,
      "epoch": 2.207073679704639,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045178858380160396,
      "loss": 5.8391,
      "mean_token_accuracy": 0.15875878036022187,
      "num_tokens": 44628760.0,
      "step": 20625
    },
    {
      "entropy": 6.469834995269776,
      "epoch": 2.207608753812403,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004517649311112426,
      "loss": 5.8924,
      "mean_token_accuracy": 0.16008990854024888,
      "num_tokens": 44641206.0,
      "step": 20630
    },
    {
      "entropy": 6.492072010040284,
      "epoch": 2.208143827920167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004517412733168187,
      "loss": 5.8207,
      "mean_token_accuracy": 0.16758359521627425,
      "num_tokens": 44651923.0,
      "step": 20635
    },
    {
      "entropy": 6.420538854598999,
      "epoch": 2.208678902027931,
      "grad_norm": 1.125,
      "learning_rate": 0.00045171761041901533,
      "loss": 5.8259,
      "mean_token_accuracy": 0.1637924700975418,
      "num_tokens": 44662375.0,
      "step": 20640
    },
    {
      "entropy": 6.301134347915649,
      "epoch": 2.2092139761356946,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004516939424185158,
      "loss": 5.7504,
      "mean_token_accuracy": 0.17519594132900237,
      "num_tokens": 44673201.0,
      "step": 20645
    },
    {
      "entropy": 6.433808708190918,
      "epoch": 2.2097490502434587,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004516702693160035,
      "loss": 5.8229,
      "mean_token_accuracy": 0.16367985308170319,
      "num_tokens": 44684300.0,
      "step": 20650
    },
    {
      "entropy": 6.499784755706787,
      "epoch": 2.210284124351223,
      "grad_norm": 1.0,
      "learning_rate": 0.0004516465911121621,
      "loss": 5.8834,
      "mean_token_accuracy": 0.16349094063043595,
      "num_tokens": 44694603.0,
      "step": 20655
    },
    {
      "entropy": 6.435208034515381,
      "epoch": 2.2108191984589864,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004516229078076751,
      "loss": 5.8557,
      "mean_token_accuracy": 0.16794978231191635,
      "num_tokens": 44704461.0,
      "step": 20660
    },
    {
      "entropy": 6.456049013137817,
      "epoch": 2.2113542725667505,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004515992194032265,
      "loss": 5.9062,
      "mean_token_accuracy": 0.1645127311348915,
      "num_tokens": 44715784.0,
      "step": 20665
    },
    {
      "entropy": 6.442699003219604,
      "epoch": 2.211889346674514,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045157552589950023,
      "loss": 5.7846,
      "mean_token_accuracy": 0.1749928191304207,
      "num_tokens": 44726025.0,
      "step": 20670
    },
    {
      "entropy": 6.353748321533203,
      "epoch": 2.2124244207822783,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004515518272971805,
      "loss": 5.7631,
      "mean_token_accuracy": 0.182698056101799,
      "num_tokens": 44736602.0,
      "step": 20675
    },
    {
      "entropy": 6.4581544399261475,
      "epoch": 2.2129594948900424,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004515281235969515,
      "loss": 5.9457,
      "mean_token_accuracy": 0.15686723440885544,
      "num_tokens": 44746260.0,
      "step": 20680
    },
    {
      "entropy": 6.448196220397949,
      "epoch": 2.213494568997806,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004515044147994978,
      "loss": 5.8184,
      "mean_token_accuracy": 0.16907912194728852,
      "num_tokens": 44756894.0,
      "step": 20685
    },
    {
      "entropy": 6.440889930725097,
      "epoch": 2.21402964310557,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004514807009055039,
      "loss": 5.8766,
      "mean_token_accuracy": 0.16600103229284285,
      "num_tokens": 44768483.0,
      "step": 20690
    },
    {
      "entropy": 6.393667268753052,
      "epoch": 2.2145647172133343,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045145698191565456,
      "loss": 5.8126,
      "mean_token_accuracy": 0.16544695496559142,
      "num_tokens": 44778843.0,
      "step": 20695
    },
    {
      "entropy": 6.445354413986206,
      "epoch": 2.215099791321098,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045143325783063466,
      "loss": 5.7778,
      "mean_token_accuracy": 0.1730131447315216,
      "num_tokens": 44788659.0,
      "step": 20700
    },
    {
      "entropy": 6.460368537902832,
      "epoch": 2.215634865428862,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004514095286511293,
      "loss": 5.8655,
      "mean_token_accuracy": 0.16774796098470687,
      "num_tokens": 44799730.0,
      "step": 20705
    },
    {
      "entropy": 6.411682796478272,
      "epoch": 2.2161699395366257,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004513857943778235,
      "loss": 5.8207,
      "mean_token_accuracy": 0.1782499819993973,
      "num_tokens": 44811679.0,
      "step": 20710
    },
    {
      "entropy": 6.427549123764038,
      "epoch": 2.2167050136443898,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004513620550114027,
      "loss": 5.8415,
      "mean_token_accuracy": 0.1663948953151703,
      "num_tokens": 44821996.0,
      "step": 20715
    },
    {
      "entropy": 6.470567893981934,
      "epoch": 2.217240087752154,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045133831055255234,
      "loss": 5.894,
      "mean_token_accuracy": 0.167454794049263,
      "num_tokens": 44833713.0,
      "step": 20720
    },
    {
      "entropy": 6.474922132492066,
      "epoch": 2.2177751618599175,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004513145610019581,
      "loss": 5.8722,
      "mean_token_accuracy": 0.16825508326292038,
      "num_tokens": 44844751.0,
      "step": 20725
    },
    {
      "entropy": 6.42686505317688,
      "epoch": 2.2183102359676816,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045129080636030554,
      "loss": 5.9023,
      "mean_token_accuracy": 0.16507235169410706,
      "num_tokens": 44856089.0,
      "step": 20730
    },
    {
      "entropy": 6.441141366958618,
      "epoch": 2.2188453100754453,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004512670466282808,
      "loss": 5.8702,
      "mean_token_accuracy": 0.16796425580978394,
      "num_tokens": 44867313.0,
      "step": 20735
    },
    {
      "entropy": 6.485146427154541,
      "epoch": 2.2193803841832094,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045124328180656993,
      "loss": 5.8529,
      "mean_token_accuracy": 0.16030625998973846,
      "num_tokens": 44878430.0,
      "step": 20740
    },
    {
      "entropy": 6.41808671951294,
      "epoch": 2.2199154582909735,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045121951189585897,
      "loss": 5.7993,
      "mean_token_accuracy": 0.16671218276023864,
      "num_tokens": 44888746.0,
      "step": 20745
    },
    {
      "entropy": 6.471944427490234,
      "epoch": 2.220450532398737,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045119573689683445,
      "loss": 5.9181,
      "mean_token_accuracy": 0.16117974519729614,
      "num_tokens": 44899412.0,
      "step": 20750
    },
    {
      "entropy": 6.4456969738006595,
      "epoch": 2.2209856065065012,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045117195681018273,
      "loss": 5.8274,
      "mean_token_accuracy": 0.16702663004398347,
      "num_tokens": 44910362.0,
      "step": 20755
    },
    {
      "entropy": 6.432430410385132,
      "epoch": 2.221520680614265,
      "grad_norm": 1.125,
      "learning_rate": 0.0004511481716365905,
      "loss": 5.8134,
      "mean_token_accuracy": 0.1654360771179199,
      "num_tokens": 44920552.0,
      "step": 20760
    },
    {
      "entropy": 6.406115293502808,
      "epoch": 2.222055754722029,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004511243813767446,
      "loss": 5.821,
      "mean_token_accuracy": 0.16179666221141814,
      "num_tokens": 44932090.0,
      "step": 20765
    },
    {
      "entropy": 6.516959476470947,
      "epoch": 2.222590828829793,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000451100586031332,
      "loss": 5.8549,
      "mean_token_accuracy": 0.16495976001024246,
      "num_tokens": 44944122.0,
      "step": 20770
    },
    {
      "entropy": 6.475701427459716,
      "epoch": 2.2231259029375567,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045107678560103953,
      "loss": 5.8602,
      "mean_token_accuracy": 0.1628498837351799,
      "num_tokens": 44955104.0,
      "step": 20775
    },
    {
      "entropy": 6.437215900421142,
      "epoch": 2.223660977045321,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045105298008655485,
      "loss": 5.8504,
      "mean_token_accuracy": 0.16345833539962767,
      "num_tokens": 44965495.0,
      "step": 20780
    },
    {
      "entropy": 6.414291191101074,
      "epoch": 2.2241960511530845,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045102916948856493,
      "loss": 5.8916,
      "mean_token_accuracy": 0.16180150359869003,
      "num_tokens": 44976627.0,
      "step": 20785
    },
    {
      "entropy": 6.481628322601319,
      "epoch": 2.2247311252608486,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004510053538077576,
      "loss": 5.9057,
      "mean_token_accuracy": 0.1656995050609112,
      "num_tokens": 44987955.0,
      "step": 20790
    },
    {
      "entropy": 6.49583249092102,
      "epoch": 2.2252661993686127,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004509815330448203,
      "loss": 5.8714,
      "mean_token_accuracy": 0.16917153298854828,
      "num_tokens": 44998124.0,
      "step": 20795
    },
    {
      "entropy": 6.476839971542359,
      "epoch": 2.2258012734763764,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000450957707200441,
      "loss": 5.9148,
      "mean_token_accuracy": 0.1624368891119957,
      "num_tokens": 45007881.0,
      "step": 20800
    },
    {
      "entropy": 6.442478179931641,
      "epoch": 2.2263363475841405,
      "grad_norm": 1.0,
      "learning_rate": 0.0004509338762753076,
      "loss": 5.8378,
      "mean_token_accuracy": 0.17196283340454102,
      "num_tokens": 45018994.0,
      "step": 20805
    },
    {
      "entropy": 6.482221174240112,
      "epoch": 2.2268714216919046,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004509100402701083,
      "loss": 5.8996,
      "mean_token_accuracy": 0.1667150780558586,
      "num_tokens": 45030109.0,
      "step": 20810
    },
    {
      "entropy": 6.339191818237305,
      "epoch": 2.227406495799668,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045088619918553136,
      "loss": 5.7849,
      "mean_token_accuracy": 0.18101093620061875,
      "num_tokens": 45039872.0,
      "step": 20815
    },
    {
      "entropy": 6.466827154159546,
      "epoch": 2.2279415699074323,
      "grad_norm": 1.125,
      "learning_rate": 0.0004508623530222651,
      "loss": 5.7918,
      "mean_token_accuracy": 0.17870845198631286,
      "num_tokens": 45049216.0,
      "step": 20820
    },
    {
      "entropy": 6.386727380752563,
      "epoch": 2.228476644015196,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045083850178099805,
      "loss": 5.7742,
      "mean_token_accuracy": 0.1740702286362648,
      "num_tokens": 45059191.0,
      "step": 20825
    },
    {
      "entropy": 6.37715973854065,
      "epoch": 2.22901171812296,
      "grad_norm": 1.078125,
      "learning_rate": 0.000450814645462419,
      "loss": 5.8079,
      "mean_token_accuracy": 0.17061534523963928,
      "num_tokens": 45069212.0,
      "step": 20830
    },
    {
      "entropy": 6.42135796546936,
      "epoch": 2.229546792230724,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045079078406721676,
      "loss": 5.7981,
      "mean_token_accuracy": 0.1667047455906868,
      "num_tokens": 45080248.0,
      "step": 20835
    },
    {
      "entropy": 6.438169384002686,
      "epoch": 2.230081866338488,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045076691759608035,
      "loss": 5.808,
      "mean_token_accuracy": 0.16439444273710252,
      "num_tokens": 45091315.0,
      "step": 20840
    },
    {
      "entropy": 6.410001182556153,
      "epoch": 2.230616940446252,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045074304604969886,
      "loss": 5.8382,
      "mean_token_accuracy": 0.16643691658973694,
      "num_tokens": 45102674.0,
      "step": 20845
    },
    {
      "entropy": 6.4012268543243405,
      "epoch": 2.2311520145540156,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045071916942876163,
      "loss": 5.8226,
      "mean_token_accuracy": 0.16765946745872498,
      "num_tokens": 45113171.0,
      "step": 20850
    },
    {
      "entropy": 6.42986650466919,
      "epoch": 2.2316870886617797,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004506952877339581,
      "loss": 5.8178,
      "mean_token_accuracy": 0.16588219702243806,
      "num_tokens": 45122912.0,
      "step": 20855
    },
    {
      "entropy": 6.482905769348145,
      "epoch": 2.2322221627695438,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004506714009659778,
      "loss": 5.8425,
      "mean_token_accuracy": 0.17204290628433228,
      "num_tokens": 45133839.0,
      "step": 20860
    },
    {
      "entropy": 6.440778589248657,
      "epoch": 2.2327572368773074,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004506475091255105,
      "loss": 5.8425,
      "mean_token_accuracy": 0.16942700892686843,
      "num_tokens": 45144914.0,
      "step": 20865
    },
    {
      "entropy": 6.407672357559204,
      "epoch": 2.2332923109850715,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000450623612213246,
      "loss": 5.8378,
      "mean_token_accuracy": 0.1650838851928711,
      "num_tokens": 45155194.0,
      "step": 20870
    },
    {
      "entropy": 6.490751647949219,
      "epoch": 2.233827385092835,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045059971022987446,
      "loss": 5.943,
      "mean_token_accuracy": 0.15966898575425148,
      "num_tokens": 45166771.0,
      "step": 20875
    },
    {
      "entropy": 6.497862052917481,
      "epoch": 2.2343624592005993,
      "grad_norm": 1.125,
      "learning_rate": 0.00045057580317608585,
      "loss": 5.7931,
      "mean_token_accuracy": 0.16808295994997025,
      "num_tokens": 45177236.0,
      "step": 20880
    },
    {
      "entropy": 6.452960681915283,
      "epoch": 2.2348975333083634,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045055189105257064,
      "loss": 5.8171,
      "mean_token_accuracy": 0.16003155708312988,
      "num_tokens": 45187382.0,
      "step": 20885
    },
    {
      "entropy": 6.440205240249634,
      "epoch": 2.235432607416127,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045052797386001936,
      "loss": 5.8587,
      "mean_token_accuracy": 0.16732324063777923,
      "num_tokens": 45197287.0,
      "step": 20890
    },
    {
      "entropy": 6.503533554077149,
      "epoch": 2.235967681523891,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004505040515991224,
      "loss": 5.805,
      "mean_token_accuracy": 0.1725912094116211,
      "num_tokens": 45208737.0,
      "step": 20895
    },
    {
      "entropy": 6.502198982238769,
      "epoch": 2.236502755631655,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045048012427057057,
      "loss": 5.9338,
      "mean_token_accuracy": 0.15847990661859512,
      "num_tokens": 45219586.0,
      "step": 20900
    },
    {
      "entropy": 6.494469547271729,
      "epoch": 2.237037829739419,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004504561918750549,
      "loss": 5.9163,
      "mean_token_accuracy": 0.15536018013954161,
      "num_tokens": 45230730.0,
      "step": 20905
    },
    {
      "entropy": 6.4553038597106935,
      "epoch": 2.237572903847183,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004504322544132663,
      "loss": 5.7455,
      "mean_token_accuracy": 0.181632137298584,
      "num_tokens": 45241330.0,
      "step": 20910
    },
    {
      "entropy": 6.371163654327392,
      "epoch": 2.2381079779549466,
      "grad_norm": 1.09375,
      "learning_rate": 0.000450408311885896,
      "loss": 5.8579,
      "mean_token_accuracy": 0.166632542014122,
      "num_tokens": 45251439.0,
      "step": 20915
    },
    {
      "entropy": 6.367977952957153,
      "epoch": 2.2386430520627107,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045038436429363537,
      "loss": 5.6603,
      "mean_token_accuracy": 0.17886811047792434,
      "num_tokens": 45262078.0,
      "step": 20920
    },
    {
      "entropy": 6.472654008865357,
      "epoch": 2.2391781261704744,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045036041163717586,
      "loss": 5.8668,
      "mean_token_accuracy": 0.17053494304418565,
      "num_tokens": 45272466.0,
      "step": 20925
    },
    {
      "entropy": 6.456154727935791,
      "epoch": 2.2397132002782385,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004503364539172091,
      "loss": 5.7916,
      "mean_token_accuracy": 0.16948889642953874,
      "num_tokens": 45281830.0,
      "step": 20930
    },
    {
      "entropy": 6.385391712188721,
      "epoch": 2.2402482743860026,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004503124911344268,
      "loss": 5.8758,
      "mean_token_accuracy": 0.15645403191447257,
      "num_tokens": 45293026.0,
      "step": 20935
    },
    {
      "entropy": 6.4380992412567135,
      "epoch": 2.2407833484937663,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045028852328952106,
      "loss": 5.846,
      "mean_token_accuracy": 0.16659156531095504,
      "num_tokens": 45303532.0,
      "step": 20940
    },
    {
      "entropy": 6.38600435256958,
      "epoch": 2.2413184226015304,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004502645503831838,
      "loss": 5.7231,
      "mean_token_accuracy": 0.17625644207000732,
      "num_tokens": 45314448.0,
      "step": 20945
    },
    {
      "entropy": 6.438287353515625,
      "epoch": 2.241853496709294,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045024057241610727,
      "loss": 5.8707,
      "mean_token_accuracy": 0.154929082095623,
      "num_tokens": 45323194.0,
      "step": 20950
    },
    {
      "entropy": 6.423833513259888,
      "epoch": 2.242388570817058,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004502165893889839,
      "loss": 5.7362,
      "mean_token_accuracy": 0.1858434185385704,
      "num_tokens": 45333464.0,
      "step": 20955
    },
    {
      "entropy": 6.4138407707214355,
      "epoch": 2.242923644924822,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004501926013025061,
      "loss": 5.8661,
      "mean_token_accuracy": 0.1673394963145256,
      "num_tokens": 45344364.0,
      "step": 20960
    },
    {
      "entropy": 6.435335111618042,
      "epoch": 2.243458719032586,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045016860815736646,
      "loss": 5.7902,
      "mean_token_accuracy": 0.17355676591396332,
      "num_tokens": 45355202.0,
      "step": 20965
    },
    {
      "entropy": 6.401474523544311,
      "epoch": 2.24399379314035,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004501446099542579,
      "loss": 5.7623,
      "mean_token_accuracy": 0.17118320018053054,
      "num_tokens": 45365696.0,
      "step": 20970
    },
    {
      "entropy": 6.382426118850708,
      "epoch": 2.244528867248114,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004501206066938734,
      "loss": 5.7492,
      "mean_token_accuracy": 0.17957000732421874,
      "num_tokens": 45376609.0,
      "step": 20975
    },
    {
      "entropy": 6.4876604080200195,
      "epoch": 2.2450639413558777,
      "grad_norm": 1.125,
      "learning_rate": 0.00045009659837690606,
      "loss": 5.9126,
      "mean_token_accuracy": 0.15603215098381043,
      "num_tokens": 45386158.0,
      "step": 20980
    },
    {
      "entropy": 6.509737491607666,
      "epoch": 2.245599015463642,
      "grad_norm": 1.0625,
      "learning_rate": 0.000450072585004049,
      "loss": 5.9136,
      "mean_token_accuracy": 0.1651718258857727,
      "num_tokens": 45396824.0,
      "step": 20985
    },
    {
      "entropy": 6.475045394897461,
      "epoch": 2.2461340895714055,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004500485665759955,
      "loss": 5.8298,
      "mean_token_accuracy": 0.1679166227579117,
      "num_tokens": 45407219.0,
      "step": 20990
    },
    {
      "entropy": 6.428127336502075,
      "epoch": 2.2466691636791696,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004500245430934394,
      "loss": 5.8639,
      "mean_token_accuracy": 0.1632233366370201,
      "num_tokens": 45418572.0,
      "step": 20995
    },
    {
      "entropy": 6.402622699737549,
      "epoch": 2.2472042377869337,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045000051455707413,
      "loss": 5.7928,
      "mean_token_accuracy": 0.16608012467622757,
      "num_tokens": 45429643.0,
      "step": 21000
    },
    {
      "epoch": 2.2472042377869337,
      "eval_entropy": 6.246916919641418,
      "eval_loss": 5.957085609436035,
      "eval_mean_token_accuracy": 0.1694445715727651,
      "eval_num_tokens": 45429643.0,
      "eval_runtime": 17.5046,
      "eval_samples_per_second": 1695.494,
      "eval_steps_per_second": 211.944,
      "step": 21000
    },
    {
      "entropy": 6.496895217895508,
      "epoch": 2.2477393118946973,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004499764809675937,
      "loss": 5.798,
      "mean_token_accuracy": 0.17210119664669038,
      "num_tokens": 45440183.0,
      "step": 21005
    },
    {
      "entropy": 6.415399074554443,
      "epoch": 2.2482743860024614,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004499524423256918,
      "loss": 5.8525,
      "mean_token_accuracy": 0.16382916867733002,
      "num_tokens": 45451252.0,
      "step": 21010
    },
    {
      "entropy": 6.35843677520752,
      "epoch": 2.248809460110225,
      "grad_norm": 1.125,
      "learning_rate": 0.00044992839863206284,
      "loss": 5.8665,
      "mean_token_accuracy": 0.16280582100152968,
      "num_tokens": 45462336.0,
      "step": 21015
    },
    {
      "entropy": 6.4185715675354,
      "epoch": 2.249344534217989,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044990434988740095,
      "loss": 5.8437,
      "mean_token_accuracy": 0.1662568300962448,
      "num_tokens": 45472526.0,
      "step": 21020
    },
    {
      "entropy": 6.502649211883545,
      "epoch": 2.2498796083257533,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004498802960924005,
      "loss": 5.8901,
      "mean_token_accuracy": 0.1589934468269348,
      "num_tokens": 45483281.0,
      "step": 21025
    },
    {
      "entropy": 6.464981889724731,
      "epoch": 2.250414682433517,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004498562372477561,
      "loss": 5.8031,
      "mean_token_accuracy": 0.17400548756122589,
      "num_tokens": 45494258.0,
      "step": 21030
    },
    {
      "entropy": 6.448374843597412,
      "epoch": 2.250949756541281,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004498321733541624,
      "loss": 5.8435,
      "mean_token_accuracy": 0.16741989105939864,
      "num_tokens": 45504233.0,
      "step": 21035
    },
    {
      "entropy": 6.408311271667481,
      "epoch": 2.2514848306490447,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044980810441231424,
      "loss": 5.8299,
      "mean_token_accuracy": 0.16709035485982895,
      "num_tokens": 45513902.0,
      "step": 21040
    },
    {
      "entropy": 6.4236959457397464,
      "epoch": 2.252019904756809,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004497840304229066,
      "loss": 5.8852,
      "mean_token_accuracy": 0.1663428857922554,
      "num_tokens": 45524111.0,
      "step": 21045
    },
    {
      "entropy": 6.417786359786987,
      "epoch": 2.252554978864573,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044975995138663466,
      "loss": 5.8207,
      "mean_token_accuracy": 0.1719718024134636,
      "num_tokens": 45534816.0,
      "step": 21050
    },
    {
      "entropy": 6.458803129196167,
      "epoch": 2.2530900529723366,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044973586730419373,
      "loss": 5.8085,
      "mean_token_accuracy": 0.17092955708503724,
      "num_tokens": 45546012.0,
      "step": 21055
    },
    {
      "entropy": 6.425588464736938,
      "epoch": 2.2536251270801007,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004497117781762792,
      "loss": 5.8087,
      "mean_token_accuracy": 0.17197504490613938,
      "num_tokens": 45555402.0,
      "step": 21060
    },
    {
      "entropy": 6.412817859649659,
      "epoch": 2.2541602011878643,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004496876840035866,
      "loss": 5.8476,
      "mean_token_accuracy": 0.16660054475069047,
      "num_tokens": 45565537.0,
      "step": 21065
    },
    {
      "entropy": 6.459649896621704,
      "epoch": 2.2546952752956284,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044966358478681163,
      "loss": 5.8789,
      "mean_token_accuracy": 0.16713212579488754,
      "num_tokens": 45576449.0,
      "step": 21070
    },
    {
      "entropy": 6.584536027908325,
      "epoch": 2.2552303494033925,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004496394805266502,
      "loss": 5.8597,
      "mean_token_accuracy": 0.15943341851234435,
      "num_tokens": 45586477.0,
      "step": 21075
    },
    {
      "entropy": 6.398723697662353,
      "epoch": 2.255765423511156,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044961537122379844,
      "loss": 5.8692,
      "mean_token_accuracy": 0.16340886056423187,
      "num_tokens": 45597279.0,
      "step": 21080
    },
    {
      "entropy": 6.43121919631958,
      "epoch": 2.2563004976189203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004495912568789522,
      "loss": 5.8861,
      "mean_token_accuracy": 0.16256649643182755,
      "num_tokens": 45608216.0,
      "step": 21085
    },
    {
      "entropy": 6.421452379226684,
      "epoch": 2.2568355717266844,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044956713749280805,
      "loss": 5.782,
      "mean_token_accuracy": 0.1687426283955574,
      "num_tokens": 45620665.0,
      "step": 21090
    },
    {
      "entropy": 6.453617525100708,
      "epoch": 2.257370645834448,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044954301306606234,
      "loss": 5.8492,
      "mean_token_accuracy": 0.16583899706602095,
      "num_tokens": 45631635.0,
      "step": 21095
    },
    {
      "entropy": 6.346903228759766,
      "epoch": 2.257905719942212,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044951888359941156,
      "loss": 5.7506,
      "mean_token_accuracy": 0.17418387830257415,
      "num_tokens": 45642507.0,
      "step": 21100
    },
    {
      "entropy": 6.464037895202637,
      "epoch": 2.2584407940499758,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044949474909355264,
      "loss": 5.8334,
      "mean_token_accuracy": 0.16723829060792922,
      "num_tokens": 45653123.0,
      "step": 21105
    },
    {
      "entropy": 6.479783201217652,
      "epoch": 2.25897586815774,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044947060954918233,
      "loss": 5.8271,
      "mean_token_accuracy": 0.17063153237104417,
      "num_tokens": 45663228.0,
      "step": 21110
    },
    {
      "entropy": 6.463861274719238,
      "epoch": 2.2595109422655035,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004494464649669976,
      "loss": 5.8263,
      "mean_token_accuracy": 0.16718240678310395,
      "num_tokens": 45673389.0,
      "step": 21115
    },
    {
      "entropy": 6.436802959442138,
      "epoch": 2.2600460163732676,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044942231534769587,
      "loss": 5.8314,
      "mean_token_accuracy": 0.16874895095825196,
      "num_tokens": 45684803.0,
      "step": 21120
    },
    {
      "entropy": 6.439955186843872,
      "epoch": 2.2605810904810317,
      "grad_norm": 1.125,
      "learning_rate": 0.00044939816069197415,
      "loss": 5.8957,
      "mean_token_accuracy": 0.17100303024053573,
      "num_tokens": 45695164.0,
      "step": 21125
    },
    {
      "entropy": 6.485132932662964,
      "epoch": 2.2611161645887954,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044937400100053017,
      "loss": 5.8762,
      "mean_token_accuracy": 0.17109508365392684,
      "num_tokens": 45706169.0,
      "step": 21130
    },
    {
      "entropy": 6.50259575843811,
      "epoch": 2.2616512386965595,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004493498362740613,
      "loss": 5.9414,
      "mean_token_accuracy": 0.15841606706380845,
      "num_tokens": 45717299.0,
      "step": 21135
    },
    {
      "entropy": 6.403090572357177,
      "epoch": 2.2621863128043236,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044932566651326546,
      "loss": 5.8163,
      "mean_token_accuracy": 0.17101982831954957,
      "num_tokens": 45728815.0,
      "step": 21140
    },
    {
      "entropy": 6.453882455825806,
      "epoch": 2.2627213869120872,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004493014917188405,
      "loss": 5.8749,
      "mean_token_accuracy": 0.16922252327203752,
      "num_tokens": 45740293.0,
      "step": 21145
    },
    {
      "entropy": 6.495658302307129,
      "epoch": 2.2632564610198513,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044927731189148447,
      "loss": 5.8193,
      "mean_token_accuracy": 0.167028246819973,
      "num_tokens": 45751243.0,
      "step": 21150
    },
    {
      "entropy": 6.3442617893219,
      "epoch": 2.263791535127615,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044925312703189546,
      "loss": 5.7993,
      "mean_token_accuracy": 0.17360675632953643,
      "num_tokens": 45762477.0,
      "step": 21155
    },
    {
      "entropy": 6.47194619178772,
      "epoch": 2.264326609235379,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000449228937140772,
      "loss": 5.8271,
      "mean_token_accuracy": 0.16666753143072127,
      "num_tokens": 45772929.0,
      "step": 21160
    },
    {
      "entropy": 6.454291915893554,
      "epoch": 2.264861683343143,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044920474221881243,
      "loss": 5.8395,
      "mean_token_accuracy": 0.16825288981199266,
      "num_tokens": 45782999.0,
      "step": 21165
    },
    {
      "entropy": 6.370670938491822,
      "epoch": 2.265396757450907,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044918054226671545,
      "loss": 5.8418,
      "mean_token_accuracy": 0.1709567129611969,
      "num_tokens": 45793020.0,
      "step": 21170
    },
    {
      "entropy": 6.433105182647705,
      "epoch": 2.265931831558671,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044915633728517975,
      "loss": 5.7846,
      "mean_token_accuracy": 0.1798429772257805,
      "num_tokens": 45803020.0,
      "step": 21175
    },
    {
      "entropy": 6.472886657714843,
      "epoch": 2.2664669056664346,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004491321272749042,
      "loss": 5.8682,
      "mean_token_accuracy": 0.16811501383781433,
      "num_tokens": 45812700.0,
      "step": 21180
    },
    {
      "entropy": 6.430380487442017,
      "epoch": 2.2670019797741987,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044910791223658805,
      "loss": 5.8952,
      "mean_token_accuracy": 0.1647849753499031,
      "num_tokens": 45823097.0,
      "step": 21185
    },
    {
      "entropy": 6.456098031997681,
      "epoch": 2.267537053881963,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044908369217093035,
      "loss": 5.819,
      "mean_token_accuracy": 0.16333086043596268,
      "num_tokens": 45832935.0,
      "step": 21190
    },
    {
      "entropy": 6.494571018218994,
      "epoch": 2.2680721279897265,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044905946707863057,
      "loss": 5.9316,
      "mean_token_accuracy": 0.15822085738182068,
      "num_tokens": 45844296.0,
      "step": 21195
    },
    {
      "entropy": 6.496649646759034,
      "epoch": 2.2686072020974906,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004490352369603881,
      "loss": 5.8515,
      "mean_token_accuracy": 0.162521068751812,
      "num_tokens": 45854942.0,
      "step": 21200
    },
    {
      "entropy": 6.44248366355896,
      "epoch": 2.2691422762052547,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044901100181690256,
      "loss": 5.8986,
      "mean_token_accuracy": 0.1692359372973442,
      "num_tokens": 45866427.0,
      "step": 21205
    },
    {
      "entropy": 6.432834959030151,
      "epoch": 2.2696773503130183,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044898676164887387,
      "loss": 5.8833,
      "mean_token_accuracy": 0.16969145983457565,
      "num_tokens": 45877955.0,
      "step": 21210
    },
    {
      "entropy": 6.376030969619751,
      "epoch": 2.2702124244207824,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044896251645700185,
      "loss": 5.745,
      "mean_token_accuracy": 0.17416718006134033,
      "num_tokens": 45888668.0,
      "step": 21215
    },
    {
      "entropy": 6.439992094039917,
      "epoch": 2.270747498528546,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004489382662419866,
      "loss": 5.8496,
      "mean_token_accuracy": 0.16741416305303575,
      "num_tokens": 45898819.0,
      "step": 21220
    },
    {
      "entropy": 6.464386749267578,
      "epoch": 2.27128257263631,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004489140110045283,
      "loss": 5.8902,
      "mean_token_accuracy": 0.16523895859718324,
      "num_tokens": 45909390.0,
      "step": 21225
    },
    {
      "entropy": 6.447940301895142,
      "epoch": 2.271817646744074,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004488897507453275,
      "loss": 5.8131,
      "mean_token_accuracy": 0.16444972306489944,
      "num_tokens": 45919697.0,
      "step": 21230
    },
    {
      "entropy": 6.448425960540772,
      "epoch": 2.272352720851838,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044886548546508453,
      "loss": 5.8987,
      "mean_token_accuracy": 0.16847995519638062,
      "num_tokens": 45929458.0,
      "step": 21235
    },
    {
      "entropy": 6.37803282737732,
      "epoch": 2.272887794959602,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004488412151645001,
      "loss": 5.7746,
      "mean_token_accuracy": 0.170806984603405,
      "num_tokens": 45939373.0,
      "step": 21240
    },
    {
      "entropy": 6.3704430103302006,
      "epoch": 2.2734228690673657,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044881693984427504,
      "loss": 5.8359,
      "mean_token_accuracy": 0.1684233769774437,
      "num_tokens": 45949954.0,
      "step": 21245
    },
    {
      "entropy": 6.464352369308472,
      "epoch": 2.27395794317513,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004487926595051102,
      "loss": 5.8897,
      "mean_token_accuracy": 0.16342808157205582,
      "num_tokens": 45960147.0,
      "step": 21250
    },
    {
      "entropy": 6.462734222412109,
      "epoch": 2.274493017282894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004487683741477068,
      "loss": 5.8001,
      "mean_token_accuracy": 0.17157927453517913,
      "num_tokens": 45970718.0,
      "step": 21255
    },
    {
      "entropy": 6.444784498214721,
      "epoch": 2.2750280913906575,
      "grad_norm": 1.0625,
      "learning_rate": 0.000448744083772766,
      "loss": 5.842,
      "mean_token_accuracy": 0.1678796112537384,
      "num_tokens": 45981631.0,
      "step": 21260
    },
    {
      "entropy": 6.4806070804595945,
      "epoch": 2.2755631654984216,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044871978838098917,
      "loss": 5.8915,
      "mean_token_accuracy": 0.1633464515209198,
      "num_tokens": 45992295.0,
      "step": 21265
    },
    {
      "entropy": 6.501303672790527,
      "epoch": 2.2760982396061853,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004486954879730778,
      "loss": 5.8867,
      "mean_token_accuracy": 0.16129858642816544,
      "num_tokens": 46002222.0,
      "step": 21270
    },
    {
      "entropy": 6.3983550548553465,
      "epoch": 2.2766333137139494,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004486711825497337,
      "loss": 5.7341,
      "mean_token_accuracy": 0.18486716151237487,
      "num_tokens": 46012393.0,
      "step": 21275
    },
    {
      "entropy": 6.4163655757904055,
      "epoch": 2.2771683878217135,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044864687211165863,
      "loss": 5.847,
      "mean_token_accuracy": 0.16614428013563157,
      "num_tokens": 46022095.0,
      "step": 21280
    },
    {
      "entropy": 6.426615285873413,
      "epoch": 2.277703461929477,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044862255665955447,
      "loss": 5.8039,
      "mean_token_accuracy": 0.16470236331224442,
      "num_tokens": 46032973.0,
      "step": 21285
    },
    {
      "entropy": 6.448262548446655,
      "epoch": 2.2782385360372412,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004485982361941234,
      "loss": 5.7974,
      "mean_token_accuracy": 0.16937133818864822,
      "num_tokens": 46043430.0,
      "step": 21290
    },
    {
      "entropy": 6.403768157958984,
      "epoch": 2.278773610145005,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004485739107160677,
      "loss": 5.8173,
      "mean_token_accuracy": 0.17114774733781815,
      "num_tokens": 46054580.0,
      "step": 21295
    },
    {
      "entropy": 6.428949880599975,
      "epoch": 2.279308684252769,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044854958022608974,
      "loss": 5.8507,
      "mean_token_accuracy": 0.17133500128984452,
      "num_tokens": 46064898.0,
      "step": 21300
    },
    {
      "entropy": 6.39233660697937,
      "epoch": 2.279843758360533,
      "grad_norm": 1.0,
      "learning_rate": 0.00044852524472489196,
      "loss": 5.8077,
      "mean_token_accuracy": 0.16954635083675385,
      "num_tokens": 46075985.0,
      "step": 21305
    },
    {
      "entropy": 6.437289381027222,
      "epoch": 2.2803788324682968,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004485009042131771,
      "loss": 5.8976,
      "mean_token_accuracy": 0.16624248921871185,
      "num_tokens": 46087544.0,
      "step": 21310
    },
    {
      "entropy": 6.478307151794434,
      "epoch": 2.280913906576061,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044847655869164813,
      "loss": 5.8414,
      "mean_token_accuracy": 0.16502964198589326,
      "num_tokens": 46097613.0,
      "step": 21315
    },
    {
      "entropy": 6.370608901977539,
      "epoch": 2.281448980683825,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004484522081610078,
      "loss": 5.7601,
      "mean_token_accuracy": 0.1811158090829849,
      "num_tokens": 46107728.0,
      "step": 21320
    },
    {
      "entropy": 6.375819349288941,
      "epoch": 2.2819840547915886,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044842785262195936,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1654544234275818,
      "num_tokens": 46118309.0,
      "step": 21325
    },
    {
      "entropy": 6.401549530029297,
      "epoch": 2.2825191288993527,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044840349207520604,
      "loss": 5.8346,
      "mean_token_accuracy": 0.1730056956410408,
      "num_tokens": 46128425.0,
      "step": 21330
    },
    {
      "entropy": 6.508475112915039,
      "epoch": 2.2830542030071164,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004483791265214513,
      "loss": 5.93,
      "mean_token_accuracy": 0.1631891831755638,
      "num_tokens": 46139506.0,
      "step": 21335
    },
    {
      "entropy": 6.446938180923462,
      "epoch": 2.2835892771148805,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044835475596139855,
      "loss": 5.8118,
      "mean_token_accuracy": 0.16921110451221466,
      "num_tokens": 46151226.0,
      "step": 21340
    },
    {
      "entropy": 6.467024517059326,
      "epoch": 2.284124351222644,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004483303803957516,
      "loss": 5.8696,
      "mean_token_accuracy": 0.1638282358646393,
      "num_tokens": 46162749.0,
      "step": 21345
    },
    {
      "entropy": 6.437602090835571,
      "epoch": 2.2846594253304082,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004483059998252144,
      "loss": 5.8718,
      "mean_token_accuracy": 0.166001196205616,
      "num_tokens": 46174802.0,
      "step": 21350
    },
    {
      "entropy": 6.452194929122925,
      "epoch": 2.2851944994381723,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044828161425049067,
      "loss": 5.8097,
      "mean_token_accuracy": 0.1713239148259163,
      "num_tokens": 46185579.0,
      "step": 21355
    },
    {
      "entropy": 6.4678308963775635,
      "epoch": 2.285729573545936,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044825722367228477,
      "loss": 5.9555,
      "mean_token_accuracy": 0.1575504347681999,
      "num_tokens": 46196399.0,
      "step": 21360
    },
    {
      "entropy": 6.463662433624267,
      "epoch": 2.2862646476537,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044823282809130084,
      "loss": 5.8643,
      "mean_token_accuracy": 0.15974075347185135,
      "num_tokens": 46206931.0,
      "step": 21365
    },
    {
      "entropy": 6.4574010372161865,
      "epoch": 2.286799721761464,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044820842750824334,
      "loss": 5.8356,
      "mean_token_accuracy": 0.17132906317710878,
      "num_tokens": 46217327.0,
      "step": 21370
    },
    {
      "entropy": 6.414389753341675,
      "epoch": 2.287334795869228,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044818402192381684,
      "loss": 5.8428,
      "mean_token_accuracy": 0.16908527165651321,
      "num_tokens": 46227981.0,
      "step": 21375
    },
    {
      "entropy": 6.427244186401367,
      "epoch": 2.287869869976992,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044815961133872606,
      "loss": 5.9015,
      "mean_token_accuracy": 0.15993898510932922,
      "num_tokens": 46238500.0,
      "step": 21380
    },
    {
      "entropy": 6.414719676971435,
      "epoch": 2.2884049440847556,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044813519575367586,
      "loss": 5.7781,
      "mean_token_accuracy": 0.172661454975605,
      "num_tokens": 46249530.0,
      "step": 21385
    },
    {
      "entropy": 6.468621730804443,
      "epoch": 2.2889400181925197,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004481107751693711,
      "loss": 5.7281,
      "mean_token_accuracy": 0.17486495524644852,
      "num_tokens": 46260272.0,
      "step": 21390
    },
    {
      "entropy": 6.383783626556396,
      "epoch": 2.2894750923002833,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004480863495865171,
      "loss": 5.8106,
      "mean_token_accuracy": 0.1652208685874939,
      "num_tokens": 46271805.0,
      "step": 21395
    },
    {
      "entropy": 6.424065780639649,
      "epoch": 2.2900101664080474,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004480619190058191,
      "loss": 5.8672,
      "mean_token_accuracy": 0.17042919844388962,
      "num_tokens": 46282211.0,
      "step": 21400
    },
    {
      "entropy": 6.3966615200042725,
      "epoch": 2.2905452405158115,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044803748342798247,
      "loss": 5.7744,
      "mean_token_accuracy": 0.16849643588066102,
      "num_tokens": 46292287.0,
      "step": 21405
    },
    {
      "entropy": 6.451309537887573,
      "epoch": 2.291080314623575,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044801304285371286,
      "loss": 5.8894,
      "mean_token_accuracy": 0.162578783929348,
      "num_tokens": 46302978.0,
      "step": 21410
    },
    {
      "entropy": 6.426165676116943,
      "epoch": 2.2916153887313393,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004479885972837159,
      "loss": 5.8556,
      "mean_token_accuracy": 0.16298803687095642,
      "num_tokens": 46313958.0,
      "step": 21415
    },
    {
      "entropy": 6.501826238632202,
      "epoch": 2.2921504628391034,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004479641467186975,
      "loss": 5.8783,
      "mean_token_accuracy": 0.16137643307447433,
      "num_tokens": 46325262.0,
      "step": 21420
    },
    {
      "entropy": 6.453494215011597,
      "epoch": 2.292685536946867,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044793969115936375,
      "loss": 5.8963,
      "mean_token_accuracy": 0.16465636640787124,
      "num_tokens": 46336773.0,
      "step": 21425
    },
    {
      "entropy": 6.420994424819947,
      "epoch": 2.293220611054631,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004479152306064207,
      "loss": 5.8438,
      "mean_token_accuracy": 0.16657519042491914,
      "num_tokens": 46346616.0,
      "step": 21430
    },
    {
      "entropy": 6.410976600646973,
      "epoch": 2.293755685162395,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004478907650605745,
      "loss": 5.8539,
      "mean_token_accuracy": 0.1691579595208168,
      "num_tokens": 46358564.0,
      "step": 21435
    },
    {
      "entropy": 6.407894086837769,
      "epoch": 2.294290759270159,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044786629452253196,
      "loss": 5.7314,
      "mean_token_accuracy": 0.1759527027606964,
      "num_tokens": 46369207.0,
      "step": 21440
    },
    {
      "entropy": 6.473242330551147,
      "epoch": 2.294825833377923,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004478418189929994,
      "loss": 5.8357,
      "mean_token_accuracy": 0.17033011615276336,
      "num_tokens": 46379382.0,
      "step": 21445
    },
    {
      "entropy": 6.399757003784179,
      "epoch": 2.2953609074856867,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044781733847268357,
      "loss": 5.9175,
      "mean_token_accuracy": 0.16238942742347717,
      "num_tokens": 46390738.0,
      "step": 21450
    },
    {
      "entropy": 6.403687524795532,
      "epoch": 2.2958959815934508,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044779285296229137,
      "loss": 5.7344,
      "mean_token_accuracy": 0.17409809082746505,
      "num_tokens": 46401740.0,
      "step": 21455
    },
    {
      "entropy": 6.4796491146087645,
      "epoch": 2.2964310557012144,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044776836246252985,
      "loss": 5.7859,
      "mean_token_accuracy": 0.16951306611299516,
      "num_tokens": 46412009.0,
      "step": 21460
    },
    {
      "entropy": 6.452200412750244,
      "epoch": 2.2969661298089785,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004477438669741061,
      "loss": 5.905,
      "mean_token_accuracy": 0.16457393169403076,
      "num_tokens": 46422550.0,
      "step": 21465
    },
    {
      "entropy": 6.460613107681274,
      "epoch": 2.2975012039167426,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004477193664977275,
      "loss": 5.8806,
      "mean_token_accuracy": 0.1676934242248535,
      "num_tokens": 46432797.0,
      "step": 21470
    },
    {
      "entropy": 6.350198602676391,
      "epoch": 2.2980362780245063,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004476948610341015,
      "loss": 5.6772,
      "mean_token_accuracy": 0.17776494473218918,
      "num_tokens": 46444638.0,
      "step": 21475
    },
    {
      "entropy": 6.4028106212615965,
      "epoch": 2.2985713521322704,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044767035058393564,
      "loss": 5.782,
      "mean_token_accuracy": 0.17503172308206558,
      "num_tokens": 46455404.0,
      "step": 21480
    },
    {
      "entropy": 6.442708539962768,
      "epoch": 2.2991064262400345,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044764583514793766,
      "loss": 5.8753,
      "mean_token_accuracy": 0.16657989621162414,
      "num_tokens": 46466821.0,
      "step": 21485
    },
    {
      "entropy": 6.479094934463501,
      "epoch": 2.299641500347798,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004476213147268154,
      "loss": 5.8425,
      "mean_token_accuracy": 0.1666167676448822,
      "num_tokens": 46477222.0,
      "step": 21490
    },
    {
      "entropy": 6.424090194702148,
      "epoch": 2.3001765744555622,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044759678932127693,
      "loss": 5.8063,
      "mean_token_accuracy": 0.1699430137872696,
      "num_tokens": 46488596.0,
      "step": 21495
    },
    {
      "entropy": 6.4277551651000975,
      "epoch": 2.300711648563326,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044757225893203047,
      "loss": 5.854,
      "mean_token_accuracy": 0.16397102326154708,
      "num_tokens": 46499799.0,
      "step": 21500
    },
    {
      "entropy": 6.4868675708770756,
      "epoch": 2.30124672267109,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004475477235597843,
      "loss": 5.8928,
      "mean_token_accuracy": 0.16855161935091018,
      "num_tokens": 46511077.0,
      "step": 21505
    },
    {
      "entropy": 6.476409816741944,
      "epoch": 2.3017817967788536,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004475231832052469,
      "loss": 5.9165,
      "mean_token_accuracy": 0.15583207905292512,
      "num_tokens": 46523517.0,
      "step": 21510
    },
    {
      "entropy": 6.426363229751587,
      "epoch": 2.3023168708866177,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004474986378691267,
      "loss": 5.8104,
      "mean_token_accuracy": 0.17498249858617781,
      "num_tokens": 46533198.0,
      "step": 21515
    },
    {
      "entropy": 6.455925369262696,
      "epoch": 2.302851944994382,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044747408755213265,
      "loss": 5.832,
      "mean_token_accuracy": 0.1665938526391983,
      "num_tokens": 46544659.0,
      "step": 21520
    },
    {
      "entropy": 6.433456373214722,
      "epoch": 2.3033870191021455,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044744953225497354,
      "loss": 5.8181,
      "mean_token_accuracy": 0.1638501226902008,
      "num_tokens": 46555897.0,
      "step": 21525
    },
    {
      "entropy": 6.427370309829712,
      "epoch": 2.3039220932099096,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044742497197835845,
      "loss": 5.8303,
      "mean_token_accuracy": 0.16559610068798064,
      "num_tokens": 46565695.0,
      "step": 21530
    },
    {
      "entropy": 6.481076002120972,
      "epoch": 2.3044571673176737,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004474004067229965,
      "loss": 5.8861,
      "mean_token_accuracy": 0.16090465039014817,
      "num_tokens": 46576620.0,
      "step": 21535
    },
    {
      "entropy": 6.545639181137085,
      "epoch": 2.3049922414254373,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000447375836489597,
      "loss": 5.9356,
      "mean_token_accuracy": 0.16195329278707504,
      "num_tokens": 46588264.0,
      "step": 21540
    },
    {
      "entropy": 6.470568990707397,
      "epoch": 2.3055273155332014,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044735126127886944,
      "loss": 5.8128,
      "mean_token_accuracy": 0.17410054355859755,
      "num_tokens": 46600180.0,
      "step": 21545
    },
    {
      "entropy": 6.440237092971802,
      "epoch": 2.306062389640965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004473266810915235,
      "loss": 5.8473,
      "mean_token_accuracy": 0.1676609694957733,
      "num_tokens": 46610391.0,
      "step": 21550
    },
    {
      "entropy": 6.4677740097045895,
      "epoch": 2.306597463748729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004473020959282688,
      "loss": 5.8959,
      "mean_token_accuracy": 0.16724032908678055,
      "num_tokens": 46622955.0,
      "step": 21555
    },
    {
      "entropy": 6.477942371368409,
      "epoch": 2.3071325378564933,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044727750578981535,
      "loss": 5.8757,
      "mean_token_accuracy": 0.16935707479715348,
      "num_tokens": 46632361.0,
      "step": 21560
    },
    {
      "entropy": 6.451447868347168,
      "epoch": 2.307667611964257,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004472529106768731,
      "loss": 5.8166,
      "mean_token_accuracy": 0.16881420463323593,
      "num_tokens": 46642693.0,
      "step": 21565
    },
    {
      "entropy": 6.3909485816955565,
      "epoch": 2.308202686072021,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004472283105901522,
      "loss": 5.7373,
      "mean_token_accuracy": 0.17096330523490905,
      "num_tokens": 46653251.0,
      "step": 21570
    },
    {
      "entropy": 6.480831813812256,
      "epoch": 2.3087377601797847,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004472037055303631,
      "loss": 5.9086,
      "mean_token_accuracy": 0.15893737599253654,
      "num_tokens": 46664751.0,
      "step": 21575
    },
    {
      "entropy": 6.406520891189575,
      "epoch": 2.309272834287549,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004471790954982162,
      "loss": 5.7635,
      "mean_token_accuracy": 0.16671787202358246,
      "num_tokens": 46676825.0,
      "step": 21580
    },
    {
      "entropy": 6.417887115478516,
      "epoch": 2.309807908395313,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004471544804944221,
      "loss": 5.7818,
      "mean_token_accuracy": 0.17565637826919556,
      "num_tokens": 46687202.0,
      "step": 21585
    },
    {
      "entropy": 6.447641658782959,
      "epoch": 2.3103429825030766,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004471298605196916,
      "loss": 5.9223,
      "mean_token_accuracy": 0.16300245523452758,
      "num_tokens": 46697687.0,
      "step": 21590
    },
    {
      "entropy": 6.436089038848877,
      "epoch": 2.3108780566108407,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044710523557473544,
      "loss": 5.7847,
      "mean_token_accuracy": 0.1744150325655937,
      "num_tokens": 46709088.0,
      "step": 21595
    },
    {
      "entropy": 6.462968397140503,
      "epoch": 2.3114131307186048,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044708060566026487,
      "loss": 5.8694,
      "mean_token_accuracy": 0.16859321892261506,
      "num_tokens": 46721675.0,
      "step": 21600
    },
    {
      "entropy": 6.464574909210205,
      "epoch": 2.3119482048263684,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000447055970776991,
      "loss": 5.8829,
      "mean_token_accuracy": 0.16767400801181792,
      "num_tokens": 46731854.0,
      "step": 21605
    },
    {
      "entropy": 6.424655294418335,
      "epoch": 2.3124832789341325,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044703133092562514,
      "loss": 5.8889,
      "mean_token_accuracy": 0.1617003098130226,
      "num_tokens": 46742815.0,
      "step": 21610
    },
    {
      "entropy": 6.414730072021484,
      "epoch": 2.313018353041896,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044700668610687884,
      "loss": 5.8436,
      "mean_token_accuracy": 0.17423504292964936,
      "num_tokens": 46753012.0,
      "step": 21615
    },
    {
      "entropy": 6.392689943313599,
      "epoch": 2.3135534271496603,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044698203632146353,
      "loss": 5.835,
      "mean_token_accuracy": 0.1717800095677376,
      "num_tokens": 46763701.0,
      "step": 21620
    },
    {
      "entropy": 6.490646457672119,
      "epoch": 2.314088501257424,
      "grad_norm": 1.125,
      "learning_rate": 0.0004469573815700911,
      "loss": 5.8989,
      "mean_token_accuracy": 0.1596858024597168,
      "num_tokens": 46775407.0,
      "step": 21625
    },
    {
      "entropy": 6.4010645866394045,
      "epoch": 2.314623575365188,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044693272185347357,
      "loss": 5.7613,
      "mean_token_accuracy": 0.1694840520620346,
      "num_tokens": 46786282.0,
      "step": 21630
    },
    {
      "entropy": 6.417765712738037,
      "epoch": 2.315158649472952,
      "grad_norm": 1.125,
      "learning_rate": 0.0004469080571723228,
      "loss": 5.8696,
      "mean_token_accuracy": 0.16838824301958083,
      "num_tokens": 46796768.0,
      "step": 21635
    },
    {
      "entropy": 6.420572090148926,
      "epoch": 2.315693723580716,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044688338752735095,
      "loss": 5.7897,
      "mean_token_accuracy": 0.16938841193914414,
      "num_tokens": 46806538.0,
      "step": 21640
    },
    {
      "entropy": 6.4180646419525145,
      "epoch": 2.31622879768848,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004468587129192705,
      "loss": 5.7565,
      "mean_token_accuracy": 0.17957530915737152,
      "num_tokens": 46816692.0,
      "step": 21645
    },
    {
      "entropy": 6.42611083984375,
      "epoch": 2.316763871796244,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044683403334879384,
      "loss": 5.8373,
      "mean_token_accuracy": 0.17771997749805452,
      "num_tokens": 46827551.0,
      "step": 21650
    },
    {
      "entropy": 6.452655410766601,
      "epoch": 2.3172989459040076,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004468093488166337,
      "loss": 5.9669,
      "mean_token_accuracy": 0.15549242198467256,
      "num_tokens": 46840168.0,
      "step": 21655
    },
    {
      "entropy": 6.444498538970947,
      "epoch": 2.3178340200117717,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044678465932350273,
      "loss": 5.8431,
      "mean_token_accuracy": 0.1674739584326744,
      "num_tokens": 46851086.0,
      "step": 21660
    },
    {
      "entropy": 6.387170362472534,
      "epoch": 2.3183690941195354,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044675996487011373,
      "loss": 5.778,
      "mean_token_accuracy": 0.17828713208436966,
      "num_tokens": 46861830.0,
      "step": 21665
    },
    {
      "entropy": 6.450907754898071,
      "epoch": 2.3189041682272995,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044673526545718,
      "loss": 5.8831,
      "mean_token_accuracy": 0.1703964203596115,
      "num_tokens": 46871348.0,
      "step": 21670
    },
    {
      "entropy": 6.408540487289429,
      "epoch": 2.3194392423350636,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004467105610854146,
      "loss": 5.8385,
      "mean_token_accuracy": 0.16815986335277558,
      "num_tokens": 46882737.0,
      "step": 21675
    },
    {
      "entropy": 6.503811931610107,
      "epoch": 2.3199743164428273,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044668585175553085,
      "loss": 5.911,
      "mean_token_accuracy": 0.15355028957128525,
      "num_tokens": 46893616.0,
      "step": 21680
    },
    {
      "entropy": 6.496567058563232,
      "epoch": 2.3205093905505914,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004466611374682423,
      "loss": 5.8415,
      "mean_token_accuracy": 0.16592923700809478,
      "num_tokens": 46904546.0,
      "step": 21685
    },
    {
      "entropy": 6.4354469776153564,
      "epoch": 2.321044464658355,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044663641822426244,
      "loss": 5.8273,
      "mean_token_accuracy": 0.1690298244357109,
      "num_tokens": 46915513.0,
      "step": 21690
    },
    {
      "entropy": 6.508310842514038,
      "epoch": 2.321579538766119,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044661169402430517,
      "loss": 5.898,
      "mean_token_accuracy": 0.1655866324901581,
      "num_tokens": 46925535.0,
      "step": 21695
    },
    {
      "entropy": 6.401183319091797,
      "epoch": 2.322114612873883,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044658696486908443,
      "loss": 5.8712,
      "mean_token_accuracy": 0.16758636385202408,
      "num_tokens": 46936222.0,
      "step": 21700
    },
    {
      "entropy": 6.425313138961792,
      "epoch": 2.322649686981647,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004465622307593141,
      "loss": 5.7587,
      "mean_token_accuracy": 0.1809204936027527,
      "num_tokens": 46946855.0,
      "step": 21705
    },
    {
      "entropy": 6.4568651676177975,
      "epoch": 2.323184761089411,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044653749169570845,
      "loss": 5.849,
      "mean_token_accuracy": 0.16760074943304062,
      "num_tokens": 46958246.0,
      "step": 21710
    },
    {
      "entropy": 6.379929304122925,
      "epoch": 2.3237198351971746,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044651274767898186,
      "loss": 5.8483,
      "mean_token_accuracy": 0.16726115494966506,
      "num_tokens": 46969237.0,
      "step": 21715
    },
    {
      "entropy": 6.505489683151245,
      "epoch": 2.3242549093049387,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004464879987098488,
      "loss": 5.909,
      "mean_token_accuracy": 0.1650969371199608,
      "num_tokens": 46979897.0,
      "step": 21720
    },
    {
      "entropy": 6.517658805847168,
      "epoch": 2.324789983412703,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004464632447890238,
      "loss": 5.7958,
      "mean_token_accuracy": 0.1745147690176964,
      "num_tokens": 46989833.0,
      "step": 21725
    },
    {
      "entropy": 6.376440572738647,
      "epoch": 2.3253250575204665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044643848591722183,
      "loss": 5.7988,
      "mean_token_accuracy": 0.17370236665010452,
      "num_tokens": 46999822.0,
      "step": 21730
    },
    {
      "entropy": 6.364861869812012,
      "epoch": 2.3258601316282306,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044641372209515764,
      "loss": 5.8175,
      "mean_token_accuracy": 0.17249365746974946,
      "num_tokens": 47010176.0,
      "step": 21735
    },
    {
      "entropy": 6.40304594039917,
      "epoch": 2.3263952057359942,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004463889533235462,
      "loss": 5.8193,
      "mean_token_accuracy": 0.17188781797885894,
      "num_tokens": 47020184.0,
      "step": 21740
    },
    {
      "entropy": 6.480036497116089,
      "epoch": 2.3269302798437583,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044636417960310294,
      "loss": 5.8967,
      "mean_token_accuracy": 0.16480078548192978,
      "num_tokens": 47032066.0,
      "step": 21745
    },
    {
      "entropy": 6.509715843200683,
      "epoch": 2.3274653539515224,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004463394009345431,
      "loss": 5.9163,
      "mean_token_accuracy": 0.16276931762695312,
      "num_tokens": 47043571.0,
      "step": 21750
    },
    {
      "entropy": 6.488185214996338,
      "epoch": 2.328000428059286,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044631461731858195,
      "loss": 5.8337,
      "mean_token_accuracy": 0.16957143396139146,
      "num_tokens": 47053813.0,
      "step": 21755
    },
    {
      "entropy": 6.44535002708435,
      "epoch": 2.32853550216705,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044628982875593546,
      "loss": 5.8841,
      "mean_token_accuracy": 0.16439603567123412,
      "num_tokens": 47063671.0,
      "step": 21760
    },
    {
      "entropy": 6.372188758850098,
      "epoch": 2.3290705762748143,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044626503524731917,
      "loss": 5.777,
      "mean_token_accuracy": 0.17434757351875305,
      "num_tokens": 47073537.0,
      "step": 21765
    },
    {
      "entropy": 6.435751676559448,
      "epoch": 2.329605650382578,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004462402367934491,
      "loss": 5.8282,
      "mean_token_accuracy": 0.1688222438097,
      "num_tokens": 47084272.0,
      "step": 21770
    },
    {
      "entropy": 6.471650123596191,
      "epoch": 2.330140724490342,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004462154333950412,
      "loss": 5.8924,
      "mean_token_accuracy": 0.16651787757873535,
      "num_tokens": 47095198.0,
      "step": 21775
    },
    {
      "entropy": 6.5011293411254885,
      "epoch": 2.3306757985981057,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004461906250528117,
      "loss": 5.9171,
      "mean_token_accuracy": 0.16287408024072647,
      "num_tokens": 47106540.0,
      "step": 21780
    },
    {
      "entropy": 6.474879169464112,
      "epoch": 2.33121087270587,
      "grad_norm": 1.125,
      "learning_rate": 0.000446165811767477,
      "loss": 5.8977,
      "mean_token_accuracy": 0.17181480675935745,
      "num_tokens": 47117301.0,
      "step": 21785
    },
    {
      "entropy": 6.421233177185059,
      "epoch": 2.3317459468136335,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004461409935397536,
      "loss": 5.8771,
      "mean_token_accuracy": 0.16813919693231583,
      "num_tokens": 47127058.0,
      "step": 21790
    },
    {
      "entropy": 6.425158166885376,
      "epoch": 2.3322810209213976,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044611617037035794,
      "loss": 5.8274,
      "mean_token_accuracy": 0.16581108272075654,
      "num_tokens": 47137182.0,
      "step": 21795
    },
    {
      "entropy": 6.464474248886108,
      "epoch": 2.3328160950291617,
      "grad_norm": 1.0,
      "learning_rate": 0.000446091342260007,
      "loss": 5.8863,
      "mean_token_accuracy": 0.1601739376783371,
      "num_tokens": 47148591.0,
      "step": 21800
    },
    {
      "entropy": 6.455053091049194,
      "epoch": 2.3333511691369253,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044606650920941753,
      "loss": 5.8528,
      "mean_token_accuracy": 0.16382087469100953,
      "num_tokens": 47159014.0,
      "step": 21805
    },
    {
      "entropy": 6.449406290054322,
      "epoch": 2.3338862432446894,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004460416712193067,
      "loss": 5.874,
      "mean_token_accuracy": 0.16721478700637818,
      "num_tokens": 47168760.0,
      "step": 21810
    },
    {
      "entropy": 6.427296257019043,
      "epoch": 2.3344213173524535,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004460168282903916,
      "loss": 5.8829,
      "mean_token_accuracy": 0.16440934240818023,
      "num_tokens": 47180902.0,
      "step": 21815
    },
    {
      "entropy": 6.403178930282593,
      "epoch": 2.334956391460217,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044599198042338963,
      "loss": 5.7502,
      "mean_token_accuracy": 0.17401932179927826,
      "num_tokens": 47190458.0,
      "step": 21820
    },
    {
      "entropy": 6.4261116027832035,
      "epoch": 2.3354914655679813,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044596712761901824,
      "loss": 5.8611,
      "mean_token_accuracy": 0.16407979130744935,
      "num_tokens": 47201099.0,
      "step": 21825
    },
    {
      "entropy": 6.398235702514649,
      "epoch": 2.336026539675745,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004459422698779951,
      "loss": 5.8282,
      "mean_token_accuracy": 0.17399132400751113,
      "num_tokens": 47211284.0,
      "step": 21830
    },
    {
      "entropy": 6.450828790664673,
      "epoch": 2.336561613783509,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004459174072010379,
      "loss": 5.9034,
      "mean_token_accuracy": 0.16285787969827653,
      "num_tokens": 47222875.0,
      "step": 21835
    },
    {
      "entropy": 6.433865642547607,
      "epoch": 2.337096687891273,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004458925395888646,
      "loss": 5.8713,
      "mean_token_accuracy": 0.16458399295806886,
      "num_tokens": 47234238.0,
      "step": 21840
    },
    {
      "entropy": 6.448133659362793,
      "epoch": 2.3376317619990368,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044586766704219326,
      "loss": 5.8777,
      "mean_token_accuracy": 0.1648955151438713,
      "num_tokens": 47244634.0,
      "step": 21845
    },
    {
      "entropy": 6.5146510124206545,
      "epoch": 2.338166836106801,
      "grad_norm": 1.0625,
      "learning_rate": 0.000445842789561742,
      "loss": 5.886,
      "mean_token_accuracy": 0.16186895966529846,
      "num_tokens": 47256538.0,
      "step": 21850
    },
    {
      "entropy": 6.375972509384155,
      "epoch": 2.3387019102145645,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004458179071482292,
      "loss": 5.7168,
      "mean_token_accuracy": 0.18118776679039,
      "num_tokens": 47266373.0,
      "step": 21855
    },
    {
      "entropy": 6.379014158248902,
      "epoch": 2.3392369843223286,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044579301980237333,
      "loss": 5.9048,
      "mean_token_accuracy": 0.1581457793712616,
      "num_tokens": 47276441.0,
      "step": 21860
    },
    {
      "entropy": 6.442761659622192,
      "epoch": 2.3397720584300927,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004457681275248931,
      "loss": 5.8924,
      "mean_token_accuracy": 0.1617085114121437,
      "num_tokens": 47287299.0,
      "step": 21865
    },
    {
      "entropy": 6.416817903518677,
      "epoch": 2.3403071325378564,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044574323031650715,
      "loss": 5.7635,
      "mean_token_accuracy": 0.16709764450788497,
      "num_tokens": 47298007.0,
      "step": 21870
    },
    {
      "entropy": 6.3689916133880615,
      "epoch": 2.3408422066456205,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004457183281779343,
      "loss": 5.7358,
      "mean_token_accuracy": 0.18001555949449538,
      "num_tokens": 47309729.0,
      "step": 21875
    },
    {
      "entropy": 6.457218837738037,
      "epoch": 2.3413772807533846,
      "grad_norm": 1.125,
      "learning_rate": 0.00044569342110989385,
      "loss": 5.8232,
      "mean_token_accuracy": 0.1661243975162506,
      "num_tokens": 47319050.0,
      "step": 21880
    },
    {
      "entropy": 6.476080083847046,
      "epoch": 2.3419123548611482,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004456685091131048,
      "loss": 5.9543,
      "mean_token_accuracy": 0.16508494913578034,
      "num_tokens": 47330365.0,
      "step": 21885
    },
    {
      "entropy": 6.525357341766357,
      "epoch": 2.3424474289689123,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044564359218828654,
      "loss": 5.8674,
      "mean_token_accuracy": 0.16277400106191636,
      "num_tokens": 47341225.0,
      "step": 21890
    },
    {
      "entropy": 6.438167953491211,
      "epoch": 2.342982503076676,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044561867033615854,
      "loss": 5.7866,
      "mean_token_accuracy": 0.17383818775415422,
      "num_tokens": 47352996.0,
      "step": 21895
    },
    {
      "entropy": 6.372847080230713,
      "epoch": 2.34351757718444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004455937435574404,
      "loss": 5.753,
      "mean_token_accuracy": 0.16943469494581223,
      "num_tokens": 47363577.0,
      "step": 21900
    },
    {
      "entropy": 6.493923759460449,
      "epoch": 2.3440526512922037,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044556881185285185,
      "loss": 5.9778,
      "mean_token_accuracy": 0.16075832545757293,
      "num_tokens": 47374389.0,
      "step": 21905
    },
    {
      "entropy": 6.408408164978027,
      "epoch": 2.344587725399968,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044554387522311286,
      "loss": 5.8005,
      "mean_token_accuracy": 0.17404745072126387,
      "num_tokens": 47385624.0,
      "step": 21910
    },
    {
      "entropy": 6.456030797958374,
      "epoch": 2.345122799507732,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004455189336689435,
      "loss": 5.8713,
      "mean_token_accuracy": 0.16448942571878433,
      "num_tokens": 47396611.0,
      "step": 21915
    },
    {
      "entropy": 6.499788093566894,
      "epoch": 2.3456578736154956,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044549398719106374,
      "loss": 5.8571,
      "mean_token_accuracy": 0.16886600404977797,
      "num_tokens": 47406693.0,
      "step": 21920
    },
    {
      "entropy": 6.408180475234985,
      "epoch": 2.3461929477232597,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004454690357901942,
      "loss": 5.7783,
      "mean_token_accuracy": 0.17017623037099838,
      "num_tokens": 47417528.0,
      "step": 21925
    },
    {
      "entropy": 6.450545120239258,
      "epoch": 2.346728021831024,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004454440794670552,
      "loss": 5.8889,
      "mean_token_accuracy": 0.16924383640289306,
      "num_tokens": 47430032.0,
      "step": 21930
    },
    {
      "entropy": 6.486557292938232,
      "epoch": 2.3472630959387875,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044541911822236724,
      "loss": 5.9263,
      "mean_token_accuracy": 0.15711178183555602,
      "num_tokens": 47441002.0,
      "step": 21935
    },
    {
      "entropy": 6.467084836959839,
      "epoch": 2.3477981700465516,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004453941520568513,
      "loss": 5.7847,
      "mean_token_accuracy": 0.17764919400215148,
      "num_tokens": 47451716.0,
      "step": 21940
    },
    {
      "entropy": 6.442506599426269,
      "epoch": 2.348333244154315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004453691809712281,
      "loss": 5.8418,
      "mean_token_accuracy": 0.16321540772914886,
      "num_tokens": 47463039.0,
      "step": 21945
    },
    {
      "entropy": 6.345868921279907,
      "epoch": 2.3488683182620793,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044534420496621877,
      "loss": 5.7526,
      "mean_token_accuracy": 0.17428285926580428,
      "num_tokens": 47473836.0,
      "step": 21950
    },
    {
      "entropy": 6.40151014328003,
      "epoch": 2.3494033923698434,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044531922404254445,
      "loss": 5.8169,
      "mean_token_accuracy": 0.170709066092968,
      "num_tokens": 47485103.0,
      "step": 21955
    },
    {
      "entropy": 6.370221996307373,
      "epoch": 2.349938466477607,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004452942382009265,
      "loss": 5.781,
      "mean_token_accuracy": 0.17253333628177642,
      "num_tokens": 47495509.0,
      "step": 21960
    },
    {
      "entropy": 6.464000034332275,
      "epoch": 2.350473540585371,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004452692474420863,
      "loss": 5.8394,
      "mean_token_accuracy": 0.1728108510375023,
      "num_tokens": 47506062.0,
      "step": 21965
    },
    {
      "entropy": 6.386452531814575,
      "epoch": 2.351008614693135,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004452442517667455,
      "loss": 5.7953,
      "mean_token_accuracy": 0.17401609867811202,
      "num_tokens": 47515632.0,
      "step": 21970
    },
    {
      "entropy": 6.364739656448364,
      "epoch": 2.351543688800899,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004452192511756259,
      "loss": 5.7411,
      "mean_token_accuracy": 0.17336448282003403,
      "num_tokens": 47525412.0,
      "step": 21975
    },
    {
      "entropy": 6.412469053268433,
      "epoch": 2.352078762908663,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044519424566944935,
      "loss": 5.8155,
      "mean_token_accuracy": 0.16784504503011705,
      "num_tokens": 47536094.0,
      "step": 21980
    },
    {
      "entropy": 6.4463988780975345,
      "epoch": 2.3526138370164267,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044516923524893777,
      "loss": 5.7941,
      "mean_token_accuracy": 0.16841497272253036,
      "num_tokens": 47546893.0,
      "step": 21985
    },
    {
      "entropy": 6.4480022430419925,
      "epoch": 2.353148911124191,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044514421991481354,
      "loss": 5.7894,
      "mean_token_accuracy": 0.17134130746126175,
      "num_tokens": 47557117.0,
      "step": 21990
    },
    {
      "entropy": 6.424397039413452,
      "epoch": 2.3536839852319544,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044511919966779885,
      "loss": 5.7774,
      "mean_token_accuracy": 0.17065892666578292,
      "num_tokens": 47567878.0,
      "step": 21995
    },
    {
      "entropy": 6.4445373058319095,
      "epoch": 2.3542190593397185,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044509417450861615,
      "loss": 5.8796,
      "mean_token_accuracy": 0.1624042272567749,
      "num_tokens": 47578185.0,
      "step": 22000
    },
    {
      "entropy": 6.496693229675293,
      "epoch": 2.3547541334474826,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004450691444379881,
      "loss": 5.9487,
      "mean_token_accuracy": 0.15874558985233306,
      "num_tokens": 47590042.0,
      "step": 22005
    },
    {
      "entropy": 6.451704740524292,
      "epoch": 2.3552892075552463,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044504410945663735,
      "loss": 5.8522,
      "mean_token_accuracy": 0.1662849634885788,
      "num_tokens": 47601225.0,
      "step": 22010
    },
    {
      "entropy": 6.4237360000610355,
      "epoch": 2.3558242816630104,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044501906956528686,
      "loss": 5.8564,
      "mean_token_accuracy": 0.1678178384900093,
      "num_tokens": 47612596.0,
      "step": 22015
    },
    {
      "entropy": 6.500827693939209,
      "epoch": 2.356359355770774,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004449940247646596,
      "loss": 5.8963,
      "mean_token_accuracy": 0.1667765870690346,
      "num_tokens": 47623156.0,
      "step": 22020
    },
    {
      "entropy": 6.4540282726287845,
      "epoch": 2.356894429878538,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044496897505547886,
      "loss": 5.816,
      "mean_token_accuracy": 0.16890366226434708,
      "num_tokens": 47633743.0,
      "step": 22025
    },
    {
      "entropy": 6.466969108581543,
      "epoch": 2.3574295039863022,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004449439204384677,
      "loss": 5.887,
      "mean_token_accuracy": 0.16450097411870956,
      "num_tokens": 47643914.0,
      "step": 22030
    },
    {
      "entropy": 6.472386693954467,
      "epoch": 2.357964578094066,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044491886091434983,
      "loss": 5.8649,
      "mean_token_accuracy": 0.1646573856472969,
      "num_tokens": 47655187.0,
      "step": 22035
    },
    {
      "entropy": 6.483145809173584,
      "epoch": 2.35849965220183,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004448937964838487,
      "loss": 5.8345,
      "mean_token_accuracy": 0.16760612428188323,
      "num_tokens": 47667193.0,
      "step": 22040
    },
    {
      "entropy": 6.431054782867432,
      "epoch": 2.359034726309594,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004448687271476881,
      "loss": 5.8843,
      "mean_token_accuracy": 0.16533220261335374,
      "num_tokens": 47678116.0,
      "step": 22045
    },
    {
      "entropy": 6.462834644317627,
      "epoch": 2.3595698004173578,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004448436529065918,
      "loss": 5.8255,
      "mean_token_accuracy": 0.1685374543070793,
      "num_tokens": 47689883.0,
      "step": 22050
    },
    {
      "entropy": 6.457907247543335,
      "epoch": 2.360104874525122,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004448185737612839,
      "loss": 5.8595,
      "mean_token_accuracy": 0.16717949211597444,
      "num_tokens": 47699939.0,
      "step": 22055
    },
    {
      "entropy": 6.355057334899902,
      "epoch": 2.3606399486328855,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044479348971248863,
      "loss": 5.7993,
      "mean_token_accuracy": 0.17200499027967453,
      "num_tokens": 47710477.0,
      "step": 22060
    },
    {
      "entropy": 6.4648374080657955,
      "epoch": 2.3611750227406496,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004447684007609301,
      "loss": 5.8471,
      "mean_token_accuracy": 0.16380867809057237,
      "num_tokens": 47720548.0,
      "step": 22065
    },
    {
      "entropy": 6.465596675872803,
      "epoch": 2.3617100968484133,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044474330690733293,
      "loss": 5.8289,
      "mean_token_accuracy": 0.1623891368508339,
      "num_tokens": 47732135.0,
      "step": 22070
    },
    {
      "entropy": 6.455597448348999,
      "epoch": 2.3622451709561774,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004447182081524216,
      "loss": 5.8567,
      "mean_token_accuracy": 0.16380796879529952,
      "num_tokens": 47742934.0,
      "step": 22075
    },
    {
      "entropy": 6.443974494934082,
      "epoch": 2.3627802450639415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044469310449692085,
      "loss": 5.8525,
      "mean_token_accuracy": 0.1649388000369072,
      "num_tokens": 47754586.0,
      "step": 22080
    },
    {
      "entropy": 6.497480154037476,
      "epoch": 2.363315319171705,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044466799594155556,
      "loss": 5.8557,
      "mean_token_accuracy": 0.16547672152519227,
      "num_tokens": 47765564.0,
      "step": 22085
    },
    {
      "entropy": 6.385167407989502,
      "epoch": 2.363850393279469,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044464288248705075,
      "loss": 5.8216,
      "mean_token_accuracy": 0.17077055275440217,
      "num_tokens": 47776191.0,
      "step": 22090
    },
    {
      "entropy": 6.506978511810303,
      "epoch": 2.3643854673872333,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044461776413413153,
      "loss": 5.9475,
      "mean_token_accuracy": 0.15783039033412932,
      "num_tokens": 47787193.0,
      "step": 22095
    },
    {
      "entropy": 6.549927854537964,
      "epoch": 2.364920541494997,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004445926408835232,
      "loss": 5.8839,
      "mean_token_accuracy": 0.16220196187496186,
      "num_tokens": 47798576.0,
      "step": 22100
    },
    {
      "entropy": 6.3720263004302975,
      "epoch": 2.365455615602761,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004445675127359511,
      "loss": 5.7685,
      "mean_token_accuracy": 0.17350470423698425,
      "num_tokens": 47809519.0,
      "step": 22105
    },
    {
      "entropy": 6.4054113864898685,
      "epoch": 2.3659906897105247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000444542379692141,
      "loss": 5.804,
      "mean_token_accuracy": 0.17124270796775817,
      "num_tokens": 47820618.0,
      "step": 22110
    },
    {
      "entropy": 6.4523578643798825,
      "epoch": 2.366525763818289,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004445172417528186,
      "loss": 5.8849,
      "mean_token_accuracy": 0.16552897095680236,
      "num_tokens": 47831756.0,
      "step": 22115
    },
    {
      "entropy": 6.5104865550994875,
      "epoch": 2.367060837926053,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044449209891870954,
      "loss": 5.8451,
      "mean_token_accuracy": 0.16619703769683838,
      "num_tokens": 47841974.0,
      "step": 22120
    },
    {
      "entropy": 6.394980382919312,
      "epoch": 2.3675959120338166,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044446695119053993,
      "loss": 5.7718,
      "mean_token_accuracy": 0.1695953607559204,
      "num_tokens": 47852194.0,
      "step": 22125
    },
    {
      "entropy": 6.442676973342896,
      "epoch": 2.3681309861415807,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000444441798569036,
      "loss": 5.921,
      "mean_token_accuracy": 0.16324637532234193,
      "num_tokens": 47863921.0,
      "step": 22130
    },
    {
      "entropy": 6.544261741638183,
      "epoch": 2.3686660602493443,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044441664105492393,
      "loss": 6.0381,
      "mean_token_accuracy": 0.15552888810634613,
      "num_tokens": 47875520.0,
      "step": 22135
    },
    {
      "entropy": 6.5026774406433105,
      "epoch": 2.3692011343571084,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004443914786489301,
      "loss": 5.7839,
      "mean_token_accuracy": 0.17120393514633178,
      "num_tokens": 47886047.0,
      "step": 22140
    },
    {
      "entropy": 6.377101278305053,
      "epoch": 2.3697362084648725,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044436631135178113,
      "loss": 5.7629,
      "mean_token_accuracy": 0.17743367552757264,
      "num_tokens": 47896449.0,
      "step": 22145
    },
    {
      "entropy": 6.491523599624633,
      "epoch": 2.370271282572636,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044434113916420377,
      "loss": 5.8724,
      "mean_token_accuracy": 0.16783621162176132,
      "num_tokens": 47908613.0,
      "step": 22150
    },
    {
      "entropy": 6.534091901779175,
      "epoch": 2.3708063566804003,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044431596208692475,
      "loss": 5.9462,
      "mean_token_accuracy": 0.15863492488861083,
      "num_tokens": 47919474.0,
      "step": 22155
    },
    {
      "entropy": 6.40099687576294,
      "epoch": 2.3713414307881644,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004442907801206712,
      "loss": 5.8252,
      "mean_token_accuracy": 0.16872695982456207,
      "num_tokens": 47930929.0,
      "step": 22160
    },
    {
      "entropy": 6.421558523178101,
      "epoch": 2.371876504895928,
      "grad_norm": 1.0,
      "learning_rate": 0.00044426559326617013,
      "loss": 5.7854,
      "mean_token_accuracy": 0.17402136623859404,
      "num_tokens": 47942050.0,
      "step": 22165
    },
    {
      "entropy": 6.496225023269654,
      "epoch": 2.372411579003692,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004442404015241488,
      "loss": 5.9016,
      "mean_token_accuracy": 0.16265181005001067,
      "num_tokens": 47953162.0,
      "step": 22170
    },
    {
      "entropy": 6.513951349258423,
      "epoch": 2.372946653111456,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044421520489533467,
      "loss": 5.8677,
      "mean_token_accuracy": 0.1657484859228134,
      "num_tokens": 47963140.0,
      "step": 22175
    },
    {
      "entropy": 6.4284868240356445,
      "epoch": 2.37348172721922,
      "grad_norm": 1.734375,
      "learning_rate": 0.00044419000338045527,
      "loss": 5.8896,
      "mean_token_accuracy": 0.1677924521267414,
      "num_tokens": 47975622.0,
      "step": 22180
    },
    {
      "entropy": 6.482190561294556,
      "epoch": 2.3740168013269836,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004441647969802383,
      "loss": 5.8523,
      "mean_token_accuracy": 0.1661618173122406,
      "num_tokens": 47986450.0,
      "step": 22185
    },
    {
      "entropy": 6.513070201873779,
      "epoch": 2.3745518754347477,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004441395856954117,
      "loss": 5.9309,
      "mean_token_accuracy": 0.15813746601343154,
      "num_tokens": 47997441.0,
      "step": 22190
    },
    {
      "entropy": 6.4650966167449955,
      "epoch": 2.3750869495425118,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044411436952670325,
      "loss": 5.8211,
      "mean_token_accuracy": 0.16200606822967528,
      "num_tokens": 48008995.0,
      "step": 22195
    },
    {
      "entropy": 6.463963460922241,
      "epoch": 2.3756220236502754,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044408914847484103,
      "loss": 5.8438,
      "mean_token_accuracy": 0.17137261778116225,
      "num_tokens": 48019975.0,
      "step": 22200
    },
    {
      "entropy": 6.483383989334106,
      "epoch": 2.3761570977580395,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004440639225405536,
      "loss": 5.8513,
      "mean_token_accuracy": 0.16690831184387206,
      "num_tokens": 48031278.0,
      "step": 22205
    },
    {
      "entropy": 6.3954887866973875,
      "epoch": 2.3766921718658036,
      "grad_norm": 1.078125,
      "learning_rate": 0.000444038691724569,
      "loss": 5.8278,
      "mean_token_accuracy": 0.17035578787326813,
      "num_tokens": 48042050.0,
      "step": 22210
    },
    {
      "entropy": 6.3397266387939455,
      "epoch": 2.3772272459735673,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044401345602761605,
      "loss": 5.7316,
      "mean_token_accuracy": 0.17713741660118104,
      "num_tokens": 48052833.0,
      "step": 22215
    },
    {
      "entropy": 6.382109117507935,
      "epoch": 2.3777623200813314,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044398821545042327,
      "loss": 5.8374,
      "mean_token_accuracy": 0.1670564278960228,
      "num_tokens": 48063425.0,
      "step": 22220
    },
    {
      "entropy": 6.36842098236084,
      "epoch": 2.378297394189095,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044396296999371946,
      "loss": 5.8015,
      "mean_token_accuracy": 0.1737656459212303,
      "num_tokens": 48074941.0,
      "step": 22225
    },
    {
      "entropy": 6.3162391662597654,
      "epoch": 2.378832468296859,
      "grad_norm": 1.84375,
      "learning_rate": 0.00044393771965823376,
      "loss": 5.6422,
      "mean_token_accuracy": 0.17923947423696518,
      "num_tokens": 48085224.0,
      "step": 22230
    },
    {
      "entropy": 6.4757218837738035,
      "epoch": 2.3793675424046232,
      "grad_norm": 1.015625,
      "learning_rate": 0.000443912464444695,
      "loss": 5.8555,
      "mean_token_accuracy": 0.16645013093948363,
      "num_tokens": 48096509.0,
      "step": 22235
    },
    {
      "entropy": 6.409917259216309,
      "epoch": 2.379902616512387,
      "grad_norm": 1.0,
      "learning_rate": 0.0004438872043538327,
      "loss": 5.8029,
      "mean_token_accuracy": 0.16911830455064775,
      "num_tokens": 48106839.0,
      "step": 22240
    },
    {
      "entropy": 6.456915760040284,
      "epoch": 2.380437690620151,
      "grad_norm": 1.078125,
      "learning_rate": 0.000443861939386376,
      "loss": 5.8713,
      "mean_token_accuracy": 0.1672685995697975,
      "num_tokens": 48118209.0,
      "step": 22245
    },
    {
      "entropy": 6.4279087543487545,
      "epoch": 2.3809727647279146,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004438366695430546,
      "loss": 5.8679,
      "mean_token_accuracy": 0.1642686530947685,
      "num_tokens": 48128651.0,
      "step": 22250
    },
    {
      "entropy": 6.480451202392578,
      "epoch": 2.3815078388356787,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044381139482459805,
      "loss": 5.8985,
      "mean_token_accuracy": 0.1647917300462723,
      "num_tokens": 48139612.0,
      "step": 22255
    },
    {
      "entropy": 6.4702154159545895,
      "epoch": 2.382042912943443,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044378611523173615,
      "loss": 5.7926,
      "mean_token_accuracy": 0.1723901018500328,
      "num_tokens": 48150557.0,
      "step": 22260
    },
    {
      "entropy": 6.393025875091553,
      "epoch": 2.3825779870512065,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004437608307651989,
      "loss": 5.7534,
      "mean_token_accuracy": 0.172428460419178,
      "num_tokens": 48160784.0,
      "step": 22265
    },
    {
      "entropy": 6.376884651184082,
      "epoch": 2.3831130611589706,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004437355414257165,
      "loss": 5.8588,
      "mean_token_accuracy": 0.16350395679473878,
      "num_tokens": 48171075.0,
      "step": 22270
    },
    {
      "entropy": 6.485073757171631,
      "epoch": 2.3836481352667347,
      "grad_norm": 1.0,
      "learning_rate": 0.00044371024721401894,
      "loss": 5.8737,
      "mean_token_accuracy": 0.16087627410888672,
      "num_tokens": 48182713.0,
      "step": 22275
    },
    {
      "entropy": 6.446030235290527,
      "epoch": 2.3841832093744983,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004436849481308367,
      "loss": 5.8007,
      "mean_token_accuracy": 0.17434695661067962,
      "num_tokens": 48194198.0,
      "step": 22280
    },
    {
      "entropy": 6.377755212783813,
      "epoch": 2.3847182834822624,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044365964417690035,
      "loss": 5.7606,
      "mean_token_accuracy": 0.17188265025615693,
      "num_tokens": 48205650.0,
      "step": 22285
    },
    {
      "entropy": 6.3504773616790775,
      "epoch": 2.385253357590026,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004436343353529404,
      "loss": 5.7928,
      "mean_token_accuracy": 0.16959455460309983,
      "num_tokens": 48216335.0,
      "step": 22290
    },
    {
      "entropy": 6.457708120346069,
      "epoch": 2.38578843169779,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044360902165968774,
      "loss": 5.8104,
      "mean_token_accuracy": 0.16982098221778869,
      "num_tokens": 48227206.0,
      "step": 22295
    },
    {
      "entropy": 6.451144361495972,
      "epoch": 2.386323505805554,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004435837030978733,
      "loss": 5.8225,
      "mean_token_accuracy": 0.17198664247989653,
      "num_tokens": 48237948.0,
      "step": 22300
    },
    {
      "entropy": 6.379450273513794,
      "epoch": 2.386858579913318,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004435583796682281,
      "loss": 5.8349,
      "mean_token_accuracy": 0.17785116136074067,
      "num_tokens": 48250086.0,
      "step": 22305
    },
    {
      "entropy": 6.437560987472534,
      "epoch": 2.387393654021082,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004435330513714834,
      "loss": 5.8557,
      "mean_token_accuracy": 0.1695139467716217,
      "num_tokens": 48261004.0,
      "step": 22310
    },
    {
      "entropy": 6.42261323928833,
      "epoch": 2.3879287281288457,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004435077182083705,
      "loss": 5.863,
      "mean_token_accuracy": 0.16872481852769852,
      "num_tokens": 48271314.0,
      "step": 22315
    },
    {
      "entropy": 6.3848401546478275,
      "epoch": 2.38846380223661,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044348238017962095,
      "loss": 5.7826,
      "mean_token_accuracy": 0.16905188113451003,
      "num_tokens": 48281899.0,
      "step": 22320
    },
    {
      "entropy": 6.42966513633728,
      "epoch": 2.388998876344374,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004434570372859663,
      "loss": 5.8439,
      "mean_token_accuracy": 0.16578342020511627,
      "num_tokens": 48292925.0,
      "step": 22325
    },
    {
      "entropy": 6.5230467319488525,
      "epoch": 2.3895339504521376,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004434316895281384,
      "loss": 5.8836,
      "mean_token_accuracy": 0.16717156171798705,
      "num_tokens": 48303138.0,
      "step": 22330
    },
    {
      "entropy": 6.384903621673584,
      "epoch": 2.3900690245599017,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004434063369068691,
      "loss": 5.7295,
      "mean_token_accuracy": 0.1852913096547127,
      "num_tokens": 48313586.0,
      "step": 22335
    },
    {
      "entropy": 6.334268093109131,
      "epoch": 2.3906040986676653,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044338097942289045,
      "loss": 5.7939,
      "mean_token_accuracy": 0.17164694517850876,
      "num_tokens": 48324756.0,
      "step": 22340
    },
    {
      "entropy": 6.426454162597656,
      "epoch": 2.3911391727754294,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004433556170769348,
      "loss": 5.877,
      "mean_token_accuracy": 0.1596416115760803,
      "num_tokens": 48336012.0,
      "step": 22345
    },
    {
      "entropy": 6.507317686080933,
      "epoch": 2.391674246883193,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004433302498697343,
      "loss": 5.904,
      "mean_token_accuracy": 0.16186148822307586,
      "num_tokens": 48348923.0,
      "step": 22350
    },
    {
      "entropy": 6.4419866561889645,
      "epoch": 2.392209320990957,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044330487780202144,
      "loss": 5.766,
      "mean_token_accuracy": 0.18468306064605713,
      "num_tokens": 48359708.0,
      "step": 22355
    },
    {
      "entropy": 6.41510763168335,
      "epoch": 2.3927443950987213,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044327950087452895,
      "loss": 5.8476,
      "mean_token_accuracy": 0.1666152998805046,
      "num_tokens": 48370497.0,
      "step": 22360
    },
    {
      "entropy": 6.374325132369995,
      "epoch": 2.393279469206485,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004432541190879895,
      "loss": 5.8407,
      "mean_token_accuracy": 0.15837531089782714,
      "num_tokens": 48381020.0,
      "step": 22365
    },
    {
      "entropy": 6.44053430557251,
      "epoch": 2.393814543314249,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044322873244313603,
      "loss": 5.7958,
      "mean_token_accuracy": 0.17205665111541749,
      "num_tokens": 48391696.0,
      "step": 22370
    },
    {
      "entropy": 6.447664499282837,
      "epoch": 2.394349617422013,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044320334094070154,
      "loss": 5.8442,
      "mean_token_accuracy": 0.1665838822722435,
      "num_tokens": 48402403.0,
      "step": 22375
    },
    {
      "entropy": 6.419774532318115,
      "epoch": 2.394884691529777,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004431779445814192,
      "loss": 5.8529,
      "mean_token_accuracy": 0.17685783207416533,
      "num_tokens": 48413579.0,
      "step": 22380
    },
    {
      "entropy": 6.496084308624267,
      "epoch": 2.395419765637541,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044315254336602236,
      "loss": 5.8993,
      "mean_token_accuracy": 0.15892360508441924,
      "num_tokens": 48423814.0,
      "step": 22385
    },
    {
      "entropy": 6.495368814468383,
      "epoch": 2.3959548397453045,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044312713729524443,
      "loss": 5.9532,
      "mean_token_accuracy": 0.16035623848438263,
      "num_tokens": 48433937.0,
      "step": 22390
    },
    {
      "entropy": 6.473110055923462,
      "epoch": 2.3964899138530686,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044310172636981905,
      "loss": 5.8389,
      "mean_token_accuracy": 0.16678206175565718,
      "num_tokens": 48444397.0,
      "step": 22395
    },
    {
      "entropy": 6.443285894393921,
      "epoch": 2.3970249879608327,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004430763105904799,
      "loss": 5.7319,
      "mean_token_accuracy": 0.17926237732172012,
      "num_tokens": 48454566.0,
      "step": 22400
    },
    {
      "entropy": 6.38153920173645,
      "epoch": 2.3975600620685964,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000443050889957961,
      "loss": 5.7776,
      "mean_token_accuracy": 0.17317631244659423,
      "num_tokens": 48465129.0,
      "step": 22405
    },
    {
      "entropy": 6.4053129196167,
      "epoch": 2.3980951361763605,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044302546447299625,
      "loss": 5.7896,
      "mean_token_accuracy": 0.16966485977172852,
      "num_tokens": 48475793.0,
      "step": 22410
    },
    {
      "entropy": 6.385374402999878,
      "epoch": 2.398630210284124,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004430000341363198,
      "loss": 5.7355,
      "mean_token_accuracy": 0.18040213882923126,
      "num_tokens": 48486543.0,
      "step": 22415
    },
    {
      "entropy": 6.460130119323731,
      "epoch": 2.3991652843918883,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004429745989486659,
      "loss": 5.8869,
      "mean_token_accuracy": 0.17039009630680085,
      "num_tokens": 48498426.0,
      "step": 22420
    },
    {
      "entropy": 6.461370277404785,
      "epoch": 2.3997003584996524,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004429491589107691,
      "loss": 5.7845,
      "mean_token_accuracy": 0.1736776202917099,
      "num_tokens": 48508763.0,
      "step": 22425
    },
    {
      "entropy": 6.4826459884643555,
      "epoch": 2.400235432607416,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004429237140233639,
      "loss": 5.8857,
      "mean_token_accuracy": 0.16338718831539153,
      "num_tokens": 48519112.0,
      "step": 22430
    },
    {
      "entropy": 6.459608459472657,
      "epoch": 2.40077050671518,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004428982642871852,
      "loss": 5.8222,
      "mean_token_accuracy": 0.17353563904762268,
      "num_tokens": 48530341.0,
      "step": 22435
    },
    {
      "entropy": 6.369563341140747,
      "epoch": 2.401305580822944,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004428728097029676,
      "loss": 5.7586,
      "mean_token_accuracy": 0.17270109206438064,
      "num_tokens": 48541962.0,
      "step": 22440
    },
    {
      "entropy": 6.482974338531494,
      "epoch": 2.401840654930708,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044284735027144624,
      "loss": 5.902,
      "mean_token_accuracy": 0.16046528220176698,
      "num_tokens": 48554206.0,
      "step": 22445
    },
    {
      "entropy": 6.392229413986206,
      "epoch": 2.402375729038472,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004428218859933562,
      "loss": 5.7977,
      "mean_token_accuracy": 0.16500746756792067,
      "num_tokens": 48563919.0,
      "step": 22450
    },
    {
      "entropy": 6.4120643615722654,
      "epoch": 2.4029108031462356,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044279641686943277,
      "loss": 5.8742,
      "mean_token_accuracy": 0.1745553806424141,
      "num_tokens": 48575388.0,
      "step": 22455
    },
    {
      "entropy": 6.500313663482666,
      "epoch": 2.4034458772539997,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044277094290041144,
      "loss": 5.8419,
      "mean_token_accuracy": 0.17162414789199829,
      "num_tokens": 48587492.0,
      "step": 22460
    },
    {
      "entropy": 6.343542098999023,
      "epoch": 2.4039809513617634,
      "grad_norm": 1.25,
      "learning_rate": 0.00044274546408702763,
      "loss": 5.6756,
      "mean_token_accuracy": 0.18850726187229155,
      "num_tokens": 48598757.0,
      "step": 22465
    },
    {
      "entropy": 6.412911748886108,
      "epoch": 2.4045160254695275,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044271998043001723,
      "loss": 5.8199,
      "mean_token_accuracy": 0.17354257255792618,
      "num_tokens": 48609912.0,
      "step": 22470
    },
    {
      "entropy": 6.512039995193481,
      "epoch": 2.4050510995772916,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004426944919301158,
      "loss": 5.9454,
      "mean_token_accuracy": 0.165751850605011,
      "num_tokens": 48621614.0,
      "step": 22475
    },
    {
      "entropy": 6.4178924560546875,
      "epoch": 2.4055861736850552,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044266899858805964,
      "loss": 5.8155,
      "mean_token_accuracy": 0.17578001469373702,
      "num_tokens": 48632539.0,
      "step": 22480
    },
    {
      "entropy": 6.52922830581665,
      "epoch": 2.4061212477928193,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044264350040458463,
      "loss": 5.9263,
      "mean_token_accuracy": 0.16308488100767135,
      "num_tokens": 48644157.0,
      "step": 22485
    },
    {
      "entropy": 6.501245212554932,
      "epoch": 2.4066563219005834,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044261799738042713,
      "loss": 5.9374,
      "mean_token_accuracy": 0.16740385591983795,
      "num_tokens": 48654928.0,
      "step": 22490
    },
    {
      "entropy": 6.495906352996826,
      "epoch": 2.407191396008347,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004425924895163235,
      "loss": 5.85,
      "mean_token_accuracy": 0.17298687547445296,
      "num_tokens": 48665183.0,
      "step": 22495
    },
    {
      "entropy": 6.411429214477539,
      "epoch": 2.407726470116111,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044256697681301034,
      "loss": 5.7768,
      "mean_token_accuracy": 0.17605870366096496,
      "num_tokens": 48675233.0,
      "step": 22500
    },
    {
      "entropy": 6.405973052978515,
      "epoch": 2.408261544223875,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004425414592712242,
      "loss": 5.844,
      "mean_token_accuracy": 0.16865808814764022,
      "num_tokens": 48685881.0,
      "step": 22505
    },
    {
      "entropy": 6.455939722061157,
      "epoch": 2.408796618331639,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044251593689170196,
      "loss": 5.8798,
      "mean_token_accuracy": 0.163880854845047,
      "num_tokens": 48696402.0,
      "step": 22510
    },
    {
      "entropy": 6.433288049697876,
      "epoch": 2.409331692439403,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044249040967518063,
      "loss": 5.8924,
      "mean_token_accuracy": 0.1701001286506653,
      "num_tokens": 48708041.0,
      "step": 22515
    },
    {
      "entropy": 6.477346515655517,
      "epoch": 2.4098667665471667,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044246487762239717,
      "loss": 5.806,
      "mean_token_accuracy": 0.16888328343629838,
      "num_tokens": 48718836.0,
      "step": 22520
    },
    {
      "entropy": 6.446034097671509,
      "epoch": 2.410401840654931,
      "grad_norm": 1.03125,
      "learning_rate": 0.000442439340734089,
      "loss": 5.842,
      "mean_token_accuracy": 0.17016484439373017,
      "num_tokens": 48729115.0,
      "step": 22525
    },
    {
      "entropy": 6.3992283821105955,
      "epoch": 2.4109369147626944,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044241379901099334,
      "loss": 5.8179,
      "mean_token_accuracy": 0.17483308762311936,
      "num_tokens": 48740037.0,
      "step": 22530
    },
    {
      "entropy": 6.425147008895874,
      "epoch": 2.4114719888704585,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044238825245384775,
      "loss": 5.8512,
      "mean_token_accuracy": 0.16596005707979203,
      "num_tokens": 48750894.0,
      "step": 22535
    },
    {
      "entropy": 6.44990086555481,
      "epoch": 2.4120070629782226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004423627010633899,
      "loss": 5.7672,
      "mean_token_accuracy": 0.17234368324279786,
      "num_tokens": 48761566.0,
      "step": 22540
    },
    {
      "entropy": 6.306667518615723,
      "epoch": 2.4125421370859863,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044233714484035765,
      "loss": 5.7383,
      "mean_token_accuracy": 0.17330043017864227,
      "num_tokens": 48772971.0,
      "step": 22545
    },
    {
      "entropy": 6.368401527404785,
      "epoch": 2.4130772111937504,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044231158378548873,
      "loss": 5.8368,
      "mean_token_accuracy": 0.17009200751781464,
      "num_tokens": 48784548.0,
      "step": 22550
    },
    {
      "entropy": 6.2891168117523195,
      "epoch": 2.4136122853015145,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004422860178995215,
      "loss": 5.6862,
      "mean_token_accuracy": 0.1785653293132782,
      "num_tokens": 48795228.0,
      "step": 22555
    },
    {
      "entropy": 6.434470176696777,
      "epoch": 2.414147359409278,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044226044718319383,
      "loss": 5.7845,
      "mean_token_accuracy": 0.1706603690981865,
      "num_tokens": 48806497.0,
      "step": 22560
    },
    {
      "entropy": 6.450307941436767,
      "epoch": 2.4146824335170423,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004422348716372444,
      "loss": 5.8128,
      "mean_token_accuracy": 0.17205536961555482,
      "num_tokens": 48816402.0,
      "step": 22565
    },
    {
      "entropy": 6.454527854919434,
      "epoch": 2.415217507624806,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004422092912624115,
      "loss": 5.7663,
      "mean_token_accuracy": 0.17233897000551224,
      "num_tokens": 48826151.0,
      "step": 22570
    },
    {
      "entropy": 6.415458297729492,
      "epoch": 2.41575258173257,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044218370605943376,
      "loss": 5.8631,
      "mean_token_accuracy": 0.16933820843696595,
      "num_tokens": 48836458.0,
      "step": 22575
    },
    {
      "entropy": 6.428048419952392,
      "epoch": 2.4162876558403337,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044215811602905003,
      "loss": 5.8564,
      "mean_token_accuracy": 0.16629236936569214,
      "num_tokens": 48845863.0,
      "step": 22580
    },
    {
      "entropy": 6.421753978729248,
      "epoch": 2.4168227299480978,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044213252117199914,
      "loss": 5.8488,
      "mean_token_accuracy": 0.16172939091920852,
      "num_tokens": 48856453.0,
      "step": 22585
    },
    {
      "entropy": 6.444147777557373,
      "epoch": 2.417357804055862,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044210692148902034,
      "loss": 5.7884,
      "mean_token_accuracy": 0.1695771634578705,
      "num_tokens": 48867013.0,
      "step": 22590
    },
    {
      "entropy": 6.426294183731079,
      "epoch": 2.4178928781636255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004420813169808525,
      "loss": 5.8278,
      "mean_token_accuracy": 0.1636970415711403,
      "num_tokens": 48876455.0,
      "step": 22595
    },
    {
      "entropy": 6.444883489608765,
      "epoch": 2.4184279522713896,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004420557076482352,
      "loss": 5.8634,
      "mean_token_accuracy": 0.16567928045988084,
      "num_tokens": 48888223.0,
      "step": 22600
    },
    {
      "entropy": 6.449288606643677,
      "epoch": 2.4189630263791537,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044203009349190784,
      "loss": 5.8363,
      "mean_token_accuracy": 0.17265767455101014,
      "num_tokens": 48899287.0,
      "step": 22605
    },
    {
      "entropy": 6.47262020111084,
      "epoch": 2.4194981004869174,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004420044745126099,
      "loss": 5.8679,
      "mean_token_accuracy": 0.16542633175849913,
      "num_tokens": 48910740.0,
      "step": 22610
    },
    {
      "entropy": 6.384039211273193,
      "epoch": 2.4200331745946815,
      "grad_norm": 1.125,
      "learning_rate": 0.00044197885071108127,
      "loss": 5.6857,
      "mean_token_accuracy": 0.17388992309570311,
      "num_tokens": 48921925.0,
      "step": 22615
    },
    {
      "entropy": 6.455592489242553,
      "epoch": 2.420568248702445,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004419532220880619,
      "loss": 5.9392,
      "mean_token_accuracy": 0.1580547034740448,
      "num_tokens": 48933164.0,
      "step": 22620
    },
    {
      "entropy": 6.472532749176025,
      "epoch": 2.4211033228102092,
      "grad_norm": 1.125,
      "learning_rate": 0.00044192758864429157,
      "loss": 5.9006,
      "mean_token_accuracy": 0.16675292402505876,
      "num_tokens": 48943285.0,
      "step": 22625
    },
    {
      "entropy": 6.471465444564819,
      "epoch": 2.421638396917973,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004419019503805106,
      "loss": 5.8536,
      "mean_token_accuracy": 0.1614159509539604,
      "num_tokens": 48952852.0,
      "step": 22630
    },
    {
      "entropy": 6.434270524978638,
      "epoch": 2.422173471025737,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044187630729745924,
      "loss": 5.9224,
      "mean_token_accuracy": 0.1595466859638691,
      "num_tokens": 48964339.0,
      "step": 22635
    },
    {
      "entropy": 6.480911302566528,
      "epoch": 2.422708545133501,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044185065939587805,
      "loss": 5.7992,
      "mean_token_accuracy": 0.1729203298687935,
      "num_tokens": 48974754.0,
      "step": 22640
    },
    {
      "entropy": 6.465394163131714,
      "epoch": 2.4232436192412647,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044182500667650746,
      "loss": 5.8554,
      "mean_token_accuracy": 0.17262777984142302,
      "num_tokens": 48985596.0,
      "step": 22645
    },
    {
      "entropy": 6.491850566864014,
      "epoch": 2.423778693349029,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004417993491400882,
      "loss": 5.884,
      "mean_token_accuracy": 0.1670263111591339,
      "num_tokens": 48997400.0,
      "step": 22650
    },
    {
      "entropy": 6.458706569671631,
      "epoch": 2.424313767456793,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044177368678736126,
      "loss": 5.8148,
      "mean_token_accuracy": 0.17140834480524064,
      "num_tokens": 49007525.0,
      "step": 22655
    },
    {
      "entropy": 6.3917176723480225,
      "epoch": 2.4248488415645566,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044174801961906744,
      "loss": 5.8398,
      "mean_token_accuracy": 0.1717819541692734,
      "num_tokens": 49018205.0,
      "step": 22660
    },
    {
      "entropy": 6.498028087615967,
      "epoch": 2.4253839156723207,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044172234763594805,
      "loss": 5.8712,
      "mean_token_accuracy": 0.16352533996105195,
      "num_tokens": 49028760.0,
      "step": 22665
    },
    {
      "entropy": 6.463360118865967,
      "epoch": 2.4259189897800844,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004416966708387443,
      "loss": 5.8327,
      "mean_token_accuracy": 0.16713132560253144,
      "num_tokens": 49040734.0,
      "step": 22670
    },
    {
      "entropy": 6.401403570175171,
      "epoch": 2.4264540638878485,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044167098922819764,
      "loss": 5.7958,
      "mean_token_accuracy": 0.16308979094028472,
      "num_tokens": 49051474.0,
      "step": 22675
    },
    {
      "entropy": 6.362638187408447,
      "epoch": 2.4269891379956126,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044164530280504954,
      "loss": 5.8013,
      "mean_token_accuracy": 0.17462510764598846,
      "num_tokens": 49061697.0,
      "step": 22680
    },
    {
      "entropy": 6.4131152629852295,
      "epoch": 2.427524212103376,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044161961157004176,
      "loss": 5.8652,
      "mean_token_accuracy": 0.1700657859444618,
      "num_tokens": 49072124.0,
      "step": 22685
    },
    {
      "entropy": 6.4388936996459964,
      "epoch": 2.4280592862111403,
      "grad_norm": 1.125,
      "learning_rate": 0.000441593915523916,
      "loss": 5.8224,
      "mean_token_accuracy": 0.1719927027821541,
      "num_tokens": 49082653.0,
      "step": 22690
    },
    {
      "entropy": 6.491871166229248,
      "epoch": 2.428594360318904,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044156821466741447,
      "loss": 5.9247,
      "mean_token_accuracy": 0.1582934617996216,
      "num_tokens": 49092481.0,
      "step": 22695
    },
    {
      "entropy": 6.458665180206299,
      "epoch": 2.429129434426668,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004415425090012792,
      "loss": 5.9361,
      "mean_token_accuracy": 0.15750405043363572,
      "num_tokens": 49103189.0,
      "step": 22700
    },
    {
      "entropy": 6.498424863815307,
      "epoch": 2.429664508534432,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044151679852625226,
      "loss": 5.8851,
      "mean_token_accuracy": 0.1610925555229187,
      "num_tokens": 49114071.0,
      "step": 22705
    },
    {
      "entropy": 6.4914905548095705,
      "epoch": 2.430199582642196,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004414910832430762,
      "loss": 5.778,
      "mean_token_accuracy": 0.17068396657705306,
      "num_tokens": 49125521.0,
      "step": 22710
    },
    {
      "entropy": 6.505469083786011,
      "epoch": 2.43073465674996,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004414653631524935,
      "loss": 5.8073,
      "mean_token_accuracy": 0.1646130457520485,
      "num_tokens": 49136162.0,
      "step": 22715
    },
    {
      "entropy": 6.492537164688111,
      "epoch": 2.431269730857724,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004414396382552468,
      "loss": 5.8323,
      "mean_token_accuracy": 0.16790963858366012,
      "num_tokens": 49146127.0,
      "step": 22720
    },
    {
      "entropy": 6.394844579696655,
      "epoch": 2.4318048049654877,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044141390855207904,
      "loss": 5.7876,
      "mean_token_accuracy": 0.17813624292612076,
      "num_tokens": 49156619.0,
      "step": 22725
    },
    {
      "entropy": 6.440625429153442,
      "epoch": 2.4323398790732518,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044138817404373295,
      "loss": 5.9239,
      "mean_token_accuracy": 0.1630412831902504,
      "num_tokens": 49168578.0,
      "step": 22730
    },
    {
      "entropy": 6.5179856300354,
      "epoch": 2.4328749531810154,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004413624347309517,
      "loss": 5.8813,
      "mean_token_accuracy": 0.17026398777961732,
      "num_tokens": 49178845.0,
      "step": 22735
    },
    {
      "entropy": 6.478231048583984,
      "epoch": 2.4334100272887795,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044133669061447855,
      "loss": 5.887,
      "mean_token_accuracy": 0.16830423176288606,
      "num_tokens": 49189067.0,
      "step": 22740
    },
    {
      "entropy": 6.46953592300415,
      "epoch": 2.433945101396543,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044131094169505694,
      "loss": 5.8798,
      "mean_token_accuracy": 0.16312731057405472,
      "num_tokens": 49200368.0,
      "step": 22745
    },
    {
      "entropy": 6.444351625442505,
      "epoch": 2.4344801755043073,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044128518797343014,
      "loss": 5.8321,
      "mean_token_accuracy": 0.16396388709545134,
      "num_tokens": 49211630.0,
      "step": 22750
    },
    {
      "entropy": 6.497300863265991,
      "epoch": 2.4350152496120714,
      "grad_norm": 1.0,
      "learning_rate": 0.000441259429450342,
      "loss": 5.9043,
      "mean_token_accuracy": 0.1656777083873749,
      "num_tokens": 49223092.0,
      "step": 22755
    },
    {
      "entropy": 6.459932088851929,
      "epoch": 2.435550323719835,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044123366612653617,
      "loss": 5.8452,
      "mean_token_accuracy": 0.17354036420583724,
      "num_tokens": 49233963.0,
      "step": 22760
    },
    {
      "entropy": 6.52847466468811,
      "epoch": 2.436085397827599,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004412078980027565,
      "loss": 5.8819,
      "mean_token_accuracy": 0.16801125556230545,
      "num_tokens": 49243834.0,
      "step": 22765
    },
    {
      "entropy": 6.3572938442230225,
      "epoch": 2.4366204719353632,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004411821250797472,
      "loss": 5.7072,
      "mean_token_accuracy": 0.17585553526878356,
      "num_tokens": 49254274.0,
      "step": 22770
    },
    {
      "entropy": 6.411256933212281,
      "epoch": 2.437155546043127,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004411563473582524,
      "loss": 5.769,
      "mean_token_accuracy": 0.1733081191778183,
      "num_tokens": 49264734.0,
      "step": 22775
    },
    {
      "entropy": 6.413739347457886,
      "epoch": 2.437690620150891,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004411305648390163,
      "loss": 5.7993,
      "mean_token_accuracy": 0.171263387799263,
      "num_tokens": 49274708.0,
      "step": 22780
    },
    {
      "entropy": 6.408799171447754,
      "epoch": 2.4382256942586547,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004411047775227836,
      "loss": 5.8418,
      "mean_token_accuracy": 0.17073815315961838,
      "num_tokens": 49286318.0,
      "step": 22785
    },
    {
      "entropy": 6.532680320739746,
      "epoch": 2.4387607683664188,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044107898541029885,
      "loss": 5.8763,
      "mean_token_accuracy": 0.16337986439466476,
      "num_tokens": 49296948.0,
      "step": 22790
    },
    {
      "entropy": 6.464037036895752,
      "epoch": 2.439295842474183,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004410531885023066,
      "loss": 5.7538,
      "mean_token_accuracy": 0.17381113916635513,
      "num_tokens": 49307060.0,
      "step": 22795
    },
    {
      "entropy": 6.43543586730957,
      "epoch": 2.4398309165819465,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044102738679955195,
      "loss": 5.8343,
      "mean_token_accuracy": 0.16511150598526,
      "num_tokens": 49318449.0,
      "step": 22800
    },
    {
      "entropy": 6.432837867736817,
      "epoch": 2.4403659906897106,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004410015803027798,
      "loss": 5.8644,
      "mean_token_accuracy": 0.1667453721165657,
      "num_tokens": 49329567.0,
      "step": 22805
    },
    {
      "entropy": 6.5052801132202145,
      "epoch": 2.4409010647974743,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044097576901273535,
      "loss": 5.8682,
      "mean_token_accuracy": 0.16873008757829666,
      "num_tokens": 49340854.0,
      "step": 22810
    },
    {
      "entropy": 6.466220474243164,
      "epoch": 2.4414361389052384,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004409499529301639,
      "loss": 5.8836,
      "mean_token_accuracy": 0.16360816806554795,
      "num_tokens": 49352749.0,
      "step": 22815
    },
    {
      "entropy": 6.4640195846557615,
      "epoch": 2.4419712130130025,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004409241320558109,
      "loss": 5.8904,
      "mean_token_accuracy": 0.1647409737110138,
      "num_tokens": 49364703.0,
      "step": 22820
    },
    {
      "entropy": 6.441115617752075,
      "epoch": 2.442506287120766,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044089830639042183,
      "loss": 5.7483,
      "mean_token_accuracy": 0.1689605101943016,
      "num_tokens": 49375873.0,
      "step": 22825
    },
    {
      "entropy": 6.440145349502563,
      "epoch": 2.44304136122853,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044087247593474255,
      "loss": 5.9222,
      "mean_token_accuracy": 0.16508698910474778,
      "num_tokens": 49386578.0,
      "step": 22830
    },
    {
      "entropy": 6.437005233764649,
      "epoch": 2.4435764353362943,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004408466406895188,
      "loss": 5.8361,
      "mean_token_accuracy": 0.16970042884349823,
      "num_tokens": 49398394.0,
      "step": 22835
    },
    {
      "entropy": 6.401193141937256,
      "epoch": 2.444111509444058,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004408208006554966,
      "loss": 5.7805,
      "mean_token_accuracy": 0.18244958370923997,
      "num_tokens": 49408758.0,
      "step": 22840
    },
    {
      "entropy": 6.432500553131104,
      "epoch": 2.444646583551822,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004407949558334221,
      "loss": 5.8015,
      "mean_token_accuracy": 0.16965993046760558,
      "num_tokens": 49420447.0,
      "step": 22845
    },
    {
      "entropy": 6.449878358840943,
      "epoch": 2.4451816576595857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044076910622404153,
      "loss": 5.8748,
      "mean_token_accuracy": 0.17048958837985992,
      "num_tokens": 49430246.0,
      "step": 22850
    },
    {
      "entropy": 6.4524036884307865,
      "epoch": 2.44571673176735,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044074325182810135,
      "loss": 5.8228,
      "mean_token_accuracy": 0.17306440472602844,
      "num_tokens": 49441272.0,
      "step": 22855
    },
    {
      "entropy": 6.436959171295166,
      "epoch": 2.4462518058751135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044071739264634803,
      "loss": 5.8663,
      "mean_token_accuracy": 0.15897945463657379,
      "num_tokens": 49452778.0,
      "step": 22860
    },
    {
      "entropy": 6.45834755897522,
      "epoch": 2.4467868799828776,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004406915286795283,
      "loss": 5.8387,
      "mean_token_accuracy": 0.17067958861589433,
      "num_tokens": 49463594.0,
      "step": 22865
    },
    {
      "entropy": 6.477951240539551,
      "epoch": 2.4473219540906417,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004406656599283889,
      "loss": 5.8668,
      "mean_token_accuracy": 0.16553113907575606,
      "num_tokens": 49473649.0,
      "step": 22870
    },
    {
      "entropy": 6.40130844116211,
      "epoch": 2.4478570281984053,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004406397863936769,
      "loss": 5.7935,
      "mean_token_accuracy": 0.17198419570922852,
      "num_tokens": 49485112.0,
      "step": 22875
    },
    {
      "entropy": 6.453299188613892,
      "epoch": 2.4483921023061694,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004406139080761394,
      "loss": 5.9002,
      "mean_token_accuracy": 0.1676807150244713,
      "num_tokens": 49495455.0,
      "step": 22880
    },
    {
      "entropy": 6.433226537704468,
      "epoch": 2.4489271764139335,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004405880249765235,
      "loss": 5.8434,
      "mean_token_accuracy": 0.17308484762907028,
      "num_tokens": 49506495.0,
      "step": 22885
    },
    {
      "entropy": 6.431468391418457,
      "epoch": 2.449462250521697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044056213709557667,
      "loss": 5.8795,
      "mean_token_accuracy": 0.1636393815279007,
      "num_tokens": 49516384.0,
      "step": 22890
    },
    {
      "entropy": 6.420477104187012,
      "epoch": 2.4499973246294613,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044053624443404646,
      "loss": 5.8194,
      "mean_token_accuracy": 0.17804911583662034,
      "num_tokens": 49527972.0,
      "step": 22895
    },
    {
      "entropy": 6.431223344802857,
      "epoch": 2.450532398737225,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004405103469926803,
      "loss": 5.7975,
      "mean_token_accuracy": 0.16770053058862686,
      "num_tokens": 49540643.0,
      "step": 22900
    },
    {
      "entropy": 6.438940000534058,
      "epoch": 2.451067472844989,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044048444477222626,
      "loss": 5.7627,
      "mean_token_accuracy": 0.17842579185962676,
      "num_tokens": 49550968.0,
      "step": 22905
    },
    {
      "entropy": 6.435456895828247,
      "epoch": 2.451602546952753,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004404585377734321,
      "loss": 5.7824,
      "mean_token_accuracy": 0.17122637182474137,
      "num_tokens": 49562381.0,
      "step": 22910
    },
    {
      "entropy": 6.43230791091919,
      "epoch": 2.452137621060517,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004404326259970459,
      "loss": 5.8718,
      "mean_token_accuracy": 0.1657914251089096,
      "num_tokens": 49573414.0,
      "step": 22915
    },
    {
      "entropy": 6.441785049438477,
      "epoch": 2.452672695168281,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004404067094438159,
      "loss": 5.8766,
      "mean_token_accuracy": 0.17254668623209,
      "num_tokens": 49583772.0,
      "step": 22920
    },
    {
      "entropy": 6.480189466476441,
      "epoch": 2.4532077692760446,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004403807881144903,
      "loss": 5.9046,
      "mean_token_accuracy": 0.16144364476203918,
      "num_tokens": 49594751.0,
      "step": 22925
    },
    {
      "entropy": 6.427809572219848,
      "epoch": 2.4537428433838087,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044035486200981786,
      "loss": 5.797,
      "mean_token_accuracy": 0.1704736053943634,
      "num_tokens": 49605048.0,
      "step": 22930
    },
    {
      "entropy": 6.496890830993652,
      "epoch": 2.4542779174915728,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044032893113054683,
      "loss": 5.9157,
      "mean_token_accuracy": 0.1632806733250618,
      "num_tokens": 49616582.0,
      "step": 22935
    },
    {
      "entropy": 6.47621898651123,
      "epoch": 2.4548129915993364,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004403029954774263,
      "loss": 5.8292,
      "mean_token_accuracy": 0.16872851252555848,
      "num_tokens": 49626974.0,
      "step": 22940
    },
    {
      "entropy": 6.430427837371826,
      "epoch": 2.4553480657071005,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004402770550512049,
      "loss": 5.8245,
      "mean_token_accuracy": 0.16588395684957505,
      "num_tokens": 49637580.0,
      "step": 22945
    },
    {
      "entropy": 6.512602758407593,
      "epoch": 2.455883139814864,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004402511098526318,
      "loss": 5.8826,
      "mean_token_accuracy": 0.16988828778266907,
      "num_tokens": 49648543.0,
      "step": 22950
    },
    {
      "entropy": 6.502292966842651,
      "epoch": 2.4564182139226283,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004402251598824561,
      "loss": 5.8823,
      "mean_token_accuracy": 0.1697079285979271,
      "num_tokens": 49659057.0,
      "step": 22955
    },
    {
      "entropy": 6.4229694366455075,
      "epoch": 2.4569532880303924,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004401992051414272,
      "loss": 5.8181,
      "mean_token_accuracy": 0.16836649924516678,
      "num_tokens": 49671226.0,
      "step": 22960
    },
    {
      "entropy": 6.413279008865357,
      "epoch": 2.457488362138156,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044017324563029435,
      "loss": 5.838,
      "mean_token_accuracy": 0.17039430290460586,
      "num_tokens": 49682063.0,
      "step": 22965
    },
    {
      "entropy": 6.360893630981446,
      "epoch": 2.45802343624592,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004401472813498074,
      "loss": 5.7917,
      "mean_token_accuracy": 0.178825481235981,
      "num_tokens": 49693768.0,
      "step": 22970
    },
    {
      "entropy": 6.475201892852783,
      "epoch": 2.458558510353684,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044012131230071575,
      "loss": 5.851,
      "mean_token_accuracy": 0.16313184946775436,
      "num_tokens": 49703807.0,
      "step": 22975
    },
    {
      "entropy": 6.46417555809021,
      "epoch": 2.459093584461448,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004400953384837695,
      "loss": 5.831,
      "mean_token_accuracy": 0.16573918461799622,
      "num_tokens": 49714945.0,
      "step": 22980
    },
    {
      "entropy": 6.455793046951294,
      "epoch": 2.459628658569212,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004400693598997185,
      "loss": 5.7637,
      "mean_token_accuracy": 0.1828693062067032,
      "num_tokens": 49727267.0,
      "step": 22985
    },
    {
      "entropy": 6.422571897506714,
      "epoch": 2.4601637326769756,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044004337654931293,
      "loss": 5.795,
      "mean_token_accuracy": 0.16650591045618057,
      "num_tokens": 49738370.0,
      "step": 22990
    },
    {
      "entropy": 6.454456186294555,
      "epoch": 2.4606988067847397,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044001738843330306,
      "loss": 5.8533,
      "mean_token_accuracy": 0.16749309301376342,
      "num_tokens": 49750113.0,
      "step": 22995
    },
    {
      "entropy": 6.453921127319336,
      "epoch": 2.461233880892504,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043999139555243936,
      "loss": 5.9326,
      "mean_token_accuracy": 0.1573762223124504,
      "num_tokens": 49761163.0,
      "step": 23000
    },
    {
      "entropy": 6.4532609462738035,
      "epoch": 2.4617689550002675,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004399653979074722,
      "loss": 5.855,
      "mean_token_accuracy": 0.16376174837350846,
      "num_tokens": 49771185.0,
      "step": 23005
    },
    {
      "entropy": 6.494804096221924,
      "epoch": 2.4623040291080316,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043993939549915245,
      "loss": 5.8533,
      "mean_token_accuracy": 0.16994985193014145,
      "num_tokens": 49781202.0,
      "step": 23010
    },
    {
      "entropy": 6.4602264881134035,
      "epoch": 2.4628391032157952,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043991338832823074,
      "loss": 5.8308,
      "mean_token_accuracy": 0.16554597169160842,
      "num_tokens": 49790971.0,
      "step": 23015
    },
    {
      "entropy": 6.407478952407837,
      "epoch": 2.4633741773235593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004398873763954582,
      "loss": 5.7756,
      "mean_token_accuracy": 0.1718434453010559,
      "num_tokens": 49800942.0,
      "step": 23020
    },
    {
      "entropy": 6.362863159179687,
      "epoch": 2.463909251431323,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043986135970158573,
      "loss": 5.7869,
      "mean_token_accuracy": 0.17284876257181167,
      "num_tokens": 49812043.0,
      "step": 23025
    },
    {
      "entropy": 6.347718954086304,
      "epoch": 2.464444325539087,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043983533824736476,
      "loss": 5.7745,
      "mean_token_accuracy": 0.18140257000923157,
      "num_tokens": 49823045.0,
      "step": 23030
    },
    {
      "entropy": 6.494714641571045,
      "epoch": 2.464979399646851,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004398093120335465,
      "loss": 5.9077,
      "mean_token_accuracy": 0.164957432448864,
      "num_tokens": 49833014.0,
      "step": 23035
    },
    {
      "entropy": 6.477995252609253,
      "epoch": 2.465514473754615,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043978328106088254,
      "loss": 5.807,
      "mean_token_accuracy": 0.1656717151403427,
      "num_tokens": 49843171.0,
      "step": 23040
    },
    {
      "entropy": 6.428487873077392,
      "epoch": 2.466049547862379,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004397572453301246,
      "loss": 5.7197,
      "mean_token_accuracy": 0.17601820230484008,
      "num_tokens": 49853772.0,
      "step": 23045
    },
    {
      "entropy": 6.461433410644531,
      "epoch": 2.466584621970143,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043973120484202416,
      "loss": 5.8559,
      "mean_token_accuracy": 0.1681311696767807,
      "num_tokens": 49865379.0,
      "step": 23050
    },
    {
      "entropy": 6.475361680984497,
      "epoch": 2.4671196960779067,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043970515959733343,
      "loss": 5.9003,
      "mean_token_accuracy": 0.16229844093322754,
      "num_tokens": 49876453.0,
      "step": 23055
    },
    {
      "entropy": 6.384465599060059,
      "epoch": 2.467654770185671,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043967910959680435,
      "loss": 5.7673,
      "mean_token_accuracy": 0.18023217767477034,
      "num_tokens": 49886675.0,
      "step": 23060
    },
    {
      "entropy": 6.381332302093506,
      "epoch": 2.4681898442934345,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004396530548411891,
      "loss": 5.7104,
      "mean_token_accuracy": 0.1726376697421074,
      "num_tokens": 49896946.0,
      "step": 23065
    },
    {
      "entropy": 6.448998928070068,
      "epoch": 2.4687249184011986,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004396269953312401,
      "loss": 5.9138,
      "mean_token_accuracy": 0.16117580533027648,
      "num_tokens": 49906989.0,
      "step": 23070
    },
    {
      "entropy": 6.32019829750061,
      "epoch": 2.4692599925089627,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043960093106770964,
      "loss": 5.69,
      "mean_token_accuracy": 0.17354349195957183,
      "num_tokens": 49917861.0,
      "step": 23075
    },
    {
      "entropy": 6.396671390533447,
      "epoch": 2.4697950666167263,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00043957486205135047,
      "loss": 5.8472,
      "mean_token_accuracy": 0.1697103664278984,
      "num_tokens": 49930529.0,
      "step": 23080
    },
    {
      "entropy": 6.439856433868409,
      "epoch": 2.4703301407244904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004395487882829153,
      "loss": 5.7881,
      "mean_token_accuracy": 0.17276947647333146,
      "num_tokens": 49943067.0,
      "step": 23085
    },
    {
      "entropy": 6.456288385391235,
      "epoch": 2.470865214832254,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043952270976315707,
      "loss": 5.8442,
      "mean_token_accuracy": 0.16697200536727905,
      "num_tokens": 49953648.0,
      "step": 23090
    },
    {
      "entropy": 6.506127882003784,
      "epoch": 2.471400288940018,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043949662649282853,
      "loss": 5.9123,
      "mean_token_accuracy": 0.15927519798278808,
      "num_tokens": 49963571.0,
      "step": 23095
    },
    {
      "entropy": 6.427070808410645,
      "epoch": 2.4719353630477823,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043947053847268313,
      "loss": 5.8272,
      "mean_token_accuracy": 0.17722609788179397,
      "num_tokens": 49974627.0,
      "step": 23100
    },
    {
      "entropy": 6.484214544296265,
      "epoch": 2.472470437155546,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000439444445703474,
      "loss": 5.8768,
      "mean_token_accuracy": 0.17038066387176515,
      "num_tokens": 49986659.0,
      "step": 23105
    },
    {
      "entropy": 6.449531507492066,
      "epoch": 2.47300551126331,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043941834818595464,
      "loss": 5.8379,
      "mean_token_accuracy": 0.17107661813497543,
      "num_tokens": 49997099.0,
      "step": 23110
    },
    {
      "entropy": 6.469316005706787,
      "epoch": 2.473540585371074,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043939224592087864,
      "loss": 5.813,
      "mean_token_accuracy": 0.17255081087350846,
      "num_tokens": 50008148.0,
      "step": 23115
    },
    {
      "entropy": 6.506937170028687,
      "epoch": 2.474075659478838,
      "grad_norm": 1.0,
      "learning_rate": 0.00043936613890899955,
      "loss": 5.9427,
      "mean_token_accuracy": 0.1588141456246376,
      "num_tokens": 50019335.0,
      "step": 23120
    },
    {
      "entropy": 6.417962503433228,
      "epoch": 2.474610733586602,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004393400271510713,
      "loss": 5.7814,
      "mean_token_accuracy": 0.16806404739618303,
      "num_tokens": 50029513.0,
      "step": 23125
    },
    {
      "entropy": 6.471676635742187,
      "epoch": 2.4751458076943655,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043931391064784786,
      "loss": 5.9111,
      "mean_token_accuracy": 0.16460796147584916,
      "num_tokens": 50039353.0,
      "step": 23130
    },
    {
      "entropy": 6.4554585933685305,
      "epoch": 2.4756808818021296,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043928778940008334,
      "loss": 5.8866,
      "mean_token_accuracy": 0.1699664607644081,
      "num_tokens": 50050260.0,
      "step": 23135
    },
    {
      "entropy": 6.395396900177002,
      "epoch": 2.4762159559098933,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043926166340853203,
      "loss": 5.7189,
      "mean_token_accuracy": 0.16833817288279534,
      "num_tokens": 50061384.0,
      "step": 23140
    },
    {
      "entropy": 6.3981616497039795,
      "epoch": 2.4767510300176574,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004392355326739483,
      "loss": 5.7943,
      "mean_token_accuracy": 0.17432668954133987,
      "num_tokens": 50071984.0,
      "step": 23145
    },
    {
      "entropy": 6.401925563812256,
      "epoch": 2.4772861041254215,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043920939719708656,
      "loss": 5.775,
      "mean_token_accuracy": 0.172234009206295,
      "num_tokens": 50083225.0,
      "step": 23150
    },
    {
      "entropy": 6.424034690856933,
      "epoch": 2.477821178233185,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004391832569787016,
      "loss": 5.799,
      "mean_token_accuracy": 0.1773081064224243,
      "num_tokens": 50093240.0,
      "step": 23155
    },
    {
      "entropy": 6.445444011688233,
      "epoch": 2.4783562523409493,
      "grad_norm": 1.0,
      "learning_rate": 0.0004391571120195481,
      "loss": 5.8635,
      "mean_token_accuracy": 0.16529409289360047,
      "num_tokens": 50103870.0,
      "step": 23160
    },
    {
      "entropy": 6.435125207901001,
      "epoch": 2.4788913264487134,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004391309623203811,
      "loss": 5.8949,
      "mean_token_accuracy": 0.16386534720659257,
      "num_tokens": 50115231.0,
      "step": 23165
    },
    {
      "entropy": 6.428262996673584,
      "epoch": 2.479426400556477,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004391048078819556,
      "loss": 5.7381,
      "mean_token_accuracy": 0.17548946887254716,
      "num_tokens": 50125592.0,
      "step": 23170
    },
    {
      "entropy": 6.38242621421814,
      "epoch": 2.479961474664241,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004390786487050269,
      "loss": 5.7636,
      "mean_token_accuracy": 0.17282581478357315,
      "num_tokens": 50136713.0,
      "step": 23175
    },
    {
      "entropy": 6.48370327949524,
      "epoch": 2.4804965487720048,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004390524847903502,
      "loss": 5.8339,
      "mean_token_accuracy": 0.17604341804981233,
      "num_tokens": 50148645.0,
      "step": 23180
    },
    {
      "entropy": 6.483289003372192,
      "epoch": 2.481031622879769,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004390263161386811,
      "loss": 5.8043,
      "mean_token_accuracy": 0.17394268959760667,
      "num_tokens": 50159973.0,
      "step": 23185
    },
    {
      "entropy": 6.46394190788269,
      "epoch": 2.481566696987533,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043900014275077517,
      "loss": 5.8353,
      "mean_token_accuracy": 0.16769794672727584,
      "num_tokens": 50170591.0,
      "step": 23190
    },
    {
      "entropy": 6.416004419326782,
      "epoch": 2.4821017710952966,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043897396462738817,
      "loss": 5.8522,
      "mean_token_accuracy": 0.1729460746049881,
      "num_tokens": 50182414.0,
      "step": 23195
    },
    {
      "entropy": 6.504813623428345,
      "epoch": 2.4826368452030607,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004389477817692759,
      "loss": 5.8969,
      "mean_token_accuracy": 0.16446331143379211,
      "num_tokens": 50192567.0,
      "step": 23200
    },
    {
      "entropy": 6.5144908905029295,
      "epoch": 2.4831719193108244,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004389215941771946,
      "loss": 5.8678,
      "mean_token_accuracy": 0.16388643234968187,
      "num_tokens": 50203142.0,
      "step": 23205
    },
    {
      "entropy": 6.452277565002442,
      "epoch": 2.4837069934185885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043889540185190024,
      "loss": 5.8467,
      "mean_token_accuracy": 0.16256500631570817,
      "num_tokens": 50214854.0,
      "step": 23210
    },
    {
      "entropy": 6.46497745513916,
      "epoch": 2.4842420675263526,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004388692047941492,
      "loss": 5.7979,
      "mean_token_accuracy": 0.1675039127469063,
      "num_tokens": 50225999.0,
      "step": 23215
    },
    {
      "entropy": 6.5630241394042965,
      "epoch": 2.4847771416341162,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004388430030046978,
      "loss": 5.9087,
      "mean_token_accuracy": 0.160951891541481,
      "num_tokens": 50236657.0,
      "step": 23220
    },
    {
      "entropy": 6.494233894348144,
      "epoch": 2.4853122157418803,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004388167964843029,
      "loss": 5.8421,
      "mean_token_accuracy": 0.17180456668138505,
      "num_tokens": 50246457.0,
      "step": 23225
    },
    {
      "entropy": 6.427776718139649,
      "epoch": 2.485847289849644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004387905852337209,
      "loss": 5.8681,
      "mean_token_accuracy": 0.16295687556266786,
      "num_tokens": 50257020.0,
      "step": 23230
    },
    {
      "entropy": 6.469898414611817,
      "epoch": 2.486382363957408,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043876436925370873,
      "loss": 5.7566,
      "mean_token_accuracy": 0.16937834918498992,
      "num_tokens": 50267173.0,
      "step": 23235
    },
    {
      "entropy": 6.43467903137207,
      "epoch": 2.486917438065172,
      "grad_norm": 1.125,
      "learning_rate": 0.0004387381485450235,
      "loss": 5.8871,
      "mean_token_accuracy": 0.16911047399044038,
      "num_tokens": 50278352.0,
      "step": 23240
    },
    {
      "entropy": 6.489453458786011,
      "epoch": 2.487452512172936,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043871192310842216,
      "loss": 5.9478,
      "mean_token_accuracy": 0.15971819758415223,
      "num_tokens": 50290301.0,
      "step": 23245
    },
    {
      "entropy": 6.546316289901734,
      "epoch": 2.4879875862807,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004386856929446621,
      "loss": 5.9188,
      "mean_token_accuracy": 0.15853563994169234,
      "num_tokens": 50301707.0,
      "step": 23250
    },
    {
      "entropy": 6.484736490249634,
      "epoch": 2.4885226603884636,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004386594580545006,
      "loss": 5.8627,
      "mean_token_accuracy": 0.16689854562282563,
      "num_tokens": 50313334.0,
      "step": 23255
    },
    {
      "entropy": 6.470256996154785,
      "epoch": 2.4890577344962277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004386332184386953,
      "loss": 5.854,
      "mean_token_accuracy": 0.16873007863759995,
      "num_tokens": 50324228.0,
      "step": 23260
    },
    {
      "entropy": 6.361552286148071,
      "epoch": 2.489592808603992,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043860697409800383,
      "loss": 5.6836,
      "mean_token_accuracy": 0.1776217356324196,
      "num_tokens": 50335001.0,
      "step": 23265
    },
    {
      "entropy": 6.422144985198974,
      "epoch": 2.4901278827117554,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043858072503318393,
      "loss": 5.8716,
      "mean_token_accuracy": 0.1701727643609047,
      "num_tokens": 50346415.0,
      "step": 23270
    },
    {
      "entropy": 6.4336690425872805,
      "epoch": 2.4906629568195195,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043855447124499356,
      "loss": 5.8821,
      "mean_token_accuracy": 0.16563731878995896,
      "num_tokens": 50357625.0,
      "step": 23275
    },
    {
      "entropy": 6.4249656200408936,
      "epoch": 2.4911980309272836,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043852821273419087,
      "loss": 5.7715,
      "mean_token_accuracy": 0.17904828786849974,
      "num_tokens": 50367672.0,
      "step": 23280
    },
    {
      "entropy": 6.426271438598633,
      "epoch": 2.4917331050350473,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000438501949501534,
      "loss": 5.7966,
      "mean_token_accuracy": 0.17345048785209655,
      "num_tokens": 50378220.0,
      "step": 23285
    },
    {
      "entropy": 6.453832960128784,
      "epoch": 2.4922681791428114,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043847568154778134,
      "loss": 5.7256,
      "mean_token_accuracy": 0.1753590077161789,
      "num_tokens": 50389517.0,
      "step": 23290
    },
    {
      "entropy": 6.453168106079102,
      "epoch": 2.492803253250575,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004384494088736913,
      "loss": 5.8473,
      "mean_token_accuracy": 0.16991420239210128,
      "num_tokens": 50401155.0,
      "step": 23295
    },
    {
      "entropy": 6.4063325881958,
      "epoch": 2.493338327358339,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004384231314800226,
      "loss": 5.83,
      "mean_token_accuracy": 0.16939754039049149,
      "num_tokens": 50412609.0,
      "step": 23300
    },
    {
      "entropy": 6.4934791088104244,
      "epoch": 2.493873401466103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004383968493675339,
      "loss": 5.821,
      "mean_token_accuracy": 0.17014556527137756,
      "num_tokens": 50422665.0,
      "step": 23305
    },
    {
      "entropy": 6.477301454544067,
      "epoch": 2.494408475573867,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004383705625369841,
      "loss": 5.8617,
      "mean_token_accuracy": 0.16748461723327637,
      "num_tokens": 50434012.0,
      "step": 23310
    },
    {
      "entropy": 6.487029981613159,
      "epoch": 2.494943549681631,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004383442709891323,
      "loss": 5.8428,
      "mean_token_accuracy": 0.1643723025918007,
      "num_tokens": 50444270.0,
      "step": 23315
    },
    {
      "entropy": 6.412434482574463,
      "epoch": 2.4954786237893947,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043831797472473763,
      "loss": 5.7015,
      "mean_token_accuracy": 0.17971059083938598,
      "num_tokens": 50455236.0,
      "step": 23320
    },
    {
      "entropy": 6.424883508682251,
      "epoch": 2.4960136978971588,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004382916737445594,
      "loss": 5.8529,
      "mean_token_accuracy": 0.16655295938253403,
      "num_tokens": 50466431.0,
      "step": 23325
    },
    {
      "entropy": 6.483275842666626,
      "epoch": 2.496548772004923,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000438265368049357,
      "loss": 5.9068,
      "mean_token_accuracy": 0.15946445614099503,
      "num_tokens": 50477211.0,
      "step": 23330
    },
    {
      "entropy": 6.447190809249878,
      "epoch": 2.4970838461126865,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043823905763989003,
      "loss": 5.8104,
      "mean_token_accuracy": 0.1648933172225952,
      "num_tokens": 50488668.0,
      "step": 23335
    },
    {
      "entropy": 6.449949026107788,
      "epoch": 2.4976189202204506,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043821274251691823,
      "loss": 5.7759,
      "mean_token_accuracy": 0.16681517213582991,
      "num_tokens": 50498031.0,
      "step": 23340
    },
    {
      "entropy": 6.377142715454101,
      "epoch": 2.4981539943282143,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043818642268120153,
      "loss": 5.8074,
      "mean_token_accuracy": 0.1686095491051674,
      "num_tokens": 50508567.0,
      "step": 23345
    },
    {
      "entropy": 6.4168843746185305,
      "epoch": 2.4986890684359784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043816009813349964,
      "loss": 5.766,
      "mean_token_accuracy": 0.17968478798866272,
      "num_tokens": 50519057.0,
      "step": 23350
    },
    {
      "entropy": 6.479688692092895,
      "epoch": 2.4992241425437425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000438133768874573,
      "loss": 5.8412,
      "mean_token_accuracy": 0.16937787532806398,
      "num_tokens": 50530572.0,
      "step": 23355
    },
    {
      "entropy": 6.447816944122314,
      "epoch": 2.499759216651506,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043810743490518155,
      "loss": 5.8749,
      "mean_token_accuracy": 0.16440805345773696,
      "num_tokens": 50541195.0,
      "step": 23360
    },
    {
      "entropy": 6.489673185348511,
      "epoch": 2.5002942907592702,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043808109622608596,
      "loss": 5.9432,
      "mean_token_accuracy": 0.1590045690536499,
      "num_tokens": 50552655.0,
      "step": 23365
    },
    {
      "entropy": 6.493535566329956,
      "epoch": 2.500829364867034,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043805475283804657,
      "loss": 5.8832,
      "mean_token_accuracy": 0.16261259019374846,
      "num_tokens": 50562848.0,
      "step": 23370
    },
    {
      "entropy": 6.439454841613769,
      "epoch": 2.501364438974798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004380284047418241,
      "loss": 5.8296,
      "mean_token_accuracy": 0.17312100976705552,
      "num_tokens": 50572690.0,
      "step": 23375
    },
    {
      "entropy": 6.367242240905762,
      "epoch": 2.501899513082562,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004380020519381794,
      "loss": 5.7592,
      "mean_token_accuracy": 0.17836537957191467,
      "num_tokens": 50583826.0,
      "step": 23380
    },
    {
      "entropy": 6.372872161865234,
      "epoch": 2.5024345871903257,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043797569442787334,
      "loss": 5.7623,
      "mean_token_accuracy": 0.17213914394378663,
      "num_tokens": 50595148.0,
      "step": 23385
    },
    {
      "entropy": 6.471211385726929,
      "epoch": 2.50296966129809,
      "grad_norm": 1.0,
      "learning_rate": 0.000437949332211667,
      "loss": 5.9359,
      "mean_token_accuracy": 0.1627334624528885,
      "num_tokens": 50606568.0,
      "step": 23390
    },
    {
      "entropy": 6.524767017364502,
      "epoch": 2.503504735405854,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043792296529032156,
      "loss": 5.8595,
      "mean_token_accuracy": 0.16299757063388826,
      "num_tokens": 50618144.0,
      "step": 23395
    },
    {
      "entropy": 6.472301054000854,
      "epoch": 2.5040398095136176,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004378965936645984,
      "loss": 5.8102,
      "mean_token_accuracy": 0.17338538467884063,
      "num_tokens": 50629721.0,
      "step": 23400
    },
    {
      "entropy": 6.428107118606567,
      "epoch": 2.5045748836213817,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043787021733525904,
      "loss": 5.8837,
      "mean_token_accuracy": 0.1648982971906662,
      "num_tokens": 50640482.0,
      "step": 23405
    },
    {
      "entropy": 6.475130414962768,
      "epoch": 2.5051099577291454,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000437843836303065,
      "loss": 5.9069,
      "mean_token_accuracy": 0.15661599338054658,
      "num_tokens": 50652261.0,
      "step": 23410
    },
    {
      "entropy": 6.403379774093628,
      "epoch": 2.5056450318369095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004378174505687781,
      "loss": 5.7029,
      "mean_token_accuracy": 0.17376570850610734,
      "num_tokens": 50662397.0,
      "step": 23415
    },
    {
      "entropy": 6.39830322265625,
      "epoch": 2.506180105944673,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004377910601331601,
      "loss": 5.7968,
      "mean_token_accuracy": 0.16959466338157653,
      "num_tokens": 50673494.0,
      "step": 23420
    },
    {
      "entropy": 6.35598931312561,
      "epoch": 2.506715180052437,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004377646649969733,
      "loss": 5.7586,
      "mean_token_accuracy": 0.1781459629535675,
      "num_tokens": 50683963.0,
      "step": 23425
    },
    {
      "entropy": 6.417419385910034,
      "epoch": 2.5072502541602013,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043773826516097945,
      "loss": 5.8412,
      "mean_token_accuracy": 0.16739363521337508,
      "num_tokens": 50694449.0,
      "step": 23430
    },
    {
      "entropy": 6.457337045669556,
      "epoch": 2.507785328267965,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043771186062594114,
      "loss": 5.8061,
      "mean_token_accuracy": 0.17668635547161102,
      "num_tokens": 50705877.0,
      "step": 23435
    },
    {
      "entropy": 6.447578477859497,
      "epoch": 2.508320402375729,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043768545139262076,
      "loss": 5.8439,
      "mean_token_accuracy": 0.16916287690401077,
      "num_tokens": 50716581.0,
      "step": 23440
    },
    {
      "entropy": 6.458028078079224,
      "epoch": 2.508855476483493,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004376590374617809,
      "loss": 5.8216,
      "mean_token_accuracy": 0.16911400407552718,
      "num_tokens": 50727948.0,
      "step": 23445
    },
    {
      "entropy": 6.304333162307739,
      "epoch": 2.509390550591257,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043763261883418396,
      "loss": 5.6885,
      "mean_token_accuracy": 0.18946111500263213,
      "num_tokens": 50739189.0,
      "step": 23450
    },
    {
      "entropy": 6.367064189910889,
      "epoch": 2.509925624699021,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004376061955105932,
      "loss": 5.8097,
      "mean_token_accuracy": 0.17116736471652985,
      "num_tokens": 50749439.0,
      "step": 23455
    },
    {
      "entropy": 6.446600866317749,
      "epoch": 2.5104606988067846,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043757976749177137,
      "loss": 5.9111,
      "mean_token_accuracy": 0.16651144176721572,
      "num_tokens": 50761463.0,
      "step": 23460
    },
    {
      "entropy": 6.493279504776001,
      "epoch": 2.5109957729145487,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004375533347784816,
      "loss": 5.9242,
      "mean_token_accuracy": 0.1622188702225685,
      "num_tokens": 50771359.0,
      "step": 23465
    },
    {
      "entropy": 6.466159439086914,
      "epoch": 2.5115308470223123,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043752689737148705,
      "loss": 5.8474,
      "mean_token_accuracy": 0.16398313790559768,
      "num_tokens": 50781669.0,
      "step": 23470
    },
    {
      "entropy": 6.474824714660644,
      "epoch": 2.5120659211300764,
      "grad_norm": 1.59375,
      "learning_rate": 0.00043750045527155126,
      "loss": 5.7712,
      "mean_token_accuracy": 0.1701836422085762,
      "num_tokens": 50792217.0,
      "step": 23475
    },
    {
      "entropy": 6.45610933303833,
      "epoch": 2.5126009952378405,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004374740084794376,
      "loss": 5.8565,
      "mean_token_accuracy": 0.1657190889120102,
      "num_tokens": 50805288.0,
      "step": 23480
    },
    {
      "entropy": 6.387471723556518,
      "epoch": 2.513136069345604,
      "grad_norm": 1.0,
      "learning_rate": 0.0004374475569959098,
      "loss": 5.7915,
      "mean_token_accuracy": 0.17089854031801224,
      "num_tokens": 50815563.0,
      "step": 23485
    },
    {
      "entropy": 6.422699594497681,
      "epoch": 2.5136711434533683,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004374211008217316,
      "loss": 5.7848,
      "mean_token_accuracy": 0.173959943652153,
      "num_tokens": 50826108.0,
      "step": 23490
    },
    {
      "entropy": 6.40271029472351,
      "epoch": 2.5142062175611324,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043739463995766696,
      "loss": 5.788,
      "mean_token_accuracy": 0.17170742750167847,
      "num_tokens": 50837180.0,
      "step": 23495
    },
    {
      "entropy": 6.4022603034973145,
      "epoch": 2.514741291668896,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004373681744044799,
      "loss": 5.7999,
      "mean_token_accuracy": 0.17160523533821107,
      "num_tokens": 50847430.0,
      "step": 23500
    },
    {
      "entropy": 6.483809518814087,
      "epoch": 2.51527636577666,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043734170416293467,
      "loss": 5.8388,
      "mean_token_accuracy": 0.16768257617950438,
      "num_tokens": 50858100.0,
      "step": 23505
    },
    {
      "entropy": 6.476467657089233,
      "epoch": 2.5158114398844242,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043731522923379554,
      "loss": 5.8792,
      "mean_token_accuracy": 0.16560536473989487,
      "num_tokens": 50868549.0,
      "step": 23510
    },
    {
      "entropy": 6.475049829483032,
      "epoch": 2.516346513992188,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004372887496178269,
      "loss": 5.8304,
      "mean_token_accuracy": 0.16881112158298492,
      "num_tokens": 50879522.0,
      "step": 23515
    },
    {
      "entropy": 6.4281618118286135,
      "epoch": 2.516881588099952,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004372622653157935,
      "loss": 5.8281,
      "mean_token_accuracy": 0.17058689147233963,
      "num_tokens": 50891139.0,
      "step": 23520
    },
    {
      "entropy": 6.413624620437622,
      "epoch": 2.5174166622077156,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004372357763284599,
      "loss": 5.8691,
      "mean_token_accuracy": 0.15801187753677368,
      "num_tokens": 50901267.0,
      "step": 23525
    },
    {
      "entropy": 6.459148979187011,
      "epoch": 2.5179517363154797,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004372092826565911,
      "loss": 5.7575,
      "mean_token_accuracy": 0.1759127974510193,
      "num_tokens": 50912137.0,
      "step": 23530
    },
    {
      "entropy": 6.449383354187011,
      "epoch": 2.5184868104232434,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004371827843009521,
      "loss": 5.8665,
      "mean_token_accuracy": 0.1692303776741028,
      "num_tokens": 50922724.0,
      "step": 23535
    },
    {
      "entropy": 6.447641754150391,
      "epoch": 2.5190218845310075,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043715628126230795,
      "loss": 5.853,
      "mean_token_accuracy": 0.16304496675729752,
      "num_tokens": 50934617.0,
      "step": 23540
    },
    {
      "entropy": 6.488238763809204,
      "epoch": 2.5195569586387716,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004371297735414239,
      "loss": 5.8721,
      "mean_token_accuracy": 0.1622187927365303,
      "num_tokens": 50945179.0,
      "step": 23545
    },
    {
      "entropy": 6.46587290763855,
      "epoch": 2.5200920327465353,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043710326113906555,
      "loss": 5.8755,
      "mean_token_accuracy": 0.17131611555814744,
      "num_tokens": 50955383.0,
      "step": 23550
    },
    {
      "entropy": 6.464931154251099,
      "epoch": 2.5206271068542994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004370767440559983,
      "loss": 5.7797,
      "mean_token_accuracy": 0.1746327519416809,
      "num_tokens": 50966107.0,
      "step": 23555
    },
    {
      "entropy": 6.336547946929931,
      "epoch": 2.5211621809620635,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043705022229298775,
      "loss": 5.717,
      "mean_token_accuracy": 0.19079294949769973,
      "num_tokens": 50977042.0,
      "step": 23560
    },
    {
      "entropy": 6.4310201644897464,
      "epoch": 2.521697255069827,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004370236958507998,
      "loss": 5.892,
      "mean_token_accuracy": 0.1699656456708908,
      "num_tokens": 50986999.0,
      "step": 23565
    },
    {
      "entropy": 6.442835140228271,
      "epoch": 2.522232329177591,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004369971647302005,
      "loss": 5.8174,
      "mean_token_accuracy": 0.1668142110109329,
      "num_tokens": 50997824.0,
      "step": 23570
    },
    {
      "entropy": 6.551996660232544,
      "epoch": 2.522767403285355,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004369706289319558,
      "loss": 5.8913,
      "mean_token_accuracy": 0.16107341796159744,
      "num_tokens": 51009001.0,
      "step": 23575
    },
    {
      "entropy": 6.476627969741822,
      "epoch": 2.523302477393119,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043694408845683185,
      "loss": 5.8043,
      "mean_token_accuracy": 0.1726684272289276,
      "num_tokens": 51019222.0,
      "step": 23580
    },
    {
      "entropy": 6.458505725860595,
      "epoch": 2.5238375515008826,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004369175433055952,
      "loss": 5.8018,
      "mean_token_accuracy": 0.16967400908470154,
      "num_tokens": 51029473.0,
      "step": 23585
    },
    {
      "entropy": 6.4428849697113035,
      "epoch": 2.5243726256086467,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043689099347901227,
      "loss": 5.8408,
      "mean_token_accuracy": 0.16933100074529647,
      "num_tokens": 51040667.0,
      "step": 23590
    },
    {
      "entropy": 6.416579103469848,
      "epoch": 2.524907699716411,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043686443897784954,
      "loss": 5.7351,
      "mean_token_accuracy": 0.1749102830886841,
      "num_tokens": 51050386.0,
      "step": 23595
    },
    {
      "entropy": 6.420112943649292,
      "epoch": 2.5254427738241745,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000436837879802874,
      "loss": 5.8579,
      "mean_token_accuracy": 0.16801944077014924,
      "num_tokens": 51060752.0,
      "step": 23600
    },
    {
      "entropy": 6.387187862396241,
      "epoch": 2.5259778479319386,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043681131595485234,
      "loss": 5.8591,
      "mean_token_accuracy": 0.16453633457422256,
      "num_tokens": 51072467.0,
      "step": 23605
    },
    {
      "entropy": 6.540108489990234,
      "epoch": 2.5265129220397027,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004367847474345517,
      "loss": 5.9164,
      "mean_token_accuracy": 0.16510888934135437,
      "num_tokens": 51083404.0,
      "step": 23610
    },
    {
      "entropy": 6.492024183273315,
      "epoch": 2.5270479961474663,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043675817424273925,
      "loss": 5.8224,
      "mean_token_accuracy": 0.1736280605196953,
      "num_tokens": 51094392.0,
      "step": 23615
    },
    {
      "entropy": 6.492126226425171,
      "epoch": 2.5275830702552304,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043673159638018215,
      "loss": 5.8339,
      "mean_token_accuracy": 0.1681446313858032,
      "num_tokens": 51104706.0,
      "step": 23620
    },
    {
      "entropy": 6.437421894073486,
      "epoch": 2.5281181443629945,
      "grad_norm": 1.125,
      "learning_rate": 0.000436705013847648,
      "loss": 5.8839,
      "mean_token_accuracy": 0.16368280947208405,
      "num_tokens": 51114881.0,
      "step": 23625
    },
    {
      "entropy": 6.493309450149536,
      "epoch": 2.528653218470758,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004366784266459043,
      "loss": 5.9025,
      "mean_token_accuracy": 0.16390820294618608,
      "num_tokens": 51125705.0,
      "step": 23630
    },
    {
      "entropy": 6.5060052394866945,
      "epoch": 2.5291882925785223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043665183477571874,
      "loss": 5.8981,
      "mean_token_accuracy": 0.1617528200149536,
      "num_tokens": 51136460.0,
      "step": 23635
    },
    {
      "entropy": 6.4168013572692875,
      "epoch": 2.529723366686286,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004366252382378592,
      "loss": 5.8323,
      "mean_token_accuracy": 0.17082517445087433,
      "num_tokens": 51146206.0,
      "step": 23640
    },
    {
      "entropy": 6.438094806671143,
      "epoch": 2.53025844079405,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004365986370330935,
      "loss": 5.9304,
      "mean_token_accuracy": 0.16151246875524522,
      "num_tokens": 51156650.0,
      "step": 23645
    },
    {
      "entropy": 6.461386871337891,
      "epoch": 2.5307935149018137,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043657203116218994,
      "loss": 5.9054,
      "mean_token_accuracy": 0.1636781707406044,
      "num_tokens": 51168401.0,
      "step": 23650
    },
    {
      "entropy": 6.497487735748291,
      "epoch": 2.531328589009578,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043654542062591674,
      "loss": 5.8669,
      "mean_token_accuracy": 0.1578354448080063,
      "num_tokens": 51178571.0,
      "step": 23655
    },
    {
      "entropy": 6.466913509368896,
      "epoch": 2.531863663117342,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043651880542504214,
      "loss": 5.9076,
      "mean_token_accuracy": 0.16343983262777328,
      "num_tokens": 51189143.0,
      "step": 23660
    },
    {
      "entropy": 6.4277726173400875,
      "epoch": 2.5323987372251056,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043649218556033475,
      "loss": 5.873,
      "mean_token_accuracy": 0.1654747188091278,
      "num_tokens": 51200552.0,
      "step": 23665
    },
    {
      "entropy": 6.478661727905274,
      "epoch": 2.5329338113328697,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004364655610325632,
      "loss": 5.8801,
      "mean_token_accuracy": 0.1660689726471901,
      "num_tokens": 51210493.0,
      "step": 23670
    },
    {
      "entropy": 6.467978477478027,
      "epoch": 2.5334688854406338,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004364389318424962,
      "loss": 5.8596,
      "mean_token_accuracy": 0.16343122571706772,
      "num_tokens": 51220535.0,
      "step": 23675
    },
    {
      "entropy": 6.456945514678955,
      "epoch": 2.5340039595483974,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043641229799090285,
      "loss": 5.9254,
      "mean_token_accuracy": 0.1584337666630745,
      "num_tokens": 51232078.0,
      "step": 23680
    },
    {
      "entropy": 6.4547418594360355,
      "epoch": 2.5345390336561615,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000436385659478552,
      "loss": 5.8644,
      "mean_token_accuracy": 0.16946864873170853,
      "num_tokens": 51242750.0,
      "step": 23685
    },
    {
      "entropy": 6.5304957866668705,
      "epoch": 2.535074107763925,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043635901630621293,
      "loss": 5.8772,
      "mean_token_accuracy": 0.16203753054141998,
      "num_tokens": 51254295.0,
      "step": 23690
    },
    {
      "entropy": 6.457598686218262,
      "epoch": 2.5356091818716893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043633236847465497,
      "loss": 5.8383,
      "mean_token_accuracy": 0.17026639133691787,
      "num_tokens": 51265060.0,
      "step": 23695
    },
    {
      "entropy": 6.385176420211792,
      "epoch": 2.536144255979453,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004363057159846475,
      "loss": 5.7653,
      "mean_token_accuracy": 0.1685713529586792,
      "num_tokens": 51275623.0,
      "step": 23700
    },
    {
      "entropy": 6.427573490142822,
      "epoch": 2.536679330087217,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004362790588369602,
      "loss": 5.7445,
      "mean_token_accuracy": 0.18103867918252944,
      "num_tokens": 51285830.0,
      "step": 23705
    },
    {
      "entropy": 6.4926313877105715,
      "epoch": 2.537214404194981,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004362523970323627,
      "loss": 5.8582,
      "mean_token_accuracy": 0.17031506299972535,
      "num_tokens": 51297047.0,
      "step": 23710
    },
    {
      "entropy": 6.442020654678345,
      "epoch": 2.5377494783027448,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043622573057162484,
      "loss": 5.7941,
      "mean_token_accuracy": 0.17245282232761383,
      "num_tokens": 51307254.0,
      "step": 23715
    },
    {
      "entropy": 6.467164087295532,
      "epoch": 2.538284552410509,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004361990594555167,
      "loss": 5.9579,
      "mean_token_accuracy": 0.1555894210934639,
      "num_tokens": 51317796.0,
      "step": 23720
    },
    {
      "entropy": 6.486655807495117,
      "epoch": 2.538819626518273,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043617238368480845,
      "loss": 5.9385,
      "mean_token_accuracy": 0.16319254636764527,
      "num_tokens": 51330172.0,
      "step": 23725
    },
    {
      "entropy": 6.471320676803589,
      "epoch": 2.5393547006260366,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043614570326027015,
      "loss": 5.7886,
      "mean_token_accuracy": 0.1688579186797142,
      "num_tokens": 51340844.0,
      "step": 23730
    },
    {
      "entropy": 6.470260143280029,
      "epoch": 2.5398897747338007,
      "grad_norm": 1.125,
      "learning_rate": 0.0004361190181826724,
      "loss": 5.8188,
      "mean_token_accuracy": 0.1693990096449852,
      "num_tokens": 51352078.0,
      "step": 23735
    },
    {
      "entropy": 6.441140413284302,
      "epoch": 2.540424848841565,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043609232845278565,
      "loss": 5.9322,
      "mean_token_accuracy": 0.15433687269687651,
      "num_tokens": 51364121.0,
      "step": 23740
    },
    {
      "entropy": 6.46961350440979,
      "epoch": 2.5409599229493285,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004360656340713805,
      "loss": 5.8684,
      "mean_token_accuracy": 0.17891923040151597,
      "num_tokens": 51375046.0,
      "step": 23745
    },
    {
      "entropy": 6.448200988769531,
      "epoch": 2.541494997057092,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004360389350392278,
      "loss": 5.7739,
      "mean_token_accuracy": 0.1667317718267441,
      "num_tokens": 51386363.0,
      "step": 23750
    },
    {
      "entropy": 6.37206711769104,
      "epoch": 2.5420300711648562,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043601223135709846,
      "loss": 5.7622,
      "mean_token_accuracy": 0.1763673320412636,
      "num_tokens": 51398036.0,
      "step": 23755
    },
    {
      "entropy": 6.443735122680664,
      "epoch": 2.5425651452726203,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004359855230257636,
      "loss": 5.8455,
      "mean_token_accuracy": 0.1647912234067917,
      "num_tokens": 51408769.0,
      "step": 23760
    },
    {
      "entropy": 6.474340438842773,
      "epoch": 2.543100219380384,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043595881004599444,
      "loss": 5.8768,
      "mean_token_accuracy": 0.1703938990831375,
      "num_tokens": 51420315.0,
      "step": 23765
    },
    {
      "entropy": 6.463323974609375,
      "epoch": 2.543635293488148,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043593209241856216,
      "loss": 5.7999,
      "mean_token_accuracy": 0.1745542123913765,
      "num_tokens": 51432283.0,
      "step": 23770
    },
    {
      "entropy": 6.456630802154541,
      "epoch": 2.544170367595912,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043590537014423833,
      "loss": 5.8322,
      "mean_token_accuracy": 0.17336047142744065,
      "num_tokens": 51443151.0,
      "step": 23775
    },
    {
      "entropy": 6.367748069763183,
      "epoch": 2.544705441703676,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004358786432237946,
      "loss": 5.7676,
      "mean_token_accuracy": 0.16979248076677322,
      "num_tokens": 51453964.0,
      "step": 23780
    },
    {
      "entropy": 6.46165280342102,
      "epoch": 2.54524051581144,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004358519116580026,
      "loss": 5.8001,
      "mean_token_accuracy": 0.16945938915014266,
      "num_tokens": 51464236.0,
      "step": 23785
    },
    {
      "entropy": 6.383675956726075,
      "epoch": 2.545775589919204,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004358251754476344,
      "loss": 5.7931,
      "mean_token_accuracy": 0.1724151387810707,
      "num_tokens": 51474388.0,
      "step": 23790
    },
    {
      "entropy": 6.421171569824219,
      "epoch": 2.5463106640269677,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004357984345934617,
      "loss": 5.8211,
      "mean_token_accuracy": 0.1592171758413315,
      "num_tokens": 51485209.0,
      "step": 23795
    },
    {
      "entropy": 6.375242042541504,
      "epoch": 2.546845738134732,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004357716890962568,
      "loss": 5.7827,
      "mean_token_accuracy": 0.17202549278736115,
      "num_tokens": 51495993.0,
      "step": 23800
    },
    {
      "entropy": 6.436607837677002,
      "epoch": 2.5473808122424955,
      "grad_norm": 1.15625,
      "learning_rate": 0.000435744938956792,
      "loss": 5.778,
      "mean_token_accuracy": 0.17447566688060762,
      "num_tokens": 51506282.0,
      "step": 23805
    },
    {
      "entropy": 6.467179155349731,
      "epoch": 2.5479158863502596,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043571818417583976,
      "loss": 5.8852,
      "mean_token_accuracy": 0.16808252185583114,
      "num_tokens": 51517728.0,
      "step": 23810
    },
    {
      "entropy": 6.4342920780181885,
      "epoch": 2.548450960458023,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004356914247541724,
      "loss": 5.809,
      "mean_token_accuracy": 0.17132724970579147,
      "num_tokens": 51528769.0,
      "step": 23815
    },
    {
      "entropy": 6.444615030288697,
      "epoch": 2.5489860345657873,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043566466069256283,
      "loss": 5.8076,
      "mean_token_accuracy": 0.1738339066505432,
      "num_tokens": 51540148.0,
      "step": 23820
    },
    {
      "entropy": 6.447171926498413,
      "epoch": 2.5495211086735514,
      "grad_norm": 1.125,
      "learning_rate": 0.0004356378919917837,
      "loss": 5.7965,
      "mean_token_accuracy": 0.17074039578437805,
      "num_tokens": 51550657.0,
      "step": 23825
    },
    {
      "entropy": 6.475306177139283,
      "epoch": 2.550056182781315,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004356111186526081,
      "loss": 5.8343,
      "mean_token_accuracy": 0.16904133558273315,
      "num_tokens": 51560399.0,
      "step": 23830
    },
    {
      "entropy": 6.433189010620117,
      "epoch": 2.550591256889079,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043558434067580894,
      "loss": 5.8742,
      "mean_token_accuracy": 0.1664064809679985,
      "num_tokens": 51571463.0,
      "step": 23835
    },
    {
      "entropy": 6.430970954895019,
      "epoch": 2.5511263309968433,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00043555755806215946,
      "loss": 5.7776,
      "mean_token_accuracy": 0.17547609508037568,
      "num_tokens": 51583136.0,
      "step": 23840
    },
    {
      "entropy": 6.494675922393799,
      "epoch": 2.551661405104607,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00043553077081243314,
      "loss": 5.8842,
      "mean_token_accuracy": 0.16515042036771774,
      "num_tokens": 51594397.0,
      "step": 23845
    },
    {
      "entropy": 6.511730766296386,
      "epoch": 2.552196479212371,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004355039789274034,
      "loss": 5.9258,
      "mean_token_accuracy": 0.163612762093544,
      "num_tokens": 51605788.0,
      "step": 23850
    },
    {
      "entropy": 6.514416980743408,
      "epoch": 2.5527315533201347,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004354771824078437,
      "loss": 5.8244,
      "mean_token_accuracy": 0.16936957389116286,
      "num_tokens": 51615693.0,
      "step": 23855
    },
    {
      "entropy": 6.386693096160888,
      "epoch": 2.553266627427899,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00043545038125452793,
      "loss": 5.7557,
      "mean_token_accuracy": 0.1710447609424591,
      "num_tokens": 51626607.0,
      "step": 23860
    },
    {
      "entropy": 6.415850210189819,
      "epoch": 2.5538017015356624,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043542357546823003,
      "loss": 5.8373,
      "mean_token_accuracy": 0.17098686695098878,
      "num_tokens": 51636905.0,
      "step": 23865
    },
    {
      "entropy": 6.426484727859497,
      "epoch": 2.5543367756434265,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043539676504972393,
      "loss": 5.813,
      "mean_token_accuracy": 0.1699366346001625,
      "num_tokens": 51648142.0,
      "step": 23870
    },
    {
      "entropy": 6.534699821472168,
      "epoch": 2.5548718497511906,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004353699499997837,
      "loss": 5.8065,
      "mean_token_accuracy": 0.16456719785928725,
      "num_tokens": 51659583.0,
      "step": 23875
    },
    {
      "entropy": 6.4629209518432615,
      "epoch": 2.5554069238589543,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004353431303191837,
      "loss": 5.9154,
      "mean_token_accuracy": 0.1651685655117035,
      "num_tokens": 51670582.0,
      "step": 23880
    },
    {
      "entropy": 6.532435894012451,
      "epoch": 2.5559419979667184,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043531630600869846,
      "loss": 5.884,
      "mean_token_accuracy": 0.15909160375595094,
      "num_tokens": 51680773.0,
      "step": 23885
    },
    {
      "entropy": 6.492989826202392,
      "epoch": 2.5564770720744825,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004352894770691024,
      "loss": 5.8679,
      "mean_token_accuracy": 0.16574910879135132,
      "num_tokens": 51690680.0,
      "step": 23890
    },
    {
      "entropy": 6.50074954032898,
      "epoch": 2.557012146182246,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004352626435011701,
      "loss": 5.8379,
      "mean_token_accuracy": 0.16873417794704437,
      "num_tokens": 51700910.0,
      "step": 23895
    },
    {
      "entropy": 6.462224721908569,
      "epoch": 2.5575472202900102,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043523580530567654,
      "loss": 5.8299,
      "mean_token_accuracy": 0.16913949996232985,
      "num_tokens": 51712331.0,
      "step": 23900
    },
    {
      "entropy": 6.357103061676026,
      "epoch": 2.5580822943977743,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004352089624833966,
      "loss": 5.7466,
      "mean_token_accuracy": 0.17619720548391343,
      "num_tokens": 51722805.0,
      "step": 23905
    },
    {
      "entropy": 6.4146585941314695,
      "epoch": 2.558617368505538,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043518211503510535,
      "loss": 5.8074,
      "mean_token_accuracy": 0.1773345336318016,
      "num_tokens": 51733909.0,
      "step": 23910
    },
    {
      "entropy": 6.4401908874511715,
      "epoch": 2.559152442613302,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043515526296157813,
      "loss": 5.8793,
      "mean_token_accuracy": 0.16183648854494095,
      "num_tokens": 51745062.0,
      "step": 23915
    },
    {
      "entropy": 6.46946349143982,
      "epoch": 2.5596875167210658,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004351284062635901,
      "loss": 5.8981,
      "mean_token_accuracy": 0.16717908531427383,
      "num_tokens": 51756235.0,
      "step": 23920
    },
    {
      "entropy": 6.504279756546021,
      "epoch": 2.56022259082883,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000435101544941917,
      "loss": 5.8355,
      "mean_token_accuracy": 0.1666359320282936,
      "num_tokens": 51765689.0,
      "step": 23925
    },
    {
      "entropy": 6.42838191986084,
      "epoch": 2.5607576649365935,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004350746789973341,
      "loss": 5.695,
      "mean_token_accuracy": 0.17084926813840867,
      "num_tokens": 51777090.0,
      "step": 23930
    },
    {
      "entropy": 6.449845695495606,
      "epoch": 2.5612927390443576,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004350478084306174,
      "loss": 5.8637,
      "mean_token_accuracy": 0.16560357213020324,
      "num_tokens": 51787360.0,
      "step": 23935
    },
    {
      "entropy": 6.388508605957031,
      "epoch": 2.5618278131521217,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004350209332425428,
      "loss": 5.7795,
      "mean_token_accuracy": 0.17110726684331895,
      "num_tokens": 51799075.0,
      "step": 23940
    },
    {
      "entropy": 6.333416604995728,
      "epoch": 2.5623628872598854,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043499405343388613,
      "loss": 5.6764,
      "mean_token_accuracy": 0.1872022345662117,
      "num_tokens": 51810024.0,
      "step": 23945
    },
    {
      "entropy": 6.445424604415893,
      "epoch": 2.5628979613676495,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004349671690054238,
      "loss": 5.7879,
      "mean_token_accuracy": 0.16610025316476823,
      "num_tokens": 51820710.0,
      "step": 23950
    },
    {
      "entropy": 6.464283418655396,
      "epoch": 2.5634330354754136,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043494027995793176,
      "loss": 5.8151,
      "mean_token_accuracy": 0.1701695591211319,
      "num_tokens": 51831451.0,
      "step": 23955
    },
    {
      "entropy": 6.316296148300171,
      "epoch": 2.5639681095831772,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004349133862921867,
      "loss": 5.7627,
      "mean_token_accuracy": 0.17691007107496262,
      "num_tokens": 51841483.0,
      "step": 23960
    },
    {
      "entropy": 6.402124881744385,
      "epoch": 2.5645031836909413,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043488648800896516,
      "loss": 5.7981,
      "mean_token_accuracy": 0.17550319582223892,
      "num_tokens": 51852838.0,
      "step": 23965
    },
    {
      "entropy": 6.393918848037719,
      "epoch": 2.565038257798705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004348595851090436,
      "loss": 5.845,
      "mean_token_accuracy": 0.16831499338150024,
      "num_tokens": 51863851.0,
      "step": 23970
    },
    {
      "entropy": 6.437416934967041,
      "epoch": 2.565573331906469,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043483267759319907,
      "loss": 5.7759,
      "mean_token_accuracy": 0.17235716581344604,
      "num_tokens": 51874380.0,
      "step": 23975
    },
    {
      "entropy": 6.507405948638916,
      "epoch": 2.5661084060142327,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004348057654622084,
      "loss": 5.8238,
      "mean_token_accuracy": 0.1684352785348892,
      "num_tokens": 51884594.0,
      "step": 23980
    },
    {
      "entropy": 6.478658962249756,
      "epoch": 2.566643480121997,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004347788487168487,
      "loss": 5.9115,
      "mean_token_accuracy": 0.16663961708545685,
      "num_tokens": 51897679.0,
      "step": 23985
    },
    {
      "entropy": 6.499426031112671,
      "epoch": 2.567178554229761,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004347519273578972,
      "loss": 5.9753,
      "mean_token_accuracy": 0.1618331864476204,
      "num_tokens": 51907501.0,
      "step": 23990
    },
    {
      "entropy": 6.443717098236084,
      "epoch": 2.5677136283375246,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043472500138613124,
      "loss": 5.8384,
      "mean_token_accuracy": 0.16931139081716537,
      "num_tokens": 51917732.0,
      "step": 23995
    },
    {
      "entropy": 6.474013710021973,
      "epoch": 2.5682487024452887,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004346980708023284,
      "loss": 5.8897,
      "mean_token_accuracy": 0.16652529090642929,
      "num_tokens": 51929038.0,
      "step": 24000
    },
    {
      "epoch": 2.5682487024452887,
      "eval_entropy": 6.244062622281097,
      "eval_loss": 5.929659843444824,
      "eval_mean_token_accuracy": 0.17219330300890773,
      "eval_num_tokens": 51929038.0,
      "eval_runtime": 17.4605,
      "eval_samples_per_second": 1699.778,
      "eval_steps_per_second": 212.479,
      "step": 24000
    },
    {
      "entropy": 6.522502851486206,
      "epoch": 2.568783776553053,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004346711356072661,
      "loss": 5.9282,
      "mean_token_accuracy": 0.1610226333141327,
      "num_tokens": 51939481.0,
      "step": 24005
    },
    {
      "entropy": 6.562824058532715,
      "epoch": 2.5693188506608164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004346441958017222,
      "loss": 5.8884,
      "mean_token_accuracy": 0.16422121822834015,
      "num_tokens": 51951393.0,
      "step": 24010
    },
    {
      "entropy": 6.547173118591308,
      "epoch": 2.5698539247685805,
      "grad_norm": 1.125,
      "learning_rate": 0.0004346172513864746,
      "loss": 5.9312,
      "mean_token_accuracy": 0.16037261337041855,
      "num_tokens": 51962897.0,
      "step": 24015
    },
    {
      "entropy": 6.474521350860596,
      "epoch": 2.5703889988763446,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043459030236230137,
      "loss": 5.8534,
      "mean_token_accuracy": 0.16470853984355927,
      "num_tokens": 51974443.0,
      "step": 24020
    },
    {
      "entropy": 6.403206634521484,
      "epoch": 2.5709240729841083,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004345633487299805,
      "loss": 5.8492,
      "mean_token_accuracy": 0.16993072777986526,
      "num_tokens": 51985158.0,
      "step": 24025
    },
    {
      "entropy": 6.530393075942993,
      "epoch": 2.571459147091872,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043453639049029035,
      "loss": 5.9621,
      "mean_token_accuracy": 0.15907127857208253,
      "num_tokens": 51994956.0,
      "step": 24030
    },
    {
      "entropy": 6.499341011047363,
      "epoch": 2.571994221199636,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043450942764400937,
      "loss": 5.7362,
      "mean_token_accuracy": 0.1712825268507004,
      "num_tokens": 52004922.0,
      "step": 24035
    },
    {
      "entropy": 6.4011430740356445,
      "epoch": 2.5725292953074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004344824601919161,
      "loss": 5.785,
      "mean_token_accuracy": 0.17704152911901475,
      "num_tokens": 52014684.0,
      "step": 24040
    },
    {
      "entropy": 6.465809726715088,
      "epoch": 2.573064369415164,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043445548813478914,
      "loss": 5.8123,
      "mean_token_accuracy": 0.17359094470739364,
      "num_tokens": 52025283.0,
      "step": 24045
    },
    {
      "entropy": 6.497803544998169,
      "epoch": 2.573599443522928,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004344285114734074,
      "loss": 5.8918,
      "mean_token_accuracy": 0.1694222182035446,
      "num_tokens": 52036513.0,
      "step": 24050
    },
    {
      "entropy": 6.487059736251831,
      "epoch": 2.574134517630692,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043440153020854987,
      "loss": 5.7739,
      "mean_token_accuracy": 0.17726627886295318,
      "num_tokens": 52047230.0,
      "step": 24055
    },
    {
      "entropy": 6.471822261810303,
      "epoch": 2.5746695917384557,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004343745443409954,
      "loss": 5.8403,
      "mean_token_accuracy": 0.16966150104999542,
      "num_tokens": 52058889.0,
      "step": 24060
    },
    {
      "entropy": 6.409003114700317,
      "epoch": 2.5752046658462198,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043434755387152354,
      "loss": 5.8062,
      "mean_token_accuracy": 0.17280226796865464,
      "num_tokens": 52068478.0,
      "step": 24065
    },
    {
      "entropy": 6.401686334609986,
      "epoch": 2.575739739953984,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004343205588009133,
      "loss": 5.8352,
      "mean_token_accuracy": 0.1683609291911125,
      "num_tokens": 52079756.0,
      "step": 24070
    },
    {
      "entropy": 6.506153297424317,
      "epoch": 2.5762748140617475,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043429355912994437,
      "loss": 5.8196,
      "mean_token_accuracy": 0.17615403681993486,
      "num_tokens": 52090452.0,
      "step": 24075
    },
    {
      "entropy": 6.425630950927735,
      "epoch": 2.5768098881695116,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004342665548593964,
      "loss": 5.7634,
      "mean_token_accuracy": 0.16616225093603135,
      "num_tokens": 52101134.0,
      "step": 24080
    },
    {
      "entropy": 6.432850646972656,
      "epoch": 2.5773449622772753,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004342395459900489,
      "loss": 5.788,
      "mean_token_accuracy": 0.17184942364692687,
      "num_tokens": 52111529.0,
      "step": 24085
    },
    {
      "entropy": 6.379417943954468,
      "epoch": 2.5778800363850394,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043421253252268194,
      "loss": 5.7693,
      "mean_token_accuracy": 0.17021328210830688,
      "num_tokens": 52122779.0,
      "step": 24090
    },
    {
      "entropy": 6.446373987197876,
      "epoch": 2.578415110492803,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004341855144580754,
      "loss": 5.7954,
      "mean_token_accuracy": 0.16753811091184617,
      "num_tokens": 52134140.0,
      "step": 24095
    },
    {
      "entropy": 6.4210878849029545,
      "epoch": 2.578950184600567,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043415849179700956,
      "loss": 5.7972,
      "mean_token_accuracy": 0.16611952632665633,
      "num_tokens": 52146033.0,
      "step": 24100
    },
    {
      "entropy": 6.418718719482422,
      "epoch": 2.5794852587083312,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004341314645402647,
      "loss": 5.8025,
      "mean_token_accuracy": 0.16686518639326095,
      "num_tokens": 52157516.0,
      "step": 24105
    },
    {
      "entropy": 6.4422938346862795,
      "epoch": 2.580020332816095,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004341044326886211,
      "loss": 5.7733,
      "mean_token_accuracy": 0.1811210662126541,
      "num_tokens": 52167523.0,
      "step": 24110
    },
    {
      "entropy": 6.359648609161377,
      "epoch": 2.580555406923859,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004340773962428593,
      "loss": 5.7335,
      "mean_token_accuracy": 0.17815857976675034,
      "num_tokens": 52178235.0,
      "step": 24115
    },
    {
      "entropy": 6.401646852493286,
      "epoch": 2.581090481031623,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043405035520376013,
      "loss": 5.7161,
      "mean_token_accuracy": 0.1780039682984352,
      "num_tokens": 52188586.0,
      "step": 24120
    },
    {
      "entropy": 6.452901792526245,
      "epoch": 2.5816255551393867,
      "grad_norm": 1.125,
      "learning_rate": 0.0004340233095721043,
      "loss": 5.8262,
      "mean_token_accuracy": 0.17175919115543364,
      "num_tokens": 52199081.0,
      "step": 24125
    },
    {
      "entropy": 6.44770622253418,
      "epoch": 2.582160629247151,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004339962593486727,
      "loss": 5.8643,
      "mean_token_accuracy": 0.16861597895622255,
      "num_tokens": 52210324.0,
      "step": 24130
    },
    {
      "entropy": 6.4853517532348635,
      "epoch": 2.5826957033549145,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00043396920453424656,
      "loss": 5.8536,
      "mean_token_accuracy": 0.16487945467233658,
      "num_tokens": 52221973.0,
      "step": 24135
    },
    {
      "entropy": 6.4158460140228275,
      "epoch": 2.5832307774626786,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043394214512960683,
      "loss": 5.7868,
      "mean_token_accuracy": 0.16782636046409607,
      "num_tokens": 52232560.0,
      "step": 24140
    },
    {
      "entropy": 6.388903522491455,
      "epoch": 2.5837658515704423,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004339150811355351,
      "loss": 5.7331,
      "mean_token_accuracy": 0.18017787784337996,
      "num_tokens": 52243416.0,
      "step": 24145
    },
    {
      "entropy": 6.391653633117675,
      "epoch": 2.5843009256782064,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004338880125528127,
      "loss": 5.8089,
      "mean_token_accuracy": 0.17448946088552475,
      "num_tokens": 52254213.0,
      "step": 24150
    },
    {
      "entropy": 6.3710792541503904,
      "epoch": 2.5848359997859705,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043386093938222126,
      "loss": 5.7748,
      "mean_token_accuracy": 0.1710189864039421,
      "num_tokens": 52265381.0,
      "step": 24155
    },
    {
      "entropy": 6.4578227519989015,
      "epoch": 2.585371073893734,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004338338616245425,
      "loss": 5.7399,
      "mean_token_accuracy": 0.17466310709714888,
      "num_tokens": 52276407.0,
      "step": 24160
    },
    {
      "entropy": 6.4348060131073,
      "epoch": 2.585906148001498,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043380677928055834,
      "loss": 5.7989,
      "mean_token_accuracy": 0.17190621346235274,
      "num_tokens": 52286484.0,
      "step": 24165
    },
    {
      "entropy": 6.433262157440185,
      "epoch": 2.5864412221092623,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043377969235105075,
      "loss": 5.7701,
      "mean_token_accuracy": 0.1791671022772789,
      "num_tokens": 52297674.0,
      "step": 24170
    },
    {
      "entropy": 6.501714992523193,
      "epoch": 2.586976296217026,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004337526008368018,
      "loss": 5.9622,
      "mean_token_accuracy": 0.15873754769563675,
      "num_tokens": 52309100.0,
      "step": 24175
    },
    {
      "entropy": 6.501345634460449,
      "epoch": 2.58751137032479,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004337255047385938,
      "loss": 5.9036,
      "mean_token_accuracy": 0.1666340172290802,
      "num_tokens": 52320002.0,
      "step": 24180
    },
    {
      "entropy": 6.475076484680176,
      "epoch": 2.588046444432554,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043369840405720923,
      "loss": 5.7892,
      "mean_token_accuracy": 0.17012355029582976,
      "num_tokens": 52330073.0,
      "step": 24185
    },
    {
      "entropy": 6.480060291290283,
      "epoch": 2.588581518540318,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043367129879343054,
      "loss": 5.8428,
      "mean_token_accuracy": 0.16450835168361663,
      "num_tokens": 52340472.0,
      "step": 24190
    },
    {
      "entropy": 6.412850284576416,
      "epoch": 2.589116592648082,
      "grad_norm": 1.125,
      "learning_rate": 0.00043364418894804033,
      "loss": 5.8575,
      "mean_token_accuracy": 0.16781412810087204,
      "num_tokens": 52350654.0,
      "step": 24195
    },
    {
      "entropy": 6.391303205490113,
      "epoch": 2.5896516667558456,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043361707452182146,
      "loss": 5.7947,
      "mean_token_accuracy": 0.17011244595050812,
      "num_tokens": 52361412.0,
      "step": 24200
    },
    {
      "entropy": 6.4793055057525635,
      "epoch": 2.5901867408636097,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043358995551555693,
      "loss": 5.7895,
      "mean_token_accuracy": 0.17831196188926696,
      "num_tokens": 52372030.0,
      "step": 24205
    },
    {
      "entropy": 6.462514591217041,
      "epoch": 2.5907218149713733,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043356283193002966,
      "loss": 5.8521,
      "mean_token_accuracy": 0.1704242318868637,
      "num_tokens": 52382965.0,
      "step": 24210
    },
    {
      "entropy": 6.414969253540039,
      "epoch": 2.5912568890791374,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004335357037660229,
      "loss": 5.7663,
      "mean_token_accuracy": 0.17729626446962357,
      "num_tokens": 52393959.0,
      "step": 24215
    },
    {
      "entropy": 6.477960681915283,
      "epoch": 2.5917919631869015,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043350857102432,
      "loss": 5.8916,
      "mean_token_accuracy": 0.1637155756354332,
      "num_tokens": 52405142.0,
      "step": 24220
    },
    {
      "entropy": 6.440229320526123,
      "epoch": 2.592327037294665,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004334814337057043,
      "loss": 5.8595,
      "mean_token_accuracy": 0.16320289224386214,
      "num_tokens": 52415927.0,
      "step": 24225
    },
    {
      "entropy": 6.440408706665039,
      "epoch": 2.5928621114024293,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043345429181095957,
      "loss": 5.7683,
      "mean_token_accuracy": 0.1733090192079544,
      "num_tokens": 52426475.0,
      "step": 24230
    },
    {
      "entropy": 6.456961917877197,
      "epoch": 2.5933971855101934,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043342714534086945,
      "loss": 5.8,
      "mean_token_accuracy": 0.17235849499702455,
      "num_tokens": 52437239.0,
      "step": 24235
    },
    {
      "entropy": 6.4544930934906,
      "epoch": 2.593932259617957,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004333999942962177,
      "loss": 5.8624,
      "mean_token_accuracy": 0.15809603929519653,
      "num_tokens": 52447323.0,
      "step": 24240
    },
    {
      "entropy": 6.479000616073608,
      "epoch": 2.594467333725721,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043337283867778836,
      "loss": 5.8872,
      "mean_token_accuracy": 0.16996936351060868,
      "num_tokens": 52458465.0,
      "step": 24245
    },
    {
      "entropy": 6.446727895736695,
      "epoch": 2.595002407833485,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043334567848636563,
      "loss": 5.87,
      "mean_token_accuracy": 0.16617592573165893,
      "num_tokens": 52468809.0,
      "step": 24250
    },
    {
      "entropy": 6.4774548530578615,
      "epoch": 2.595537481941249,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043331851372273365,
      "loss": 5.8569,
      "mean_token_accuracy": 0.16399066150188446,
      "num_tokens": 52479456.0,
      "step": 24255
    },
    {
      "entropy": 6.431064081192017,
      "epoch": 2.5960725560490125,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043329134438767683,
      "loss": 5.7794,
      "mean_token_accuracy": 0.1719185620546341,
      "num_tokens": 52490146.0,
      "step": 24260
    },
    {
      "entropy": 6.481818199157715,
      "epoch": 2.5966076301567766,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004332641704819797,
      "loss": 5.8924,
      "mean_token_accuracy": 0.16406404078006745,
      "num_tokens": 52499875.0,
      "step": 24265
    },
    {
      "entropy": 6.421052122116089,
      "epoch": 2.5971427042645407,
      "grad_norm": 1.125,
      "learning_rate": 0.00043323699200642686,
      "loss": 5.7176,
      "mean_token_accuracy": 0.17414314448833465,
      "num_tokens": 52510297.0,
      "step": 24270
    },
    {
      "entropy": 6.260036659240723,
      "epoch": 2.5976777783723044,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004332098089618031,
      "loss": 5.6723,
      "mean_token_accuracy": 0.18745329976081848,
      "num_tokens": 52523111.0,
      "step": 24275
    },
    {
      "entropy": 6.408556222915649,
      "epoch": 2.5982128524800685,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043318262134889334,
      "loss": 5.7921,
      "mean_token_accuracy": 0.17890815883874894,
      "num_tokens": 52534097.0,
      "step": 24280
    },
    {
      "entropy": 6.453095626831055,
      "epoch": 2.5987479265878326,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004331554291684827,
      "loss": 5.8402,
      "mean_token_accuracy": 0.17172944396734238,
      "num_tokens": 52544985.0,
      "step": 24285
    },
    {
      "entropy": 6.484789133071899,
      "epoch": 2.5992830006955963,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004331282324213561,
      "loss": 5.8471,
      "mean_token_accuracy": 0.1695963606238365,
      "num_tokens": 52555772.0,
      "step": 24290
    },
    {
      "entropy": 6.434114742279053,
      "epoch": 2.5998180748033604,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004331010311082992,
      "loss": 5.8482,
      "mean_token_accuracy": 0.16421815752983093,
      "num_tokens": 52566578.0,
      "step": 24295
    },
    {
      "entropy": 6.4282794952392575,
      "epoch": 2.6003531489111245,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043307382523009717,
      "loss": 5.8958,
      "mean_token_accuracy": 0.16458659768104553,
      "num_tokens": 52578551.0,
      "step": 24300
    },
    {
      "entropy": 6.484844493865967,
      "epoch": 2.600888223018888,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004330466147875356,
      "loss": 5.799,
      "mean_token_accuracy": 0.17603293359279631,
      "num_tokens": 52589225.0,
      "step": 24305
    },
    {
      "entropy": 6.469280099868774,
      "epoch": 2.6014232971266518,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043301939978140024,
      "loss": 5.74,
      "mean_token_accuracy": 0.17997439801692963,
      "num_tokens": 52599572.0,
      "step": 24310
    },
    {
      "entropy": 6.501903104782104,
      "epoch": 2.601958371234416,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043299218021247693,
      "loss": 5.8592,
      "mean_token_accuracy": 0.16623570322990416,
      "num_tokens": 52610562.0,
      "step": 24315
    },
    {
      "entropy": 6.480174207687378,
      "epoch": 2.60249344534218,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004329649560815517,
      "loss": 5.905,
      "mean_token_accuracy": 0.16309404969215394,
      "num_tokens": 52621078.0,
      "step": 24320
    },
    {
      "entropy": 6.448821067810059,
      "epoch": 2.6030285194499436,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004329377273894104,
      "loss": 5.8227,
      "mean_token_accuracy": 0.1738484248518944,
      "num_tokens": 52632336.0,
      "step": 24325
    },
    {
      "entropy": 6.3790301322937015,
      "epoch": 2.6035635935577077,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004329104941368396,
      "loss": 5.715,
      "mean_token_accuracy": 0.1782524675130844,
      "num_tokens": 52643701.0,
      "step": 24330
    },
    {
      "entropy": 6.335832214355468,
      "epoch": 2.604098667665472,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004328832563246254,
      "loss": 5.7493,
      "mean_token_accuracy": 0.1845685735344887,
      "num_tokens": 52654052.0,
      "step": 24335
    },
    {
      "entropy": 6.3761434078216555,
      "epoch": 2.6046337417732355,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043285601395355426,
      "loss": 5.7259,
      "mean_token_accuracy": 0.1797483742237091,
      "num_tokens": 52665033.0,
      "step": 24340
    },
    {
      "entropy": 6.464943981170654,
      "epoch": 2.6051688158809996,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043282876702441296,
      "loss": 5.8755,
      "mean_token_accuracy": 0.16545893698930741,
      "num_tokens": 52674726.0,
      "step": 24345
    },
    {
      "entropy": 6.4300065517425535,
      "epoch": 2.6057038899887637,
      "grad_norm": 1.0,
      "learning_rate": 0.00043280151553798824,
      "loss": 5.8399,
      "mean_token_accuracy": 0.17543198466300963,
      "num_tokens": 52687221.0,
      "step": 24350
    },
    {
      "entropy": 6.433019828796387,
      "epoch": 2.6062389640965273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004327742594950669,
      "loss": 5.8732,
      "mean_token_accuracy": 0.16775529831647873,
      "num_tokens": 52697902.0,
      "step": 24355
    },
    {
      "entropy": 6.464955854415893,
      "epoch": 2.6067740382042914,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000432746998896436,
      "loss": 5.8074,
      "mean_token_accuracy": 0.16740150153636932,
      "num_tokens": 52707845.0,
      "step": 24360
    },
    {
      "entropy": 6.391427898406983,
      "epoch": 2.607309112312055,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004327197337428827,
      "loss": 5.7563,
      "mean_token_accuracy": 0.16954540014266967,
      "num_tokens": 52719254.0,
      "step": 24365
    },
    {
      "entropy": 6.446796941757202,
      "epoch": 2.607844186419819,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004326924640351942,
      "loss": 5.8604,
      "mean_token_accuracy": 0.16562969237565994,
      "num_tokens": 52729765.0,
      "step": 24370
    },
    {
      "entropy": 6.485124158859253,
      "epoch": 2.608379260527583,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043266518977415797,
      "loss": 5.7934,
      "mean_token_accuracy": 0.17366866320371627,
      "num_tokens": 52740997.0,
      "step": 24375
    },
    {
      "entropy": 6.502767515182495,
      "epoch": 2.608914334635347,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004326379109605615,
      "loss": 5.9208,
      "mean_token_accuracy": 0.16227914541959762,
      "num_tokens": 52753107.0,
      "step": 24380
    },
    {
      "entropy": 6.367576313018799,
      "epoch": 2.609449408743111,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004326106275951926,
      "loss": 5.7513,
      "mean_token_accuracy": 0.16398344337940216,
      "num_tokens": 52765453.0,
      "step": 24385
    },
    {
      "entropy": 6.459119558334351,
      "epoch": 2.6099844828508747,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004325833396788389,
      "loss": 5.8522,
      "mean_token_accuracy": 0.17112422734498978,
      "num_tokens": 52776145.0,
      "step": 24390
    },
    {
      "entropy": 6.5094801902771,
      "epoch": 2.610519556958639,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004325560472122884,
      "loss": 5.8634,
      "mean_token_accuracy": 0.161820188164711,
      "num_tokens": 52788404.0,
      "step": 24395
    },
    {
      "entropy": 6.495896577835083,
      "epoch": 2.611054631066403,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004325287501963293,
      "loss": 5.8672,
      "mean_token_accuracy": 0.1708492636680603,
      "num_tokens": 52800481.0,
      "step": 24400
    },
    {
      "entropy": 6.451962804794311,
      "epoch": 2.6115897051741666,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004325014486317495,
      "loss": 5.8203,
      "mean_token_accuracy": 0.17177785634994508,
      "num_tokens": 52811539.0,
      "step": 24405
    },
    {
      "entropy": 6.463801336288452,
      "epoch": 2.6121247792819307,
      "grad_norm": 1.125,
      "learning_rate": 0.0004324741425193376,
      "loss": 5.8492,
      "mean_token_accuracy": 0.17786547243595124,
      "num_tokens": 52822330.0,
      "step": 24410
    },
    {
      "entropy": 6.49189829826355,
      "epoch": 2.6126598533896943,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000432446831859882,
      "loss": 5.8372,
      "mean_token_accuracy": 0.1649198204278946,
      "num_tokens": 52832447.0,
      "step": 24415
    },
    {
      "entropy": 6.422043800354004,
      "epoch": 2.6131949274974584,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004324195166541712,
      "loss": 5.7737,
      "mean_token_accuracy": 0.17375193536281586,
      "num_tokens": 52843089.0,
      "step": 24420
    },
    {
      "entropy": 6.501670789718628,
      "epoch": 2.613730001605222,
      "grad_norm": 1.125,
      "learning_rate": 0.000432392196902994,
      "loss": 5.8354,
      "mean_token_accuracy": 0.17070451825857164,
      "num_tokens": 52854235.0,
      "step": 24425
    },
    {
      "entropy": 6.481780290603638,
      "epoch": 2.614265075712986,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004323648726071392,
      "loss": 5.8995,
      "mean_token_accuracy": 0.16882860660552979,
      "num_tokens": 52864693.0,
      "step": 24430
    },
    {
      "entropy": 6.455888128280639,
      "epoch": 2.6148001498207503,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043233754376739583,
      "loss": 5.8684,
      "mean_token_accuracy": 0.1621541202068329,
      "num_tokens": 52876410.0,
      "step": 24435
    },
    {
      "entropy": 6.47052149772644,
      "epoch": 2.615335223928514,
      "grad_norm": 1.125,
      "learning_rate": 0.00043231021038455297,
      "loss": 5.8514,
      "mean_token_accuracy": 0.16740756928920747,
      "num_tokens": 52887416.0,
      "step": 24440
    },
    {
      "entropy": 6.491811609268188,
      "epoch": 2.615870298036278,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043228287245939995,
      "loss": 5.8175,
      "mean_token_accuracy": 0.16909861862659453,
      "num_tokens": 52899191.0,
      "step": 24445
    },
    {
      "entropy": 6.46834168434143,
      "epoch": 2.616405372144042,
      "grad_norm": 1.125,
      "learning_rate": 0.000432255529992726,
      "loss": 5.8767,
      "mean_token_accuracy": 0.16728740334510803,
      "num_tokens": 52910479.0,
      "step": 24450
    },
    {
      "entropy": 6.488281917572022,
      "epoch": 2.6169404462518058,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043222818298532074,
      "loss": 5.799,
      "mean_token_accuracy": 0.17055121958255767,
      "num_tokens": 52921121.0,
      "step": 24455
    },
    {
      "entropy": 6.442463827133179,
      "epoch": 2.61747552035957,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043220083143797385,
      "loss": 5.8491,
      "mean_token_accuracy": 0.16266627833247185,
      "num_tokens": 52931604.0,
      "step": 24460
    },
    {
      "entropy": 6.416452646255493,
      "epoch": 2.618010594467334,
      "grad_norm": 1.109375,
      "learning_rate": 0.000432173475351475,
      "loss": 5.8245,
      "mean_token_accuracy": 0.1623881384730339,
      "num_tokens": 52942200.0,
      "step": 24465
    },
    {
      "entropy": 6.443952369689941,
      "epoch": 2.6185456685750976,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004321461147266141,
      "loss": 5.8117,
      "mean_token_accuracy": 0.1691933661699295,
      "num_tokens": 52954350.0,
      "step": 24470
    },
    {
      "entropy": 6.462041139602661,
      "epoch": 2.6190807426828617,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043211874956418134,
      "loss": 5.8273,
      "mean_token_accuracy": 0.18141989260911942,
      "num_tokens": 52965683.0,
      "step": 24475
    },
    {
      "entropy": 6.442408847808838,
      "epoch": 2.6196158167906254,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004320913798649667,
      "loss": 5.8298,
      "mean_token_accuracy": 0.17099935710430145,
      "num_tokens": 52975463.0,
      "step": 24480
    },
    {
      "entropy": 6.397478628158569,
      "epoch": 2.6201508908983895,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004320640056297605,
      "loss": 5.8297,
      "mean_token_accuracy": 0.1666795566678047,
      "num_tokens": 52986115.0,
      "step": 24485
    },
    {
      "entropy": 6.479571437835693,
      "epoch": 2.620685965006153,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004320366268593532,
      "loss": 5.8197,
      "mean_token_accuracy": 0.16897168159484863,
      "num_tokens": 52995770.0,
      "step": 24490
    },
    {
      "entropy": 6.457843065261841,
      "epoch": 2.6212210391139172,
      "grad_norm": 1.125,
      "learning_rate": 0.00043200924355453544,
      "loss": 5.8108,
      "mean_token_accuracy": 0.17015268057584762,
      "num_tokens": 53006490.0,
      "step": 24495
    },
    {
      "entropy": 6.4328361511230465,
      "epoch": 2.6217561132216813,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004319818557160978,
      "loss": 5.714,
      "mean_token_accuracy": 0.17988343983888627,
      "num_tokens": 53016129.0,
      "step": 24500
    },
    {
      "entropy": 6.432383441925049,
      "epoch": 2.622291187329445,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043195446334483116,
      "loss": 5.8181,
      "mean_token_accuracy": 0.16969407796859742,
      "num_tokens": 53026568.0,
      "step": 24505
    },
    {
      "entropy": 6.43871808052063,
      "epoch": 2.622826261437209,
      "grad_norm": 1.125,
      "learning_rate": 0.00043192706644152646,
      "loss": 5.7781,
      "mean_token_accuracy": 0.173400217294693,
      "num_tokens": 53036488.0,
      "step": 24510
    },
    {
      "entropy": 6.479575681686401,
      "epoch": 2.623361335544973,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004318996650069747,
      "loss": 5.8549,
      "mean_token_accuracy": 0.1698138400912285,
      "num_tokens": 53047195.0,
      "step": 24515
    },
    {
      "entropy": 6.480584812164307,
      "epoch": 2.623896409652737,
      "grad_norm": 1.0,
      "learning_rate": 0.0004318722590419672,
      "loss": 5.9245,
      "mean_token_accuracy": 0.1565020889043808,
      "num_tokens": 53058602.0,
      "step": 24520
    },
    {
      "entropy": 6.501772069931031,
      "epoch": 2.624431483760501,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004318448485472953,
      "loss": 5.8953,
      "mean_token_accuracy": 0.16519170701503755,
      "num_tokens": 53070822.0,
      "step": 24525
    },
    {
      "entropy": 6.448581409454346,
      "epoch": 2.6249665578682646,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043181743352375035,
      "loss": 5.8155,
      "mean_token_accuracy": 0.1700589567422867,
      "num_tokens": 53082166.0,
      "step": 24530
    },
    {
      "entropy": 6.499645137786866,
      "epoch": 2.6255016319760287,
      "grad_norm": 1.015625,
      "learning_rate": 0.000431790013972124,
      "loss": 5.9895,
      "mean_token_accuracy": 0.15900662839412688,
      "num_tokens": 53093688.0,
      "step": 24535
    },
    {
      "entropy": 6.47393569946289,
      "epoch": 2.6260367060837924,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004317625898932082,
      "loss": 5.8251,
      "mean_token_accuracy": 0.1710244208574295,
      "num_tokens": 53104793.0,
      "step": 24540
    },
    {
      "entropy": 6.451453638076782,
      "epoch": 2.6265717801915565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004317351612877946,
      "loss": 5.8473,
      "mean_token_accuracy": 0.16726931184530258,
      "num_tokens": 53115835.0,
      "step": 24545
    },
    {
      "entropy": 6.401977920532227,
      "epoch": 2.6271068542993206,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004317077281566751,
      "loss": 5.6966,
      "mean_token_accuracy": 0.1789967119693756,
      "num_tokens": 53127153.0,
      "step": 24550
    },
    {
      "entropy": 6.411933135986328,
      "epoch": 2.627641928407084,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043168029050064214,
      "loss": 5.7474,
      "mean_token_accuracy": 0.17534516602754593,
      "num_tokens": 53137323.0,
      "step": 24555
    },
    {
      "entropy": 6.480818128585815,
      "epoch": 2.6281770025148483,
      "grad_norm": 1.25,
      "learning_rate": 0.00043165284832048773,
      "loss": 5.9113,
      "mean_token_accuracy": 0.1650221213698387,
      "num_tokens": 53146858.0,
      "step": 24560
    },
    {
      "entropy": 6.412811803817749,
      "epoch": 2.6287120766226124,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043162540161700434,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16923512816429137,
      "num_tokens": 53157225.0,
      "step": 24565
    },
    {
      "entropy": 6.44229097366333,
      "epoch": 2.629247150730376,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043159795039098453,
      "loss": 5.8267,
      "mean_token_accuracy": 0.17323046922683716,
      "num_tokens": 53167571.0,
      "step": 24570
    },
    {
      "entropy": 6.452820539474487,
      "epoch": 2.62978222483814,
      "grad_norm": 1.125,
      "learning_rate": 0.0004315704946432209,
      "loss": 5.7911,
      "mean_token_accuracy": 0.1738385185599327,
      "num_tokens": 53178562.0,
      "step": 24575
    },
    {
      "entropy": 6.507671213150024,
      "epoch": 2.6303172989459043,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043154303437450617,
      "loss": 5.9036,
      "mean_token_accuracy": 0.1629347026348114,
      "num_tokens": 53188833.0,
      "step": 24580
    },
    {
      "entropy": 6.4435022354125975,
      "epoch": 2.630852373053668,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043151556958563337,
      "loss": 5.8419,
      "mean_token_accuracy": 0.1699503019452095,
      "num_tokens": 53199596.0,
      "step": 24585
    },
    {
      "entropy": 6.492887496948242,
      "epoch": 2.6313874471614316,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004314881002773955,
      "loss": 5.8286,
      "mean_token_accuracy": 0.1700746551156044,
      "num_tokens": 53209942.0,
      "step": 24590
    },
    {
      "entropy": 6.521886110305786,
      "epoch": 2.6319225212691957,
      "grad_norm": 1.0,
      "learning_rate": 0.0004314606264505857,
      "loss": 5.9234,
      "mean_token_accuracy": 0.16213921904563905,
      "num_tokens": 53221546.0,
      "step": 24595
    },
    {
      "entropy": 6.490083456039429,
      "epoch": 2.63245759537696,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004314331481059973,
      "loss": 5.8688,
      "mean_token_accuracy": 0.16733311712741852,
      "num_tokens": 53232343.0,
      "step": 24600
    },
    {
      "entropy": 6.422762393951416,
      "epoch": 2.6329926694847234,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004314056652444237,
      "loss": 5.8527,
      "mean_token_accuracy": 0.17834841907024385,
      "num_tokens": 53242820.0,
      "step": 24605
    },
    {
      "entropy": 6.537918663024902,
      "epoch": 2.6335277435924875,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043137817786665863,
      "loss": 5.8478,
      "mean_token_accuracy": 0.17067694216966628,
      "num_tokens": 53253503.0,
      "step": 24610
    },
    {
      "entropy": 6.525417470932007,
      "epoch": 2.6340628177002516,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043135068597349545,
      "loss": 5.8543,
      "mean_token_accuracy": 0.1705772951245308,
      "num_tokens": 53264160.0,
      "step": 24615
    },
    {
      "entropy": 6.442506170272827,
      "epoch": 2.6345978918080153,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004313231895657283,
      "loss": 5.8458,
      "mean_token_accuracy": 0.16806966066360474,
      "num_tokens": 53275002.0,
      "step": 24620
    },
    {
      "entropy": 6.392654323577881,
      "epoch": 2.6351329659157794,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043129568864415103,
      "loss": 5.7397,
      "mean_token_accuracy": 0.17496725767850876,
      "num_tokens": 53284216.0,
      "step": 24625
    },
    {
      "entropy": 6.443193197250366,
      "epoch": 2.6356680400235435,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043126818320955766,
      "loss": 5.773,
      "mean_token_accuracy": 0.1765234127640724,
      "num_tokens": 53294277.0,
      "step": 24630
    },
    {
      "entropy": 6.46633734703064,
      "epoch": 2.636203114131307,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004312406732627425,
      "loss": 5.8391,
      "mean_token_accuracy": 0.1686805009841919,
      "num_tokens": 53304448.0,
      "step": 24635
    },
    {
      "entropy": 6.4417150020599365,
      "epoch": 2.6367381882390712,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043121315880449983,
      "loss": 5.801,
      "mean_token_accuracy": 0.16806722134351731,
      "num_tokens": 53315765.0,
      "step": 24640
    },
    {
      "entropy": 6.352233409881592,
      "epoch": 2.637273262346835,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043118563983562415,
      "loss": 5.6962,
      "mean_token_accuracy": 0.18003236055374144,
      "num_tokens": 53326529.0,
      "step": 24645
    },
    {
      "entropy": 6.384424829483033,
      "epoch": 2.637808336454599,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043115811635691006,
      "loss": 5.7069,
      "mean_token_accuracy": 0.17665813565254213,
      "num_tokens": 53336132.0,
      "step": 24650
    },
    {
      "entropy": 6.407582807540893,
      "epoch": 2.6383434105623627,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004311305883691524,
      "loss": 5.8427,
      "mean_token_accuracy": 0.1719327375292778,
      "num_tokens": 53347284.0,
      "step": 24655
    },
    {
      "entropy": 6.426649379730224,
      "epoch": 2.6388784846701268,
      "grad_norm": 1.125,
      "learning_rate": 0.0004311030558731458,
      "loss": 5.8564,
      "mean_token_accuracy": 0.17096722573041917,
      "num_tokens": 53357872.0,
      "step": 24660
    },
    {
      "entropy": 6.474016952514648,
      "epoch": 2.639413558777891,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043107551886968545,
      "loss": 5.7834,
      "mean_token_accuracy": 0.17290949076414108,
      "num_tokens": 53370311.0,
      "step": 24665
    },
    {
      "entropy": 6.428773069381714,
      "epoch": 2.6399486328856545,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043104797735956643,
      "loss": 5.6804,
      "mean_token_accuracy": 0.1908307209610939,
      "num_tokens": 53380320.0,
      "step": 24670
    },
    {
      "entropy": 6.4640764713287355,
      "epoch": 2.6404837069934186,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000431020431343584,
      "loss": 5.9353,
      "mean_token_accuracy": 0.16229447722434998,
      "num_tokens": 53391311.0,
      "step": 24675
    },
    {
      "entropy": 6.413698530197143,
      "epoch": 2.6410187811011827,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004309928808225335,
      "loss": 5.9133,
      "mean_token_accuracy": 0.16719427555799485,
      "num_tokens": 53402367.0,
      "step": 24680
    },
    {
      "entropy": 6.440971851348877,
      "epoch": 2.6415538552089464,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004309653257972105,
      "loss": 5.8225,
      "mean_token_accuracy": 0.16705271750688552,
      "num_tokens": 53414253.0,
      "step": 24685
    },
    {
      "entropy": 6.513284015655517,
      "epoch": 2.6420889293167105,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043093776626841075,
      "loss": 5.8205,
      "mean_token_accuracy": 0.17728720158338546,
      "num_tokens": 53425727.0,
      "step": 24690
    },
    {
      "entropy": 6.465902328491211,
      "epoch": 2.6426240034244746,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043091020223692974,
      "loss": 5.834,
      "mean_token_accuracy": 0.17062907218933104,
      "num_tokens": 53435402.0,
      "step": 24695
    },
    {
      "entropy": 6.35536379814148,
      "epoch": 2.643159077532238,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043088263370356364,
      "loss": 5.742,
      "mean_token_accuracy": 0.1796056717634201,
      "num_tokens": 53447421.0,
      "step": 24700
    },
    {
      "entropy": 6.419341278076172,
      "epoch": 2.643694151640002,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004308550606691084,
      "loss": 5.7112,
      "mean_token_accuracy": 0.1818605676293373,
      "num_tokens": 53458346.0,
      "step": 24705
    },
    {
      "entropy": 6.439358758926391,
      "epoch": 2.644229225747766,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004308274831343601,
      "loss": 5.8319,
      "mean_token_accuracy": 0.17426709234714508,
      "num_tokens": 53469638.0,
      "step": 24710
    },
    {
      "entropy": 6.519730949401856,
      "epoch": 2.64476429985553,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043079990110011525,
      "loss": 5.8849,
      "mean_token_accuracy": 0.16534826308488845,
      "num_tokens": 53480645.0,
      "step": 24715
    },
    {
      "entropy": 6.469854879379272,
      "epoch": 2.6452993739632937,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043077231456717006,
      "loss": 5.8125,
      "mean_token_accuracy": 0.17016890794038772,
      "num_tokens": 53491958.0,
      "step": 24720
    },
    {
      "entropy": 6.435523891448975,
      "epoch": 2.645834448071058,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043074472353632124,
      "loss": 5.784,
      "mean_token_accuracy": 0.17269142121076583,
      "num_tokens": 53502183.0,
      "step": 24725
    },
    {
      "entropy": 6.425982284545898,
      "epoch": 2.646369522178822,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004307171280083653,
      "loss": 5.7947,
      "mean_token_accuracy": 0.16722883731126786,
      "num_tokens": 53513373.0,
      "step": 24730
    },
    {
      "entropy": 6.404912424087525,
      "epoch": 2.6469045962865856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004306895279840992,
      "loss": 5.7208,
      "mean_token_accuracy": 0.17724404186010362,
      "num_tokens": 53523668.0,
      "step": 24735
    },
    {
      "entropy": 6.486309957504273,
      "epoch": 2.6474396703943497,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004306619234643199,
      "loss": 5.8512,
      "mean_token_accuracy": 0.16618050932884215,
      "num_tokens": 53535642.0,
      "step": 24740
    },
    {
      "entropy": 6.30486912727356,
      "epoch": 2.647974744502114,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043063431444982447,
      "loss": 5.5994,
      "mean_token_accuracy": 0.19146352857351304,
      "num_tokens": 53546192.0,
      "step": 24745
    },
    {
      "entropy": 6.421440458297729,
      "epoch": 2.6485098186098774,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043060670094141005,
      "loss": 5.8102,
      "mean_token_accuracy": 0.17130795121192932,
      "num_tokens": 53557375.0,
      "step": 24750
    },
    {
      "entropy": 6.515907192230225,
      "epoch": 2.6490448927176415,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043057908293987404,
      "loss": 5.8839,
      "mean_token_accuracy": 0.1653410479426384,
      "num_tokens": 53568434.0,
      "step": 24755
    },
    {
      "entropy": 6.432010078430176,
      "epoch": 2.649579966825405,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004305514604460138,
      "loss": 5.8072,
      "mean_token_accuracy": 0.17700727432966232,
      "num_tokens": 53579095.0,
      "step": 24760
    },
    {
      "entropy": 6.413215780258179,
      "epoch": 2.6501150409331693,
      "grad_norm": 1.140625,
      "learning_rate": 0.000430523833460627,
      "loss": 5.8224,
      "mean_token_accuracy": 0.16994547098875046,
      "num_tokens": 53590082.0,
      "step": 24765
    },
    {
      "entropy": 6.432853269577026,
      "epoch": 2.650650115040933,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043049620198451147,
      "loss": 5.8883,
      "mean_token_accuracy": 0.16787045001983641,
      "num_tokens": 53600421.0,
      "step": 24770
    },
    {
      "entropy": 6.367246437072754,
      "epoch": 2.651185189148697,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043046856601846483,
      "loss": 5.7497,
      "mean_token_accuracy": 0.18025726675987244,
      "num_tokens": 53610994.0,
      "step": 24775
    },
    {
      "entropy": 6.479902362823486,
      "epoch": 2.651720263256461,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043044092556328517,
      "loss": 5.8478,
      "mean_token_accuracy": 0.1709132745862007,
      "num_tokens": 53622205.0,
      "step": 24780
    },
    {
      "entropy": 6.434491443634033,
      "epoch": 2.652255337364225,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043041328061977073,
      "loss": 5.7817,
      "mean_token_accuracy": 0.17370395362377167,
      "num_tokens": 53632440.0,
      "step": 24785
    },
    {
      "entropy": 6.397217702865601,
      "epoch": 2.652790411471989,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043038563118871965,
      "loss": 5.7922,
      "mean_token_accuracy": 0.16892056167125702,
      "num_tokens": 53642443.0,
      "step": 24790
    },
    {
      "entropy": 6.524038410186767,
      "epoch": 2.653325485579753,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004303579772709303,
      "loss": 5.9632,
      "mean_token_accuracy": 0.1587246298789978,
      "num_tokens": 53653671.0,
      "step": 24795
    },
    {
      "entropy": 6.465799474716187,
      "epoch": 2.6538605596875167,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043033031886720107,
      "loss": 5.7527,
      "mean_token_accuracy": 0.1770178735256195,
      "num_tokens": 53664874.0,
      "step": 24800
    },
    {
      "entropy": 6.473470830917359,
      "epoch": 2.6543956337952808,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004303026559783309,
      "loss": 5.8425,
      "mean_token_accuracy": 0.16595245003700257,
      "num_tokens": 53675623.0,
      "step": 24805
    },
    {
      "entropy": 6.458851957321167,
      "epoch": 2.6549307079030444,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004302749886051182,
      "loss": 5.8304,
      "mean_token_accuracy": 0.16906968504190445,
      "num_tokens": 53685754.0,
      "step": 24810
    },
    {
      "entropy": 6.429592275619507,
      "epoch": 2.6554657820108085,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004302473167483621,
      "loss": 5.7255,
      "mean_token_accuracy": 0.17613309472799302,
      "num_tokens": 53696925.0,
      "step": 24815
    },
    {
      "entropy": 6.459819412231445,
      "epoch": 2.656000856118572,
      "grad_norm": 1.125,
      "learning_rate": 0.00043021964040886154,
      "loss": 5.7945,
      "mean_token_accuracy": 0.16905299425125123,
      "num_tokens": 53708046.0,
      "step": 24820
    },
    {
      "entropy": 6.3388978958129885,
      "epoch": 2.6565359302263363,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004301919595874156,
      "loss": 5.6771,
      "mean_token_accuracy": 0.1831871196627617,
      "num_tokens": 53719179.0,
      "step": 24825
    },
    {
      "entropy": 6.439083242416382,
      "epoch": 2.6570710043341004,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004301642742848237,
      "loss": 5.8458,
      "mean_token_accuracy": 0.17032753080129623,
      "num_tokens": 53731523.0,
      "step": 24830
    },
    {
      "entropy": 6.510975360870361,
      "epoch": 2.657606078441864,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004301365845018853,
      "loss": 5.9518,
      "mean_token_accuracy": 0.1568746343255043,
      "num_tokens": 53742072.0,
      "step": 24835
    },
    {
      "entropy": 6.507861137390137,
      "epoch": 2.658141152549628,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004301088902393996,
      "loss": 5.8501,
      "mean_token_accuracy": 0.16987327188253404,
      "num_tokens": 53754695.0,
      "step": 24840
    },
    {
      "entropy": 6.400056982040406,
      "epoch": 2.6586762266573922,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004300811914981666,
      "loss": 5.7322,
      "mean_token_accuracy": 0.1801162138581276,
      "num_tokens": 53764587.0,
      "step": 24845
    },
    {
      "entropy": 6.506076192855835,
      "epoch": 2.659211300765156,
      "grad_norm": 1.15625,
      "learning_rate": 0.000430053488278986,
      "loss": 5.9146,
      "mean_token_accuracy": 0.15570441037416458,
      "num_tokens": 53775609.0,
      "step": 24850
    },
    {
      "entropy": 6.427083253860474,
      "epoch": 2.65974637487292,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043002578058265776,
      "loss": 5.8147,
      "mean_token_accuracy": 0.17026914954185485,
      "num_tokens": 53786495.0,
      "step": 24855
    },
    {
      "entropy": 6.439125919342041,
      "epoch": 2.660281448980684,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004299980684099818,
      "loss": 5.8112,
      "mean_token_accuracy": 0.17032919973134994,
      "num_tokens": 53797058.0,
      "step": 24860
    },
    {
      "entropy": 6.4773850440979,
      "epoch": 2.6608165230884477,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042997035176175834,
      "loss": 5.8481,
      "mean_token_accuracy": 0.16986709833145142,
      "num_tokens": 53808148.0,
      "step": 24865
    },
    {
      "entropy": 6.441238451004028,
      "epoch": 2.661351597196212,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004299426306387879,
      "loss": 5.7622,
      "mean_token_accuracy": 0.18048732876777648,
      "num_tokens": 53818375.0,
      "step": 24870
    },
    {
      "entropy": 6.488950586318969,
      "epoch": 2.6618866713039755,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004299149050418707,
      "loss": 5.8668,
      "mean_token_accuracy": 0.16638995856046676,
      "num_tokens": 53831273.0,
      "step": 24875
    },
    {
      "entropy": 6.436811208724976,
      "epoch": 2.6624217454117396,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042988717497180733,
      "loss": 5.7823,
      "mean_token_accuracy": 0.17514830976724624,
      "num_tokens": 53842492.0,
      "step": 24880
    },
    {
      "entropy": 6.462623262405396,
      "epoch": 2.6629568195195032,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042985944042939867,
      "loss": 5.815,
      "mean_token_accuracy": 0.16838769912719725,
      "num_tokens": 53853017.0,
      "step": 24885
    },
    {
      "entropy": 6.525988483428955,
      "epoch": 2.6634918936272673,
      "grad_norm": 1.25,
      "learning_rate": 0.00042983170141544535,
      "loss": 5.9108,
      "mean_token_accuracy": 0.16826982647180558,
      "num_tokens": 53863606.0,
      "step": 24890
    },
    {
      "entropy": 6.462504577636719,
      "epoch": 2.6640269677350314,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004298039579307484,
      "loss": 5.8342,
      "mean_token_accuracy": 0.17039280086755754,
      "num_tokens": 53874689.0,
      "step": 24895
    },
    {
      "entropy": 6.512143802642822,
      "epoch": 2.664562041842795,
      "grad_norm": 1.109375,
      "learning_rate": 0.000429776209976109,
      "loss": 5.798,
      "mean_token_accuracy": 0.1739049658179283,
      "num_tokens": 53885247.0,
      "step": 24900
    },
    {
      "entropy": 6.446283054351807,
      "epoch": 2.665097115950559,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004297484575523282,
      "loss": 5.8738,
      "mean_token_accuracy": 0.16764336675405503,
      "num_tokens": 53896829.0,
      "step": 24905
    },
    {
      "entropy": 6.407952737808228,
      "epoch": 2.6656321900583233,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042972070066020745,
      "loss": 5.7724,
      "mean_token_accuracy": 0.17526051253080369,
      "num_tokens": 53907236.0,
      "step": 24910
    },
    {
      "entropy": 6.463792085647583,
      "epoch": 2.666167264166087,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004296929393005482,
      "loss": 5.8339,
      "mean_token_accuracy": 0.16849246323108674,
      "num_tokens": 53918420.0,
      "step": 24915
    },
    {
      "entropy": 6.406598424911499,
      "epoch": 2.666702338273851,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004296651734741521,
      "loss": 5.7622,
      "mean_token_accuracy": 0.18343898952007293,
      "num_tokens": 53928949.0,
      "step": 24920
    },
    {
      "entropy": 6.462457609176636,
      "epoch": 2.6672374123816147,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042963740318182085,
      "loss": 5.8862,
      "mean_token_accuracy": 0.16612108945846557,
      "num_tokens": 53939332.0,
      "step": 24925
    },
    {
      "entropy": 6.411294555664062,
      "epoch": 2.667772486489379,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004296096284243563,
      "loss": 5.8122,
      "mean_token_accuracy": 0.16959052085876464,
      "num_tokens": 53950703.0,
      "step": 24930
    },
    {
      "entropy": 6.5040459632873535,
      "epoch": 2.6683075605971425,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004295818492025605,
      "loss": 5.8825,
      "mean_token_accuracy": 0.16431426405906677,
      "num_tokens": 53964061.0,
      "step": 24935
    },
    {
      "entropy": 6.451537466049194,
      "epoch": 2.6688426347049066,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004295540655172355,
      "loss": 5.8919,
      "mean_token_accuracy": 0.17233105450868608,
      "num_tokens": 53974980.0,
      "step": 24940
    },
    {
      "entropy": 6.395253133773804,
      "epoch": 2.6693777088126707,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004295262773691835,
      "loss": 5.7669,
      "mean_token_accuracy": 0.1778295159339905,
      "num_tokens": 53985437.0,
      "step": 24945
    },
    {
      "entropy": 6.491303730010986,
      "epoch": 2.6699127829204343,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042949848475920706,
      "loss": 5.8199,
      "mean_token_accuracy": 0.16982153207063674,
      "num_tokens": 53996952.0,
      "step": 24950
    },
    {
      "entropy": 6.490273380279541,
      "epoch": 2.6704478570281984,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004294706876881085,
      "loss": 5.7929,
      "mean_token_accuracy": 0.17697339206933976,
      "num_tokens": 54007727.0,
      "step": 24955
    },
    {
      "entropy": 6.319596958160401,
      "epoch": 2.6709829311359625,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042944288615669055,
      "loss": 5.6845,
      "mean_token_accuracy": 0.17902670949697494,
      "num_tokens": 54018275.0,
      "step": 24960
    },
    {
      "entropy": 6.431728315353394,
      "epoch": 2.671518005243726,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042941508016575597,
      "loss": 5.9123,
      "mean_token_accuracy": 0.16717791706323623,
      "num_tokens": 54030376.0,
      "step": 24965
    },
    {
      "entropy": 6.4254310131073,
      "epoch": 2.6720530793514903,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004293872697161077,
      "loss": 5.7948,
      "mean_token_accuracy": 0.16999568045139313,
      "num_tokens": 54041487.0,
      "step": 24970
    },
    {
      "entropy": 6.461040115356445,
      "epoch": 2.6725881534592544,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042935945480854866,
      "loss": 5.785,
      "mean_token_accuracy": 0.18070854991674423,
      "num_tokens": 54051273.0,
      "step": 24975
    },
    {
      "entropy": 6.4249639987945555,
      "epoch": 2.673123227567018,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004293316354438821,
      "loss": 5.7681,
      "mean_token_accuracy": 0.17518055737018584,
      "num_tokens": 54061910.0,
      "step": 24980
    },
    {
      "entropy": 6.453087234497071,
      "epoch": 2.6736583016747817,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004293038116229112,
      "loss": 5.879,
      "mean_token_accuracy": 0.165846386551857,
      "num_tokens": 54072980.0,
      "step": 24985
    },
    {
      "entropy": 6.414533948898315,
      "epoch": 2.674193375782546,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004292759833464394,
      "loss": 5.8029,
      "mean_token_accuracy": 0.1759504571557045,
      "num_tokens": 54083027.0,
      "step": 24990
    },
    {
      "entropy": 6.362003231048584,
      "epoch": 2.67472844989031,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042924815061527033,
      "loss": 5.753,
      "mean_token_accuracy": 0.1710598051548004,
      "num_tokens": 54094068.0,
      "step": 24995
    },
    {
      "entropy": 6.482504272460938,
      "epoch": 2.6752635239980735,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004292203134302076,
      "loss": 5.815,
      "mean_token_accuracy": 0.16888271123170853,
      "num_tokens": 54104683.0,
      "step": 25000
    },
    {
      "entropy": 6.438959836959839,
      "epoch": 2.6757985981058376,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000429192471792055,
      "loss": 5.7738,
      "mean_token_accuracy": 0.17088697701692582,
      "num_tokens": 54115357.0,
      "step": 25005
    },
    {
      "entropy": 6.466482400894165,
      "epoch": 2.6763336722136017,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004291646257016163,
      "loss": 5.8246,
      "mean_token_accuracy": 0.1695661425590515,
      "num_tokens": 54126410.0,
      "step": 25010
    },
    {
      "entropy": 6.399688625335694,
      "epoch": 2.6768687463213654,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042913677515969586,
      "loss": 5.7482,
      "mean_token_accuracy": 0.1759047344326973,
      "num_tokens": 54137279.0,
      "step": 25015
    },
    {
      "entropy": 6.463199424743652,
      "epoch": 2.6774038204291295,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004291089201670976,
      "loss": 5.8474,
      "mean_token_accuracy": 0.17263779044151306,
      "num_tokens": 54146521.0,
      "step": 25020
    },
    {
      "entropy": 6.493490695953369,
      "epoch": 2.6779388945368936,
      "grad_norm": 1.125,
      "learning_rate": 0.00042908106072462595,
      "loss": 5.8893,
      "mean_token_accuracy": 0.1589824691414833,
      "num_tokens": 54157277.0,
      "step": 25025
    },
    {
      "entropy": 6.493120241165161,
      "epoch": 2.6784739686446573,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004290531968330853,
      "loss": 5.9117,
      "mean_token_accuracy": 0.1629042848944664,
      "num_tokens": 54168078.0,
      "step": 25030
    },
    {
      "entropy": 6.519865274429321,
      "epoch": 2.6790090427524214,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004290253284932803,
      "loss": 5.8742,
      "mean_token_accuracy": 0.16669410169124604,
      "num_tokens": 54178567.0,
      "step": 25035
    },
    {
      "entropy": 6.459631538391113,
      "epoch": 2.679544116860185,
      "grad_norm": 1.125,
      "learning_rate": 0.0004289974557060155,
      "loss": 5.8133,
      "mean_token_accuracy": 0.16972366273403167,
      "num_tokens": 54189485.0,
      "step": 25040
    },
    {
      "entropy": 6.536159992218018,
      "epoch": 2.680079190967949,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042896957847209593,
      "loss": 5.9177,
      "mean_token_accuracy": 0.168818636238575,
      "num_tokens": 54199351.0,
      "step": 25045
    },
    {
      "entropy": 6.491835737228394,
      "epoch": 2.6806142650757128,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004289416967923264,
      "loss": 5.8501,
      "mean_token_accuracy": 0.16874407827854157,
      "num_tokens": 54209001.0,
      "step": 25050
    },
    {
      "entropy": 6.51226577758789,
      "epoch": 2.681149339183477,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042891381066751186,
      "loss": 5.8893,
      "mean_token_accuracy": 0.16996545791625978,
      "num_tokens": 54219783.0,
      "step": 25055
    },
    {
      "entropy": 6.49288969039917,
      "epoch": 2.681684413291241,
      "grad_norm": 1.125,
      "learning_rate": 0.00042888592009845776,
      "loss": 5.8202,
      "mean_token_accuracy": 0.172476689517498,
      "num_tokens": 54231337.0,
      "step": 25060
    },
    {
      "entropy": 6.449243116378784,
      "epoch": 2.6822194873990046,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004288580250859694,
      "loss": 5.8617,
      "mean_token_accuracy": 0.16993087381124497,
      "num_tokens": 54241650.0,
      "step": 25065
    },
    {
      "entropy": 6.440430927276611,
      "epoch": 2.6827545615067687,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004288301256308521,
      "loss": 5.8739,
      "mean_token_accuracy": 0.1602565750479698,
      "num_tokens": 54252223.0,
      "step": 25070
    },
    {
      "entropy": 6.465013551712036,
      "epoch": 2.683289635614533,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042880222173391155,
      "loss": 5.7709,
      "mean_token_accuracy": 0.1790974721312523,
      "num_tokens": 54262469.0,
      "step": 25075
    },
    {
      "entropy": 6.480777406692505,
      "epoch": 2.6838247097222965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004287743133959535,
      "loss": 5.8064,
      "mean_token_accuracy": 0.16949342638254167,
      "num_tokens": 54272885.0,
      "step": 25080
    },
    {
      "entropy": 6.476455307006836,
      "epoch": 2.6843597838300606,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004287464006177837,
      "loss": 5.8408,
      "mean_token_accuracy": 0.16763137578964232,
      "num_tokens": 54283327.0,
      "step": 25085
    },
    {
      "entropy": 6.42815842628479,
      "epoch": 2.6848948579378242,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004287184834002082,
      "loss": 5.7663,
      "mean_token_accuracy": 0.1754793107509613,
      "num_tokens": 54294761.0,
      "step": 25090
    },
    {
      "entropy": 6.500039672851562,
      "epoch": 2.6854299320455883,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042869056174403303,
      "loss": 5.8368,
      "mean_token_accuracy": 0.16968054473400115,
      "num_tokens": 54305205.0,
      "step": 25095
    },
    {
      "entropy": 6.483677244186401,
      "epoch": 2.685965006153352,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004286626356500647,
      "loss": 5.8632,
      "mean_token_accuracy": 0.16853777170181275,
      "num_tokens": 54315902.0,
      "step": 25100
    },
    {
      "entropy": 6.485949230194092,
      "epoch": 2.686500080261116,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004286347051191091,
      "loss": 5.8467,
      "mean_token_accuracy": 0.17048720568418502,
      "num_tokens": 54327345.0,
      "step": 25105
    },
    {
      "entropy": 6.512165164947509,
      "epoch": 2.68703515436888,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004286067701519731,
      "loss": 5.9014,
      "mean_token_accuracy": 0.16638799607753754,
      "num_tokens": 54337548.0,
      "step": 25110
    },
    {
      "entropy": 6.459859561920166,
      "epoch": 2.687570228476644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042857883074946324,
      "loss": 5.8602,
      "mean_token_accuracy": 0.1632643461227417,
      "num_tokens": 54348384.0,
      "step": 25115
    },
    {
      "entropy": 6.501186895370483,
      "epoch": 2.688105302584408,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004285508869123862,
      "loss": 5.8294,
      "mean_token_accuracy": 0.16407384723424911,
      "num_tokens": 54358245.0,
      "step": 25120
    },
    {
      "entropy": 6.404818773269653,
      "epoch": 2.688640376692172,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042852293864154885,
      "loss": 5.778,
      "mean_token_accuracy": 0.17113247513771057,
      "num_tokens": 54370421.0,
      "step": 25125
    },
    {
      "entropy": 6.4910118103027346,
      "epoch": 2.6891754507999357,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042849498593775827,
      "loss": 5.7801,
      "mean_token_accuracy": 0.16793135553598404,
      "num_tokens": 54381368.0,
      "step": 25130
    },
    {
      "entropy": 6.4969565868377686,
      "epoch": 2.6897105249077,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042846702880182147,
      "loss": 5.832,
      "mean_token_accuracy": 0.16521817147731782,
      "num_tokens": 54392226.0,
      "step": 25135
    },
    {
      "entropy": 6.435066747665405,
      "epoch": 2.690245599015464,
      "grad_norm": 1.125,
      "learning_rate": 0.0004284390672345458,
      "loss": 5.8199,
      "mean_token_accuracy": 0.16536412835121156,
      "num_tokens": 54402295.0,
      "step": 25140
    },
    {
      "entropy": 6.471045589447021,
      "epoch": 2.6907806731232276,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004284111012367387,
      "loss": 5.8844,
      "mean_token_accuracy": 0.1629992350935936,
      "num_tokens": 54413808.0,
      "step": 25145
    },
    {
      "entropy": 6.478885078430176,
      "epoch": 2.6913157472309917,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042838313080920755,
      "loss": 5.8811,
      "mean_token_accuracy": 0.16799707114696502,
      "num_tokens": 54425048.0,
      "step": 25150
    },
    {
      "entropy": 6.522646236419678,
      "epoch": 2.6918508213387553,
      "grad_norm": 1.125,
      "learning_rate": 0.00042835515595276,
      "loss": 5.8966,
      "mean_token_accuracy": 0.16508034616708755,
      "num_tokens": 54435908.0,
      "step": 25155
    },
    {
      "entropy": 6.399665069580078,
      "epoch": 2.6923858954465194,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004283271766682039,
      "loss": 5.826,
      "mean_token_accuracy": 0.17247272282838821,
      "num_tokens": 54447594.0,
      "step": 25160
    },
    {
      "entropy": 6.420532703399658,
      "epoch": 2.692920969554283,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004282991929563471,
      "loss": 5.7585,
      "mean_token_accuracy": 0.1725929081439972,
      "num_tokens": 54458918.0,
      "step": 25165
    },
    {
      "entropy": 6.424621295928955,
      "epoch": 2.693456043662047,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004282712048179977,
      "loss": 5.7707,
      "mean_token_accuracy": 0.17404652684926986,
      "num_tokens": 54469423.0,
      "step": 25170
    },
    {
      "entropy": 6.4649632453918455,
      "epoch": 2.6939911177698113,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004282432122539638,
      "loss": 5.9012,
      "mean_token_accuracy": 0.16889043003320695,
      "num_tokens": 54481895.0,
      "step": 25175
    },
    {
      "entropy": 6.474952173233032,
      "epoch": 2.694526191877575,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042821521526505363,
      "loss": 5.7913,
      "mean_token_accuracy": 0.1684905543923378,
      "num_tokens": 54492233.0,
      "step": 25180
    },
    {
      "entropy": 6.487647199630738,
      "epoch": 2.695061265985339,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004281872138520757,
      "loss": 5.7901,
      "mean_token_accuracy": 0.1751820668578148,
      "num_tokens": 54502366.0,
      "step": 25185
    },
    {
      "entropy": 6.396858835220337,
      "epoch": 2.695596340093103,
      "grad_norm": 1.125,
      "learning_rate": 0.00042815920801583845,
      "loss": 5.8269,
      "mean_token_accuracy": 0.1700294330716133,
      "num_tokens": 54512565.0,
      "step": 25190
    },
    {
      "entropy": 6.431601190567017,
      "epoch": 2.6961314142008668,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042813119775715066,
      "loss": 5.813,
      "mean_token_accuracy": 0.17214077562093735,
      "num_tokens": 54522999.0,
      "step": 25195
    },
    {
      "entropy": 6.455900859832764,
      "epoch": 2.696666488308631,
      "grad_norm": 1.25,
      "learning_rate": 0.00042810318307682095,
      "loss": 5.8267,
      "mean_token_accuracy": 0.17201851010322572,
      "num_tokens": 54533776.0,
      "step": 25200
    },
    {
      "entropy": 6.44603328704834,
      "epoch": 2.6972015624163945,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042807516397565845,
      "loss": 5.8223,
      "mean_token_accuracy": 0.1704736202955246,
      "num_tokens": 54544972.0,
      "step": 25205
    },
    {
      "entropy": 6.4355762004852295,
      "epoch": 2.6977366365241586,
      "grad_norm": 1.125,
      "learning_rate": 0.0004280471404544721,
      "loss": 5.8229,
      "mean_token_accuracy": 0.1683978721499443,
      "num_tokens": 54556034.0,
      "step": 25210
    },
    {
      "entropy": 6.456488561630249,
      "epoch": 2.6982717106319223,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042801911251407095,
      "loss": 5.7702,
      "mean_token_accuracy": 0.1768851712346077,
      "num_tokens": 54566662.0,
      "step": 25215
    },
    {
      "entropy": 6.445424365997314,
      "epoch": 2.6988067847396864,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004279910801552645,
      "loss": 5.8142,
      "mean_token_accuracy": 0.17107101678848266,
      "num_tokens": 54578889.0,
      "step": 25220
    },
    {
      "entropy": 6.504504442214966,
      "epoch": 2.6993418588474505,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004279630433788621,
      "loss": 5.899,
      "mean_token_accuracy": 0.1650948479771614,
      "num_tokens": 54590456.0,
      "step": 25225
    },
    {
      "entropy": 6.504015350341797,
      "epoch": 2.699876932955214,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004279350021856734,
      "loss": 5.8533,
      "mean_token_accuracy": 0.1658875212073326,
      "num_tokens": 54602327.0,
      "step": 25230
    },
    {
      "entropy": 6.531709146499634,
      "epoch": 2.7004120070629782,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042790695657650787,
      "loss": 5.8086,
      "mean_token_accuracy": 0.1715105414390564,
      "num_tokens": 54612999.0,
      "step": 25235
    },
    {
      "entropy": 6.4551318168640135,
      "epoch": 2.7009470811707423,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042787890655217553,
      "loss": 5.8072,
      "mean_token_accuracy": 0.165636545419693,
      "num_tokens": 54624332.0,
      "step": 25240
    },
    {
      "entropy": 6.380843591690064,
      "epoch": 2.701482155278506,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042785085211348624,
      "loss": 5.7776,
      "mean_token_accuracy": 0.1734172821044922,
      "num_tokens": 54635679.0,
      "step": 25245
    },
    {
      "entropy": 6.437270975112915,
      "epoch": 2.70201722938627,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042782279326125005,
      "loss": 5.8415,
      "mean_token_accuracy": 0.17035256922245026,
      "num_tokens": 54646344.0,
      "step": 25250
    },
    {
      "entropy": 6.522429370880127,
      "epoch": 2.702552303494034,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042779472999627724,
      "loss": 5.851,
      "mean_token_accuracy": 0.16565621197223662,
      "num_tokens": 54657395.0,
      "step": 25255
    },
    {
      "entropy": 6.424329614639282,
      "epoch": 2.703087377601798,
      "grad_norm": 1.125,
      "learning_rate": 0.000427766662319378,
      "loss": 5.8019,
      "mean_token_accuracy": 0.17603667378425597,
      "num_tokens": 54668755.0,
      "step": 25260
    },
    {
      "entropy": 6.4714983940124515,
      "epoch": 2.7036224517095615,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042773859023136285,
      "loss": 5.809,
      "mean_token_accuracy": 0.17232057601213455,
      "num_tokens": 54679170.0,
      "step": 25265
    },
    {
      "entropy": 6.441302394866943,
      "epoch": 2.7041575258173256,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004277105137330424,
      "loss": 5.8395,
      "mean_token_accuracy": 0.16444401144981385,
      "num_tokens": 54688728.0,
      "step": 25270
    },
    {
      "entropy": 6.431235027313233,
      "epoch": 2.7046925999250897,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042768243282522735,
      "loss": 5.8344,
      "mean_token_accuracy": 0.17216030210256578,
      "num_tokens": 54700236.0,
      "step": 25275
    },
    {
      "entropy": 6.468287706375122,
      "epoch": 2.7052276740328534,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042765434750872847,
      "loss": 5.775,
      "mean_token_accuracy": 0.17214246094226837,
      "num_tokens": 54710586.0,
      "step": 25280
    },
    {
      "entropy": 6.453936243057251,
      "epoch": 2.7057627481406175,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042762625778435676,
      "loss": 5.7754,
      "mean_token_accuracy": 0.1746569886803627,
      "num_tokens": 54721159.0,
      "step": 25285
    },
    {
      "entropy": 6.430946111679077,
      "epoch": 2.7062978222483816,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004275981636529233,
      "loss": 5.8337,
      "mean_token_accuracy": 0.1726752981543541,
      "num_tokens": 54731593.0,
      "step": 25290
    },
    {
      "entropy": 6.444863986968994,
      "epoch": 2.706832896356145,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004275700651152393,
      "loss": 5.7567,
      "mean_token_accuracy": 0.1762674331665039,
      "num_tokens": 54742890.0,
      "step": 25295
    },
    {
      "entropy": 6.376671361923218,
      "epoch": 2.7073679704639093,
      "grad_norm": 1.25,
      "learning_rate": 0.00042754196217211616,
      "loss": 5.8514,
      "mean_token_accuracy": 0.17293013334274293,
      "num_tokens": 54753344.0,
      "step": 25300
    },
    {
      "entropy": 6.4662456035614015,
      "epoch": 2.7079030445716734,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042751385482436523,
      "loss": 5.9108,
      "mean_token_accuracy": 0.1654234617948532,
      "num_tokens": 54763157.0,
      "step": 25305
    },
    {
      "entropy": 6.505421781539917,
      "epoch": 2.708438118679437,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042748574307279824,
      "loss": 5.803,
      "mean_token_accuracy": 0.1673574611544609,
      "num_tokens": 54772515.0,
      "step": 25310
    },
    {
      "entropy": 6.453595304489136,
      "epoch": 2.708973192787201,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004274576269182268,
      "loss": 5.7819,
      "mean_token_accuracy": 0.17233504503965377,
      "num_tokens": 54783178.0,
      "step": 25315
    },
    {
      "entropy": 6.4772439956665036,
      "epoch": 2.709508266894965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004274295063614629,
      "loss": 5.8847,
      "mean_token_accuracy": 0.16831636428833008,
      "num_tokens": 54794709.0,
      "step": 25320
    },
    {
      "entropy": 6.493694210052491,
      "epoch": 2.710043341002729,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004274013814033184,
      "loss": 5.8407,
      "mean_token_accuracy": 0.16986371725797653,
      "num_tokens": 54804821.0,
      "step": 25325
    },
    {
      "entropy": 6.502756500244141,
      "epoch": 2.7105784151104926,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042737325204460523,
      "loss": 5.8728,
      "mean_token_accuracy": 0.16646825075149535,
      "num_tokens": 54815291.0,
      "step": 25330
    },
    {
      "entropy": 6.267319869995117,
      "epoch": 2.7111134892182567,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000427345118286136,
      "loss": 5.6532,
      "mean_token_accuracy": 0.1861624911427498,
      "num_tokens": 54826891.0,
      "step": 25335
    },
    {
      "entropy": 6.48826231956482,
      "epoch": 2.7116485633260208,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004273169801287228,
      "loss": 5.8315,
      "mean_token_accuracy": 0.17124121636152267,
      "num_tokens": 54837545.0,
      "step": 25340
    },
    {
      "entropy": 6.505214548110962,
      "epoch": 2.7121836374337844,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042728883757317825,
      "loss": 5.9349,
      "mean_token_accuracy": 0.16102031022310256,
      "num_tokens": 54848028.0,
      "step": 25345
    },
    {
      "entropy": 6.468828344345093,
      "epoch": 2.7127187115415485,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042726069062031495,
      "loss": 5.8745,
      "mean_token_accuracy": 0.16409480422735215,
      "num_tokens": 54858679.0,
      "step": 25350
    },
    {
      "entropy": 6.456073522567749,
      "epoch": 2.7132537856493126,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042723253927094565,
      "loss": 5.7486,
      "mean_token_accuracy": 0.17322767823934554,
      "num_tokens": 54870892.0,
      "step": 25355
    },
    {
      "entropy": 6.440365362167358,
      "epoch": 2.7137888597570763,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000427204383525883,
      "loss": 5.7482,
      "mean_token_accuracy": 0.1723293513059616,
      "num_tokens": 54881131.0,
      "step": 25360
    },
    {
      "entropy": 6.444860601425171,
      "epoch": 2.7143239338648404,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004271762233859403,
      "loss": 5.8607,
      "mean_token_accuracy": 0.16640077233314515,
      "num_tokens": 54891532.0,
      "step": 25365
    },
    {
      "entropy": 6.425418281555176,
      "epoch": 2.714859007972604,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042714805885193046,
      "loss": 5.7983,
      "mean_token_accuracy": 0.17340895980596543,
      "num_tokens": 54902945.0,
      "step": 25370
    },
    {
      "entropy": 6.474296283721924,
      "epoch": 2.715394082080368,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042711988992466684,
      "loss": 5.8433,
      "mean_token_accuracy": 0.16776596754789352,
      "num_tokens": 54913943.0,
      "step": 25375
    },
    {
      "entropy": 6.506180000305176,
      "epoch": 2.715929156188132,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042709171660496275,
      "loss": 5.879,
      "mean_token_accuracy": 0.1733723521232605,
      "num_tokens": 54924241.0,
      "step": 25380
    },
    {
      "entropy": 6.453070402145386,
      "epoch": 2.716464230295896,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004270635388936317,
      "loss": 5.8134,
      "mean_token_accuracy": 0.17376256138086318,
      "num_tokens": 54934083.0,
      "step": 25385
    },
    {
      "entropy": 6.493812847137451,
      "epoch": 2.71699930440366,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042703535679148734,
      "loss": 5.9075,
      "mean_token_accuracy": 0.1614554539322853,
      "num_tokens": 54944928.0,
      "step": 25390
    },
    {
      "entropy": 6.390759468078613,
      "epoch": 2.7175343785114237,
      "grad_norm": 1.125,
      "learning_rate": 0.00042700717029934345,
      "loss": 5.7122,
      "mean_token_accuracy": 0.17519088387489318,
      "num_tokens": 54955715.0,
      "step": 25395
    },
    {
      "entropy": 6.455529069900512,
      "epoch": 2.7180694526191878,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004269789794180139,
      "loss": 5.8881,
      "mean_token_accuracy": 0.16690321266651154,
      "num_tokens": 54966149.0,
      "step": 25400
    },
    {
      "entropy": 6.461684274673462,
      "epoch": 2.718604526726952,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042695078414831263,
      "loss": 5.8516,
      "mean_token_accuracy": 0.1678011193871498,
      "num_tokens": 54976854.0,
      "step": 25405
    },
    {
      "entropy": 6.48012490272522,
      "epoch": 2.7191396008347155,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004269225844910538,
      "loss": 5.8608,
      "mean_token_accuracy": 0.1676257312297821,
      "num_tokens": 54987377.0,
      "step": 25410
    },
    {
      "entropy": 6.471225833892822,
      "epoch": 2.7196746749424796,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004268943804470518,
      "loss": 5.8582,
      "mean_token_accuracy": 0.16692287772893905,
      "num_tokens": 54999488.0,
      "step": 25415
    },
    {
      "entropy": 6.557813930511474,
      "epoch": 2.7202097490502437,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042686617201712074,
      "loss": 5.9239,
      "mean_token_accuracy": 0.15887483060359955,
      "num_tokens": 55011517.0,
      "step": 25420
    },
    {
      "entropy": 6.527052593231201,
      "epoch": 2.7207448231580074,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004268379592020754,
      "loss": 5.8518,
      "mean_token_accuracy": 0.16792556196451186,
      "num_tokens": 55022133.0,
      "step": 25425
    },
    {
      "entropy": 6.419015169143677,
      "epoch": 2.7212798972657715,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042680974200273036,
      "loss": 5.82,
      "mean_token_accuracy": 0.1723191425204277,
      "num_tokens": 55032684.0,
      "step": 25430
    },
    {
      "entropy": 6.546659231185913,
      "epoch": 2.721814971373535,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004267815204199003,
      "loss": 5.9033,
      "mean_token_accuracy": 0.17117930352687835,
      "num_tokens": 55042943.0,
      "step": 25435
    },
    {
      "entropy": 6.483098793029785,
      "epoch": 2.722350045481299,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004267532944544002,
      "loss": 5.8474,
      "mean_token_accuracy": 0.16500110030174256,
      "num_tokens": 55053805.0,
      "step": 25440
    },
    {
      "entropy": 6.380464649200439,
      "epoch": 2.722885119589063,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042672506410704495,
      "loss": 5.8291,
      "mean_token_accuracy": 0.16930353194475173,
      "num_tokens": 55066611.0,
      "step": 25445
    },
    {
      "entropy": 6.476042604446411,
      "epoch": 2.723420193696827,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004266968293786499,
      "loss": 5.8577,
      "mean_token_accuracy": 0.1675291106104851,
      "num_tokens": 55077954.0,
      "step": 25450
    },
    {
      "entropy": 6.472329139709473,
      "epoch": 2.723955267804591,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004266685902700302,
      "loss": 5.8584,
      "mean_token_accuracy": 0.17777703255414962,
      "num_tokens": 55090296.0,
      "step": 25455
    },
    {
      "entropy": 6.532119607925415,
      "epoch": 2.7244903419123547,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004266403467820013,
      "loss": 5.8492,
      "mean_token_accuracy": 0.16520194113254547,
      "num_tokens": 55100664.0,
      "step": 25460
    },
    {
      "entropy": 6.41826171875,
      "epoch": 2.725025416020119,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004266120989153786,
      "loss": 5.7386,
      "mean_token_accuracy": 0.17147511392831802,
      "num_tokens": 55111094.0,
      "step": 25465
    },
    {
      "entropy": 6.49174690246582,
      "epoch": 2.725560490127883,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042658384667097787,
      "loss": 5.8694,
      "mean_token_accuracy": 0.16757980734109879,
      "num_tokens": 55121877.0,
      "step": 25470
    },
    {
      "entropy": 6.410719919204712,
      "epoch": 2.7260955642356466,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004265555900496149,
      "loss": 5.7674,
      "mean_token_accuracy": 0.16902597695589067,
      "num_tokens": 55133087.0,
      "step": 25475
    },
    {
      "entropy": 6.4846961975097654,
      "epoch": 2.7266306383434107,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042652732905210555,
      "loss": 5.876,
      "mean_token_accuracy": 0.16199003607034684,
      "num_tokens": 55144257.0,
      "step": 25480
    },
    {
      "entropy": 6.4511054992675785,
      "epoch": 2.7271657124511743,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004264990636792659,
      "loss": 5.7396,
      "mean_token_accuracy": 0.18135810494422913,
      "num_tokens": 55154389.0,
      "step": 25485
    },
    {
      "entropy": 6.383771753311157,
      "epoch": 2.7277007865589384,
      "grad_norm": 1.109375,
      "learning_rate": 0.000426470793931912,
      "loss": 5.7443,
      "mean_token_accuracy": 0.18193051517009734,
      "num_tokens": 55165734.0,
      "step": 25490
    },
    {
      "entropy": 6.331616449356079,
      "epoch": 2.728235860666702,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004264425198108603,
      "loss": 5.7378,
      "mean_token_accuracy": 0.17816318720579147,
      "num_tokens": 55176448.0,
      "step": 25495
    },
    {
      "entropy": 6.423361968994141,
      "epoch": 2.728770934774466,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042641424131692707,
      "loss": 5.7819,
      "mean_token_accuracy": 0.17122667729854585,
      "num_tokens": 55187515.0,
      "step": 25500
    },
    {
      "entropy": 6.5026247024536135,
      "epoch": 2.7293060088822303,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004263859584509289,
      "loss": 5.8903,
      "mean_token_accuracy": 0.16885142475366594,
      "num_tokens": 55199236.0,
      "step": 25505
    },
    {
      "entropy": 6.430768013000488,
      "epoch": 2.729841082989994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004263576712136825,
      "loss": 5.8339,
      "mean_token_accuracy": 0.1734820395708084,
      "num_tokens": 55210151.0,
      "step": 25510
    },
    {
      "entropy": 6.4220703601837155,
      "epoch": 2.730376157097758,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042632937960600454,
      "loss": 5.8017,
      "mean_token_accuracy": 0.1655623123049736,
      "num_tokens": 55221322.0,
      "step": 25515
    },
    {
      "entropy": 6.416780138015747,
      "epoch": 2.730911231205522,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042630108362871205,
      "loss": 5.7474,
      "mean_token_accuracy": 0.18318707942962648,
      "num_tokens": 55231901.0,
      "step": 25520
    },
    {
      "entropy": 6.474559640884399,
      "epoch": 2.731446305313286,
      "grad_norm": 1.125,
      "learning_rate": 0.00042627278328262203,
      "loss": 5.7919,
      "mean_token_accuracy": 0.1746283233165741,
      "num_tokens": 55243154.0,
      "step": 25525
    },
    {
      "entropy": 6.432326936721802,
      "epoch": 2.73198137942105,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042624447856855164,
      "loss": 5.8071,
      "mean_token_accuracy": 0.17177635431289673,
      "num_tokens": 55253485.0,
      "step": 25530
    },
    {
      "entropy": 6.497948884963989,
      "epoch": 2.732516453528814,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042621616948731816,
      "loss": 5.8706,
      "mean_token_accuracy": 0.16470078080892564,
      "num_tokens": 55264936.0,
      "step": 25535
    },
    {
      "entropy": 6.470392560958862,
      "epoch": 2.7330515276365777,
      "grad_norm": 1.25,
      "learning_rate": 0.0004261878560397391,
      "loss": 5.8392,
      "mean_token_accuracy": 0.16869108825922013,
      "num_tokens": 55276138.0,
      "step": 25540
    },
    {
      "entropy": 6.515729379653931,
      "epoch": 2.7335866017443413,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042615953822663195,
      "loss": 5.8683,
      "mean_token_accuracy": 0.16806696057319642,
      "num_tokens": 55287398.0,
      "step": 25545
    },
    {
      "entropy": 6.570737028121949,
      "epoch": 2.7341216758521054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042613121604881426,
      "loss": 5.9011,
      "mean_token_accuracy": 0.16318141520023347,
      "num_tokens": 55298806.0,
      "step": 25550
    },
    {
      "entropy": 6.434325742721557,
      "epoch": 2.7346567499598695,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004261028895071041,
      "loss": 5.7841,
      "mean_token_accuracy": 0.17625530511140824,
      "num_tokens": 55309976.0,
      "step": 25555
    },
    {
      "entropy": 6.456106233596802,
      "epoch": 2.735191824067633,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042607455860231916,
      "loss": 5.8639,
      "mean_token_accuracy": 0.16404063403606414,
      "num_tokens": 55320806.0,
      "step": 25560
    },
    {
      "entropy": 6.461478900909424,
      "epoch": 2.7357268981753973,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004260462233352775,
      "loss": 5.8732,
      "mean_token_accuracy": 0.1695462554693222,
      "num_tokens": 55332153.0,
      "step": 25565
    },
    {
      "entropy": 6.495965528488159,
      "epoch": 2.7362619722831614,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042601788370679746,
      "loss": 5.8452,
      "mean_token_accuracy": 0.1649416595697403,
      "num_tokens": 55343595.0,
      "step": 25570
    },
    {
      "entropy": 6.478165054321289,
      "epoch": 2.736797046390925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042598953971769723,
      "loss": 5.8675,
      "mean_token_accuracy": 0.16082160025835038,
      "num_tokens": 55353891.0,
      "step": 25575
    },
    {
      "entropy": 6.479030466079712,
      "epoch": 2.737332120498689,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004259611913687952,
      "loss": 5.7832,
      "mean_token_accuracy": 0.17106559872627258,
      "num_tokens": 55364743.0,
      "step": 25580
    },
    {
      "entropy": 6.539567565917968,
      "epoch": 2.7378671946064532,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042593283866091,
      "loss": 5.9015,
      "mean_token_accuracy": 0.15888729095458984,
      "num_tokens": 55375122.0,
      "step": 25585
    },
    {
      "entropy": 6.445958948135376,
      "epoch": 2.738402268714217,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042590448159486035,
      "loss": 5.8389,
      "mean_token_accuracy": 0.1658326953649521,
      "num_tokens": 55387281.0,
      "step": 25590
    },
    {
      "entropy": 6.491234540939331,
      "epoch": 2.738937342821981,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042587612017146496,
      "loss": 5.8472,
      "mean_token_accuracy": 0.15737028568983077,
      "num_tokens": 55397859.0,
      "step": 25595
    },
    {
      "entropy": 6.475116014480591,
      "epoch": 2.7394724169297446,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042584775439154276,
      "loss": 5.7991,
      "mean_token_accuracy": 0.17171958833932877,
      "num_tokens": 55408169.0,
      "step": 25600
    },
    {
      "entropy": 6.4353853225708,
      "epoch": 2.7400074910375087,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004258193842559129,
      "loss": 5.7669,
      "mean_token_accuracy": 0.16976314336061477,
      "num_tokens": 55419342.0,
      "step": 25605
    },
    {
      "entropy": 6.355871868133545,
      "epoch": 2.7405425651452724,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004257910097653945,
      "loss": 5.7401,
      "mean_token_accuracy": 0.17486047893762588,
      "num_tokens": 55430271.0,
      "step": 25610
    },
    {
      "entropy": 6.5034730434417725,
      "epoch": 2.7410776392530365,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004257626309208069,
      "loss": 5.7814,
      "mean_token_accuracy": 0.17025331109762193,
      "num_tokens": 55441901.0,
      "step": 25615
    },
    {
      "entropy": 6.451963567733765,
      "epoch": 2.7416127133608006,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004257342477229695,
      "loss": 5.8358,
      "mean_token_accuracy": 0.17120668590068816,
      "num_tokens": 55451743.0,
      "step": 25620
    },
    {
      "entropy": 6.480689334869385,
      "epoch": 2.7421477874685642,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042570586017270196,
      "loss": 5.7974,
      "mean_token_accuracy": 0.17127918750047683,
      "num_tokens": 55462805.0,
      "step": 25625
    },
    {
      "entropy": 6.405721378326416,
      "epoch": 2.7426828615763283,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042567746827082377,
      "loss": 5.8786,
      "mean_token_accuracy": 0.17034100592136384,
      "num_tokens": 55474277.0,
      "step": 25630
    },
    {
      "entropy": 6.469999933242798,
      "epoch": 2.7432179356840924,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004256490720181549,
      "loss": 5.8043,
      "mean_token_accuracy": 0.16280681043863296,
      "num_tokens": 55485257.0,
      "step": 25635
    },
    {
      "entropy": 6.374029684066772,
      "epoch": 2.743753009791856,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042562067141551523,
      "loss": 5.7468,
      "mean_token_accuracy": 0.17351360619068146,
      "num_tokens": 55495835.0,
      "step": 25640
    },
    {
      "entropy": 6.5328452587127686,
      "epoch": 2.74428808389962,
      "grad_norm": 1.125,
      "learning_rate": 0.00042559226646372486,
      "loss": 5.94,
      "mean_token_accuracy": 0.16532618552446365,
      "num_tokens": 55506256.0,
      "step": 25645
    },
    {
      "entropy": 6.489005088806152,
      "epoch": 2.744823158007384,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000425563857163604,
      "loss": 5.7818,
      "mean_token_accuracy": 0.18599790334701538,
      "num_tokens": 55516821.0,
      "step": 25650
    },
    {
      "entropy": 6.455095672607422,
      "epoch": 2.745358232115148,
      "grad_norm": 1.125,
      "learning_rate": 0.00042553544351597286,
      "loss": 5.837,
      "mean_token_accuracy": 0.16232891678810119,
      "num_tokens": 55527153.0,
      "step": 25655
    },
    {
      "entropy": 6.456127166748047,
      "epoch": 2.7458933062229116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042550702552165195,
      "loss": 5.899,
      "mean_token_accuracy": 0.16093794852495194,
      "num_tokens": 55538219.0,
      "step": 25660
    },
    {
      "entropy": 6.515220689773559,
      "epoch": 2.7464283803306757,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042547860318146183,
      "loss": 5.837,
      "mean_token_accuracy": 0.16682583093643188,
      "num_tokens": 55548733.0,
      "step": 25665
    },
    {
      "entropy": 6.511574029922485,
      "epoch": 2.74696345443844,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042545017649622314,
      "loss": 5.8197,
      "mean_token_accuracy": 0.17100900113582612,
      "num_tokens": 55559746.0,
      "step": 25670
    },
    {
      "entropy": 6.48558087348938,
      "epoch": 2.7474985285462035,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004254217454667569,
      "loss": 5.798,
      "mean_token_accuracy": 0.17104122191667556,
      "num_tokens": 55570322.0,
      "step": 25675
    },
    {
      "entropy": 6.356256341934204,
      "epoch": 2.7480336026539676,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004253933100938837,
      "loss": 5.7298,
      "mean_token_accuracy": 0.17463203370571137,
      "num_tokens": 55580715.0,
      "step": 25680
    },
    {
      "entropy": 6.459119987487793,
      "epoch": 2.7485686767617317,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004253648703784249,
      "loss": 5.7599,
      "mean_token_accuracy": 0.18006957173347474,
      "num_tokens": 55591685.0,
      "step": 25685
    },
    {
      "entropy": 6.488014125823975,
      "epoch": 2.7491037508694953,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004253364263212016,
      "loss": 5.8022,
      "mean_token_accuracy": 0.16619215458631514,
      "num_tokens": 55602288.0,
      "step": 25690
    },
    {
      "entropy": 6.467062711715698,
      "epoch": 2.7496388249772594,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004253079779230352,
      "loss": 5.8929,
      "mean_token_accuracy": 0.16972920000553132,
      "num_tokens": 55613011.0,
      "step": 25695
    },
    {
      "entropy": 6.410215568542481,
      "epoch": 2.7501738990850235,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004252795251847469,
      "loss": 5.8206,
      "mean_token_accuracy": 0.17466325908899308,
      "num_tokens": 55623630.0,
      "step": 25700
    },
    {
      "entropy": 6.39742078781128,
      "epoch": 2.750708973192787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042525106810715845,
      "loss": 5.7704,
      "mean_token_accuracy": 0.17414259910583496,
      "num_tokens": 55634764.0,
      "step": 25705
    },
    {
      "entropy": 6.555657291412354,
      "epoch": 2.7512440473005513,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042522260669109157,
      "loss": 5.8521,
      "mean_token_accuracy": 0.1689560279250145,
      "num_tokens": 55645085.0,
      "step": 25710
    },
    {
      "entropy": 6.514989852905273,
      "epoch": 2.751779121408315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000425194140937368,
      "loss": 5.9201,
      "mean_token_accuracy": 0.15902235358953476,
      "num_tokens": 55656249.0,
      "step": 25715
    },
    {
      "entropy": 6.488837623596192,
      "epoch": 2.752314195516079,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042516567084680977,
      "loss": 5.8549,
      "mean_token_accuracy": 0.17407193183898925,
      "num_tokens": 55668138.0,
      "step": 25720
    },
    {
      "entropy": 6.525090980529785,
      "epoch": 2.7528492696238427,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004251371964202389,
      "loss": 5.8117,
      "mean_token_accuracy": 0.17391688078641893,
      "num_tokens": 55678747.0,
      "step": 25725
    },
    {
      "entropy": 6.4792896747589115,
      "epoch": 2.753384343731607,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004251087176584775,
      "loss": 5.8856,
      "mean_token_accuracy": 0.1676110491156578,
      "num_tokens": 55690266.0,
      "step": 25730
    },
    {
      "entropy": 6.45226559638977,
      "epoch": 2.753919417839371,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042508023456234805,
      "loss": 5.8341,
      "mean_token_accuracy": 0.165319000184536,
      "num_tokens": 55700229.0,
      "step": 25735
    },
    {
      "entropy": 6.458577489852905,
      "epoch": 2.7544544919471345,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042505174713267286,
      "loss": 5.8549,
      "mean_token_accuracy": 0.16421509981155397,
      "num_tokens": 55711564.0,
      "step": 25740
    },
    {
      "entropy": 6.468289947509765,
      "epoch": 2.7549895660548986,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004250232553702746,
      "loss": 5.7937,
      "mean_token_accuracy": 0.17414254546165467,
      "num_tokens": 55722618.0,
      "step": 25745
    },
    {
      "entropy": 6.421773099899292,
      "epoch": 2.7555246401626627,
      "grad_norm": 1.125,
      "learning_rate": 0.0004249947592759759,
      "loss": 5.7783,
      "mean_token_accuracy": 0.17552315592765808,
      "num_tokens": 55732828.0,
      "step": 25750
    },
    {
      "entropy": 6.4087584018707275,
      "epoch": 2.7560597142704264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004249662588505996,
      "loss": 5.7905,
      "mean_token_accuracy": 0.16677282005548477,
      "num_tokens": 55742283.0,
      "step": 25755
    },
    {
      "entropy": 6.463822317123413,
      "epoch": 2.7565947883781905,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042493775409496867,
      "loss": 5.8183,
      "mean_token_accuracy": 0.17059487700462342,
      "num_tokens": 55752883.0,
      "step": 25760
    },
    {
      "entropy": 6.527851295471192,
      "epoch": 2.757129862485954,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004249092450099062,
      "loss": 5.851,
      "mean_token_accuracy": 0.16927163302898407,
      "num_tokens": 55762730.0,
      "step": 25765
    },
    {
      "entropy": 6.36800103187561,
      "epoch": 2.7576649365937183,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004248807315962353,
      "loss": 5.792,
      "mean_token_accuracy": 0.17454822808504106,
      "num_tokens": 55774068.0,
      "step": 25770
    },
    {
      "entropy": 6.4587695598602295,
      "epoch": 2.758200010701482,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004248522138547793,
      "loss": 5.8469,
      "mean_token_accuracy": 0.16078879535198212,
      "num_tokens": 55784896.0,
      "step": 25775
    },
    {
      "entropy": 6.441817903518677,
      "epoch": 2.758735084809246,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004248236917863617,
      "loss": 5.8052,
      "mean_token_accuracy": 0.17552930116653442,
      "num_tokens": 55796186.0,
      "step": 25780
    },
    {
      "entropy": 6.422281742095947,
      "epoch": 2.75927015891701,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042479516539180605,
      "loss": 5.7555,
      "mean_token_accuracy": 0.18010578900575638,
      "num_tokens": 55806439.0,
      "step": 25785
    },
    {
      "entropy": 6.436918020248413,
      "epoch": 2.7598052330247738,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000424766634671936,
      "loss": 5.8435,
      "mean_token_accuracy": 0.1630051001906395,
      "num_tokens": 55817955.0,
      "step": 25790
    },
    {
      "entropy": 6.462068891525268,
      "epoch": 2.760340307132538,
      "grad_norm": 1.125,
      "learning_rate": 0.00042473809962757553,
      "loss": 5.8059,
      "mean_token_accuracy": 0.17338243126869202,
      "num_tokens": 55828876.0,
      "step": 25795
    },
    {
      "entropy": 6.461212491989135,
      "epoch": 2.760875381240302,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004247095602595485,
      "loss": 5.7421,
      "mean_token_accuracy": 0.1780534952878952,
      "num_tokens": 55839721.0,
      "step": 25800
    },
    {
      "entropy": 6.382177686691284,
      "epoch": 2.7614104553480656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004246810165686788,
      "loss": 5.8483,
      "mean_token_accuracy": 0.1714698851108551,
      "num_tokens": 55850427.0,
      "step": 25805
    },
    {
      "entropy": 6.498190879821777,
      "epoch": 2.7619455294558297,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042465246855579093,
      "loss": 5.8777,
      "mean_token_accuracy": 0.16088765263557434,
      "num_tokens": 55860245.0,
      "step": 25810
    },
    {
      "entropy": 6.503045177459716,
      "epoch": 2.762480603563594,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042462391622170897,
      "loss": 5.8588,
      "mean_token_accuracy": 0.17062283009290696,
      "num_tokens": 55869957.0,
      "step": 25815
    },
    {
      "entropy": 6.476087808609009,
      "epoch": 2.7630156776713575,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004245953595672575,
      "loss": 5.9013,
      "mean_token_accuracy": 0.15957446545362472,
      "num_tokens": 55881159.0,
      "step": 25820
    },
    {
      "entropy": 6.589024782180786,
      "epoch": 2.7635507517791216,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000424566798593261,
      "loss": 5.8583,
      "mean_token_accuracy": 0.16577644646167755,
      "num_tokens": 55891830.0,
      "step": 25825
    },
    {
      "entropy": 6.515221643447876,
      "epoch": 2.7640858258868852,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042453823330054427,
      "loss": 5.8548,
      "mean_token_accuracy": 0.16769682615995407,
      "num_tokens": 55901904.0,
      "step": 25830
    },
    {
      "entropy": 6.447419500350952,
      "epoch": 2.7646208999946493,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000424509663689932,
      "loss": 5.7691,
      "mean_token_accuracy": 0.16778302043676377,
      "num_tokens": 55913808.0,
      "step": 25835
    },
    {
      "entropy": 6.403764915466309,
      "epoch": 2.765155974102413,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004244810897622492,
      "loss": 5.8008,
      "mean_token_accuracy": 0.17449028939008712,
      "num_tokens": 55924997.0,
      "step": 25840
    },
    {
      "entropy": 6.499400329589844,
      "epoch": 2.765691048210177,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000424452511518321,
      "loss": 5.9233,
      "mean_token_accuracy": 0.16195693910121917,
      "num_tokens": 55935623.0,
      "step": 25845
    },
    {
      "entropy": 6.503520965576172,
      "epoch": 2.766226122317941,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004244239289589724,
      "loss": 5.8728,
      "mean_token_accuracy": 0.16624418497085572,
      "num_tokens": 55946733.0,
      "step": 25850
    },
    {
      "entropy": 6.483802747726441,
      "epoch": 2.766761196425705,
      "grad_norm": 1.109375,
      "learning_rate": 0.000424395342085029,
      "loss": 5.9255,
      "mean_token_accuracy": 0.1647105485200882,
      "num_tokens": 55956158.0,
      "step": 25855
    },
    {
      "entropy": 6.466374254226684,
      "epoch": 2.767296270533469,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042436675089731603,
      "loss": 5.8009,
      "mean_token_accuracy": 0.1741783544421196,
      "num_tokens": 55966442.0,
      "step": 25860
    },
    {
      "entropy": 6.494933700561523,
      "epoch": 2.767831344641233,
      "grad_norm": 1.125,
      "learning_rate": 0.0004243381553966591,
      "loss": 5.8625,
      "mean_token_accuracy": 0.16725597232580186,
      "num_tokens": 55977314.0,
      "step": 25865
    },
    {
      "entropy": 6.45178189277649,
      "epoch": 2.7683664187489967,
      "grad_norm": 1.53125,
      "learning_rate": 0.00042430955558388396,
      "loss": 5.7732,
      "mean_token_accuracy": 0.17609888762235643,
      "num_tokens": 55987457.0,
      "step": 25870
    },
    {
      "entropy": 6.456023454666138,
      "epoch": 2.768901492856761,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042428095145981634,
      "loss": 5.796,
      "mean_token_accuracy": 0.1772090807557106,
      "num_tokens": 55998319.0,
      "step": 25875
    },
    {
      "entropy": 6.415425872802734,
      "epoch": 2.7694365669645244,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042425234302528224,
      "loss": 5.7022,
      "mean_token_accuracy": 0.17721234858036042,
      "num_tokens": 56008654.0,
      "step": 25880
    },
    {
      "entropy": 6.474841547012329,
      "epoch": 2.7699716410722885,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004242237302811077,
      "loss": 5.8938,
      "mean_token_accuracy": 0.16309083998203278,
      "num_tokens": 56019793.0,
      "step": 25885
    },
    {
      "entropy": 6.480176305770874,
      "epoch": 2.770506715180052,
      "grad_norm": 1.0625,
      "learning_rate": 0.000424195113228119,
      "loss": 5.8265,
      "mean_token_accuracy": 0.17035767138004304,
      "num_tokens": 56031051.0,
      "step": 25890
    },
    {
      "entropy": 6.509409952163696,
      "epoch": 2.7710417892878163,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004241664918671423,
      "loss": 5.929,
      "mean_token_accuracy": 0.1634727194905281,
      "num_tokens": 56041210.0,
      "step": 25895
    },
    {
      "entropy": 6.437812471389771,
      "epoch": 2.7715768633955804,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000424137866199004,
      "loss": 5.7971,
      "mean_token_accuracy": 0.1686531960964203,
      "num_tokens": 56051601.0,
      "step": 25900
    },
    {
      "entropy": 6.4550598621368405,
      "epoch": 2.772111937503344,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004241092362245309,
      "loss": 5.7322,
      "mean_token_accuracy": 0.17782142907381057,
      "num_tokens": 56062669.0,
      "step": 25905
    },
    {
      "entropy": 6.407167387008667,
      "epoch": 2.772647011611108,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042408060194454956,
      "loss": 5.8685,
      "mean_token_accuracy": 0.1704555481672287,
      "num_tokens": 56074414.0,
      "step": 25910
    },
    {
      "entropy": 6.481329011917114,
      "epoch": 2.7731820857188723,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042405196335988676,
      "loss": 5.8269,
      "mean_token_accuracy": 0.1686076045036316,
      "num_tokens": 56085428.0,
      "step": 25915
    },
    {
      "entropy": 6.48163595199585,
      "epoch": 2.773717159826636,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042402332047136943,
      "loss": 5.8398,
      "mean_token_accuracy": 0.16365103870630265,
      "num_tokens": 56097403.0,
      "step": 25920
    },
    {
      "entropy": 6.404376268386841,
      "epoch": 2.7742522339344,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004239946732798247,
      "loss": 5.7455,
      "mean_token_accuracy": 0.17978830486536027,
      "num_tokens": 56107781.0,
      "step": 25925
    },
    {
      "entropy": 6.366741037368774,
      "epoch": 2.774787308042164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042396602178607973,
      "loss": 5.7133,
      "mean_token_accuracy": 0.18355431109666825,
      "num_tokens": 56118698.0,
      "step": 25930
    },
    {
      "entropy": 6.43956732749939,
      "epoch": 2.7753223821499278,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004239373659909618,
      "loss": 5.8244,
      "mean_token_accuracy": 0.16843785643577575,
      "num_tokens": 56128397.0,
      "step": 25935
    },
    {
      "entropy": 6.432868766784668,
      "epoch": 2.7758574562576914,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004239087058952984,
      "loss": 5.8116,
      "mean_token_accuracy": 0.17392237037420272,
      "num_tokens": 56138798.0,
      "step": 25940
    },
    {
      "entropy": 6.473392343521118,
      "epoch": 2.7763925303654555,
      "grad_norm": 1.078125,
      "learning_rate": 0.000423880041499917,
      "loss": 5.7558,
      "mean_token_accuracy": 0.1807953491806984,
      "num_tokens": 56149936.0,
      "step": 25945
    },
    {
      "entropy": 6.471223974227906,
      "epoch": 2.7769276044732196,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004238513728056453,
      "loss": 5.8385,
      "mean_token_accuracy": 0.1677446722984314,
      "num_tokens": 56161433.0,
      "step": 25950
    },
    {
      "entropy": 6.482017612457275,
      "epoch": 2.7774626785809833,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042382269981331116,
      "loss": 5.8815,
      "mean_token_accuracy": 0.16234824806451797,
      "num_tokens": 56172312.0,
      "step": 25955
    },
    {
      "entropy": 6.447206163406372,
      "epoch": 2.7779977526887474,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042379402252374244,
      "loss": 5.7662,
      "mean_token_accuracy": 0.17454010844230652,
      "num_tokens": 56183887.0,
      "step": 25960
    },
    {
      "entropy": 6.433726787567139,
      "epoch": 2.7785328267965115,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042376534093776727,
      "loss": 5.8001,
      "mean_token_accuracy": 0.17445893734693527,
      "num_tokens": 56194172.0,
      "step": 25965
    },
    {
      "entropy": 6.457703399658203,
      "epoch": 2.779067900904275,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004237366550562137,
      "loss": 5.9378,
      "mean_token_accuracy": 0.1615542307496071,
      "num_tokens": 56204569.0,
      "step": 25970
    },
    {
      "entropy": 6.455522203445435,
      "epoch": 2.7796029750120392,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042370796487991013,
      "loss": 5.813,
      "mean_token_accuracy": 0.16980332732200623,
      "num_tokens": 56215875.0,
      "step": 25975
    },
    {
      "entropy": 6.497547626495361,
      "epoch": 2.7801380491198033,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042367927040968514,
      "loss": 5.8643,
      "mean_token_accuracy": 0.16200816929340361,
      "num_tokens": 56227633.0,
      "step": 25980
    },
    {
      "entropy": 6.334244346618652,
      "epoch": 2.780673123227567,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004236505716463669,
      "loss": 5.7263,
      "mean_token_accuracy": 0.1779629722237587,
      "num_tokens": 56239420.0,
      "step": 25985
    },
    {
      "entropy": 6.434944438934326,
      "epoch": 2.781208197335331,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004236218685907844,
      "loss": 5.7834,
      "mean_token_accuracy": 0.16941509395837784,
      "num_tokens": 56250538.0,
      "step": 25990
    },
    {
      "entropy": 6.452974510192871,
      "epoch": 2.7817432714430947,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004235931612437663,
      "loss": 5.8611,
      "mean_token_accuracy": 0.16806139200925826,
      "num_tokens": 56261423.0,
      "step": 25995
    },
    {
      "entropy": 6.449911308288574,
      "epoch": 2.782278345550859,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004235644496061416,
      "loss": 5.8629,
      "mean_token_accuracy": 0.16811030507087707,
      "num_tokens": 56272581.0,
      "step": 26000
    },
    {
      "entropy": 6.445717811584473,
      "epoch": 2.7828134196586225,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004235357336787392,
      "loss": 5.7911,
      "mean_token_accuracy": 0.17547626197338104,
      "num_tokens": 56283703.0,
      "step": 26005
    },
    {
      "entropy": 6.5109014987945555,
      "epoch": 2.7833484937663866,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042350701346238853,
      "loss": 5.8737,
      "mean_token_accuracy": 0.17125126719474792,
      "num_tokens": 56294657.0,
      "step": 26010
    },
    {
      "entropy": 6.435869359970093,
      "epoch": 2.7838835678741507,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004234782889579185,
      "loss": 5.7516,
      "mean_token_accuracy": 0.17440682649612427,
      "num_tokens": 56305555.0,
      "step": 26015
    },
    {
      "entropy": 6.499895000457764,
      "epoch": 2.7844186419819144,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042344956016615885,
      "loss": 5.8724,
      "mean_token_accuracy": 0.17056342214345932,
      "num_tokens": 56316716.0,
      "step": 26020
    },
    {
      "entropy": 6.4929300308227536,
      "epoch": 2.7849537160896785,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004234208270879389,
      "loss": 5.7536,
      "mean_token_accuracy": 0.16976161748170854,
      "num_tokens": 56326582.0,
      "step": 26025
    },
    {
      "entropy": 6.4922693252563475,
      "epoch": 2.7854887901974426,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042339208972408865,
      "loss": 5.8562,
      "mean_token_accuracy": 0.1758776366710663,
      "num_tokens": 56336687.0,
      "step": 26030
    },
    {
      "entropy": 6.355084609985352,
      "epoch": 2.786023864305206,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004233633480754375,
      "loss": 5.6551,
      "mean_token_accuracy": 0.185956272482872,
      "num_tokens": 56347493.0,
      "step": 26035
    },
    {
      "entropy": 6.432875347137451,
      "epoch": 2.7865589384129703,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042333460214281557,
      "loss": 5.778,
      "mean_token_accuracy": 0.17220503389835357,
      "num_tokens": 56359259.0,
      "step": 26040
    },
    {
      "entropy": 6.471282863616944,
      "epoch": 2.787094012520734,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042330585192705284,
      "loss": 5.8081,
      "mean_token_accuracy": 0.1641317918896675,
      "num_tokens": 56370502.0,
      "step": 26045
    },
    {
      "entropy": 6.436733722686768,
      "epoch": 2.787629086628498,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042327709742897944,
      "loss": 5.8368,
      "mean_token_accuracy": 0.16978551000356673,
      "num_tokens": 56381047.0,
      "step": 26050
    },
    {
      "entropy": 6.407622528076172,
      "epoch": 2.7881641607362617,
      "grad_norm": 1.125,
      "learning_rate": 0.00042324833864942576,
      "loss": 5.748,
      "mean_token_accuracy": 0.1757706433534622,
      "num_tokens": 56392304.0,
      "step": 26055
    },
    {
      "entropy": 6.443569898605347,
      "epoch": 2.788699234844026,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004232195755892221,
      "loss": 5.756,
      "mean_token_accuracy": 0.178072190284729,
      "num_tokens": 56402480.0,
      "step": 26060
    },
    {
      "entropy": 6.441643571853637,
      "epoch": 2.78923430895179,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042319080824919895,
      "loss": 5.8243,
      "mean_token_accuracy": 0.17225735485553742,
      "num_tokens": 56413659.0,
      "step": 26065
    },
    {
      "entropy": 6.4760956287384035,
      "epoch": 2.7897693830595536,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004231620366301871,
      "loss": 5.8815,
      "mean_token_accuracy": 0.16634269803762436,
      "num_tokens": 56425272.0,
      "step": 26070
    },
    {
      "entropy": 6.5351568222045895,
      "epoch": 2.7903044571673177,
      "grad_norm": 1.125,
      "learning_rate": 0.00042313326073301733,
      "loss": 5.8364,
      "mean_token_accuracy": 0.1680128738284111,
      "num_tokens": 56436696.0,
      "step": 26075
    },
    {
      "entropy": 6.514017581939697,
      "epoch": 2.7908395312750818,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004231044805585204,
      "loss": 5.883,
      "mean_token_accuracy": 0.16779196560382842,
      "num_tokens": 56446666.0,
      "step": 26080
    },
    {
      "entropy": 6.415836429595947,
      "epoch": 2.7913746053828454,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004230756961075274,
      "loss": 5.8337,
      "mean_token_accuracy": 0.17225100100040436,
      "num_tokens": 56457082.0,
      "step": 26085
    },
    {
      "entropy": 6.418101024627686,
      "epoch": 2.7919096794906095,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042304690738086944,
      "loss": 5.7683,
      "mean_token_accuracy": 0.17269551008939743,
      "num_tokens": 56468492.0,
      "step": 26090
    },
    {
      "entropy": 6.490676498413086,
      "epoch": 2.7924447535983736,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004230181143793779,
      "loss": 5.8479,
      "mean_token_accuracy": 0.16599166989326478,
      "num_tokens": 56479353.0,
      "step": 26095
    },
    {
      "entropy": 6.48487024307251,
      "epoch": 2.7929798277061373,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042298931710388414,
      "loss": 5.844,
      "mean_token_accuracy": 0.1719767525792122,
      "num_tokens": 56490105.0,
      "step": 26100
    },
    {
      "entropy": 6.516659069061279,
      "epoch": 2.7935149018139014,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004229605155552195,
      "loss": 5.8672,
      "mean_token_accuracy": 0.16689072996377946,
      "num_tokens": 56501528.0,
      "step": 26105
    },
    {
      "entropy": 6.52530837059021,
      "epoch": 2.794049975921665,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042293170973421593,
      "loss": 5.9336,
      "mean_token_accuracy": 0.1608760878443718,
      "num_tokens": 56512952.0,
      "step": 26110
    },
    {
      "entropy": 6.500452089309692,
      "epoch": 2.794585050029429,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042290289964170495,
      "loss": 5.8211,
      "mean_token_accuracy": 0.17102095037698745,
      "num_tokens": 56523332.0,
      "step": 26115
    },
    {
      "entropy": 6.481412029266357,
      "epoch": 2.795120124137193,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004228740852785185,
      "loss": 5.8209,
      "mean_token_accuracy": 0.17031925916671753,
      "num_tokens": 56534777.0,
      "step": 26120
    },
    {
      "entropy": 6.467931127548217,
      "epoch": 2.795655198244957,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042284526664548866,
      "loss": 5.8231,
      "mean_token_accuracy": 0.1652720734477043,
      "num_tokens": 56545706.0,
      "step": 26125
    },
    {
      "entropy": 6.4550035953521725,
      "epoch": 2.796190272352721,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042281644374344754,
      "loss": 5.819,
      "mean_token_accuracy": 0.1647439032793045,
      "num_tokens": 56557014.0,
      "step": 26130
    },
    {
      "entropy": 6.464187669754028,
      "epoch": 2.7967253464604847,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042278761657322736,
      "loss": 5.8494,
      "mean_token_accuracy": 0.1702168881893158,
      "num_tokens": 56569707.0,
      "step": 26135
    },
    {
      "entropy": 6.476227045059204,
      "epoch": 2.7972604205682488,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004227587851356605,
      "loss": 5.7837,
      "mean_token_accuracy": 0.17326206415891648,
      "num_tokens": 56579924.0,
      "step": 26140
    },
    {
      "entropy": 6.500516605377197,
      "epoch": 2.797795494676013,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004227299494315795,
      "loss": 5.8601,
      "mean_token_accuracy": 0.16955725848674774,
      "num_tokens": 56590301.0,
      "step": 26145
    },
    {
      "entropy": 6.481996488571167,
      "epoch": 2.7983305687837765,
      "grad_norm": 1.125,
      "learning_rate": 0.00042270110946181693,
      "loss": 5.8095,
      "mean_token_accuracy": 0.16885800957679747,
      "num_tokens": 56600717.0,
      "step": 26150
    },
    {
      "entropy": 6.538051891326904,
      "epoch": 2.7988656428915406,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042267226522720556,
      "loss": 5.8438,
      "mean_token_accuracy": 0.16856716573238373,
      "num_tokens": 56611547.0,
      "step": 26155
    },
    {
      "entropy": 6.47439227104187,
      "epoch": 2.7994007169993043,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004226434167285784,
      "loss": 5.7518,
      "mean_token_accuracy": 0.17362110167741776,
      "num_tokens": 56622542.0,
      "step": 26160
    },
    {
      "entropy": 6.463205480575562,
      "epoch": 2.7999357911070684,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004226145639667683,
      "loss": 5.8056,
      "mean_token_accuracy": 0.17939109951257706,
      "num_tokens": 56633359.0,
      "step": 26165
    },
    {
      "entropy": 6.40850338935852,
      "epoch": 2.800470865214832,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004225857069426083,
      "loss": 5.693,
      "mean_token_accuracy": 0.1847177267074585,
      "num_tokens": 56644101.0,
      "step": 26170
    },
    {
      "entropy": 6.46279878616333,
      "epoch": 2.801005939322596,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042255684565693183,
      "loss": 5.7617,
      "mean_token_accuracy": 0.16719890534877777,
      "num_tokens": 56654630.0,
      "step": 26175
    },
    {
      "entropy": 6.45663480758667,
      "epoch": 2.80154101343036,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004225279801105722,
      "loss": 5.7828,
      "mean_token_accuracy": 0.1720054566860199,
      "num_tokens": 56665826.0,
      "step": 26180
    },
    {
      "entropy": 6.453690767288208,
      "epoch": 2.802076087538124,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004224991103043628,
      "loss": 5.8276,
      "mean_token_accuracy": 0.17133368998765947,
      "num_tokens": 56676431.0,
      "step": 26185
    },
    {
      "entropy": 6.496699285507202,
      "epoch": 2.802611161645888,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042247023623913745,
      "loss": 5.8112,
      "mean_token_accuracy": 0.17278799116611482,
      "num_tokens": 56687297.0,
      "step": 26190
    },
    {
      "entropy": 6.446616458892822,
      "epoch": 2.803146235753652,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004224413579157296,
      "loss": 5.8507,
      "mean_token_accuracy": 0.17175252437591554,
      "num_tokens": 56698392.0,
      "step": 26195
    },
    {
      "entropy": 6.479364538192749,
      "epoch": 2.8036813098614157,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004224124753349733,
      "loss": 5.8626,
      "mean_token_accuracy": 0.17010470628738403,
      "num_tokens": 56710588.0,
      "step": 26200
    },
    {
      "entropy": 6.391730451583863,
      "epoch": 2.80421638396918,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004223835884977025,
      "loss": 5.7041,
      "mean_token_accuracy": 0.17623843997716904,
      "num_tokens": 56723143.0,
      "step": 26205
    },
    {
      "entropy": 6.494681119918823,
      "epoch": 2.804751458076944,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004223546974047513,
      "loss": 5.7816,
      "mean_token_accuracy": 0.17336915135383607,
      "num_tokens": 56734097.0,
      "step": 26210
    },
    {
      "entropy": 6.521607971191406,
      "epoch": 2.8052865321847076,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004223258020569539,
      "loss": 5.8371,
      "mean_token_accuracy": 0.16809049546718596,
      "num_tokens": 56745311.0,
      "step": 26215
    },
    {
      "entropy": 6.542292308807373,
      "epoch": 2.8058216062924712,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004222969024551447,
      "loss": 5.9206,
      "mean_token_accuracy": 0.17003602981567384,
      "num_tokens": 56755495.0,
      "step": 26220
    },
    {
      "entropy": 6.465632104873658,
      "epoch": 2.8063566804002353,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042226799860015817,
      "loss": 5.7945,
      "mean_token_accuracy": 0.17656239122152328,
      "num_tokens": 56766107.0,
      "step": 26225
    },
    {
      "entropy": 6.4124387264251705,
      "epoch": 2.8068917545079994,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004222390904928288,
      "loss": 5.7723,
      "mean_token_accuracy": 0.17852274775505067,
      "num_tokens": 56776877.0,
      "step": 26230
    },
    {
      "entropy": 6.479631137847901,
      "epoch": 2.807426828615763,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042221017813399146,
      "loss": 5.8645,
      "mean_token_accuracy": 0.17444402873516082,
      "num_tokens": 56786439.0,
      "step": 26235
    },
    {
      "entropy": 6.449943208694458,
      "epoch": 2.807961902723527,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004221812615244809,
      "loss": 5.8704,
      "mean_token_accuracy": 0.16309298872947692,
      "num_tokens": 56797433.0,
      "step": 26240
    },
    {
      "entropy": 6.554708099365234,
      "epoch": 2.8084969768312913,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042215234066513203,
      "loss": 5.8938,
      "mean_token_accuracy": 0.16272909492254256,
      "num_tokens": 56808270.0,
      "step": 26245
    },
    {
      "entropy": 6.553452968597412,
      "epoch": 2.809032050939055,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042212341555678,
      "loss": 5.832,
      "mean_token_accuracy": 0.16483175307512282,
      "num_tokens": 56818590.0,
      "step": 26250
    },
    {
      "entropy": 6.4799981117248535,
      "epoch": 2.809567125046819,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004220944862002601,
      "loss": 5.8641,
      "mean_token_accuracy": 0.1636879786849022,
      "num_tokens": 56828799.0,
      "step": 26255
    },
    {
      "entropy": 6.4302849769592285,
      "epoch": 2.810102199154583,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042206555259640744,
      "loss": 5.8619,
      "mean_token_accuracy": 0.16516897231340408,
      "num_tokens": 56839409.0,
      "step": 26260
    },
    {
      "entropy": 6.416729307174682,
      "epoch": 2.810637273262347,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042203661474605775,
      "loss": 5.7657,
      "mean_token_accuracy": 0.17169727981090546,
      "num_tokens": 56849650.0,
      "step": 26265
    },
    {
      "entropy": 6.454157447814941,
      "epoch": 2.811172347370111,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004220076726500464,
      "loss": 5.8546,
      "mean_token_accuracy": 0.16697733104228973,
      "num_tokens": 56860437.0,
      "step": 26270
    },
    {
      "entropy": 6.468470048904419,
      "epoch": 2.8117074214778746,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042197872630920927,
      "loss": 5.8369,
      "mean_token_accuracy": 0.17603850364685059,
      "num_tokens": 56870314.0,
      "step": 26275
    },
    {
      "entropy": 6.503255033493042,
      "epoch": 2.8122424955856387,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042194977572438194,
      "loss": 5.8411,
      "mean_token_accuracy": 0.17215453386306762,
      "num_tokens": 56881675.0,
      "step": 26280
    },
    {
      "entropy": 6.499451589584351,
      "epoch": 2.8127775696934023,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042192082089640055,
      "loss": 5.9118,
      "mean_token_accuracy": 0.1609245166182518,
      "num_tokens": 56893148.0,
      "step": 26285
    },
    {
      "entropy": 6.48630576133728,
      "epoch": 2.8133126438011664,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004218918618261011,
      "loss": 5.7626,
      "mean_token_accuracy": 0.17753468900918962,
      "num_tokens": 56904008.0,
      "step": 26290
    },
    {
      "entropy": 6.465262985229492,
      "epoch": 2.8138477179089305,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042186289851431976,
      "loss": 5.7989,
      "mean_token_accuracy": 0.17429958283901215,
      "num_tokens": 56914731.0,
      "step": 26295
    },
    {
      "entropy": 6.5429328918457035,
      "epoch": 2.814382792016694,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042183393096189285,
      "loss": 5.9304,
      "mean_token_accuracy": 0.16464067697525026,
      "num_tokens": 56926037.0,
      "step": 26300
    },
    {
      "entropy": 6.406075859069825,
      "epoch": 2.8149178661244583,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042180495916965686,
      "loss": 5.7363,
      "mean_token_accuracy": 0.17645036578178405,
      "num_tokens": 56936456.0,
      "step": 26305
    },
    {
      "entropy": 6.433007717132568,
      "epoch": 2.8154529402322224,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004217759831384483,
      "loss": 5.7644,
      "mean_token_accuracy": 0.1767757683992386,
      "num_tokens": 56947929.0,
      "step": 26310
    },
    {
      "entropy": 6.4891478538513185,
      "epoch": 2.815988014339986,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042174700286910376,
      "loss": 5.8036,
      "mean_token_accuracy": 0.17371415197849274,
      "num_tokens": 56958153.0,
      "step": 26315
    },
    {
      "entropy": 6.509412527084351,
      "epoch": 2.81652308844775,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004217180183624602,
      "loss": 5.8657,
      "mean_token_accuracy": 0.16785891652107238,
      "num_tokens": 56968609.0,
      "step": 26320
    },
    {
      "entropy": 6.44675235748291,
      "epoch": 2.817058162555514,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004216890296193545,
      "loss": 5.856,
      "mean_token_accuracy": 0.17289304435253144,
      "num_tokens": 56978434.0,
      "step": 26325
    },
    {
      "entropy": 6.467288827896118,
      "epoch": 2.817593236663278,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004216600366406236,
      "loss": 5.818,
      "mean_token_accuracy": 0.17440975904464723,
      "num_tokens": 56988826.0,
      "step": 26330
    },
    {
      "entropy": 6.452823495864868,
      "epoch": 2.8181283107710415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042163103942710487,
      "loss": 5.8454,
      "mean_token_accuracy": 0.1719656154513359,
      "num_tokens": 57000385.0,
      "step": 26335
    },
    {
      "entropy": 6.400611734390258,
      "epoch": 2.8186633848788056,
      "grad_norm": 1.75,
      "learning_rate": 0.0004216020379796354,
      "loss": 5.7998,
      "mean_token_accuracy": 0.18056640625,
      "num_tokens": 57012137.0,
      "step": 26340
    },
    {
      "entropy": 6.508644437789917,
      "epoch": 2.8191984589865697,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042157303229905277,
      "loss": 5.8883,
      "mean_token_accuracy": 0.16500847935676574,
      "num_tokens": 57022533.0,
      "step": 26345
    },
    {
      "entropy": 6.485795497894287,
      "epoch": 2.8197335330943334,
      "grad_norm": 1.25,
      "learning_rate": 0.00042154402238619434,
      "loss": 5.8397,
      "mean_token_accuracy": 0.16529279053211213,
      "num_tokens": 57034438.0,
      "step": 26350
    },
    {
      "entropy": 6.497997426986695,
      "epoch": 2.8202686072020975,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042151500824189785,
      "loss": 5.8873,
      "mean_token_accuracy": 0.16473207622766495,
      "num_tokens": 57046343.0,
      "step": 26355
    },
    {
      "entropy": 6.512043046951294,
      "epoch": 2.8208036813098616,
      "grad_norm": 0.984375,
      "learning_rate": 0.00042148598986700117,
      "loss": 5.8613,
      "mean_token_accuracy": 0.16752277314662933,
      "num_tokens": 57058269.0,
      "step": 26360
    },
    {
      "entropy": 6.481241130828858,
      "epoch": 2.8213387554176252,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042145696726234207,
      "loss": 5.8061,
      "mean_token_accuracy": 0.17570967078208924,
      "num_tokens": 57069192.0,
      "step": 26365
    },
    {
      "entropy": 6.5066690921783445,
      "epoch": 2.8218738295253893,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004214279404287586,
      "loss": 5.8596,
      "mean_token_accuracy": 0.1708306446671486,
      "num_tokens": 57080021.0,
      "step": 26370
    },
    {
      "entropy": 6.44480299949646,
      "epoch": 2.8224089036331534,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004213989093670889,
      "loss": 5.7884,
      "mean_token_accuracy": 0.17007977217435838,
      "num_tokens": 57090336.0,
      "step": 26375
    },
    {
      "entropy": 6.450229692459106,
      "epoch": 2.822943977740917,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004213698740781713,
      "loss": 5.8103,
      "mean_token_accuracy": 0.17503671497106552,
      "num_tokens": 57101942.0,
      "step": 26380
    },
    {
      "entropy": 6.453200626373291,
      "epoch": 2.823479051848681,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004213408345628442,
      "loss": 5.766,
      "mean_token_accuracy": 0.17287904173135757,
      "num_tokens": 57112353.0,
      "step": 26385
    },
    {
      "entropy": 6.357598114013672,
      "epoch": 2.824014125956445,
      "grad_norm": 1.5,
      "learning_rate": 0.000421311790821946,
      "loss": 5.687,
      "mean_token_accuracy": 0.19179102331399916,
      "num_tokens": 57123796.0,
      "step": 26390
    },
    {
      "entropy": 6.460034942626953,
      "epoch": 2.824549200064209,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042128274285631545,
      "loss": 5.8154,
      "mean_token_accuracy": 0.1691857635974884,
      "num_tokens": 57134463.0,
      "step": 26395
    },
    {
      "entropy": 6.443429660797119,
      "epoch": 2.8250842741719726,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042125369066679124,
      "loss": 5.7547,
      "mean_token_accuracy": 0.1730487585067749,
      "num_tokens": 57146964.0,
      "step": 26400
    },
    {
      "entropy": 6.493409156799316,
      "epoch": 2.8256193482797367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042122463425421224,
      "loss": 5.8062,
      "mean_token_accuracy": 0.16484901905059815,
      "num_tokens": 57157322.0,
      "step": 26405
    },
    {
      "entropy": 6.404153299331665,
      "epoch": 2.826154422387501,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004211955736194175,
      "loss": 5.7438,
      "mean_token_accuracy": 0.1791894793510437,
      "num_tokens": 57168758.0,
      "step": 26410
    },
    {
      "entropy": 6.457229661941528,
      "epoch": 2.8266894964952645,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042116650876324616,
      "loss": 5.7919,
      "mean_token_accuracy": 0.1744456559419632,
      "num_tokens": 57180322.0,
      "step": 26415
    },
    {
      "entropy": 6.500288200378418,
      "epoch": 2.8272245706030286,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004211374396865373,
      "loss": 5.8127,
      "mean_token_accuracy": 0.16510481983423234,
      "num_tokens": 57190836.0,
      "step": 26420
    },
    {
      "entropy": 6.419603633880615,
      "epoch": 2.8277596447107927,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004211083663901305,
      "loss": 5.7833,
      "mean_token_accuracy": 0.17021819949150085,
      "num_tokens": 57202439.0,
      "step": 26425
    },
    {
      "entropy": 6.534043693542481,
      "epoch": 2.8282947188185563,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004210792888748651,
      "loss": 5.9459,
      "mean_token_accuracy": 0.159040467441082,
      "num_tokens": 57213335.0,
      "step": 26430
    },
    {
      "entropy": 6.522109270095825,
      "epoch": 2.8288297929263204,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042105020714158076,
      "loss": 5.8536,
      "mean_token_accuracy": 0.1677111342549324,
      "num_tokens": 57223840.0,
      "step": 26435
    },
    {
      "entropy": 6.477433395385742,
      "epoch": 2.829364867034084,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042102112119111725,
      "loss": 5.8409,
      "mean_token_accuracy": 0.1655831128358841,
      "num_tokens": 57234702.0,
      "step": 26440
    },
    {
      "entropy": 6.368810319900513,
      "epoch": 2.829899941141848,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004209920310243144,
      "loss": 5.7581,
      "mean_token_accuracy": 0.17686099857091903,
      "num_tokens": 57247415.0,
      "step": 26445
    },
    {
      "entropy": 6.415661811828613,
      "epoch": 2.830435015249612,
      "grad_norm": 1.25,
      "learning_rate": 0.0004209629366420121,
      "loss": 5.7649,
      "mean_token_accuracy": 0.1737866073846817,
      "num_tokens": 57257768.0,
      "step": 26450
    },
    {
      "entropy": 6.550294399261475,
      "epoch": 2.830970089357376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004209338380450506,
      "loss": 5.954,
      "mean_token_accuracy": 0.1642732009291649,
      "num_tokens": 57268114.0,
      "step": 26455
    },
    {
      "entropy": 6.530879974365234,
      "epoch": 2.83150516346514,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042090473523426994,
      "loss": 5.8996,
      "mean_token_accuracy": 0.16444089263677597,
      "num_tokens": 57278680.0,
      "step": 26460
    },
    {
      "entropy": 6.5446939945220945,
      "epoch": 2.8320402375729037,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042087562821051055,
      "loss": 5.884,
      "mean_token_accuracy": 0.16575006097555162,
      "num_tokens": 57289823.0,
      "step": 26465
    },
    {
      "entropy": 6.4836217880249025,
      "epoch": 2.832575311680668,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042084651697461297,
      "loss": 5.8104,
      "mean_token_accuracy": 0.16465361565351486,
      "num_tokens": 57300137.0,
      "step": 26470
    },
    {
      "entropy": 6.438157749176026,
      "epoch": 2.833110385788432,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042081740152741763,
      "loss": 5.748,
      "mean_token_accuracy": 0.17575358897447585,
      "num_tokens": 57311160.0,
      "step": 26475
    },
    {
      "entropy": 6.432830333709717,
      "epoch": 2.8336454598961955,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004207882818697654,
      "loss": 5.8135,
      "mean_token_accuracy": 0.17397969514131545,
      "num_tokens": 57322708.0,
      "step": 26480
    },
    {
      "entropy": 6.439680910110473,
      "epoch": 2.8341805340039596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000420759158002497,
      "loss": 5.7766,
      "mean_token_accuracy": 0.17299417704343795,
      "num_tokens": 57333788.0,
      "step": 26485
    },
    {
      "entropy": 6.503746366500854,
      "epoch": 2.8347156081117237,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042073002992645337,
      "loss": 5.8219,
      "mean_token_accuracy": 0.17127186805009842,
      "num_tokens": 57344806.0,
      "step": 26490
    },
    {
      "entropy": 6.489925146102905,
      "epoch": 2.8352506822194874,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004207008976424756,
      "loss": 5.8799,
      "mean_token_accuracy": 0.1673963710665703,
      "num_tokens": 57355086.0,
      "step": 26495
    },
    {
      "entropy": 6.40111141204834,
      "epoch": 2.835785756327251,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004206717611514049,
      "loss": 5.7141,
      "mean_token_accuracy": 0.18211520314216614,
      "num_tokens": 57364837.0,
      "step": 26500
    },
    {
      "entropy": 6.4496077537536625,
      "epoch": 2.836320830435015,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004206426204540825,
      "loss": 5.833,
      "mean_token_accuracy": 0.16985072195529938,
      "num_tokens": 57376049.0,
      "step": 26505
    },
    {
      "entropy": 6.495711088180542,
      "epoch": 2.8368559045427792,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042061347555135,
      "loss": 5.8484,
      "mean_token_accuracy": 0.1634271189570427,
      "num_tokens": 57387829.0,
      "step": 26510
    },
    {
      "entropy": 6.5073769092559814,
      "epoch": 2.837390978650543,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004205843264440488,
      "loss": 5.812,
      "mean_token_accuracy": 0.17338123470544814,
      "num_tokens": 57397165.0,
      "step": 26515
    },
    {
      "entropy": 6.451508092880249,
      "epoch": 2.837926052758307,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042055517313302064,
      "loss": 5.7908,
      "mean_token_accuracy": 0.1734862133860588,
      "num_tokens": 57408602.0,
      "step": 26520
    },
    {
      "entropy": 6.4287886142730715,
      "epoch": 2.838461126866071,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004205260156191073,
      "loss": 5.7313,
      "mean_token_accuracy": 0.17845809310674668,
      "num_tokens": 57419393.0,
      "step": 26525
    },
    {
      "entropy": 6.443080759048462,
      "epoch": 2.8389962009738348,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042049685390315074,
      "loss": 5.8367,
      "mean_token_accuracy": 0.16856845319271088,
      "num_tokens": 57431238.0,
      "step": 26530
    },
    {
      "entropy": 6.485386037826538,
      "epoch": 2.839531275081599,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00042046768798599293,
      "loss": 5.7776,
      "mean_token_accuracy": 0.17763406187295913,
      "num_tokens": 57442080.0,
      "step": 26535
    },
    {
      "entropy": 6.475267362594605,
      "epoch": 2.840066349189363,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004204385178684761,
      "loss": 5.8157,
      "mean_token_accuracy": 0.16742378622293472,
      "num_tokens": 57452539.0,
      "step": 26540
    },
    {
      "entropy": 6.497808599472046,
      "epoch": 2.8406014232971266,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042040934355144245,
      "loss": 5.926,
      "mean_token_accuracy": 0.1630005270242691,
      "num_tokens": 57464936.0,
      "step": 26545
    },
    {
      "entropy": 6.465134525299073,
      "epoch": 2.8411364974048907,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004203801650357346,
      "loss": 5.7961,
      "mean_token_accuracy": 0.16723983883857726,
      "num_tokens": 57477460.0,
      "step": 26550
    },
    {
      "entropy": 6.483828449249268,
      "epoch": 2.8416715715126544,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004203509823221947,
      "loss": 5.8373,
      "mean_token_accuracy": 0.17065078765153885,
      "num_tokens": 57487707.0,
      "step": 26555
    },
    {
      "entropy": 6.45072832107544,
      "epoch": 2.8422066456204185,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004203217954116657,
      "loss": 5.8769,
      "mean_token_accuracy": 0.16688260734081267,
      "num_tokens": 57498107.0,
      "step": 26560
    },
    {
      "entropy": 6.464191341400147,
      "epoch": 2.842741719728182,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042029260430499033,
      "loss": 5.7823,
      "mean_token_accuracy": 0.16672473102808,
      "num_tokens": 57509035.0,
      "step": 26565
    },
    {
      "entropy": 6.497441625595092,
      "epoch": 2.8432767938359462,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00042026340900301135,
      "loss": 5.8701,
      "mean_token_accuracy": 0.16703203767538072,
      "num_tokens": 57521234.0,
      "step": 26570
    },
    {
      "entropy": 6.471724510192871,
      "epoch": 2.8438118679437103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004202342095065719,
      "loss": 5.7996,
      "mean_token_accuracy": 0.17439806908369065,
      "num_tokens": 57532294.0,
      "step": 26575
    },
    {
      "entropy": 6.489065170288086,
      "epoch": 2.844346942051474,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000420205005816515,
      "loss": 5.818,
      "mean_token_accuracy": 0.17718469649553298,
      "num_tokens": 57542047.0,
      "step": 26580
    },
    {
      "entropy": 6.415178298950195,
      "epoch": 2.844882016159238,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000420175797933684,
      "loss": 5.8181,
      "mean_token_accuracy": 0.1711246207356453,
      "num_tokens": 57553254.0,
      "step": 26585
    },
    {
      "entropy": 6.479855298995972,
      "epoch": 2.845417090267002,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042014658585892223,
      "loss": 5.7682,
      "mean_token_accuracy": 0.18370454311370848,
      "num_tokens": 57564321.0,
      "step": 26590
    },
    {
      "entropy": 6.424128627777099,
      "epoch": 2.845952164374766,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042011736959307323,
      "loss": 5.7343,
      "mean_token_accuracy": 0.17164987921714783,
      "num_tokens": 57575018.0,
      "step": 26595
    },
    {
      "entropy": 6.405411434173584,
      "epoch": 2.84648723848253,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042008814913698054,
      "loss": 5.7953,
      "mean_token_accuracy": 0.17005515545606614,
      "num_tokens": 57585698.0,
      "step": 26600
    },
    {
      "entropy": 6.457719135284424,
      "epoch": 2.8470223125902936,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004200589244914879,
      "loss": 5.7798,
      "mean_token_accuracy": 0.17542052567005156,
      "num_tokens": 57595363.0,
      "step": 26605
    },
    {
      "entropy": 6.4409153938293455,
      "epoch": 2.8475573866980577,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042002969565743925,
      "loss": 5.7861,
      "mean_token_accuracy": 0.1807983100414276,
      "num_tokens": 57605908.0,
      "step": 26610
    },
    {
      "entropy": 6.384437322616577,
      "epoch": 2.8480924608058213,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004200004626356785,
      "loss": 5.7344,
      "mean_token_accuracy": 0.1803414598107338,
      "num_tokens": 57616172.0,
      "step": 26615
    },
    {
      "entropy": 6.4571808815002445,
      "epoch": 2.8486275349135854,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041997122542704974,
      "loss": 5.7877,
      "mean_token_accuracy": 0.17260331362485887,
      "num_tokens": 57626770.0,
      "step": 26620
    },
    {
      "entropy": 6.420671367645264,
      "epoch": 2.8491626090213495,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004199419840323972,
      "loss": 5.7143,
      "mean_token_accuracy": 0.18028916269540787,
      "num_tokens": 57636576.0,
      "step": 26625
    },
    {
      "entropy": 6.382935285568237,
      "epoch": 2.849697683129113,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041991273845256536,
      "loss": 5.8036,
      "mean_token_accuracy": 0.1740951806306839,
      "num_tokens": 57646431.0,
      "step": 26630
    },
    {
      "entropy": 6.395848512649536,
      "epoch": 2.8502327572368773,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041988348868839836,
      "loss": 5.794,
      "mean_token_accuracy": 0.1750243902206421,
      "num_tokens": 57657412.0,
      "step": 26635
    },
    {
      "entropy": 6.527501392364502,
      "epoch": 2.8507678313446414,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004198542347407412,
      "loss": 5.8248,
      "mean_token_accuracy": 0.17517351508140563,
      "num_tokens": 57667434.0,
      "step": 26640
    },
    {
      "entropy": 6.465921306610108,
      "epoch": 2.851302905452405,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004198249766104382,
      "loss": 5.8311,
      "mean_token_accuracy": 0.17623837888240815,
      "num_tokens": 57677640.0,
      "step": 26645
    },
    {
      "entropy": 6.448953485488891,
      "epoch": 2.851837979560169,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004197957142983344,
      "loss": 5.8447,
      "mean_token_accuracy": 0.1700751319527626,
      "num_tokens": 57687909.0,
      "step": 26650
    },
    {
      "entropy": 6.519346284866333,
      "epoch": 2.8523730536679333,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004197664478052747,
      "loss": 5.9333,
      "mean_token_accuracy": 0.1589268907904625,
      "num_tokens": 57700374.0,
      "step": 26655
    },
    {
      "entropy": 6.54340443611145,
      "epoch": 2.852908127775697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004197371771321042,
      "loss": 5.8125,
      "mean_token_accuracy": 0.17370715290307998,
      "num_tokens": 57711577.0,
      "step": 26660
    },
    {
      "entropy": 6.4163007736206055,
      "epoch": 2.853443201883461,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041970790227966804,
      "loss": 5.7478,
      "mean_token_accuracy": 0.17904950827360153,
      "num_tokens": 57721259.0,
      "step": 26665
    },
    {
      "entropy": 6.386013507843018,
      "epoch": 2.8539782759912247,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041967862324881154,
      "loss": 5.7867,
      "mean_token_accuracy": 0.17278318703174592,
      "num_tokens": 57731499.0,
      "step": 26670
    },
    {
      "entropy": 6.504896593093872,
      "epoch": 2.8545133500989888,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004196493400403801,
      "loss": 5.9015,
      "mean_token_accuracy": 0.1612050399184227,
      "num_tokens": 57742343.0,
      "step": 26675
    },
    {
      "entropy": 6.482236337661743,
      "epoch": 2.8550484242067524,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004196200526552193,
      "loss": 5.8213,
      "mean_token_accuracy": 0.17872020453214646,
      "num_tokens": 57752942.0,
      "step": 26680
    },
    {
      "entropy": 6.497006464004516,
      "epoch": 2.8555834983145165,
      "grad_norm": 1.125,
      "learning_rate": 0.0004195907610941748,
      "loss": 5.8297,
      "mean_token_accuracy": 0.16921166181564332,
      "num_tokens": 57763623.0,
      "step": 26685
    },
    {
      "entropy": 6.419076633453369,
      "epoch": 2.8561185724222806,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004195614653580924,
      "loss": 5.8252,
      "mean_token_accuracy": 0.1702154517173767,
      "num_tokens": 57774538.0,
      "step": 26690
    },
    {
      "entropy": 6.453876399993897,
      "epoch": 2.8566536465300443,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041953216544781796,
      "loss": 5.8213,
      "mean_token_accuracy": 0.1706807166337967,
      "num_tokens": 57785420.0,
      "step": 26695
    },
    {
      "entropy": 6.46655011177063,
      "epoch": 2.8571887206378084,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041950286136419756,
      "loss": 5.8097,
      "mean_token_accuracy": 0.16172026395797728,
      "num_tokens": 57796488.0,
      "step": 26700
    },
    {
      "entropy": 6.485834360122681,
      "epoch": 2.8577237947455725,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041947355310807737,
      "loss": 5.7981,
      "mean_token_accuracy": 0.16996438652276993,
      "num_tokens": 57806391.0,
      "step": 26705
    },
    {
      "entropy": 6.453691291809082,
      "epoch": 2.858258868853336,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041944424068030364,
      "loss": 5.8223,
      "mean_token_accuracy": 0.1683011069893837,
      "num_tokens": 57817410.0,
      "step": 26710
    },
    {
      "entropy": 6.4903747081756595,
      "epoch": 2.8587939429611002,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041941492408172277,
      "loss": 5.824,
      "mean_token_accuracy": 0.16963643878698348,
      "num_tokens": 57828582.0,
      "step": 26715
    },
    {
      "entropy": 6.403255748748779,
      "epoch": 2.859329017068864,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041938560331318124,
      "loss": 5.7716,
      "mean_token_accuracy": 0.1789974495768547,
      "num_tokens": 57838962.0,
      "step": 26720
    },
    {
      "entropy": 6.5032611846923825,
      "epoch": 2.859864091176628,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004193562783755257,
      "loss": 5.8138,
      "mean_token_accuracy": 0.1751888856291771,
      "num_tokens": 57849335.0,
      "step": 26725
    },
    {
      "entropy": 6.4968232154846195,
      "epoch": 2.8603991652843916,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004193269492696029,
      "loss": 5.8214,
      "mean_token_accuracy": 0.16886212825775146,
      "num_tokens": 57860904.0,
      "step": 26730
    },
    {
      "entropy": 6.480409955978393,
      "epoch": 2.8609342393921557,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004192976159962597,
      "loss": 5.8262,
      "mean_token_accuracy": 0.16820450872182846,
      "num_tokens": 57870861.0,
      "step": 26735
    },
    {
      "entropy": 6.455139255523681,
      "epoch": 2.86146931349992,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004192682785563432,
      "loss": 5.8535,
      "mean_token_accuracy": 0.16414737999439238,
      "num_tokens": 57881590.0,
      "step": 26740
    },
    {
      "entropy": 6.453397130966186,
      "epoch": 2.8620043876076835,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004192389369507004,
      "loss": 5.7781,
      "mean_token_accuracy": 0.17582362294197082,
      "num_tokens": 57892736.0,
      "step": 26745
    },
    {
      "entropy": 6.512140226364136,
      "epoch": 2.8625394617154476,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004192095911801785,
      "loss": 5.8596,
      "mean_token_accuracy": 0.16790420860052108,
      "num_tokens": 57904340.0,
      "step": 26750
    },
    {
      "entropy": 6.486385202407837,
      "epoch": 2.8630745358232117,
      "grad_norm": 1.046875,
      "learning_rate": 0.000419180241245625,
      "loss": 5.8283,
      "mean_token_accuracy": 0.17195628434419633,
      "num_tokens": 57916369.0,
      "step": 26755
    },
    {
      "entropy": 6.4878654956817625,
      "epoch": 2.8636096099309754,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004191508871478873,
      "loss": 5.8856,
      "mean_token_accuracy": 0.16352616995573044,
      "num_tokens": 57926867.0,
      "step": 26760
    },
    {
      "entropy": 6.466889953613281,
      "epoch": 2.8641446840387395,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041912152888781283,
      "loss": 5.7628,
      "mean_token_accuracy": 0.16692820489406585,
      "num_tokens": 57938408.0,
      "step": 26765
    },
    {
      "entropy": 6.470924282073975,
      "epoch": 2.8646797581465036,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041909216646624967,
      "loss": 5.7712,
      "mean_token_accuracy": 0.18296070098876954,
      "num_tokens": 57948821.0,
      "step": 26770
    },
    {
      "entropy": 6.51012921333313,
      "epoch": 2.865214832254267,
      "grad_norm": 1.0,
      "learning_rate": 0.0004190627998840453,
      "loss": 5.896,
      "mean_token_accuracy": 0.16444231569766998,
      "num_tokens": 57959589.0,
      "step": 26775
    },
    {
      "entropy": 6.452295017242432,
      "epoch": 2.8657499063620313,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004190334291420479,
      "loss": 5.8016,
      "mean_token_accuracy": 0.16970878541469575,
      "num_tokens": 57970508.0,
      "step": 26780
    },
    {
      "entropy": 6.47591986656189,
      "epoch": 2.866284980469795,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004190040542411055,
      "loss": 5.8103,
      "mean_token_accuracy": 0.17332662045955657,
      "num_tokens": 57980411.0,
      "step": 26785
    },
    {
      "entropy": 6.465128421783447,
      "epoch": 2.866820054577559,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041897467518206614,
      "loss": 5.8373,
      "mean_token_accuracy": 0.17022701352834702,
      "num_tokens": 57991457.0,
      "step": 26790
    },
    {
      "entropy": 6.515887641906739,
      "epoch": 2.8673551286853227,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004189452919657784,
      "loss": 5.8746,
      "mean_token_accuracy": 0.16018275767564774,
      "num_tokens": 58001860.0,
      "step": 26795
    },
    {
      "entropy": 6.508437824249268,
      "epoch": 2.867890202793087,
      "grad_norm": 1.125,
      "learning_rate": 0.00041891590459309055,
      "loss": 5.8849,
      "mean_token_accuracy": 0.1620408520102501,
      "num_tokens": 58011958.0,
      "step": 26800
    },
    {
      "entropy": 6.456497097015381,
      "epoch": 2.868425276900851,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041888651306485117,
      "loss": 5.8033,
      "mean_token_accuracy": 0.17302217185497284,
      "num_tokens": 58022216.0,
      "step": 26805
    },
    {
      "entropy": 6.4861448287963865,
      "epoch": 2.8689603510086146,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041885711738190895,
      "loss": 5.8112,
      "mean_token_accuracy": 0.1695413902401924,
      "num_tokens": 58032502.0,
      "step": 26810
    },
    {
      "entropy": 6.404025983810425,
      "epoch": 2.8694954251163787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004188277175451126,
      "loss": 5.8561,
      "mean_token_accuracy": 0.16837769448757173,
      "num_tokens": 58043944.0,
      "step": 26815
    },
    {
      "entropy": 6.462481927871704,
      "epoch": 2.8700304992241428,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041879831355531123,
      "loss": 5.8948,
      "mean_token_accuracy": 0.1699540928006172,
      "num_tokens": 58055857.0,
      "step": 26820
    },
    {
      "entropy": 6.467721462249756,
      "epoch": 2.8705655733319064,
      "grad_norm": 1.125,
      "learning_rate": 0.0004187689054133537,
      "loss": 5.8357,
      "mean_token_accuracy": 0.17062511444091796,
      "num_tokens": 58067070.0,
      "step": 26825
    },
    {
      "entropy": 6.43928713798523,
      "epoch": 2.8711006474396705,
      "grad_norm": 1.25,
      "learning_rate": 0.0004187394931200893,
      "loss": 5.6715,
      "mean_token_accuracy": 0.18161547929048538,
      "num_tokens": 58077922.0,
      "step": 26830
    },
    {
      "entropy": 6.483482313156128,
      "epoch": 2.871635721547434,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004187100766763671,
      "loss": 5.7791,
      "mean_token_accuracy": 0.17461425215005874,
      "num_tokens": 58089698.0,
      "step": 26835
    },
    {
      "entropy": 6.493060874938965,
      "epoch": 2.8721707956551983,
      "grad_norm": 1.125,
      "learning_rate": 0.00041868065608303674,
      "loss": 5.7825,
      "mean_token_accuracy": 0.17359798699617385,
      "num_tokens": 58100640.0,
      "step": 26840
    },
    {
      "entropy": 6.437036657333374,
      "epoch": 2.872705869762962,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004186512313409476,
      "loss": 5.8504,
      "mean_token_accuracy": 0.16769282072782515,
      "num_tokens": 58112810.0,
      "step": 26845
    },
    {
      "entropy": 6.5224333763122555,
      "epoch": 2.873240943870726,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004186218024509493,
      "loss": 5.8765,
      "mean_token_accuracy": 0.16851792186498643,
      "num_tokens": 58123818.0,
      "step": 26850
    },
    {
      "entropy": 6.515582752227783,
      "epoch": 2.87377601797849,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041859236941389175,
      "loss": 5.842,
      "mean_token_accuracy": 0.16800140142440795,
      "num_tokens": 58134785.0,
      "step": 26855
    },
    {
      "entropy": 6.480053186416626,
      "epoch": 2.874311092086254,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004185629322306246,
      "loss": 5.8337,
      "mean_token_accuracy": 0.16924772560596466,
      "num_tokens": 58144987.0,
      "step": 26860
    },
    {
      "entropy": 6.466507291793823,
      "epoch": 2.874846166194018,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041853349090199805,
      "loss": 5.7774,
      "mean_token_accuracy": 0.17792009860277175,
      "num_tokens": 58155031.0,
      "step": 26865
    },
    {
      "entropy": 6.419298839569092,
      "epoch": 2.875381240301782,
      "grad_norm": 1.171875,
      "learning_rate": 0.000418504045428862,
      "loss": 5.7814,
      "mean_token_accuracy": 0.17632722854614258,
      "num_tokens": 58165216.0,
      "step": 26870
    },
    {
      "entropy": 6.472931957244873,
      "epoch": 2.8759163144095456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004184745958120669,
      "loss": 5.8231,
      "mean_token_accuracy": 0.1767644166946411,
      "num_tokens": 58176329.0,
      "step": 26875
    },
    {
      "entropy": 6.513908958435058,
      "epoch": 2.8764513885173097,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041844514205246293,
      "loss": 5.8432,
      "mean_token_accuracy": 0.16909378618001938,
      "num_tokens": 58187058.0,
      "step": 26880
    },
    {
      "entropy": 6.488415479660034,
      "epoch": 2.876986462625074,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004184156841509007,
      "loss": 5.8192,
      "mean_token_accuracy": 0.17203661650419236,
      "num_tokens": 58196579.0,
      "step": 26885
    },
    {
      "entropy": 6.34435248374939,
      "epoch": 2.8775215367328375,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041838622210823076,
      "loss": 5.7616,
      "mean_token_accuracy": 0.1723661795258522,
      "num_tokens": 58206689.0,
      "step": 26890
    },
    {
      "entropy": 6.45076994895935,
      "epoch": 2.878056610840601,
      "grad_norm": 1.125,
      "learning_rate": 0.0004183567559253037,
      "loss": 5.8194,
      "mean_token_accuracy": 0.17260395735502243,
      "num_tokens": 58218093.0,
      "step": 26895
    },
    {
      "entropy": 6.491899871826172,
      "epoch": 2.8785916849483653,
      "grad_norm": 1.125,
      "learning_rate": 0.00041832728560297055,
      "loss": 5.8519,
      "mean_token_accuracy": 0.16608088463544846,
      "num_tokens": 58228587.0,
      "step": 26900
    },
    {
      "entropy": 6.448890924453735,
      "epoch": 2.8791267590561294,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041829781114208214,
      "loss": 5.7354,
      "mean_token_accuracy": 0.174269337952137,
      "num_tokens": 58240500.0,
      "step": 26905
    },
    {
      "entropy": 6.4216859340667725,
      "epoch": 2.879661833163893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041826833254348956,
      "loss": 5.7849,
      "mean_token_accuracy": 0.1705540508031845,
      "num_tokens": 58252253.0,
      "step": 26910
    },
    {
      "entropy": 6.4954084873199465,
      "epoch": 2.880196907271657,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041823884980804406,
      "loss": 5.8802,
      "mean_token_accuracy": 0.1667529121041298,
      "num_tokens": 58263356.0,
      "step": 26915
    },
    {
      "entropy": 6.381589460372925,
      "epoch": 2.880731981379421,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041820936293659685,
      "loss": 5.6827,
      "mean_token_accuracy": 0.17650352865457536,
      "num_tokens": 58273270.0,
      "step": 26920
    },
    {
      "entropy": 6.4515797138214115,
      "epoch": 2.881267055487185,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041817987192999947,
      "loss": 5.8268,
      "mean_token_accuracy": 0.16515176147222518,
      "num_tokens": 58284669.0,
      "step": 26925
    },
    {
      "entropy": 6.507050037384033,
      "epoch": 2.881802129594949,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041815037678910334,
      "loss": 5.8956,
      "mean_token_accuracy": 0.1675328105688095,
      "num_tokens": 58294791.0,
      "step": 26930
    },
    {
      "entropy": 6.482274389266967,
      "epoch": 2.882337203702713,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004181208775147602,
      "loss": 5.8511,
      "mean_token_accuracy": 0.17185698598623275,
      "num_tokens": 58305789.0,
      "step": 26935
    },
    {
      "entropy": 6.483967399597168,
      "epoch": 2.8828722778104767,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004180913741078218,
      "loss": 5.8235,
      "mean_token_accuracy": 0.16828051954507828,
      "num_tokens": 58317547.0,
      "step": 26940
    },
    {
      "entropy": 6.468969678878784,
      "epoch": 2.883407351918241,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004180618665691402,
      "loss": 5.7839,
      "mean_token_accuracy": 0.17038003206253052,
      "num_tokens": 58328408.0,
      "step": 26945
    },
    {
      "entropy": 6.314954519271851,
      "epoch": 2.8839424260260045,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004180323548995672,
      "loss": 5.6977,
      "mean_token_accuracy": 0.1877925291657448,
      "num_tokens": 58339356.0,
      "step": 26950
    },
    {
      "entropy": 6.383652257919311,
      "epoch": 2.8844775001337686,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041800283909995505,
      "loss": 5.7542,
      "mean_token_accuracy": 0.1818198636174202,
      "num_tokens": 58349016.0,
      "step": 26955
    },
    {
      "entropy": 6.49000244140625,
      "epoch": 2.8850125742415322,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004179733191711561,
      "loss": 5.896,
      "mean_token_accuracy": 0.16129131615161896,
      "num_tokens": 58360188.0,
      "step": 26960
    },
    {
      "entropy": 6.546307802200317,
      "epoch": 2.8855476483492963,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041794379511402263,
      "loss": 5.8703,
      "mean_token_accuracy": 0.16784733384847642,
      "num_tokens": 58370733.0,
      "step": 26965
    },
    {
      "entropy": 6.460375642776489,
      "epoch": 2.8860827224570604,
      "grad_norm": 1.125,
      "learning_rate": 0.0004179142669294071,
      "loss": 5.7893,
      "mean_token_accuracy": 0.1680884838104248,
      "num_tokens": 58382249.0,
      "step": 26970
    },
    {
      "entropy": 6.437125873565674,
      "epoch": 2.886617796564824,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041788473461816224,
      "loss": 5.7672,
      "mean_token_accuracy": 0.1764753744006157,
      "num_tokens": 58393598.0,
      "step": 26975
    },
    {
      "entropy": 6.470301008224487,
      "epoch": 2.887152870672588,
      "grad_norm": 1.125,
      "learning_rate": 0.00041785519818114073,
      "loss": 5.8312,
      "mean_token_accuracy": 0.16676746010780336,
      "num_tokens": 58403920.0,
      "step": 26980
    },
    {
      "entropy": 6.410873746871948,
      "epoch": 2.8876879447803523,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041782565761919546,
      "loss": 5.8386,
      "mean_token_accuracy": 0.1655145451426506,
      "num_tokens": 58413691.0,
      "step": 26985
    },
    {
      "entropy": 6.512901496887207,
      "epoch": 2.888223018888116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041779611293317936,
      "loss": 5.8365,
      "mean_token_accuracy": 0.1713208317756653,
      "num_tokens": 58425995.0,
      "step": 26990
    },
    {
      "entropy": 6.47371187210083,
      "epoch": 2.88875809299588,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004177665641239456,
      "loss": 5.8183,
      "mean_token_accuracy": 0.17210733890533447,
      "num_tokens": 58435984.0,
      "step": 26995
    },
    {
      "entropy": 6.47396240234375,
      "epoch": 2.8892931671036437,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041773701119234734,
      "loss": 5.8545,
      "mean_token_accuracy": 0.1672710120677948,
      "num_tokens": 58447630.0,
      "step": 27000
    },
    {
      "epoch": 2.8892931671036437,
      "eval_entropy": 6.260813719628635,
      "eval_loss": 5.9033002853393555,
      "eval_mean_token_accuracy": 0.17514751864361233,
      "eval_num_tokens": 58447630.0,
      "eval_runtime": 17.6434,
      "eval_samples_per_second": 1682.159,
      "eval_steps_per_second": 210.277,
      "step": 27000
    }
  ],
  "logging_steps": 5,
  "max_steps": 93440,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.274155940524032e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}