{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.2472042377869337,
  "eval_steps": 3000,
  "global_step": 21000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.722395038604736,
      "epoch": 0.0005350741077639254,
      "grad_norm": 19.75,
      "learning_rate": 2e-06,
      "loss": 14.1573,
      "mean_token_accuracy": 0.0019991567474789917,
      "num_tokens": 10038.0,
      "step": 5
    },
    {
      "entropy": 4.733135223388672,
      "epoch": 0.0010701482155278507,
      "grad_norm": 22.5,
      "learning_rate": 4.5e-06,
      "loss": 14.1399,
      "mean_token_accuracy": 0.002229393390007317,
      "num_tokens": 21916.0,
      "step": 10
    },
    {
      "entropy": 4.8563759326934814,
      "epoch": 0.0016052223232917758,
      "grad_norm": 30.875,
      "learning_rate": 7e-06,
      "loss": 13.8613,
      "mean_token_accuracy": 0.0014959125372115522,
      "num_tokens": 33434.0,
      "step": 15
    },
    {
      "entropy": 5.691608190536499,
      "epoch": 0.0021402964310557014,
      "grad_norm": 39.0,
      "learning_rate": 9.5e-06,
      "loss": 12.8416,
      "mean_token_accuracy": 0.0012791280925739557,
      "num_tokens": 44216.0,
      "step": 20
    },
    {
      "entropy": 9.404048442840576,
      "epoch": 0.0026753705388196263,
      "grad_norm": 3.578125,
      "learning_rate": 1.2e-05,
      "loss": 11.1148,
      "mean_token_accuracy": 0.0004811433027498424,
      "num_tokens": 54874.0,
      "step": 25
    },
    {
      "entropy": 10.681069660186768,
      "epoch": 0.0032104446465835517,
      "grad_norm": 3.109375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.7614,
      "mean_token_accuracy": 0.004937578848330304,
      "num_tokens": 65085.0,
      "step": 30
    },
    {
      "entropy": 10.695120239257813,
      "epoch": 0.003745518754347477,
      "grad_norm": 3.03125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.4993,
      "mean_token_accuracy": 0.02851757351309061,
      "num_tokens": 75601.0,
      "step": 35
    },
    {
      "entropy": 10.643773174285888,
      "epoch": 0.004280592862111403,
      "grad_norm": 2.40625,
      "learning_rate": 1.95e-05,
      "loss": 10.2473,
      "mean_token_accuracy": 0.029688681289553643,
      "num_tokens": 86193.0,
      "step": 40
    },
    {
      "entropy": 10.502691459655761,
      "epoch": 0.004815666969875327,
      "grad_norm": 1.8203125,
      "learning_rate": 2.2e-05,
      "loss": 10.0423,
      "mean_token_accuracy": 0.045137681812047956,
      "num_tokens": 96957.0,
      "step": 45
    },
    {
      "entropy": 10.450238704681396,
      "epoch": 0.005350741077639253,
      "grad_norm": 1.875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.9263,
      "mean_token_accuracy": 0.04649358205497265,
      "num_tokens": 108408.0,
      "step": 50
    },
    {
      "entropy": 10.498290061950684,
      "epoch": 0.005885815185403178,
      "grad_norm": 1.609375,
      "learning_rate": 2.7e-05,
      "loss": 9.8446,
      "mean_token_accuracy": 0.04394957646727562,
      "num_tokens": 119273.0,
      "step": 55
    },
    {
      "entropy": 10.508341979980468,
      "epoch": 0.006420889293167103,
      "grad_norm": 1.6796875,
      "learning_rate": 2.95e-05,
      "loss": 9.7839,
      "mean_token_accuracy": 0.04651816934347153,
      "num_tokens": 130817.0,
      "step": 60
    },
    {
      "entropy": 10.453034210205079,
      "epoch": 0.006955963400931029,
      "grad_norm": 1.5859375,
      "learning_rate": 3.2e-05,
      "loss": 9.7212,
      "mean_token_accuracy": 0.04715100526809692,
      "num_tokens": 142599.0,
      "step": 65
    },
    {
      "entropy": 10.463440990447998,
      "epoch": 0.007491037508694954,
      "grad_norm": 1.703125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.6449,
      "mean_token_accuracy": 0.04722196869552135,
      "num_tokens": 153617.0,
      "step": 70
    },
    {
      "entropy": 10.457964038848877,
      "epoch": 0.00802611161645888,
      "grad_norm": 1.453125,
      "learning_rate": 3.7e-05,
      "loss": 9.6007,
      "mean_token_accuracy": 0.053398077934980394,
      "num_tokens": 165280.0,
      "step": 75
    },
    {
      "entropy": 10.41784963607788,
      "epoch": 0.008561185724222806,
      "grad_norm": 1.328125,
      "learning_rate": 3.95e-05,
      "loss": 9.5362,
      "mean_token_accuracy": 0.053571393340826036,
      "num_tokens": 176896.0,
      "step": 80
    },
    {
      "entropy": 10.401211261749268,
      "epoch": 0.00909625983198673,
      "grad_norm": 1.453125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.4668,
      "mean_token_accuracy": 0.05559399500489235,
      "num_tokens": 188000.0,
      "step": 85
    },
    {
      "entropy": 10.360631847381592,
      "epoch": 0.009631333939750655,
      "grad_norm": 1.5390625,
      "learning_rate": 4.45e-05,
      "loss": 9.4151,
      "mean_token_accuracy": 0.055846284702420236,
      "num_tokens": 199360.0,
      "step": 90
    },
    {
      "entropy": 10.300343799591065,
      "epoch": 0.01016640804751458,
      "grad_norm": 1.2890625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.3631,
      "mean_token_accuracy": 0.05220822170376778,
      "num_tokens": 210214.0,
      "step": 95
    },
    {
      "entropy": 10.266947269439697,
      "epoch": 0.010701482155278505,
      "grad_norm": 1.421875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.2166,
      "mean_token_accuracy": 0.05733812190592289,
      "num_tokens": 220567.0,
      "step": 100
    },
    {
      "entropy": 10.19696626663208,
      "epoch": 0.011236556263042432,
      "grad_norm": 1.2734375,
      "learning_rate": 5.2e-05,
      "loss": 9.1178,
      "mean_token_accuracy": 0.05813850089907646,
      "num_tokens": 230585.0,
      "step": 105
    },
    {
      "entropy": 10.136646366119384,
      "epoch": 0.011771630370806356,
      "grad_norm": 1.296875,
      "learning_rate": 5.45e-05,
      "loss": 9.0634,
      "mean_token_accuracy": 0.06136388815939427,
      "num_tokens": 241996.0,
      "step": 110
    },
    {
      "entropy": 10.102339267730713,
      "epoch": 0.012306704478570282,
      "grad_norm": 1.25,
      "learning_rate": 5.7e-05,
      "loss": 9.0193,
      "mean_token_accuracy": 0.0571962833404541,
      "num_tokens": 253348.0,
      "step": 115
    },
    {
      "entropy": 10.066826057434081,
      "epoch": 0.012841778586334207,
      "grad_norm": 1.359375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.9111,
      "mean_token_accuracy": 0.055212068185210225,
      "num_tokens": 264239.0,
      "step": 120
    },
    {
      "entropy": 9.894806575775146,
      "epoch": 0.013376852694098133,
      "grad_norm": 1.0703125,
      "learning_rate": 6.2e-05,
      "loss": 8.8161,
      "mean_token_accuracy": 0.05635458119213581,
      "num_tokens": 274522.0,
      "step": 125
    },
    {
      "entropy": 9.833152961730956,
      "epoch": 0.013911926801862057,
      "grad_norm": 1.46875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.7223,
      "mean_token_accuracy": 0.0572861559689045,
      "num_tokens": 284492.0,
      "step": 130
    },
    {
      "entropy": 9.669886493682862,
      "epoch": 0.014447000909625984,
      "grad_norm": 1.0390625,
      "learning_rate": 6.7e-05,
      "loss": 8.6657,
      "mean_token_accuracy": 0.05727625191211701,
      "num_tokens": 294796.0,
      "step": 135
    },
    {
      "entropy": 9.54542293548584,
      "epoch": 0.014982075017389908,
      "grad_norm": 0.88671875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.6381,
      "mean_token_accuracy": 0.055407172068953514,
      "num_tokens": 305252.0,
      "step": 140
    },
    {
      "entropy": 9.40982427597046,
      "epoch": 0.015517149125153834,
      "grad_norm": 0.96484375,
      "learning_rate": 7.2e-05,
      "loss": 8.541,
      "mean_token_accuracy": 0.0527797307819128,
      "num_tokens": 315505.0,
      "step": 145
    },
    {
      "entropy": 9.249729537963868,
      "epoch": 0.01605222323291776,
      "grad_norm": 1.0234375,
      "learning_rate": 7.45e-05,
      "loss": 8.5348,
      "mean_token_accuracy": 0.05866497568786144,
      "num_tokens": 326997.0,
      "step": 150
    },
    {
      "entropy": 9.123694038391113,
      "epoch": 0.016587297340681685,
      "grad_norm": 0.84375,
      "learning_rate": 7.7e-05,
      "loss": 8.4691,
      "mean_token_accuracy": 0.05534619092941284,
      "num_tokens": 338089.0,
      "step": 155
    },
    {
      "entropy": 8.934013175964356,
      "epoch": 0.01712237144844561,
      "grad_norm": 0.79296875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.4569,
      "mean_token_accuracy": 0.055205514281988145,
      "num_tokens": 348536.0,
      "step": 160
    },
    {
      "entropy": 8.928122615814209,
      "epoch": 0.017657445556209534,
      "grad_norm": 0.9375,
      "learning_rate": 8.2e-05,
      "loss": 8.472,
      "mean_token_accuracy": 0.05698439441621304,
      "num_tokens": 359776.0,
      "step": 165
    },
    {
      "entropy": 8.888623428344726,
      "epoch": 0.01819251966397346,
      "grad_norm": 0.9375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 8.4013,
      "mean_token_accuracy": 0.05928085558116436,
      "num_tokens": 370810.0,
      "step": 170
    },
    {
      "entropy": 8.746195697784424,
      "epoch": 0.018727593771737387,
      "grad_norm": 0.77734375,
      "learning_rate": 8.7e-05,
      "loss": 8.4049,
      "mean_token_accuracy": 0.05481472611427307,
      "num_tokens": 381788.0,
      "step": 175
    },
    {
      "entropy": 8.743162918090821,
      "epoch": 0.01926266787950131,
      "grad_norm": 0.84375,
      "learning_rate": 8.95e-05,
      "loss": 8.3926,
      "mean_token_accuracy": 0.055552761256694796,
      "num_tokens": 391984.0,
      "step": 180
    },
    {
      "entropy": 8.635337162017823,
      "epoch": 0.019797741987265235,
      "grad_norm": 0.921875,
      "learning_rate": 9.2e-05,
      "loss": 8.3947,
      "mean_token_accuracy": 0.05619353819638491,
      "num_tokens": 403430.0,
      "step": 185
    },
    {
      "entropy": 8.695697593688966,
      "epoch": 0.02033281609502916,
      "grad_norm": 0.81640625,
      "learning_rate": 9.45e-05,
      "loss": 8.3824,
      "mean_token_accuracy": 0.0553983062505722,
      "num_tokens": 414467.0,
      "step": 190
    },
    {
      "entropy": 8.570184803009033,
      "epoch": 0.020867890202793088,
      "grad_norm": 0.87890625,
      "learning_rate": 9.7e-05,
      "loss": 8.3479,
      "mean_token_accuracy": 0.05803927257657051,
      "num_tokens": 425752.0,
      "step": 195
    },
    {
      "entropy": 8.64119758605957,
      "epoch": 0.02140296431055701,
      "grad_norm": 0.93359375,
      "learning_rate": 9.95e-05,
      "loss": 8.3272,
      "mean_token_accuracy": 0.05978759676218033,
      "num_tokens": 435908.0,
      "step": 200
    },
    {
      "entropy": 8.563984680175782,
      "epoch": 0.021938038418320937,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000102,
      "loss": 8.3718,
      "mean_token_accuracy": 0.05641747824847698,
      "num_tokens": 446424.0,
      "step": 205
    },
    {
      "entropy": 8.567623805999755,
      "epoch": 0.022473112526084863,
      "grad_norm": 0.796875,
      "learning_rate": 0.00010449999999999999,
      "loss": 8.3392,
      "mean_token_accuracy": 0.06241393908858299,
      "num_tokens": 456854.0,
      "step": 210
    },
    {
      "entropy": 8.583000755310058,
      "epoch": 0.02300818663384879,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000107,
      "loss": 8.3556,
      "mean_token_accuracy": 0.06351997442543507,
      "num_tokens": 467899.0,
      "step": 215
    },
    {
      "entropy": 8.510110855102539,
      "epoch": 0.023543260741612712,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001095,
      "loss": 8.3013,
      "mean_token_accuracy": 0.061149004846811295,
      "num_tokens": 478683.0,
      "step": 220
    },
    {
      "entropy": 8.565245342254638,
      "epoch": 0.02407833484937664,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000112,
      "loss": 8.3457,
      "mean_token_accuracy": 0.05644676610827446,
      "num_tokens": 488745.0,
      "step": 225
    },
    {
      "entropy": 8.513838863372802,
      "epoch": 0.024613408957140565,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0001145,
      "loss": 8.4193,
      "mean_token_accuracy": 0.055473582446575166,
      "num_tokens": 499448.0,
      "step": 230
    },
    {
      "entropy": 8.656694984436035,
      "epoch": 0.02514848306490449,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00011700000000000001,
      "loss": 8.3499,
      "mean_token_accuracy": 0.05867874249815941,
      "num_tokens": 510148.0,
      "step": 235
    },
    {
      "entropy": 8.570264339447021,
      "epoch": 0.025683557172668414,
      "grad_norm": 0.84375,
      "learning_rate": 0.00011949999999999999,
      "loss": 8.2806,
      "mean_token_accuracy": 0.0545498002320528,
      "num_tokens": 522093.0,
      "step": 240
    },
    {
      "entropy": 8.575170516967773,
      "epoch": 0.02621863128043234,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000122,
      "loss": 8.3034,
      "mean_token_accuracy": 0.06172005347907543,
      "num_tokens": 532598.0,
      "step": 245
    },
    {
      "entropy": 8.435428237915039,
      "epoch": 0.026753705388196266,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001245,
      "loss": 8.2791,
      "mean_token_accuracy": 0.06071905605494976,
      "num_tokens": 542652.0,
      "step": 250
    },
    {
      "entropy": 8.516559886932374,
      "epoch": 0.027288779495960192,
      "grad_norm": 0.73828125,
      "learning_rate": 0.000127,
      "loss": 8.3286,
      "mean_token_accuracy": 0.062316672131419185,
      "num_tokens": 552526.0,
      "step": 255
    },
    {
      "entropy": 8.463175582885743,
      "epoch": 0.027823853603724115,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0001295,
      "loss": 8.3292,
      "mean_token_accuracy": 0.059035299345850945,
      "num_tokens": 563143.0,
      "step": 260
    },
    {
      "entropy": 8.509436798095702,
      "epoch": 0.02835892771148804,
      "grad_norm": 0.84375,
      "learning_rate": 0.000132,
      "loss": 8.3044,
      "mean_token_accuracy": 0.05978103019297123,
      "num_tokens": 573814.0,
      "step": 265
    },
    {
      "entropy": 8.468083572387695,
      "epoch": 0.028894001819251967,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00013450000000000002,
      "loss": 8.2633,
      "mean_token_accuracy": 0.06531455405056477,
      "num_tokens": 585367.0,
      "step": 270
    },
    {
      "entropy": 8.460441875457764,
      "epoch": 0.02942907592701589,
      "grad_norm": 2.34375,
      "learning_rate": 0.00013700000000000002,
      "loss": 8.3529,
      "mean_token_accuracy": 0.061703698709607124,
      "num_tokens": 595902.0,
      "step": 275
    },
    {
      "entropy": 8.537551498413086,
      "epoch": 0.029964150034779816,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001395,
      "loss": 8.2341,
      "mean_token_accuracy": 0.05600880384445191,
      "num_tokens": 607178.0,
      "step": 280
    },
    {
      "entropy": 8.52383508682251,
      "epoch": 0.030499224142543743,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00014199999999999998,
      "loss": 8.2256,
      "mean_token_accuracy": 0.061789267137646674,
      "num_tokens": 618604.0,
      "step": 285
    },
    {
      "entropy": 8.46062650680542,
      "epoch": 0.03103429825030767,
      "grad_norm": 0.84375,
      "learning_rate": 0.0001445,
      "loss": 8.2826,
      "mean_token_accuracy": 0.06835589855909348,
      "num_tokens": 628111.0,
      "step": 290
    },
    {
      "entropy": 8.490731143951416,
      "epoch": 0.03156937235807159,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000147,
      "loss": 8.3204,
      "mean_token_accuracy": 0.06516639403998851,
      "num_tokens": 639358.0,
      "step": 295
    },
    {
      "entropy": 8.466291427612305,
      "epoch": 0.03210444646583552,
      "grad_norm": 0.9375,
      "learning_rate": 0.0001495,
      "loss": 8.201,
      "mean_token_accuracy": 0.0651388991624117,
      "num_tokens": 649504.0,
      "step": 300
    },
    {
      "entropy": 8.41495246887207,
      "epoch": 0.032639520573599444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000152,
      "loss": 8.2145,
      "mean_token_accuracy": 0.06336927972733974,
      "num_tokens": 659924.0,
      "step": 305
    },
    {
      "entropy": 8.46861686706543,
      "epoch": 0.03317459468136337,
      "grad_norm": 0.96875,
      "learning_rate": 0.00015450000000000001,
      "loss": 8.2479,
      "mean_token_accuracy": 0.05956652238965034,
      "num_tokens": 671120.0,
      "step": 310
    },
    {
      "entropy": 8.482641410827636,
      "epoch": 0.033709668789127296,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000157,
      "loss": 8.2983,
      "mean_token_accuracy": 0.06287384144961834,
      "num_tokens": 681956.0,
      "step": 315
    },
    {
      "entropy": 8.419728565216065,
      "epoch": 0.03424474289689122,
      "grad_norm": 0.90625,
      "learning_rate": 0.0001595,
      "loss": 8.2718,
      "mean_token_accuracy": 0.06350533962249756,
      "num_tokens": 693226.0,
      "step": 320
    },
    {
      "entropy": 8.438337421417236,
      "epoch": 0.03477981700465514,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000162,
      "loss": 8.2865,
      "mean_token_accuracy": 0.0667075652629137,
      "num_tokens": 704691.0,
      "step": 325
    },
    {
      "entropy": 8.398693466186524,
      "epoch": 0.03531489111241907,
      "grad_norm": 0.9375,
      "learning_rate": 0.00016450000000000001,
      "loss": 8.2436,
      "mean_token_accuracy": 0.05939121432602405,
      "num_tokens": 716000.0,
      "step": 330
    },
    {
      "entropy": 8.420809078216553,
      "epoch": 0.035849965220182994,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016700000000000002,
      "loss": 8.1626,
      "mean_token_accuracy": 0.06278250962495804,
      "num_tokens": 726723.0,
      "step": 335
    },
    {
      "entropy": 8.330788707733154,
      "epoch": 0.03638503932794692,
      "grad_norm": 1.0625,
      "learning_rate": 0.00016950000000000003,
      "loss": 8.174,
      "mean_token_accuracy": 0.06592709310352803,
      "num_tokens": 736740.0,
      "step": 340
    },
    {
      "entropy": 8.379256629943848,
      "epoch": 0.03692011343571085,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00017199999999999998,
      "loss": 8.2427,
      "mean_token_accuracy": 0.06097902096807957,
      "num_tokens": 747567.0,
      "step": 345
    },
    {
      "entropy": 8.421171760559082,
      "epoch": 0.03745518754347477,
      "grad_norm": 1.0,
      "learning_rate": 0.00017449999999999999,
      "loss": 8.1672,
      "mean_token_accuracy": 0.06565289981663228,
      "num_tokens": 757937.0,
      "step": 350
    },
    {
      "entropy": 8.351040077209472,
      "epoch": 0.0379902616512387,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000177,
      "loss": 8.1967,
      "mean_token_accuracy": 0.06799040511250495,
      "num_tokens": 768003.0,
      "step": 355
    },
    {
      "entropy": 8.385930728912353,
      "epoch": 0.03852533575900262,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0001795,
      "loss": 8.2486,
      "mean_token_accuracy": 0.06156148537993431,
      "num_tokens": 779098.0,
      "step": 360
    },
    {
      "entropy": 8.474120998382569,
      "epoch": 0.039060409866766545,
      "grad_norm": 1.328125,
      "learning_rate": 0.000182,
      "loss": 8.216,
      "mean_token_accuracy": 0.06129729710519314,
      "num_tokens": 790002.0,
      "step": 365
    },
    {
      "entropy": 8.336334323883056,
      "epoch": 0.03959548397453047,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0001845,
      "loss": 8.1953,
      "mean_token_accuracy": 0.062206690385937694,
      "num_tokens": 800444.0,
      "step": 370
    },
    {
      "entropy": 8.424775123596191,
      "epoch": 0.0401305580822944,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000187,
      "loss": 8.1423,
      "mean_token_accuracy": 0.06721410900354385,
      "num_tokens": 810912.0,
      "step": 375
    },
    {
      "entropy": 8.392704105377197,
      "epoch": 0.04066563219005832,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0001895,
      "loss": 8.2261,
      "mean_token_accuracy": 0.05968447625637054,
      "num_tokens": 822809.0,
      "step": 380
    },
    {
      "entropy": 8.346403789520263,
      "epoch": 0.04120070629782225,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000192,
      "loss": 8.1395,
      "mean_token_accuracy": 0.06515709534287453,
      "num_tokens": 833362.0,
      "step": 385
    },
    {
      "entropy": 8.385848045349121,
      "epoch": 0.041735780405586176,
      "grad_norm": 0.984375,
      "learning_rate": 0.0001945,
      "loss": 8.1498,
      "mean_token_accuracy": 0.06862472333014011,
      "num_tokens": 844239.0,
      "step": 390
    },
    {
      "entropy": 8.250670433044434,
      "epoch": 0.0422708545133501,
      "grad_norm": 0.96875,
      "learning_rate": 0.00019700000000000002,
      "loss": 8.1561,
      "mean_token_accuracy": 0.060210096463561055,
      "num_tokens": 855178.0,
      "step": 395
    },
    {
      "entropy": 8.319873905181884,
      "epoch": 0.04280592862111402,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019950000000000002,
      "loss": 8.1566,
      "mean_token_accuracy": 0.06562219709157943,
      "num_tokens": 866120.0,
      "step": 400
    },
    {
      "entropy": 8.30604248046875,
      "epoch": 0.04334100272887795,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000202,
      "loss": 8.1382,
      "mean_token_accuracy": 0.0657996479421854,
      "num_tokens": 876569.0,
      "step": 405
    },
    {
      "entropy": 8.446116733551026,
      "epoch": 0.043876076836641874,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00020449999999999998,
      "loss": 8.2014,
      "mean_token_accuracy": 0.0606235396116972,
      "num_tokens": 887580.0,
      "step": 410
    },
    {
      "entropy": 8.289253234863281,
      "epoch": 0.0444111509444058,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000207,
      "loss": 8.1351,
      "mean_token_accuracy": 0.06126284152269364,
      "num_tokens": 897922.0,
      "step": 415
    },
    {
      "entropy": 8.343159866333007,
      "epoch": 0.044946225052169726,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002095,
      "loss": 8.0903,
      "mean_token_accuracy": 0.06723331548273563,
      "num_tokens": 908522.0,
      "step": 420
    },
    {
      "entropy": 8.330431842803955,
      "epoch": 0.04548129915993365,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000212,
      "loss": 8.1108,
      "mean_token_accuracy": 0.06402806043624878,
      "num_tokens": 918253.0,
      "step": 425
    },
    {
      "entropy": 8.292218208312988,
      "epoch": 0.04601637326769758,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0002145,
      "loss": 8.106,
      "mean_token_accuracy": 0.06909515634179116,
      "num_tokens": 929150.0,
      "step": 430
    },
    {
      "entropy": 8.331517314910888,
      "epoch": 0.0465514473754615,
      "grad_norm": 0.82421875,
      "learning_rate": 0.00021700000000000002,
      "loss": 8.1406,
      "mean_token_accuracy": 0.06766720302402973,
      "num_tokens": 941041.0,
      "step": 435
    },
    {
      "entropy": 8.226575946807861,
      "epoch": 0.047086521483225424,
      "grad_norm": 1.671875,
      "learning_rate": 0.0002195,
      "loss": 8.0896,
      "mean_token_accuracy": 0.07020636461675167,
      "num_tokens": 952714.0,
      "step": 440
    },
    {
      "entropy": 8.243675708770752,
      "epoch": 0.04762159559098935,
      "grad_norm": 1.265625,
      "learning_rate": 0.000222,
      "loss": 8.1592,
      "mean_token_accuracy": 0.060818764939904214,
      "num_tokens": 964330.0,
      "step": 445
    },
    {
      "entropy": 8.260573291778565,
      "epoch": 0.04815666969875328,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002245,
      "loss": 8.0846,
      "mean_token_accuracy": 0.0660694733262062,
      "num_tokens": 975257.0,
      "step": 450
    },
    {
      "entropy": 8.280968284606933,
      "epoch": 0.0486917438065172,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00022700000000000002,
      "loss": 8.11,
      "mean_token_accuracy": 0.06583500131964684,
      "num_tokens": 986335.0,
      "step": 455
    },
    {
      "entropy": 8.23782262802124,
      "epoch": 0.04922681791428113,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.9957,
      "mean_token_accuracy": 0.07163017019629478,
      "num_tokens": 997521.0,
      "step": 460
    },
    {
      "entropy": 8.19036931991577,
      "epoch": 0.049761892022045055,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00023200000000000003,
      "loss": 8.0475,
      "mean_token_accuracy": 0.06431003175675869,
      "num_tokens": 1008383.0,
      "step": 465
    },
    {
      "entropy": 8.257700347900391,
      "epoch": 0.05029696612980898,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023449999999999998,
      "loss": 8.1057,
      "mean_token_accuracy": 0.06629102565348148,
      "num_tokens": 1020552.0,
      "step": 470
    },
    {
      "entropy": 8.202971076965332,
      "epoch": 0.0508320402375729,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000237,
      "loss": 8.0756,
      "mean_token_accuracy": 0.06509139761328697,
      "num_tokens": 1031127.0,
      "step": 475
    },
    {
      "entropy": 8.153694152832031,
      "epoch": 0.05136711434533683,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0002395,
      "loss": 7.9914,
      "mean_token_accuracy": 0.07003656215965748,
      "num_tokens": 1043209.0,
      "step": 480
    },
    {
      "entropy": 8.294580459594727,
      "epoch": 0.05190218845310075,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000242,
      "loss": 8.1227,
      "mean_token_accuracy": 0.06236105561256409,
      "num_tokens": 1054032.0,
      "step": 485
    },
    {
      "entropy": 8.169051265716552,
      "epoch": 0.05243726256086468,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0002445,
      "loss": 7.9544,
      "mean_token_accuracy": 0.07054617665708066,
      "num_tokens": 1064565.0,
      "step": 490
    },
    {
      "entropy": 8.221727657318116,
      "epoch": 0.052972336668628606,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000247,
      "loss": 8.096,
      "mean_token_accuracy": 0.07002902217209339,
      "num_tokens": 1076352.0,
      "step": 495
    },
    {
      "entropy": 8.255829620361329,
      "epoch": 0.05350741077639253,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002495,
      "loss": 8.0241,
      "mean_token_accuracy": 0.07085983417928218,
      "num_tokens": 1087697.0,
      "step": 500
    },
    {
      "entropy": 8.139251327514648,
      "epoch": 0.05404248488415646,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000252,
      "loss": 7.9711,
      "mean_token_accuracy": 0.07000853382050991,
      "num_tokens": 1098056.0,
      "step": 505
    },
    {
      "entropy": 8.11564884185791,
      "epoch": 0.054577558991920384,
      "grad_norm": 0.890625,
      "learning_rate": 0.0002545,
      "loss": 7.9527,
      "mean_token_accuracy": 0.07442603223025798,
      "num_tokens": 1109219.0,
      "step": 510
    },
    {
      "entropy": 8.148744678497314,
      "epoch": 0.055112633099684304,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000257,
      "loss": 8.0254,
      "mean_token_accuracy": 0.0679584439843893,
      "num_tokens": 1120248.0,
      "step": 515
    },
    {
      "entropy": 8.103824901580811,
      "epoch": 0.05564770720744823,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002595,
      "loss": 8.035,
      "mean_token_accuracy": 0.06900341361761093,
      "num_tokens": 1131278.0,
      "step": 520
    },
    {
      "entropy": 8.184957122802734,
      "epoch": 0.056182781315212156,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000262,
      "loss": 8.053,
      "mean_token_accuracy": 0.06036060117185116,
      "num_tokens": 1143147.0,
      "step": 525
    },
    {
      "entropy": 8.217153358459473,
      "epoch": 0.05671785542297608,
      "grad_norm": 0.890625,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.9545,
      "mean_token_accuracy": 0.06593920513987542,
      "num_tokens": 1154316.0,
      "step": 530
    },
    {
      "entropy": 8.083821010589599,
      "epoch": 0.05725292953074001,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00026700000000000004,
      "loss": 8.0045,
      "mean_token_accuracy": 0.06556588783860207,
      "num_tokens": 1164696.0,
      "step": 535
    },
    {
      "entropy": 8.10741205215454,
      "epoch": 0.057788003638503935,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.9876,
      "mean_token_accuracy": 0.07419452145695686,
      "num_tokens": 1176129.0,
      "step": 540
    },
    {
      "entropy": 8.21908884048462,
      "epoch": 0.05832307774626786,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.9966,
      "mean_token_accuracy": 0.06897441558539867,
      "num_tokens": 1186499.0,
      "step": 545
    },
    {
      "entropy": 8.077346086502075,
      "epoch": 0.05885815185403178,
      "grad_norm": 0.90625,
      "learning_rate": 0.0002745,
      "loss": 7.9598,
      "mean_token_accuracy": 0.06709535419940948,
      "num_tokens": 1196841.0,
      "step": 550
    },
    {
      "entropy": 8.092563199996949,
      "epoch": 0.059393225961795706,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000277,
      "loss": 7.9145,
      "mean_token_accuracy": 0.07183470316231251,
      "num_tokens": 1207647.0,
      "step": 555
    },
    {
      "entropy": 8.107578134536743,
      "epoch": 0.05992830006955963,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002795,
      "loss": 7.9071,
      "mean_token_accuracy": 0.07336057722568512,
      "num_tokens": 1217072.0,
      "step": 560
    },
    {
      "entropy": 8.063977479934692,
      "epoch": 0.06046337417732356,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028199999999999997,
      "loss": 8.0262,
      "mean_token_accuracy": 0.06556856445968151,
      "num_tokens": 1228564.0,
      "step": 565
    },
    {
      "entropy": 8.046744298934936,
      "epoch": 0.060998448285087485,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002845,
      "loss": 7.952,
      "mean_token_accuracy": 0.07086653858423234,
      "num_tokens": 1238755.0,
      "step": 570
    },
    {
      "entropy": 8.117069578170776,
      "epoch": 0.06153352239285141,
      "grad_norm": 1.109375,
      "learning_rate": 0.000287,
      "loss": 7.9119,
      "mean_token_accuracy": 0.0674261998385191,
      "num_tokens": 1248968.0,
      "step": 575
    },
    {
      "entropy": 8.066808080673217,
      "epoch": 0.06206859650061534,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0002895,
      "loss": 7.8725,
      "mean_token_accuracy": 0.06969248615205288,
      "num_tokens": 1259256.0,
      "step": 580
    },
    {
      "entropy": 7.942452716827392,
      "epoch": 0.06260367060837926,
      "grad_norm": 1.046875,
      "learning_rate": 0.000292,
      "loss": 7.9218,
      "mean_token_accuracy": 0.07239063829183578,
      "num_tokens": 1270469.0,
      "step": 585
    },
    {
      "entropy": 8.091417694091797,
      "epoch": 0.06313874471614318,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002945,
      "loss": 7.8805,
      "mean_token_accuracy": 0.06797636225819588,
      "num_tokens": 1280558.0,
      "step": 590
    },
    {
      "entropy": 8.002246284484864,
      "epoch": 0.06367381882390712,
      "grad_norm": 0.921875,
      "learning_rate": 0.000297,
      "loss": 7.9341,
      "mean_token_accuracy": 0.06753262616693974,
      "num_tokens": 1291372.0,
      "step": 595
    },
    {
      "entropy": 8.119279384613037,
      "epoch": 0.06420889293167104,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0002995,
      "loss": 8.0125,
      "mean_token_accuracy": 0.0642974954098463,
      "num_tokens": 1302521.0,
      "step": 600
    },
    {
      "entropy": 8.015694999694825,
      "epoch": 0.06474396703943497,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000302,
      "loss": 7.8392,
      "mean_token_accuracy": 0.06977246440947056,
      "num_tokens": 1313632.0,
      "step": 605
    },
    {
      "entropy": 8.086053228378296,
      "epoch": 0.06527904114719889,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003045,
      "loss": 7.9187,
      "mean_token_accuracy": 0.07155244462192059,
      "num_tokens": 1325247.0,
      "step": 610
    },
    {
      "entropy": 7.980141305923462,
      "epoch": 0.06581411525496281,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000307,
      "loss": 7.8969,
      "mean_token_accuracy": 0.07443655468523502,
      "num_tokens": 1336026.0,
      "step": 615
    },
    {
      "entropy": 8.025841999053956,
      "epoch": 0.06634918936272674,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0003095,
      "loss": 7.8245,
      "mean_token_accuracy": 0.07328929230570794,
      "num_tokens": 1346920.0,
      "step": 620
    },
    {
      "entropy": 7.913932514190674,
      "epoch": 0.06688426347049066,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000312,
      "loss": 7.9712,
      "mean_token_accuracy": 0.07009200640022754,
      "num_tokens": 1359752.0,
      "step": 625
    },
    {
      "entropy": 8.165118312835693,
      "epoch": 0.06741933757825459,
      "grad_norm": 0.90625,
      "learning_rate": 0.0003145,
      "loss": 7.964,
      "mean_token_accuracy": 0.06890517212450505,
      "num_tokens": 1370874.0,
      "step": 630
    },
    {
      "entropy": 7.918849992752075,
      "epoch": 0.06795441168601851,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000317,
      "loss": 7.8755,
      "mean_token_accuracy": 0.07662892192602158,
      "num_tokens": 1382316.0,
      "step": 635
    },
    {
      "entropy": 7.990935611724853,
      "epoch": 0.06848948579378245,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003195,
      "loss": 7.8866,
      "mean_token_accuracy": 0.06935312673449516,
      "num_tokens": 1394656.0,
      "step": 640
    },
    {
      "entropy": 7.928519058227539,
      "epoch": 0.06902455990154636,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000322,
      "loss": 7.8415,
      "mean_token_accuracy": 0.07609727829694748,
      "num_tokens": 1405061.0,
      "step": 645
    },
    {
      "entropy": 7.969374084472657,
      "epoch": 0.06955963400931028,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.8124,
      "mean_token_accuracy": 0.07858087830245494,
      "num_tokens": 1415573.0,
      "step": 650
    },
    {
      "entropy": 7.936450529098511,
      "epoch": 0.07009470811707422,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.837,
      "mean_token_accuracy": 0.07200551256537438,
      "num_tokens": 1426103.0,
      "step": 655
    },
    {
      "entropy": 7.920168352127075,
      "epoch": 0.07062978222483814,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.867,
      "mean_token_accuracy": 0.06607943587005138,
      "num_tokens": 1437256.0,
      "step": 660
    },
    {
      "entropy": 8.067060947418213,
      "epoch": 0.07116485633260207,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.8179,
      "mean_token_accuracy": 0.07232290767133236,
      "num_tokens": 1446975.0,
      "step": 665
    },
    {
      "entropy": 7.911382961273193,
      "epoch": 0.07169993044036599,
      "grad_norm": 0.921875,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.8317,
      "mean_token_accuracy": 0.07128445580601692,
      "num_tokens": 1457751.0,
      "step": 670
    },
    {
      "entropy": 7.850908708572388,
      "epoch": 0.07223500454812992,
      "grad_norm": 1.03125,
      "learning_rate": 0.000337,
      "loss": 7.807,
      "mean_token_accuracy": 0.07180027030408383,
      "num_tokens": 1468186.0,
      "step": 675
    },
    {
      "entropy": 7.946501731872559,
      "epoch": 0.07277007865589384,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0003395,
      "loss": 7.8495,
      "mean_token_accuracy": 0.0691140852868557,
      "num_tokens": 1479780.0,
      "step": 680
    },
    {
      "entropy": 7.927443170547486,
      "epoch": 0.07330515276365776,
      "grad_norm": 0.890625,
      "learning_rate": 0.000342,
      "loss": 7.7733,
      "mean_token_accuracy": 0.07666020393371582,
      "num_tokens": 1490565.0,
      "step": 685
    },
    {
      "entropy": 7.999663305282593,
      "epoch": 0.0738402268714217,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.8189,
      "mean_token_accuracy": 0.07104030475020409,
      "num_tokens": 1501552.0,
      "step": 690
    },
    {
      "entropy": 7.861136484146118,
      "epoch": 0.07437530097918561,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000347,
      "loss": 7.8092,
      "mean_token_accuracy": 0.0690252747386694,
      "num_tokens": 1513311.0,
      "step": 695
    },
    {
      "entropy": 7.929182291030884,
      "epoch": 0.07491037508694955,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003495,
      "loss": 7.8493,
      "mean_token_accuracy": 0.07585305124521255,
      "num_tokens": 1524142.0,
      "step": 700
    },
    {
      "entropy": 7.9166868209838865,
      "epoch": 0.07544544919471347,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000352,
      "loss": 7.8032,
      "mean_token_accuracy": 0.07204971313476563,
      "num_tokens": 1535571.0,
      "step": 705
    },
    {
      "entropy": 7.963001823425293,
      "epoch": 0.0759805233024774,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003545,
      "loss": 7.8184,
      "mean_token_accuracy": 0.07111813239753247,
      "num_tokens": 1546528.0,
      "step": 710
    },
    {
      "entropy": 7.771885633468628,
      "epoch": 0.07651559741024132,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000357,
      "loss": 7.8303,
      "mean_token_accuracy": 0.06822719126939773,
      "num_tokens": 1557264.0,
      "step": 715
    },
    {
      "entropy": 7.922857713699341,
      "epoch": 0.07705067151800524,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003595,
      "loss": 7.7899,
      "mean_token_accuracy": 0.0768304031342268,
      "num_tokens": 1566773.0,
      "step": 720
    },
    {
      "entropy": 7.871666193008423,
      "epoch": 0.07758574562576917,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000362,
      "loss": 7.7434,
      "mean_token_accuracy": 0.07752364575862884,
      "num_tokens": 1578173.0,
      "step": 725
    },
    {
      "entropy": 7.79464282989502,
      "epoch": 0.07812081973353309,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003645,
      "loss": 7.7138,
      "mean_token_accuracy": 0.07739724665880203,
      "num_tokens": 1588537.0,
      "step": 730
    },
    {
      "entropy": 7.797579956054688,
      "epoch": 0.07865589384129702,
      "grad_norm": 1.0,
      "learning_rate": 0.000367,
      "loss": 7.7126,
      "mean_token_accuracy": 0.08267080709338188,
      "num_tokens": 1599159.0,
      "step": 735
    },
    {
      "entropy": 7.877125597000122,
      "epoch": 0.07919096794906094,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003695,
      "loss": 7.7908,
      "mean_token_accuracy": 0.07560665309429168,
      "num_tokens": 1610378.0,
      "step": 740
    },
    {
      "entropy": 7.83719277381897,
      "epoch": 0.07972604205682488,
      "grad_norm": 1.046875,
      "learning_rate": 0.000372,
      "loss": 7.8064,
      "mean_token_accuracy": 0.07094332799315453,
      "num_tokens": 1621215.0,
      "step": 745
    },
    {
      "entropy": 7.8825953006744385,
      "epoch": 0.0802611161645888,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003745,
      "loss": 7.7372,
      "mean_token_accuracy": 0.07185808084905147,
      "num_tokens": 1631170.0,
      "step": 750
    },
    {
      "entropy": 7.881593084335327,
      "epoch": 0.08079619027235273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000377,
      "loss": 7.7573,
      "mean_token_accuracy": 0.07288290858268738,
      "num_tokens": 1641897.0,
      "step": 755
    },
    {
      "entropy": 7.814771223068237,
      "epoch": 0.08133126438011665,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003795,
      "loss": 7.8249,
      "mean_token_accuracy": 0.07262703813612462,
      "num_tokens": 1653592.0,
      "step": 760
    },
    {
      "entropy": 7.764955615997314,
      "epoch": 0.08186633848788057,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000382,
      "loss": 7.7359,
      "mean_token_accuracy": 0.07105322629213333,
      "num_tokens": 1665760.0,
      "step": 765
    },
    {
      "entropy": 7.860084962844849,
      "epoch": 0.0824014125956445,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0003845,
      "loss": 7.7313,
      "mean_token_accuracy": 0.07819439396262169,
      "num_tokens": 1676211.0,
      "step": 770
    },
    {
      "entropy": 7.864154577255249,
      "epoch": 0.08293648670340842,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.7604,
      "mean_token_accuracy": 0.07544596865773201,
      "num_tokens": 1686405.0,
      "step": 775
    },
    {
      "entropy": 7.842417526245117,
      "epoch": 0.08347156081117235,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.7456,
      "mean_token_accuracy": 0.07263315543532371,
      "num_tokens": 1697340.0,
      "step": 780
    },
    {
      "entropy": 7.7913618087768555,
      "epoch": 0.08400663491893627,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.6556,
      "mean_token_accuracy": 0.0825340673327446,
      "num_tokens": 1707250.0,
      "step": 785
    },
    {
      "entropy": 7.6649213314056395,
      "epoch": 0.0845417090267002,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.6374,
      "mean_token_accuracy": 0.07561526447534561,
      "num_tokens": 1718333.0,
      "step": 790
    },
    {
      "entropy": 7.79561710357666,
      "epoch": 0.08507678313446412,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.737,
      "mean_token_accuracy": 0.06993279084563256,
      "num_tokens": 1730520.0,
      "step": 795
    },
    {
      "entropy": 7.776207351684571,
      "epoch": 0.08561185724222804,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003995,
      "loss": 7.7826,
      "mean_token_accuracy": 0.07513744048774243,
      "num_tokens": 1741898.0,
      "step": 800
    },
    {
      "entropy": 7.806885290145874,
      "epoch": 0.08614693134999198,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000402,
      "loss": 7.7283,
      "mean_token_accuracy": 0.07864765003323555,
      "num_tokens": 1751671.0,
      "step": 805
    },
    {
      "entropy": 7.722856760025024,
      "epoch": 0.0866820054577559,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004045,
      "loss": 7.7421,
      "mean_token_accuracy": 0.07978087440133094,
      "num_tokens": 1762076.0,
      "step": 810
    },
    {
      "entropy": 7.788973808288574,
      "epoch": 0.08721707956551983,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.7164,
      "mean_token_accuracy": 0.07577739097177982,
      "num_tokens": 1772666.0,
      "step": 815
    },
    {
      "entropy": 7.822106456756591,
      "epoch": 0.08775215367328375,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004095,
      "loss": 7.7369,
      "mean_token_accuracy": 0.07942529432475567,
      "num_tokens": 1783621.0,
      "step": 820
    },
    {
      "entropy": 7.778071069717408,
      "epoch": 0.08828722778104768,
      "grad_norm": 1.5625,
      "learning_rate": 0.000412,
      "loss": 7.8027,
      "mean_token_accuracy": 0.0739357691258192,
      "num_tokens": 1794654.0,
      "step": 825
    },
    {
      "entropy": 7.724924182891845,
      "epoch": 0.0888223018888116,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004145,
      "loss": 7.6881,
      "mean_token_accuracy": 0.08523635193705559,
      "num_tokens": 1805853.0,
      "step": 830
    },
    {
      "entropy": 7.641312026977539,
      "epoch": 0.08935737599657552,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000417,
      "loss": 7.6675,
      "mean_token_accuracy": 0.07578803896903992,
      "num_tokens": 1817006.0,
      "step": 835
    },
    {
      "entropy": 7.8067771911621096,
      "epoch": 0.08989245010433945,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004195,
      "loss": 7.662,
      "mean_token_accuracy": 0.07342330142855644,
      "num_tokens": 1827166.0,
      "step": 840
    },
    {
      "entropy": 7.841296720504761,
      "epoch": 0.09042752421210337,
      "grad_norm": 1.125,
      "learning_rate": 0.000422,
      "loss": 7.7485,
      "mean_token_accuracy": 0.07532198280096054,
      "num_tokens": 1837182.0,
      "step": 845
    },
    {
      "entropy": 7.612701177597046,
      "epoch": 0.0909625983198673,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004245,
      "loss": 7.6955,
      "mean_token_accuracy": 0.07896317429840564,
      "num_tokens": 1849034.0,
      "step": 850
    },
    {
      "entropy": 7.698033094406128,
      "epoch": 0.09149767242763122,
      "grad_norm": 0.984375,
      "learning_rate": 0.000427,
      "loss": 7.5761,
      "mean_token_accuracy": 0.08696590065956115,
      "num_tokens": 1859033.0,
      "step": 855
    },
    {
      "entropy": 7.688354253768921,
      "epoch": 0.09203274653539516,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004295,
      "loss": 7.6849,
      "mean_token_accuracy": 0.07928116545081139,
      "num_tokens": 1870274.0,
      "step": 860
    },
    {
      "entropy": 7.66806640625,
      "epoch": 0.09256782064315908,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000432,
      "loss": 7.6252,
      "mean_token_accuracy": 0.08459196016192436,
      "num_tokens": 1881306.0,
      "step": 865
    },
    {
      "entropy": 7.665304851531983,
      "epoch": 0.093102894750923,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004345,
      "loss": 7.6184,
      "mean_token_accuracy": 0.07928042747080326,
      "num_tokens": 1892464.0,
      "step": 870
    },
    {
      "entropy": 7.751847410202027,
      "epoch": 0.09363796885868693,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000437,
      "loss": 7.6891,
      "mean_token_accuracy": 0.07430651262402535,
      "num_tokens": 1903743.0,
      "step": 875
    },
    {
      "entropy": 7.656659269332886,
      "epoch": 0.09417304296645085,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004395,
      "loss": 7.6452,
      "mean_token_accuracy": 0.08042613193392753,
      "num_tokens": 1915253.0,
      "step": 880
    },
    {
      "entropy": 7.739474582672119,
      "epoch": 0.09470811707421478,
      "grad_norm": 1.234375,
      "learning_rate": 0.000442,
      "loss": 7.6105,
      "mean_token_accuracy": 0.07811787836253643,
      "num_tokens": 1926113.0,
      "step": 885
    },
    {
      "entropy": 7.660387325286865,
      "epoch": 0.0952431911819787,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004445,
      "loss": 7.709,
      "mean_token_accuracy": 0.07902690842747688,
      "num_tokens": 1937718.0,
      "step": 890
    },
    {
      "entropy": 7.725072240829467,
      "epoch": 0.09577826528974263,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000447,
      "loss": 7.6993,
      "mean_token_accuracy": 0.07840467914938927,
      "num_tokens": 1948307.0,
      "step": 895
    },
    {
      "entropy": 7.573423147201538,
      "epoch": 0.09631333939750655,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.5078,
      "mean_token_accuracy": 0.0859848216176033,
      "num_tokens": 1957861.0,
      "step": 900
    },
    {
      "entropy": 7.603833866119385,
      "epoch": 0.09684841350527049,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.5528,
      "mean_token_accuracy": 0.08676589056849479,
      "num_tokens": 1967984.0,
      "step": 905
    },
    {
      "entropy": 7.597575378417969,
      "epoch": 0.0973834876130344,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.5639,
      "mean_token_accuracy": 0.08211242109537124,
      "num_tokens": 1978608.0,
      "step": 910
    },
    {
      "entropy": 7.650602197647094,
      "epoch": 0.09791856172079832,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.6146,
      "mean_token_accuracy": 0.08076006174087524,
      "num_tokens": 1989355.0,
      "step": 915
    },
    {
      "entropy": 7.582733201980591,
      "epoch": 0.09845363582856226,
      "grad_norm": 0.953125,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.6068,
      "mean_token_accuracy": 0.08393601626157761,
      "num_tokens": 1999754.0,
      "step": 920
    },
    {
      "entropy": 7.575737714767456,
      "epoch": 0.09898870993632618,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000462,
      "loss": 7.5896,
      "mean_token_accuracy": 0.0805906105786562,
      "num_tokens": 2010877.0,
      "step": 925
    },
    {
      "entropy": 7.744453620910645,
      "epoch": 0.09952378404409011,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004645,
      "loss": 7.6979,
      "mean_token_accuracy": 0.07580301314592361,
      "num_tokens": 2022301.0,
      "step": 930
    },
    {
      "entropy": 7.62055778503418,
      "epoch": 0.10005885815185403,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000467,
      "loss": 7.5905,
      "mean_token_accuracy": 0.07995235547423363,
      "num_tokens": 2033537.0,
      "step": 935
    },
    {
      "entropy": 7.555106735229492,
      "epoch": 0.10059393225961796,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004695,
      "loss": 7.5269,
      "mean_token_accuracy": 0.0813826397061348,
      "num_tokens": 2044360.0,
      "step": 940
    },
    {
      "entropy": 7.654840993881225,
      "epoch": 0.10112900636738188,
      "grad_norm": 0.90625,
      "learning_rate": 0.000472,
      "loss": 7.6073,
      "mean_token_accuracy": 0.08286485522985458,
      "num_tokens": 2055000.0,
      "step": 945
    },
    {
      "entropy": 7.554987287521362,
      "epoch": 0.1016640804751458,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004745,
      "loss": 7.6055,
      "mean_token_accuracy": 0.08866458013653755,
      "num_tokens": 2064807.0,
      "step": 950
    },
    {
      "entropy": 7.591622447967529,
      "epoch": 0.10219915458290973,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000477,
      "loss": 7.5605,
      "mean_token_accuracy": 0.07844374999403954,
      "num_tokens": 2075618.0,
      "step": 955
    },
    {
      "entropy": 7.59330415725708,
      "epoch": 0.10273422869067365,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004795,
      "loss": 7.5705,
      "mean_token_accuracy": 0.07968904972076415,
      "num_tokens": 2086226.0,
      "step": 960
    },
    {
      "entropy": 7.618208169937134,
      "epoch": 0.10326930279843759,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000482,
      "loss": 7.5604,
      "mean_token_accuracy": 0.07846201322972775,
      "num_tokens": 2097222.0,
      "step": 965
    },
    {
      "entropy": 7.5985006332397464,
      "epoch": 0.1038043769062015,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004845,
      "loss": 7.6154,
      "mean_token_accuracy": 0.080904171615839,
      "num_tokens": 2108919.0,
      "step": 970
    },
    {
      "entropy": 7.526676940917969,
      "epoch": 0.10433945101396544,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000487,
      "loss": 7.5506,
      "mean_token_accuracy": 0.08453056961297989,
      "num_tokens": 2120215.0,
      "step": 975
    },
    {
      "entropy": 7.601911354064941,
      "epoch": 0.10487452512172936,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004895,
      "loss": 7.5917,
      "mean_token_accuracy": 0.07946700900793076,
      "num_tokens": 2131503.0,
      "step": 980
    },
    {
      "entropy": 7.553497743606568,
      "epoch": 0.10540959922949328,
      "grad_norm": 1.0,
      "learning_rate": 0.000492,
      "loss": 7.536,
      "mean_token_accuracy": 0.08508892878890037,
      "num_tokens": 2141551.0,
      "step": 985
    },
    {
      "entropy": 7.530622768402099,
      "epoch": 0.10594467333725721,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004945,
      "loss": 7.5234,
      "mean_token_accuracy": 0.08779416754841804,
      "num_tokens": 2151821.0,
      "step": 990
    },
    {
      "entropy": 7.575374460220337,
      "epoch": 0.10647974744502113,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000497,
      "loss": 7.569,
      "mean_token_accuracy": 0.08431452587246895,
      "num_tokens": 2162474.0,
      "step": 995
    },
    {
      "entropy": 7.60914421081543,
      "epoch": 0.10701482155278506,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995,
      "loss": 7.6061,
      "mean_token_accuracy": 0.0851107444614172,
      "num_tokens": 2173985.0,
      "step": 1000
    },
    {
      "entropy": 7.586228322982788,
      "epoch": 0.10754989566054898,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999999979210101,
      "loss": 7.6445,
      "mean_token_accuracy": 0.0784386482089758,
      "num_tokens": 2186432.0,
      "step": 1005
    },
    {
      "entropy": 7.589601135253906,
      "epoch": 0.10808496976831292,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999999894751138,
      "loss": 7.4898,
      "mean_token_accuracy": 0.08204622492194176,
      "num_tokens": 2197085.0,
      "step": 1010
    },
    {
      "entropy": 7.4365442276000975,
      "epoch": 0.10862004387607684,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999999745323744,
      "loss": 7.4981,
      "mean_token_accuracy": 0.08734163418412208,
      "num_tokens": 2207162.0,
      "step": 1015
    },
    {
      "entropy": 7.579596948623657,
      "epoch": 0.10915511798384077,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999999530927923,
      "loss": 7.6013,
      "mean_token_accuracy": 0.07841602377593518,
      "num_tokens": 2218714.0,
      "step": 1020
    },
    {
      "entropy": 7.562452411651611,
      "epoch": 0.10969019209160469,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999999251563682,
      "loss": 7.5478,
      "mean_token_accuracy": 0.08398598507046699,
      "num_tokens": 2229711.0,
      "step": 1025
    },
    {
      "entropy": 7.566829061508178,
      "epoch": 0.11022526619936861,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999998907231029,
      "loss": 7.5443,
      "mean_token_accuracy": 0.08293986842036247,
      "num_tokens": 2239885.0,
      "step": 1030
    },
    {
      "entropy": 7.510043811798096,
      "epoch": 0.11076034030713254,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999998497929974,
      "loss": 7.5413,
      "mean_token_accuracy": 0.08550122715532779,
      "num_tokens": 2250221.0,
      "step": 1035
    },
    {
      "entropy": 7.535279560089111,
      "epoch": 0.11129541441489646,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999998023660527,
      "loss": 7.6213,
      "mean_token_accuracy": 0.07655972242355347,
      "num_tokens": 2261059.0,
      "step": 1040
    },
    {
      "entropy": 7.53675012588501,
      "epoch": 0.11183048852266039,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999997484422705,
      "loss": 7.4938,
      "mean_token_accuracy": 0.08518914058804512,
      "num_tokens": 2271174.0,
      "step": 1045
    },
    {
      "entropy": 7.458289527893067,
      "epoch": 0.11236556263042431,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999996880216521,
      "loss": 7.5496,
      "mean_token_accuracy": 0.08104559034109116,
      "num_tokens": 2282431.0,
      "step": 1050
    },
    {
      "entropy": 7.545944452285767,
      "epoch": 0.11290063673818825,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999996211041994,
      "loss": 7.5439,
      "mean_token_accuracy": 0.08365396112203598,
      "num_tokens": 2292891.0,
      "step": 1055
    },
    {
      "entropy": 7.627215766906739,
      "epoch": 0.11343571084595216,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999995476899141,
      "loss": 7.5633,
      "mean_token_accuracy": 0.08242802582681179,
      "num_tokens": 2303764.0,
      "step": 1060
    },
    {
      "entropy": 7.38291015625,
      "epoch": 0.11397078495371608,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999994677787987,
      "loss": 7.4539,
      "mean_token_accuracy": 0.08884900882840156,
      "num_tokens": 2314172.0,
      "step": 1065
    },
    {
      "entropy": 7.587633895874023,
      "epoch": 0.11450585906148002,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999993813708551,
      "loss": 7.4497,
      "mean_token_accuracy": 0.09005523398518563,
      "num_tokens": 2324848.0,
      "step": 1070
    },
    {
      "entropy": 7.366536426544189,
      "epoch": 0.11504093316924394,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499999288466086,
      "loss": 7.4785,
      "mean_token_accuracy": 0.08457461893558502,
      "num_tokens": 2336601.0,
      "step": 1075
    },
    {
      "entropy": 7.547029972076416,
      "epoch": 0.11557600727700787,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999991890644941,
      "loss": 7.4505,
      "mean_token_accuracy": 0.08462237492203713,
      "num_tokens": 2347281.0,
      "step": 1080
    },
    {
      "entropy": 7.441406202316284,
      "epoch": 0.11611108138477179,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999990831660822,
      "loss": 7.4898,
      "mean_token_accuracy": 0.09125454500317573,
      "num_tokens": 2358199.0,
      "step": 1085
    },
    {
      "entropy": 7.519572830200195,
      "epoch": 0.11664615549253572,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999989707708534,
      "loss": 7.5368,
      "mean_token_accuracy": 0.08167605474591255,
      "num_tokens": 2369092.0,
      "step": 1090
    },
    {
      "entropy": 7.438222408294678,
      "epoch": 0.11718122960029964,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999988518788111,
      "loss": 7.5142,
      "mean_token_accuracy": 0.08107429631054401,
      "num_tokens": 2378849.0,
      "step": 1095
    },
    {
      "entropy": 7.512506532669067,
      "epoch": 0.11771630370806356,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999987264899583,
      "loss": 7.5032,
      "mean_token_accuracy": 0.08989172950387,
      "num_tokens": 2389587.0,
      "step": 1100
    },
    {
      "entropy": 7.559766674041748,
      "epoch": 0.1182513778158275,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499998594604299,
      "loss": 7.4806,
      "mean_token_accuracy": 0.0849291667342186,
      "num_tokens": 2400442.0,
      "step": 1105
    },
    {
      "entropy": 7.420989179611206,
      "epoch": 0.11878645192359141,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499998456221837,
      "loss": 7.4684,
      "mean_token_accuracy": 0.08736843839287758,
      "num_tokens": 2410771.0,
      "step": 1110
    },
    {
      "entropy": 7.410952568054199,
      "epoch": 0.11932152603135535,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999983113425762,
      "loss": 7.4693,
      "mean_token_accuracy": 0.09239947721362114,
      "num_tokens": 2421112.0,
      "step": 1115
    },
    {
      "entropy": 7.545989656448365,
      "epoch": 0.11985660013911927,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999981599665207,
      "loss": 7.4512,
      "mean_token_accuracy": 0.08613410592079163,
      "num_tokens": 2431891.0,
      "step": 1120
    },
    {
      "entropy": 7.4298975467681885,
      "epoch": 0.1203916742468832,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999980020936748,
      "loss": 7.4131,
      "mean_token_accuracy": 0.08798880130052567,
      "num_tokens": 2442987.0,
      "step": 1125
    },
    {
      "entropy": 7.39614634513855,
      "epoch": 0.12092674835464712,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999978377240434,
      "loss": 7.4519,
      "mean_token_accuracy": 0.08737303614616394,
      "num_tokens": 2452703.0,
      "step": 1130
    },
    {
      "entropy": 7.4548506259918215,
      "epoch": 0.12146182246241104,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499997666857631,
      "loss": 7.5001,
      "mean_token_accuracy": 0.08815332502126694,
      "num_tokens": 2462975.0,
      "step": 1135
    },
    {
      "entropy": 7.4424488067626955,
      "epoch": 0.12199689657017497,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999974894944426,
      "loss": 7.4515,
      "mean_token_accuracy": 0.084615208953619,
      "num_tokens": 2473853.0,
      "step": 1140
    },
    {
      "entropy": 7.427521181106568,
      "epoch": 0.12253197067793889,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999973056344832,
      "loss": 7.4324,
      "mean_token_accuracy": 0.08493943735957146,
      "num_tokens": 2485633.0,
      "step": 1145
    },
    {
      "entropy": 7.469012641906739,
      "epoch": 0.12306704478570282,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999971152777583,
      "loss": 7.3906,
      "mean_token_accuracy": 0.08597971722483636,
      "num_tokens": 2496922.0,
      "step": 1150
    },
    {
      "entropy": 7.443756341934204,
      "epoch": 0.12360211889346674,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004999969184242733,
      "loss": 7.4573,
      "mean_token_accuracy": 0.08639369383454323,
      "num_tokens": 2507890.0,
      "step": 1155
    },
    {
      "entropy": 7.414251852035522,
      "epoch": 0.12413719300123068,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499996715074034,
      "loss": 7.4679,
      "mean_token_accuracy": 0.08895903378725052,
      "num_tokens": 2519749.0,
      "step": 1160
    },
    {
      "entropy": 7.369732522964478,
      "epoch": 0.1246722671089946,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999965052270461,
      "loss": 7.4587,
      "mean_token_accuracy": 0.08741387501358985,
      "num_tokens": 2530666.0,
      "step": 1165
    },
    {
      "entropy": 7.455218648910522,
      "epoch": 0.12520734121675853,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999962888833157,
      "loss": 7.4517,
      "mean_token_accuracy": 0.087038604170084,
      "num_tokens": 2541594.0,
      "step": 1170
    },
    {
      "entropy": 7.468952941894531,
      "epoch": 0.12574241532452243,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999960660428491,
      "loss": 7.4548,
      "mean_token_accuracy": 0.08958195969462394,
      "num_tokens": 2552330.0,
      "step": 1175
    },
    {
      "entropy": 7.457394981384278,
      "epoch": 0.12627748943228637,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999958367056526,
      "loss": 7.5094,
      "mean_token_accuracy": 0.0814846083521843,
      "num_tokens": 2564601.0,
      "step": 1180
    },
    {
      "entropy": 7.5275298118591305,
      "epoch": 0.1268125635400503,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499995600871733,
      "loss": 7.4705,
      "mean_token_accuracy": 0.08419692143797874,
      "num_tokens": 2575539.0,
      "step": 1185
    },
    {
      "entropy": 7.352044630050659,
      "epoch": 0.12734763764781423,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999953585410971,
      "loss": 7.3968,
      "mean_token_accuracy": 0.08637173548340797,
      "num_tokens": 2586635.0,
      "step": 1190
    },
    {
      "entropy": 7.49582724571228,
      "epoch": 0.12788271175557814,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999951097137518,
      "loss": 7.4108,
      "mean_token_accuracy": 0.09154140949249268,
      "num_tokens": 2596814.0,
      "step": 1195
    },
    {
      "entropy": 7.325179576873779,
      "epoch": 0.12841778586334207,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999948543897044,
      "loss": 7.4522,
      "mean_token_accuracy": 0.08449748009443284,
      "num_tokens": 2608087.0,
      "step": 1200
    },
    {
      "entropy": 7.586294317245484,
      "epoch": 0.128952859971106,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999945925689621,
      "loss": 7.4593,
      "mean_token_accuracy": 0.08878617733716965,
      "num_tokens": 2619377.0,
      "step": 1205
    },
    {
      "entropy": 7.303364086151123,
      "epoch": 0.12948793407886994,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999943242515325,
      "loss": 7.3505,
      "mean_token_accuracy": 0.09070543050765992,
      "num_tokens": 2629089.0,
      "step": 1210
    },
    {
      "entropy": 7.450505971908569,
      "epoch": 0.13002300818663384,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999940494374236,
      "loss": 7.4299,
      "mean_token_accuracy": 0.08928245157003403,
      "num_tokens": 2639743.0,
      "step": 1215
    },
    {
      "entropy": 7.489728927612305,
      "epoch": 0.13055808229439778,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499993768126643,
      "loss": 7.4676,
      "mean_token_accuracy": 0.08282669633626938,
      "num_tokens": 2650631.0,
      "step": 1220
    },
    {
      "entropy": 7.376281833648681,
      "epoch": 0.1310931564021617,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499993480319199,
      "loss": 7.4305,
      "mean_token_accuracy": 0.08671566098928452,
      "num_tokens": 2660197.0,
      "step": 1225
    },
    {
      "entropy": 7.40994963645935,
      "epoch": 0.13162823050992561,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999931860150999,
      "loss": 7.3856,
      "mean_token_accuracy": 0.08833521455526352,
      "num_tokens": 2670162.0,
      "step": 1230
    },
    {
      "entropy": 7.446912240982056,
      "epoch": 0.13216330461768955,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999928852143541,
      "loss": 7.3825,
      "mean_token_accuracy": 0.0878111682832241,
      "num_tokens": 2681233.0,
      "step": 1235
    },
    {
      "entropy": 7.31794056892395,
      "epoch": 0.13269837872545348,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999925779169703,
      "loss": 7.3545,
      "mean_token_accuracy": 0.08922291100025177,
      "num_tokens": 2692145.0,
      "step": 1240
    },
    {
      "entropy": 7.363669300079346,
      "epoch": 0.13323345283321741,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999922641229575,
      "loss": 7.3874,
      "mean_token_accuracy": 0.09013602733612061,
      "num_tokens": 2703192.0,
      "step": 1245
    },
    {
      "entropy": 7.490590524673462,
      "epoch": 0.13376852694098132,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999919438323248,
      "loss": 7.4102,
      "mean_token_accuracy": 0.08662581890821457,
      "num_tokens": 2714804.0,
      "step": 1250
    },
    {
      "entropy": 7.302708768844605,
      "epoch": 0.13430360104874525,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999916170450812,
      "loss": 7.3626,
      "mean_token_accuracy": 0.09281876906752587,
      "num_tokens": 2724340.0,
      "step": 1255
    },
    {
      "entropy": 7.357819366455078,
      "epoch": 0.13483867515650919,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999912837612364,
      "loss": 7.3677,
      "mean_token_accuracy": 0.0910140611231327,
      "num_tokens": 2735767.0,
      "step": 1260
    },
    {
      "entropy": 7.488349437713623,
      "epoch": 0.1353737492642731,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999909439807997,
      "loss": 7.4274,
      "mean_token_accuracy": 0.08374015688896179,
      "num_tokens": 2747658.0,
      "step": 1265
    },
    {
      "entropy": 7.304438734054566,
      "epoch": 0.13590882337203702,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999905977037813,
      "loss": 7.2895,
      "mean_token_accuracy": 0.09504042863845825,
      "num_tokens": 2757901.0,
      "step": 1270
    },
    {
      "entropy": 7.333110237121582,
      "epoch": 0.13644389747980096,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999902449301909,
      "loss": 7.3426,
      "mean_token_accuracy": 0.08784788250923156,
      "num_tokens": 2768945.0,
      "step": 1275
    },
    {
      "entropy": 7.313035011291504,
      "epoch": 0.1369789715875649,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004999898856600387,
      "loss": 7.3162,
      "mean_token_accuracy": 0.10365832224488258,
      "num_tokens": 2780276.0,
      "step": 1280
    },
    {
      "entropy": 7.380517864227295,
      "epoch": 0.1375140456953288,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999895198933354,
      "loss": 7.3902,
      "mean_token_accuracy": 0.09485040828585625,
      "num_tokens": 2791183.0,
      "step": 1285
    },
    {
      "entropy": 7.4089750289917,
      "epoch": 0.13804911980309273,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999891476300911,
      "loss": 7.3059,
      "mean_token_accuracy": 0.08951659351587296,
      "num_tokens": 2801928.0,
      "step": 1290
    },
    {
      "entropy": 7.355879497528076,
      "epoch": 0.13858419391085666,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000499988768870317,
      "loss": 7.3872,
      "mean_token_accuracy": 0.08165468871593476,
      "num_tokens": 2813415.0,
      "step": 1295
    },
    {
      "entropy": 7.355454301834106,
      "epoch": 0.13911926801862057,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999883836140236,
      "loss": 7.3923,
      "mean_token_accuracy": 0.08718730732798577,
      "num_tokens": 2825288.0,
      "step": 1300
    },
    {
      "entropy": 7.375982999801636,
      "epoch": 0.1396543421263845,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999879918612224,
      "loss": 7.3466,
      "mean_token_accuracy": 0.09210592210292816,
      "num_tokens": 2836298.0,
      "step": 1305
    },
    {
      "entropy": 7.382136964797974,
      "epoch": 0.14018941623414843,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999875936119245,
      "loss": 7.325,
      "mean_token_accuracy": 0.09450321719050407,
      "num_tokens": 2846367.0,
      "step": 1310
    },
    {
      "entropy": 7.285005521774292,
      "epoch": 0.14072449034191237,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999871888661415,
      "loss": 7.3419,
      "mean_token_accuracy": 0.08686081692576408,
      "num_tokens": 2856832.0,
      "step": 1315
    },
    {
      "entropy": 7.406700801849365,
      "epoch": 0.14125956444967627,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499986777623885,
      "loss": 7.3557,
      "mean_token_accuracy": 0.08402335196733475,
      "num_tokens": 2867954.0,
      "step": 1320
    },
    {
      "entropy": 7.433336400985718,
      "epoch": 0.1417946385574402,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999863598851669,
      "loss": 7.4023,
      "mean_token_accuracy": 0.0905662328004837,
      "num_tokens": 2878714.0,
      "step": 1325
    },
    {
      "entropy": 7.216334629058838,
      "epoch": 0.14232971266520414,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999859356499992,
      "loss": 7.3287,
      "mean_token_accuracy": 0.09552604258060456,
      "num_tokens": 2890697.0,
      "step": 1330
    },
    {
      "entropy": 7.416978931427002,
      "epoch": 0.14286478677296804,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999855049183943,
      "loss": 7.3428,
      "mean_token_accuracy": 0.08655953332781792,
      "num_tokens": 2901468.0,
      "step": 1335
    },
    {
      "entropy": 7.319197273254394,
      "epoch": 0.14339986088073198,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999850676903646,
      "loss": 7.3074,
      "mean_token_accuracy": 0.0925314411520958,
      "num_tokens": 2913466.0,
      "step": 1340
    },
    {
      "entropy": 7.3042479991912845,
      "epoch": 0.1439349349884959,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999846239659226,
      "loss": 7.2803,
      "mean_token_accuracy": 0.09751186668872833,
      "num_tokens": 2923638.0,
      "step": 1345
    },
    {
      "entropy": 7.315991830825806,
      "epoch": 0.14447000909625984,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999841737450812,
      "loss": 7.3706,
      "mean_token_accuracy": 0.09228342920541763,
      "num_tokens": 2934790.0,
      "step": 1350
    },
    {
      "entropy": 7.371410799026489,
      "epoch": 0.14500508320402375,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999837170278535,
      "loss": 7.3076,
      "mean_token_accuracy": 0.08736552372574806,
      "num_tokens": 2946320.0,
      "step": 1355
    },
    {
      "entropy": 7.350737285614014,
      "epoch": 0.14554015731178768,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999832538142526,
      "loss": 7.2741,
      "mean_token_accuracy": 0.09060471281409263,
      "num_tokens": 2957077.0,
      "step": 1360
    },
    {
      "entropy": 7.227811241149903,
      "epoch": 0.14607523141955162,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999827841042917,
      "loss": 7.2953,
      "mean_token_accuracy": 0.08847967609763145,
      "num_tokens": 2967286.0,
      "step": 1365
    },
    {
      "entropy": 7.3782144546508786,
      "epoch": 0.14661030552731552,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004999823078979846,
      "loss": 7.3385,
      "mean_token_accuracy": 0.09329641237854958,
      "num_tokens": 2979516.0,
      "step": 1370
    },
    {
      "entropy": 7.3981129169464115,
      "epoch": 0.14714537963507945,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999818251953449,
      "loss": 7.4533,
      "mean_token_accuracy": 0.08806639239192009,
      "num_tokens": 2991594.0,
      "step": 1375
    },
    {
      "entropy": 7.355781364440918,
      "epoch": 0.1476804537428434,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999813359963867,
      "loss": 7.3319,
      "mean_token_accuracy": 0.09073294252157212,
      "num_tokens": 3002487.0,
      "step": 1380
    },
    {
      "entropy": 7.387376403808593,
      "epoch": 0.14821552785060732,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999808403011241,
      "loss": 7.3475,
      "mean_token_accuracy": 0.09293655529618264,
      "num_tokens": 3013197.0,
      "step": 1385
    },
    {
      "entropy": 7.207638359069824,
      "epoch": 0.14875060195837123,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999803381095712,
      "loss": 7.3669,
      "mean_token_accuracy": 0.08983460813760757,
      "num_tokens": 3024512.0,
      "step": 1390
    },
    {
      "entropy": 7.475574684143067,
      "epoch": 0.14928567606613516,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999798294217428,
      "loss": 7.3365,
      "mean_token_accuracy": 0.08949591889977455,
      "num_tokens": 3035621.0,
      "step": 1395
    },
    {
      "entropy": 7.273073291778564,
      "epoch": 0.1498207501738991,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999793142376533,
      "loss": 7.328,
      "mean_token_accuracy": 0.09089694023132325,
      "num_tokens": 3046492.0,
      "step": 1400
    },
    {
      "entropy": 7.384609603881836,
      "epoch": 0.150355824281663,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999787925573177,
      "loss": 7.3486,
      "mean_token_accuracy": 0.09108722358942031,
      "num_tokens": 3057715.0,
      "step": 1405
    },
    {
      "entropy": 7.4058764457702635,
      "epoch": 0.15089089838942693,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999782643807513,
      "loss": 7.2967,
      "mean_token_accuracy": 0.08750828057527542,
      "num_tokens": 3068344.0,
      "step": 1410
    },
    {
      "entropy": 7.250711297988891,
      "epoch": 0.15142597249719086,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999777297079689,
      "loss": 7.2772,
      "mean_token_accuracy": 0.08728645220398903,
      "num_tokens": 3079576.0,
      "step": 1415
    },
    {
      "entropy": 7.336335229873657,
      "epoch": 0.1519610466049548,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999771885389863,
      "loss": 7.2531,
      "mean_token_accuracy": 0.0913350023329258,
      "num_tokens": 3090201.0,
      "step": 1420
    },
    {
      "entropy": 7.224132394790649,
      "epoch": 0.1524961207127187,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999766408738189,
      "loss": 7.2227,
      "mean_token_accuracy": 0.09826805666089058,
      "num_tokens": 3099824.0,
      "step": 1425
    },
    {
      "entropy": 7.277629327774048,
      "epoch": 0.15303119482048264,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999760867124827,
      "loss": 7.3072,
      "mean_token_accuracy": 0.08773592785000801,
      "num_tokens": 3109948.0,
      "step": 1430
    },
    {
      "entropy": 7.388230419158935,
      "epoch": 0.15356626892824657,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999755260549937,
      "loss": 7.2946,
      "mean_token_accuracy": 0.09707588031888008,
      "num_tokens": 3121492.0,
      "step": 1435
    },
    {
      "entropy": 7.204993820190429,
      "epoch": 0.15410134303601047,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004999749589013677,
      "loss": 7.0761,
      "mean_token_accuracy": 0.11195724084973335,
      "num_tokens": 3131990.0,
      "step": 1440
    },
    {
      "entropy": 7.1843321323394775,
      "epoch": 0.1546364171437744,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999743852516217,
      "loss": 7.2552,
      "mean_token_accuracy": 0.09591018855571747,
      "num_tokens": 3142567.0,
      "step": 1445
    },
    {
      "entropy": 7.3704992771148685,
      "epoch": 0.15517149125153834,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004999738051057717,
      "loss": 7.2903,
      "mean_token_accuracy": 0.09068220183253288,
      "num_tokens": 3153083.0,
      "step": 1450
    },
    {
      "entropy": 7.274682903289795,
      "epoch": 0.15570656535930227,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999732184638347,
      "loss": 7.2761,
      "mean_token_accuracy": 0.09170850291848183,
      "num_tokens": 3163985.0,
      "step": 1455
    },
    {
      "entropy": 7.231216716766357,
      "epoch": 0.15624163946706618,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999726253258278,
      "loss": 7.2721,
      "mean_token_accuracy": 0.09016884565353393,
      "num_tokens": 3175259.0,
      "step": 1460
    },
    {
      "entropy": 7.324146890640259,
      "epoch": 0.1567767135748301,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999720256917678,
      "loss": 7.2402,
      "mean_token_accuracy": 0.09691082388162613,
      "num_tokens": 3185487.0,
      "step": 1465
    },
    {
      "entropy": 7.17336688041687,
      "epoch": 0.15731178768259405,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999714195616723,
      "loss": 7.226,
      "mean_token_accuracy": 0.0957397349178791,
      "num_tokens": 3196200.0,
      "step": 1470
    },
    {
      "entropy": 7.437542724609375,
      "epoch": 0.15784686179035795,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999708069355586,
      "loss": 7.3376,
      "mean_token_accuracy": 0.08450718447566033,
      "num_tokens": 3207675.0,
      "step": 1475
    },
    {
      "entropy": 7.145585060119629,
      "epoch": 0.15838193589812188,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999701878134445,
      "loss": 7.2091,
      "mean_token_accuracy": 0.09381906762719154,
      "num_tokens": 3218117.0,
      "step": 1480
    },
    {
      "entropy": 7.420627593994141,
      "epoch": 0.15891701000588582,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999695621953477,
      "loss": 7.2853,
      "mean_token_accuracy": 0.10199789702892303,
      "num_tokens": 3229208.0,
      "step": 1485
    },
    {
      "entropy": 7.2098907947540285,
      "epoch": 0.15945208411364975,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999689300812866,
      "loss": 7.184,
      "mean_token_accuracy": 0.09962581098079681,
      "num_tokens": 3239586.0,
      "step": 1490
    },
    {
      "entropy": 7.2988934993743895,
      "epoch": 0.15998715822141366,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004999682914712791,
      "loss": 7.2533,
      "mean_token_accuracy": 0.09887138307094574,
      "num_tokens": 3250103.0,
      "step": 1495
    },
    {
      "entropy": 7.312133359909057,
      "epoch": 0.1605222323291776,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999676463653439,
      "loss": 7.2385,
      "mean_token_accuracy": 0.09936083406209946,
      "num_tokens": 3261844.0,
      "step": 1500
    },
    {
      "entropy": 7.241543960571289,
      "epoch": 0.16105730643694152,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999669947634996,
      "loss": 7.1859,
      "mean_token_accuracy": 0.09886201843619347,
      "num_tokens": 3272073.0,
      "step": 1505
    },
    {
      "entropy": 7.265321111679077,
      "epoch": 0.16159238054470546,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999663366657647,
      "loss": 7.1839,
      "mean_token_accuracy": 0.09600849002599716,
      "num_tokens": 3281725.0,
      "step": 1510
    },
    {
      "entropy": 7.194610452651977,
      "epoch": 0.16212745465246936,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999656720721586,
      "loss": 7.3232,
      "mean_token_accuracy": 0.09642767012119294,
      "num_tokens": 3292708.0,
      "step": 1515
    },
    {
      "entropy": 7.37993426322937,
      "epoch": 0.1626625287602333,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999650009827004,
      "loss": 7.3017,
      "mean_token_accuracy": 0.0933539777994156,
      "num_tokens": 3303548.0,
      "step": 1520
    },
    {
      "entropy": 7.182165908813476,
      "epoch": 0.16319760286799723,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999643233974092,
      "loss": 7.1686,
      "mean_token_accuracy": 0.09983909949660301,
      "num_tokens": 3314261.0,
      "step": 1525
    },
    {
      "entropy": 7.272876930236817,
      "epoch": 0.16373267697576113,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999636393163049,
      "loss": 7.1939,
      "mean_token_accuracy": 0.09785049334168434,
      "num_tokens": 3324833.0,
      "step": 1530
    },
    {
      "entropy": 7.116202354431152,
      "epoch": 0.16426775108352507,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999629487394071,
      "loss": 7.0996,
      "mean_token_accuracy": 0.10248475447297097,
      "num_tokens": 3334790.0,
      "step": 1535
    },
    {
      "entropy": 7.299325656890869,
      "epoch": 0.164802825191289,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999622516667358,
      "loss": 7.229,
      "mean_token_accuracy": 0.09786866381764411,
      "num_tokens": 3345115.0,
      "step": 1540
    },
    {
      "entropy": 7.127289056777954,
      "epoch": 0.16533789929905293,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000499961548098311,
      "loss": 7.1158,
      "mean_token_accuracy": 0.10143255740404129,
      "num_tokens": 3356187.0,
      "step": 1545
    },
    {
      "entropy": 7.1759922981262205,
      "epoch": 0.16587297340681684,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999608380341533,
      "loss": 7.1422,
      "mean_token_accuracy": 0.10127759128808975,
      "num_tokens": 3367218.0,
      "step": 1550
    },
    {
      "entropy": 7.264885663986206,
      "epoch": 0.16640804751458077,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999601214742829,
      "loss": 7.285,
      "mean_token_accuracy": 0.08926323279738427,
      "num_tokens": 3377557.0,
      "step": 1555
    },
    {
      "entropy": 7.265757751464844,
      "epoch": 0.1669431216223447,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999593984187206,
      "loss": 7.2802,
      "mean_token_accuracy": 0.09445410892367363,
      "num_tokens": 3387402.0,
      "step": 1560
    },
    {
      "entropy": 7.219360876083374,
      "epoch": 0.1674781957301086,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999586688674872,
      "loss": 7.1539,
      "mean_token_accuracy": 0.09557267501950265,
      "num_tokens": 3397924.0,
      "step": 1565
    },
    {
      "entropy": 7.275343990325927,
      "epoch": 0.16801326983787254,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499957932820604,
      "loss": 7.2363,
      "mean_token_accuracy": 0.0918659195303917,
      "num_tokens": 3408622.0,
      "step": 1570
    },
    {
      "entropy": 7.273813772201538,
      "epoch": 0.16854834394563648,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499957190278092,
      "loss": 7.1887,
      "mean_token_accuracy": 0.09335600733757018,
      "num_tokens": 3419692.0,
      "step": 1575
    },
    {
      "entropy": 7.148430204391479,
      "epoch": 0.1690834180534004,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999564412399728,
      "loss": 7.2385,
      "mean_token_accuracy": 0.09515550211071969,
      "num_tokens": 3430986.0,
      "step": 1580
    },
    {
      "entropy": 7.262103271484375,
      "epoch": 0.1696184921611643,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499955685706268,
      "loss": 7.1655,
      "mean_token_accuracy": 0.09615519493818284,
      "num_tokens": 3442211.0,
      "step": 1585
    },
    {
      "entropy": 7.084865522384644,
      "epoch": 0.17015356626892825,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999549236769993,
      "loss": 7.144,
      "mean_token_accuracy": 0.10298331528902054,
      "num_tokens": 3453585.0,
      "step": 1590
    },
    {
      "entropy": 7.217924642562866,
      "epoch": 0.17068864037669218,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999541551521888,
      "loss": 7.1997,
      "mean_token_accuracy": 0.09542482867836952,
      "num_tokens": 3464156.0,
      "step": 1595
    },
    {
      "entropy": 7.321245956420898,
      "epoch": 0.17122371448445609,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999533801318587,
      "loss": 7.2598,
      "mean_token_accuracy": 0.0914094440639019,
      "num_tokens": 3474853.0,
      "step": 1600
    },
    {
      "entropy": 7.274269437789917,
      "epoch": 0.17175878859222002,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999525986160313,
      "loss": 7.2159,
      "mean_token_accuracy": 0.09993736371397972,
      "num_tokens": 3486841.0,
      "step": 1605
    },
    {
      "entropy": 7.151181650161743,
      "epoch": 0.17229386269998395,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999518106047293,
      "loss": 7.2369,
      "mean_token_accuracy": 0.09555753394961357,
      "num_tokens": 3498735.0,
      "step": 1610
    },
    {
      "entropy": 7.3050923347473145,
      "epoch": 0.17282893680774788,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004999510160979754,
      "loss": 7.2628,
      "mean_token_accuracy": 0.09753414988517761,
      "num_tokens": 3508829.0,
      "step": 1615
    },
    {
      "entropy": 7.1783294677734375,
      "epoch": 0.1733640109155118,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999502150957925,
      "loss": 7.1788,
      "mean_token_accuracy": 0.09674842134118081,
      "num_tokens": 3519465.0,
      "step": 1620
    },
    {
      "entropy": 7.303341913223266,
      "epoch": 0.17389908502327572,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999494075982037,
      "loss": 7.1351,
      "mean_token_accuracy": 0.10047532543540001,
      "num_tokens": 3530401.0,
      "step": 1625
    },
    {
      "entropy": 7.150378227233887,
      "epoch": 0.17443415913103966,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999485936052324,
      "loss": 7.214,
      "mean_token_accuracy": 0.09674938917160034,
      "num_tokens": 3541136.0,
      "step": 1630
    },
    {
      "entropy": 7.273062419891358,
      "epoch": 0.17496923323880356,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499947773116902,
      "loss": 7.1883,
      "mean_token_accuracy": 0.09701626598834992,
      "num_tokens": 3552525.0,
      "step": 1635
    },
    {
      "entropy": 7.137057590484619,
      "epoch": 0.1755043073465675,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999469461332365,
      "loss": 7.1541,
      "mean_token_accuracy": 0.09842078015208244,
      "num_tokens": 3563028.0,
      "step": 1640
    },
    {
      "entropy": 7.233504962921143,
      "epoch": 0.17603938145433143,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999461126542592,
      "loss": 7.108,
      "mean_token_accuracy": 0.1023821361362934,
      "num_tokens": 3573354.0,
      "step": 1645
    },
    {
      "entropy": 7.027111387252807,
      "epoch": 0.17657445556209536,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999452726799947,
      "loss": 7.1025,
      "mean_token_accuracy": 0.09993429109454155,
      "num_tokens": 3585576.0,
      "step": 1650
    },
    {
      "entropy": 7.312371683120728,
      "epoch": 0.17710952966985927,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999444262104671,
      "loss": 7.233,
      "mean_token_accuracy": 0.08933234214782715,
      "num_tokens": 3596478.0,
      "step": 1655
    },
    {
      "entropy": 7.192934226989746,
      "epoch": 0.1776446037776232,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999435732457007,
      "loss": 7.1824,
      "mean_token_accuracy": 0.09323065280914307,
      "num_tokens": 3608014.0,
      "step": 1660
    },
    {
      "entropy": 7.241013860702514,
      "epoch": 0.17817967788538713,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999427137857203,
      "loss": 7.2286,
      "mean_token_accuracy": 0.09294094443321228,
      "num_tokens": 3620120.0,
      "step": 1665
    },
    {
      "entropy": 7.1006693840026855,
      "epoch": 0.17871475199315104,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999418478305506,
      "loss": 7.1279,
      "mean_token_accuracy": 0.099704759567976,
      "num_tokens": 3632578.0,
      "step": 1670
    },
    {
      "entropy": 7.275180101394653,
      "epoch": 0.17924982610091497,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999409753802167,
      "loss": 7.1468,
      "mean_token_accuracy": 0.10483805015683174,
      "num_tokens": 3643245.0,
      "step": 1675
    },
    {
      "entropy": 7.124938106536865,
      "epoch": 0.1797849002086789,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999400964347437,
      "loss": 7.1487,
      "mean_token_accuracy": 0.09849804788827896,
      "num_tokens": 3655832.0,
      "step": 1680
    },
    {
      "entropy": 7.222979402542114,
      "epoch": 0.18031997431644284,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999392109941571,
      "loss": 7.1295,
      "mean_token_accuracy": 0.10116728618741036,
      "num_tokens": 3666566.0,
      "step": 1685
    },
    {
      "entropy": 7.194202947616577,
      "epoch": 0.18085504842420674,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999383190584822,
      "loss": 7.1611,
      "mean_token_accuracy": 0.10110493823885917,
      "num_tokens": 3677572.0,
      "step": 1690
    },
    {
      "entropy": 7.103847599029541,
      "epoch": 0.18139012253197068,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999374206277451,
      "loss": 7.0891,
      "mean_token_accuracy": 0.09825331270694733,
      "num_tokens": 3687116.0,
      "step": 1695
    },
    {
      "entropy": 7.121718025207519,
      "epoch": 0.1819251966397346,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999365157019717,
      "loss": 7.176,
      "mean_token_accuracy": 0.10047127306461334,
      "num_tokens": 3697961.0,
      "step": 1700
    },
    {
      "entropy": 7.246279668807984,
      "epoch": 0.18246027074749852,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999356042811878,
      "loss": 7.1016,
      "mean_token_accuracy": 0.10373674482107162,
      "num_tokens": 3708834.0,
      "step": 1705
    },
    {
      "entropy": 7.094297456741333,
      "epoch": 0.18299534485526245,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00049993468636542,
      "loss": 7.1584,
      "mean_token_accuracy": 0.09685484990477562,
      "num_tokens": 3719835.0,
      "step": 1710
    },
    {
      "entropy": 7.173304510116577,
      "epoch": 0.18353041896302638,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999337619546948,
      "loss": 7.0544,
      "mean_token_accuracy": 0.10510629564523696,
      "num_tokens": 3730408.0,
      "step": 1715
    },
    {
      "entropy": 7.064211463928222,
      "epoch": 0.18406549307079031,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999328310490387,
      "loss": 7.1446,
      "mean_token_accuracy": 0.10267272144556046,
      "num_tokens": 3741257.0,
      "step": 1720
    },
    {
      "entropy": 7.2027805805206295,
      "epoch": 0.18460056717855422,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999318936484789,
      "loss": 7.1282,
      "mean_token_accuracy": 0.10227805599570275,
      "num_tokens": 3752615.0,
      "step": 1725
    },
    {
      "entropy": 7.191047954559326,
      "epoch": 0.18513564128631815,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999309497530422,
      "loss": 7.1568,
      "mean_token_accuracy": 0.0937609352171421,
      "num_tokens": 3763656.0,
      "step": 1730
    },
    {
      "entropy": 7.051603937149048,
      "epoch": 0.1856707153940821,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999299993627558,
      "loss": 7.1315,
      "mean_token_accuracy": 0.10040904507040978,
      "num_tokens": 3775257.0,
      "step": 1735
    },
    {
      "entropy": 7.149741744995117,
      "epoch": 0.186205789501846,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999290424776475,
      "loss": 7.0984,
      "mean_token_accuracy": 0.10568056777119636,
      "num_tokens": 3786094.0,
      "step": 1740
    },
    {
      "entropy": 7.321368026733398,
      "epoch": 0.18674086360960993,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999280790977445,
      "loss": 7.1886,
      "mean_token_accuracy": 0.09335805177688598,
      "num_tokens": 3796672.0,
      "step": 1745
    },
    {
      "entropy": 7.137669038772583,
      "epoch": 0.18727593771737386,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499927109223075,
      "loss": 7.1558,
      "mean_token_accuracy": 0.10123137831687927,
      "num_tokens": 3807910.0,
      "step": 1750
    },
    {
      "entropy": 7.183064413070679,
      "epoch": 0.1878110118251378,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999261328536667,
      "loss": 7.1356,
      "mean_token_accuracy": 0.09966104477643967,
      "num_tokens": 3820792.0,
      "step": 1755
    },
    {
      "entropy": 7.04319133758545,
      "epoch": 0.1883460859329017,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999251499895479,
      "loss": 7.038,
      "mean_token_accuracy": 0.10163617879152298,
      "num_tokens": 3831126.0,
      "step": 1760
    },
    {
      "entropy": 7.234713125228882,
      "epoch": 0.18888116004066563,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999241606307471,
      "loss": 7.1611,
      "mean_token_accuracy": 0.09807982221245766,
      "num_tokens": 3841923.0,
      "step": 1765
    },
    {
      "entropy": 7.183645105361938,
      "epoch": 0.18941623414842956,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999231647772927,
      "loss": 7.0516,
      "mean_token_accuracy": 0.1046050101518631,
      "num_tokens": 3852504.0,
      "step": 1770
    },
    {
      "entropy": 6.985522317886352,
      "epoch": 0.1899513082561935,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999221624292136,
      "loss": 7.0732,
      "mean_token_accuracy": 0.10412966907024383,
      "num_tokens": 3862761.0,
      "step": 1775
    },
    {
      "entropy": 7.208767414093018,
      "epoch": 0.1904863823639574,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999211535865385,
      "loss": 7.0701,
      "mean_token_accuracy": 0.09672508612275124,
      "num_tokens": 3873662.0,
      "step": 1780
    },
    {
      "entropy": 7.0239424228668215,
      "epoch": 0.19102145647172133,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999201382492969,
      "loss": 7.0227,
      "mean_token_accuracy": 0.10187353417277337,
      "num_tokens": 3884835.0,
      "step": 1785
    },
    {
      "entropy": 7.158309888839722,
      "epoch": 0.19155653057948527,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999191164175178,
      "loss": 7.1636,
      "mean_token_accuracy": 0.09919204637408256,
      "num_tokens": 3895073.0,
      "step": 1790
    },
    {
      "entropy": 7.206078290939331,
      "epoch": 0.19209160468724917,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999180880912309,
      "loss": 7.1267,
      "mean_token_accuracy": 0.09839320108294487,
      "num_tokens": 3905440.0,
      "step": 1795
    },
    {
      "entropy": 7.085821580886841,
      "epoch": 0.1926266787950131,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999170532704657,
      "loss": 7.125,
      "mean_token_accuracy": 0.09741980135440827,
      "num_tokens": 3917135.0,
      "step": 1800
    },
    {
      "entropy": 7.156965684890747,
      "epoch": 0.19316175290277704,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999160119552523,
      "loss": 7.1303,
      "mean_token_accuracy": 0.10348564088344574,
      "num_tokens": 3928247.0,
      "step": 1805
    },
    {
      "entropy": 7.109297609329223,
      "epoch": 0.19369682701054097,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999149641456206,
      "loss": 7.1139,
      "mean_token_accuracy": 0.09793074429035187,
      "num_tokens": 3938609.0,
      "step": 1810
    },
    {
      "entropy": 7.246528387069702,
      "epoch": 0.19423190111830488,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499913909841601,
      "loss": 7.098,
      "mean_token_accuracy": 0.10168406963348389,
      "num_tokens": 3949221.0,
      "step": 1815
    },
    {
      "entropy": 6.995891475677491,
      "epoch": 0.1947669752260688,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999128490432238,
      "loss": 7.0711,
      "mean_token_accuracy": 0.09396483302116394,
      "num_tokens": 3959475.0,
      "step": 1820
    },
    {
      "entropy": 7.154359769821167,
      "epoch": 0.19530204933383274,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999117817505196,
      "loss": 7.0738,
      "mean_token_accuracy": 0.11302891224622727,
      "num_tokens": 3969638.0,
      "step": 1825
    },
    {
      "entropy": 7.10610032081604,
      "epoch": 0.19583712344159665,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999107079635194,
      "loss": 7.123,
      "mean_token_accuracy": 0.10322771817445756,
      "num_tokens": 3980543.0,
      "step": 1830
    },
    {
      "entropy": 7.329257583618164,
      "epoch": 0.19637219754936058,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499909627682254,
      "loss": 7.1756,
      "mean_token_accuracy": 0.08275276124477386,
      "num_tokens": 3993518.0,
      "step": 1835
    },
    {
      "entropy": 7.120573711395264,
      "epoch": 0.19690727165712452,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999085409067549,
      "loss": 7.155,
      "mean_token_accuracy": 0.10039021968841552,
      "num_tokens": 4002940.0,
      "step": 1840
    },
    {
      "entropy": 7.034014892578125,
      "epoch": 0.19744234576488845,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999074476370531,
      "loss": 7.0947,
      "mean_token_accuracy": 0.10023277997970581,
      "num_tokens": 4013982.0,
      "step": 1845
    },
    {
      "entropy": 7.101888799667359,
      "epoch": 0.19797741987265235,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999063478731805,
      "loss": 7.0607,
      "mean_token_accuracy": 0.10425660833716392,
      "num_tokens": 4025607.0,
      "step": 1850
    },
    {
      "entropy": 7.142320871353149,
      "epoch": 0.1985124939804163,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999052416151687,
      "loss": 7.1317,
      "mean_token_accuracy": 0.09948315396904946,
      "num_tokens": 4037467.0,
      "step": 1855
    },
    {
      "entropy": 7.155796337127685,
      "epoch": 0.19904756808818022,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999041288630496,
      "loss": 7.1317,
      "mean_token_accuracy": 0.10021365135908127,
      "num_tokens": 4049146.0,
      "step": 1860
    },
    {
      "entropy": 7.215651226043701,
      "epoch": 0.19958264219594413,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999030096168553,
      "loss": 7.1688,
      "mean_token_accuracy": 0.09719501882791519,
      "num_tokens": 4060500.0,
      "step": 1865
    },
    {
      "entropy": 7.176360416412353,
      "epoch": 0.20011771630370806,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999018838766182,
      "loss": 7.121,
      "mean_token_accuracy": 0.10209911167621613,
      "num_tokens": 4070553.0,
      "step": 1870
    },
    {
      "entropy": 7.173916673660278,
      "epoch": 0.200652790411472,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499900751642371,
      "loss": 7.0344,
      "mean_token_accuracy": 0.11530624777078628,
      "num_tokens": 4079653.0,
      "step": 1875
    },
    {
      "entropy": 7.061277198791504,
      "epoch": 0.20118786451923593,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998996129141461,
      "loss": 7.135,
      "mean_token_accuracy": 0.09668742194771766,
      "num_tokens": 4089912.0,
      "step": 1880
    },
    {
      "entropy": 7.230644083023071,
      "epoch": 0.20172293862699983,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998984676919764,
      "loss": 7.0917,
      "mean_token_accuracy": 0.0961770847439766,
      "num_tokens": 4100577.0,
      "step": 1885
    },
    {
      "entropy": 7.1247584342956545,
      "epoch": 0.20225801273476376,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998973159758952,
      "loss": 7.1167,
      "mean_token_accuracy": 0.10018153339624405,
      "num_tokens": 4111752.0,
      "step": 1890
    },
    {
      "entropy": 7.063771104812622,
      "epoch": 0.2027930868425277,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998961577659355,
      "loss": 7.0671,
      "mean_token_accuracy": 0.10595502480864524,
      "num_tokens": 4122616.0,
      "step": 1895
    },
    {
      "entropy": 7.2158653259277346,
      "epoch": 0.2033281609502916,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998949930621309,
      "loss": 7.1747,
      "mean_token_accuracy": 0.1001809298992157,
      "num_tokens": 4134275.0,
      "step": 1900
    },
    {
      "entropy": 7.102860450744629,
      "epoch": 0.20386323505805554,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998938218645149,
      "loss": 7.0642,
      "mean_token_accuracy": 0.10334794074296952,
      "num_tokens": 4146412.0,
      "step": 1905
    },
    {
      "entropy": 7.073519086837768,
      "epoch": 0.20439830916581947,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998926441731215,
      "loss": 7.0558,
      "mean_token_accuracy": 0.10119741782546043,
      "num_tokens": 4157592.0,
      "step": 1910
    },
    {
      "entropy": 7.074370908737182,
      "epoch": 0.2049333832735834,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998914599879845,
      "loss": 7.0511,
      "mean_token_accuracy": 0.10084806829690933,
      "num_tokens": 4168637.0,
      "step": 1915
    },
    {
      "entropy": 7.115645503997802,
      "epoch": 0.2054684573813473,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998902693091382,
      "loss": 7.0857,
      "mean_token_accuracy": 0.09849425479769706,
      "num_tokens": 4179951.0,
      "step": 1920
    },
    {
      "entropy": 7.090283823013306,
      "epoch": 0.20600353148911124,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998890721366171,
      "loss": 6.9867,
      "mean_token_accuracy": 0.10344830304384231,
      "num_tokens": 4189895.0,
      "step": 1925
    },
    {
      "entropy": 7.081069898605347,
      "epoch": 0.20653860559687517,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998878684704556,
      "loss": 7.0339,
      "mean_token_accuracy": 0.10446125566959381,
      "num_tokens": 4199796.0,
      "step": 1930
    },
    {
      "entropy": 7.079880237579346,
      "epoch": 0.20707367970463908,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998866583106884,
      "loss": 7.0653,
      "mean_token_accuracy": 0.10093458741903305,
      "num_tokens": 4210428.0,
      "step": 1935
    },
    {
      "entropy": 7.160430908203125,
      "epoch": 0.207608753812403,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998854416573506,
      "loss": 7.1286,
      "mean_token_accuracy": 0.10443870946764947,
      "num_tokens": 4221515.0,
      "step": 1940
    },
    {
      "entropy": 7.142211771011352,
      "epoch": 0.20814382792016695,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998842185104772,
      "loss": 7.1474,
      "mean_token_accuracy": 0.0981269545853138,
      "num_tokens": 4232580.0,
      "step": 1945
    },
    {
      "entropy": 7.054647779464721,
      "epoch": 0.20867890202793088,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998829888701036,
      "loss": 7.0565,
      "mean_token_accuracy": 0.10753207132220269,
      "num_tokens": 4242871.0,
      "step": 1950
    },
    {
      "entropy": 7.156534862518311,
      "epoch": 0.20921397613569478,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998817527362654,
      "loss": 7.0786,
      "mean_token_accuracy": 0.09486880823969841,
      "num_tokens": 4253419.0,
      "step": 1955
    },
    {
      "entropy": 7.098293828964233,
      "epoch": 0.20974905024345872,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998805101089982,
      "loss": 7.0551,
      "mean_token_accuracy": 0.10472317561507224,
      "num_tokens": 4264327.0,
      "step": 1960
    },
    {
      "entropy": 7.143654727935791,
      "epoch": 0.21028412435122265,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998792609883378,
      "loss": 7.0423,
      "mean_token_accuracy": 0.10480301082134247,
      "num_tokens": 4274568.0,
      "step": 1965
    },
    {
      "entropy": 7.062982511520386,
      "epoch": 0.21081919845898656,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998780053743203,
      "loss": 7.0376,
      "mean_token_accuracy": 0.10589545667171478,
      "num_tokens": 4284895.0,
      "step": 1970
    },
    {
      "entropy": 7.148151302337647,
      "epoch": 0.2113542725667505,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004998767432669822,
      "loss": 7.0787,
      "mean_token_accuracy": 0.102964149415493,
      "num_tokens": 4297258.0,
      "step": 1975
    },
    {
      "entropy": 7.025489664077758,
      "epoch": 0.21188934667451442,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998754746663595,
      "loss": 6.9864,
      "mean_token_accuracy": 0.10545038804411888,
      "num_tokens": 4308307.0,
      "step": 1980
    },
    {
      "entropy": 7.0815025806427006,
      "epoch": 0.21242442078227836,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998741995724892,
      "loss": 7.0615,
      "mean_token_accuracy": 0.10010265707969665,
      "num_tokens": 4318801.0,
      "step": 1985
    },
    {
      "entropy": 7.061044788360595,
      "epoch": 0.21295949489004226,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998729179854079,
      "loss": 7.0328,
      "mean_token_accuracy": 0.1036236509680748,
      "num_tokens": 4329531.0,
      "step": 1990
    },
    {
      "entropy": 7.0804825782775875,
      "epoch": 0.2134945689978062,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998716299051527,
      "loss": 7.0256,
      "mean_token_accuracy": 0.10520645901560784,
      "num_tokens": 4341203.0,
      "step": 1995
    },
    {
      "entropy": 7.035392093658447,
      "epoch": 0.21402964310557013,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998703353317609,
      "loss": 7.0754,
      "mean_token_accuracy": 0.09851410016417503,
      "num_tokens": 4352426.0,
      "step": 2000
    },
    {
      "entropy": 7.181427955627441,
      "epoch": 0.21456471721333403,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998690342652697,
      "loss": 7.0155,
      "mean_token_accuracy": 0.10717839673161507,
      "num_tokens": 4363413.0,
      "step": 2005
    },
    {
      "entropy": 6.992197751998901,
      "epoch": 0.21509979132109797,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004998677267057166,
      "loss": 6.9503,
      "mean_token_accuracy": 0.10650679394602776,
      "num_tokens": 4373470.0,
      "step": 2010
    },
    {
      "entropy": 7.0062854290008545,
      "epoch": 0.2156348654288619,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998664126531397,
      "loss": 7.0237,
      "mean_token_accuracy": 0.10450349599123002,
      "num_tokens": 4384438.0,
      "step": 2015
    },
    {
      "entropy": 7.172478914260864,
      "epoch": 0.21616993953662583,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998650921075766,
      "loss": 6.9991,
      "mean_token_accuracy": 0.10896808505058289,
      "num_tokens": 4395100.0,
      "step": 2020
    },
    {
      "entropy": 7.004340648651123,
      "epoch": 0.21670501364438974,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998637650690656,
      "loss": 7.1196,
      "mean_token_accuracy": 0.09563550055027008,
      "num_tokens": 4407131.0,
      "step": 2025
    },
    {
      "entropy": 7.083667659759522,
      "epoch": 0.21724008775215367,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499862431537645,
      "loss": 6.9577,
      "mean_token_accuracy": 0.10268971174955369,
      "num_tokens": 4417701.0,
      "step": 2030
    },
    {
      "entropy": 7.023851299285889,
      "epoch": 0.2177751618599176,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998610915133533,
      "loss": 7.0216,
      "mean_token_accuracy": 0.10145122185349464,
      "num_tokens": 4429309.0,
      "step": 2035
    },
    {
      "entropy": 7.0214043140411375,
      "epoch": 0.21831023596768154,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998597449962292,
      "loss": 7.0769,
      "mean_token_accuracy": 0.10378911569714547,
      "num_tokens": 4440850.0,
      "step": 2040
    },
    {
      "entropy": 7.054154539108277,
      "epoch": 0.21884531007544544,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998583919863115,
      "loss": 7.054,
      "mean_token_accuracy": 0.09939034059643745,
      "num_tokens": 4451995.0,
      "step": 2045
    },
    {
      "entropy": 7.043123245239258,
      "epoch": 0.21938038418320938,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998570324836393,
      "loss": 6.972,
      "mean_token_accuracy": 0.10836615711450577,
      "num_tokens": 4461916.0,
      "step": 2050
    },
    {
      "entropy": 7.089596509933472,
      "epoch": 0.2199154582909733,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499855666488252,
      "loss": 7.0501,
      "mean_token_accuracy": 0.10234842598438262,
      "num_tokens": 4471603.0,
      "step": 2055
    },
    {
      "entropy": 7.144379091262818,
      "epoch": 0.22045053239873721,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998542940001889,
      "loss": 7.0775,
      "mean_token_accuracy": 0.10257849842309952,
      "num_tokens": 4483166.0,
      "step": 2060
    },
    {
      "entropy": 7.052639961242676,
      "epoch": 0.22098560650650115,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998529150194895,
      "loss": 7.0006,
      "mean_token_accuracy": 0.10316628888249398,
      "num_tokens": 4493824.0,
      "step": 2065
    },
    {
      "entropy": 7.0311055183410645,
      "epoch": 0.22152068061426508,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998515295461939,
      "loss": 6.9058,
      "mean_token_accuracy": 0.12331796064972878,
      "num_tokens": 4505674.0,
      "step": 2070
    },
    {
      "entropy": 6.943395900726318,
      "epoch": 0.22205575472202901,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499850137580342,
      "loss": 7.0,
      "mean_token_accuracy": 0.10888025835156441,
      "num_tokens": 4516516.0,
      "step": 2075
    },
    {
      "entropy": 7.085627651214599,
      "epoch": 0.22259082882979292,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998487391219739,
      "loss": 6.9976,
      "mean_token_accuracy": 0.10101775825023651,
      "num_tokens": 4527811.0,
      "step": 2080
    },
    {
      "entropy": 7.010357475280761,
      "epoch": 0.22312590293755685,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998473341711301,
      "loss": 6.9383,
      "mean_token_accuracy": 0.10195653289556503,
      "num_tokens": 4538988.0,
      "step": 2085
    },
    {
      "entropy": 7.025439023971558,
      "epoch": 0.22366097704532079,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998459227278512,
      "loss": 6.9788,
      "mean_token_accuracy": 0.10650492385029793,
      "num_tokens": 4550049.0,
      "step": 2090
    },
    {
      "entropy": 7.1111595153808596,
      "epoch": 0.2241960511530847,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998445047921777,
      "loss": 7.0758,
      "mean_token_accuracy": 0.10203522294759751,
      "num_tokens": 4559784.0,
      "step": 2095
    },
    {
      "entropy": 7.0501655578613285,
      "epoch": 0.22473112526084862,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998430803641509,
      "loss": 6.9376,
      "mean_token_accuracy": 0.10454667881131172,
      "num_tokens": 4569982.0,
      "step": 2100
    },
    {
      "entropy": 6.932292032241821,
      "epoch": 0.22526619936861256,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998416494438118,
      "loss": 6.9489,
      "mean_token_accuracy": 0.10733313858509064,
      "num_tokens": 4581117.0,
      "step": 2105
    },
    {
      "entropy": 7.042090797424317,
      "epoch": 0.2258012734763765,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998402120312014,
      "loss": 7.0174,
      "mean_token_accuracy": 0.10961635783314705,
      "num_tokens": 4592012.0,
      "step": 2110
    },
    {
      "entropy": 7.025863790512085,
      "epoch": 0.2263363475841404,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998387681263616,
      "loss": 7.0634,
      "mean_token_accuracy": 0.10254996567964554,
      "num_tokens": 4602591.0,
      "step": 2115
    },
    {
      "entropy": 7.054075098037719,
      "epoch": 0.22687142169190433,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499837317729334,
      "loss": 6.9882,
      "mean_token_accuracy": 0.10465681180357933,
      "num_tokens": 4613418.0,
      "step": 2120
    },
    {
      "entropy": 6.957354021072388,
      "epoch": 0.22740649579966826,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004998358608401603,
      "loss": 6.9169,
      "mean_token_accuracy": 0.1147283211350441,
      "num_tokens": 4625156.0,
      "step": 2125
    },
    {
      "entropy": 7.111089897155762,
      "epoch": 0.22794156990743217,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499834397458883,
      "loss": 6.9472,
      "mean_token_accuracy": 0.10765151679515839,
      "num_tokens": 4634523.0,
      "step": 2130
    },
    {
      "entropy": 6.936869287490845,
      "epoch": 0.2284766440151961,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998329275855438,
      "loss": 6.9341,
      "mean_token_accuracy": 0.11054303497076035,
      "num_tokens": 4646292.0,
      "step": 2135
    },
    {
      "entropy": 7.02635498046875,
      "epoch": 0.22901171812296003,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998314512201855,
      "loss": 6.9176,
      "mean_token_accuracy": 0.10452343374490738,
      "num_tokens": 4656417.0,
      "step": 2140
    },
    {
      "entropy": 6.944140291213989,
      "epoch": 0.22954679223072397,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998299683628507,
      "loss": 6.9361,
      "mean_token_accuracy": 0.10723940208554268,
      "num_tokens": 4666933.0,
      "step": 2145
    },
    {
      "entropy": 7.071186447143555,
      "epoch": 0.23008186633848787,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998284790135821,
      "loss": 7.1086,
      "mean_token_accuracy": 0.09997956231236457,
      "num_tokens": 4678482.0,
      "step": 2150
    },
    {
      "entropy": 7.071504402160644,
      "epoch": 0.2306169404462518,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998269831724228,
      "loss": 6.943,
      "mean_token_accuracy": 0.1055432416498661,
      "num_tokens": 4688454.0,
      "step": 2155
    },
    {
      "entropy": 6.989952278137207,
      "epoch": 0.23115201455401574,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998254808394159,
      "loss": 6.9886,
      "mean_token_accuracy": 0.10683377385139466,
      "num_tokens": 4699545.0,
      "step": 2160
    },
    {
      "entropy": 7.105897760391235,
      "epoch": 0.23168708866177964,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004998239720146048,
      "loss": 7.0006,
      "mean_token_accuracy": 0.10335398241877555,
      "num_tokens": 4711149.0,
      "step": 2165
    },
    {
      "entropy": 6.922723054885864,
      "epoch": 0.23222216276954358,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998224566980332,
      "loss": 6.996,
      "mean_token_accuracy": 0.10218452662229538,
      "num_tokens": 4722651.0,
      "step": 2170
    },
    {
      "entropy": 7.068100786209106,
      "epoch": 0.2327572368773075,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998209348897447,
      "loss": 6.9957,
      "mean_token_accuracy": 0.09393229782581329,
      "num_tokens": 4733883.0,
      "step": 2175
    },
    {
      "entropy": 7.1184226989746096,
      "epoch": 0.23329231098507144,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998194065897833,
      "loss": 6.9513,
      "mean_token_accuracy": 0.11117385998368264,
      "num_tokens": 4744517.0,
      "step": 2180
    },
    {
      "entropy": 6.966147184371948,
      "epoch": 0.23382738509283535,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499817871798193,
      "loss": 6.9746,
      "mean_token_accuracy": 0.11019051298499108,
      "num_tokens": 4755481.0,
      "step": 2185
    },
    {
      "entropy": 6.96983757019043,
      "epoch": 0.23436245920059928,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998163305150185,
      "loss": 6.9042,
      "mean_token_accuracy": 0.11359247267246246,
      "num_tokens": 4765535.0,
      "step": 2190
    },
    {
      "entropy": 7.0188885688781735,
      "epoch": 0.23489753330836322,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998147827403038,
      "loss": 7.0168,
      "mean_token_accuracy": 0.10479639694094658,
      "num_tokens": 4777045.0,
      "step": 2195
    },
    {
      "entropy": 7.045113372802734,
      "epoch": 0.23543260741612712,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499813228474094,
      "loss": 6.9067,
      "mean_token_accuracy": 0.11008131876587868,
      "num_tokens": 4787958.0,
      "step": 2200
    },
    {
      "entropy": 6.917256498336792,
      "epoch": 0.23596768152389105,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998116677164338,
      "loss": 6.9399,
      "mean_token_accuracy": 0.1049346685409546,
      "num_tokens": 4798502.0,
      "step": 2205
    },
    {
      "entropy": 6.958500146865845,
      "epoch": 0.236502755631655,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998101004673682,
      "loss": 6.9403,
      "mean_token_accuracy": 0.1110065832734108,
      "num_tokens": 4809179.0,
      "step": 2210
    },
    {
      "entropy": 7.1306075096130375,
      "epoch": 0.23703782973941892,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998085267269426,
      "loss": 7.0169,
      "mean_token_accuracy": 0.09675444290041924,
      "num_tokens": 4819803.0,
      "step": 2215
    },
    {
      "entropy": 7.033319044113159,
      "epoch": 0.23757290384718283,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998069464952025,
      "loss": 6.9958,
      "mean_token_accuracy": 0.1008896566927433,
      "num_tokens": 4831226.0,
      "step": 2220
    },
    {
      "entropy": 7.051976871490479,
      "epoch": 0.23810797795494676,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998053597721932,
      "loss": 7.0115,
      "mean_token_accuracy": 0.10829942151904107,
      "num_tokens": 4842168.0,
      "step": 2225
    },
    {
      "entropy": 7.034261083602905,
      "epoch": 0.2386430520627107,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998037665579608,
      "loss": 6.9697,
      "mean_token_accuracy": 0.10675674006342888,
      "num_tokens": 4852655.0,
      "step": 2230
    },
    {
      "entropy": 6.881174850463867,
      "epoch": 0.2391781261704746,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998021668525513,
      "loss": 6.8537,
      "mean_token_accuracy": 0.11227660775184631,
      "num_tokens": 4862164.0,
      "step": 2235
    },
    {
      "entropy": 7.07776198387146,
      "epoch": 0.23971320027823853,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998005606560107,
      "loss": 7.0126,
      "mean_token_accuracy": 0.11018440797924996,
      "num_tokens": 4874233.0,
      "step": 2240
    },
    {
      "entropy": 6.984408712387085,
      "epoch": 0.24024827438600246,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997989479683856,
      "loss": 6.9927,
      "mean_token_accuracy": 0.10390212759375572,
      "num_tokens": 4885948.0,
      "step": 2245
    },
    {
      "entropy": 6.981641530990601,
      "epoch": 0.2407833484937664,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997973287897224,
      "loss": 6.9739,
      "mean_token_accuracy": 0.10715372562408447,
      "num_tokens": 4897128.0,
      "step": 2250
    },
    {
      "entropy": 7.011375617980957,
      "epoch": 0.2413184226015303,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997957031200681,
      "loss": 6.939,
      "mean_token_accuracy": 0.11313669458031654,
      "num_tokens": 4907699.0,
      "step": 2255
    },
    {
      "entropy": 6.93510890007019,
      "epoch": 0.24185349670929424,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997940709594692,
      "loss": 6.875,
      "mean_token_accuracy": 0.10799874365329742,
      "num_tokens": 4918772.0,
      "step": 2260
    },
    {
      "entropy": 7.044176292419434,
      "epoch": 0.24238857081705817,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997924323079733,
      "loss": 6.9352,
      "mean_token_accuracy": 0.10844795927405357,
      "num_tokens": 4930046.0,
      "step": 2265
    },
    {
      "entropy": 6.897712516784668,
      "epoch": 0.24292364492482207,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997907871656273,
      "loss": 6.9542,
      "mean_token_accuracy": 0.10848618820309638,
      "num_tokens": 4940609.0,
      "step": 2270
    },
    {
      "entropy": 7.033410692214966,
      "epoch": 0.243458719032586,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997891355324791,
      "loss": 6.9538,
      "mean_token_accuracy": 0.10902024582028388,
      "num_tokens": 4951339.0,
      "step": 2275
    },
    {
      "entropy": 6.970147943496704,
      "epoch": 0.24399379314034994,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997874774085761,
      "loss": 6.8889,
      "mean_token_accuracy": 0.10065512731671333,
      "num_tokens": 4961083.0,
      "step": 2280
    },
    {
      "entropy": 6.938036298751831,
      "epoch": 0.24452886724811387,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997858127939662,
      "loss": 6.8926,
      "mean_token_accuracy": 0.10896471291780471,
      "num_tokens": 4971399.0,
      "step": 2285
    },
    {
      "entropy": 6.986947107315063,
      "epoch": 0.24506394135587778,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997841416886976,
      "loss": 6.9635,
      "mean_token_accuracy": 0.1085882268846035,
      "num_tokens": 4982429.0,
      "step": 2290
    },
    {
      "entropy": 7.061171579360962,
      "epoch": 0.2455990154636417,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997824640928186,
      "loss": 7.0195,
      "mean_token_accuracy": 0.10628353208303451,
      "num_tokens": 4993706.0,
      "step": 2295
    },
    {
      "entropy": 6.986022043228149,
      "epoch": 0.24613408957140565,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997807800063774,
      "loss": 6.9109,
      "mean_token_accuracy": 0.11161794662475585,
      "num_tokens": 5005411.0,
      "step": 2300
    },
    {
      "entropy": 7.0011697769165036,
      "epoch": 0.24666916367916958,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499779089429423,
      "loss": 7.0298,
      "mean_token_accuracy": 0.10325584262609482,
      "num_tokens": 5016011.0,
      "step": 2305
    },
    {
      "entropy": 7.0111462593078615,
      "epoch": 0.24720423778693348,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004997773923620037,
      "loss": 6.9998,
      "mean_token_accuracy": 0.10303730890154839,
      "num_tokens": 5027841.0,
      "step": 2310
    },
    {
      "entropy": 7.078068637847901,
      "epoch": 0.24773931189469742,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997756888041689,
      "loss": 6.9792,
      "mean_token_accuracy": 0.09846872761845589,
      "num_tokens": 5038640.0,
      "step": 2315
    },
    {
      "entropy": 6.975053834915161,
      "epoch": 0.24827438600246135,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997739787559677,
      "loss": 6.9227,
      "mean_token_accuracy": 0.10773401260375977,
      "num_tokens": 5050423.0,
      "step": 2320
    },
    {
      "entropy": 6.994720554351806,
      "epoch": 0.24880946011022526,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997722622174494,
      "loss": 6.9495,
      "mean_token_accuracy": 0.10427153035998345,
      "num_tokens": 5060654.0,
      "step": 2325
    },
    {
      "entropy": 6.94322247505188,
      "epoch": 0.2493445342179892,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997705391886636,
      "loss": 6.8533,
      "mean_token_accuracy": 0.11165059134364128,
      "num_tokens": 5070165.0,
      "step": 2330
    },
    {
      "entropy": 6.8971030712127686,
      "epoch": 0.24987960832575312,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997688096696601,
      "loss": 6.9014,
      "mean_token_accuracy": 0.10189634934067726,
      "num_tokens": 5080937.0,
      "step": 2335
    },
    {
      "entropy": 6.940387964248657,
      "epoch": 0.25041468243351706,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997670736604888,
      "loss": 6.8601,
      "mean_token_accuracy": 0.11307725235819817,
      "num_tokens": 5090955.0,
      "step": 2340
    },
    {
      "entropy": 7.025439310073852,
      "epoch": 0.250949756541281,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004997653311611998,
      "loss": 7.012,
      "mean_token_accuracy": 0.1010633334517479,
      "num_tokens": 5102201.0,
      "step": 2345
    },
    {
      "entropy": 6.99841890335083,
      "epoch": 0.25148483064904487,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997635821718435,
      "loss": 6.935,
      "mean_token_accuracy": 0.11045385003089905,
      "num_tokens": 5112235.0,
      "step": 2350
    },
    {
      "entropy": 6.925424194335937,
      "epoch": 0.2520199047568088,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997618266924704,
      "loss": 6.7969,
      "mean_token_accuracy": 0.10532390475273132,
      "num_tokens": 5122360.0,
      "step": 2355
    },
    {
      "entropy": 6.942410039901733,
      "epoch": 0.25255497886457273,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499760064723131,
      "loss": 6.9771,
      "mean_token_accuracy": 0.10276588127017021,
      "num_tokens": 5133084.0,
      "step": 2360
    },
    {
      "entropy": 6.947847557067871,
      "epoch": 0.25309005297233667,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997582962638764,
      "loss": 6.9291,
      "mean_token_accuracy": 0.10205074176192283,
      "num_tokens": 5144495.0,
      "step": 2365
    },
    {
      "entropy": 7.063449239730835,
      "epoch": 0.2536251270801006,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997565213147575,
      "loss": 6.9216,
      "mean_token_accuracy": 0.11108831390738487,
      "num_tokens": 5154824.0,
      "step": 2370
    },
    {
      "entropy": 6.933940172195435,
      "epoch": 0.25416020118786453,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997547398758257,
      "loss": 6.9012,
      "mean_token_accuracy": 0.10534899979829788,
      "num_tokens": 5165061.0,
      "step": 2375
    },
    {
      "entropy": 6.941448068618774,
      "epoch": 0.25469527529562846,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997529519471323,
      "loss": 6.8652,
      "mean_token_accuracy": 0.11675361394882203,
      "num_tokens": 5174964.0,
      "step": 2380
    },
    {
      "entropy": 6.894770765304566,
      "epoch": 0.25523034940339234,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997511575287291,
      "loss": 6.9139,
      "mean_token_accuracy": 0.10526548847556114,
      "num_tokens": 5185458.0,
      "step": 2385
    },
    {
      "entropy": 7.003988265991211,
      "epoch": 0.2557654235111563,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997493566206677,
      "loss": 6.8953,
      "mean_token_accuracy": 0.10926351025700569,
      "num_tokens": 5196193.0,
      "step": 2390
    },
    {
      "entropy": 6.910524320602417,
      "epoch": 0.2563004976189202,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004997475492230003,
      "loss": 7.0084,
      "mean_token_accuracy": 0.09937358275055885,
      "num_tokens": 5208541.0,
      "step": 2395
    },
    {
      "entropy": 7.02528567314148,
      "epoch": 0.25683557172668414,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499745735335779,
      "loss": 6.8456,
      "mean_token_accuracy": 0.11003420725464821,
      "num_tokens": 5218567.0,
      "step": 2400
    },
    {
      "entropy": 6.922707414627075,
      "epoch": 0.2573706458344481,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997439149590561,
      "loss": 6.9484,
      "mean_token_accuracy": 0.10656592324376106,
      "num_tokens": 5228293.0,
      "step": 2405
    },
    {
      "entropy": 6.966908025741577,
      "epoch": 0.257905719942212,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997420880928843,
      "loss": 6.8958,
      "mean_token_accuracy": 0.11399568244814873,
      "num_tokens": 5238880.0,
      "step": 2410
    },
    {
      "entropy": 6.962517213821411,
      "epoch": 0.25844079404997594,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997402547373162,
      "loss": 6.9943,
      "mean_token_accuracy": 0.10480297654867173,
      "num_tokens": 5250104.0,
      "step": 2415
    },
    {
      "entropy": 7.100482559204101,
      "epoch": 0.2589758681577399,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004997384148924049,
      "loss": 7.011,
      "mean_token_accuracy": 0.10677094757556915,
      "num_tokens": 5260355.0,
      "step": 2420
    },
    {
      "entropy": 6.962700939178466,
      "epoch": 0.25951094226550375,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997365685582035,
      "loss": 6.929,
      "mean_token_accuracy": 0.1068173035979271,
      "num_tokens": 5272117.0,
      "step": 2425
    },
    {
      "entropy": 7.024838829040528,
      "epoch": 0.2600460163732677,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004997347157347651,
      "loss": 7.0198,
      "mean_token_accuracy": 0.09878586754202842,
      "num_tokens": 5284583.0,
      "step": 2430
    },
    {
      "entropy": 6.973494672775269,
      "epoch": 0.2605810904810316,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004997328564221435,
      "loss": 6.8048,
      "mean_token_accuracy": 0.11668524593114853,
      "num_tokens": 5295846.0,
      "step": 2435
    },
    {
      "entropy": 6.864496850967408,
      "epoch": 0.26111616458879555,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997309906203922,
      "loss": 6.9651,
      "mean_token_accuracy": 0.10438142120838165,
      "num_tokens": 5307802.0,
      "step": 2440
    },
    {
      "entropy": 7.095636558532715,
      "epoch": 0.2616512386965595,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997291183295652,
      "loss": 6.8785,
      "mean_token_accuracy": 0.11019614040851593,
      "num_tokens": 5318263.0,
      "step": 2445
    },
    {
      "entropy": 6.985703229904175,
      "epoch": 0.2621863128043234,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997272395497164,
      "loss": 6.9369,
      "mean_token_accuracy": 0.10467563569545746,
      "num_tokens": 5328555.0,
      "step": 2450
    },
    {
      "entropy": 6.877440309524536,
      "epoch": 0.26272138691208735,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004997253542809002,
      "loss": 6.8952,
      "mean_token_accuracy": 0.11062222197651864,
      "num_tokens": 5339954.0,
      "step": 2455
    },
    {
      "entropy": 7.098154449462891,
      "epoch": 0.26325646101985123,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004997234625231709,
      "loss": 6.9431,
      "mean_token_accuracy": 0.10974880829453468,
      "num_tokens": 5350919.0,
      "step": 2460
    },
    {
      "entropy": 6.838466548919678,
      "epoch": 0.26379153512761516,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004997215642765832,
      "loss": 6.9072,
      "mean_token_accuracy": 0.11278821527957916,
      "num_tokens": 5361844.0,
      "step": 2465
    },
    {
      "entropy": 6.999466514587402,
      "epoch": 0.2643266092353791,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499719659541192,
      "loss": 6.9529,
      "mean_token_accuracy": 0.10501818954944611,
      "num_tokens": 5372026.0,
      "step": 2470
    },
    {
      "entropy": 7.002458000183106,
      "epoch": 0.26486168334314303,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004997177483170521,
      "loss": 6.9913,
      "mean_token_accuracy": 0.11029187738895416,
      "num_tokens": 5382965.0,
      "step": 2475
    },
    {
      "entropy": 6.9821876049041744,
      "epoch": 0.26539675745090696,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997158306042188,
      "loss": 6.9353,
      "mean_token_accuracy": 0.11200979501008987,
      "num_tokens": 5393738.0,
      "step": 2480
    },
    {
      "entropy": 6.932433700561523,
      "epoch": 0.2659318315586709,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997139064027476,
      "loss": 6.8079,
      "mean_token_accuracy": 0.1185495488345623,
      "num_tokens": 5404059.0,
      "step": 2485
    },
    {
      "entropy": 6.902970170974731,
      "epoch": 0.26646690566643483,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997119757126938,
      "loss": 6.8471,
      "mean_token_accuracy": 0.10944071337580681,
      "num_tokens": 5414152.0,
      "step": 2490
    },
    {
      "entropy": 7.007512903213501,
      "epoch": 0.2670019797741987,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997100385341133,
      "loss": 7.0207,
      "mean_token_accuracy": 0.1047422781586647,
      "num_tokens": 5425227.0,
      "step": 2495
    },
    {
      "entropy": 6.9481219291687015,
      "epoch": 0.26753705388196264,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997080948670621,
      "loss": 6.851,
      "mean_token_accuracy": 0.11013932898640633,
      "num_tokens": 5435716.0,
      "step": 2500
    },
    {
      "entropy": 6.94979419708252,
      "epoch": 0.26807212798972657,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997061447115962,
      "loss": 6.9098,
      "mean_token_accuracy": 0.10717173442244529,
      "num_tokens": 5446519.0,
      "step": 2505
    },
    {
      "entropy": 6.94394063949585,
      "epoch": 0.2686072020974905,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997041880677719,
      "loss": 6.9125,
      "mean_token_accuracy": 0.10961175858974456,
      "num_tokens": 5457496.0,
      "step": 2510
    },
    {
      "entropy": 6.833197116851807,
      "epoch": 0.26914227620525444,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997022249356457,
      "loss": 6.7977,
      "mean_token_accuracy": 0.11055489629507065,
      "num_tokens": 5469059.0,
      "step": 2515
    },
    {
      "entropy": 6.931081962585449,
      "epoch": 0.26967735031301837,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004997002553152745,
      "loss": 6.9011,
      "mean_token_accuracy": 0.10881304815411567,
      "num_tokens": 5480324.0,
      "step": 2520
    },
    {
      "entropy": 6.962761163711548,
      "epoch": 0.2702124244207823,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004996982792067148,
      "loss": 6.9194,
      "mean_token_accuracy": 0.10171629562973976,
      "num_tokens": 5491513.0,
      "step": 2525
    },
    {
      "entropy": 6.862958478927612,
      "epoch": 0.2707474985285462,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499696296610024,
      "loss": 6.8224,
      "mean_token_accuracy": 0.11024987697601318,
      "num_tokens": 5502414.0,
      "step": 2530
    },
    {
      "entropy": 6.903598022460938,
      "epoch": 0.2712825726363101,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996943075252592,
      "loss": 6.9627,
      "mean_token_accuracy": 0.10326235815882683,
      "num_tokens": 5512962.0,
      "step": 2535
    },
    {
      "entropy": 6.995188236236572,
      "epoch": 0.27181764674407405,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996923119524777,
      "loss": 6.8347,
      "mean_token_accuracy": 0.10915073826909065,
      "num_tokens": 5524648.0,
      "step": 2540
    },
    {
      "entropy": 6.832237482070923,
      "epoch": 0.272352720851838,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004996903098917375,
      "loss": 6.9081,
      "mean_token_accuracy": 0.10347842946648597,
      "num_tokens": 5536158.0,
      "step": 2545
    },
    {
      "entropy": 6.879280519485474,
      "epoch": 0.2728877949596019,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499688301343096,
      "loss": 6.8113,
      "mean_token_accuracy": 0.11145112216472626,
      "num_tokens": 5547374.0,
      "step": 2550
    },
    {
      "entropy": 6.9910749912261965,
      "epoch": 0.27342286906736585,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996862863066114,
      "loss": 6.9016,
      "mean_token_accuracy": 0.10962290540337563,
      "num_tokens": 5559212.0,
      "step": 2555
    },
    {
      "entropy": 6.935283899307251,
      "epoch": 0.2739579431751298,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996842647823419,
      "loss": 6.862,
      "mean_token_accuracy": 0.11124518364667893,
      "num_tokens": 5569302.0,
      "step": 2560
    },
    {
      "entropy": 6.867130374908447,
      "epoch": 0.27449301728289366,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996822367703458,
      "loss": 6.8764,
      "mean_token_accuracy": 0.1044128842651844,
      "num_tokens": 5581688.0,
      "step": 2565
    },
    {
      "entropy": 6.930262565612793,
      "epoch": 0.2750280913906576,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996802022706817,
      "loss": 6.8049,
      "mean_token_accuracy": 0.1141422875225544,
      "num_tokens": 5591376.0,
      "step": 2570
    },
    {
      "entropy": 6.87796802520752,
      "epoch": 0.2755631654984215,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996781612834083,
      "loss": 6.8474,
      "mean_token_accuracy": 0.11154236644506454,
      "num_tokens": 5601510.0,
      "step": 2575
    },
    {
      "entropy": 6.987272834777832,
      "epoch": 0.27609823960618546,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996761138085846,
      "loss": 6.9155,
      "mean_token_accuracy": 0.10548463389277458,
      "num_tokens": 5611436.0,
      "step": 2580
    },
    {
      "entropy": 6.904627418518066,
      "epoch": 0.2766333137139494,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996740598462697,
      "loss": 6.919,
      "mean_token_accuracy": 0.10808756351470947,
      "num_tokens": 5620470.0,
      "step": 2585
    },
    {
      "entropy": 6.84314284324646,
      "epoch": 0.2771683878217133,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996719993965229,
      "loss": 6.8253,
      "mean_token_accuracy": 0.11080246791243553,
      "num_tokens": 5630740.0,
      "step": 2590
    },
    {
      "entropy": 6.915116310119629,
      "epoch": 0.27770346192947726,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996699324594036,
      "loss": 6.8227,
      "mean_token_accuracy": 0.11591969728469849,
      "num_tokens": 5641138.0,
      "step": 2595
    },
    {
      "entropy": 6.922396898269653,
      "epoch": 0.27823853603724114,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996678590349717,
      "loss": 6.9565,
      "mean_token_accuracy": 0.10815248861908913,
      "num_tokens": 5652474.0,
      "step": 2600
    },
    {
      "entropy": 6.979134368896484,
      "epoch": 0.27877361014500507,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996657791232868,
      "loss": 6.8568,
      "mean_token_accuracy": 0.11182226091623307,
      "num_tokens": 5663178.0,
      "step": 2605
    },
    {
      "entropy": 6.886240863800049,
      "epoch": 0.279308684252769,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996636927244092,
      "loss": 6.8834,
      "mean_token_accuracy": 0.10366401001811028,
      "num_tokens": 5673565.0,
      "step": 2610
    },
    {
      "entropy": 7.031818151473999,
      "epoch": 0.27984375836053293,
      "grad_norm": 1.0,
      "learning_rate": 0.000499661599838399,
      "loss": 6.924,
      "mean_token_accuracy": 0.10450949743390084,
      "num_tokens": 5684651.0,
      "step": 2615
    },
    {
      "entropy": 6.81337537765503,
      "epoch": 0.28037883246829687,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996595004653167,
      "loss": 6.8068,
      "mean_token_accuracy": 0.1102476328611374,
      "num_tokens": 5696081.0,
      "step": 2620
    },
    {
      "entropy": 6.930059099197388,
      "epoch": 0.2809139065760608,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996573946052229,
      "loss": 6.8431,
      "mean_token_accuracy": 0.11160285472869873,
      "num_tokens": 5706830.0,
      "step": 2625
    },
    {
      "entropy": 6.902744102478027,
      "epoch": 0.28144898068382473,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996552822581783,
      "loss": 6.8702,
      "mean_token_accuracy": 0.11438541486859322,
      "num_tokens": 5717261.0,
      "step": 2630
    },
    {
      "entropy": 6.92958779335022,
      "epoch": 0.2819840547915886,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499653163424244,
      "loss": 6.8333,
      "mean_token_accuracy": 0.10897459760308266,
      "num_tokens": 5727280.0,
      "step": 2635
    },
    {
      "entropy": 6.897455167770386,
      "epoch": 0.28251912889935255,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996510381034814,
      "loss": 6.9035,
      "mean_token_accuracy": 0.10450076460838317,
      "num_tokens": 5737485.0,
      "step": 2640
    },
    {
      "entropy": 6.918154382705689,
      "epoch": 0.2830542030071165,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996489062959515,
      "loss": 6.8543,
      "mean_token_accuracy": 0.10870546996593475,
      "num_tokens": 5747683.0,
      "step": 2645
    },
    {
      "entropy": 6.8486976146698,
      "epoch": 0.2835892771148804,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996467680017159,
      "loss": 6.927,
      "mean_token_accuracy": 0.10556823164224624,
      "num_tokens": 5758148.0,
      "step": 2650
    },
    {
      "entropy": 6.931982326507568,
      "epoch": 0.28412435122264434,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996446232208365,
      "loss": 6.855,
      "mean_token_accuracy": 0.11552600264549255,
      "num_tokens": 5769109.0,
      "step": 2655
    },
    {
      "entropy": 6.934611415863037,
      "epoch": 0.2846594253304083,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996424719533753,
      "loss": 6.8814,
      "mean_token_accuracy": 0.10460555031895638,
      "num_tokens": 5780344.0,
      "step": 2660
    },
    {
      "entropy": 6.882219982147217,
      "epoch": 0.2851944994381722,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996403141993941,
      "loss": 6.6908,
      "mean_token_accuracy": 0.11797674894332885,
      "num_tokens": 5790167.0,
      "step": 2665
    },
    {
      "entropy": 6.848588514328003,
      "epoch": 0.2857295735459361,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996381499589555,
      "loss": 6.8693,
      "mean_token_accuracy": 0.10816706344485283,
      "num_tokens": 5800076.0,
      "step": 2670
    },
    {
      "entropy": 6.981981515884399,
      "epoch": 0.2862646476537,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996359792321219,
      "loss": 6.806,
      "mean_token_accuracy": 0.11308679804205894,
      "num_tokens": 5810547.0,
      "step": 2675
    },
    {
      "entropy": 6.848021030426025,
      "epoch": 0.28679972176146395,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499633802018956,
      "loss": 6.8335,
      "mean_token_accuracy": 0.11338982209563256,
      "num_tokens": 5821210.0,
      "step": 2680
    },
    {
      "entropy": 6.866777563095093,
      "epoch": 0.2873347958692279,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996316183195204,
      "loss": 6.8726,
      "mean_token_accuracy": 0.10846167802810669,
      "num_tokens": 5832361.0,
      "step": 2685
    },
    {
      "entropy": 6.973077297210693,
      "epoch": 0.2878698699769918,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996294281338786,
      "loss": 6.8657,
      "mean_token_accuracy": 0.10616375505924225,
      "num_tokens": 5843242.0,
      "step": 2690
    },
    {
      "entropy": 6.948590850830078,
      "epoch": 0.28840494408475575,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996272314620936,
      "loss": 6.9116,
      "mean_token_accuracy": 0.10822648853063584,
      "num_tokens": 5855163.0,
      "step": 2695
    },
    {
      "entropy": 6.8517499446868895,
      "epoch": 0.2889400181925197,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996250283042288,
      "loss": 6.8609,
      "mean_token_accuracy": 0.11147657707333565,
      "num_tokens": 5866908.0,
      "step": 2700
    },
    {
      "entropy": 6.9437988758087155,
      "epoch": 0.28947509230028357,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499622818660348,
      "loss": 6.8965,
      "mean_token_accuracy": 0.10642576739192008,
      "num_tokens": 5877342.0,
      "step": 2705
    },
    {
      "entropy": 6.871459531784057,
      "epoch": 0.2900101664080475,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996206025305147,
      "loss": 6.8371,
      "mean_token_accuracy": 0.1101390264928341,
      "num_tokens": 5887960.0,
      "step": 2710
    },
    {
      "entropy": 6.852532005310058,
      "epoch": 0.29054524051581143,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996183799147932,
      "loss": 6.8293,
      "mean_token_accuracy": 0.1123297668993473,
      "num_tokens": 5899887.0,
      "step": 2715
    },
    {
      "entropy": 6.906511831283569,
      "epoch": 0.29108031462357536,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996161508132475,
      "loss": 6.8399,
      "mean_token_accuracy": 0.10181801989674569,
      "num_tokens": 5911022.0,
      "step": 2720
    },
    {
      "entropy": 6.972908449172974,
      "epoch": 0.2916153887313393,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499613915225942,
      "loss": 6.9042,
      "mean_token_accuracy": 0.11424257606267929,
      "num_tokens": 5922780.0,
      "step": 2725
    },
    {
      "entropy": 6.888608407974243,
      "epoch": 0.29215046283910323,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996116731529412,
      "loss": 6.9206,
      "mean_token_accuracy": 0.10672475844621658,
      "num_tokens": 5934112.0,
      "step": 2730
    },
    {
      "entropy": 6.909975528717041,
      "epoch": 0.29268553694686716,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049960942459431,
      "loss": 6.8719,
      "mean_token_accuracy": 0.1075200729072094,
      "num_tokens": 5944386.0,
      "step": 2735
    },
    {
      "entropy": 6.823930931091309,
      "epoch": 0.29322061105463104,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996071695501132,
      "loss": 6.795,
      "mean_token_accuracy": 0.11570547893643379,
      "num_tokens": 5955552.0,
      "step": 2740
    },
    {
      "entropy": 6.929108047485352,
      "epoch": 0.293755685162395,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004996049080204159,
      "loss": 6.7992,
      "mean_token_accuracy": 0.11161436066031456,
      "num_tokens": 5966790.0,
      "step": 2745
    },
    {
      "entropy": 6.947720003128052,
      "epoch": 0.2942907592701589,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996026400052835,
      "loss": 6.8082,
      "mean_token_accuracy": 0.10977331325411796,
      "num_tokens": 5977704.0,
      "step": 2750
    },
    {
      "entropy": 6.746592235565186,
      "epoch": 0.29482583337792284,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996003655047814,
      "loss": 6.8204,
      "mean_token_accuracy": 0.11269323304295539,
      "num_tokens": 5989221.0,
      "step": 2755
    },
    {
      "entropy": 6.8818727970123295,
      "epoch": 0.2953609074856868,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004995980845189753,
      "loss": 6.8632,
      "mean_token_accuracy": 0.10988379120826722,
      "num_tokens": 6000906.0,
      "step": 2760
    },
    {
      "entropy": 6.933154582977295,
      "epoch": 0.2958959815934507,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995957970479312,
      "loss": 6.8819,
      "mean_token_accuracy": 0.10542053505778312,
      "num_tokens": 6012039.0,
      "step": 2765
    },
    {
      "entropy": 6.859393644332886,
      "epoch": 0.29643105570121464,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004995935030917148,
      "loss": 6.7583,
      "mean_token_accuracy": 0.1107563003897667,
      "num_tokens": 6024383.0,
      "step": 2770
    },
    {
      "entropy": 6.944475984573364,
      "epoch": 0.2969661298089785,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995912026503926,
      "loss": 6.8297,
      "mean_token_accuracy": 0.10689945966005325,
      "num_tokens": 6034904.0,
      "step": 2775
    },
    {
      "entropy": 6.822006130218506,
      "epoch": 0.29750120391674245,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499588895724031,
      "loss": 6.8446,
      "mean_token_accuracy": 0.10969873294234275,
      "num_tokens": 6044449.0,
      "step": 2780
    },
    {
      "entropy": 6.967623519897461,
      "epoch": 0.2980362780245064,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995865823126968,
      "loss": 6.8635,
      "mean_token_accuracy": 0.10977466106414795,
      "num_tokens": 6054465.0,
      "step": 2785
    },
    {
      "entropy": 6.842376041412353,
      "epoch": 0.2985713521322703,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995842624164564,
      "loss": 6.8659,
      "mean_token_accuracy": 0.11244026795029641,
      "num_tokens": 6065539.0,
      "step": 2790
    },
    {
      "entropy": 6.909923028945923,
      "epoch": 0.29910642624003425,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499581936035377,
      "loss": 6.8292,
      "mean_token_accuracy": 0.11023836135864258,
      "num_tokens": 6076135.0,
      "step": 2795
    },
    {
      "entropy": 6.862402391433716,
      "epoch": 0.2996415003477982,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995796031695257,
      "loss": 6.8321,
      "mean_token_accuracy": 0.11241180747747422,
      "num_tokens": 6086604.0,
      "step": 2800
    },
    {
      "entropy": 6.898853063583374,
      "epoch": 0.3001765744555621,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00049957726381897,
      "loss": 6.8524,
      "mean_token_accuracy": 0.110887211561203,
      "num_tokens": 6095672.0,
      "step": 2805
    },
    {
      "entropy": 6.832514142990112,
      "epoch": 0.300711648563326,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004995749179837773,
      "loss": 6.7318,
      "mean_token_accuracy": 0.1211558923125267,
      "num_tokens": 6105652.0,
      "step": 2810
    },
    {
      "entropy": 6.821787977218628,
      "epoch": 0.30124672267108993,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004995725656640155,
      "loss": 6.7662,
      "mean_token_accuracy": 0.11450377255678176,
      "num_tokens": 6116930.0,
      "step": 2815
    },
    {
      "entropy": 6.7775352001190186,
      "epoch": 0.30178179677885386,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995702068597524,
      "loss": 6.7879,
      "mean_token_accuracy": 0.11996725052595139,
      "num_tokens": 6126659.0,
      "step": 2820
    },
    {
      "entropy": 6.892369937896729,
      "epoch": 0.3023168708866178,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004995678415710561,
      "loss": 6.8129,
      "mean_token_accuracy": 0.11141618713736534,
      "num_tokens": 6137227.0,
      "step": 2825
    },
    {
      "entropy": 6.947154092788696,
      "epoch": 0.3028519449943817,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499565469797995,
      "loss": 6.8657,
      "mean_token_accuracy": 0.10855738222599029,
      "num_tokens": 6147800.0,
      "step": 2830
    },
    {
      "entropy": 6.7966372013092045,
      "epoch": 0.30338701910214566,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995630915406374,
      "loss": 6.7189,
      "mean_token_accuracy": 0.12241782546043396,
      "num_tokens": 6157671.0,
      "step": 2835
    },
    {
      "entropy": 6.899184989929199,
      "epoch": 0.3039220932099096,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004995607067990521,
      "loss": 6.8556,
      "mean_token_accuracy": 0.10919948741793632,
      "num_tokens": 6168651.0,
      "step": 2840
    },
    {
      "entropy": 6.825665998458862,
      "epoch": 0.30445716731767347,
      "grad_norm": 1.3125,
      "learning_rate": 0.000499558315573308,
      "loss": 6.8348,
      "mean_token_accuracy": 0.11045416444540024,
      "num_tokens": 6179925.0,
      "step": 2845
    },
    {
      "entropy": 6.824082469940185,
      "epoch": 0.3049922414254374,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995559178634741,
      "loss": 6.8225,
      "mean_token_accuracy": 0.1148190289735794,
      "num_tokens": 6190541.0,
      "step": 2850
    },
    {
      "entropy": 6.998980903625489,
      "epoch": 0.30552731553320134,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995535136696196,
      "loss": 6.8724,
      "mean_token_accuracy": 0.10492667257785797,
      "num_tokens": 6201491.0,
      "step": 2855
    },
    {
      "entropy": 6.7254845142364506,
      "epoch": 0.30606238964096527,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499551102991814,
      "loss": 6.6706,
      "mean_token_accuracy": 0.11548433601856231,
      "num_tokens": 6212443.0,
      "step": 2860
    },
    {
      "entropy": 6.859250020980835,
      "epoch": 0.3065974637487292,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995486858301268,
      "loss": 6.8688,
      "mean_token_accuracy": 0.10495678186416627,
      "num_tokens": 6222623.0,
      "step": 2865
    },
    {
      "entropy": 6.866484022140503,
      "epoch": 0.30713253785649314,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004995462621846279,
      "loss": 6.7644,
      "mean_token_accuracy": 0.11185173094272613,
      "num_tokens": 6233712.0,
      "step": 2870
    },
    {
      "entropy": 6.8427448749542235,
      "epoch": 0.30766761196425707,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995438320553871,
      "loss": 6.8319,
      "mean_token_accuracy": 0.11026796624064446,
      "num_tokens": 6244254.0,
      "step": 2875
    },
    {
      "entropy": 6.854343605041504,
      "epoch": 0.30820268607202095,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995413954424748,
      "loss": 6.7313,
      "mean_token_accuracy": 0.11795990094542504,
      "num_tokens": 6254202.0,
      "step": 2880
    },
    {
      "entropy": 6.881530380249023,
      "epoch": 0.3087377601797849,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995389523459612,
      "loss": 6.7989,
      "mean_token_accuracy": 0.11081418693065644,
      "num_tokens": 6264496.0,
      "step": 2885
    },
    {
      "entropy": 6.803116035461426,
      "epoch": 0.3092728342875488,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499536502765917,
      "loss": 6.8187,
      "mean_token_accuracy": 0.1121898703277111,
      "num_tokens": 6275835.0,
      "step": 2890
    },
    {
      "entropy": 6.885844755172729,
      "epoch": 0.30980790839531275,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995340467024128,
      "loss": 6.7583,
      "mean_token_accuracy": 0.11417267769575119,
      "num_tokens": 6286805.0,
      "step": 2895
    },
    {
      "entropy": 6.834712362289428,
      "epoch": 0.3103429825030767,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004995315841555195,
      "loss": 6.7758,
      "mean_token_accuracy": 0.10655382797122001,
      "num_tokens": 6298444.0,
      "step": 2900
    },
    {
      "entropy": 6.729912948608399,
      "epoch": 0.3108780566108406,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995291151253083,
      "loss": 6.7163,
      "mean_token_accuracy": 0.11894652470946313,
      "num_tokens": 6308259.0,
      "step": 2905
    },
    {
      "entropy": 6.859683704376221,
      "epoch": 0.31141313071860455,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995266396118505,
      "loss": 6.7759,
      "mean_token_accuracy": 0.10650198608636856,
      "num_tokens": 6320058.0,
      "step": 2910
    },
    {
      "entropy": 6.822231721878052,
      "epoch": 0.3119482048263684,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995241576152174,
      "loss": 6.7593,
      "mean_token_accuracy": 0.11011653542518615,
      "num_tokens": 6331203.0,
      "step": 2915
    },
    {
      "entropy": 6.908184051513672,
      "epoch": 0.31248327893413236,
      "grad_norm": 1.0,
      "learning_rate": 0.000499521669135481,
      "loss": 6.8635,
      "mean_token_accuracy": 0.1086197093129158,
      "num_tokens": 6342027.0,
      "step": 2920
    },
    {
      "entropy": 6.910853099822998,
      "epoch": 0.3130183530418963,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004995191741727128,
      "loss": 6.7739,
      "mean_token_accuracy": 0.11335675492882728,
      "num_tokens": 6352580.0,
      "step": 2925
    },
    {
      "entropy": 6.797909688949585,
      "epoch": 0.3135534271496602,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000499516672726985,
      "loss": 6.7823,
      "mean_token_accuracy": 0.1125186413526535,
      "num_tokens": 6363997.0,
      "step": 2930
    },
    {
      "entropy": 6.844564914703369,
      "epoch": 0.31408850125742416,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995141647983699,
      "loss": 6.8312,
      "mean_token_accuracy": 0.11025090962648391,
      "num_tokens": 6375284.0,
      "step": 2935
    },
    {
      "entropy": 6.801123428344726,
      "epoch": 0.3146235753651881,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995116503869398,
      "loss": 6.7135,
      "mean_token_accuracy": 0.11932718157768249,
      "num_tokens": 6386698.0,
      "step": 2940
    },
    {
      "entropy": 6.840458106994629,
      "epoch": 0.315158649472952,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995091294927673,
      "loss": 6.8021,
      "mean_token_accuracy": 0.10923359990119934,
      "num_tokens": 6397237.0,
      "step": 2945
    },
    {
      "entropy": 6.926814460754395,
      "epoch": 0.3156937235807159,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995066021159253,
      "loss": 6.9187,
      "mean_token_accuracy": 0.10698705762624741,
      "num_tokens": 6408266.0,
      "step": 2950
    },
    {
      "entropy": 6.794530534744263,
      "epoch": 0.31622879768847983,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995040682564867,
      "loss": 6.8007,
      "mean_token_accuracy": 0.11740597262978554,
      "num_tokens": 6419563.0,
      "step": 2955
    },
    {
      "entropy": 6.8177916526794435,
      "epoch": 0.31676387179624377,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995015279145247,
      "loss": 6.7606,
      "mean_token_accuracy": 0.11514141336083412,
      "num_tokens": 6430220.0,
      "step": 2960
    },
    {
      "entropy": 6.885624027252197,
      "epoch": 0.3172989459040077,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994989810901126,
      "loss": 6.7825,
      "mean_token_accuracy": 0.10802810341119766,
      "num_tokens": 6440157.0,
      "step": 2965
    },
    {
      "entropy": 6.80977611541748,
      "epoch": 0.31783402001177163,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994964277833239,
      "loss": 6.8243,
      "mean_token_accuracy": 0.1138004906475544,
      "num_tokens": 6449511.0,
      "step": 2970
    },
    {
      "entropy": 6.902962112426758,
      "epoch": 0.31836909411953557,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004994938679942324,
      "loss": 6.8115,
      "mean_token_accuracy": 0.10856785252690315,
      "num_tokens": 6460608.0,
      "step": 2975
    },
    {
      "entropy": 6.849463319778442,
      "epoch": 0.3189041682272995,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994913017229121,
      "loss": 6.8086,
      "mean_token_accuracy": 0.11735697165131569,
      "num_tokens": 6472737.0,
      "step": 2980
    },
    {
      "entropy": 6.767067813873291,
      "epoch": 0.31943924233506343,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994887289694371,
      "loss": 6.6996,
      "mean_token_accuracy": 0.11470640897750854,
      "num_tokens": 6482993.0,
      "step": 2985
    },
    {
      "entropy": 6.941344833374023,
      "epoch": 0.3199743164428273,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994861497338814,
      "loss": 6.9096,
      "mean_token_accuracy": 0.10804323926568031,
      "num_tokens": 6494530.0,
      "step": 2990
    },
    {
      "entropy": 6.845445013046264,
      "epoch": 0.32050939055059124,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994835640163198,
      "loss": 6.729,
      "mean_token_accuracy": 0.10808207094669342,
      "num_tokens": 6504691.0,
      "step": 2995
    },
    {
      "entropy": 6.832252359390258,
      "epoch": 0.3210444646583552,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994809718168267,
      "loss": 6.7681,
      "mean_token_accuracy": 0.11426141634583473,
      "num_tokens": 6514924.0,
      "step": 3000
    },
    {
      "epoch": 0.3210444646583552,
      "eval_entropy": 6.628516179339262,
      "eval_loss": 6.879497051239014,
      "eval_mean_token_accuracy": 0.1148492504127143,
      "eval_num_tokens": 6514924.0,
      "eval_runtime": 23.2795,
      "eval_samples_per_second": 1274.897,
      "eval_steps_per_second": 159.367,
      "step": 3000
    },
    {
      "entropy": 6.749145030975342,
      "epoch": 0.3215795387661191,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994783731354771,
      "loss": 6.7273,
      "mean_token_accuracy": 0.11545848771929741,
      "num_tokens": 6526097.0,
      "step": 3005
    },
    {
      "entropy": 6.898501777648926,
      "epoch": 0.32211461287388304,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994757679723461,
      "loss": 6.7879,
      "mean_token_accuracy": 0.11010189205408097,
      "num_tokens": 6536384.0,
      "step": 3010
    },
    {
      "entropy": 6.875375509262085,
      "epoch": 0.322649686981647,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994731563275088,
      "loss": 6.8397,
      "mean_token_accuracy": 0.10491524860262871,
      "num_tokens": 6547161.0,
      "step": 3015
    },
    {
      "entropy": 6.734170627593994,
      "epoch": 0.3231847610894109,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994705382010406,
      "loss": 6.7589,
      "mean_token_accuracy": 0.1133797138929367,
      "num_tokens": 6556751.0,
      "step": 3020
    },
    {
      "entropy": 6.82257308959961,
      "epoch": 0.3237198351971748,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994679135930174,
      "loss": 6.826,
      "mean_token_accuracy": 0.11575602516531944,
      "num_tokens": 6567655.0,
      "step": 3025
    },
    {
      "entropy": 6.946610784530639,
      "epoch": 0.3242549093049387,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994652825035144,
      "loss": 6.8511,
      "mean_token_accuracy": 0.11181805208325386,
      "num_tokens": 6579127.0,
      "step": 3030
    },
    {
      "entropy": 6.837035274505615,
      "epoch": 0.32478998341270265,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994626449326081,
      "loss": 6.8227,
      "mean_token_accuracy": 0.11550880670547485,
      "num_tokens": 6590779.0,
      "step": 3035
    },
    {
      "entropy": 6.8091630935668945,
      "epoch": 0.3253250575204666,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994600008803745,
      "loss": 6.7243,
      "mean_token_accuracy": 0.11760424971580505,
      "num_tokens": 6600849.0,
      "step": 3040
    },
    {
      "entropy": 6.80384407043457,
      "epoch": 0.3258601316282305,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994573503468899,
      "loss": 6.7253,
      "mean_token_accuracy": 0.12053586021065713,
      "num_tokens": 6610633.0,
      "step": 3045
    },
    {
      "entropy": 6.780997371673584,
      "epoch": 0.32639520573599445,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004994546933322308,
      "loss": 6.7579,
      "mean_token_accuracy": 0.11172309741377831,
      "num_tokens": 6621327.0,
      "step": 3050
    },
    {
      "entropy": 6.871892499923706,
      "epoch": 0.3269302798437584,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499452029836474,
      "loss": 6.72,
      "mean_token_accuracy": 0.11655889376997948,
      "num_tokens": 6632267.0,
      "step": 3055
    },
    {
      "entropy": 6.787418460845947,
      "epoch": 0.32746535395152226,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994493598596963,
      "loss": 6.7415,
      "mean_token_accuracy": 0.11163913160562515,
      "num_tokens": 6642714.0,
      "step": 3060
    },
    {
      "entropy": 6.833496141433716,
      "epoch": 0.3280004280592862,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994466834019749,
      "loss": 6.7629,
      "mean_token_accuracy": 0.11880431026220321,
      "num_tokens": 6652353.0,
      "step": 3065
    },
    {
      "entropy": 6.796854448318482,
      "epoch": 0.32853550216705013,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994440004633871,
      "loss": 6.7317,
      "mean_token_accuracy": 0.11276261731982232,
      "num_tokens": 6663133.0,
      "step": 3070
    },
    {
      "entropy": 6.724622488021851,
      "epoch": 0.32907057627481406,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994413110440103,
      "loss": 6.6964,
      "mean_token_accuracy": 0.1199144259095192,
      "num_tokens": 6673984.0,
      "step": 3075
    },
    {
      "entropy": 6.838273334503174,
      "epoch": 0.329605650382578,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004994386151439222,
      "loss": 6.78,
      "mean_token_accuracy": 0.10911644920706749,
      "num_tokens": 6686171.0,
      "step": 3080
    },
    {
      "entropy": 6.856180095672608,
      "epoch": 0.33014072449034193,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004994359127632006,
      "loss": 6.8237,
      "mean_token_accuracy": 0.11093982830643653,
      "num_tokens": 6696609.0,
      "step": 3085
    },
    {
      "entropy": 6.86229100227356,
      "epoch": 0.33067579859810586,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994332039019235,
      "loss": 6.7534,
      "mean_token_accuracy": 0.11561796367168427,
      "num_tokens": 6706188.0,
      "step": 3090
    },
    {
      "entropy": 6.672589731216431,
      "epoch": 0.33121087270586974,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994304885601694,
      "loss": 6.7374,
      "mean_token_accuracy": 0.11004470363259315,
      "num_tokens": 6716276.0,
      "step": 3095
    },
    {
      "entropy": 6.78862738609314,
      "epoch": 0.3317459468136337,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994277667380162,
      "loss": 6.7064,
      "mean_token_accuracy": 0.12015045955777168,
      "num_tokens": 6727176.0,
      "step": 3100
    },
    {
      "entropy": 6.838768720626831,
      "epoch": 0.3322810209213976,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994250384355429,
      "loss": 6.7934,
      "mean_token_accuracy": 0.11404274702072144,
      "num_tokens": 6737768.0,
      "step": 3105
    },
    {
      "entropy": 6.753113698959351,
      "epoch": 0.33281609502916154,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994223036528282,
      "loss": 6.7369,
      "mean_token_accuracy": 0.11214326322078705,
      "num_tokens": 6747962.0,
      "step": 3110
    },
    {
      "entropy": 6.861884021759034,
      "epoch": 0.3333511691369255,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499419562389951,
      "loss": 6.727,
      "mean_token_accuracy": 0.11631539016962052,
      "num_tokens": 6758400.0,
      "step": 3115
    },
    {
      "entropy": 6.746067714691162,
      "epoch": 0.3338862432446894,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994168146469904,
      "loss": 6.7155,
      "mean_token_accuracy": 0.11617984026670455,
      "num_tokens": 6768855.0,
      "step": 3120
    },
    {
      "entropy": 6.8038405418396,
      "epoch": 0.33442131735245334,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994140604240259,
      "loss": 6.6931,
      "mean_token_accuracy": 0.11366450041532516,
      "num_tokens": 6778656.0,
      "step": 3125
    },
    {
      "entropy": 6.80411696434021,
      "epoch": 0.3349563914602172,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004994112997211369,
      "loss": 6.7853,
      "mean_token_accuracy": 0.11631628945469856,
      "num_tokens": 6789960.0,
      "step": 3130
    },
    {
      "entropy": 6.721054792404175,
      "epoch": 0.33549146556798115,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994085325384031,
      "loss": 6.7672,
      "mean_token_accuracy": 0.11114905029535294,
      "num_tokens": 6800188.0,
      "step": 3135
    },
    {
      "entropy": 6.8586445331573485,
      "epoch": 0.3360265396757451,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994057588759045,
      "loss": 6.7148,
      "mean_token_accuracy": 0.11809506490826607,
      "num_tokens": 6811032.0,
      "step": 3140
    },
    {
      "entropy": 6.737908458709716,
      "epoch": 0.336561613783509,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994029787337212,
      "loss": 6.7373,
      "mean_token_accuracy": 0.11105087548494338,
      "num_tokens": 6823060.0,
      "step": 3145
    },
    {
      "entropy": 6.82881531715393,
      "epoch": 0.33709668789127295,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994001921119333,
      "loss": 6.7996,
      "mean_token_accuracy": 0.11395466029644012,
      "num_tokens": 6833559.0,
      "step": 3150
    },
    {
      "entropy": 6.8296314716339115,
      "epoch": 0.3376317619990369,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993973990106215,
      "loss": 6.7678,
      "mean_token_accuracy": 0.11352298334240914,
      "num_tokens": 6843657.0,
      "step": 3155
    },
    {
      "entropy": 6.717380046844482,
      "epoch": 0.3381668361068008,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993945994298662,
      "loss": 6.7039,
      "mean_token_accuracy": 0.10965939536690712,
      "num_tokens": 6853615.0,
      "step": 3160
    },
    {
      "entropy": 6.852908182144165,
      "epoch": 0.3387019102145647,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993917933697484,
      "loss": 6.8052,
      "mean_token_accuracy": 0.11230268105864524,
      "num_tokens": 6865256.0,
      "step": 3165
    },
    {
      "entropy": 6.965103816986084,
      "epoch": 0.3392369843223286,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993889808303493,
      "loss": 6.8339,
      "mean_token_accuracy": 0.11253256648778916,
      "num_tokens": 6875358.0,
      "step": 3170
    },
    {
      "entropy": 6.713176727294922,
      "epoch": 0.33977205843009256,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993861618117498,
      "loss": 6.7396,
      "mean_token_accuracy": 0.11107463911175727,
      "num_tokens": 6886584.0,
      "step": 3175
    },
    {
      "entropy": 6.785865211486817,
      "epoch": 0.3403071325378565,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993833363140314,
      "loss": 6.6709,
      "mean_token_accuracy": 0.12084045559167862,
      "num_tokens": 6896958.0,
      "step": 3180
    },
    {
      "entropy": 6.753049325942993,
      "epoch": 0.3408422066456204,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993805043372756,
      "loss": 6.7497,
      "mean_token_accuracy": 0.11840963140130042,
      "num_tokens": 6907944.0,
      "step": 3185
    },
    {
      "entropy": 6.82925009727478,
      "epoch": 0.34137728075338436,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993776658815644,
      "loss": 6.8358,
      "mean_token_accuracy": 0.10783932730555534,
      "num_tokens": 6919486.0,
      "step": 3190
    },
    {
      "entropy": 6.919765424728394,
      "epoch": 0.3419123548611483,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993748209469796,
      "loss": 6.8202,
      "mean_token_accuracy": 0.11589022427797317,
      "num_tokens": 6930650.0,
      "step": 3195
    },
    {
      "entropy": 6.816604948043823,
      "epoch": 0.34244742896891217,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993719695336033,
      "loss": 6.7463,
      "mean_token_accuracy": 0.11324085146188737,
      "num_tokens": 6941519.0,
      "step": 3200
    },
    {
      "entropy": 6.772028493881225,
      "epoch": 0.3429825030766761,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993691116415179,
      "loss": 6.7962,
      "mean_token_accuracy": 0.11355016529560089,
      "num_tokens": 6952952.0,
      "step": 3205
    },
    {
      "entropy": 6.861077785491943,
      "epoch": 0.34351757718444004,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499366247270806,
      "loss": 6.7533,
      "mean_token_accuracy": 0.11717762500047683,
      "num_tokens": 6962867.0,
      "step": 3210
    },
    {
      "entropy": 6.7836878299713135,
      "epoch": 0.34405265129220397,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993633764215502,
      "loss": 6.6925,
      "mean_token_accuracy": 0.11943860650062561,
      "num_tokens": 6973434.0,
      "step": 3215
    },
    {
      "entropy": 6.810595512390137,
      "epoch": 0.3445877253999679,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993604990938336,
      "loss": 6.6933,
      "mean_token_accuracy": 0.1249708317220211,
      "num_tokens": 6984161.0,
      "step": 3220
    },
    {
      "entropy": 6.757954692840576,
      "epoch": 0.34512279950773184,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993576152877389,
      "loss": 6.6774,
      "mean_token_accuracy": 0.11473060622811318,
      "num_tokens": 6994700.0,
      "step": 3225
    },
    {
      "entropy": 6.822511672973633,
      "epoch": 0.34565787361549577,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993547250033496,
      "loss": 6.831,
      "mean_token_accuracy": 0.10570734962821007,
      "num_tokens": 7006497.0,
      "step": 3230
    },
    {
      "entropy": 6.826636171340942,
      "epoch": 0.34619294772325965,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993518282407493,
      "loss": 6.7574,
      "mean_token_accuracy": 0.11926060020923615,
      "num_tokens": 7016415.0,
      "step": 3235
    },
    {
      "entropy": 6.757991743087769,
      "epoch": 0.3467280218310236,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993489250000213,
      "loss": 6.7025,
      "mean_token_accuracy": 0.11810342445969582,
      "num_tokens": 7026610.0,
      "step": 3240
    },
    {
      "entropy": 6.803502607345581,
      "epoch": 0.3472630959387875,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993460152812497,
      "loss": 6.7172,
      "mean_token_accuracy": 0.11101711913943291,
      "num_tokens": 7038236.0,
      "step": 3245
    },
    {
      "entropy": 6.742218255996704,
      "epoch": 0.34779817004655145,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993430990845184,
      "loss": 6.7625,
      "mean_token_accuracy": 0.12092217653989792,
      "num_tokens": 7049438.0,
      "step": 3250
    },
    {
      "entropy": 6.787970876693725,
      "epoch": 0.3483332441543154,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993401764099118,
      "loss": 6.7329,
      "mean_token_accuracy": 0.11177100390195846,
      "num_tokens": 7061001.0,
      "step": 3255
    },
    {
      "entropy": 6.80476942062378,
      "epoch": 0.3488683182620793,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004993372472575139,
      "loss": 6.7011,
      "mean_token_accuracy": 0.11599645167589187,
      "num_tokens": 7071328.0,
      "step": 3260
    },
    {
      "entropy": 6.850950717926025,
      "epoch": 0.34940339236984325,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004993343116274096,
      "loss": 6.8958,
      "mean_token_accuracy": 0.10798017606139183,
      "num_tokens": 7082597.0,
      "step": 3265
    },
    {
      "entropy": 6.789200735092163,
      "epoch": 0.3499384664776071,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004993313695196836,
      "loss": 6.7096,
      "mean_token_accuracy": 0.11781042814254761,
      "num_tokens": 7093463.0,
      "step": 3270
    },
    {
      "entropy": 6.761033916473389,
      "epoch": 0.35047354058537106,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004993284209344208,
      "loss": 6.7864,
      "mean_token_accuracy": 0.12042865455150605,
      "num_tokens": 7104350.0,
      "step": 3275
    },
    {
      "entropy": 6.915784788131714,
      "epoch": 0.351008614693135,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993254658717065,
      "loss": 6.7369,
      "mean_token_accuracy": 0.11913842409849167,
      "num_tokens": 7114851.0,
      "step": 3280
    },
    {
      "entropy": 6.787514400482178,
      "epoch": 0.3515436888008989,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004993225043316257,
      "loss": 6.7874,
      "mean_token_accuracy": 0.11141818836331367,
      "num_tokens": 7126440.0,
      "step": 3285
    },
    {
      "entropy": 6.858088207244873,
      "epoch": 0.35207876290866286,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993195363142643,
      "loss": 6.8004,
      "mean_token_accuracy": 0.11125228255987167,
      "num_tokens": 7137884.0,
      "step": 3290
    },
    {
      "entropy": 6.84716763496399,
      "epoch": 0.3526138370164268,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993165618197076,
      "loss": 6.7796,
      "mean_token_accuracy": 0.10891105383634567,
      "num_tokens": 7147930.0,
      "step": 3295
    },
    {
      "entropy": 6.7377028465271,
      "epoch": 0.3531489111241907,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004993135808480418,
      "loss": 6.6485,
      "mean_token_accuracy": 0.12461603209376335,
      "num_tokens": 7159200.0,
      "step": 3300
    },
    {
      "entropy": 6.83154182434082,
      "epoch": 0.3536839852319546,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993105933993529,
      "loss": 6.7354,
      "mean_token_accuracy": 0.11935730800032615,
      "num_tokens": 7169601.0,
      "step": 3305
    },
    {
      "entropy": 6.692518615722657,
      "epoch": 0.35421905933971853,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993075994737271,
      "loss": 6.6574,
      "mean_token_accuracy": 0.12401366010308265,
      "num_tokens": 7180465.0,
      "step": 3310
    },
    {
      "entropy": 6.799501037597656,
      "epoch": 0.35475413344748247,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993045990712508,
      "loss": 6.6867,
      "mean_token_accuracy": 0.11757550165057182,
      "num_tokens": 7190603.0,
      "step": 3315
    },
    {
      "entropy": 6.695236730575561,
      "epoch": 0.3552892075552464,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993015921920109,
      "loss": 6.6407,
      "mean_token_accuracy": 0.1223471999168396,
      "num_tokens": 7201017.0,
      "step": 3320
    },
    {
      "entropy": 6.663977766036988,
      "epoch": 0.35582428166301033,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499298578836094,
      "loss": 6.6613,
      "mean_token_accuracy": 0.11793281361460686,
      "num_tokens": 7210908.0,
      "step": 3325
    },
    {
      "entropy": 6.856654691696167,
      "epoch": 0.35635935577077427,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499295559003587,
      "loss": 6.6847,
      "mean_token_accuracy": 0.11869573444128037,
      "num_tokens": 7221306.0,
      "step": 3330
    },
    {
      "entropy": 6.756274557113647,
      "epoch": 0.3568944298785382,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992925326945774,
      "loss": 6.8152,
      "mean_token_accuracy": 0.11084710285067559,
      "num_tokens": 7233076.0,
      "step": 3335
    },
    {
      "entropy": 6.821779584884643,
      "epoch": 0.3574295039863021,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992894999091524,
      "loss": 6.809,
      "mean_token_accuracy": 0.11464324221014977,
      "num_tokens": 7244703.0,
      "step": 3340
    },
    {
      "entropy": 6.871327877044678,
      "epoch": 0.357964578094066,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992864606473996,
      "loss": 6.7702,
      "mean_token_accuracy": 0.10874314606189728,
      "num_tokens": 7255438.0,
      "step": 3345
    },
    {
      "entropy": 6.789708518981934,
      "epoch": 0.35849965220182994,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992834149094069,
      "loss": 6.7225,
      "mean_token_accuracy": 0.1161846548318863,
      "num_tokens": 7265720.0,
      "step": 3350
    },
    {
      "entropy": 6.778335046768189,
      "epoch": 0.3590347263095939,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004992803626952619,
      "loss": 6.7794,
      "mean_token_accuracy": 0.11164675280451775,
      "num_tokens": 7277791.0,
      "step": 3355
    },
    {
      "entropy": 6.721046304702758,
      "epoch": 0.3595698004173578,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499277304005053,
      "loss": 6.6411,
      "mean_token_accuracy": 0.1210532508790493,
      "num_tokens": 7288521.0,
      "step": 3360
    },
    {
      "entropy": 6.812202024459839,
      "epoch": 0.36010487452512174,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992742388388686,
      "loss": 6.8254,
      "mean_token_accuracy": 0.10808679983019828,
      "num_tokens": 7299298.0,
      "step": 3365
    },
    {
      "entropy": 6.759737586975097,
      "epoch": 0.3606399486328857,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992711671967968,
      "loss": 6.6621,
      "mean_token_accuracy": 0.11631347015500068,
      "num_tokens": 7310271.0,
      "step": 3370
    },
    {
      "entropy": 6.777987098693847,
      "epoch": 0.36117502274064955,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004992680890789268,
      "loss": 6.7087,
      "mean_token_accuracy": 0.11685341075062752,
      "num_tokens": 7321112.0,
      "step": 3375
    },
    {
      "entropy": 6.706979513168335,
      "epoch": 0.3617100968484135,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992650044853471,
      "loss": 6.6774,
      "mean_token_accuracy": 0.11788536980748177,
      "num_tokens": 7331976.0,
      "step": 3380
    },
    {
      "entropy": 6.912421607971192,
      "epoch": 0.3622451709561774,
      "grad_norm": 1.0,
      "learning_rate": 0.000499261913416147,
      "loss": 6.7944,
      "mean_token_accuracy": 0.1156319186091423,
      "num_tokens": 7342800.0,
      "step": 3385
    },
    {
      "entropy": 6.759701156616211,
      "epoch": 0.36278024506394135,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992588158714157,
      "loss": 6.6172,
      "mean_token_accuracy": 0.11470383033156395,
      "num_tokens": 7354213.0,
      "step": 3390
    },
    {
      "entropy": 6.719116735458374,
      "epoch": 0.3633153191717053,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992557118512424,
      "loss": 6.6924,
      "mean_token_accuracy": 0.11977254748344421,
      "num_tokens": 7363250.0,
      "step": 3395
    },
    {
      "entropy": 6.741212654113769,
      "epoch": 0.3638503932794692,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992526013557171,
      "loss": 6.7299,
      "mean_token_accuracy": 0.1131053201854229,
      "num_tokens": 7374185.0,
      "step": 3400
    },
    {
      "entropy": 6.722179841995239,
      "epoch": 0.36438546738723315,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992494843849294,
      "loss": 6.6811,
      "mean_token_accuracy": 0.11220811381936073,
      "num_tokens": 7384579.0,
      "step": 3405
    },
    {
      "entropy": 6.872027826309204,
      "epoch": 0.36492054149499703,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992463609389694,
      "loss": 6.7122,
      "mean_token_accuracy": 0.11460142880678177,
      "num_tokens": 7395277.0,
      "step": 3410
    },
    {
      "entropy": 6.7189747333526615,
      "epoch": 0.36545561560276096,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004992432310179272,
      "loss": 6.702,
      "mean_token_accuracy": 0.11632506772875786,
      "num_tokens": 7406688.0,
      "step": 3415
    },
    {
      "entropy": 6.68887767791748,
      "epoch": 0.3659906897105249,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992400946218932,
      "loss": 6.7531,
      "mean_token_accuracy": 0.11685344353318214,
      "num_tokens": 7417378.0,
      "step": 3420
    },
    {
      "entropy": 6.928713035583496,
      "epoch": 0.36652576381828883,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004992369517509581,
      "loss": 6.7899,
      "mean_token_accuracy": 0.1146116465330124,
      "num_tokens": 7427675.0,
      "step": 3425
    },
    {
      "entropy": 6.749551439285279,
      "epoch": 0.36706083792605276,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992338024052123,
      "loss": 6.8196,
      "mean_token_accuracy": 0.11129243597388268,
      "num_tokens": 7438083.0,
      "step": 3430
    },
    {
      "entropy": 6.729024171829224,
      "epoch": 0.3675959120338167,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992306465847473,
      "loss": 6.6599,
      "mean_token_accuracy": 0.1189745843410492,
      "num_tokens": 7449051.0,
      "step": 3435
    },
    {
      "entropy": 6.778530502319336,
      "epoch": 0.36813098614158063,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004992274842896536,
      "loss": 6.6903,
      "mean_token_accuracy": 0.12192284166812897,
      "num_tokens": 7459074.0,
      "step": 3440
    },
    {
      "entropy": 6.739143228530883,
      "epoch": 0.3686660602493445,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992243155200229,
      "loss": 6.75,
      "mean_token_accuracy": 0.11641542166471482,
      "num_tokens": 7470472.0,
      "step": 3445
    },
    {
      "entropy": 6.795718145370484,
      "epoch": 0.36920113435710844,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992211402759466,
      "loss": 6.7236,
      "mean_token_accuracy": 0.12090882286429405,
      "num_tokens": 7481477.0,
      "step": 3450
    },
    {
      "entropy": 6.791615772247314,
      "epoch": 0.3697362084648724,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992179585575164,
      "loss": 6.6776,
      "mean_token_accuracy": 0.12016240730881692,
      "num_tokens": 7491906.0,
      "step": 3455
    },
    {
      "entropy": 6.725628423690796,
      "epoch": 0.3702712825726363,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499214770364824,
      "loss": 6.6118,
      "mean_token_accuracy": 0.11341834589838981,
      "num_tokens": 7502447.0,
      "step": 3460
    },
    {
      "entropy": 6.695222520828247,
      "epoch": 0.37080635668040024,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992115756979617,
      "loss": 6.6113,
      "mean_token_accuracy": 0.12414827719330787,
      "num_tokens": 7513505.0,
      "step": 3465
    },
    {
      "entropy": 6.70502028465271,
      "epoch": 0.3713414307881642,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992083745570218,
      "loss": 6.6382,
      "mean_token_accuracy": 0.12110274732112884,
      "num_tokens": 7524884.0,
      "step": 3470
    },
    {
      "entropy": 6.719266605377197,
      "epoch": 0.3718765048959281,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992051669420964,
      "loss": 6.7074,
      "mean_token_accuracy": 0.11903871521353722,
      "num_tokens": 7535942.0,
      "step": 3475
    },
    {
      "entropy": 6.779046154022216,
      "epoch": 0.372411579003692,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992019528532785,
      "loss": 6.6768,
      "mean_token_accuracy": 0.12179931253194809,
      "num_tokens": 7545506.0,
      "step": 3480
    },
    {
      "entropy": 6.800584745407105,
      "epoch": 0.3729466531114559,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991987322906605,
      "loss": 6.7635,
      "mean_token_accuracy": 0.11009820327162742,
      "num_tokens": 7555328.0,
      "step": 3485
    },
    {
      "entropy": 6.754938220977783,
      "epoch": 0.37348172721921985,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991955052543357,
      "loss": 6.6491,
      "mean_token_accuracy": 0.11945150792598724,
      "num_tokens": 7565762.0,
      "step": 3490
    },
    {
      "entropy": 6.766927289962768,
      "epoch": 0.3740168013269838,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991922717443972,
      "loss": 6.6786,
      "mean_token_accuracy": 0.12708448022603988,
      "num_tokens": 7575884.0,
      "step": 3495
    },
    {
      "entropy": 6.688040733337402,
      "epoch": 0.3745518754347477,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991890317609384,
      "loss": 6.7177,
      "mean_token_accuracy": 0.11786145195364953,
      "num_tokens": 7586100.0,
      "step": 3500
    },
    {
      "entropy": 6.804274797439575,
      "epoch": 0.37508694954251165,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991857853040528,
      "loss": 6.7334,
      "mean_token_accuracy": 0.11635208576917648,
      "num_tokens": 7597137.0,
      "step": 3505
    },
    {
      "entropy": 6.722698783874511,
      "epoch": 0.3756220236502756,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991825323738341,
      "loss": 6.7415,
      "mean_token_accuracy": 0.11808021515607833,
      "num_tokens": 7607640.0,
      "step": 3510
    },
    {
      "entropy": 6.792010879516601,
      "epoch": 0.3761570977580395,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991792729703763,
      "loss": 6.8155,
      "mean_token_accuracy": 0.1162738896906376,
      "num_tokens": 7618319.0,
      "step": 3515
    },
    {
      "entropy": 6.81560378074646,
      "epoch": 0.3766921718658034,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004991760070937734,
      "loss": 6.6739,
      "mean_token_accuracy": 0.11905946955084801,
      "num_tokens": 7628350.0,
      "step": 3520
    },
    {
      "entropy": 6.782265710830688,
      "epoch": 0.3772272459735673,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991727347441199,
      "loss": 6.7034,
      "mean_token_accuracy": 0.11543130278587341,
      "num_tokens": 7637857.0,
      "step": 3525
    },
    {
      "entropy": 6.806895208358765,
      "epoch": 0.37776232008133126,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991694559215102,
      "loss": 6.7241,
      "mean_token_accuracy": 0.11398787200450897,
      "num_tokens": 7648242.0,
      "step": 3530
    },
    {
      "entropy": 6.744486331939697,
      "epoch": 0.3782973941890952,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000499166170626039,
      "loss": 6.733,
      "mean_token_accuracy": 0.1127483144402504,
      "num_tokens": 7660565.0,
      "step": 3535
    },
    {
      "entropy": 6.713475036621094,
      "epoch": 0.3788324682968591,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499162878857801,
      "loss": 6.7323,
      "mean_token_accuracy": 0.11836331784725189,
      "num_tokens": 7671600.0,
      "step": 3540
    },
    {
      "entropy": 6.793336200714111,
      "epoch": 0.37936754240462306,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991595806168915,
      "loss": 6.7383,
      "mean_token_accuracy": 0.11187790781259536,
      "num_tokens": 7682453.0,
      "step": 3545
    },
    {
      "entropy": 6.815833520889282,
      "epoch": 0.379902616512387,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004991562759034055,
      "loss": 6.6809,
      "mean_token_accuracy": 0.1189245343208313,
      "num_tokens": 7693490.0,
      "step": 3550
    },
    {
      "entropy": 6.729942464828492,
      "epoch": 0.38043769062015087,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991529647174386,
      "loss": 6.6362,
      "mean_token_accuracy": 0.11553772613406181,
      "num_tokens": 7703478.0,
      "step": 3555
    },
    {
      "entropy": 6.693486452102661,
      "epoch": 0.3809727647279148,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991496470590864,
      "loss": 6.7356,
      "mean_token_accuracy": 0.11734711453318596,
      "num_tokens": 7714647.0,
      "step": 3560
    },
    {
      "entropy": 6.791001510620117,
      "epoch": 0.38150783883567874,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004991463229284445,
      "loss": 6.6783,
      "mean_token_accuracy": 0.1179485410451889,
      "num_tokens": 7725528.0,
      "step": 3565
    },
    {
      "entropy": 6.70664644241333,
      "epoch": 0.38204291294344267,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991429923256091,
      "loss": 6.7271,
      "mean_token_accuracy": 0.11620324924588203,
      "num_tokens": 7737708.0,
      "step": 3570
    },
    {
      "entropy": 6.844662094116211,
      "epoch": 0.3825779870512066,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004991396552506762,
      "loss": 6.7292,
      "mean_token_accuracy": 0.11200911179184914,
      "num_tokens": 7749093.0,
      "step": 3575
    },
    {
      "entropy": 6.649820470809937,
      "epoch": 0.38311306115897054,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991363117037423,
      "loss": 6.6586,
      "mean_token_accuracy": 0.11758191138505936,
      "num_tokens": 7760024.0,
      "step": 3580
    },
    {
      "entropy": 6.750805044174195,
      "epoch": 0.38364813526673447,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004991329616849039,
      "loss": 6.7035,
      "mean_token_accuracy": 0.12209635972976685,
      "num_tokens": 7770801.0,
      "step": 3585
    },
    {
      "entropy": 6.781581115722656,
      "epoch": 0.38418320937449835,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991296051942578,
      "loss": 6.7629,
      "mean_token_accuracy": 0.11573808416724204,
      "num_tokens": 7782257.0,
      "step": 3590
    },
    {
      "entropy": 6.727374076843262,
      "epoch": 0.3847182834822623,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991262422319008,
      "loss": 6.6751,
      "mean_token_accuracy": 0.11982549130916595,
      "num_tokens": 7793016.0,
      "step": 3595
    },
    {
      "entropy": 6.711760234832764,
      "epoch": 0.3852533575900262,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00049912287279793,
      "loss": 6.6033,
      "mean_token_accuracy": 0.12885212972760202,
      "num_tokens": 7802207.0,
      "step": 3600
    },
    {
      "entropy": 6.767945432662964,
      "epoch": 0.38578843169779015,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991194968924429,
      "loss": 6.7279,
      "mean_token_accuracy": 0.11415286064147949,
      "num_tokens": 7813487.0,
      "step": 3605
    },
    {
      "entropy": 6.792677783966065,
      "epoch": 0.3863235058055541,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991161145155367,
      "loss": 6.7103,
      "mean_token_accuracy": 0.11249904707074165,
      "num_tokens": 7824587.0,
      "step": 3610
    },
    {
      "entropy": 6.755476427078247,
      "epoch": 0.386858579913318,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004991127256673091,
      "loss": 6.6914,
      "mean_token_accuracy": 0.11287119016051292,
      "num_tokens": 7836240.0,
      "step": 3615
    },
    {
      "entropy": 6.688715505599975,
      "epoch": 0.38739365402108195,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991093303478582,
      "loss": 6.6489,
      "mean_token_accuracy": 0.11667624786496163,
      "num_tokens": 7847014.0,
      "step": 3620
    },
    {
      "entropy": 6.784205055236816,
      "epoch": 0.3879287281288458,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004991059285572818,
      "loss": 6.7614,
      "mean_token_accuracy": 0.12089815065264702,
      "num_tokens": 7857899.0,
      "step": 3625
    },
    {
      "entropy": 6.8409124374389645,
      "epoch": 0.38846380223660976,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991025202956781,
      "loss": 6.6477,
      "mean_token_accuracy": 0.11972974091768265,
      "num_tokens": 7868987.0,
      "step": 3630
    },
    {
      "entropy": 6.740195369720459,
      "epoch": 0.3889988763443737,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004990991055631457,
      "loss": 6.7674,
      "mean_token_accuracy": 0.11278079152107238,
      "num_tokens": 7879547.0,
      "step": 3635
    },
    {
      "entropy": 6.698840999603272,
      "epoch": 0.3895339504521376,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990956843597832,
      "loss": 6.6347,
      "mean_token_accuracy": 0.12292784750461579,
      "num_tokens": 7889824.0,
      "step": 3640
    },
    {
      "entropy": 6.655135917663574,
      "epoch": 0.39006902455990156,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004990922566856892,
      "loss": 6.6254,
      "mean_token_accuracy": 0.1167909786105156,
      "num_tokens": 7901156.0,
      "step": 3645
    },
    {
      "entropy": 6.787825727462769,
      "epoch": 0.3906040986676655,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004990888225409629,
      "loss": 6.6672,
      "mean_token_accuracy": 0.1191375233232975,
      "num_tokens": 7911695.0,
      "step": 3650
    },
    {
      "entropy": 6.661014366149902,
      "epoch": 0.3911391727754294,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004990853819257033,
      "loss": 6.6564,
      "mean_token_accuracy": 0.11809485256671906,
      "num_tokens": 7923559.0,
      "step": 3655
    },
    {
      "entropy": 6.777256345748901,
      "epoch": 0.3916742468831933,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990819348400097,
      "loss": 6.6877,
      "mean_token_accuracy": 0.1148997962474823,
      "num_tokens": 7935792.0,
      "step": 3660
    },
    {
      "entropy": 6.745595169067383,
      "epoch": 0.39220932099095723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004990784812839818,
      "loss": 6.6675,
      "mean_token_accuracy": 0.12436324208974839,
      "num_tokens": 7946987.0,
      "step": 3665
    },
    {
      "entropy": 6.624452543258667,
      "epoch": 0.39274439509872117,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990750212577192,
      "loss": 6.6066,
      "mean_token_accuracy": 0.12105611711740494,
      "num_tokens": 7957160.0,
      "step": 3670
    },
    {
      "entropy": 6.770350170135498,
      "epoch": 0.3932794692064851,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990715547613219,
      "loss": 6.6924,
      "mean_token_accuracy": 0.1102707713842392,
      "num_tokens": 7968833.0,
      "step": 3675
    },
    {
      "entropy": 6.818655633926392,
      "epoch": 0.39381454331424903,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004990680817948898,
      "loss": 6.7448,
      "mean_token_accuracy": 0.1078771933913231,
      "num_tokens": 7980661.0,
      "step": 3680
    },
    {
      "entropy": 6.783357191085815,
      "epoch": 0.39434961742201297,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990646023585234,
      "loss": 6.6646,
      "mean_token_accuracy": 0.11455314755439758,
      "num_tokens": 7992677.0,
      "step": 3685
    },
    {
      "entropy": 6.646414422988892,
      "epoch": 0.3948846915297769,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990611164523231,
      "loss": 6.636,
      "mean_token_accuracy": 0.11985439360141754,
      "num_tokens": 8002796.0,
      "step": 3690
    },
    {
      "entropy": 6.6637656688690186,
      "epoch": 0.3954197656375408,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990576240763896,
      "loss": 6.6766,
      "mean_token_accuracy": 0.12152916267514229,
      "num_tokens": 8012779.0,
      "step": 3695
    },
    {
      "entropy": 6.814187288284302,
      "epoch": 0.3959548397453047,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004990541252308237,
      "loss": 6.6628,
      "mean_token_accuracy": 0.1231559731066227,
      "num_tokens": 8023442.0,
      "step": 3700
    },
    {
      "entropy": 6.747038173675537,
      "epoch": 0.39648991385306864,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990506199157263,
      "loss": 6.73,
      "mean_token_accuracy": 0.11592325270175934,
      "num_tokens": 8035163.0,
      "step": 3705
    },
    {
      "entropy": 6.762581205368042,
      "epoch": 0.3970249879608326,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990471081311988,
      "loss": 6.665,
      "mean_token_accuracy": 0.11645122244954109,
      "num_tokens": 8045330.0,
      "step": 3710
    },
    {
      "entropy": 6.774192762374878,
      "epoch": 0.3975600620685965,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990435898773426,
      "loss": 6.6854,
      "mean_token_accuracy": 0.11946498751640319,
      "num_tokens": 8056154.0,
      "step": 3715
    },
    {
      "entropy": 6.73372073173523,
      "epoch": 0.39809513617636044,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990400651542592,
      "loss": 6.6467,
      "mean_token_accuracy": 0.1226908691227436,
      "num_tokens": 8066101.0,
      "step": 3720
    },
    {
      "entropy": 6.738417625427246,
      "epoch": 0.3986302102841244,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990365339620504,
      "loss": 6.7145,
      "mean_token_accuracy": 0.11076832339167594,
      "num_tokens": 8077249.0,
      "step": 3725
    },
    {
      "entropy": 6.767896938323974,
      "epoch": 0.39916528439188825,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499032996300818,
      "loss": 6.6649,
      "mean_token_accuracy": 0.11845851466059684,
      "num_tokens": 8088532.0,
      "step": 3730
    },
    {
      "entropy": 6.78744330406189,
      "epoch": 0.3997003584996522,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004990294521706645,
      "loss": 6.7122,
      "mean_token_accuracy": 0.10873547717928886,
      "num_tokens": 8099541.0,
      "step": 3735
    },
    {
      "entropy": 6.71293625831604,
      "epoch": 0.4002354326074161,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499025901571692,
      "loss": 6.7521,
      "mean_token_accuracy": 0.1155206136405468,
      "num_tokens": 8110532.0,
      "step": 3740
    },
    {
      "entropy": 6.76421160697937,
      "epoch": 0.40077050671518005,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004990223445040031,
      "loss": 6.6845,
      "mean_token_accuracy": 0.1239987462759018,
      "num_tokens": 8120991.0,
      "step": 3745
    },
    {
      "entropy": 6.7578223705291744,
      "epoch": 0.401305580822944,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004990187809677004,
      "loss": 6.6057,
      "mean_token_accuracy": 0.11667410209774971,
      "num_tokens": 8130737.0,
      "step": 3750
    },
    {
      "entropy": 6.620117902755737,
      "epoch": 0.4018406549307079,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990152109628869,
      "loss": 6.6132,
      "mean_token_accuracy": 0.1167342483997345,
      "num_tokens": 8142315.0,
      "step": 3755
    },
    {
      "entropy": 6.748623418807983,
      "epoch": 0.40237572903847185,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004990116344896657,
      "loss": 6.7076,
      "mean_token_accuracy": 0.11085205748677254,
      "num_tokens": 8153618.0,
      "step": 3760
    },
    {
      "entropy": 6.682367515563965,
      "epoch": 0.40291080314623573,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00049900805154814,
      "loss": 6.6437,
      "mean_token_accuracy": 0.1233924686908722,
      "num_tokens": 8164304.0,
      "step": 3765
    },
    {
      "entropy": 6.741749382019043,
      "epoch": 0.40344587725399966,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990044621384134,
      "loss": 6.6478,
      "mean_token_accuracy": 0.1144657127559185,
      "num_tokens": 8175289.0,
      "step": 3770
    },
    {
      "entropy": 6.778554677963257,
      "epoch": 0.4039809513617636,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990008662605893,
      "loss": 6.6962,
      "mean_token_accuracy": 0.11412339508533478,
      "num_tokens": 8186274.0,
      "step": 3775
    },
    {
      "entropy": 6.7074426174163815,
      "epoch": 0.40451602546952753,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989972639147717,
      "loss": 6.6278,
      "mean_token_accuracy": 0.12370904386043549,
      "num_tokens": 8196380.0,
      "step": 3780
    },
    {
      "entropy": 6.701422214508057,
      "epoch": 0.40505109957729146,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989936551010646,
      "loss": 6.7017,
      "mean_token_accuracy": 0.1167987897992134,
      "num_tokens": 8207204.0,
      "step": 3785
    },
    {
      "entropy": 6.665639448165893,
      "epoch": 0.4055861736850554,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989900398195722,
      "loss": 6.598,
      "mean_token_accuracy": 0.11927853375673295,
      "num_tokens": 8217745.0,
      "step": 3790
    },
    {
      "entropy": 6.699862337112426,
      "epoch": 0.40612124779281933,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989864180703988,
      "loss": 6.603,
      "mean_token_accuracy": 0.12265799641609192,
      "num_tokens": 8228171.0,
      "step": 3795
    },
    {
      "entropy": 6.688190937042236,
      "epoch": 0.4066563219005832,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989827898536491,
      "loss": 6.6512,
      "mean_token_accuracy": 0.11492432728409767,
      "num_tokens": 8239220.0,
      "step": 3800
    },
    {
      "entropy": 6.772840118408203,
      "epoch": 0.40719139600834714,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989791551694279,
      "loss": 6.6927,
      "mean_token_accuracy": 0.11572175249457359,
      "num_tokens": 8250363.0,
      "step": 3805
    },
    {
      "entropy": 6.7018718242645265,
      "epoch": 0.4077264701161111,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989755140178398,
      "loss": 6.6535,
      "mean_token_accuracy": 0.11787640154361725,
      "num_tokens": 8261207.0,
      "step": 3810
    },
    {
      "entropy": 6.627443408966064,
      "epoch": 0.408261544223875,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989718663989905,
      "loss": 6.5546,
      "mean_token_accuracy": 0.11985465362668038,
      "num_tokens": 8272343.0,
      "step": 3815
    },
    {
      "entropy": 6.661830902099609,
      "epoch": 0.40879661833163894,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004989682123129849,
      "loss": 6.6075,
      "mean_token_accuracy": 0.11918459832668304,
      "num_tokens": 8282677.0,
      "step": 3820
    },
    {
      "entropy": 6.636364889144898,
      "epoch": 0.40933169243940287,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989645517599286,
      "loss": 6.6211,
      "mean_token_accuracy": 0.11509314477443695,
      "num_tokens": 8292753.0,
      "step": 3825
    },
    {
      "entropy": 6.685767602920532,
      "epoch": 0.4098667665471668,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989608847399275,
      "loss": 6.61,
      "mean_token_accuracy": 0.12080298736691475,
      "num_tokens": 8302747.0,
      "step": 3830
    },
    {
      "entropy": 6.7413169860839846,
      "epoch": 0.4104018406549307,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004989572112530871,
      "loss": 6.6778,
      "mean_token_accuracy": 0.11377226114273072,
      "num_tokens": 8314766.0,
      "step": 3835
    },
    {
      "entropy": 6.787960433959961,
      "epoch": 0.4109369147626946,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989535312995139,
      "loss": 6.7178,
      "mean_token_accuracy": 0.11318623274564743,
      "num_tokens": 8325791.0,
      "step": 3840
    },
    {
      "entropy": 6.717939376831055,
      "epoch": 0.41147198887045855,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989498448793139,
      "loss": 6.6361,
      "mean_token_accuracy": 0.12173747569322586,
      "num_tokens": 8335809.0,
      "step": 3845
    },
    {
      "entropy": 6.658769178390503,
      "epoch": 0.4120070629782225,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004989461519925935,
      "loss": 6.6329,
      "mean_token_accuracy": 0.11577236950397492,
      "num_tokens": 8347505.0,
      "step": 3850
    },
    {
      "entropy": 6.760444831848145,
      "epoch": 0.4125421370859864,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989424526394596,
      "loss": 6.6476,
      "mean_token_accuracy": 0.11038320958614349,
      "num_tokens": 8358363.0,
      "step": 3855
    },
    {
      "entropy": 6.685798072814942,
      "epoch": 0.41307721119375035,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989387468200187,
      "loss": 6.6586,
      "mean_token_accuracy": 0.11995591446757317,
      "num_tokens": 8369076.0,
      "step": 3860
    },
    {
      "entropy": 6.792741823196411,
      "epoch": 0.4136122853015143,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498935034534378,
      "loss": 6.7066,
      "mean_token_accuracy": 0.11812848448753357,
      "num_tokens": 8380071.0,
      "step": 3865
    },
    {
      "entropy": 6.7582135677337645,
      "epoch": 0.41414735940927816,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004989313157826446,
      "loss": 6.6634,
      "mean_token_accuracy": 0.12095108553767205,
      "num_tokens": 8390574.0,
      "step": 3870
    },
    {
      "entropy": 6.685640001296997,
      "epoch": 0.4146824335170421,
      "grad_norm": 0.953125,
      "learning_rate": 0.000498927590564926,
      "loss": 6.6488,
      "mean_token_accuracy": 0.11926722228527069,
      "num_tokens": 8401154.0,
      "step": 3875
    },
    {
      "entropy": 6.767927932739258,
      "epoch": 0.415217507624806,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989238588813296,
      "loss": 6.726,
      "mean_token_accuracy": 0.11190465018153191,
      "num_tokens": 8412113.0,
      "step": 3880
    },
    {
      "entropy": 6.811918687820435,
      "epoch": 0.41575258173256996,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989201207319634,
      "loss": 6.7108,
      "mean_token_accuracy": 0.1182281218469143,
      "num_tokens": 8422734.0,
      "step": 3885
    },
    {
      "entropy": 6.656246137619019,
      "epoch": 0.4162876558403339,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004989163761169351,
      "loss": 6.6486,
      "mean_token_accuracy": 0.11386847794055939,
      "num_tokens": 8433647.0,
      "step": 3890
    },
    {
      "entropy": 6.706199264526367,
      "epoch": 0.4168227299480978,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989126250363529,
      "loss": 6.6691,
      "mean_token_accuracy": 0.12189342752099037,
      "num_tokens": 8444726.0,
      "step": 3895
    },
    {
      "entropy": 6.775026035308838,
      "epoch": 0.41735780405586176,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989088674903251,
      "loss": 6.6902,
      "mean_token_accuracy": 0.11433882266283035,
      "num_tokens": 8455351.0,
      "step": 3900
    },
    {
      "entropy": 6.726575613021851,
      "epoch": 0.41789287816362564,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989051034789602,
      "loss": 6.6622,
      "mean_token_accuracy": 0.11935925707221032,
      "num_tokens": 8466423.0,
      "step": 3905
    },
    {
      "entropy": 6.712535429000854,
      "epoch": 0.41842795227138957,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498901333002367,
      "loss": 6.6157,
      "mean_token_accuracy": 0.11526683047413826,
      "num_tokens": 8475841.0,
      "step": 3910
    },
    {
      "entropy": 6.743713092803955,
      "epoch": 0.4189630263791535,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004988975560606541,
      "loss": 6.7144,
      "mean_token_accuracy": 0.1172465056180954,
      "num_tokens": 8487656.0,
      "step": 3915
    },
    {
      "entropy": 6.729172039031982,
      "epoch": 0.41949810048691744,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988937726539308,
      "loss": 6.721,
      "mean_token_accuracy": 0.10986425504088401,
      "num_tokens": 8497959.0,
      "step": 3920
    },
    {
      "entropy": 6.711235523223877,
      "epoch": 0.42003317459468137,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988899827823062,
      "loss": 6.6334,
      "mean_token_accuracy": 0.11989354267716408,
      "num_tokens": 8508228.0,
      "step": 3925
    },
    {
      "entropy": 6.709075260162353,
      "epoch": 0.4205682487024453,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049888618644589,
      "loss": 6.6763,
      "mean_token_accuracy": 0.11326484829187393,
      "num_tokens": 8518999.0,
      "step": 3930
    },
    {
      "entropy": 6.6963389873504635,
      "epoch": 0.42110332281020924,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988823836447916,
      "loss": 6.7018,
      "mean_token_accuracy": 0.11377114206552505,
      "num_tokens": 8529583.0,
      "step": 3935
    },
    {
      "entropy": 6.884630489349365,
      "epoch": 0.4216383969179731,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004988785743791207,
      "loss": 6.7854,
      "mean_token_accuracy": 0.10914561599493026,
      "num_tokens": 8540069.0,
      "step": 3940
    },
    {
      "entropy": 6.764514493942261,
      "epoch": 0.42217347102573705,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988747586489876,
      "loss": 6.6535,
      "mean_token_accuracy": 0.1135537788271904,
      "num_tokens": 8550441.0,
      "step": 3945
    },
    {
      "entropy": 6.6675217628479,
      "epoch": 0.422708545133501,
      "grad_norm": 1.0,
      "learning_rate": 0.000498870936454502,
      "loss": 6.6104,
      "mean_token_accuracy": 0.12095493003726006,
      "num_tokens": 8560257.0,
      "step": 3950
    },
    {
      "entropy": 6.727188062667847,
      "epoch": 0.4232436192412649,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004988671077957749,
      "loss": 6.6945,
      "mean_token_accuracy": 0.12016920894384384,
      "num_tokens": 8570991.0,
      "step": 3955
    },
    {
      "entropy": 6.795679950714112,
      "epoch": 0.42377869334902885,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004988632726729164,
      "loss": 6.7556,
      "mean_token_accuracy": 0.11323562562465668,
      "num_tokens": 8583668.0,
      "step": 3960
    },
    {
      "entropy": 6.731791210174561,
      "epoch": 0.4243137674567928,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004988594310860374,
      "loss": 6.6781,
      "mean_token_accuracy": 0.11760099306702614,
      "num_tokens": 8593727.0,
      "step": 3965
    },
    {
      "entropy": 6.690351915359497,
      "epoch": 0.4248488415645567,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004988555830352485,
      "loss": 6.6778,
      "mean_token_accuracy": 0.11991618275642395,
      "num_tokens": 8605470.0,
      "step": 3970
    },
    {
      "entropy": 6.747353982925415,
      "epoch": 0.4253839156723206,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004988517285206614,
      "loss": 6.6079,
      "mean_token_accuracy": 0.11389391347765923,
      "num_tokens": 8617176.0,
      "step": 3975
    },
    {
      "entropy": 6.691808223724365,
      "epoch": 0.4259189897800845,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004988478675423871,
      "loss": 6.6936,
      "mean_token_accuracy": 0.11446112543344497,
      "num_tokens": 8627756.0,
      "step": 3980
    },
    {
      "entropy": 6.723490619659424,
      "epoch": 0.42645406388784846,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988440001005369,
      "loss": 6.633,
      "mean_token_accuracy": 0.12306867241859436,
      "num_tokens": 8637766.0,
      "step": 3985
    },
    {
      "entropy": 6.736693811416626,
      "epoch": 0.4269891379956124,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988401261952227,
      "loss": 6.664,
      "mean_token_accuracy": 0.1152922384440899,
      "num_tokens": 8648562.0,
      "step": 3990
    },
    {
      "entropy": 6.774680089950562,
      "epoch": 0.4275242121033763,
      "grad_norm": 1.125,
      "learning_rate": 0.0004988362458265565,
      "loss": 6.6814,
      "mean_token_accuracy": 0.1093427799642086,
      "num_tokens": 8658393.0,
      "step": 3995
    },
    {
      "entropy": 6.665417051315307,
      "epoch": 0.42805928621114026,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988323589946499,
      "loss": 6.6165,
      "mean_token_accuracy": 0.12214137688279152,
      "num_tokens": 8668365.0,
      "step": 4000
    },
    {
      "entropy": 6.708557367324829,
      "epoch": 0.4285943603189042,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988284656996156,
      "loss": 6.645,
      "mean_token_accuracy": 0.1173097126185894,
      "num_tokens": 8679228.0,
      "step": 4005
    },
    {
      "entropy": 6.636530923843384,
      "epoch": 0.42912943442666807,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004988245659415656,
      "loss": 6.5261,
      "mean_token_accuracy": 0.1273254171013832,
      "num_tokens": 8688940.0,
      "step": 4010
    },
    {
      "entropy": 6.769261598587036,
      "epoch": 0.429664508534432,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988206597206129,
      "loss": 6.6923,
      "mean_token_accuracy": 0.11572074741125107,
      "num_tokens": 8699923.0,
      "step": 4015
    },
    {
      "entropy": 6.650954866409302,
      "epoch": 0.43019958264219593,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00049881674703687,
      "loss": 6.5266,
      "mean_token_accuracy": 0.1199740432202816,
      "num_tokens": 8710445.0,
      "step": 4020
    },
    {
      "entropy": 6.641194677352905,
      "epoch": 0.43073465674995987,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988128278904499,
      "loss": 6.6582,
      "mean_token_accuracy": 0.11788298562169075,
      "num_tokens": 8720686.0,
      "step": 4025
    },
    {
      "entropy": 6.765309906005859,
      "epoch": 0.4312697308577238,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498808902281466,
      "loss": 6.6738,
      "mean_token_accuracy": 0.11736497506499291,
      "num_tokens": 8732122.0,
      "step": 4030
    },
    {
      "entropy": 6.677246761322022,
      "epoch": 0.43180480496548773,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988049702100315,
      "loss": 6.638,
      "mean_token_accuracy": 0.12187011390924454,
      "num_tokens": 8744122.0,
      "step": 4035
    },
    {
      "entropy": 6.78152813911438,
      "epoch": 0.43233987907325166,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004988010316762598,
      "loss": 6.6428,
      "mean_token_accuracy": 0.11561979800462723,
      "num_tokens": 8754758.0,
      "step": 4040
    },
    {
      "entropy": 6.664593458175659,
      "epoch": 0.4328749531810156,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004987970866802649,
      "loss": 6.6774,
      "mean_token_accuracy": 0.10897876471281051,
      "num_tokens": 8766577.0,
      "step": 4045
    },
    {
      "entropy": 6.668974208831787,
      "epoch": 0.4334100272887795,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987931352221605,
      "loss": 6.6509,
      "mean_token_accuracy": 0.1155616320669651,
      "num_tokens": 8777698.0,
      "step": 4050
    },
    {
      "entropy": 6.748575258255005,
      "epoch": 0.4339451013965434,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987891773020606,
      "loss": 6.6637,
      "mean_token_accuracy": 0.12043780609965324,
      "num_tokens": 8788637.0,
      "step": 4055
    },
    {
      "entropy": 6.692825508117676,
      "epoch": 0.43448017550430734,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987852129200799,
      "loss": 6.6415,
      "mean_token_accuracy": 0.12397622093558311,
      "num_tokens": 8799709.0,
      "step": 4060
    },
    {
      "entropy": 6.712762689590454,
      "epoch": 0.4350152496120713,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987812420763326,
      "loss": 6.6496,
      "mean_token_accuracy": 0.1190851904451847,
      "num_tokens": 8810469.0,
      "step": 4065
    },
    {
      "entropy": 6.746799421310425,
      "epoch": 0.4355503237198352,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987772647709333,
      "loss": 6.6276,
      "mean_token_accuracy": 0.11508499830961227,
      "num_tokens": 8821142.0,
      "step": 4070
    },
    {
      "entropy": 6.6189371109008786,
      "epoch": 0.43608539782759914,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498773281003997,
      "loss": 6.5837,
      "mean_token_accuracy": 0.12859731987118722,
      "num_tokens": 8832951.0,
      "step": 4075
    },
    {
      "entropy": 6.702429962158203,
      "epoch": 0.4366204719353631,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987692907756385,
      "loss": 6.6863,
      "mean_token_accuracy": 0.11083661466836929,
      "num_tokens": 8843579.0,
      "step": 4080
    },
    {
      "entropy": 6.699003410339356,
      "epoch": 0.43715554604312695,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004987652940859733,
      "loss": 6.6164,
      "mean_token_accuracy": 0.11203439235687256,
      "num_tokens": 8854612.0,
      "step": 4085
    },
    {
      "entropy": 6.685548162460327,
      "epoch": 0.4376906201508909,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987612909351167,
      "loss": 6.5709,
      "mean_token_accuracy": 0.12011634036898614,
      "num_tokens": 8865584.0,
      "step": 4090
    },
    {
      "entropy": 6.605876493453979,
      "epoch": 0.4382256942586548,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004987572813231842,
      "loss": 6.6125,
      "mean_token_accuracy": 0.12365005165338516,
      "num_tokens": 8876662.0,
      "step": 4095
    },
    {
      "entropy": 6.744110584259033,
      "epoch": 0.43876076836641875,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004987532652502917,
      "loss": 6.6406,
      "mean_token_accuracy": 0.1150108441710472,
      "num_tokens": 8887249.0,
      "step": 4100
    },
    {
      "entropy": 6.658827400207519,
      "epoch": 0.4392958424741827,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987492427165551,
      "loss": 6.4757,
      "mean_token_accuracy": 0.13154868632555008,
      "num_tokens": 8898635.0,
      "step": 4105
    },
    {
      "entropy": 6.6048284530639645,
      "epoch": 0.4398309165819466,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004987452137220906,
      "loss": 6.6346,
      "mean_token_accuracy": 0.11840783655643464,
      "num_tokens": 8908794.0,
      "step": 4110
    },
    {
      "entropy": 6.831158971786499,
      "epoch": 0.44036599068971055,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004987411782670144,
      "loss": 6.6518,
      "mean_token_accuracy": 0.11594994887709617,
      "num_tokens": 8919015.0,
      "step": 4115
    },
    {
      "entropy": 6.770243310928345,
      "epoch": 0.44090106479747443,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987371363514431,
      "loss": 6.6761,
      "mean_token_accuracy": 0.11354120001196862,
      "num_tokens": 8929028.0,
      "step": 4120
    },
    {
      "entropy": 6.587651777267456,
      "epoch": 0.44143613890523836,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987330879754933,
      "loss": 6.5756,
      "mean_token_accuracy": 0.11564656272530556,
      "num_tokens": 8939789.0,
      "step": 4125
    },
    {
      "entropy": 6.635728883743286,
      "epoch": 0.4419712130130023,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004987290331392823,
      "loss": 6.5662,
      "mean_token_accuracy": 0.12269725501537324,
      "num_tokens": 8950657.0,
      "step": 4130
    },
    {
      "entropy": 6.808789396286011,
      "epoch": 0.44250628712076623,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004987249718429269,
      "loss": 6.6695,
      "mean_token_accuracy": 0.12260328158736229,
      "num_tokens": 8960744.0,
      "step": 4135
    },
    {
      "entropy": 6.666414737701416,
      "epoch": 0.44304136122853016,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004987209040865441,
      "loss": 6.5333,
      "mean_token_accuracy": 0.12561455368995667,
      "num_tokens": 8972411.0,
      "step": 4140
    },
    {
      "entropy": 6.680865144729614,
      "epoch": 0.4435764353362941,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987168298702519,
      "loss": 6.6873,
      "mean_token_accuracy": 0.11710079982876778,
      "num_tokens": 8982818.0,
      "step": 4145
    },
    {
      "entropy": 6.726581954956055,
      "epoch": 0.44411150944405803,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987127491941675,
      "loss": 6.6092,
      "mean_token_accuracy": 0.11893833950161933,
      "num_tokens": 8993227.0,
      "step": 4150
    },
    {
      "entropy": 6.6443657875061035,
      "epoch": 0.4446465835518219,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987086620584091,
      "loss": 6.5626,
      "mean_token_accuracy": 0.11763859167695045,
      "num_tokens": 9003796.0,
      "step": 4155
    },
    {
      "entropy": 6.694578695297241,
      "epoch": 0.44518165765958584,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004987045684630944,
      "loss": 6.574,
      "mean_token_accuracy": 0.12663180232048035,
      "num_tokens": 9014931.0,
      "step": 4160
    },
    {
      "entropy": 6.681905269622803,
      "epoch": 0.44571673176734977,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004987004684083417,
      "loss": 6.6337,
      "mean_token_accuracy": 0.11897012591362,
      "num_tokens": 9025312.0,
      "step": 4165
    },
    {
      "entropy": 6.681381607055664,
      "epoch": 0.4462518058751137,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986963618942694,
      "loss": 6.617,
      "mean_token_accuracy": 0.12766270712018013,
      "num_tokens": 9035563.0,
      "step": 4170
    },
    {
      "entropy": 6.783052968978882,
      "epoch": 0.44678687998287764,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004986922489209961,
      "loss": 6.6507,
      "mean_token_accuracy": 0.1229518175125122,
      "num_tokens": 9046065.0,
      "step": 4175
    },
    {
      "entropy": 6.676540994644165,
      "epoch": 0.44732195409064157,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986881294886406,
      "loss": 6.5382,
      "mean_token_accuracy": 0.12728339359164237,
      "num_tokens": 9056696.0,
      "step": 4180
    },
    {
      "entropy": 6.611935377120972,
      "epoch": 0.4478570281984055,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986840035973218,
      "loss": 6.6113,
      "mean_token_accuracy": 0.11595813184976578,
      "num_tokens": 9067035.0,
      "step": 4185
    },
    {
      "entropy": 6.672677946090698,
      "epoch": 0.4483921023061694,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986798712471588,
      "loss": 6.5888,
      "mean_token_accuracy": 0.12024698033928871,
      "num_tokens": 9078253.0,
      "step": 4190
    },
    {
      "entropy": 6.76561918258667,
      "epoch": 0.4489271764139333,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004986757324382709,
      "loss": 6.6607,
      "mean_token_accuracy": 0.11117863357067108,
      "num_tokens": 9088184.0,
      "step": 4195
    },
    {
      "entropy": 6.667329740524292,
      "epoch": 0.44946225052169725,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986715871707777,
      "loss": 6.7281,
      "mean_token_accuracy": 0.11160746067762375,
      "num_tokens": 9099321.0,
      "step": 4200
    },
    {
      "entropy": 6.786033248901367,
      "epoch": 0.4499973246294612,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004986674354447988,
      "loss": 6.7537,
      "mean_token_accuracy": 0.11719949394464493,
      "num_tokens": 9111410.0,
      "step": 4205
    },
    {
      "entropy": 6.679773759841919,
      "epoch": 0.4505323987372251,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986632772604543,
      "loss": 6.5738,
      "mean_token_accuracy": 0.12075933143496513,
      "num_tokens": 9121932.0,
      "step": 4210
    },
    {
      "entropy": 6.589503049850464,
      "epoch": 0.45106747284498905,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004986591126178639,
      "loss": 6.6564,
      "mean_token_accuracy": 0.12272690162062645,
      "num_tokens": 9132686.0,
      "step": 4215
    },
    {
      "entropy": 6.6449024200439455,
      "epoch": 0.451602546952753,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004986549415171482,
      "loss": 6.5829,
      "mean_token_accuracy": 0.11985766738653184,
      "num_tokens": 9143582.0,
      "step": 4220
    },
    {
      "entropy": 6.822705459594727,
      "epoch": 0.45213762106051686,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004986507639584274,
      "loss": 6.6585,
      "mean_token_accuracy": 0.11077262237668037,
      "num_tokens": 9156173.0,
      "step": 4225
    },
    {
      "entropy": 6.608674240112305,
      "epoch": 0.4526726951682808,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986465799418223,
      "loss": 6.6443,
      "mean_token_accuracy": 0.11647051796317101,
      "num_tokens": 9166882.0,
      "step": 4230
    },
    {
      "entropy": 6.6760540962219235,
      "epoch": 0.4532077692760447,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004986423894674534,
      "loss": 6.6421,
      "mean_token_accuracy": 0.11714837476611137,
      "num_tokens": 9177525.0,
      "step": 4235
    },
    {
      "entropy": 6.776611709594727,
      "epoch": 0.45374284338380866,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986381925354422,
      "loss": 6.6064,
      "mean_token_accuracy": 0.1193653829395771,
      "num_tokens": 9188586.0,
      "step": 4240
    },
    {
      "entropy": 6.5767412185668945,
      "epoch": 0.4542779174915726,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004986339891459094,
      "loss": 6.5167,
      "mean_token_accuracy": 0.12630144357681275,
      "num_tokens": 9199920.0,
      "step": 4245
    },
    {
      "entropy": 6.593917989730835,
      "epoch": 0.4548129915993365,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004986297792989768,
      "loss": 6.5148,
      "mean_token_accuracy": 0.12705099806189538,
      "num_tokens": 9210923.0,
      "step": 4250
    },
    {
      "entropy": 6.754072189331055,
      "epoch": 0.45534806570710046,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004986255629947655,
      "loss": 6.6896,
      "mean_token_accuracy": 0.11421982869505883,
      "num_tokens": 9221459.0,
      "step": 4255
    },
    {
      "entropy": 6.70556755065918,
      "epoch": 0.45588313981486434,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986213402333978,
      "loss": 6.6306,
      "mean_token_accuracy": 0.11420354172587395,
      "num_tokens": 9231768.0,
      "step": 4260
    },
    {
      "entropy": 6.6920239448547365,
      "epoch": 0.45641821392262827,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986171110149951,
      "loss": 6.619,
      "mean_token_accuracy": 0.11584020033478737,
      "num_tokens": 9242230.0,
      "step": 4265
    },
    {
      "entropy": 6.715697002410889,
      "epoch": 0.4569532880303922,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986128753396798,
      "loss": 6.5343,
      "mean_token_accuracy": 0.1253846324980259,
      "num_tokens": 9252485.0,
      "step": 4270
    },
    {
      "entropy": 6.545043516159057,
      "epoch": 0.45748836213815613,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986086332075742,
      "loss": 6.5936,
      "mean_token_accuracy": 0.12571235075592996,
      "num_tokens": 9263014.0,
      "step": 4275
    },
    {
      "entropy": 6.685902500152588,
      "epoch": 0.45802343624592007,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004986043846188006,
      "loss": 6.6307,
      "mean_token_accuracy": 0.12278592735528945,
      "num_tokens": 9273777.0,
      "step": 4280
    },
    {
      "entropy": 6.657036352157593,
      "epoch": 0.458558510353684,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000498600129573482,
      "loss": 6.573,
      "mean_token_accuracy": 0.1176757849752903,
      "num_tokens": 9284738.0,
      "step": 4285
    },
    {
      "entropy": 6.7254064083099365,
      "epoch": 0.45909358446144793,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498595868071741,
      "loss": 6.6216,
      "mean_token_accuracy": 0.11621444001793861,
      "num_tokens": 9295921.0,
      "step": 4290
    },
    {
      "entropy": 6.663456153869629,
      "epoch": 0.4596286585692118,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004985916001137006,
      "loss": 6.6041,
      "mean_token_accuracy": 0.11686597913503646,
      "num_tokens": 9306568.0,
      "step": 4295
    },
    {
      "entropy": 6.680031871795654,
      "epoch": 0.46016373267697575,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004985873256994843,
      "loss": 6.5667,
      "mean_token_accuracy": 0.11980162784457207,
      "num_tokens": 9316468.0,
      "step": 4300
    },
    {
      "entropy": 6.6148521900177,
      "epoch": 0.4606988067847397,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985830448292154,
      "loss": 6.6454,
      "mean_token_accuracy": 0.11909077167510987,
      "num_tokens": 9328185.0,
      "step": 4305
    },
    {
      "entropy": 6.648210954666138,
      "epoch": 0.4612338808925036,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004985787575030175,
      "loss": 6.6032,
      "mean_token_accuracy": 0.11979346349835396,
      "num_tokens": 9338591.0,
      "step": 4310
    },
    {
      "entropy": 6.6767906665802,
      "epoch": 0.46176895500026754,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004985744637210144,
      "loss": 6.6112,
      "mean_token_accuracy": 0.12049485743045807,
      "num_tokens": 9349286.0,
      "step": 4315
    },
    {
      "entropy": 6.680856657028198,
      "epoch": 0.4623040291080315,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049857016348333,
      "loss": 6.592,
      "mean_token_accuracy": 0.1270862713456154,
      "num_tokens": 9359752.0,
      "step": 4320
    },
    {
      "entropy": 6.7019250869750975,
      "epoch": 0.4628391032157954,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004985658567900886,
      "loss": 6.6666,
      "mean_token_accuracy": 0.11103762239217758,
      "num_tokens": 9371823.0,
      "step": 4325
    },
    {
      "entropy": 6.595825672149658,
      "epoch": 0.4633741773235593,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004985615436414145,
      "loss": 6.5099,
      "mean_token_accuracy": 0.11884014680981636,
      "num_tokens": 9383127.0,
      "step": 4330
    },
    {
      "entropy": 6.646753454208374,
      "epoch": 0.4639092514313232,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004985572240374321,
      "loss": 6.6249,
      "mean_token_accuracy": 0.12329683154821396,
      "num_tokens": 9394296.0,
      "step": 4335
    },
    {
      "entropy": 6.715232181549072,
      "epoch": 0.46444432553908715,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004985528979782665,
      "loss": 6.595,
      "mean_token_accuracy": 0.11608856394886971,
      "num_tokens": 9405988.0,
      "step": 4340
    },
    {
      "entropy": 6.672331953048706,
      "epoch": 0.4649793996468511,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985485654640423,
      "loss": 6.6095,
      "mean_token_accuracy": 0.12022607624530793,
      "num_tokens": 9417479.0,
      "step": 4345
    },
    {
      "entropy": 6.625867748260498,
      "epoch": 0.465514473754615,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985442264948847,
      "loss": 6.5421,
      "mean_token_accuracy": 0.12694113552570344,
      "num_tokens": 9428050.0,
      "step": 4350
    },
    {
      "entropy": 6.547886896133423,
      "epoch": 0.46604954786237895,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004985398810709189,
      "loss": 6.5174,
      "mean_token_accuracy": 0.12760872766375542,
      "num_tokens": 9438229.0,
      "step": 4355
    },
    {
      "entropy": 6.648811626434326,
      "epoch": 0.4665846219701429,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004985355291922704,
      "loss": 6.5934,
      "mean_token_accuracy": 0.11865689232945442,
      "num_tokens": 9449611.0,
      "step": 4360
    },
    {
      "entropy": 6.6590087890625,
      "epoch": 0.46711969607790677,
      "grad_norm": 1.0625,
      "learning_rate": 0.000498531170859065,
      "loss": 6.5649,
      "mean_token_accuracy": 0.12043324112892151,
      "num_tokens": 9460234.0,
      "step": 4365
    },
    {
      "entropy": 6.696133613586426,
      "epoch": 0.4676547701856707,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985268060714285,
      "loss": 6.6336,
      "mean_token_accuracy": 0.11819689199328423,
      "num_tokens": 9472136.0,
      "step": 4370
    },
    {
      "entropy": 6.585649061203003,
      "epoch": 0.46818984429343463,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985224348294868,
      "loss": 6.5709,
      "mean_token_accuracy": 0.12418014109134674,
      "num_tokens": 9482602.0,
      "step": 4375
    },
    {
      "entropy": 6.748106908798218,
      "epoch": 0.46872491840119856,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004985180571333663,
      "loss": 6.5689,
      "mean_token_accuracy": 0.12485300302505493,
      "num_tokens": 9493782.0,
      "step": 4380
    },
    {
      "entropy": 6.540544319152832,
      "epoch": 0.4692599925089625,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004985136729831932,
      "loss": 6.5438,
      "mean_token_accuracy": 0.12441813126206398,
      "num_tokens": 9504389.0,
      "step": 4385
    },
    {
      "entropy": 6.671164274215698,
      "epoch": 0.46979506661672643,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004985092823790942,
      "loss": 6.6814,
      "mean_token_accuracy": 0.119224763661623,
      "num_tokens": 9515166.0,
      "step": 4390
    },
    {
      "entropy": 6.73720269203186,
      "epoch": 0.47033014072449036,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004985048853211961,
      "loss": 6.5746,
      "mean_token_accuracy": 0.12142504379153252,
      "num_tokens": 9525566.0,
      "step": 4395
    },
    {
      "entropy": 6.675395488739014,
      "epoch": 0.47086521483225424,
      "grad_norm": 1.0625,
      "learning_rate": 0.000498500481809626,
      "loss": 6.4819,
      "mean_token_accuracy": 0.13088136091828345,
      "num_tokens": 9536021.0,
      "step": 4400
    },
    {
      "entropy": 6.533463430404663,
      "epoch": 0.4714002889400182,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004984960718445107,
      "loss": 6.6052,
      "mean_token_accuracy": 0.1171295128762722,
      "num_tokens": 9546827.0,
      "step": 4405
    },
    {
      "entropy": 6.67804708480835,
      "epoch": 0.4719353630477821,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004984916554259778,
      "loss": 6.7149,
      "mean_token_accuracy": 0.11228696927428246,
      "num_tokens": 9558104.0,
      "step": 4410
    },
    {
      "entropy": 6.8455202102661135,
      "epoch": 0.47247043715554604,
      "grad_norm": 1.0,
      "learning_rate": 0.0004984872325541547,
      "loss": 6.6141,
      "mean_token_accuracy": 0.12217790111899376,
      "num_tokens": 9568487.0,
      "step": 4415
    },
    {
      "entropy": 6.582124662399292,
      "epoch": 0.47300551126331,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984828032291692,
      "loss": 6.5365,
      "mean_token_accuracy": 0.1269690603017807,
      "num_tokens": 9579509.0,
      "step": 4420
    },
    {
      "entropy": 6.651116609573364,
      "epoch": 0.4735405853710739,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984783674511492,
      "loss": 6.5747,
      "mean_token_accuracy": 0.12385042384266853,
      "num_tokens": 9589931.0,
      "step": 4425
    },
    {
      "entropy": 6.6950758934021,
      "epoch": 0.47407565947883784,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004984739252202227,
      "loss": 6.6175,
      "mean_token_accuracy": 0.12152868285775184,
      "num_tokens": 9599764.0,
      "step": 4430
    },
    {
      "entropy": 6.623296928405762,
      "epoch": 0.4746107335866017,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000498469476536518,
      "loss": 6.599,
      "mean_token_accuracy": 0.12063135579228401,
      "num_tokens": 9611145.0,
      "step": 4435
    },
    {
      "entropy": 6.677924060821534,
      "epoch": 0.47514580769436565,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984650214001636,
      "loss": 6.5834,
      "mean_token_accuracy": 0.12878919318318366,
      "num_tokens": 9621441.0,
      "step": 4440
    },
    {
      "entropy": 6.647787189483642,
      "epoch": 0.4756808818021296,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984605598112881,
      "loss": 6.6236,
      "mean_token_accuracy": 0.1271962121129036,
      "num_tokens": 9631874.0,
      "step": 4445
    },
    {
      "entropy": 6.630089616775512,
      "epoch": 0.4762159559098935,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984560917700204,
      "loss": 6.5916,
      "mean_token_accuracy": 0.11591947823762894,
      "num_tokens": 9643006.0,
      "step": 4450
    },
    {
      "entropy": 6.6707008361816404,
      "epoch": 0.47675103001765745,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004984516172764894,
      "loss": 6.4951,
      "mean_token_accuracy": 0.12678682953119277,
      "num_tokens": 9653897.0,
      "step": 4455
    },
    {
      "entropy": 6.604708528518676,
      "epoch": 0.4772861041254214,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004984471363308243,
      "loss": 6.5619,
      "mean_token_accuracy": 0.12140805572271347,
      "num_tokens": 9664460.0,
      "step": 4460
    },
    {
      "entropy": 6.661033821105957,
      "epoch": 0.4778211782331853,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004984426489331546,
      "loss": 6.5736,
      "mean_token_accuracy": 0.1274728685617447,
      "num_tokens": 9674534.0,
      "step": 4465
    },
    {
      "entropy": 6.736891746520996,
      "epoch": 0.4783562523409492,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984381550836097,
      "loss": 6.6574,
      "mean_token_accuracy": 0.11825382485985755,
      "num_tokens": 9684828.0,
      "step": 4470
    },
    {
      "entropy": 6.716025924682617,
      "epoch": 0.47889132644871313,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984336547823196,
      "loss": 6.6238,
      "mean_token_accuracy": 0.11967333108186722,
      "num_tokens": 9694536.0,
      "step": 4475
    },
    {
      "entropy": 6.664143896102905,
      "epoch": 0.47942640055647706,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000498429148029414,
      "loss": 6.483,
      "mean_token_accuracy": 0.12928671464323999,
      "num_tokens": 9705715.0,
      "step": 4480
    },
    {
      "entropy": 6.578292751312256,
      "epoch": 0.479961474664241,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984246348250233,
      "loss": 6.5722,
      "mean_token_accuracy": 0.12405260875821114,
      "num_tokens": 9717505.0,
      "step": 4485
    },
    {
      "entropy": 6.650796604156494,
      "epoch": 0.4804965487720049,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004984201151692775,
      "loss": 6.4341,
      "mean_token_accuracy": 0.12355747520923614,
      "num_tokens": 9727910.0,
      "step": 4490
    },
    {
      "entropy": 6.5766956329345705,
      "epoch": 0.48103162287976886,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004984155890623074,
      "loss": 6.5412,
      "mean_token_accuracy": 0.12643233090639114,
      "num_tokens": 9738528.0,
      "step": 4495
    },
    {
      "entropy": 6.597425556182861,
      "epoch": 0.4815666969875328,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004984110565042435,
      "loss": 6.5686,
      "mean_token_accuracy": 0.12338585183024406,
      "num_tokens": 9749744.0,
      "step": 4500
    },
    {
      "entropy": 6.71343297958374,
      "epoch": 0.48210177109529667,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004984065174952168,
      "loss": 6.5795,
      "mean_token_accuracy": 0.11854353696107864,
      "num_tokens": 9759968.0,
      "step": 4505
    },
    {
      "entropy": 6.6341510772705075,
      "epoch": 0.4826368452030606,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984019720353582,
      "loss": 6.5911,
      "mean_token_accuracy": 0.11985000744462013,
      "num_tokens": 9769791.0,
      "step": 4510
    },
    {
      "entropy": 6.594589281082153,
      "epoch": 0.48317191931082454,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004983974201247991,
      "loss": 6.5564,
      "mean_token_accuracy": 0.1293729916214943,
      "num_tokens": 9780369.0,
      "step": 4515
    },
    {
      "entropy": 6.650420045852661,
      "epoch": 0.48370699341858847,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004983928617636709,
      "loss": 6.6077,
      "mean_token_accuracy": 0.11779389381408692,
      "num_tokens": 9791125.0,
      "step": 4520
    },
    {
      "entropy": 6.741210079193115,
      "epoch": 0.4842420675263524,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983882969521052,
      "loss": 6.6039,
      "mean_token_accuracy": 0.12302820384502411,
      "num_tokens": 9801697.0,
      "step": 4525
    },
    {
      "entropy": 6.714509773254394,
      "epoch": 0.48477714163411634,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983837256902338,
      "loss": 6.6394,
      "mean_token_accuracy": 0.12414250075817108,
      "num_tokens": 9812171.0,
      "step": 4530
    },
    {
      "entropy": 6.612561130523682,
      "epoch": 0.48531221574188027,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983791479781887,
      "loss": 6.4865,
      "mean_token_accuracy": 0.12814272344112396,
      "num_tokens": 9821596.0,
      "step": 4535
    },
    {
      "entropy": 6.6429736614227295,
      "epoch": 0.48584728984964415,
      "grad_norm": 1.0,
      "learning_rate": 0.0004983745638161021,
      "loss": 6.5942,
      "mean_token_accuracy": 0.11311374977231026,
      "num_tokens": 9831634.0,
      "step": 4540
    },
    {
      "entropy": 6.680133295059204,
      "epoch": 0.4863823639574081,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004983699732041064,
      "loss": 6.5608,
      "mean_token_accuracy": 0.12484560459852219,
      "num_tokens": 9841662.0,
      "step": 4545
    },
    {
      "entropy": 6.665322494506836,
      "epoch": 0.486917438065172,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000498365376142334,
      "loss": 6.5451,
      "mean_token_accuracy": 0.12128180265426636,
      "num_tokens": 9853797.0,
      "step": 4550
    },
    {
      "entropy": 6.616853523254394,
      "epoch": 0.48745251217293595,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004983607726309179,
      "loss": 6.597,
      "mean_token_accuracy": 0.12437746599316597,
      "num_tokens": 9865245.0,
      "step": 4555
    },
    {
      "entropy": 6.767341279983521,
      "epoch": 0.4879875862806999,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004983561626699908,
      "loss": 6.674,
      "mean_token_accuracy": 0.11401030793786049,
      "num_tokens": 9876207.0,
      "step": 4560
    },
    {
      "entropy": 6.63656063079834,
      "epoch": 0.4885226603884638,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004983515462596859,
      "loss": 6.5099,
      "mean_token_accuracy": 0.12511964961886407,
      "num_tokens": 9887210.0,
      "step": 4565
    },
    {
      "entropy": 6.625952911376953,
      "epoch": 0.48905773449622775,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983469234001365,
      "loss": 6.6882,
      "mean_token_accuracy": 0.11470266729593277,
      "num_tokens": 9899607.0,
      "step": 4570
    },
    {
      "entropy": 6.559663248062134,
      "epoch": 0.4895928086039917,
      "grad_norm": 1.171875,
      "learning_rate": 0.000498342294091476,
      "loss": 6.466,
      "mean_token_accuracy": 0.13191099166870118,
      "num_tokens": 9909910.0,
      "step": 4575
    },
    {
      "entropy": 6.581341791152954,
      "epoch": 0.49012788271175556,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983376583338382,
      "loss": 6.5956,
      "mean_token_accuracy": 0.11748891472816467,
      "num_tokens": 9922137.0,
      "step": 4580
    },
    {
      "entropy": 6.6524818420410154,
      "epoch": 0.4906629568195195,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983330161273569,
      "loss": 6.5779,
      "mean_token_accuracy": 0.12338467091321945,
      "num_tokens": 9932514.0,
      "step": 4585
    },
    {
      "entropy": 6.672944164276123,
      "epoch": 0.4911980309272834,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004983283674721662,
      "loss": 6.6145,
      "mean_token_accuracy": 0.12371377795934677,
      "num_tokens": 9943054.0,
      "step": 4590
    },
    {
      "entropy": 6.685930919647217,
      "epoch": 0.49173310503504736,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004983237123684002,
      "loss": 6.583,
      "mean_token_accuracy": 0.11797807738184929,
      "num_tokens": 9953791.0,
      "step": 4595
    },
    {
      "entropy": 6.595043659210205,
      "epoch": 0.4922681791428113,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983190508161934,
      "loss": 6.5268,
      "mean_token_accuracy": 0.12165041193366051,
      "num_tokens": 9964782.0,
      "step": 4600
    },
    {
      "entropy": 6.681060695648194,
      "epoch": 0.4928032532505752,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004983143828156804,
      "loss": 6.6055,
      "mean_token_accuracy": 0.1215230606496334,
      "num_tokens": 9975843.0,
      "step": 4605
    },
    {
      "entropy": 6.689913415908814,
      "epoch": 0.49333832735833916,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983097083669959,
      "loss": 6.5777,
      "mean_token_accuracy": 0.12538425028324127,
      "num_tokens": 9987060.0,
      "step": 4610
    },
    {
      "entropy": 6.666298151016235,
      "epoch": 0.49387340146610303,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983050274702751,
      "loss": 6.5579,
      "mean_token_accuracy": 0.1193149946630001,
      "num_tokens": 9997380.0,
      "step": 4615
    },
    {
      "entropy": 6.622103214263916,
      "epoch": 0.49440847557386697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004983003401256529,
      "loss": 6.6041,
      "mean_token_accuracy": 0.1213637612760067,
      "num_tokens": 10007955.0,
      "step": 4620
    },
    {
      "entropy": 6.796403980255127,
      "epoch": 0.4949435496816309,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004982956463332647,
      "loss": 6.6785,
      "mean_token_accuracy": 0.12040725573897362,
      "num_tokens": 10017645.0,
      "step": 4625
    },
    {
      "entropy": 6.627289819717407,
      "epoch": 0.49547862378939483,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004982909460932463,
      "loss": 6.6096,
      "mean_token_accuracy": 0.12429628074169159,
      "num_tokens": 10027619.0,
      "step": 4630
    },
    {
      "entropy": 6.590019702911377,
      "epoch": 0.49601369789715877,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498286239405733,
      "loss": 6.4729,
      "mean_token_accuracy": 0.13149576336145402,
      "num_tokens": 10037551.0,
      "step": 4635
    },
    {
      "entropy": 6.616154289245605,
      "epoch": 0.4965487720049227,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498281526270861,
      "loss": 6.5099,
      "mean_token_accuracy": 0.13509751111268997,
      "num_tokens": 10047771.0,
      "step": 4640
    },
    {
      "entropy": 6.767517042160034,
      "epoch": 0.49708384611268663,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004982768066887663,
      "loss": 6.6369,
      "mean_token_accuracy": 0.11773812174797058,
      "num_tokens": 10058771.0,
      "step": 4645
    },
    {
      "entropy": 6.55728063583374,
      "epoch": 0.4976189202204505,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982720806595851,
      "loss": 6.5014,
      "mean_token_accuracy": 0.12969422712922096,
      "num_tokens": 10070318.0,
      "step": 4650
    },
    {
      "entropy": 6.569836902618408,
      "epoch": 0.49815399432821444,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004982673481834541,
      "loss": 6.5496,
      "mean_token_accuracy": 0.1241609700024128,
      "num_tokens": 10081810.0,
      "step": 4655
    },
    {
      "entropy": 6.622956085205078,
      "epoch": 0.4986890684359784,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004982626092605097,
      "loss": 6.5524,
      "mean_token_accuracy": 0.12292885780334473,
      "num_tokens": 10091724.0,
      "step": 4660
    },
    {
      "entropy": 6.6552161693573,
      "epoch": 0.4992241425437423,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982578638908888,
      "loss": 6.5478,
      "mean_token_accuracy": 0.12208108678460121,
      "num_tokens": 10102510.0,
      "step": 4665
    },
    {
      "entropy": 6.660357189178467,
      "epoch": 0.49975921665150624,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004982531120747286,
      "loss": 6.6188,
      "mean_token_accuracy": 0.11734322607517242,
      "num_tokens": 10112636.0,
      "step": 4670
    },
    {
      "entropy": 6.636687994003296,
      "epoch": 0.5002942907592701,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498248353812166,
      "loss": 6.6127,
      "mean_token_accuracy": 0.1258271850645542,
      "num_tokens": 10123562.0,
      "step": 4675
    },
    {
      "entropy": 6.608578443527222,
      "epoch": 0.5008293648670341,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004982435891033387,
      "loss": 6.5729,
      "mean_token_accuracy": 0.11916021406650543,
      "num_tokens": 10134731.0,
      "step": 4680
    },
    {
      "entropy": 6.676313686370849,
      "epoch": 0.501364438974798,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000498238817948384,
      "loss": 6.5562,
      "mean_token_accuracy": 0.11949019730091096,
      "num_tokens": 10145232.0,
      "step": 4685
    },
    {
      "entropy": 6.622157955169678,
      "epoch": 0.501899513082562,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982340403474398,
      "loss": 6.5497,
      "mean_token_accuracy": 0.11989447847008705,
      "num_tokens": 10155857.0,
      "step": 4690
    },
    {
      "entropy": 6.658979082107544,
      "epoch": 0.5024345871903259,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004982292563006441,
      "loss": 6.5304,
      "mean_token_accuracy": 0.1236383281648159,
      "num_tokens": 10166145.0,
      "step": 4695
    },
    {
      "entropy": 6.685210704803467,
      "epoch": 0.5029696612980897,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000498224465808135,
      "loss": 6.6729,
      "mean_token_accuracy": 0.1140787348151207,
      "num_tokens": 10177086.0,
      "step": 4700
    },
    {
      "entropy": 6.685143899917603,
      "epoch": 0.5035047354058537,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982196688700509,
      "loss": 6.5409,
      "mean_token_accuracy": 0.12187634781002998,
      "num_tokens": 10187120.0,
      "step": 4705
    },
    {
      "entropy": 6.635204744338989,
      "epoch": 0.5040398095136176,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004982148654865302,
      "loss": 6.608,
      "mean_token_accuracy": 0.1216482937335968,
      "num_tokens": 10197784.0,
      "step": 4710
    },
    {
      "entropy": 6.670889568328858,
      "epoch": 0.5045748836213816,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004982100556577115,
      "loss": 6.4635,
      "mean_token_accuracy": 0.12821068465709687,
      "num_tokens": 10209373.0,
      "step": 4715
    },
    {
      "entropy": 6.615971565246582,
      "epoch": 0.5051099577291455,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004982052393837338,
      "loss": 6.5939,
      "mean_token_accuracy": 0.12902486696839333,
      "num_tokens": 10220316.0,
      "step": 4720
    },
    {
      "entropy": 6.600869941711426,
      "epoch": 0.5056450318369095,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982004166647363,
      "loss": 6.4836,
      "mean_token_accuracy": 0.13095228374004364,
      "num_tokens": 10230478.0,
      "step": 4725
    },
    {
      "entropy": 6.5474138259887695,
      "epoch": 0.5061801059446733,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981955875008582,
      "loss": 6.5734,
      "mean_token_accuracy": 0.11955861896276473,
      "num_tokens": 10240641.0,
      "step": 4730
    },
    {
      "entropy": 6.645769548416138,
      "epoch": 0.5067151800524372,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004981907518922387,
      "loss": 6.5441,
      "mean_token_accuracy": 0.13211010918021202,
      "num_tokens": 10250676.0,
      "step": 4735
    },
    {
      "entropy": 6.763848876953125,
      "epoch": 0.5072502541602012,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004981859098390177,
      "loss": 6.6238,
      "mean_token_accuracy": 0.11968207731842995,
      "num_tokens": 10262016.0,
      "step": 4740
    },
    {
      "entropy": 6.6407387256622314,
      "epoch": 0.5077853282679651,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004981810613413348,
      "loss": 6.5719,
      "mean_token_accuracy": 0.12192937582731248,
      "num_tokens": 10271985.0,
      "step": 4745
    },
    {
      "entropy": 6.645168304443359,
      "epoch": 0.5083204023757291,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004981762063993302,
      "loss": 6.5566,
      "mean_token_accuracy": 0.12744775116443635,
      "num_tokens": 10282156.0,
      "step": 4750
    },
    {
      "entropy": 6.597253704071045,
      "epoch": 0.5088554764834929,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000498171345013144,
      "loss": 6.549,
      "mean_token_accuracy": 0.11908648312091827,
      "num_tokens": 10292576.0,
      "step": 4755
    },
    {
      "entropy": 6.554513072967529,
      "epoch": 0.5093905505912569,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004981664771829165,
      "loss": 6.5546,
      "mean_token_accuracy": 0.12112470418214798,
      "num_tokens": 10303648.0,
      "step": 4760
    },
    {
      "entropy": 6.619035243988037,
      "epoch": 0.5099256246990208,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004981616029087884,
      "loss": 6.4427,
      "mean_token_accuracy": 0.13242612406611443,
      "num_tokens": 10314275.0,
      "step": 4765
    },
    {
      "entropy": 6.590420579910278,
      "epoch": 0.5104606988067847,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981567221909004,
      "loss": 6.5793,
      "mean_token_accuracy": 0.1155736081302166,
      "num_tokens": 10325386.0,
      "step": 4770
    },
    {
      "entropy": 6.649965286254883,
      "epoch": 0.5109957729145487,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004981518350293935,
      "loss": 6.5544,
      "mean_token_accuracy": 0.11998137161135673,
      "num_tokens": 10335041.0,
      "step": 4775
    },
    {
      "entropy": 6.641874313354492,
      "epoch": 0.5115308470223126,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004981469414244086,
      "loss": 6.5078,
      "mean_token_accuracy": 0.12559844851493834,
      "num_tokens": 10345746.0,
      "step": 4780
    },
    {
      "entropy": 6.562429475784302,
      "epoch": 0.5120659211300765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498142041376087,
      "loss": 6.6167,
      "mean_token_accuracy": 0.12392572686076164,
      "num_tokens": 10356995.0,
      "step": 4785
    },
    {
      "entropy": 6.646564197540283,
      "epoch": 0.5126009952378404,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004981371348845705,
      "loss": 6.521,
      "mean_token_accuracy": 0.12766078263521194,
      "num_tokens": 10368445.0,
      "step": 4790
    },
    {
      "entropy": 6.618320369720459,
      "epoch": 0.5131360693456044,
      "grad_norm": 1.0,
      "learning_rate": 0.0004981322219500006,
      "loss": 6.451,
      "mean_token_accuracy": 0.13063669577240944,
      "num_tokens": 10378256.0,
      "step": 4795
    },
    {
      "entropy": 6.618687152862549,
      "epoch": 0.5136711434533683,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498127302572519,
      "loss": 6.5504,
      "mean_token_accuracy": 0.12561604380607605,
      "num_tokens": 10388698.0,
      "step": 4800
    },
    {
      "entropy": 6.581791400909424,
      "epoch": 0.5142062175611322,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004981223767522679,
      "loss": 6.5474,
      "mean_token_accuracy": 0.12754715532064437,
      "num_tokens": 10399975.0,
      "step": 4805
    },
    {
      "entropy": 6.6619141578674315,
      "epoch": 0.5147412916688962,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004981174444893896,
      "loss": 6.5771,
      "mean_token_accuracy": 0.12108841314911842,
      "num_tokens": 10410480.0,
      "step": 4810
    },
    {
      "entropy": 6.621873950958252,
      "epoch": 0.51527636577666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004981125057840264,
      "loss": 6.5255,
      "mean_token_accuracy": 0.12419212684035301,
      "num_tokens": 10421023.0,
      "step": 4815
    },
    {
      "entropy": 6.609471273422241,
      "epoch": 0.515811439884424,
      "grad_norm": 1.328125,
      "learning_rate": 0.000498107560636321,
      "loss": 6.5677,
      "mean_token_accuracy": 0.12525818571448327,
      "num_tokens": 10431152.0,
      "step": 4820
    },
    {
      "entropy": 6.599048805236817,
      "epoch": 0.5163465139921879,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004981026090464161,
      "loss": 6.4636,
      "mean_token_accuracy": 0.1198187604546547,
      "num_tokens": 10441986.0,
      "step": 4825
    },
    {
      "entropy": 6.625163555145264,
      "epoch": 0.5168815880999519,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980976510144548,
      "loss": 6.5606,
      "mean_token_accuracy": 0.11982694864273072,
      "num_tokens": 10452932.0,
      "step": 4830
    },
    {
      "entropy": 6.682273530960083,
      "epoch": 0.5174166622077158,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980926865405801,
      "loss": 6.6225,
      "mean_token_accuracy": 0.12185526937246323,
      "num_tokens": 10463411.0,
      "step": 4835
    },
    {
      "entropy": 6.644960546493531,
      "epoch": 0.5179517363154797,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004980877156249354,
      "loss": 6.5179,
      "mean_token_accuracy": 0.1238639697432518,
      "num_tokens": 10474820.0,
      "step": 4840
    },
    {
      "entropy": 6.58879828453064,
      "epoch": 0.5184868104232436,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004980827382676643,
      "loss": 6.5409,
      "mean_token_accuracy": 0.12410817965865135,
      "num_tokens": 10485701.0,
      "step": 4845
    },
    {
      "entropy": 6.618032360076905,
      "epoch": 0.5190218845310075,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004980777544689105,
      "loss": 6.5149,
      "mean_token_accuracy": 0.12538835555315017,
      "num_tokens": 10495825.0,
      "step": 4850
    },
    {
      "entropy": 6.606825733184815,
      "epoch": 0.5195569586387715,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004980727642288178,
      "loss": 6.5051,
      "mean_token_accuracy": 0.1290580116212368,
      "num_tokens": 10505847.0,
      "step": 4855
    },
    {
      "entropy": 6.601777362823486,
      "epoch": 0.5200920327465354,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004980677675475305,
      "loss": 6.5648,
      "mean_token_accuracy": 0.12240067943930626,
      "num_tokens": 10516596.0,
      "step": 4860
    },
    {
      "entropy": 6.600971317291259,
      "epoch": 0.5206271068542994,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980627644251926,
      "loss": 6.4615,
      "mean_token_accuracy": 0.13353928551077843,
      "num_tokens": 10528175.0,
      "step": 4865
    },
    {
      "entropy": 6.54287166595459,
      "epoch": 0.5211621809620632,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980577548619489,
      "loss": 6.5695,
      "mean_token_accuracy": 0.12125243842601777,
      "num_tokens": 10540076.0,
      "step": 4870
    },
    {
      "entropy": 6.631709575653076,
      "epoch": 0.5216972550698272,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980527388579437,
      "loss": 6.5854,
      "mean_token_accuracy": 0.12283660843968391,
      "num_tokens": 10552025.0,
      "step": 4875
    },
    {
      "entropy": 6.649728012084961,
      "epoch": 0.5222323291775911,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980477164133221,
      "loss": 6.5795,
      "mean_token_accuracy": 0.12160530313849449,
      "num_tokens": 10562535.0,
      "step": 4880
    },
    {
      "entropy": 6.612128973007202,
      "epoch": 0.522767403285355,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000498042687528229,
      "loss": 6.5045,
      "mean_token_accuracy": 0.12807957753539084,
      "num_tokens": 10572989.0,
      "step": 4885
    },
    {
      "entropy": 6.628624200820923,
      "epoch": 0.523302477393119,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004980376522028098,
      "loss": 6.5237,
      "mean_token_accuracy": 0.12581543624401093,
      "num_tokens": 10583033.0,
      "step": 4890
    },
    {
      "entropy": 6.641993856430053,
      "epoch": 0.5238375515008828,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004980326104372095,
      "loss": 6.5893,
      "mean_token_accuracy": 0.12306767404079437,
      "num_tokens": 10595205.0,
      "step": 4895
    },
    {
      "entropy": 6.61875,
      "epoch": 0.5243726256086468,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000498027562231574,
      "loss": 6.4629,
      "mean_token_accuracy": 0.12763072922825813,
      "num_tokens": 10605244.0,
      "step": 4900
    },
    {
      "entropy": 6.586288261413574,
      "epoch": 0.5249076997164107,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498022507586049,
      "loss": 6.5499,
      "mean_token_accuracy": 0.12274945080280304,
      "num_tokens": 10615817.0,
      "step": 4905
    },
    {
      "entropy": 6.616854333877564,
      "epoch": 0.5254427738241747,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980174465007804,
      "loss": 6.5486,
      "mean_token_accuracy": 0.12240270897746086,
      "num_tokens": 10627160.0,
      "step": 4910
    },
    {
      "entropy": 6.639895439147949,
      "epoch": 0.5259778479319386,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004980123789759145,
      "loss": 6.5643,
      "mean_token_accuracy": 0.12684085741639137,
      "num_tokens": 10637552.0,
      "step": 4915
    },
    {
      "entropy": 6.608394575119019,
      "epoch": 0.5265129220397025,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004980073050115973,
      "loss": 6.5872,
      "mean_token_accuracy": 0.12828944325447084,
      "num_tokens": 10648773.0,
      "step": 4920
    },
    {
      "entropy": 6.694124412536621,
      "epoch": 0.5270479961474664,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980022246079755,
      "loss": 6.5759,
      "mean_token_accuracy": 0.12196737080812455,
      "num_tokens": 10659537.0,
      "step": 4925
    },
    {
      "entropy": 6.687956190109253,
      "epoch": 0.5275830702552303,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000497997137765196,
      "loss": 6.6109,
      "mean_token_accuracy": 0.11862236484885216,
      "num_tokens": 10671390.0,
      "step": 4930
    },
    {
      "entropy": 6.562083435058594,
      "epoch": 0.5281181443629943,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004979920444834054,
      "loss": 6.4562,
      "mean_token_accuracy": 0.13357677832245826,
      "num_tokens": 10682389.0,
      "step": 4935
    },
    {
      "entropy": 6.591794204711914,
      "epoch": 0.5286532184707582,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004979869447627508,
      "loss": 6.5156,
      "mean_token_accuracy": 0.12352539300918579,
      "num_tokens": 10692400.0,
      "step": 4940
    },
    {
      "entropy": 6.567591524124145,
      "epoch": 0.5291882925785222,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004979818386033795,
      "loss": 6.4585,
      "mean_token_accuracy": 0.12442874908447266,
      "num_tokens": 10702396.0,
      "step": 4945
    },
    {
      "entropy": 6.578789091110229,
      "epoch": 0.5297233666862861,
      "grad_norm": 1.171875,
      "learning_rate": 0.000497976726005439,
      "loss": 6.5628,
      "mean_token_accuracy": 0.12618231922388076,
      "num_tokens": 10712863.0,
      "step": 4950
    },
    {
      "entropy": 6.701795530319214,
      "epoch": 0.5302584407940499,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004979716069690768,
      "loss": 6.5763,
      "mean_token_accuracy": 0.12288771569728851,
      "num_tokens": 10724234.0,
      "step": 4955
    },
    {
      "entropy": 6.726573753356933,
      "epoch": 0.5307935149018139,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004979664814944409,
      "loss": 6.6071,
      "mean_token_accuracy": 0.11938823983073235,
      "num_tokens": 10734405.0,
      "step": 4960
    },
    {
      "entropy": 6.736481618881226,
      "epoch": 0.5313285890095778,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004979613495816791,
      "loss": 6.5528,
      "mean_token_accuracy": 0.11843769699335098,
      "num_tokens": 10744573.0,
      "step": 4965
    },
    {
      "entropy": 6.623452425003052,
      "epoch": 0.5318636631173418,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004979562112309397,
      "loss": 6.6006,
      "mean_token_accuracy": 0.12156387493014335,
      "num_tokens": 10755555.0,
      "step": 4970
    },
    {
      "entropy": 6.588447856903076,
      "epoch": 0.5323987372251057,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000497951066442371,
      "loss": 6.6136,
      "mean_token_accuracy": 0.11861244738101959,
      "num_tokens": 10766493.0,
      "step": 4975
    },
    {
      "entropy": 6.716310358047485,
      "epoch": 0.5329338113328697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004979459152161215,
      "loss": 6.5755,
      "mean_token_accuracy": 0.11967594027519227,
      "num_tokens": 10776735.0,
      "step": 4980
    },
    {
      "entropy": 6.614170980453491,
      "epoch": 0.5334688854406335,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004979407575523402,
      "loss": 6.5516,
      "mean_token_accuracy": 0.11883142217993736,
      "num_tokens": 10787500.0,
      "step": 4985
    },
    {
      "entropy": 6.567416620254517,
      "epoch": 0.5340039595483974,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004979355934511757,
      "loss": 6.5498,
      "mean_token_accuracy": 0.12720205932855605,
      "num_tokens": 10798002.0,
      "step": 4990
    },
    {
      "entropy": 6.599498462677002,
      "epoch": 0.5345390336561614,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979304229127773,
      "loss": 6.5352,
      "mean_token_accuracy": 0.12421434372663498,
      "num_tokens": 10808982.0,
      "step": 4995
    },
    {
      "entropy": 6.6084154605865475,
      "epoch": 0.5350741077639253,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004979252459372944,
      "loss": 6.4957,
      "mean_token_accuracy": 0.12441762760281563,
      "num_tokens": 10821619.0,
      "step": 5000
    },
    {
      "entropy": 6.626148271560669,
      "epoch": 0.5356091818716893,
      "grad_norm": 1.0,
      "learning_rate": 0.0004979200625248761,
      "loss": 6.5066,
      "mean_token_accuracy": 0.12545772269368172,
      "num_tokens": 10832344.0,
      "step": 5005
    },
    {
      "entropy": 6.571537780761719,
      "epoch": 0.5361442559794531,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004979148726756725,
      "loss": 6.473,
      "mean_token_accuracy": 0.1212929643690586,
      "num_tokens": 10843149.0,
      "step": 5010
    },
    {
      "entropy": 6.5270911693573,
      "epoch": 0.5366793300872171,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004979096763898333,
      "loss": 6.5281,
      "mean_token_accuracy": 0.12871197313070298,
      "num_tokens": 10854590.0,
      "step": 5015
    },
    {
      "entropy": 6.68946328163147,
      "epoch": 0.537214404194981,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004979044736675084,
      "loss": 6.4971,
      "mean_token_accuracy": 0.12162916138768196,
      "num_tokens": 10865500.0,
      "step": 5020
    },
    {
      "entropy": 6.583962106704712,
      "epoch": 0.5377494783027449,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004978992645088483,
      "loss": 6.5357,
      "mean_token_accuracy": 0.1269901439547539,
      "num_tokens": 10876866.0,
      "step": 5025
    },
    {
      "entropy": 6.472790622711182,
      "epoch": 0.5382845524105089,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000497894048914003,
      "loss": 6.4848,
      "mean_token_accuracy": 0.1286119908094406,
      "num_tokens": 10888627.0,
      "step": 5030
    },
    {
      "entropy": 6.6368138790130615,
      "epoch": 0.5388196265182728,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004978888268831236,
      "loss": 6.603,
      "mean_token_accuracy": 0.11341231167316437,
      "num_tokens": 10900992.0,
      "step": 5035
    },
    {
      "entropy": 6.712962627410889,
      "epoch": 0.5393547006260367,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004978835984163606,
      "loss": 6.6514,
      "mean_token_accuracy": 0.11450790166854859,
      "num_tokens": 10912323.0,
      "step": 5040
    },
    {
      "entropy": 6.680682039260864,
      "epoch": 0.5398897747338006,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004978783635138649,
      "loss": 6.5843,
      "mean_token_accuracy": 0.12086773961782456,
      "num_tokens": 10923991.0,
      "step": 5045
    },
    {
      "entropy": 6.537887716293335,
      "epoch": 0.5404248488415646,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004978731221757878,
      "loss": 6.4308,
      "mean_token_accuracy": 0.13115262985229492,
      "num_tokens": 10934081.0,
      "step": 5050
    },
    {
      "entropy": 6.602122402191162,
      "epoch": 0.5409599229493285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004978678744022808,
      "loss": 6.5121,
      "mean_token_accuracy": 0.1250887967646122,
      "num_tokens": 10944729.0,
      "step": 5055
    },
    {
      "entropy": 6.644594287872314,
      "epoch": 0.5414949970570924,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000497862620193495,
      "loss": 6.5758,
      "mean_token_accuracy": 0.12274185344576835,
      "num_tokens": 10955708.0,
      "step": 5060
    },
    {
      "entropy": 6.630021095275879,
      "epoch": 0.5420300711648564,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004978573595495826,
      "loss": 6.5363,
      "mean_token_accuracy": 0.1229963481426239,
      "num_tokens": 10966427.0,
      "step": 5065
    },
    {
      "entropy": 6.6108145236969,
      "epoch": 0.5425651452726202,
      "grad_norm": 1.0625,
      "learning_rate": 0.000497852092470695,
      "loss": 6.5493,
      "mean_token_accuracy": 0.1274164654314518,
      "num_tokens": 10977741.0,
      "step": 5070
    },
    {
      "entropy": 6.687741661071778,
      "epoch": 0.5431002193803842,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004978468189569847,
      "loss": 6.5148,
      "mean_token_accuracy": 0.12572550773620605,
      "num_tokens": 10989127.0,
      "step": 5075
    },
    {
      "entropy": 6.562218809127808,
      "epoch": 0.5436352934881481,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004978415390086038,
      "loss": 6.5308,
      "mean_token_accuracy": 0.12041111961007119,
      "num_tokens": 10999791.0,
      "step": 5080
    },
    {
      "entropy": 6.656288146972656,
      "epoch": 0.5441703675959121,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004978362526257047,
      "loss": 6.5999,
      "mean_token_accuracy": 0.11804638057947159,
      "num_tokens": 11010045.0,
      "step": 5085
    },
    {
      "entropy": 6.667946529388428,
      "epoch": 0.544705441703676,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004978309598084403,
      "loss": 6.5242,
      "mean_token_accuracy": 0.12063587233424186,
      "num_tokens": 11020729.0,
      "step": 5090
    },
    {
      "entropy": 6.624614810943603,
      "epoch": 0.5452405158114398,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004978256605569631,
      "loss": 6.5328,
      "mean_token_accuracy": 0.1263642780482769,
      "num_tokens": 11030641.0,
      "step": 5095
    },
    {
      "entropy": 6.63623456954956,
      "epoch": 0.5457755899192038,
      "grad_norm": 1.0,
      "learning_rate": 0.0004978203548714262,
      "loss": 6.5354,
      "mean_token_accuracy": 0.12168349623680115,
      "num_tokens": 11041533.0,
      "step": 5100
    },
    {
      "entropy": 6.604489040374756,
      "epoch": 0.5463106640269677,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000497815042751983,
      "loss": 6.4924,
      "mean_token_accuracy": 0.12448608353734017,
      "num_tokens": 11052170.0,
      "step": 5105
    },
    {
      "entropy": 6.556596899032593,
      "epoch": 0.5468457381347317,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004978097241987868,
      "loss": 6.4707,
      "mean_token_accuracy": 0.12820492386817933,
      "num_tokens": 11062872.0,
      "step": 5110
    },
    {
      "entropy": 6.61461443901062,
      "epoch": 0.5473808122424956,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000497804399211991,
      "loss": 6.5735,
      "mean_token_accuracy": 0.1245592214167118,
      "num_tokens": 11074199.0,
      "step": 5115
    },
    {
      "entropy": 6.614849328994751,
      "epoch": 0.5479158863502596,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004977990677917495,
      "loss": 6.5189,
      "mean_token_accuracy": 0.1276326894760132,
      "num_tokens": 11083881.0,
      "step": 5120
    },
    {
      "entropy": 6.584414482116699,
      "epoch": 0.5484509604580234,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004977937299382162,
      "loss": 6.5657,
      "mean_token_accuracy": 0.12166793793439865,
      "num_tokens": 11094374.0,
      "step": 5125
    },
    {
      "entropy": 6.642978191375732,
      "epoch": 0.5489860345657873,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004977883856515453,
      "loss": 6.4124,
      "mean_token_accuracy": 0.13538418114185333,
      "num_tokens": 11105174.0,
      "step": 5130
    },
    {
      "entropy": 6.577599334716797,
      "epoch": 0.5495211086735513,
      "grad_norm": 1.109375,
      "learning_rate": 0.000497783034931891,
      "loss": 6.575,
      "mean_token_accuracy": 0.1258682705461979,
      "num_tokens": 11117223.0,
      "step": 5135
    },
    {
      "entropy": 6.601963710784912,
      "epoch": 0.5500561827813152,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497777677779408,
      "loss": 6.4921,
      "mean_token_accuracy": 0.1251679018139839,
      "num_tokens": 11127240.0,
      "step": 5140
    },
    {
      "entropy": 6.6047278881073,
      "epoch": 0.5505912568890792,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004977723141942509,
      "loss": 6.4896,
      "mean_token_accuracy": 0.13126735910773277,
      "num_tokens": 11137287.0,
      "step": 5145
    },
    {
      "entropy": 6.602693653106689,
      "epoch": 0.551126330996843,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004977669441765743,
      "loss": 6.6306,
      "mean_token_accuracy": 0.11842129975557328,
      "num_tokens": 11148988.0,
      "step": 5150
    },
    {
      "entropy": 6.683191967010498,
      "epoch": 0.551661405104607,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004977615677265336,
      "loss": 6.5534,
      "mean_token_accuracy": 0.12478872463107109,
      "num_tokens": 11159874.0,
      "step": 5155
    },
    {
      "entropy": 6.675896072387696,
      "epoch": 0.5521964792123709,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000497756184844284,
      "loss": 6.5035,
      "mean_token_accuracy": 0.12875596135854722,
      "num_tokens": 11170703.0,
      "step": 5160
    },
    {
      "entropy": 6.495262527465821,
      "epoch": 0.5527315533201348,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004977507955299809,
      "loss": 6.4019,
      "mean_token_accuracy": 0.135747130215168,
      "num_tokens": 11182230.0,
      "step": 5165
    },
    {
      "entropy": 6.565120220184326,
      "epoch": 0.5532666274278988,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004977453997837797,
      "loss": 6.4878,
      "mean_token_accuracy": 0.1157505251467228,
      "num_tokens": 11192897.0,
      "step": 5170
    },
    {
      "entropy": 6.5999902248382565,
      "epoch": 0.5538017015356627,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004977399976058366,
      "loss": 6.5729,
      "mean_token_accuracy": 0.12268262952566147,
      "num_tokens": 11203942.0,
      "step": 5175
    },
    {
      "entropy": 6.653809833526611,
      "epoch": 0.5543367756434266,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004977345889963072,
      "loss": 6.6084,
      "mean_token_accuracy": 0.11838528290390968,
      "num_tokens": 11214361.0,
      "step": 5180
    },
    {
      "entropy": 6.599223709106445,
      "epoch": 0.5548718497511905,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004977291739553479,
      "loss": 6.5105,
      "mean_token_accuracy": 0.12759996205568314,
      "num_tokens": 11225928.0,
      "step": 5185
    },
    {
      "entropy": 6.576416730880737,
      "epoch": 0.5554069238589545,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004977237524831149,
      "loss": 6.4664,
      "mean_token_accuracy": 0.11715415790677071,
      "num_tokens": 11236660.0,
      "step": 5190
    },
    {
      "entropy": 6.529564666748047,
      "epoch": 0.5559419979667184,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004977183245797649,
      "loss": 6.4845,
      "mean_token_accuracy": 0.12914775237441062,
      "num_tokens": 11247272.0,
      "step": 5195
    },
    {
      "entropy": 6.670500707626343,
      "epoch": 0.5564770720744823,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004977128902454547,
      "loss": 6.5896,
      "mean_token_accuracy": 0.11652119681239129,
      "num_tokens": 11258472.0,
      "step": 5200
    },
    {
      "entropy": 6.642084646224975,
      "epoch": 0.5570121461822463,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004977074494803409,
      "loss": 6.4799,
      "mean_token_accuracy": 0.13261723890900612,
      "num_tokens": 11268344.0,
      "step": 5205
    },
    {
      "entropy": 6.603341007232666,
      "epoch": 0.5575472202900101,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004977020022845809,
      "loss": 6.5984,
      "mean_token_accuracy": 0.11450821608304977,
      "num_tokens": 11279931.0,
      "step": 5210
    },
    {
      "entropy": 6.650809907913208,
      "epoch": 0.5580822943977741,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004976965486583318,
      "loss": 6.5288,
      "mean_token_accuracy": 0.12144831568002701,
      "num_tokens": 11290595.0,
      "step": 5215
    },
    {
      "entropy": 6.600453662872314,
      "epoch": 0.558617368505538,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004976910886017513,
      "loss": 6.5573,
      "mean_token_accuracy": 0.12385655269026756,
      "num_tokens": 11300774.0,
      "step": 5220
    },
    {
      "entropy": 6.630309820175171,
      "epoch": 0.559152442613302,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004976856221149969,
      "loss": 6.4384,
      "mean_token_accuracy": 0.13099406361579896,
      "num_tokens": 11310288.0,
      "step": 5225
    },
    {
      "entropy": 6.519909429550171,
      "epoch": 0.5596875167210659,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004976801491982263,
      "loss": 6.5191,
      "mean_token_accuracy": 0.12184159904718399,
      "num_tokens": 11321599.0,
      "step": 5230
    },
    {
      "entropy": 6.661324596405029,
      "epoch": 0.5602225908288297,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004976746698515977,
      "loss": 6.5663,
      "mean_token_accuracy": 0.12378089874982834,
      "num_tokens": 11331383.0,
      "step": 5235
    },
    {
      "entropy": 6.58859395980835,
      "epoch": 0.5607576649365937,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004976691840752694,
      "loss": 6.5245,
      "mean_token_accuracy": 0.12175474986433983,
      "num_tokens": 11342085.0,
      "step": 5240
    },
    {
      "entropy": 6.664286851882935,
      "epoch": 0.5612927390443576,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004976636918693996,
      "loss": 6.5803,
      "mean_token_accuracy": 0.1158666305243969,
      "num_tokens": 11354061.0,
      "step": 5245
    },
    {
      "entropy": 6.676495313644409,
      "epoch": 0.5618278131521216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004976581932341471,
      "loss": 6.4956,
      "mean_token_accuracy": 0.12871850952506064,
      "num_tokens": 11364448.0,
      "step": 5250
    },
    {
      "entropy": 6.619661426544189,
      "epoch": 0.5623628872598855,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004976526881696706,
      "loss": 6.4729,
      "mean_token_accuracy": 0.12930259183049203,
      "num_tokens": 11374838.0,
      "step": 5255
    },
    {
      "entropy": 6.506873607635498,
      "epoch": 0.5628979613676495,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004976471766761288,
      "loss": 6.531,
      "mean_token_accuracy": 0.12527473792433738,
      "num_tokens": 11385310.0,
      "step": 5260
    },
    {
      "entropy": 6.650757884979248,
      "epoch": 0.5634330354754133,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004976416587536812,
      "loss": 6.5386,
      "mean_token_accuracy": 0.1180022731423378,
      "num_tokens": 11396590.0,
      "step": 5265
    },
    {
      "entropy": 6.6104088306427,
      "epoch": 0.5639681095831772,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004976361344024871,
      "loss": 6.4781,
      "mean_token_accuracy": 0.12395978718996048,
      "num_tokens": 11407218.0,
      "step": 5270
    },
    {
      "entropy": 6.484217166900635,
      "epoch": 0.5645031836909412,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976306036227056,
      "loss": 6.5003,
      "mean_token_accuracy": 0.1241976909339428,
      "num_tokens": 11417716.0,
      "step": 5275
    },
    {
      "entropy": 6.66533055305481,
      "epoch": 0.5650382577987051,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000497625066414497,
      "loss": 6.5995,
      "mean_token_accuracy": 0.11878701969981194,
      "num_tokens": 11428295.0,
      "step": 5280
    },
    {
      "entropy": 6.716561126708984,
      "epoch": 0.5655733319064691,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004976195227780207,
      "loss": 6.5096,
      "mean_token_accuracy": 0.12021773606538773,
      "num_tokens": 11439044.0,
      "step": 5285
    },
    {
      "entropy": 6.623074865341186,
      "epoch": 0.566108406014233,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004976139727134371,
      "loss": 6.4704,
      "mean_token_accuracy": 0.12361141070723533,
      "num_tokens": 11450029.0,
      "step": 5290
    },
    {
      "entropy": 6.6110166072845455,
      "epoch": 0.566643480121997,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004976084162209062,
      "loss": 6.6368,
      "mean_token_accuracy": 0.11236973628401756,
      "num_tokens": 11461190.0,
      "step": 5295
    },
    {
      "entropy": 6.638668632507324,
      "epoch": 0.5671785542297608,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004976028533005886,
      "loss": 6.5039,
      "mean_token_accuracy": 0.12434427589178085,
      "num_tokens": 11472143.0,
      "step": 5300
    },
    {
      "entropy": 6.608167266845703,
      "epoch": 0.5677136283375247,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004975972839526449,
      "loss": 6.6195,
      "mean_token_accuracy": 0.11615654602646827,
      "num_tokens": 11483745.0,
      "step": 5305
    },
    {
      "entropy": 6.641917085647583,
      "epoch": 0.5682487024452887,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004975917081772358,
      "loss": 6.4895,
      "mean_token_accuracy": 0.12567954733967782,
      "num_tokens": 11495695.0,
      "step": 5310
    },
    {
      "entropy": 6.587403202056885,
      "epoch": 0.5687837765530526,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004975861259745225,
      "loss": 6.5281,
      "mean_token_accuracy": 0.12585749477148056,
      "num_tokens": 11505576.0,
      "step": 5315
    },
    {
      "entropy": 6.560363531112671,
      "epoch": 0.5693188506608166,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004975805373446662,
      "loss": 6.3828,
      "mean_token_accuracy": 0.12965528070926666,
      "num_tokens": 11516614.0,
      "step": 5320
    },
    {
      "entropy": 6.607542037963867,
      "epoch": 0.5698539247685804,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000497574942287828,
      "loss": 6.5164,
      "mean_token_accuracy": 0.11931004524230956,
      "num_tokens": 11528032.0,
      "step": 5325
    },
    {
      "entropy": 6.558800792694091,
      "epoch": 0.5703889988763444,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004975693408041697,
      "loss": 6.5303,
      "mean_token_accuracy": 0.11935954615473747,
      "num_tokens": 11538822.0,
      "step": 5330
    },
    {
      "entropy": 6.6300797939300535,
      "epoch": 0.5709240729841083,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975637328938529,
      "loss": 6.5425,
      "mean_token_accuracy": 0.11745326891541481,
      "num_tokens": 11548895.0,
      "step": 5335
    },
    {
      "entropy": 6.636452150344849,
      "epoch": 0.5714591470918722,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004975581185570398,
      "loss": 6.4338,
      "mean_token_accuracy": 0.12448979616165161,
      "num_tokens": 11559613.0,
      "step": 5340
    },
    {
      "entropy": 6.526908922195434,
      "epoch": 0.5719942211996362,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004975524977938921,
      "loss": 6.457,
      "mean_token_accuracy": 0.13034842759370804,
      "num_tokens": 11571092.0,
      "step": 5345
    },
    {
      "entropy": 6.594049692153931,
      "epoch": 0.5725292953074,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004975468706045722,
      "loss": 6.5988,
      "mean_token_accuracy": 0.1227944739162922,
      "num_tokens": 11581448.0,
      "step": 5350
    },
    {
      "entropy": 6.661578178405762,
      "epoch": 0.573064369415164,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975412369892429,
      "loss": 6.4963,
      "mean_token_accuracy": 0.13152960166335106,
      "num_tokens": 11592273.0,
      "step": 5355
    },
    {
      "entropy": 6.598143672943115,
      "epoch": 0.5735994435229279,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004975355969480665,
      "loss": 6.4556,
      "mean_token_accuracy": 0.12935501858592033,
      "num_tokens": 11602759.0,
      "step": 5360
    },
    {
      "entropy": 6.557130956649781,
      "epoch": 0.5741345176306919,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004975299504812061,
      "loss": 6.5054,
      "mean_token_accuracy": 0.1214064471423626,
      "num_tokens": 11613977.0,
      "step": 5365
    },
    {
      "entropy": 6.639095115661621,
      "epoch": 0.5746695917384558,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004975242975888246,
      "loss": 6.5483,
      "mean_token_accuracy": 0.12487179338932038,
      "num_tokens": 11625425.0,
      "step": 5370
    },
    {
      "entropy": 6.5668292999267575,
      "epoch": 0.5752046658462197,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004975186382710854,
      "loss": 6.5128,
      "mean_token_accuracy": 0.12402708381414414,
      "num_tokens": 11636530.0,
      "step": 5375
    },
    {
      "entropy": 6.627521133422851,
      "epoch": 0.5757397399539836,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004975129725281516,
      "loss": 6.5037,
      "mean_token_accuracy": 0.12818056493997573,
      "num_tokens": 11646516.0,
      "step": 5380
    },
    {
      "entropy": 6.602155256271362,
      "epoch": 0.5762748140617475,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004975073003601871,
      "loss": 6.4906,
      "mean_token_accuracy": 0.1249205619096756,
      "num_tokens": 11657424.0,
      "step": 5385
    },
    {
      "entropy": 6.5941839694976805,
      "epoch": 0.5768098881695115,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975016217673556,
      "loss": 6.5204,
      "mean_token_accuracy": 0.1244162991642952,
      "num_tokens": 11669447.0,
      "step": 5390
    },
    {
      "entropy": 6.563456344604492,
      "epoch": 0.5773449622772754,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497495936749821,
      "loss": 6.4915,
      "mean_token_accuracy": 0.13013387992978095,
      "num_tokens": 11679612.0,
      "step": 5395
    },
    {
      "entropy": 6.548102712631225,
      "epoch": 0.5778800363850394,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004974902453077474,
      "loss": 6.4639,
      "mean_token_accuracy": 0.12684713825583457,
      "num_tokens": 11689527.0,
      "step": 5400
    },
    {
      "entropy": 6.665154981613159,
      "epoch": 0.5784151104928033,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004974845474412993,
      "loss": 6.499,
      "mean_token_accuracy": 0.127774278819561,
      "num_tokens": 11700452.0,
      "step": 5405
    },
    {
      "entropy": 6.550641870498657,
      "epoch": 0.5789501846005671,
      "grad_norm": 1.125,
      "learning_rate": 0.0004974788431506412,
      "loss": 6.5022,
      "mean_token_accuracy": 0.12524343207478522,
      "num_tokens": 11710902.0,
      "step": 5410
    },
    {
      "entropy": 6.631369781494141,
      "epoch": 0.5794852587083311,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004974731324359378,
      "loss": 6.5351,
      "mean_token_accuracy": 0.11656294539570808,
      "num_tokens": 11721901.0,
      "step": 5415
    },
    {
      "entropy": 6.547674751281738,
      "epoch": 0.580020332816095,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004974674152973538,
      "loss": 6.5057,
      "mean_token_accuracy": 0.1236759714782238,
      "num_tokens": 11733622.0,
      "step": 5420
    },
    {
      "entropy": 6.636887931823731,
      "epoch": 0.580555406923859,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004974616917350545,
      "loss": 6.5098,
      "mean_token_accuracy": 0.12396662756800651,
      "num_tokens": 11745116.0,
      "step": 5425
    },
    {
      "entropy": 6.514854383468628,
      "epoch": 0.5810904810316229,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974559617492052,
      "loss": 6.4501,
      "mean_token_accuracy": 0.13217924833297728,
      "num_tokens": 11756203.0,
      "step": 5430
    },
    {
      "entropy": 6.6254181385040285,
      "epoch": 0.5816255551393869,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004974502253399712,
      "loss": 6.5112,
      "mean_token_accuracy": 0.12464274242520332,
      "num_tokens": 11766797.0,
      "step": 5435
    },
    {
      "entropy": 6.658445978164673,
      "epoch": 0.5821606292471507,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004974444825075182,
      "loss": 6.5877,
      "mean_token_accuracy": 0.12322907596826553,
      "num_tokens": 11778110.0,
      "step": 5440
    },
    {
      "entropy": 6.535740184783935,
      "epoch": 0.5826957033549146,
      "grad_norm": 1.109375,
      "learning_rate": 0.000497438733252012,
      "loss": 6.4443,
      "mean_token_accuracy": 0.1271793082356453,
      "num_tokens": 11788396.0,
      "step": 5445
    },
    {
      "entropy": 6.619189882278443,
      "epoch": 0.5832307774626786,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004974329775736188,
      "loss": 6.5334,
      "mean_token_accuracy": 0.11946209296584129,
      "num_tokens": 11798691.0,
      "step": 5450
    },
    {
      "entropy": 6.666010951995849,
      "epoch": 0.5837658515704425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004974272154725044,
      "loss": 6.5826,
      "mean_token_accuracy": 0.12276230081915855,
      "num_tokens": 11808965.0,
      "step": 5455
    },
    {
      "entropy": 6.570070552825928,
      "epoch": 0.5843009256782065,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004974214469488355,
      "loss": 6.4135,
      "mean_token_accuracy": 0.1297299839556217,
      "num_tokens": 11819932.0,
      "step": 5460
    },
    {
      "entropy": 6.589346218109131,
      "epoch": 0.5848359997859703,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004974156720027786,
      "loss": 6.5191,
      "mean_token_accuracy": 0.12162868157029152,
      "num_tokens": 11831332.0,
      "step": 5465
    },
    {
      "entropy": 6.5868500709533695,
      "epoch": 0.5853710738937343,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974098906345004,
      "loss": 6.4911,
      "mean_token_accuracy": 0.12161843627691268,
      "num_tokens": 11840947.0,
      "step": 5470
    },
    {
      "entropy": 6.560147905349732,
      "epoch": 0.5859061480014982,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004974041028441679,
      "loss": 6.4379,
      "mean_token_accuracy": 0.1253178521990776,
      "num_tokens": 11851640.0,
      "step": 5475
    },
    {
      "entropy": 6.6644033908844,
      "epoch": 0.5864412221092621,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004973983086319481,
      "loss": 6.5712,
      "mean_token_accuracy": 0.11793641969561577,
      "num_tokens": 11863608.0,
      "step": 5480
    },
    {
      "entropy": 6.634626054763794,
      "epoch": 0.5869762962170261,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004973925079980085,
      "loss": 6.4778,
      "mean_token_accuracy": 0.12400742322206497,
      "num_tokens": 11874707.0,
      "step": 5485
    },
    {
      "entropy": 6.506961536407471,
      "epoch": 0.58751137032479,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004973867009425164,
      "loss": 6.564,
      "mean_token_accuracy": 0.12651642858982087,
      "num_tokens": 11885118.0,
      "step": 5490
    },
    {
      "entropy": 6.70133867263794,
      "epoch": 0.5880464444325539,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004973808874656396,
      "loss": 6.5287,
      "mean_token_accuracy": 0.1264993041753769,
      "num_tokens": 11895859.0,
      "step": 5495
    },
    {
      "entropy": 6.647668409347534,
      "epoch": 0.5885815185403178,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004973750675675458,
      "loss": 6.5292,
      "mean_token_accuracy": 0.12409508153796196,
      "num_tokens": 11907019.0,
      "step": 5500
    },
    {
      "entropy": 6.589305782318116,
      "epoch": 0.5891165926480818,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004973692412484034,
      "loss": 6.4863,
      "mean_token_accuracy": 0.11921481937170028,
      "num_tokens": 11918257.0,
      "step": 5505
    },
    {
      "entropy": 6.55865330696106,
      "epoch": 0.5896516667558457,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004973634085083802,
      "loss": 6.4835,
      "mean_token_accuracy": 0.13126906454563142,
      "num_tokens": 11929003.0,
      "step": 5510
    },
    {
      "entropy": 6.645260190963745,
      "epoch": 0.5901867408636096,
      "grad_norm": 1.09375,
      "learning_rate": 0.000497357569347645,
      "loss": 6.6359,
      "mean_token_accuracy": 0.11833299845457076,
      "num_tokens": 11939786.0,
      "step": 5515
    },
    {
      "entropy": 6.657250452041626,
      "epoch": 0.5907218149713735,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004973517237663661,
      "loss": 6.5478,
      "mean_token_accuracy": 0.11691000610589981,
      "num_tokens": 11951042.0,
      "step": 5520
    },
    {
      "entropy": 6.605471992492676,
      "epoch": 0.5912568890791374,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004973458717647124,
      "loss": 6.4439,
      "mean_token_accuracy": 0.13294798955321313,
      "num_tokens": 11961982.0,
      "step": 5525
    },
    {
      "entropy": 6.551524877548218,
      "epoch": 0.5917919631869014,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000497340013342853,
      "loss": 6.4861,
      "mean_token_accuracy": 0.1264800451695919,
      "num_tokens": 11972078.0,
      "step": 5530
    },
    {
      "entropy": 6.605613803863525,
      "epoch": 0.5923270372946653,
      "grad_norm": 1.078125,
      "learning_rate": 0.000497334148500957,
      "loss": 6.5494,
      "mean_token_accuracy": 0.12491055428981782,
      "num_tokens": 11984037.0,
      "step": 5535
    },
    {
      "entropy": 6.667060375213623,
      "epoch": 0.5928621114024293,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004973282772391936,
      "loss": 6.4877,
      "mean_token_accuracy": 0.12558887004852295,
      "num_tokens": 11993688.0,
      "step": 5540
    },
    {
      "entropy": 6.489277362823486,
      "epoch": 0.5933971855101932,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004973223995577324,
      "loss": 6.4494,
      "mean_token_accuracy": 0.12596305087208748,
      "num_tokens": 12003759.0,
      "step": 5545
    },
    {
      "entropy": 6.603683185577393,
      "epoch": 0.593932259617957,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004973165154567432,
      "loss": 6.486,
      "mean_token_accuracy": 0.12778774201869963,
      "num_tokens": 12015030.0,
      "step": 5550
    },
    {
      "entropy": 6.636406421661377,
      "epoch": 0.594467333725721,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004973106249363959,
      "loss": 6.5651,
      "mean_token_accuracy": 0.1225056804716587,
      "num_tokens": 12026181.0,
      "step": 5555
    },
    {
      "entropy": 6.483244705200195,
      "epoch": 0.5950024078334849,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004973047279968606,
      "loss": 6.3287,
      "mean_token_accuracy": 0.14041824042797088,
      "num_tokens": 12037336.0,
      "step": 5560
    },
    {
      "entropy": 6.525764036178589,
      "epoch": 0.5955374819412489,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004972988246383074,
      "loss": 6.394,
      "mean_token_accuracy": 0.12839040011167527,
      "num_tokens": 12048425.0,
      "step": 5565
    },
    {
      "entropy": 6.46997389793396,
      "epoch": 0.5960725560490128,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004972929148609068,
      "loss": 6.3676,
      "mean_token_accuracy": 0.1441659763455391,
      "num_tokens": 12059814.0,
      "step": 5570
    },
    {
      "entropy": 6.503654193878174,
      "epoch": 0.5966076301567768,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004972869986648296,
      "loss": 6.4419,
      "mean_token_accuracy": 0.13450188264250756,
      "num_tokens": 12069377.0,
      "step": 5575
    },
    {
      "entropy": 6.620211458206176,
      "epoch": 0.5971427042645406,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004972810760502467,
      "loss": 6.5783,
      "mean_token_accuracy": 0.1187084436416626,
      "num_tokens": 12080277.0,
      "step": 5580
    },
    {
      "entropy": 6.658747053146362,
      "epoch": 0.5976777783723045,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004972751470173287,
      "loss": 6.4505,
      "mean_token_accuracy": 0.1255773939192295,
      "num_tokens": 12091375.0,
      "step": 5585
    },
    {
      "entropy": 6.507024526596069,
      "epoch": 0.5982128524800685,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004972692115662473,
      "loss": 6.5013,
      "mean_token_accuracy": 0.1259841337800026,
      "num_tokens": 12103110.0,
      "step": 5590
    },
    {
      "entropy": 6.637642526626587,
      "epoch": 0.5987479265878324,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004972632696971735,
      "loss": 6.5436,
      "mean_token_accuracy": 0.12355464100837707,
      "num_tokens": 12112454.0,
      "step": 5595
    },
    {
      "entropy": 6.56805100440979,
      "epoch": 0.5992830006955964,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004972573214102791,
      "loss": 6.4271,
      "mean_token_accuracy": 0.125191880017519,
      "num_tokens": 12125140.0,
      "step": 5600
    },
    {
      "entropy": 6.567241144180298,
      "epoch": 0.5998180748033602,
      "grad_norm": 1.125,
      "learning_rate": 0.0004972513667057358,
      "loss": 6.4568,
      "mean_token_accuracy": 0.12768169343471528,
      "num_tokens": 12136061.0,
      "step": 5605
    },
    {
      "entropy": 6.628464031219482,
      "epoch": 0.6003531489111242,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004972454055837155,
      "loss": 6.4822,
      "mean_token_accuracy": 0.13015633150935174,
      "num_tokens": 12146654.0,
      "step": 5610
    },
    {
      "entropy": 6.6061324119567875,
      "epoch": 0.6008882230188881,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004972394380443903,
      "loss": 6.4842,
      "mean_token_accuracy": 0.12277491986751557,
      "num_tokens": 12157881.0,
      "step": 5615
    },
    {
      "entropy": 6.549179172515869,
      "epoch": 0.601423297126652,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004972334640879325,
      "loss": 6.5163,
      "mean_token_accuracy": 0.12600575461983682,
      "num_tokens": 12168534.0,
      "step": 5620
    },
    {
      "entropy": 6.557713556289673,
      "epoch": 0.601958371234416,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004972274837145147,
      "loss": 6.4212,
      "mean_token_accuracy": 0.12794229462742807,
      "num_tokens": 12178773.0,
      "step": 5625
    },
    {
      "entropy": 6.617846250534058,
      "epoch": 0.6024934453421799,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004972214969243096,
      "loss": 6.4851,
      "mean_token_accuracy": 0.12551011741161347,
      "num_tokens": 12188819.0,
      "step": 5630
    },
    {
      "entropy": 6.57431697845459,
      "epoch": 0.6030285194499438,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049721550371749,
      "loss": 6.4516,
      "mean_token_accuracy": 0.13068530559539795,
      "num_tokens": 12199198.0,
      "step": 5635
    },
    {
      "entropy": 6.579136323928833,
      "epoch": 0.6035635935577077,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004972095040942289,
      "loss": 6.4676,
      "mean_token_accuracy": 0.1252219207584858,
      "num_tokens": 12210765.0,
      "step": 5640
    },
    {
      "entropy": 6.541816520690918,
      "epoch": 0.6040986676654717,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004972034980546996,
      "loss": 6.52,
      "mean_token_accuracy": 0.11947290003299713,
      "num_tokens": 12221067.0,
      "step": 5645
    },
    {
      "entropy": 6.568389987945556,
      "epoch": 0.6046337417732356,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004971974855990755,
      "loss": 6.5253,
      "mean_token_accuracy": 0.12596880793571472,
      "num_tokens": 12232994.0,
      "step": 5650
    },
    {
      "entropy": 6.694991207122802,
      "epoch": 0.6051688158809995,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004971914667275302,
      "loss": 6.466,
      "mean_token_accuracy": 0.1270422488451004,
      "num_tokens": 12243327.0,
      "step": 5655
    },
    {
      "entropy": 6.519532155990601,
      "epoch": 0.6057038899887635,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004971854414402377,
      "loss": 6.367,
      "mean_token_accuracy": 0.13553767204284667,
      "num_tokens": 12254249.0,
      "step": 5660
    },
    {
      "entropy": 6.532563400268555,
      "epoch": 0.6062389640965273,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004971794097373716,
      "loss": 6.4209,
      "mean_token_accuracy": 0.1289665073156357,
      "num_tokens": 12264487.0,
      "step": 5665
    },
    {
      "entropy": 6.446199893951416,
      "epoch": 0.6067740382042913,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004971733716191063,
      "loss": 6.4233,
      "mean_token_accuracy": 0.13476498275995255,
      "num_tokens": 12274801.0,
      "step": 5670
    },
    {
      "entropy": 6.541055154800415,
      "epoch": 0.6073091123120552,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004971673270856162,
      "loss": 6.3849,
      "mean_token_accuracy": 0.12896822169423103,
      "num_tokens": 12284647.0,
      "step": 5675
    },
    {
      "entropy": 6.55897159576416,
      "epoch": 0.6078441864198192,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004971612761370756,
      "loss": 6.5065,
      "mean_token_accuracy": 0.12677512541413308,
      "num_tokens": 12295215.0,
      "step": 5680
    },
    {
      "entropy": 6.557615089416504,
      "epoch": 0.6083792605275831,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004971552187736596,
      "loss": 6.4284,
      "mean_token_accuracy": 0.1286513738334179,
      "num_tokens": 12305324.0,
      "step": 5685
    },
    {
      "entropy": 6.543941259384155,
      "epoch": 0.6089143346353469,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004971491549955427,
      "loss": 6.433,
      "mean_token_accuracy": 0.13123927116394044,
      "num_tokens": 12316399.0,
      "step": 5690
    },
    {
      "entropy": 6.456441211700439,
      "epoch": 0.6094494087431109,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004971430848029002,
      "loss": 6.4119,
      "mean_token_accuracy": 0.13755123019218446,
      "num_tokens": 12328033.0,
      "step": 5695
    },
    {
      "entropy": 6.5196257591247555,
      "epoch": 0.6099844828508748,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004971370081959073,
      "loss": 6.4475,
      "mean_token_accuracy": 0.1325080767273903,
      "num_tokens": 12338962.0,
      "step": 5700
    },
    {
      "entropy": 6.596745729446411,
      "epoch": 0.6105195569586388,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004971309251747396,
      "loss": 6.4461,
      "mean_token_accuracy": 0.12643044963479041,
      "num_tokens": 12350216.0,
      "step": 5705
    },
    {
      "entropy": 6.610299110412598,
      "epoch": 0.6110546310664027,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004971248357395726,
      "loss": 6.47,
      "mean_token_accuracy": 0.12886352986097335,
      "num_tokens": 12361277.0,
      "step": 5710
    },
    {
      "entropy": 6.433839178085327,
      "epoch": 0.6115897051741667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004971187398905821,
      "loss": 6.3915,
      "mean_token_accuracy": 0.1265926755964756,
      "num_tokens": 12373308.0,
      "step": 5715
    },
    {
      "entropy": 6.634950113296509,
      "epoch": 0.6121247792819305,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004971126376279443,
      "loss": 6.621,
      "mean_token_accuracy": 0.11679412126541137,
      "num_tokens": 12383943.0,
      "step": 5720
    },
    {
      "entropy": 6.612603712081909,
      "epoch": 0.6126598533896944,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004971065289518354,
      "loss": 6.3772,
      "mean_token_accuracy": 0.12417749539017678,
      "num_tokens": 12394575.0,
      "step": 5725
    },
    {
      "entropy": 6.550524425506592,
      "epoch": 0.6131949274974584,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004971004138624315,
      "loss": 6.4989,
      "mean_token_accuracy": 0.12523326948285102,
      "num_tokens": 12405536.0,
      "step": 5730
    },
    {
      "entropy": 6.48319935798645,
      "epoch": 0.6137300016052223,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004970942923599094,
      "loss": 6.3823,
      "mean_token_accuracy": 0.1334580399096012,
      "num_tokens": 12416148.0,
      "step": 5735
    },
    {
      "entropy": 6.5434459209442135,
      "epoch": 0.6142650757129863,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004970881644444459,
      "loss": 6.3795,
      "mean_token_accuracy": 0.13239484578371047,
      "num_tokens": 12426182.0,
      "step": 5740
    },
    {
      "entropy": 6.552867317199707,
      "epoch": 0.6148001498207502,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004970820301162178,
      "loss": 6.566,
      "mean_token_accuracy": 0.12335467711091042,
      "num_tokens": 12437548.0,
      "step": 5745
    },
    {
      "entropy": 6.632025623321534,
      "epoch": 0.6153352239285141,
      "grad_norm": 1.0,
      "learning_rate": 0.0004970758893754022,
      "loss": 6.5069,
      "mean_token_accuracy": 0.12346102967858315,
      "num_tokens": 12449720.0,
      "step": 5750
    },
    {
      "entropy": 6.63417387008667,
      "epoch": 0.615870298036278,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004970697422221767,
      "loss": 6.479,
      "mean_token_accuracy": 0.1252536728978157,
      "num_tokens": 12460666.0,
      "step": 5755
    },
    {
      "entropy": 6.534920263290405,
      "epoch": 0.6164053721440419,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004970635886567185,
      "loss": 6.5012,
      "mean_token_accuracy": 0.12670053765177727,
      "num_tokens": 12471472.0,
      "step": 5760
    },
    {
      "entropy": 6.519795703887939,
      "epoch": 0.6169404462518059,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970574286792054,
      "loss": 6.4881,
      "mean_token_accuracy": 0.1228668674826622,
      "num_tokens": 12482127.0,
      "step": 5765
    },
    {
      "entropy": 6.56483211517334,
      "epoch": 0.6174755203595698,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004970512622898152,
      "loss": 6.4786,
      "mean_token_accuracy": 0.1271592140197754,
      "num_tokens": 12493680.0,
      "step": 5770
    },
    {
      "entropy": 6.5475750923156735,
      "epoch": 0.6180105944673338,
      "grad_norm": 1.046875,
      "learning_rate": 0.000497045089488726,
      "loss": 6.4289,
      "mean_token_accuracy": 0.12926682755351065,
      "num_tokens": 12505838.0,
      "step": 5775
    },
    {
      "entropy": 6.605856084823609,
      "epoch": 0.6185456685750976,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004970389102761161,
      "loss": 6.4203,
      "mean_token_accuracy": 0.1347701705992222,
      "num_tokens": 12516334.0,
      "step": 5780
    },
    {
      "entropy": 6.502717208862305,
      "epoch": 0.6190807426828616,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004970327246521638,
      "loss": 6.4981,
      "mean_token_accuracy": 0.1220128633081913,
      "num_tokens": 12526440.0,
      "step": 5785
    },
    {
      "entropy": 6.527550458908081,
      "epoch": 0.6196158167906255,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004970265326170478,
      "loss": 6.4207,
      "mean_token_accuracy": 0.12335163801908493,
      "num_tokens": 12538127.0,
      "step": 5790
    },
    {
      "entropy": 6.654475927352905,
      "epoch": 0.6201508908983894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004970203341709469,
      "loss": 6.5003,
      "mean_token_accuracy": 0.12200881615281105,
      "num_tokens": 12548459.0,
      "step": 5795
    },
    {
      "entropy": 6.5456257343292235,
      "epoch": 0.6206859650061534,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00049701412931404,
      "loss": 6.4473,
      "mean_token_accuracy": 0.1271664597094059,
      "num_tokens": 12558989.0,
      "step": 5800
    },
    {
      "entropy": 6.526596260070801,
      "epoch": 0.6212210391139172,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004970079180465064,
      "loss": 6.3775,
      "mean_token_accuracy": 0.13296955302357674,
      "num_tokens": 12569449.0,
      "step": 5805
    },
    {
      "entropy": 6.510442543029785,
      "epoch": 0.6217561132216812,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004970017003685253,
      "loss": 6.437,
      "mean_token_accuracy": 0.12862144485116006,
      "num_tokens": 12580733.0,
      "step": 5810
    },
    {
      "entropy": 6.555939054489135,
      "epoch": 0.6222911873294451,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004969954762802763,
      "loss": 6.4541,
      "mean_token_accuracy": 0.1280243769288063,
      "num_tokens": 12590869.0,
      "step": 5815
    },
    {
      "entropy": 6.626267814636231,
      "epoch": 0.6228262614372091,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000496989245781939,
      "loss": 6.5169,
      "mean_token_accuracy": 0.1257978305220604,
      "num_tokens": 12602109.0,
      "step": 5820
    },
    {
      "entropy": 6.565516376495362,
      "epoch": 0.623361335544973,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004969830088736937,
      "loss": 6.4519,
      "mean_token_accuracy": 0.12484232187271119,
      "num_tokens": 12612234.0,
      "step": 5825
    },
    {
      "entropy": 6.609433937072754,
      "epoch": 0.6238964096527368,
      "grad_norm": 1.125,
      "learning_rate": 0.00049697676555572,
      "loss": 6.5979,
      "mean_token_accuracy": 0.11923756599426269,
      "num_tokens": 12623252.0,
      "step": 5830
    },
    {
      "entropy": 6.633053159713745,
      "epoch": 0.6244314837605008,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004969705158281985,
      "loss": 6.352,
      "mean_token_accuracy": 0.13244080618023873,
      "num_tokens": 12632796.0,
      "step": 5835
    },
    {
      "entropy": 6.526743745803833,
      "epoch": 0.6249665578682647,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004969642596913095,
      "loss": 6.4529,
      "mean_token_accuracy": 0.12618629336357118,
      "num_tokens": 12643278.0,
      "step": 5840
    },
    {
      "entropy": 6.527894830703735,
      "epoch": 0.6255016319760287,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004969579971452337,
      "loss": 6.5187,
      "mean_token_accuracy": 0.12627973407506943,
      "num_tokens": 12654446.0,
      "step": 5845
    },
    {
      "entropy": 6.6252518653869625,
      "epoch": 0.6260367060837926,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004969517281901519,
      "loss": 6.4955,
      "mean_token_accuracy": 0.13113024607300758,
      "num_tokens": 12664281.0,
      "step": 5850
    },
    {
      "entropy": 6.5939366817474365,
      "epoch": 0.6265717801915566,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004969454528262451,
      "loss": 6.4836,
      "mean_token_accuracy": 0.12383594810962677,
      "num_tokens": 12675252.0,
      "step": 5855
    },
    {
      "entropy": 6.531436395645142,
      "epoch": 0.6271068542993204,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004969391710536946,
      "loss": 6.4625,
      "mean_token_accuracy": 0.12760263681411743,
      "num_tokens": 12685764.0,
      "step": 5860
    },
    {
      "entropy": 6.495209646224976,
      "epoch": 0.6276419284070843,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004969328828726817,
      "loss": 6.3636,
      "mean_token_accuracy": 0.13051239922642707,
      "num_tokens": 12696974.0,
      "step": 5865
    },
    {
      "entropy": 6.6419761180877686,
      "epoch": 0.6281770025148483,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004969265882833879,
      "loss": 6.5443,
      "mean_token_accuracy": 0.11920846477150918,
      "num_tokens": 12708217.0,
      "step": 5870
    },
    {
      "entropy": 6.548468828201294,
      "epoch": 0.6287120766226122,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004969202872859952,
      "loss": 6.401,
      "mean_token_accuracy": 0.13056466579437256,
      "num_tokens": 12717840.0,
      "step": 5875
    },
    {
      "entropy": 6.53254451751709,
      "epoch": 0.6292471507303762,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004969139798806854,
      "loss": 6.5009,
      "mean_token_accuracy": 0.11990201696753502,
      "num_tokens": 12728390.0,
      "step": 5880
    },
    {
      "entropy": 6.52357702255249,
      "epoch": 0.6297822248381401,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004969076660676405,
      "loss": 6.494,
      "mean_token_accuracy": 0.12666737139225007,
      "num_tokens": 12738650.0,
      "step": 5885
    },
    {
      "entropy": 6.59197096824646,
      "epoch": 0.630317298945904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000496901345847043,
      "loss": 6.5044,
      "mean_token_accuracy": 0.12580613866448404,
      "num_tokens": 12748989.0,
      "step": 5890
    },
    {
      "entropy": 6.614646244049072,
      "epoch": 0.6308523730536679,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004968950192190752,
      "loss": 6.4448,
      "mean_token_accuracy": 0.1283442422747612,
      "num_tokens": 12761132.0,
      "step": 5895
    },
    {
      "entropy": 6.532540225982666,
      "epoch": 0.6313874471614318,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049688868618392,
      "loss": 6.47,
      "mean_token_accuracy": 0.13095390945672988,
      "num_tokens": 12771478.0,
      "step": 5900
    },
    {
      "entropy": 6.583694553375244,
      "epoch": 0.6319225212691958,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004968823467417602,
      "loss": 6.525,
      "mean_token_accuracy": 0.1221963606774807,
      "num_tokens": 12782096.0,
      "step": 5905
    },
    {
      "entropy": 6.577635097503662,
      "epoch": 0.6324575953769597,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004968760008927788,
      "loss": 6.4173,
      "mean_token_accuracy": 0.13012173324823378,
      "num_tokens": 12792467.0,
      "step": 5910
    },
    {
      "entropy": 6.602513790130615,
      "epoch": 0.6329926694847237,
      "grad_norm": 1.0625,
      "learning_rate": 0.000496869648637159,
      "loss": 6.4317,
      "mean_token_accuracy": 0.12972439229488372,
      "num_tokens": 12803248.0,
      "step": 5915
    },
    {
      "entropy": 6.5703932762146,
      "epoch": 0.6335277435924875,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004968632899750844,
      "loss": 6.4584,
      "mean_token_accuracy": 0.12943121418356895,
      "num_tokens": 12813701.0,
      "step": 5920
    },
    {
      "entropy": 6.566262197494507,
      "epoch": 0.6340628177002515,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004968569249067383,
      "loss": 6.5179,
      "mean_token_accuracy": 0.12049417719244956,
      "num_tokens": 12824945.0,
      "step": 5925
    },
    {
      "entropy": 6.5732197761535645,
      "epoch": 0.6345978918080154,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004968505534323047,
      "loss": 6.4803,
      "mean_token_accuracy": 0.11932752430438995,
      "num_tokens": 12836699.0,
      "step": 5930
    },
    {
      "entropy": 6.5687562942504885,
      "epoch": 0.6351329659157794,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004968441755519677,
      "loss": 6.4482,
      "mean_token_accuracy": 0.12387168183922767,
      "num_tokens": 12846905.0,
      "step": 5935
    },
    {
      "entropy": 6.562500810623169,
      "epoch": 0.6356680400235433,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004968377912659112,
      "loss": 6.3881,
      "mean_token_accuracy": 0.13033246025443077,
      "num_tokens": 12856919.0,
      "step": 5940
    },
    {
      "entropy": 6.408761262893677,
      "epoch": 0.6362031141313071,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004968314005743196,
      "loss": 6.3532,
      "mean_token_accuracy": 0.13002551943063737,
      "num_tokens": 12867863.0,
      "step": 5945
    },
    {
      "entropy": 6.438524532318115,
      "epoch": 0.6367381882390711,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004968250034773776,
      "loss": 6.4448,
      "mean_token_accuracy": 0.12836804687976838,
      "num_tokens": 12877877.0,
      "step": 5950
    },
    {
      "entropy": 6.656329107284546,
      "epoch": 0.637273262346835,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004968185999752697,
      "loss": 6.5395,
      "mean_token_accuracy": 0.11811442151665688,
      "num_tokens": 12887954.0,
      "step": 5955
    },
    {
      "entropy": 6.5940252304077145,
      "epoch": 0.637808336454599,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004968121900681809,
      "loss": 6.4526,
      "mean_token_accuracy": 0.12516535967588424,
      "num_tokens": 12897744.0,
      "step": 5960
    },
    {
      "entropy": 6.553208208084106,
      "epoch": 0.6383434105623629,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004968057737562964,
      "loss": 6.5265,
      "mean_token_accuracy": 0.12683621421456337,
      "num_tokens": 12909193.0,
      "step": 5965
    },
    {
      "entropy": 6.5956096172332765,
      "epoch": 0.6388784846701269,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004967993510398012,
      "loss": 6.4035,
      "mean_token_accuracy": 0.12933284789323807,
      "num_tokens": 12919879.0,
      "step": 5970
    },
    {
      "entropy": 6.508267831802368,
      "epoch": 0.6394135587778907,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967929219188809,
      "loss": 6.4386,
      "mean_token_accuracy": 0.12756086364388466,
      "num_tokens": 12930297.0,
      "step": 5975
    },
    {
      "entropy": 6.508812427520752,
      "epoch": 0.6399486328856546,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004967864863937212,
      "loss": 6.4152,
      "mean_token_accuracy": 0.13640334010124205,
      "num_tokens": 12940467.0,
      "step": 5980
    },
    {
      "entropy": 6.467658615112304,
      "epoch": 0.6404837069934186,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004967800444645079,
      "loss": 6.422,
      "mean_token_accuracy": 0.12693843767046928,
      "num_tokens": 12950417.0,
      "step": 5985
    },
    {
      "entropy": 6.483503913879394,
      "epoch": 0.6410187811011825,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004967735961314269,
      "loss": 6.3835,
      "mean_token_accuracy": 0.1324019141495228,
      "num_tokens": 12960957.0,
      "step": 5990
    },
    {
      "entropy": 6.573039913177491,
      "epoch": 0.6415538552089465,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004967671413946646,
      "loss": 6.4813,
      "mean_token_accuracy": 0.12658655494451523,
      "num_tokens": 12971194.0,
      "step": 5995
    },
    {
      "entropy": 6.54908390045166,
      "epoch": 0.6420889293167104,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004967606802544071,
      "loss": 6.4631,
      "mean_token_accuracy": 0.12189773693680764,
      "num_tokens": 12982479.0,
      "step": 6000
    },
    {
      "epoch": 0.6420889293167104,
      "eval_entropy": 6.440853921036836,
      "eval_loss": 6.532153606414795,
      "eval_mean_token_accuracy": 0.1298162946979593,
      "eval_num_tokens": 12982479.0,
      "eval_runtime": 22.8018,
      "eval_samples_per_second": 1301.609,
      "eval_steps_per_second": 162.707,
      "step": 6000
    },
    {
      "entropy": 6.576130056381226,
      "epoch": 0.6426240034244743,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004967542127108412,
      "loss": 6.5215,
      "mean_token_accuracy": 0.12136223763227463,
      "num_tokens": 12994119.0,
      "step": 6005
    },
    {
      "entropy": 6.61212010383606,
      "epoch": 0.6431590775322382,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004967477387641537,
      "loss": 6.4399,
      "mean_token_accuracy": 0.12868218570947648,
      "num_tokens": 13005437.0,
      "step": 6010
    },
    {
      "entropy": 6.518870067596436,
      "epoch": 0.6436941516400021,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004967412584145312,
      "loss": 6.4608,
      "mean_token_accuracy": 0.12810143455863,
      "num_tokens": 13017278.0,
      "step": 6015
    },
    {
      "entropy": 6.509357690811157,
      "epoch": 0.6442292257477661,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004967347716621611,
      "loss": 6.4248,
      "mean_token_accuracy": 0.12839114665985107,
      "num_tokens": 13028010.0,
      "step": 6020
    },
    {
      "entropy": 6.536433267593384,
      "epoch": 0.64476429985553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004967282785072306,
      "loss": 6.4226,
      "mean_token_accuracy": 0.13216758370399476,
      "num_tokens": 13038740.0,
      "step": 6025
    },
    {
      "entropy": 6.545450019836426,
      "epoch": 0.645299373963294,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004967217789499272,
      "loss": 6.4419,
      "mean_token_accuracy": 0.1310186579823494,
      "num_tokens": 13049921.0,
      "step": 6030
    },
    {
      "entropy": 6.474092245101929,
      "epoch": 0.6458344480710578,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004967152729904386,
      "loss": 6.3825,
      "mean_token_accuracy": 0.12997894808650018,
      "num_tokens": 13060907.0,
      "step": 6035
    },
    {
      "entropy": 6.5579814434051515,
      "epoch": 0.6463695221788218,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004967087606289527,
      "loss": 6.4041,
      "mean_token_accuracy": 0.13156920149922371,
      "num_tokens": 13071720.0,
      "step": 6040
    },
    {
      "entropy": 6.522566843032837,
      "epoch": 0.6469045962865857,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004967022418656575,
      "loss": 6.4581,
      "mean_token_accuracy": 0.13345534577965737,
      "num_tokens": 13082474.0,
      "step": 6045
    },
    {
      "entropy": 6.4885979175567625,
      "epoch": 0.6474396703943496,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004966957167007412,
      "loss": 6.3627,
      "mean_token_accuracy": 0.12854093685746193,
      "num_tokens": 13092777.0,
      "step": 6050
    },
    {
      "entropy": 6.5061181545257565,
      "epoch": 0.6479747445021136,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004966891851343922,
      "loss": 6.3575,
      "mean_token_accuracy": 0.13643522262573243,
      "num_tokens": 13104019.0,
      "step": 6055
    },
    {
      "entropy": 6.520139169692993,
      "epoch": 0.6485098186098774,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004966826471667992,
      "loss": 6.4017,
      "mean_token_accuracy": 0.13252074345946313,
      "num_tokens": 13115741.0,
      "step": 6060
    },
    {
      "entropy": 6.516830205917358,
      "epoch": 0.6490448927176414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004966761027981509,
      "loss": 6.3909,
      "mean_token_accuracy": 0.1338729202747345,
      "num_tokens": 13126838.0,
      "step": 6065
    },
    {
      "entropy": 6.546872282028199,
      "epoch": 0.6495799668254053,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004966695520286362,
      "loss": 6.5141,
      "mean_token_accuracy": 0.12625649571418762,
      "num_tokens": 13137485.0,
      "step": 6070
    },
    {
      "entropy": 6.532670974731445,
      "epoch": 0.6501150409331693,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004966629948584444,
      "loss": 6.3701,
      "mean_token_accuracy": 0.1363312631845474,
      "num_tokens": 13147631.0,
      "step": 6075
    },
    {
      "entropy": 6.420833253860474,
      "epoch": 0.6506501150409332,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004966564312877647,
      "loss": 6.4457,
      "mean_token_accuracy": 0.13039978444576264,
      "num_tokens": 13158468.0,
      "step": 6080
    },
    {
      "entropy": 6.611920595169067,
      "epoch": 0.651185189148697,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004966498613167868,
      "loss": 6.4736,
      "mean_token_accuracy": 0.12443958222866058,
      "num_tokens": 13169263.0,
      "step": 6085
    },
    {
      "entropy": 6.5496948719024655,
      "epoch": 0.651720263256461,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004966432849457002,
      "loss": 6.4006,
      "mean_token_accuracy": 0.13697725608944894,
      "num_tokens": 13178647.0,
      "step": 6090
    },
    {
      "entropy": 6.429669809341431,
      "epoch": 0.6522553373642249,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000496636702174695,
      "loss": 6.381,
      "mean_token_accuracy": 0.13246307298541068,
      "num_tokens": 13189883.0,
      "step": 6095
    },
    {
      "entropy": 6.51705493927002,
      "epoch": 0.6527904114719889,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004966301130039611,
      "loss": 6.5076,
      "mean_token_accuracy": 0.128587756305933,
      "num_tokens": 13200920.0,
      "step": 6100
    },
    {
      "entropy": 6.580085420608521,
      "epoch": 0.6533254855797528,
      "grad_norm": 1.125,
      "learning_rate": 0.0004966235174336889,
      "loss": 6.5074,
      "mean_token_accuracy": 0.1210002399981022,
      "num_tokens": 13211765.0,
      "step": 6105
    },
    {
      "entropy": 6.55566725730896,
      "epoch": 0.6538605596875168,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004966169154640687,
      "loss": 6.4223,
      "mean_token_accuracy": 0.12444746568799019,
      "num_tokens": 13221768.0,
      "step": 6110
    },
    {
      "entropy": 6.504034042358398,
      "epoch": 0.6543956337952807,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004966103070952912,
      "loss": 6.4289,
      "mean_token_accuracy": 0.13048909232020378,
      "num_tokens": 13233545.0,
      "step": 6115
    },
    {
      "entropy": 6.537092208862305,
      "epoch": 0.6549307079030445,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004966036923275472,
      "loss": 6.4739,
      "mean_token_accuracy": 0.1343090645968914,
      "num_tokens": 13245018.0,
      "step": 6120
    },
    {
      "entropy": 6.530694627761841,
      "epoch": 0.6554657820108085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004965970711610278,
      "loss": 6.3863,
      "mean_token_accuracy": 0.1258619748055935,
      "num_tokens": 13255772.0,
      "step": 6125
    },
    {
      "entropy": 6.47897310256958,
      "epoch": 0.6560008561185724,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004965904435959241,
      "loss": 6.4744,
      "mean_token_accuracy": 0.12190289348363877,
      "num_tokens": 13267271.0,
      "step": 6130
    },
    {
      "entropy": 6.541171455383301,
      "epoch": 0.6565359302263364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004965838096324274,
      "loss": 6.4661,
      "mean_token_accuracy": 0.13229011446237565,
      "num_tokens": 13278467.0,
      "step": 6135
    },
    {
      "entropy": 6.608281469345092,
      "epoch": 0.6570710043341003,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004965771692707294,
      "loss": 6.5245,
      "mean_token_accuracy": 0.12438037320971489,
      "num_tokens": 13288522.0,
      "step": 6140
    },
    {
      "entropy": 6.602456378936767,
      "epoch": 0.6576060784418643,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004965705225110218,
      "loss": 6.5136,
      "mean_token_accuracy": 0.11394706219434739,
      "num_tokens": 13299355.0,
      "step": 6145
    },
    {
      "entropy": 6.597771215438843,
      "epoch": 0.6581411525496281,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004965638693534964,
      "loss": 6.4405,
      "mean_token_accuracy": 0.12820237800478934,
      "num_tokens": 13310334.0,
      "step": 6150
    },
    {
      "entropy": 6.606806373596191,
      "epoch": 0.658676226657392,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004965572097983455,
      "loss": 6.4768,
      "mean_token_accuracy": 0.12380762472748756,
      "num_tokens": 13320917.0,
      "step": 6155
    },
    {
      "entropy": 6.542260456085205,
      "epoch": 0.659211300765156,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004965505438457613,
      "loss": 6.3845,
      "mean_token_accuracy": 0.13052801191806793,
      "num_tokens": 13331707.0,
      "step": 6160
    },
    {
      "entropy": 6.517107105255127,
      "epoch": 0.6597463748729199,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004965438714959363,
      "loss": 6.5393,
      "mean_token_accuracy": 0.12245380133390427,
      "num_tokens": 13341784.0,
      "step": 6165
    },
    {
      "entropy": 6.621573448181152,
      "epoch": 0.6602814489806839,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000496537192749063,
      "loss": 6.4898,
      "mean_token_accuracy": 0.12351065501570702,
      "num_tokens": 13351633.0,
      "step": 6170
    },
    {
      "entropy": 6.585510158538819,
      "epoch": 0.6608165230884477,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004965305076053345,
      "loss": 6.381,
      "mean_token_accuracy": 0.13753572180867196,
      "num_tokens": 13362444.0,
      "step": 6175
    },
    {
      "entropy": 6.490717458724975,
      "epoch": 0.6613515971962117,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004965238160649437,
      "loss": 6.507,
      "mean_token_accuracy": 0.12068307399749756,
      "num_tokens": 13374322.0,
      "step": 6180
    },
    {
      "entropy": 6.57393159866333,
      "epoch": 0.6618866713039756,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004965171181280838,
      "loss": 6.4723,
      "mean_token_accuracy": 0.12478531822562218,
      "num_tokens": 13384188.0,
      "step": 6185
    },
    {
      "entropy": 6.602143239974976,
      "epoch": 0.6624217454117395,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965104137949483,
      "loss": 6.4734,
      "mean_token_accuracy": 0.12065911665558815,
      "num_tokens": 13395648.0,
      "step": 6190
    },
    {
      "entropy": 6.5944429397583,
      "epoch": 0.6629568195195035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004965037030657308,
      "loss": 6.4814,
      "mean_token_accuracy": 0.12683579176664353,
      "num_tokens": 13406769.0,
      "step": 6195
    },
    {
      "entropy": 6.597658920288086,
      "epoch": 0.6634918936272673,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004964969859406248,
      "loss": 6.4714,
      "mean_token_accuracy": 0.12145014703273774,
      "num_tokens": 13417509.0,
      "step": 6200
    },
    {
      "entropy": 6.529022264480591,
      "epoch": 0.6640269677350313,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004964902624198246,
      "loss": 6.3766,
      "mean_token_accuracy": 0.12588708102703094,
      "num_tokens": 13429441.0,
      "step": 6205
    },
    {
      "entropy": 6.482220697402954,
      "epoch": 0.6645620418427952,
      "grad_norm": 1.125,
      "learning_rate": 0.0004964835325035243,
      "loss": 6.4041,
      "mean_token_accuracy": 0.13321912959218024,
      "num_tokens": 13439332.0,
      "step": 6210
    },
    {
      "entropy": 6.48531584739685,
      "epoch": 0.6650971159505592,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004964767961919179,
      "loss": 6.4015,
      "mean_token_accuracy": 0.1290483571588993,
      "num_tokens": 13449400.0,
      "step": 6215
    },
    {
      "entropy": 6.59116153717041,
      "epoch": 0.6656321900583231,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004964700534852003,
      "loss": 6.4946,
      "mean_token_accuracy": 0.1224409706890583,
      "num_tokens": 13460134.0,
      "step": 6220
    },
    {
      "entropy": 6.530368375778198,
      "epoch": 0.666167264166087,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004964633043835658,
      "loss": 6.4488,
      "mean_token_accuracy": 0.12903735116124154,
      "num_tokens": 13469954.0,
      "step": 6225
    },
    {
      "entropy": 6.551534271240234,
      "epoch": 0.666702338273851,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004964565488872097,
      "loss": 6.4259,
      "mean_token_accuracy": 0.13188995644450188,
      "num_tokens": 13479866.0,
      "step": 6230
    },
    {
      "entropy": 6.509408283233642,
      "epoch": 0.6672374123816148,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004964497869963268,
      "loss": 6.3794,
      "mean_token_accuracy": 0.12967947050929068,
      "num_tokens": 13490088.0,
      "step": 6235
    },
    {
      "entropy": 6.522313356399536,
      "epoch": 0.6677724864893788,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004964430187111125,
      "loss": 6.4827,
      "mean_token_accuracy": 0.12645957320928575,
      "num_tokens": 13500898.0,
      "step": 6240
    },
    {
      "entropy": 6.530525207519531,
      "epoch": 0.6683075605971427,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004964362440317621,
      "loss": 6.4896,
      "mean_token_accuracy": 0.1254833646118641,
      "num_tokens": 13511537.0,
      "step": 6245
    },
    {
      "entropy": 6.583571100234986,
      "epoch": 0.6688426347049067,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004964294629584712,
      "loss": 6.4227,
      "mean_token_accuracy": 0.1282012976706028,
      "num_tokens": 13522916.0,
      "step": 6250
    },
    {
      "entropy": 6.549555015563965,
      "epoch": 0.6693777088126706,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004964226754914357,
      "loss": 6.3824,
      "mean_token_accuracy": 0.12852125465869904,
      "num_tokens": 13533817.0,
      "step": 6255
    },
    {
      "entropy": 6.507667160034179,
      "epoch": 0.6699127829204344,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004964158816308517,
      "loss": 6.4833,
      "mean_token_accuracy": 0.1205535277724266,
      "num_tokens": 13543723.0,
      "step": 6260
    },
    {
      "entropy": 6.505305910110474,
      "epoch": 0.6704478570281984,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004964090813769151,
      "loss": 6.3415,
      "mean_token_accuracy": 0.1334453582763672,
      "num_tokens": 13555269.0,
      "step": 6265
    },
    {
      "entropy": 6.538452768325806,
      "epoch": 0.6709829311359623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004964022747298224,
      "loss": 6.5131,
      "mean_token_accuracy": 0.12255731970071793,
      "num_tokens": 13567465.0,
      "step": 6270
    },
    {
      "entropy": 6.6572778701782225,
      "epoch": 0.6715180052437263,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004963954616897701,
      "loss": 6.4483,
      "mean_token_accuracy": 0.12406835183501244,
      "num_tokens": 13578637.0,
      "step": 6275
    },
    {
      "entropy": 6.5309834480285645,
      "epoch": 0.6720530793514902,
      "grad_norm": 1.1875,
      "learning_rate": 0.000496388642256955,
      "loss": 6.4611,
      "mean_token_accuracy": 0.12793243303894997,
      "num_tokens": 13589074.0,
      "step": 6280
    },
    {
      "entropy": 6.486390876770019,
      "epoch": 0.6725881534592542,
      "grad_norm": 1.125,
      "learning_rate": 0.000496381816431574,
      "loss": 6.4201,
      "mean_token_accuracy": 0.1265800841152668,
      "num_tokens": 13599624.0,
      "step": 6285
    },
    {
      "entropy": 6.625155448913574,
      "epoch": 0.673123227567018,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004963749842138241,
      "loss": 6.4378,
      "mean_token_accuracy": 0.12977349385619164,
      "num_tokens": 13610293.0,
      "step": 6290
    },
    {
      "entropy": 6.5692267417907715,
      "epoch": 0.6736583016747819,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004963681456039027,
      "loss": 6.4611,
      "mean_token_accuracy": 0.12518024072051048,
      "num_tokens": 13621641.0,
      "step": 6295
    },
    {
      "entropy": 6.482174062728882,
      "epoch": 0.6741933757825459,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004963613006020072,
      "loss": 6.4199,
      "mean_token_accuracy": 0.12508815601468087,
      "num_tokens": 13632083.0,
      "step": 6300
    },
    {
      "entropy": 6.500299024581909,
      "epoch": 0.6747284498903098,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004963544492083352,
      "loss": 6.3093,
      "mean_token_accuracy": 0.1482139140367508,
      "num_tokens": 13644084.0,
      "step": 6305
    },
    {
      "entropy": 6.456757640838623,
      "epoch": 0.6752635239980738,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004963475914230845,
      "loss": 6.4504,
      "mean_token_accuracy": 0.12221738547086716,
      "num_tokens": 13656231.0,
      "step": 6310
    },
    {
      "entropy": 6.576803588867188,
      "epoch": 0.6757985981058376,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004963407272464533,
      "loss": 6.4742,
      "mean_token_accuracy": 0.12403271645307541,
      "num_tokens": 13667334.0,
      "step": 6315
    },
    {
      "entropy": 6.588499879837036,
      "epoch": 0.6763336722136016,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004963338566786398,
      "loss": 6.4754,
      "mean_token_accuracy": 0.12822098508477212,
      "num_tokens": 13678615.0,
      "step": 6320
    },
    {
      "entropy": 6.571232652664184,
      "epoch": 0.6768687463213655,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004963269797198422,
      "loss": 6.48,
      "mean_token_accuracy": 0.13238448649644852,
      "num_tokens": 13689890.0,
      "step": 6325
    },
    {
      "entropy": 6.442437601089478,
      "epoch": 0.6774038204291294,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004963200963702592,
      "loss": 6.3223,
      "mean_token_accuracy": 0.13694801926612854,
      "num_tokens": 13699120.0,
      "step": 6330
    },
    {
      "entropy": 6.490188312530518,
      "epoch": 0.6779388945368934,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004963132066300894,
      "loss": 6.4212,
      "mean_token_accuracy": 0.1265653096139431,
      "num_tokens": 13710016.0,
      "step": 6335
    },
    {
      "entropy": 6.596869707107544,
      "epoch": 0.6784739686446573,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004963063104995319,
      "loss": 6.531,
      "mean_token_accuracy": 0.11942817196249962,
      "num_tokens": 13721576.0,
      "step": 6340
    },
    {
      "entropy": 6.6302014827728275,
      "epoch": 0.6790090427524212,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004962994079787859,
      "loss": 6.4978,
      "mean_token_accuracy": 0.12895067408680916,
      "num_tokens": 13732568.0,
      "step": 6345
    },
    {
      "entropy": 6.532273626327514,
      "epoch": 0.6795441168601851,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004962924990680504,
      "loss": 6.3783,
      "mean_token_accuracy": 0.13477483615279198,
      "num_tokens": 13742569.0,
      "step": 6350
    },
    {
      "entropy": 6.480876016616821,
      "epoch": 0.6800791909679491,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004962855837675252,
      "loss": 6.5065,
      "mean_token_accuracy": 0.12247845903038979,
      "num_tokens": 13752689.0,
      "step": 6355
    },
    {
      "entropy": 6.566415929794312,
      "epoch": 0.680614265075713,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004962786620774099,
      "loss": 6.4788,
      "mean_token_accuracy": 0.1281518444418907,
      "num_tokens": 13763709.0,
      "step": 6360
    },
    {
      "entropy": 6.588185787200928,
      "epoch": 0.6811493391834769,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004962717339979043,
      "loss": 6.3782,
      "mean_token_accuracy": 0.1340754009783268,
      "num_tokens": 13773726.0,
      "step": 6365
    },
    {
      "entropy": 6.480638027191162,
      "epoch": 0.6816844132912409,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004962647995292084,
      "loss": 6.4211,
      "mean_token_accuracy": 0.1272565059363842,
      "num_tokens": 13785435.0,
      "step": 6370
    },
    {
      "entropy": 6.514063596725464,
      "epoch": 0.6822194873990047,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004962578586715226,
      "loss": 6.4225,
      "mean_token_accuracy": 0.12357092872262002,
      "num_tokens": 13797104.0,
      "step": 6375
    },
    {
      "entropy": 6.617072057723999,
      "epoch": 0.6827545615067687,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004962509114250471,
      "loss": 6.5118,
      "mean_token_accuracy": 0.12474318742752075,
      "num_tokens": 13807756.0,
      "step": 6380
    },
    {
      "entropy": 6.374080753326416,
      "epoch": 0.6832896356145326,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004962439577899828,
      "loss": 6.1958,
      "mean_token_accuracy": 0.1468205489218235,
      "num_tokens": 13818490.0,
      "step": 6385
    },
    {
      "entropy": 6.543634843826294,
      "epoch": 0.6838247097222966,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004962369977665301,
      "loss": 6.4764,
      "mean_token_accuracy": 0.12152235433459282,
      "num_tokens": 13829465.0,
      "step": 6390
    },
    {
      "entropy": 6.574006509780884,
      "epoch": 0.6843597838300605,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004962300313548903,
      "loss": 6.453,
      "mean_token_accuracy": 0.13068217784166336,
      "num_tokens": 13840863.0,
      "step": 6395
    },
    {
      "entropy": 6.552519083023071,
      "epoch": 0.6848948579378243,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004962230585552645,
      "loss": 6.4432,
      "mean_token_accuracy": 0.12793402150273323,
      "num_tokens": 13851711.0,
      "step": 6400
    },
    {
      "entropy": 6.574922513961792,
      "epoch": 0.6854299320455883,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962160793678539,
      "loss": 6.4759,
      "mean_token_accuracy": 0.1269335649907589,
      "num_tokens": 13864576.0,
      "step": 6405
    },
    {
      "entropy": 6.546037006378174,
      "epoch": 0.6859650061533522,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00049620909379286,
      "loss": 6.3879,
      "mean_token_accuracy": 0.13360610902309417,
      "num_tokens": 13875150.0,
      "step": 6410
    },
    {
      "entropy": 6.522165393829345,
      "epoch": 0.6865000802611162,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004962021018304847,
      "loss": 6.4281,
      "mean_token_accuracy": 0.12703227996826172,
      "num_tokens": 13886070.0,
      "step": 6415
    },
    {
      "entropy": 6.458231115341187,
      "epoch": 0.6870351543688801,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004961951034809298,
      "loss": 6.3567,
      "mean_token_accuracy": 0.1256949707865715,
      "num_tokens": 13897898.0,
      "step": 6420
    },
    {
      "entropy": 6.60064663887024,
      "epoch": 0.6875702284766441,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004961880987443974,
      "loss": 6.4849,
      "mean_token_accuracy": 0.12602555900812148,
      "num_tokens": 13909578.0,
      "step": 6425
    },
    {
      "entropy": 6.449302625656128,
      "epoch": 0.6881053025844079,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004961810876210897,
      "loss": 6.351,
      "mean_token_accuracy": 0.1321983277797699,
      "num_tokens": 13920537.0,
      "step": 6430
    },
    {
      "entropy": 6.478025960922241,
      "epoch": 0.6886403766921718,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004961740701112091,
      "loss": 6.4042,
      "mean_token_accuracy": 0.13507199063897132,
      "num_tokens": 13931015.0,
      "step": 6435
    },
    {
      "entropy": 6.526848936080933,
      "epoch": 0.6891754507999358,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004961670462149583,
      "loss": 6.4542,
      "mean_token_accuracy": 0.12253931611776352,
      "num_tokens": 13942691.0,
      "step": 6440
    },
    {
      "entropy": 6.636831903457642,
      "epoch": 0.6897105249076997,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004961600159325403,
      "loss": 6.549,
      "mean_token_accuracy": 0.11617881655693055,
      "num_tokens": 13953945.0,
      "step": 6445
    },
    {
      "entropy": 6.6486979007720945,
      "epoch": 0.6902455990154637,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004961529792641577,
      "loss": 6.5264,
      "mean_token_accuracy": 0.12448358535766602,
      "num_tokens": 13965658.0,
      "step": 6450
    },
    {
      "entropy": 6.60369758605957,
      "epoch": 0.6907806731232276,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004961459362100141,
      "loss": 6.4957,
      "mean_token_accuracy": 0.1263794869184494,
      "num_tokens": 13976184.0,
      "step": 6455
    },
    {
      "entropy": 6.538023471832275,
      "epoch": 0.6913157472309915,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004961388867703125,
      "loss": 6.4221,
      "mean_token_accuracy": 0.12125025913119317,
      "num_tokens": 13986243.0,
      "step": 6460
    },
    {
      "entropy": 6.5492411136627195,
      "epoch": 0.6918508213387554,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004961318309452566,
      "loss": 6.4073,
      "mean_token_accuracy": 0.12485115006566047,
      "num_tokens": 13995904.0,
      "step": 6465
    },
    {
      "entropy": 6.563848781585693,
      "epoch": 0.6923858954465193,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004961247687350504,
      "loss": 6.4594,
      "mean_token_accuracy": 0.13267786130309106,
      "num_tokens": 14006711.0,
      "step": 6470
    },
    {
      "entropy": 6.4849306583404545,
      "epoch": 0.6929209695542833,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004961177001398974,
      "loss": 6.4315,
      "mean_token_accuracy": 0.1251344844698906,
      "num_tokens": 14018177.0,
      "step": 6475
    },
    {
      "entropy": 6.485970306396484,
      "epoch": 0.6934560436620472,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004961106251600018,
      "loss": 6.3815,
      "mean_token_accuracy": 0.1258880615234375,
      "num_tokens": 14028725.0,
      "step": 6480
    },
    {
      "entropy": 6.485081338882447,
      "epoch": 0.6939911177698111,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004961035437955681,
      "loss": 6.433,
      "mean_token_accuracy": 0.13642846941947936,
      "num_tokens": 14040448.0,
      "step": 6485
    },
    {
      "entropy": 6.516305828094483,
      "epoch": 0.694526191877575,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004960964560468005,
      "loss": 6.4599,
      "mean_token_accuracy": 0.12411079853773117,
      "num_tokens": 14051731.0,
      "step": 6490
    },
    {
      "entropy": 6.5961902141571045,
      "epoch": 0.695061265985339,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004960893619139039,
      "loss": 6.4722,
      "mean_token_accuracy": 0.12596373707056047,
      "num_tokens": 14062330.0,
      "step": 6495
    },
    {
      "entropy": 6.538251209259033,
      "epoch": 0.6955963400931029,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004960822613970831,
      "loss": 6.3613,
      "mean_token_accuracy": 0.1290792517364025,
      "num_tokens": 14073991.0,
      "step": 6500
    },
    {
      "entropy": 6.522839975357056,
      "epoch": 0.6961314142008668,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004960751544965431,
      "loss": 6.4051,
      "mean_token_accuracy": 0.12446002662181854,
      "num_tokens": 14085132.0,
      "step": 6505
    },
    {
      "entropy": 6.460032892227173,
      "epoch": 0.6966664883086308,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004960680412124889,
      "loss": 6.354,
      "mean_token_accuracy": 0.130805291980505,
      "num_tokens": 14095371.0,
      "step": 6510
    },
    {
      "entropy": 6.42583122253418,
      "epoch": 0.6972015624163946,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004960609215451261,
      "loss": 6.4141,
      "mean_token_accuracy": 0.1271771550178528,
      "num_tokens": 14106117.0,
      "step": 6515
    },
    {
      "entropy": 6.54330530166626,
      "epoch": 0.6977366365241586,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004960537954946604,
      "loss": 6.4278,
      "mean_token_accuracy": 0.12602221667766572,
      "num_tokens": 14117112.0,
      "step": 6520
    },
    {
      "entropy": 6.480698490142823,
      "epoch": 0.6982717106319225,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004960466630612974,
      "loss": 6.3682,
      "mean_token_accuracy": 0.14053670763969422,
      "num_tokens": 14128037.0,
      "step": 6525
    },
    {
      "entropy": 6.457269525527954,
      "epoch": 0.6988067847396865,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004960395242452429,
      "loss": 6.3635,
      "mean_token_accuracy": 0.1324993148446083,
      "num_tokens": 14140183.0,
      "step": 6530
    },
    {
      "entropy": 6.547682666778565,
      "epoch": 0.6993418588474504,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004960323790467033,
      "loss": 6.4968,
      "mean_token_accuracy": 0.12845934554934502,
      "num_tokens": 14151635.0,
      "step": 6535
    },
    {
      "entropy": 6.606826257705689,
      "epoch": 0.6998769329552142,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004960252274658849,
      "loss": 6.4554,
      "mean_token_accuracy": 0.12469615787267685,
      "num_tokens": 14162660.0,
      "step": 6540
    },
    {
      "entropy": 6.574669551849365,
      "epoch": 0.7004120070629782,
      "grad_norm": 1.125,
      "learning_rate": 0.0004960180695029939,
      "loss": 6.4298,
      "mean_token_accuracy": 0.13192950710654258,
      "num_tokens": 14173496.0,
      "step": 6545
    },
    {
      "entropy": 6.477722978591919,
      "epoch": 0.7009470811707421,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004960109051582374,
      "loss": 6.3847,
      "mean_token_accuracy": 0.12185332253575325,
      "num_tokens": 14185479.0,
      "step": 6550
    },
    {
      "entropy": 6.475646877288819,
      "epoch": 0.7014821552785061,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004960037344318221,
      "loss": 6.3488,
      "mean_token_accuracy": 0.127738206833601,
      "num_tokens": 14197441.0,
      "step": 6555
    },
    {
      "entropy": 6.422780752182007,
      "epoch": 0.70201722938627,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004959965573239548,
      "loss": 6.3166,
      "mean_token_accuracy": 0.13443872928619385,
      "num_tokens": 14207832.0,
      "step": 6560
    },
    {
      "entropy": 6.433670568466186,
      "epoch": 0.702552303494034,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004959893738348432,
      "loss": 6.3982,
      "mean_token_accuracy": 0.13432809486985206,
      "num_tokens": 14217976.0,
      "step": 6565
    },
    {
      "entropy": 6.618579244613647,
      "epoch": 0.7030873776017978,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004959821839646943,
      "loss": 6.3517,
      "mean_token_accuracy": 0.13541164621710777,
      "num_tokens": 14227557.0,
      "step": 6570
    },
    {
      "entropy": 6.5433601379394535,
      "epoch": 0.7036224517095617,
      "grad_norm": 1.28125,
      "learning_rate": 0.000495974987713716,
      "loss": 6.4823,
      "mean_token_accuracy": 0.12258463501930236,
      "num_tokens": 14238912.0,
      "step": 6575
    },
    {
      "entropy": 6.491708469390869,
      "epoch": 0.7041575258173257,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004959677850821159,
      "loss": 6.3899,
      "mean_token_accuracy": 0.13074335977435111,
      "num_tokens": 14249968.0,
      "step": 6580
    },
    {
      "entropy": 6.513847208023071,
      "epoch": 0.7046925999250896,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004959605760701021,
      "loss": 6.4593,
      "mean_token_accuracy": 0.13145442232489585,
      "num_tokens": 14260809.0,
      "step": 6585
    },
    {
      "entropy": 6.5283472537994385,
      "epoch": 0.7052276740328536,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004959533606778827,
      "loss": 6.4129,
      "mean_token_accuracy": 0.12853396609425544,
      "num_tokens": 14271971.0,
      "step": 6590
    },
    {
      "entropy": 6.607065677642822,
      "epoch": 0.7057627481406175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004959461389056661,
      "loss": 6.495,
      "mean_token_accuracy": 0.13045316487550734,
      "num_tokens": 14283550.0,
      "step": 6595
    },
    {
      "entropy": 6.525335359573364,
      "epoch": 0.7062978222483814,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004959389107536608,
      "loss": 6.3195,
      "mean_token_accuracy": 0.13225220292806625,
      "num_tokens": 14293769.0,
      "step": 6600
    },
    {
      "entropy": 6.493773126602173,
      "epoch": 0.7068328963561453,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004959316762220754,
      "loss": 6.4405,
      "mean_token_accuracy": 0.12467349618673325,
      "num_tokens": 14304668.0,
      "step": 6605
    },
    {
      "entropy": 6.491579151153564,
      "epoch": 0.7073679704639092,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000495924435311119,
      "loss": 6.5015,
      "mean_token_accuracy": 0.12518160939216613,
      "num_tokens": 14315912.0,
      "step": 6610
    },
    {
      "entropy": 6.526792812347412,
      "epoch": 0.7079030445716732,
      "grad_norm": 1.125,
      "learning_rate": 0.0004959171880210005,
      "loss": 6.4568,
      "mean_token_accuracy": 0.12405647858977317,
      "num_tokens": 14327386.0,
      "step": 6615
    },
    {
      "entropy": 6.588793849945068,
      "epoch": 0.7084381186794371,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004959099343519294,
      "loss": 6.4202,
      "mean_token_accuracy": 0.1256176069378853,
      "num_tokens": 14338629.0,
      "step": 6620
    },
    {
      "entropy": 6.604103422164917,
      "epoch": 0.7089731927872011,
      "grad_norm": 1.125,
      "learning_rate": 0.0004959026743041148,
      "loss": 6.463,
      "mean_token_accuracy": 0.12887532338500024,
      "num_tokens": 14349195.0,
      "step": 6625
    },
    {
      "entropy": 6.574780082702636,
      "epoch": 0.7095082668949649,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004958954078777665,
      "loss": 6.4751,
      "mean_token_accuracy": 0.12435831874608994,
      "num_tokens": 14359449.0,
      "step": 6630
    },
    {
      "entropy": 6.55607123374939,
      "epoch": 0.7100433410027289,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004958881350730944,
      "loss": 6.3735,
      "mean_token_accuracy": 0.13689294904470445,
      "num_tokens": 14369904.0,
      "step": 6635
    },
    {
      "entropy": 6.500827789306641,
      "epoch": 0.7105784151104928,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004958808558903085,
      "loss": 6.4095,
      "mean_token_accuracy": 0.12818561419844626,
      "num_tokens": 14382635.0,
      "step": 6640
    },
    {
      "entropy": 6.48659462928772,
      "epoch": 0.7111134892182567,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004958735703296189,
      "loss": 6.3551,
      "mean_token_accuracy": 0.13057286143302918,
      "num_tokens": 14393238.0,
      "step": 6645
    },
    {
      "entropy": 6.522722864151001,
      "epoch": 0.7116485633260207,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000495866278391236,
      "loss": 6.4118,
      "mean_token_accuracy": 0.12817465737462044,
      "num_tokens": 14403734.0,
      "step": 6650
    },
    {
      "entropy": 6.5527108669281,
      "epoch": 0.7121836374337845,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004958589800753703,
      "loss": 6.3916,
      "mean_token_accuracy": 0.12830689623951913,
      "num_tokens": 14414201.0,
      "step": 6655
    },
    {
      "entropy": 6.517422771453857,
      "epoch": 0.7127187115415485,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004958516753822326,
      "loss": 6.4104,
      "mean_token_accuracy": 0.13550665900111197,
      "num_tokens": 14425153.0,
      "step": 6660
    },
    {
      "entropy": 6.490578508377075,
      "epoch": 0.7132537856493124,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004958443643120338,
      "loss": 6.407,
      "mean_token_accuracy": 0.13405749946832657,
      "num_tokens": 14435942.0,
      "step": 6665
    },
    {
      "entropy": 6.532617378234863,
      "epoch": 0.7137888597570764,
      "grad_norm": 1.6875,
      "learning_rate": 0.000495837046864985,
      "loss": 6.378,
      "mean_token_accuracy": 0.13143783584237098,
      "num_tokens": 14446371.0,
      "step": 6670
    },
    {
      "entropy": 6.503788185119629,
      "epoch": 0.7143239338648403,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004958297230412976,
      "loss": 6.328,
      "mean_token_accuracy": 0.1338941276073456,
      "num_tokens": 14456598.0,
      "step": 6675
    },
    {
      "entropy": 6.484310913085937,
      "epoch": 0.7148590079726042,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004958223928411829,
      "loss": 6.4094,
      "mean_token_accuracy": 0.12213060334324836,
      "num_tokens": 14466775.0,
      "step": 6680
    },
    {
      "entropy": 6.506584787368775,
      "epoch": 0.7153940820803681,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004958150562648526,
      "loss": 6.3668,
      "mean_token_accuracy": 0.1280994415283203,
      "num_tokens": 14476876.0,
      "step": 6685
    },
    {
      "entropy": 6.478542423248291,
      "epoch": 0.715929156188132,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004958077133125187,
      "loss": 6.3696,
      "mean_token_accuracy": 0.13375263214111327,
      "num_tokens": 14487698.0,
      "step": 6690
    },
    {
      "entropy": 6.569677829742432,
      "epoch": 0.716464230295896,
      "grad_norm": 1.0625,
      "learning_rate": 0.000495800363984393,
      "loss": 6.4156,
      "mean_token_accuracy": 0.13004623353481293,
      "num_tokens": 14498572.0,
      "step": 6695
    },
    {
      "entropy": 6.511337184906006,
      "epoch": 0.7169993044036599,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004957930082806878,
      "loss": 6.4328,
      "mean_token_accuracy": 0.1281271144747734,
      "num_tokens": 14509673.0,
      "step": 6700
    },
    {
      "entropy": 6.578439950942993,
      "epoch": 0.7175343785114239,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004957856462016155,
      "loss": 6.504,
      "mean_token_accuracy": 0.11684139519929886,
      "num_tokens": 14519974.0,
      "step": 6705
    },
    {
      "entropy": 6.544417047500611,
      "epoch": 0.7180694526191878,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004957782777473888,
      "loss": 6.4465,
      "mean_token_accuracy": 0.12320490702986717,
      "num_tokens": 14531194.0,
      "step": 6710
    },
    {
      "entropy": 6.54014835357666,
      "epoch": 0.7186045267269516,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004957709029182203,
      "loss": 6.3993,
      "mean_token_accuracy": 0.12733219116926192,
      "num_tokens": 14542393.0,
      "step": 6715
    },
    {
      "entropy": 6.5045250415802,
      "epoch": 0.7191396008347156,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004957635217143231,
      "loss": 6.3943,
      "mean_token_accuracy": 0.13563383966684342,
      "num_tokens": 14552934.0,
      "step": 6720
    },
    {
      "entropy": 6.466321516036987,
      "epoch": 0.7196746749424795,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049575613413591,
      "loss": 6.4032,
      "mean_token_accuracy": 0.12802834659814835,
      "num_tokens": 14563643.0,
      "step": 6725
    },
    {
      "entropy": 6.515633964538575,
      "epoch": 0.7202097490502435,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004957487401831947,
      "loss": 6.3313,
      "mean_token_accuracy": 0.13340641632676126,
      "num_tokens": 14574422.0,
      "step": 6730
    },
    {
      "entropy": 6.511780023574829,
      "epoch": 0.7207448231580074,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004957413398563906,
      "loss": 6.3284,
      "mean_token_accuracy": 0.1287507437169552,
      "num_tokens": 14585002.0,
      "step": 6735
    },
    {
      "entropy": 6.5095727920532225,
      "epoch": 0.7212798972657714,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004957339331557113,
      "loss": 6.3634,
      "mean_token_accuracy": 0.12985105961561202,
      "num_tokens": 14595172.0,
      "step": 6740
    },
    {
      "entropy": 6.5062037944793705,
      "epoch": 0.7218149713735352,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004957265200813706,
      "loss": 6.468,
      "mean_token_accuracy": 0.12697600275278093,
      "num_tokens": 14606316.0,
      "step": 6745
    },
    {
      "entropy": 6.555118751525879,
      "epoch": 0.7223500454812991,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004957191006335827,
      "loss": 6.3752,
      "mean_token_accuracy": 0.13320598006248474,
      "num_tokens": 14616394.0,
      "step": 6750
    },
    {
      "entropy": 6.450709819793701,
      "epoch": 0.7228851195890631,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004957116748125618,
      "loss": 6.3835,
      "mean_token_accuracy": 0.1288460023701191,
      "num_tokens": 14627474.0,
      "step": 6755
    },
    {
      "entropy": 6.470903968811035,
      "epoch": 0.723420193696827,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004957042426185223,
      "loss": 6.3142,
      "mean_token_accuracy": 0.13408376649022102,
      "num_tokens": 14638343.0,
      "step": 6760
    },
    {
      "entropy": 6.47650294303894,
      "epoch": 0.723955267804591,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004956968040516789,
      "loss": 6.3937,
      "mean_token_accuracy": 0.13217563629150392,
      "num_tokens": 14649115.0,
      "step": 6765
    },
    {
      "entropy": 6.498498582839966,
      "epoch": 0.7244903419123548,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004956893591122461,
      "loss": 6.4599,
      "mean_token_accuracy": 0.12775713056325913,
      "num_tokens": 14659714.0,
      "step": 6770
    },
    {
      "entropy": 6.668419694900512,
      "epoch": 0.7250254160201188,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004956819078004392,
      "loss": 6.4921,
      "mean_token_accuracy": 0.12514772564172744,
      "num_tokens": 14670481.0,
      "step": 6775
    },
    {
      "entropy": 6.4725781917572025,
      "epoch": 0.7255604901278827,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004956744501164732,
      "loss": 6.3219,
      "mean_token_accuracy": 0.13545508608222007,
      "num_tokens": 14680816.0,
      "step": 6780
    },
    {
      "entropy": 6.460849714279175,
      "epoch": 0.7260955642356466,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004956669860605634,
      "loss": 6.4044,
      "mean_token_accuracy": 0.13029070869088172,
      "num_tokens": 14692723.0,
      "step": 6785
    },
    {
      "entropy": 6.538440179824829,
      "epoch": 0.7266306383434106,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004956595156329254,
      "loss": 6.4284,
      "mean_token_accuracy": 0.12866610437631607,
      "num_tokens": 14703824.0,
      "step": 6790
    },
    {
      "entropy": 6.538825750350952,
      "epoch": 0.7271657124511745,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000495652038833775,
      "loss": 6.422,
      "mean_token_accuracy": 0.13543595150113105,
      "num_tokens": 14715934.0,
      "step": 6795
    },
    {
      "entropy": 6.492854022979737,
      "epoch": 0.7277007865589384,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004956445556633279,
      "loss": 6.4226,
      "mean_token_accuracy": 0.12873160615563392,
      "num_tokens": 14725617.0,
      "step": 6800
    },
    {
      "entropy": 6.550632190704346,
      "epoch": 0.7282358606667023,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004956370661218003,
      "loss": 6.3944,
      "mean_token_accuracy": 0.13200772255659105,
      "num_tokens": 14736430.0,
      "step": 6805
    },
    {
      "entropy": 6.5043745040893555,
      "epoch": 0.7287709347744663,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004956295702094084,
      "loss": 6.4056,
      "mean_token_accuracy": 0.1324337676167488,
      "num_tokens": 14746740.0,
      "step": 6810
    },
    {
      "entropy": 6.434523057937622,
      "epoch": 0.7293060088822302,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004956220679263687,
      "loss": 6.3463,
      "mean_token_accuracy": 0.13504129871726037,
      "num_tokens": 14756884.0,
      "step": 6815
    },
    {
      "entropy": 6.51843638420105,
      "epoch": 0.7298410829899941,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004956145592728976,
      "loss": 6.3658,
      "mean_token_accuracy": 0.13665802925825118,
      "num_tokens": 14766974.0,
      "step": 6820
    },
    {
      "entropy": 6.517039251327515,
      "epoch": 0.730376157097758,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004956070442492123,
      "loss": 6.5015,
      "mean_token_accuracy": 0.12268729209899902,
      "num_tokens": 14778449.0,
      "step": 6825
    },
    {
      "entropy": 6.569549226760865,
      "epoch": 0.7309112312055219,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004955995228555296,
      "loss": 6.4535,
      "mean_token_accuracy": 0.12801005318760872,
      "num_tokens": 14788989.0,
      "step": 6830
    },
    {
      "entropy": 6.509778070449829,
      "epoch": 0.7314463053132859,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004955919950920666,
      "loss": 6.4093,
      "mean_token_accuracy": 0.13627632707357407,
      "num_tokens": 14798454.0,
      "step": 6835
    },
    {
      "entropy": 6.505553531646728,
      "epoch": 0.7319813794210498,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004955844609590408,
      "loss": 6.346,
      "mean_token_accuracy": 0.12747592255473136,
      "num_tokens": 14808920.0,
      "step": 6840
    },
    {
      "entropy": 6.468402481079101,
      "epoch": 0.7325164535288138,
      "grad_norm": 1.125,
      "learning_rate": 0.0004955769204566696,
      "loss": 6.3408,
      "mean_token_accuracy": 0.13437819182872773,
      "num_tokens": 14819250.0,
      "step": 6845
    },
    {
      "entropy": 6.550874280929565,
      "epoch": 0.7330515276365777,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004955693735851709,
      "loss": 6.4247,
      "mean_token_accuracy": 0.13173568844795228,
      "num_tokens": 14829789.0,
      "step": 6850
    },
    {
      "entropy": 6.479909467697143,
      "epoch": 0.7335866017443415,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004955618203447625,
      "loss": 6.3879,
      "mean_token_accuracy": 0.1247677482664585,
      "num_tokens": 14840527.0,
      "step": 6855
    },
    {
      "entropy": 6.52197961807251,
      "epoch": 0.7341216758521055,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004955542607356625,
      "loss": 6.4479,
      "mean_token_accuracy": 0.1325635425746441,
      "num_tokens": 14850922.0,
      "step": 6860
    },
    {
      "entropy": 6.526068162918091,
      "epoch": 0.7346567499598694,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004955466947580893,
      "loss": 6.3944,
      "mean_token_accuracy": 0.1303033322095871,
      "num_tokens": 14862226.0,
      "step": 6865
    },
    {
      "entropy": 6.53472261428833,
      "epoch": 0.7351918240676334,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004955391224122611,
      "loss": 6.3104,
      "mean_token_accuracy": 0.14006347134709357,
      "num_tokens": 14872461.0,
      "step": 6870
    },
    {
      "entropy": 6.477840423583984,
      "epoch": 0.7357268981753973,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000495531543698397,
      "loss": 6.4099,
      "mean_token_accuracy": 0.1295254684984684,
      "num_tokens": 14883088.0,
      "step": 6875
    },
    {
      "entropy": 6.536417150497437,
      "epoch": 0.7362619722831613,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004955239586167153,
      "loss": 6.4266,
      "mean_token_accuracy": 0.12837038263678552,
      "num_tokens": 14894055.0,
      "step": 6880
    },
    {
      "entropy": 6.5057226657867435,
      "epoch": 0.7367970463909251,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004955163671674354,
      "loss": 6.4112,
      "mean_token_accuracy": 0.12286904379725457,
      "num_tokens": 14904760.0,
      "step": 6885
    },
    {
      "entropy": 6.60149188041687,
      "epoch": 0.737332120498689,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004955087693507764,
      "loss": 6.5215,
      "mean_token_accuracy": 0.12230655699968337,
      "num_tokens": 14915983.0,
      "step": 6890
    },
    {
      "entropy": 6.514084672927856,
      "epoch": 0.737867194606453,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004955011651669577,
      "loss": 6.4174,
      "mean_token_accuracy": 0.12770415022969245,
      "num_tokens": 14925959.0,
      "step": 6895
    },
    {
      "entropy": 6.480762052536011,
      "epoch": 0.7384022687142169,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004954935546161988,
      "loss": 6.3638,
      "mean_token_accuracy": 0.13212289810180664,
      "num_tokens": 14936138.0,
      "step": 6900
    },
    {
      "entropy": 6.511771965026855,
      "epoch": 0.7389373428219809,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004954859376987195,
      "loss": 6.3908,
      "mean_token_accuracy": 0.1293765649199486,
      "num_tokens": 14946979.0,
      "step": 6905
    },
    {
      "entropy": 6.458079385757446,
      "epoch": 0.7394724169297447,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004954783144147397,
      "loss": 6.4498,
      "mean_token_accuracy": 0.12310977950692177,
      "num_tokens": 14958324.0,
      "step": 6910
    },
    {
      "entropy": 6.582757997512817,
      "epoch": 0.7400074910375087,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004954706847644796,
      "loss": 6.4615,
      "mean_token_accuracy": 0.12378886714577675,
      "num_tokens": 14969048.0,
      "step": 6915
    },
    {
      "entropy": 6.553678131103515,
      "epoch": 0.7405425651452726,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004954630487481594,
      "loss": 6.4465,
      "mean_token_accuracy": 0.12774492874741555,
      "num_tokens": 14979825.0,
      "step": 6920
    },
    {
      "entropy": 6.483800029754638,
      "epoch": 0.7410776392530365,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004954554063659998,
      "loss": 6.3729,
      "mean_token_accuracy": 0.13075642809271812,
      "num_tokens": 14990138.0,
      "step": 6925
    },
    {
      "entropy": 6.475497198104859,
      "epoch": 0.7416127133608005,
      "grad_norm": 1.125,
      "learning_rate": 0.0004954477576182212,
      "loss": 6.3289,
      "mean_token_accuracy": 0.1300647996366024,
      "num_tokens": 15002276.0,
      "step": 6930
    },
    {
      "entropy": 6.538154077529907,
      "epoch": 0.7421477874685644,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004954401025050445,
      "loss": 6.4744,
      "mean_token_accuracy": 0.12569414600729942,
      "num_tokens": 15013622.0,
      "step": 6935
    },
    {
      "entropy": 6.555956172943115,
      "epoch": 0.7426828615763283,
      "grad_norm": 1.125,
      "learning_rate": 0.0004954324410266909,
      "loss": 6.435,
      "mean_token_accuracy": 0.12979554608464242,
      "num_tokens": 15024067.0,
      "step": 6940
    },
    {
      "entropy": 6.550615549087524,
      "epoch": 0.7432179356840922,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004954247731833815,
      "loss": 6.3122,
      "mean_token_accuracy": 0.14097004979848862,
      "num_tokens": 15033646.0,
      "step": 6945
    },
    {
      "entropy": 6.424137783050537,
      "epoch": 0.7437530097918562,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004954170989753377,
      "loss": 6.3715,
      "mean_token_accuracy": 0.12743182927370073,
      "num_tokens": 15044655.0,
      "step": 6950
    },
    {
      "entropy": 6.581145906448365,
      "epoch": 0.7442880838996201,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004954094184027813,
      "loss": 6.4274,
      "mean_token_accuracy": 0.12889956682920456,
      "num_tokens": 15055492.0,
      "step": 6955
    },
    {
      "entropy": 6.578283596038818,
      "epoch": 0.744823158007384,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004954017314659339,
      "loss": 6.4215,
      "mean_token_accuracy": 0.12162732109427452,
      "num_tokens": 15066472.0,
      "step": 6960
    },
    {
      "entropy": 6.548455810546875,
      "epoch": 0.745358232115148,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004953940381650174,
      "loss": 6.3892,
      "mean_token_accuracy": 0.1315125435590744,
      "num_tokens": 15076851.0,
      "step": 6965
    },
    {
      "entropy": 6.401629161834717,
      "epoch": 0.7458933062229118,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004953863385002541,
      "loss": 6.3226,
      "mean_token_accuracy": 0.1303832098841667,
      "num_tokens": 15086997.0,
      "step": 6970
    },
    {
      "entropy": 6.55853042602539,
      "epoch": 0.7464283803306758,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004953786324718661,
      "loss": 6.4814,
      "mean_token_accuracy": 0.12583869323134422,
      "num_tokens": 15098625.0,
      "step": 6975
    },
    {
      "entropy": 6.437220764160156,
      "epoch": 0.7469634544384397,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004953709200800761,
      "loss": 6.2873,
      "mean_token_accuracy": 0.13968943431973457,
      "num_tokens": 15108265.0,
      "step": 6980
    },
    {
      "entropy": 6.4666337966918945,
      "epoch": 0.7474985285462037,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004953632013251068,
      "loss": 6.3984,
      "mean_token_accuracy": 0.12585160583257676,
      "num_tokens": 15118794.0,
      "step": 6985
    },
    {
      "entropy": 6.531051254272461,
      "epoch": 0.7480336026539676,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004953554762071811,
      "loss": 6.417,
      "mean_token_accuracy": 0.12451769188046455,
      "num_tokens": 15130084.0,
      "step": 6990
    },
    {
      "entropy": 6.491759157180786,
      "epoch": 0.7485686767617316,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004953477447265218,
      "loss": 6.4153,
      "mean_token_accuracy": 0.1272393710911274,
      "num_tokens": 15142193.0,
      "step": 6995
    },
    {
      "entropy": 6.559592580795288,
      "epoch": 0.7491037508694954,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004953400068833524,
      "loss": 6.5311,
      "mean_token_accuracy": 0.12169862687587737,
      "num_tokens": 15152817.0,
      "step": 7000
    },
    {
      "entropy": 6.473178482055664,
      "epoch": 0.7496388249772593,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004953322626778964,
      "loss": 6.2029,
      "mean_token_accuracy": 0.13469267189502715,
      "num_tokens": 15162352.0,
      "step": 7005
    },
    {
      "entropy": 6.5206136226654055,
      "epoch": 0.7501738990850233,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004953245121103771,
      "loss": 6.3862,
      "mean_token_accuracy": 0.13724407851696013,
      "num_tokens": 15173091.0,
      "step": 7010
    },
    {
      "entropy": 6.57209210395813,
      "epoch": 0.7507089731927872,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004953167551810185,
      "loss": 6.5461,
      "mean_token_accuracy": 0.11932987794280052,
      "num_tokens": 15184422.0,
      "step": 7015
    },
    {
      "entropy": 6.436837577819825,
      "epoch": 0.7512440473005512,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004953089918900447,
      "loss": 6.2766,
      "mean_token_accuracy": 0.1402566373348236,
      "num_tokens": 15194765.0,
      "step": 7020
    },
    {
      "entropy": 6.511735343933106,
      "epoch": 0.751779121408315,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004953012222376795,
      "loss": 6.4372,
      "mean_token_accuracy": 0.12373902946710587,
      "num_tokens": 15206007.0,
      "step": 7025
    },
    {
      "entropy": 6.521138668060303,
      "epoch": 0.752314195516079,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004952934462241476,
      "loss": 6.3877,
      "mean_token_accuracy": 0.12996315509080886,
      "num_tokens": 15216634.0,
      "step": 7030
    },
    {
      "entropy": 6.534381771087647,
      "epoch": 0.7528492696238429,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004952856638496734,
      "loss": 6.4571,
      "mean_token_accuracy": 0.12829971462488174,
      "num_tokens": 15226931.0,
      "step": 7035
    },
    {
      "entropy": 6.488435840606689,
      "epoch": 0.7533843437316068,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004952778751144817,
      "loss": 6.4234,
      "mean_token_accuracy": 0.12561191916465758,
      "num_tokens": 15239034.0,
      "step": 7040
    },
    {
      "entropy": 6.569809341430664,
      "epoch": 0.7539194178393708,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004952700800187971,
      "loss": 6.4168,
      "mean_token_accuracy": 0.12696855142712593,
      "num_tokens": 15250088.0,
      "step": 7045
    },
    {
      "entropy": 6.574549913406372,
      "epoch": 0.7544544919471347,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000495262278562845,
      "loss": 6.3615,
      "mean_token_accuracy": 0.13199912756681442,
      "num_tokens": 15261150.0,
      "step": 7050
    },
    {
      "entropy": 6.505710458755493,
      "epoch": 0.7549895660548986,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004952544707468506,
      "loss": 6.4317,
      "mean_token_accuracy": 0.12971632182598114,
      "num_tokens": 15270865.0,
      "step": 7055
    },
    {
      "entropy": 6.550460767745972,
      "epoch": 0.7555246401626625,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004952466565710391,
      "loss": 6.4439,
      "mean_token_accuracy": 0.12355516329407693,
      "num_tokens": 15282809.0,
      "step": 7060
    },
    {
      "entropy": 6.537486124038696,
      "epoch": 0.7560597142704265,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004952388360356366,
      "loss": 6.4153,
      "mean_token_accuracy": 0.13766660764813424,
      "num_tokens": 15293166.0,
      "step": 7065
    },
    {
      "entropy": 6.526575565338135,
      "epoch": 0.7565947883781904,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004952310091408685,
      "loss": 6.3898,
      "mean_token_accuracy": 0.13106199577450753,
      "num_tokens": 15302985.0,
      "step": 7070
    },
    {
      "entropy": 6.547119808197022,
      "epoch": 0.7571298624859543,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004952231758869611,
      "loss": 6.4308,
      "mean_token_accuracy": 0.13215380236506463,
      "num_tokens": 15314737.0,
      "step": 7075
    },
    {
      "entropy": 6.512074613571167,
      "epoch": 0.7576649365937183,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004952153362741403,
      "loss": 6.3748,
      "mean_token_accuracy": 0.13392539247870444,
      "num_tokens": 15325441.0,
      "step": 7080
    },
    {
      "entropy": 6.474653863906861,
      "epoch": 0.7582000107014821,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004952074903026327,
      "loss": 6.3053,
      "mean_token_accuracy": 0.13513123542070388,
      "num_tokens": 15334939.0,
      "step": 7085
    },
    {
      "entropy": 6.476730108261108,
      "epoch": 0.7587350848092461,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004951996379726647,
      "loss": 6.4629,
      "mean_token_accuracy": 0.12283681035041809,
      "num_tokens": 15345655.0,
      "step": 7090
    },
    {
      "entropy": 6.461888980865479,
      "epoch": 0.75927015891701,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004951917792844631,
      "loss": 6.2908,
      "mean_token_accuracy": 0.1339256890118122,
      "num_tokens": 15356035.0,
      "step": 7095
    },
    {
      "entropy": 6.531403064727783,
      "epoch": 0.759805233024774,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004951839142382548,
      "loss": 6.4034,
      "mean_token_accuracy": 0.12990091666579245,
      "num_tokens": 15368509.0,
      "step": 7100
    },
    {
      "entropy": 6.541523408889771,
      "epoch": 0.7603403071325379,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000495176042834267,
      "loss": 6.3865,
      "mean_token_accuracy": 0.13039031401276588,
      "num_tokens": 15379662.0,
      "step": 7105
    },
    {
      "entropy": 6.4948807716369625,
      "epoch": 0.7608753812403017,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004951681650727268,
      "loss": 6.3693,
      "mean_token_accuracy": 0.12637224569916725,
      "num_tokens": 15389698.0,
      "step": 7110
    },
    {
      "entropy": 6.577439165115356,
      "epoch": 0.7614104553480657,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004951602809538619,
      "loss": 6.4692,
      "mean_token_accuracy": 0.12436027377843857,
      "num_tokens": 15400724.0,
      "step": 7115
    },
    {
      "entropy": 6.652684736251831,
      "epoch": 0.7619455294558296,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004951523904778997,
      "loss": 6.4626,
      "mean_token_accuracy": 0.1268015682697296,
      "num_tokens": 15410732.0,
      "step": 7120
    },
    {
      "entropy": 6.485833644866943,
      "epoch": 0.7624806035635936,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004951444936450682,
      "loss": 6.3761,
      "mean_token_accuracy": 0.13191442266106607,
      "num_tokens": 15420185.0,
      "step": 7125
    },
    {
      "entropy": 6.449607181549072,
      "epoch": 0.7630156776713575,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004951365904555954,
      "loss": 6.2846,
      "mean_token_accuracy": 0.13596602231264115,
      "num_tokens": 15431904.0,
      "step": 7130
    },
    {
      "entropy": 6.468115425109863,
      "epoch": 0.7635507517791215,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004951286809097094,
      "loss": 6.4221,
      "mean_token_accuracy": 0.12676772549748422,
      "num_tokens": 15442650.0,
      "step": 7135
    },
    {
      "entropy": 6.458074331283569,
      "epoch": 0.7640858258868853,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004951207650076388,
      "loss": 6.3891,
      "mean_token_accuracy": 0.12797302827239038,
      "num_tokens": 15454273.0,
      "step": 7140
    },
    {
      "entropy": 6.5104332447052,
      "epoch": 0.7646208999946492,
      "grad_norm": 1.109375,
      "learning_rate": 0.000495112842749612,
      "loss": 6.3702,
      "mean_token_accuracy": 0.13236722573637963,
      "num_tokens": 15465751.0,
      "step": 7145
    },
    {
      "entropy": 6.512122869491577,
      "epoch": 0.7651559741024132,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004951049141358578,
      "loss": 6.4765,
      "mean_token_accuracy": 0.12459504380822181,
      "num_tokens": 15475340.0,
      "step": 7150
    },
    {
      "entropy": 6.575246095657349,
      "epoch": 0.7656910482101771,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004950969791666051,
      "loss": 6.4474,
      "mean_token_accuracy": 0.1253798261284828,
      "num_tokens": 15486246.0,
      "step": 7155
    },
    {
      "entropy": 6.595798397064209,
      "epoch": 0.7662261223179411,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004950890378420831,
      "loss": 6.4376,
      "mean_token_accuracy": 0.12804016768932341,
      "num_tokens": 15497331.0,
      "step": 7160
    },
    {
      "entropy": 6.525610780715942,
      "epoch": 0.766761196425705,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000495081090162521,
      "loss": 6.3774,
      "mean_token_accuracy": 0.13211151733994483,
      "num_tokens": 15508036.0,
      "step": 7165
    },
    {
      "entropy": 6.48620719909668,
      "epoch": 0.7672962705334689,
      "grad_norm": 1.125,
      "learning_rate": 0.0004950731361281483,
      "loss": 6.3907,
      "mean_token_accuracy": 0.12933095246553422,
      "num_tokens": 15518381.0,
      "step": 7170
    },
    {
      "entropy": 6.511308240890503,
      "epoch": 0.7678313446412328,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004950651757391948,
      "loss": 6.4011,
      "mean_token_accuracy": 0.12953920066356658,
      "num_tokens": 15528939.0,
      "step": 7175
    },
    {
      "entropy": 6.546399641036987,
      "epoch": 0.7683664187489967,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004950572089958904,
      "loss": 6.3775,
      "mean_token_accuracy": 0.12832187190651895,
      "num_tokens": 15539718.0,
      "step": 7180
    },
    {
      "entropy": 6.571915531158448,
      "epoch": 0.7689014928567607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004950492358984648,
      "loss": 6.5372,
      "mean_token_accuracy": 0.12224936410784722,
      "num_tokens": 15550970.0,
      "step": 7185
    },
    {
      "entropy": 6.539775609970093,
      "epoch": 0.7694365669645246,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004950412564471486,
      "loss": 6.3162,
      "mean_token_accuracy": 0.1328173279762268,
      "num_tokens": 15560490.0,
      "step": 7190
    },
    {
      "entropy": 6.395464038848877,
      "epoch": 0.7699716410722885,
      "grad_norm": 1.140625,
      "learning_rate": 0.000495033270642172,
      "loss": 6.3162,
      "mean_token_accuracy": 0.13221260383725167,
      "num_tokens": 15570188.0,
      "step": 7195
    },
    {
      "entropy": 6.453646039962768,
      "epoch": 0.7705067151800524,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004950252784837655,
      "loss": 6.3325,
      "mean_token_accuracy": 0.1252153255045414,
      "num_tokens": 15581446.0,
      "step": 7200
    },
    {
      "entropy": 6.546197080612183,
      "epoch": 0.7710417892878164,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004950172799721601,
      "loss": 6.4476,
      "mean_token_accuracy": 0.1269929975271225,
      "num_tokens": 15594362.0,
      "step": 7205
    },
    {
      "entropy": 6.507910299301147,
      "epoch": 0.7715768633955803,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004950092751075866,
      "loss": 6.3161,
      "mean_token_accuracy": 0.14033825621008872,
      "num_tokens": 15605531.0,
      "step": 7210
    },
    {
      "entropy": 6.468828296661377,
      "epoch": 0.7721119375033442,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004950012638902763,
      "loss": 6.3742,
      "mean_token_accuracy": 0.13031978532671928,
      "num_tokens": 15614962.0,
      "step": 7215
    },
    {
      "entropy": 6.428341960906982,
      "epoch": 0.7726470116111082,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004949932463204603,
      "loss": 6.4055,
      "mean_token_accuracy": 0.12961461320519446,
      "num_tokens": 15625475.0,
      "step": 7220
    },
    {
      "entropy": 6.4668690204620365,
      "epoch": 0.773182085718872,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004949852223983703,
      "loss": 6.3751,
      "mean_token_accuracy": 0.13136933892965316,
      "num_tokens": 15637359.0,
      "step": 7225
    },
    {
      "entropy": 6.6264551162719725,
      "epoch": 0.773717159826636,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004949771921242379,
      "loss": 6.428,
      "mean_token_accuracy": 0.12189994752407074,
      "num_tokens": 15648368.0,
      "step": 7230
    },
    {
      "entropy": 6.5627178192138675,
      "epoch": 0.7742522339343999,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004949691554982951,
      "loss": 6.3352,
      "mean_token_accuracy": 0.13558300137519835,
      "num_tokens": 15659295.0,
      "step": 7235
    },
    {
      "entropy": 6.4363445281982425,
      "epoch": 0.7747873080421639,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004949611125207737,
      "loss": 6.2911,
      "mean_token_accuracy": 0.13113539069890975,
      "num_tokens": 15669975.0,
      "step": 7240
    },
    {
      "entropy": 6.428279304504395,
      "epoch": 0.7753223821499278,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000494953063191906,
      "loss": 6.3836,
      "mean_token_accuracy": 0.12822090610861778,
      "num_tokens": 15680343.0,
      "step": 7245
    },
    {
      "entropy": 6.530182790756226,
      "epoch": 0.7758574562576916,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004949450075119247,
      "loss": 6.4536,
      "mean_token_accuracy": 0.12421686574816704,
      "num_tokens": 15692246.0,
      "step": 7250
    },
    {
      "entropy": 6.563816976547241,
      "epoch": 0.7763925303654556,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004949369454810621,
      "loss": 6.4057,
      "mean_token_accuracy": 0.13185440003871918,
      "num_tokens": 15704262.0,
      "step": 7255
    },
    {
      "entropy": 6.472266149520874,
      "epoch": 0.7769276044732195,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004949288770995512,
      "loss": 6.4006,
      "mean_token_accuracy": 0.12933578193187714,
      "num_tokens": 15715078.0,
      "step": 7260
    },
    {
      "entropy": 6.513643789291382,
      "epoch": 0.7774626785809835,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004949208023676249,
      "loss": 6.4331,
      "mean_token_accuracy": 0.13019643425941468,
      "num_tokens": 15726503.0,
      "step": 7265
    },
    {
      "entropy": 6.443303871154785,
      "epoch": 0.7779977526887474,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004949127212855161,
      "loss": 6.3139,
      "mean_token_accuracy": 0.13357823193073273,
      "num_tokens": 15738854.0,
      "step": 7270
    },
    {
      "entropy": 6.489425325393677,
      "epoch": 0.7785328267965114,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004949046338534587,
      "loss": 6.4647,
      "mean_token_accuracy": 0.1271779641509056,
      "num_tokens": 15750836.0,
      "step": 7275
    },
    {
      "entropy": 6.643733310699463,
      "epoch": 0.7790679009042752,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004948965400716857,
      "loss": 6.4848,
      "mean_token_accuracy": 0.12600264996290206,
      "num_tokens": 15762355.0,
      "step": 7280
    },
    {
      "entropy": 6.609560966491699,
      "epoch": 0.7796029750120391,
      "grad_norm": 1.125,
      "learning_rate": 0.000494888439940431,
      "loss": 6.4578,
      "mean_token_accuracy": 0.1298013560473919,
      "num_tokens": 15772952.0,
      "step": 7285
    },
    {
      "entropy": 6.449377250671387,
      "epoch": 0.7801380491198031,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004948803334599286,
      "loss": 6.2951,
      "mean_token_accuracy": 0.13864458948373795,
      "num_tokens": 15782997.0,
      "step": 7290
    },
    {
      "entropy": 6.51337080001831,
      "epoch": 0.780673123227567,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004948722206304123,
      "loss": 6.3963,
      "mean_token_accuracy": 0.13196406364440919,
      "num_tokens": 15794060.0,
      "step": 7295
    },
    {
      "entropy": 6.547667980194092,
      "epoch": 0.781208197335331,
      "grad_norm": 1.125,
      "learning_rate": 0.0004948641014521167,
      "loss": 6.4405,
      "mean_token_accuracy": 0.12708537578582763,
      "num_tokens": 15804149.0,
      "step": 7300
    },
    {
      "entropy": 6.495831680297852,
      "epoch": 0.7817432714430949,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000494855975925276,
      "loss": 6.3979,
      "mean_token_accuracy": 0.1315051406621933,
      "num_tokens": 15814643.0,
      "step": 7305
    },
    {
      "entropy": 6.498531341552734,
      "epoch": 0.7822783455508588,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004948478440501249,
      "loss": 6.3419,
      "mean_token_accuracy": 0.1307312585413456,
      "num_tokens": 15824640.0,
      "step": 7310
    },
    {
      "entropy": 6.481297731399536,
      "epoch": 0.7828134196586227,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004948397058268982,
      "loss": 6.3662,
      "mean_token_accuracy": 0.13351597711443902,
      "num_tokens": 15836160.0,
      "step": 7315
    },
    {
      "entropy": 6.51258955001831,
      "epoch": 0.7833484937663866,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004948315612558308,
      "loss": 6.4138,
      "mean_token_accuracy": 0.1286468580365181,
      "num_tokens": 15846886.0,
      "step": 7320
    },
    {
      "entropy": 6.511155891418457,
      "epoch": 0.7838835678741506,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000494823410337158,
      "loss": 6.3307,
      "mean_token_accuracy": 0.13343393728137015,
      "num_tokens": 15857309.0,
      "step": 7325
    },
    {
      "entropy": 6.498040056228637,
      "epoch": 0.7844186419819145,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004948152530711153,
      "loss": 6.4654,
      "mean_token_accuracy": 0.127084119617939,
      "num_tokens": 15869620.0,
      "step": 7330
    },
    {
      "entropy": 6.484524965286255,
      "epoch": 0.7849537160896785,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004948070894579378,
      "loss": 6.3249,
      "mean_token_accuracy": 0.13769325762987136,
      "num_tokens": 15880915.0,
      "step": 7335
    },
    {
      "entropy": 6.466250991821289,
      "epoch": 0.7854887901974423,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004947989194978616,
      "loss": 6.374,
      "mean_token_accuracy": 0.13448203429579736,
      "num_tokens": 15891150.0,
      "step": 7340
    },
    {
      "entropy": 6.501854610443115,
      "epoch": 0.7860238643052063,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004947907431911225,
      "loss": 6.3869,
      "mean_token_accuracy": 0.13170364424586295,
      "num_tokens": 15902526.0,
      "step": 7345
    },
    {
      "entropy": 6.4714038372039795,
      "epoch": 0.7865589384129702,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004947825605379566,
      "loss": 6.3306,
      "mean_token_accuracy": 0.13312023654580116,
      "num_tokens": 15913425.0,
      "step": 7350
    },
    {
      "entropy": 6.522894859313965,
      "epoch": 0.7870940125207341,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004947743715386,
      "loss": 6.3075,
      "mean_token_accuracy": 0.13080790415406227,
      "num_tokens": 15924324.0,
      "step": 7355
    },
    {
      "entropy": 6.453833723068238,
      "epoch": 0.7876290866284981,
      "grad_norm": 1.375,
      "learning_rate": 0.0004947661761932894,
      "loss": 6.3629,
      "mean_token_accuracy": 0.12942021265625953,
      "num_tokens": 15935533.0,
      "step": 7360
    },
    {
      "entropy": 6.470954418182373,
      "epoch": 0.7881641607362619,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004947579745022613,
      "loss": 6.3116,
      "mean_token_accuracy": 0.1286455772817135,
      "num_tokens": 15947007.0,
      "step": 7365
    },
    {
      "entropy": 6.504639148712158,
      "epoch": 0.7886992348440259,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004947497664657527,
      "loss": 6.3977,
      "mean_token_accuracy": 0.1285076804459095,
      "num_tokens": 15957405.0,
      "step": 7370
    },
    {
      "entropy": 6.498816251754761,
      "epoch": 0.7892343089517898,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004947415520840003,
      "loss": 6.3646,
      "mean_token_accuracy": 0.13653657957911491,
      "num_tokens": 15969895.0,
      "step": 7375
    },
    {
      "entropy": 6.485688638687134,
      "epoch": 0.7897693830595538,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004947333313572416,
      "loss": 6.2821,
      "mean_token_accuracy": 0.1394954301416874,
      "num_tokens": 15979854.0,
      "step": 7380
    },
    {
      "entropy": 6.390361785888672,
      "epoch": 0.7903044571673177,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004947251042857137,
      "loss": 6.2829,
      "mean_token_accuracy": 0.1381194122135639,
      "num_tokens": 15991341.0,
      "step": 7385
    },
    {
      "entropy": 6.399601984024048,
      "epoch": 0.7908395312750816,
      "grad_norm": 1.125,
      "learning_rate": 0.0004947168708696544,
      "loss": 6.3352,
      "mean_token_accuracy": 0.13659097105264664,
      "num_tokens": 16001944.0,
      "step": 7390
    },
    {
      "entropy": 6.412859869003296,
      "epoch": 0.7913746053828455,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004947086311093013,
      "loss": 6.3324,
      "mean_token_accuracy": 0.13536788076162337,
      "num_tokens": 16011771.0,
      "step": 7395
    },
    {
      "entropy": 6.512271595001221,
      "epoch": 0.7919096794906094,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004947003850048924,
      "loss": 6.4966,
      "mean_token_accuracy": 0.11639693081378936,
      "num_tokens": 16023901.0,
      "step": 7400
    },
    {
      "entropy": 6.560624074935913,
      "epoch": 0.7924447535983734,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004946921325566656,
      "loss": 6.4009,
      "mean_token_accuracy": 0.1253039740025997,
      "num_tokens": 16035014.0,
      "step": 7405
    },
    {
      "entropy": 6.512752962112427,
      "epoch": 0.7929798277061373,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004946838737648595,
      "loss": 6.4672,
      "mean_token_accuracy": 0.12525063306093215,
      "num_tokens": 16046927.0,
      "step": 7410
    },
    {
      "entropy": 6.544916105270386,
      "epoch": 0.7935149018139013,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004946756086297124,
      "loss": 6.3896,
      "mean_token_accuracy": 0.13451874628663063,
      "num_tokens": 16057341.0,
      "step": 7415
    },
    {
      "entropy": 6.507423257827758,
      "epoch": 0.7940499759216652,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004946673371514628,
      "loss": 6.3553,
      "mean_token_accuracy": 0.13357612043619155,
      "num_tokens": 16068408.0,
      "step": 7420
    },
    {
      "entropy": 6.472815227508545,
      "epoch": 0.794585050029429,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004946590593303499,
      "loss": 6.3391,
      "mean_token_accuracy": 0.1377520516514778,
      "num_tokens": 16078693.0,
      "step": 7425
    },
    {
      "entropy": 6.445299482345581,
      "epoch": 0.795120124137193,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004946507751666124,
      "loss": 6.3964,
      "mean_token_accuracy": 0.1341624528169632,
      "num_tokens": 16089127.0,
      "step": 7430
    },
    {
      "entropy": 6.412180757522583,
      "epoch": 0.7956551982449569,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004946424846604897,
      "loss": 6.2916,
      "mean_token_accuracy": 0.1397658295929432,
      "num_tokens": 16099726.0,
      "step": 7435
    },
    {
      "entropy": 6.492917585372925,
      "epoch": 0.7961902723527209,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004946341878122212,
      "loss": 6.3501,
      "mean_token_accuracy": 0.13139667809009553,
      "num_tokens": 16111017.0,
      "step": 7440
    },
    {
      "entropy": 6.486097431182861,
      "epoch": 0.7967253464604848,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004946258846220462,
      "loss": 6.4238,
      "mean_token_accuracy": 0.13282189816236495,
      "num_tokens": 16121406.0,
      "step": 7445
    },
    {
      "entropy": 6.492733526229858,
      "epoch": 0.7972604205682488,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004946175750902049,
      "loss": 6.3519,
      "mean_token_accuracy": 0.13248605877161027,
      "num_tokens": 16131587.0,
      "step": 7450
    },
    {
      "entropy": 6.4529060363769535,
      "epoch": 0.7977954946760126,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004946092592169368,
      "loss": 6.3117,
      "mean_token_accuracy": 0.1373440332710743,
      "num_tokens": 16142578.0,
      "step": 7455
    },
    {
      "entropy": 6.4882293224334715,
      "epoch": 0.7983305687837765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004946009370024822,
      "loss": 6.305,
      "mean_token_accuracy": 0.1364786833524704,
      "num_tokens": 16153501.0,
      "step": 7460
    },
    {
      "entropy": 6.442016315460205,
      "epoch": 0.7988656428915405,
      "grad_norm": 2.25,
      "learning_rate": 0.0004945926084470814,
      "loss": 6.3908,
      "mean_token_accuracy": 0.12975837886333466,
      "num_tokens": 16164273.0,
      "step": 7465
    },
    {
      "entropy": 6.52316370010376,
      "epoch": 0.7994007169993044,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004945842735509749,
      "loss": 6.382,
      "mean_token_accuracy": 0.1321078695356846,
      "num_tokens": 16175071.0,
      "step": 7470
    },
    {
      "entropy": 6.533743858337402,
      "epoch": 0.7999357911070684,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004945759323144034,
      "loss": 6.4076,
      "mean_token_accuracy": 0.12893555089831352,
      "num_tokens": 16185404.0,
      "step": 7475
    },
    {
      "entropy": 6.498791265487671,
      "epoch": 0.8004708652148322,
      "grad_norm": 1.25,
      "learning_rate": 0.0004945675847376077,
      "loss": 6.3906,
      "mean_token_accuracy": 0.1299203760921955,
      "num_tokens": 16196881.0,
      "step": 7480
    },
    {
      "entropy": 6.557330369949341,
      "epoch": 0.8010059393225962,
      "grad_norm": 1.125,
      "learning_rate": 0.0004945592308208288,
      "loss": 6.4316,
      "mean_token_accuracy": 0.12797078713774682,
      "num_tokens": 16207534.0,
      "step": 7485
    },
    {
      "entropy": 6.523251342773437,
      "epoch": 0.8015410134303601,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000494550870564308,
      "loss": 6.3058,
      "mean_token_accuracy": 0.13882087022066117,
      "num_tokens": 16217371.0,
      "step": 7490
    },
    {
      "entropy": 6.494204521179199,
      "epoch": 0.802076087538124,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004945425039682866,
      "loss": 6.3575,
      "mean_token_accuracy": 0.13369035199284554,
      "num_tokens": 16227898.0,
      "step": 7495
    },
    {
      "entropy": 6.437259006500244,
      "epoch": 0.802611161645888,
      "grad_norm": 1.25,
      "learning_rate": 0.0004945341310330064,
      "loss": 6.3352,
      "mean_token_accuracy": 0.13994823172688484,
      "num_tokens": 16238966.0,
      "step": 7500
    },
    {
      "entropy": 6.4564601421356205,
      "epoch": 0.8031462357536518,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004945257517587089,
      "loss": 6.3725,
      "mean_token_accuracy": 0.14170586168766022,
      "num_tokens": 16250129.0,
      "step": 7505
    },
    {
      "entropy": 6.500148057937622,
      "epoch": 0.8036813098614158,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004945173661456361,
      "loss": 6.4034,
      "mean_token_accuracy": 0.12952237501740455,
      "num_tokens": 16260877.0,
      "step": 7510
    },
    {
      "entropy": 6.561743211746216,
      "epoch": 0.8042163839691797,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004945089741940303,
      "loss": 6.3533,
      "mean_token_accuracy": 0.13671437054872512,
      "num_tokens": 16270548.0,
      "step": 7515
    },
    {
      "entropy": 6.476692771911621,
      "epoch": 0.8047514580769437,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004945005759041338,
      "loss": 6.3738,
      "mean_token_accuracy": 0.13085496798157692,
      "num_tokens": 16281014.0,
      "step": 7520
    },
    {
      "entropy": 6.519422149658203,
      "epoch": 0.8052865321847076,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004944921712761889,
      "loss": 6.4332,
      "mean_token_accuracy": 0.13116990849375726,
      "num_tokens": 16291653.0,
      "step": 7525
    },
    {
      "entropy": 6.5457864761352536,
      "epoch": 0.8058216062924715,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004944837603104383,
      "loss": 6.2911,
      "mean_token_accuracy": 0.13363028466701507,
      "num_tokens": 16301703.0,
      "step": 7530
    },
    {
      "entropy": 6.427401351928711,
      "epoch": 0.8063566804002354,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004944753430071252,
      "loss": 6.3453,
      "mean_token_accuracy": 0.13163626119494437,
      "num_tokens": 16312252.0,
      "step": 7535
    },
    {
      "entropy": 6.455341339111328,
      "epoch": 0.8068917545079993,
      "grad_norm": 1.125,
      "learning_rate": 0.0004944669193664923,
      "loss": 6.3254,
      "mean_token_accuracy": 0.12827802076935768,
      "num_tokens": 16324073.0,
      "step": 7540
    },
    {
      "entropy": 6.54691252708435,
      "epoch": 0.8074268286157633,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004944584893887829,
      "loss": 6.3391,
      "mean_token_accuracy": 0.13218172937631606,
      "num_tokens": 16334637.0,
      "step": 7545
    },
    {
      "entropy": 6.483603620529175,
      "epoch": 0.8079619027235272,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004944500530742404,
      "loss": 6.4684,
      "mean_token_accuracy": 0.12402921468019486,
      "num_tokens": 16345581.0,
      "step": 7550
    },
    {
      "entropy": 6.532661581039429,
      "epoch": 0.8084969768312912,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004944416104231086,
      "loss": 6.4471,
      "mean_token_accuracy": 0.12670645713806153,
      "num_tokens": 16356995.0,
      "step": 7555
    },
    {
      "entropy": 6.576150703430176,
      "epoch": 0.8090320509390551,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004944331614356311,
      "loss": 6.3997,
      "mean_token_accuracy": 0.1279869444668293,
      "num_tokens": 16368841.0,
      "step": 7560
    },
    {
      "entropy": 6.501333379745484,
      "epoch": 0.8095671250468189,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004944247061120519,
      "loss": 6.3241,
      "mean_token_accuracy": 0.13083065152168274,
      "num_tokens": 16379295.0,
      "step": 7565
    },
    {
      "entropy": 6.448875951766968,
      "epoch": 0.8101021991545829,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004944162444526151,
      "loss": 6.3528,
      "mean_token_accuracy": 0.1341070421040058,
      "num_tokens": 16389512.0,
      "step": 7570
    },
    {
      "entropy": 6.364917612075805,
      "epoch": 0.8106372732623468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004944077764575651,
      "loss": 6.2789,
      "mean_token_accuracy": 0.14279957339167595,
      "num_tokens": 16400720.0,
      "step": 7575
    },
    {
      "entropy": 6.418307447433472,
      "epoch": 0.8111723473701108,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004943993021271463,
      "loss": 6.3437,
      "mean_token_accuracy": 0.13257088959217073,
      "num_tokens": 16411278.0,
      "step": 7580
    },
    {
      "entropy": 6.536353874206543,
      "epoch": 0.8117074214778747,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004943908214616035,
      "loss": 6.3993,
      "mean_token_accuracy": 0.13009507954120636,
      "num_tokens": 16421665.0,
      "step": 7585
    },
    {
      "entropy": 6.564996147155762,
      "epoch": 0.8122424955856387,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004943823344611818,
      "loss": 6.4455,
      "mean_token_accuracy": 0.1258238859474659,
      "num_tokens": 16432943.0,
      "step": 7590
    },
    {
      "entropy": 6.510878992080689,
      "epoch": 0.8127775696934025,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004943738411261258,
      "loss": 6.3478,
      "mean_token_accuracy": 0.1286700777709484,
      "num_tokens": 16444506.0,
      "step": 7595
    },
    {
      "entropy": 6.421297883987426,
      "epoch": 0.8133126438011664,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004943653414566809,
      "loss": 6.3697,
      "mean_token_accuracy": 0.12976935356855393,
      "num_tokens": 16455403.0,
      "step": 7600
    },
    {
      "entropy": 6.552541732788086,
      "epoch": 0.8138477179089304,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004943568354530927,
      "loss": 6.4763,
      "mean_token_accuracy": 0.12392953410744667,
      "num_tokens": 16465616.0,
      "step": 7605
    },
    {
      "entropy": 6.59772744178772,
      "epoch": 0.8143827920166943,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004943483231156068,
      "loss": 6.4444,
      "mean_token_accuracy": 0.12751524969935418,
      "num_tokens": 16476850.0,
      "step": 7610
    },
    {
      "entropy": 6.578122758865357,
      "epoch": 0.8149178661244583,
      "grad_norm": 1.125,
      "learning_rate": 0.0004943398044444687,
      "loss": 6.3891,
      "mean_token_accuracy": 0.13252009749412536,
      "num_tokens": 16489118.0,
      "step": 7615
    },
    {
      "entropy": 6.510885381698609,
      "epoch": 0.8154529402322221,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004943312794399246,
      "loss": 6.3391,
      "mean_token_accuracy": 0.13087911158800125,
      "num_tokens": 16499336.0,
      "step": 7620
    },
    {
      "entropy": 6.494793939590454,
      "epoch": 0.8159880143399861,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004943227481022207,
      "loss": 6.4157,
      "mean_token_accuracy": 0.12642048820853233,
      "num_tokens": 16510158.0,
      "step": 7625
    },
    {
      "entropy": 6.436304807662964,
      "epoch": 0.81652308844775,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004943142104316033,
      "loss": 6.3514,
      "mean_token_accuracy": 0.13307944238185881,
      "num_tokens": 16521080.0,
      "step": 7630
    },
    {
      "entropy": 6.49186692237854,
      "epoch": 0.8170581625555139,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004943056664283189,
      "loss": 6.3717,
      "mean_token_accuracy": 0.13422367423772813,
      "num_tokens": 16531072.0,
      "step": 7635
    },
    {
      "entropy": 6.5786233901977536,
      "epoch": 0.8175932366632779,
      "grad_norm": 1.09375,
      "learning_rate": 0.000494297116092614,
      "loss": 6.4516,
      "mean_token_accuracy": 0.12318632155656814,
      "num_tokens": 16542588.0,
      "step": 7640
    },
    {
      "entropy": 6.539233589172364,
      "epoch": 0.8181283107710418,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004942885594247359,
      "loss": 6.3202,
      "mean_token_accuracy": 0.13083992674946784,
      "num_tokens": 16553466.0,
      "step": 7645
    },
    {
      "entropy": 6.409227275848389,
      "epoch": 0.8186633848788057,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004942799964249314,
      "loss": 6.3346,
      "mean_token_accuracy": 0.1320448063313961,
      "num_tokens": 16564084.0,
      "step": 7650
    },
    {
      "entropy": 6.427356386184693,
      "epoch": 0.8191984589865696,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004942714270934479,
      "loss": 6.3422,
      "mean_token_accuracy": 0.14386487230658532,
      "num_tokens": 16575665.0,
      "step": 7655
    },
    {
      "entropy": 6.495164585113526,
      "epoch": 0.8197335330943336,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004942628514305326,
      "loss": 6.3538,
      "mean_token_accuracy": 0.1278291217982769,
      "num_tokens": 16586518.0,
      "step": 7660
    },
    {
      "entropy": 6.444916820526123,
      "epoch": 0.8202686072020975,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004942542694364333,
      "loss": 6.3627,
      "mean_token_accuracy": 0.1335558131337166,
      "num_tokens": 16597345.0,
      "step": 7665
    },
    {
      "entropy": 6.5291589260101315,
      "epoch": 0.8208036813098614,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004942456811113978,
      "loss": 6.3841,
      "mean_token_accuracy": 0.13352123349905015,
      "num_tokens": 16609586.0,
      "step": 7670
    },
    {
      "entropy": 6.532260084152222,
      "epoch": 0.8213387554176254,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004942370864556741,
      "loss": 6.2514,
      "mean_token_accuracy": 0.14125269502401352,
      "num_tokens": 16620165.0,
      "step": 7675
    },
    {
      "entropy": 6.409881162643432,
      "epoch": 0.8218738295253892,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004942284854695104,
      "loss": 6.3702,
      "mean_token_accuracy": 0.13256416469812393,
      "num_tokens": 16631200.0,
      "step": 7680
    },
    {
      "entropy": 6.46362795829773,
      "epoch": 0.8224089036331532,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004942198781531549,
      "loss": 6.3989,
      "mean_token_accuracy": 0.12870605513453484,
      "num_tokens": 16642553.0,
      "step": 7685
    },
    {
      "entropy": 6.5733600616455075,
      "epoch": 0.8229439777409171,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004942112645068561,
      "loss": 6.416,
      "mean_token_accuracy": 0.12830686196684837,
      "num_tokens": 16653431.0,
      "step": 7690
    },
    {
      "entropy": 6.468099308013916,
      "epoch": 0.8234790518486811,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000494202644530863,
      "loss": 6.3484,
      "mean_token_accuracy": 0.13155423626303672,
      "num_tokens": 16664880.0,
      "step": 7695
    },
    {
      "entropy": 6.419212770462036,
      "epoch": 0.824014125956445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004941940182254244,
      "loss": 6.3568,
      "mean_token_accuracy": 0.13886465951800347,
      "num_tokens": 16676224.0,
      "step": 7700
    },
    {
      "entropy": 6.504215860366822,
      "epoch": 0.8245492000642088,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004941853855907892,
      "loss": 6.4178,
      "mean_token_accuracy": 0.12964134514331818,
      "num_tokens": 16687732.0,
      "step": 7705
    },
    {
      "entropy": 6.539362001419067,
      "epoch": 0.8250842741719728,
      "grad_norm": 1.25,
      "learning_rate": 0.0004941767466272068,
      "loss": 6.3519,
      "mean_token_accuracy": 0.135162653028965,
      "num_tokens": 16698257.0,
      "step": 7710
    },
    {
      "entropy": 6.485494470596313,
      "epoch": 0.8256193482797367,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004941681013349267,
      "loss": 6.3135,
      "mean_token_accuracy": 0.1323456011712551,
      "num_tokens": 16708860.0,
      "step": 7715
    },
    {
      "entropy": 6.445706129074097,
      "epoch": 0.8261544223875007,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004941594497141985,
      "loss": 6.3187,
      "mean_token_accuracy": 0.13786930367350578,
      "num_tokens": 16719987.0,
      "step": 7720
    },
    {
      "entropy": 6.497876167297363,
      "epoch": 0.8266894964952646,
      "grad_norm": 1.25,
      "learning_rate": 0.0004941507917652718,
      "loss": 6.4206,
      "mean_token_accuracy": 0.12651644125580788,
      "num_tokens": 16730861.0,
      "step": 7725
    },
    {
      "entropy": 6.422515773773194,
      "epoch": 0.8272245706030286,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004941421274883969,
      "loss": 6.2691,
      "mean_token_accuracy": 0.14215103685855865,
      "num_tokens": 16740968.0,
      "step": 7730
    },
    {
      "entropy": 6.47673397064209,
      "epoch": 0.8277596447107924,
      "grad_norm": 1.140625,
      "learning_rate": 0.000494133456883824,
      "loss": 6.3668,
      "mean_token_accuracy": 0.13557462096214296,
      "num_tokens": 16750700.0,
      "step": 7735
    },
    {
      "entropy": 6.496967697143555,
      "epoch": 0.8282947188185563,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004941247799518031,
      "loss": 6.3646,
      "mean_token_accuracy": 0.12662710696458818,
      "num_tokens": 16760719.0,
      "step": 7740
    },
    {
      "entropy": 6.404402160644532,
      "epoch": 0.8288297929263203,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004941160966925851,
      "loss": 6.2455,
      "mean_token_accuracy": 0.1323725014925003,
      "num_tokens": 16771331.0,
      "step": 7745
    },
    {
      "entropy": 6.342697286605835,
      "epoch": 0.8293648670340842,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004941074071064205,
      "loss": 6.3331,
      "mean_token_accuracy": 0.13557431995868682,
      "num_tokens": 16781306.0,
      "step": 7750
    },
    {
      "entropy": 6.5244416236877445,
      "epoch": 0.8298999411418482,
      "grad_norm": 1.125,
      "learning_rate": 0.0004940987111935605,
      "loss": 6.3778,
      "mean_token_accuracy": 0.13223263323307038,
      "num_tokens": 16793013.0,
      "step": 7755
    },
    {
      "entropy": 6.494056749343872,
      "epoch": 0.830435015249612,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004940900089542558,
      "loss": 6.3538,
      "mean_token_accuracy": 0.13380676209926606,
      "num_tokens": 16803987.0,
      "step": 7760
    },
    {
      "entropy": 6.42992525100708,
      "epoch": 0.830970089357376,
      "grad_norm": 1.125,
      "learning_rate": 0.000494081300388758,
      "loss": 6.3676,
      "mean_token_accuracy": 0.12486355975270272,
      "num_tokens": 16815031.0,
      "step": 7765
    },
    {
      "entropy": 6.569543313980103,
      "epoch": 0.8315051634651399,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004940725854973184,
      "loss": 6.4133,
      "mean_token_accuracy": 0.12749240547418594,
      "num_tokens": 16826313.0,
      "step": 7770
    },
    {
      "entropy": 6.568631839752197,
      "epoch": 0.8320402375729038,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004940638642801886,
      "loss": 6.4028,
      "mean_token_accuracy": 0.12546593844890594,
      "num_tokens": 16838470.0,
      "step": 7775
    },
    {
      "entropy": 6.422917079925537,
      "epoch": 0.8325753116806678,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004940551367376205,
      "loss": 6.2943,
      "mean_token_accuracy": 0.13652390614151955,
      "num_tokens": 16849266.0,
      "step": 7780
    },
    {
      "entropy": 6.4542001247406,
      "epoch": 0.8331103857884317,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004940464028698662,
      "loss": 6.4218,
      "mean_token_accuracy": 0.1295318268239498,
      "num_tokens": 16860720.0,
      "step": 7785
    },
    {
      "entropy": 6.499501037597656,
      "epoch": 0.8336454598961957,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004940376626771779,
      "loss": 6.3853,
      "mean_token_accuracy": 0.13086575269699097,
      "num_tokens": 16872004.0,
      "step": 7790
    },
    {
      "entropy": 6.547130680084228,
      "epoch": 0.8341805340039595,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004940289161598076,
      "loss": 6.4118,
      "mean_token_accuracy": 0.12446997463703155,
      "num_tokens": 16882898.0,
      "step": 7795
    },
    {
      "entropy": 6.506767797470093,
      "epoch": 0.8347156081117235,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004940201633180084,
      "loss": 6.3353,
      "mean_token_accuracy": 0.12109738364815711,
      "num_tokens": 16894331.0,
      "step": 7800
    },
    {
      "entropy": 6.570382738113404,
      "epoch": 0.8352506822194874,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004940114041520326,
      "loss": 6.4212,
      "mean_token_accuracy": 0.13175138607621192,
      "num_tokens": 16903656.0,
      "step": 7805
    },
    {
      "entropy": 6.450027179718018,
      "epoch": 0.8357857563272513,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004940026386621333,
      "loss": 6.3951,
      "mean_token_accuracy": 0.13099692463874818,
      "num_tokens": 16915878.0,
      "step": 7810
    },
    {
      "entropy": 6.57182354927063,
      "epoch": 0.8363208304350153,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004939938668485636,
      "loss": 6.4308,
      "mean_token_accuracy": 0.1262585252523422,
      "num_tokens": 16926577.0,
      "step": 7815
    },
    {
      "entropy": 6.441868734359741,
      "epoch": 0.8368559045427791,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004939850887115768,
      "loss": 6.2887,
      "mean_token_accuracy": 0.13820819780230523,
      "num_tokens": 16938012.0,
      "step": 7820
    },
    {
      "entropy": 6.467156171798706,
      "epoch": 0.8373909786505431,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004939763042514263,
      "loss": 6.2928,
      "mean_token_accuracy": 0.12899075001478194,
      "num_tokens": 16949007.0,
      "step": 7825
    },
    {
      "entropy": 6.345366525650024,
      "epoch": 0.837926052758307,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004939675134683658,
      "loss": 6.2732,
      "mean_token_accuracy": 0.13780242428183556,
      "num_tokens": 16959425.0,
      "step": 7830
    },
    {
      "entropy": 6.519599437713623,
      "epoch": 0.838461126866071,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004939587163626491,
      "loss": 6.378,
      "mean_token_accuracy": 0.1242096595466137,
      "num_tokens": 16970002.0,
      "step": 7835
    },
    {
      "entropy": 6.493680238723755,
      "epoch": 0.8389962009738349,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004939499129345302,
      "loss": 6.4116,
      "mean_token_accuracy": 0.1285258114337921,
      "num_tokens": 16980739.0,
      "step": 7840
    },
    {
      "entropy": 6.495212459564209,
      "epoch": 0.8395312750815987,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004939411031842633,
      "loss": 6.3575,
      "mean_token_accuracy": 0.13084009438753127,
      "num_tokens": 16991643.0,
      "step": 7845
    },
    {
      "entropy": 6.362844181060791,
      "epoch": 0.8400663491893627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000493932287112103,
      "loss": 6.2058,
      "mean_token_accuracy": 0.14883958995342256,
      "num_tokens": 17001975.0,
      "step": 7850
    },
    {
      "entropy": 6.506348896026611,
      "epoch": 0.8406014232971266,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004939234647183035,
      "loss": 6.361,
      "mean_token_accuracy": 0.12864112481474876,
      "num_tokens": 17012658.0,
      "step": 7855
    },
    {
      "entropy": 6.485020494461059,
      "epoch": 0.8411364974048906,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004939146360031197,
      "loss": 6.3807,
      "mean_token_accuracy": 0.1270829275250435,
      "num_tokens": 17023196.0,
      "step": 7860
    },
    {
      "entropy": 6.537542486190796,
      "epoch": 0.8416715715126545,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004939058009668067,
      "loss": 6.4101,
      "mean_token_accuracy": 0.1271943613886833,
      "num_tokens": 17034589.0,
      "step": 7865
    },
    {
      "entropy": 6.557751131057739,
      "epoch": 0.8422066456204185,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004938969596096194,
      "loss": 6.3624,
      "mean_token_accuracy": 0.13300060480833054,
      "num_tokens": 17046139.0,
      "step": 7870
    },
    {
      "entropy": 6.440458726882935,
      "epoch": 0.8427417197281823,
      "grad_norm": 1.046875,
      "learning_rate": 0.000493888111931813,
      "loss": 6.3777,
      "mean_token_accuracy": 0.12570422664284706,
      "num_tokens": 17056926.0,
      "step": 7875
    },
    {
      "entropy": 6.362991523742676,
      "epoch": 0.8432767938359462,
      "grad_norm": 1.125,
      "learning_rate": 0.0004938792579336433,
      "loss": 6.3286,
      "mean_token_accuracy": 0.13219914138317107,
      "num_tokens": 17067339.0,
      "step": 7880
    },
    {
      "entropy": 6.541076803207398,
      "epoch": 0.8438118679437102,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004938703976153657,
      "loss": 6.3343,
      "mean_token_accuracy": 0.13340948671102523,
      "num_tokens": 17078037.0,
      "step": 7885
    },
    {
      "entropy": 6.536648845672607,
      "epoch": 0.8443469420514741,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004938615309772362,
      "loss": 6.3719,
      "mean_token_accuracy": 0.12455196157097817,
      "num_tokens": 17089069.0,
      "step": 7890
    },
    {
      "entropy": 6.475170755386353,
      "epoch": 0.8448820161592381,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004938526580195107,
      "loss": 6.3517,
      "mean_token_accuracy": 0.13798089027404786,
      "num_tokens": 17100006.0,
      "step": 7895
    },
    {
      "entropy": 6.496241331100464,
      "epoch": 0.845417090267002,
      "grad_norm": 1.125,
      "learning_rate": 0.0004938437787424454,
      "loss": 6.3776,
      "mean_token_accuracy": 0.13163897916674613,
      "num_tokens": 17111896.0,
      "step": 7900
    },
    {
      "entropy": 6.465775442123413,
      "epoch": 0.845952164374766,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004938348931462969,
      "loss": 6.3721,
      "mean_token_accuracy": 0.13953188359737395,
      "num_tokens": 17123202.0,
      "step": 7905
    },
    {
      "entropy": 6.478828382492066,
      "epoch": 0.8464872384825298,
      "grad_norm": 1.375,
      "learning_rate": 0.0004938260012313215,
      "loss": 6.3348,
      "mean_token_accuracy": 0.1360417015850544,
      "num_tokens": 17134013.0,
      "step": 7910
    },
    {
      "entropy": 6.391561985015869,
      "epoch": 0.8470223125902937,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004938171029977761,
      "loss": 6.337,
      "mean_token_accuracy": 0.13235071823000907,
      "num_tokens": 17144041.0,
      "step": 7915
    },
    {
      "entropy": 6.445134162902832,
      "epoch": 0.8475573866980577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004938081984459176,
      "loss": 6.3361,
      "mean_token_accuracy": 0.13340756967663764,
      "num_tokens": 17155334.0,
      "step": 7920
    },
    {
      "entropy": 6.487791061401367,
      "epoch": 0.8480924608058216,
      "grad_norm": 1.078125,
      "learning_rate": 0.000493799287576003,
      "loss": 6.2834,
      "mean_token_accuracy": 0.13395375832915307,
      "num_tokens": 17166243.0,
      "step": 7925
    },
    {
      "entropy": 6.45171971321106,
      "epoch": 0.8486275349135856,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004937903703882898,
      "loss": 6.348,
      "mean_token_accuracy": 0.13165194243192674,
      "num_tokens": 17177709.0,
      "step": 7930
    },
    {
      "entropy": 6.438591194152832,
      "epoch": 0.8491626090213494,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004937814468830353,
      "loss": 6.3169,
      "mean_token_accuracy": 0.13306962847709655,
      "num_tokens": 17188275.0,
      "step": 7935
    },
    {
      "entropy": 6.510176277160644,
      "epoch": 0.8496976831291134,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004937725170604975,
      "loss": 6.4401,
      "mean_token_accuracy": 0.13184208944439887,
      "num_tokens": 17199527.0,
      "step": 7940
    },
    {
      "entropy": 6.581670665740967,
      "epoch": 0.8502327572368773,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004937635809209339,
      "loss": 6.3622,
      "mean_token_accuracy": 0.12923471257090569,
      "num_tokens": 17209500.0,
      "step": 7945
    },
    {
      "entropy": 6.476590251922607,
      "epoch": 0.8507678313446412,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004937546384646027,
      "loss": 6.3079,
      "mean_token_accuracy": 0.1286625549197197,
      "num_tokens": 17219589.0,
      "step": 7950
    },
    {
      "entropy": 6.489587450027466,
      "epoch": 0.8513029054524052,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004937456896917619,
      "loss": 6.3425,
      "mean_token_accuracy": 0.13386864438652993,
      "num_tokens": 17230508.0,
      "step": 7955
    },
    {
      "entropy": 6.410209989547729,
      "epoch": 0.851837979560169,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004937367346026702,
      "loss": 6.3672,
      "mean_token_accuracy": 0.13320907056331635,
      "num_tokens": 17240886.0,
      "step": 7960
    },
    {
      "entropy": 6.483451175689697,
      "epoch": 0.852373053667933,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004937277731975858,
      "loss": 6.3814,
      "mean_token_accuracy": 0.13769503235816954,
      "num_tokens": 17252247.0,
      "step": 7965
    },
    {
      "entropy": 6.491815853118896,
      "epoch": 0.8529081277756969,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004937188054767679,
      "loss": 6.3294,
      "mean_token_accuracy": 0.13983974382281303,
      "num_tokens": 17262845.0,
      "step": 7970
    },
    {
      "entropy": 6.481371831893921,
      "epoch": 0.8534432018834609,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004937098314404751,
      "loss": 6.3496,
      "mean_token_accuracy": 0.13282273933291436,
      "num_tokens": 17274247.0,
      "step": 7975
    },
    {
      "entropy": 6.498018503189087,
      "epoch": 0.8539782759912248,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004937008510889668,
      "loss": 6.3695,
      "mean_token_accuracy": 0.13182224780321122,
      "num_tokens": 17285471.0,
      "step": 7980
    },
    {
      "entropy": 6.4872630596160885,
      "epoch": 0.8545133500989887,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004936918644225021,
      "loss": 6.3825,
      "mean_token_accuracy": 0.13239139690995216,
      "num_tokens": 17296243.0,
      "step": 7985
    },
    {
      "entropy": 6.444990968704223,
      "epoch": 0.8550484242067526,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004936828714413405,
      "loss": 6.3578,
      "mean_token_accuracy": 0.1335774205625057,
      "num_tokens": 17307301.0,
      "step": 7990
    },
    {
      "entropy": 6.417161989212036,
      "epoch": 0.8555834983145165,
      "grad_norm": 1.375,
      "learning_rate": 0.0004936738721457417,
      "loss": 6.2729,
      "mean_token_accuracy": 0.13876613676548005,
      "num_tokens": 17318550.0,
      "step": 7995
    },
    {
      "entropy": 6.515401363372803,
      "epoch": 0.8561185724222805,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004936648665359657,
      "loss": 6.3532,
      "mean_token_accuracy": 0.13077180162072183,
      "num_tokens": 17328577.0,
      "step": 8000
    },
    {
      "entropy": 6.489120054244995,
      "epoch": 0.8566536465300444,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004936558546122723,
      "loss": 6.3889,
      "mean_token_accuracy": 0.12584815546870232,
      "num_tokens": 17341013.0,
      "step": 8005
    },
    {
      "entropy": 6.4292481422424315,
      "epoch": 0.8571887206378084,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004936468363749217,
      "loss": 6.3538,
      "mean_token_accuracy": 0.12572975009679793,
      "num_tokens": 17351956.0,
      "step": 8010
    },
    {
      "entropy": 6.495111608505249,
      "epoch": 0.8577237947455723,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004936378118241746,
      "loss": 6.3288,
      "mean_token_accuracy": 0.1259569115936756,
      "num_tokens": 17363465.0,
      "step": 8015
    },
    {
      "entropy": 6.463568639755249,
      "epoch": 0.8582588688533361,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004936287809602913,
      "loss": 6.3725,
      "mean_token_accuracy": 0.13539832085371017,
      "num_tokens": 17374458.0,
      "step": 8020
    },
    {
      "entropy": 6.45210018157959,
      "epoch": 0.8587939429611001,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004936197437835329,
      "loss": 6.3936,
      "mean_token_accuracy": 0.12702601328492164,
      "num_tokens": 17385492.0,
      "step": 8025
    },
    {
      "entropy": 6.53994779586792,
      "epoch": 0.859329017068864,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004936107002941599,
      "loss": 6.3409,
      "mean_token_accuracy": 0.13330572098493576,
      "num_tokens": 17395640.0,
      "step": 8030
    },
    {
      "entropy": 6.469773578643799,
      "epoch": 0.859864091176628,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004936016504924337,
      "loss": 6.354,
      "mean_token_accuracy": 0.13245750740170478,
      "num_tokens": 17405211.0,
      "step": 8035
    },
    {
      "entropy": 6.434872341156006,
      "epoch": 0.8603991652843919,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004935925943786155,
      "loss": 6.3174,
      "mean_token_accuracy": 0.13540192246437072,
      "num_tokens": 17415421.0,
      "step": 8040
    },
    {
      "entropy": 6.464983558654785,
      "epoch": 0.8609342393921559,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004935835319529669,
      "loss": 6.3758,
      "mean_token_accuracy": 0.13090842664241792,
      "num_tokens": 17426040.0,
      "step": 8045
    },
    {
      "entropy": 6.4260272026062015,
      "epoch": 0.8614693134999197,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004935744632157496,
      "loss": 6.3577,
      "mean_token_accuracy": 0.13038558065891265,
      "num_tokens": 17436305.0,
      "step": 8050
    },
    {
      "entropy": 6.511065244674683,
      "epoch": 0.8620043876076836,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004935653881672253,
      "loss": 6.418,
      "mean_token_accuracy": 0.1279979854822159,
      "num_tokens": 17446786.0,
      "step": 8055
    },
    {
      "entropy": 6.497061395645142,
      "epoch": 0.8625394617154476,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004935563068076561,
      "loss": 6.2898,
      "mean_token_accuracy": 0.1439591020345688,
      "num_tokens": 17456613.0,
      "step": 8060
    },
    {
      "entropy": 6.363898181915284,
      "epoch": 0.8630745358232115,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004935472191373044,
      "loss": 6.2623,
      "mean_token_accuracy": 0.13567324951291085,
      "num_tokens": 17467281.0,
      "step": 8065
    },
    {
      "entropy": 6.434315633773804,
      "epoch": 0.8636096099309755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004935381251564324,
      "loss": 6.3585,
      "mean_token_accuracy": 0.12976556569337844,
      "num_tokens": 17478271.0,
      "step": 8070
    },
    {
      "entropy": 6.496790647506714,
      "epoch": 0.8641446840387393,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004935290248653027,
      "loss": 6.3514,
      "mean_token_accuracy": 0.131693996489048,
      "num_tokens": 17489519.0,
      "step": 8075
    },
    {
      "entropy": 6.497652578353882,
      "epoch": 0.8646797581465033,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004935199182641781,
      "loss": 6.3359,
      "mean_token_accuracy": 0.12923235818743706,
      "num_tokens": 17500076.0,
      "step": 8080
    },
    {
      "entropy": 6.498486042022705,
      "epoch": 0.8652148322542672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004935108053533216,
      "loss": 6.3359,
      "mean_token_accuracy": 0.13047051727771758,
      "num_tokens": 17509898.0,
      "step": 8085
    },
    {
      "entropy": 6.405987119674682,
      "epoch": 0.8657499063620312,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004935016861329964,
      "loss": 6.2848,
      "mean_token_accuracy": 0.1348872035741806,
      "num_tokens": 17519890.0,
      "step": 8090
    },
    {
      "entropy": 6.416878890991211,
      "epoch": 0.8662849804697951,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004934925606034656,
      "loss": 6.3079,
      "mean_token_accuracy": 0.1306733675301075,
      "num_tokens": 17530927.0,
      "step": 8095
    },
    {
      "entropy": 6.545966386795044,
      "epoch": 0.866820054577559,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004934834287649928,
      "loss": 6.3951,
      "mean_token_accuracy": 0.13003761693835258,
      "num_tokens": 17541174.0,
      "step": 8100
    },
    {
      "entropy": 6.510114955902099,
      "epoch": 0.8673551286853229,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004934742906178417,
      "loss": 6.3669,
      "mean_token_accuracy": 0.128059920668602,
      "num_tokens": 17552460.0,
      "step": 8105
    },
    {
      "entropy": 6.468778133392334,
      "epoch": 0.8678902027930868,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004934651461622762,
      "loss": 6.3766,
      "mean_token_accuracy": 0.13208626359701156,
      "num_tokens": 17563908.0,
      "step": 8110
    },
    {
      "entropy": 6.46075701713562,
      "epoch": 0.8684252769008508,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004934559953985603,
      "loss": 6.3836,
      "mean_token_accuracy": 0.13392364606261253,
      "num_tokens": 17574314.0,
      "step": 8115
    },
    {
      "entropy": 6.500995635986328,
      "epoch": 0.8689603510086147,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004934468383269582,
      "loss": 6.3306,
      "mean_token_accuracy": 0.1337294913828373,
      "num_tokens": 17583719.0,
      "step": 8120
    },
    {
      "entropy": 6.3925378799438475,
      "epoch": 0.8694954251163787,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004934376749477344,
      "loss": 6.3166,
      "mean_token_accuracy": 0.13932424336671828,
      "num_tokens": 17594522.0,
      "step": 8125
    },
    {
      "entropy": 6.4486431121826175,
      "epoch": 0.8700304992241426,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004934285052611533,
      "loss": 6.2773,
      "mean_token_accuracy": 0.13956039696931838,
      "num_tokens": 17604632.0,
      "step": 8130
    },
    {
      "entropy": 6.57233772277832,
      "epoch": 0.8705655733319064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00049341932926748,
      "loss": 6.4078,
      "mean_token_accuracy": 0.1298700749874115,
      "num_tokens": 17615459.0,
      "step": 8135
    },
    {
      "entropy": 6.499594449996948,
      "epoch": 0.8711006474396704,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004934101469669791,
      "loss": 6.3548,
      "mean_token_accuracy": 0.13289933204650878,
      "num_tokens": 17625800.0,
      "step": 8140
    },
    {
      "entropy": 6.357038974761963,
      "epoch": 0.8716357215474343,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004934009583599159,
      "loss": 6.1715,
      "mean_token_accuracy": 0.14703518375754357,
      "num_tokens": 17637264.0,
      "step": 8145
    },
    {
      "entropy": 6.4010762691497805,
      "epoch": 0.8721707956551983,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004933917634465558,
      "loss": 6.2419,
      "mean_token_accuracy": 0.13697617501020432,
      "num_tokens": 17648256.0,
      "step": 8150
    },
    {
      "entropy": 6.368176364898682,
      "epoch": 0.8727058697629622,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004933825622271642,
      "loss": 6.3702,
      "mean_token_accuracy": 0.12936635091900825,
      "num_tokens": 17659096.0,
      "step": 8155
    },
    {
      "entropy": 6.525013446807861,
      "epoch": 0.8732409438707261,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004933733547020069,
      "loss": 6.3434,
      "mean_token_accuracy": 0.13006690070033072,
      "num_tokens": 17669808.0,
      "step": 8160
    },
    {
      "entropy": 6.4963397026062015,
      "epoch": 0.87377601797849,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004933641408713495,
      "loss": 6.3071,
      "mean_token_accuracy": 0.13472993671894073,
      "num_tokens": 17680375.0,
      "step": 8165
    },
    {
      "entropy": 6.357452487945556,
      "epoch": 0.8743110920862539,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004933549207354582,
      "loss": 6.2572,
      "mean_token_accuracy": 0.13582748398184777,
      "num_tokens": 17690780.0,
      "step": 8170
    },
    {
      "entropy": 6.4800115585327145,
      "epoch": 0.8748461661940179,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004933456942945994,
      "loss": 6.3806,
      "mean_token_accuracy": 0.13594808131456376,
      "num_tokens": 17702170.0,
      "step": 8175
    },
    {
      "entropy": 6.49486141204834,
      "epoch": 0.8753812403017818,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004933364615490393,
      "loss": 6.3182,
      "mean_token_accuracy": 0.1333539716899395,
      "num_tokens": 17712631.0,
      "step": 8180
    },
    {
      "entropy": 6.545153522491455,
      "epoch": 0.8759163144095458,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004933272224990445,
      "loss": 6.4073,
      "mean_token_accuracy": 0.12732836753129959,
      "num_tokens": 17723692.0,
      "step": 8185
    },
    {
      "entropy": 6.452640056610107,
      "epoch": 0.8764513885173096,
      "grad_norm": 1.25,
      "learning_rate": 0.0004933179771448819,
      "loss": 6.2961,
      "mean_token_accuracy": 0.14339506179094313,
      "num_tokens": 17735264.0,
      "step": 8190
    },
    {
      "entropy": 6.385478401184082,
      "epoch": 0.8769864626250736,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004933087254868183,
      "loss": 6.2144,
      "mean_token_accuracy": 0.14313811659812928,
      "num_tokens": 17747060.0,
      "step": 8195
    },
    {
      "entropy": 6.389611864089966,
      "epoch": 0.8775215367328375,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000493299467525121,
      "loss": 6.333,
      "mean_token_accuracy": 0.12955316826701163,
      "num_tokens": 17758159.0,
      "step": 8200
    },
    {
      "entropy": 6.47101902961731,
      "epoch": 0.8780566108406014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004932902032600573,
      "loss": 6.3413,
      "mean_token_accuracy": 0.13438796624541283,
      "num_tokens": 17768444.0,
      "step": 8205
    },
    {
      "entropy": 6.418268489837646,
      "epoch": 0.8785916849483654,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004932809326918945,
      "loss": 6.2262,
      "mean_token_accuracy": 0.1454801969230175,
      "num_tokens": 17779303.0,
      "step": 8210
    },
    {
      "entropy": 6.4551633358001705,
      "epoch": 0.8791267590561292,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004932716558209005,
      "loss": 6.4096,
      "mean_token_accuracy": 0.1273011937737465,
      "num_tokens": 17789853.0,
      "step": 8215
    },
    {
      "entropy": 6.549696636199951,
      "epoch": 0.8796618331638932,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004932623726473432,
      "loss": 6.2755,
      "mean_token_accuracy": 0.1448992371559143,
      "num_tokens": 17799232.0,
      "step": 8220
    },
    {
      "entropy": 6.378572559356689,
      "epoch": 0.8801969072716571,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004932530831714904,
      "loss": 6.2809,
      "mean_token_accuracy": 0.13865395933389663,
      "num_tokens": 17808870.0,
      "step": 8225
    },
    {
      "entropy": 6.366944742202759,
      "epoch": 0.8807319813794211,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004932437873936107,
      "loss": 6.3503,
      "mean_token_accuracy": 0.12971489354968072,
      "num_tokens": 17819394.0,
      "step": 8230
    },
    {
      "entropy": 6.53240475654602,
      "epoch": 0.881267055487185,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004932344853139721,
      "loss": 6.3558,
      "mean_token_accuracy": 0.13249184787273408,
      "num_tokens": 17830220.0,
      "step": 8235
    },
    {
      "entropy": 6.455821895599366,
      "epoch": 0.8818021295949489,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004932251769328436,
      "loss": 6.3021,
      "mean_token_accuracy": 0.12674521878361703,
      "num_tokens": 17841971.0,
      "step": 8240
    },
    {
      "entropy": 6.460284090042114,
      "epoch": 0.8823372037027128,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004932158622504937,
      "loss": 6.2861,
      "mean_token_accuracy": 0.1391370452940464,
      "num_tokens": 17851890.0,
      "step": 8245
    },
    {
      "entropy": 6.392160558700562,
      "epoch": 0.8828722778104767,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004932065412671915,
      "loss": 6.3022,
      "mean_token_accuracy": 0.13761991634964943,
      "num_tokens": 17862855.0,
      "step": 8250
    },
    {
      "entropy": 6.52732515335083,
      "epoch": 0.8834073519182407,
      "grad_norm": 1.25,
      "learning_rate": 0.0004931972139832058,
      "loss": 6.3043,
      "mean_token_accuracy": 0.12973644435405732,
      "num_tokens": 17873550.0,
      "step": 8255
    },
    {
      "entropy": 6.460469436645508,
      "epoch": 0.8839424260260046,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004931878803988066,
      "loss": 6.3215,
      "mean_token_accuracy": 0.1367967113852501,
      "num_tokens": 17884067.0,
      "step": 8260
    },
    {
      "entropy": 6.383226919174194,
      "epoch": 0.8844775001337686,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004931785405142627,
      "loss": 6.3328,
      "mean_token_accuracy": 0.13421533703804017,
      "num_tokens": 17894587.0,
      "step": 8265
    },
    {
      "entropy": 6.4551536560058596,
      "epoch": 0.8850125742415325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004931691943298443,
      "loss": 6.3289,
      "mean_token_accuracy": 0.13369324803352356,
      "num_tokens": 17905297.0,
      "step": 8270
    },
    {
      "entropy": 6.373410844802857,
      "epoch": 0.8855476483492963,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000493159841845821,
      "loss": 6.2429,
      "mean_token_accuracy": 0.13628267273306846,
      "num_tokens": 17916225.0,
      "step": 8275
    },
    {
      "entropy": 6.5151769638061525,
      "epoch": 0.8860827224570603,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004931504830624629,
      "loss": 6.2933,
      "mean_token_accuracy": 0.13161173984408378,
      "num_tokens": 17925795.0,
      "step": 8280
    },
    {
      "entropy": 6.391347217559814,
      "epoch": 0.8866177965648242,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004931411179800403,
      "loss": 6.3019,
      "mean_token_accuracy": 0.1317742332816124,
      "num_tokens": 17936134.0,
      "step": 8285
    },
    {
      "entropy": 6.410823774337769,
      "epoch": 0.8871528706725882,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004931317465988235,
      "loss": 6.3772,
      "mean_token_accuracy": 0.12903313040733339,
      "num_tokens": 17946393.0,
      "step": 8290
    },
    {
      "entropy": 6.466726589202881,
      "epoch": 0.8876879447803521,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004931223689190832,
      "loss": 6.3845,
      "mean_token_accuracy": 0.1275428466498852,
      "num_tokens": 17957527.0,
      "step": 8295
    },
    {
      "entropy": 6.513217782974243,
      "epoch": 0.8882230188881161,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004931129849410903,
      "loss": 6.3626,
      "mean_token_accuracy": 0.13466710597276688,
      "num_tokens": 17968390.0,
      "step": 8300
    },
    {
      "entropy": 6.514923715591431,
      "epoch": 0.8887580929958799,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004931035946651154,
      "loss": 6.2961,
      "mean_token_accuracy": 0.14052446484565734,
      "num_tokens": 17978667.0,
      "step": 8305
    },
    {
      "entropy": 6.452924728393555,
      "epoch": 0.8892931671036438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004930941980914299,
      "loss": 6.3666,
      "mean_token_accuracy": 0.12860609516501426,
      "num_tokens": 17989008.0,
      "step": 8310
    },
    {
      "entropy": 6.400901556015015,
      "epoch": 0.8898282412114078,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004930847952203052,
      "loss": 6.3026,
      "mean_token_accuracy": 0.12873381301760672,
      "num_tokens": 18000199.0,
      "step": 8315
    },
    {
      "entropy": 6.479356575012207,
      "epoch": 0.8903633153191717,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004930753860520126,
      "loss": 6.2453,
      "mean_token_accuracy": 0.13265913128852844,
      "num_tokens": 18010941.0,
      "step": 8320
    },
    {
      "entropy": 6.503233289718628,
      "epoch": 0.8908983894269357,
      "grad_norm": 1.375,
      "learning_rate": 0.0004930659705868237,
      "loss": 6.3551,
      "mean_token_accuracy": 0.13169527053833008,
      "num_tokens": 18021519.0,
      "step": 8325
    },
    {
      "entropy": 6.426452779769898,
      "epoch": 0.8914334635346995,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004930565488250108,
      "loss": 6.3055,
      "mean_token_accuracy": 0.13103703036904335,
      "num_tokens": 18032773.0,
      "step": 8330
    },
    {
      "entropy": 6.379181051254273,
      "epoch": 0.8919685376424635,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004930471207668456,
      "loss": 6.2673,
      "mean_token_accuracy": 0.13982483744621277,
      "num_tokens": 18043250.0,
      "step": 8335
    },
    {
      "entropy": 6.41217007637024,
      "epoch": 0.8925036117502274,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004930376864126005,
      "loss": 6.2887,
      "mean_token_accuracy": 0.13771607205271721,
      "num_tokens": 18055165.0,
      "step": 8340
    },
    {
      "entropy": 6.409035301208496,
      "epoch": 0.8930386858579913,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004930282457625476,
      "loss": 6.2407,
      "mean_token_accuracy": 0.14460903480648996,
      "num_tokens": 18065296.0,
      "step": 8345
    },
    {
      "entropy": 6.39350004196167,
      "epoch": 0.8935737599657553,
      "grad_norm": 1.125,
      "learning_rate": 0.00049301879881696,
      "loss": 6.2569,
      "mean_token_accuracy": 0.1321134440600872,
      "num_tokens": 18075536.0,
      "step": 8350
    },
    {
      "entropy": 6.4311154842376705,
      "epoch": 0.8941088340735192,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004930093455761101,
      "loss": 6.3602,
      "mean_token_accuracy": 0.1396454744040966,
      "num_tokens": 18086253.0,
      "step": 8355
    },
    {
      "entropy": 6.449955940246582,
      "epoch": 0.8946439081812831,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000492999886040271,
      "loss": 6.3554,
      "mean_token_accuracy": 0.13434803187847139,
      "num_tokens": 18097335.0,
      "step": 8360
    },
    {
      "entropy": 6.441750335693359,
      "epoch": 0.895178982289047,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004929904202097159,
      "loss": 6.2801,
      "mean_token_accuracy": 0.13604716956615448,
      "num_tokens": 18107293.0,
      "step": 8365
    },
    {
      "entropy": 6.429928255081177,
      "epoch": 0.895714056396811,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004929809480847179,
      "loss": 6.4181,
      "mean_token_accuracy": 0.12502148151397705,
      "num_tokens": 18117705.0,
      "step": 8370
    },
    {
      "entropy": 6.531365060806275,
      "epoch": 0.8962491305045749,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004929714696655507,
      "loss": 6.3031,
      "mean_token_accuracy": 0.1303057961165905,
      "num_tokens": 18129003.0,
      "step": 8375
    },
    {
      "entropy": 6.434348201751709,
      "epoch": 0.8967842046123388,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000492961984952488,
      "loss": 6.2578,
      "mean_token_accuracy": 0.14130010455846786,
      "num_tokens": 18140843.0,
      "step": 8380
    },
    {
      "entropy": 6.376867628097534,
      "epoch": 0.8973192787201028,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004929524939458037,
      "loss": 6.344,
      "mean_token_accuracy": 0.1332057908177376,
      "num_tokens": 18151476.0,
      "step": 8385
    },
    {
      "entropy": 6.442126178741455,
      "epoch": 0.8978543528278666,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004929429966457716,
      "loss": 6.374,
      "mean_token_accuracy": 0.12659687623381616,
      "num_tokens": 18161304.0,
      "step": 8390
    },
    {
      "entropy": 6.496311855316162,
      "epoch": 0.8983894269356306,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004929334930526662,
      "loss": 6.3053,
      "mean_token_accuracy": 0.13693815767765044,
      "num_tokens": 18173696.0,
      "step": 8395
    },
    {
      "entropy": 6.481581306457519,
      "epoch": 0.8989245010433945,
      "grad_norm": 1.25,
      "learning_rate": 0.0004929239831667617,
      "loss": 6.3315,
      "mean_token_accuracy": 0.12777818590402604,
      "num_tokens": 18186024.0,
      "step": 8400
    },
    {
      "entropy": 6.491195344924927,
      "epoch": 0.8994595751511585,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004929144669883329,
      "loss": 6.356,
      "mean_token_accuracy": 0.14038287550210954,
      "num_tokens": 18196976.0,
      "step": 8405
    },
    {
      "entropy": 6.433043909072876,
      "epoch": 0.8999946492589224,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004929049445176545,
      "loss": 6.3359,
      "mean_token_accuracy": 0.12853538244962692,
      "num_tokens": 18208086.0,
      "step": 8410
    },
    {
      "entropy": 6.412654304504395,
      "epoch": 0.9005297233666862,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004928954157550013,
      "loss": 6.2842,
      "mean_token_accuracy": 0.14256626963615418,
      "num_tokens": 18218468.0,
      "step": 8415
    },
    {
      "entropy": 6.499575519561768,
      "epoch": 0.9010647974744502,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004928858807006488,
      "loss": 6.3986,
      "mean_token_accuracy": 0.1332329712808132,
      "num_tokens": 18230489.0,
      "step": 8420
    },
    {
      "entropy": 6.47307276725769,
      "epoch": 0.9015998715822141,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004928763393548721,
      "loss": 6.3311,
      "mean_token_accuracy": 0.13532978817820548,
      "num_tokens": 18239968.0,
      "step": 8425
    },
    {
      "entropy": 6.438126707077027,
      "epoch": 0.9021349456899781,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004928667917179465,
      "loss": 6.3648,
      "mean_token_accuracy": 0.12859018445014953,
      "num_tokens": 18250902.0,
      "step": 8430
    },
    {
      "entropy": 6.492179822921753,
      "epoch": 0.902670019797742,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004928572377901481,
      "loss": 6.3618,
      "mean_token_accuracy": 0.13571698144078254,
      "num_tokens": 18262094.0,
      "step": 8435
    },
    {
      "entropy": 6.494809627532959,
      "epoch": 0.903205093905506,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004928476775717524,
      "loss": 6.3337,
      "mean_token_accuracy": 0.130261592566967,
      "num_tokens": 18272765.0,
      "step": 8440
    },
    {
      "entropy": 6.471274375915527,
      "epoch": 0.9037401680132698,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004928381110630358,
      "loss": 6.2367,
      "mean_token_accuracy": 0.14439094737172126,
      "num_tokens": 18284570.0,
      "step": 8445
    },
    {
      "entropy": 6.584997320175171,
      "epoch": 0.9042752421210337,
      "grad_norm": 1.125,
      "learning_rate": 0.0004928285382642742,
      "loss": 6.3627,
      "mean_token_accuracy": 0.12972358614206314,
      "num_tokens": 18295170.0,
      "step": 8450
    },
    {
      "entropy": 6.399804878234863,
      "epoch": 0.9048103162287977,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004928189591757443,
      "loss": 6.3022,
      "mean_token_accuracy": 0.1307802267372608,
      "num_tokens": 18306388.0,
      "step": 8455
    },
    {
      "entropy": 6.40751085281372,
      "epoch": 0.9053453903365616,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004928093737977225,
      "loss": 6.3128,
      "mean_token_accuracy": 0.13409198671579362,
      "num_tokens": 18316363.0,
      "step": 8460
    },
    {
      "entropy": 6.37100739479065,
      "epoch": 0.9058804644443256,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004927997821304856,
      "loss": 6.2601,
      "mean_token_accuracy": 0.13586740121245383,
      "num_tokens": 18327397.0,
      "step": 8465
    },
    {
      "entropy": 6.482768392562866,
      "epoch": 0.9064155385520895,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004927901841743107,
      "loss": 6.377,
      "mean_token_accuracy": 0.13369872719049453,
      "num_tokens": 18338909.0,
      "step": 8470
    },
    {
      "entropy": 6.5202696323394775,
      "epoch": 0.9069506126598534,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004927805799294747,
      "loss": 6.3478,
      "mean_token_accuracy": 0.1307516872882843,
      "num_tokens": 18350110.0,
      "step": 8475
    },
    {
      "entropy": 6.503915548324585,
      "epoch": 0.9074856867676173,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004927709693962551,
      "loss": 6.3828,
      "mean_token_accuracy": 0.12589319944381713,
      "num_tokens": 18361590.0,
      "step": 8480
    },
    {
      "entropy": 6.433329391479492,
      "epoch": 0.9080207608753812,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004927613525749295,
      "loss": 6.3316,
      "mean_token_accuracy": 0.13395185098052026,
      "num_tokens": 18372602.0,
      "step": 8485
    },
    {
      "entropy": 6.458200407028198,
      "epoch": 0.9085558349831452,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004927517294657753,
      "loss": 6.3803,
      "mean_token_accuracy": 0.1316978432238102,
      "num_tokens": 18384653.0,
      "step": 8490
    },
    {
      "entropy": 6.572974061965942,
      "epoch": 0.9090909090909091,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004927421000690705,
      "loss": 6.3467,
      "mean_token_accuracy": 0.13045882061123848,
      "num_tokens": 18395379.0,
      "step": 8495
    },
    {
      "entropy": 6.489434146881104,
      "epoch": 0.909625983198673,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004927324643850933,
      "loss": 6.3116,
      "mean_token_accuracy": 0.1332891933619976,
      "num_tokens": 18405915.0,
      "step": 8500
    },
    {
      "entropy": 6.366228199005127,
      "epoch": 0.9101610573064369,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004927228224141217,
      "loss": 6.2074,
      "mean_token_accuracy": 0.15189967304468155,
      "num_tokens": 18416656.0,
      "step": 8505
    },
    {
      "entropy": 6.447817325592041,
      "epoch": 0.9106961314142009,
      "grad_norm": 3.125,
      "learning_rate": 0.0004927131741564342,
      "loss": 6.3302,
      "mean_token_accuracy": 0.1295878879725933,
      "num_tokens": 18427141.0,
      "step": 8510
    },
    {
      "entropy": 6.4294486999511715,
      "epoch": 0.9112312055219648,
      "grad_norm": 1.5,
      "learning_rate": 0.0004927035196123094,
      "loss": 6.3667,
      "mean_token_accuracy": 0.13632506504654884,
      "num_tokens": 18437135.0,
      "step": 8515
    },
    {
      "entropy": 6.607345724105835,
      "epoch": 0.9117662796297287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000492693858782026,
      "loss": 6.4496,
      "mean_token_accuracy": 0.12477857396006584,
      "num_tokens": 18448567.0,
      "step": 8520
    },
    {
      "entropy": 6.584646558761596,
      "epoch": 0.9123013537374927,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000492684191665863,
      "loss": 6.3946,
      "mean_token_accuracy": 0.1252623051404953,
      "num_tokens": 18459385.0,
      "step": 8525
    },
    {
      "entropy": 6.429875802993775,
      "epoch": 0.9128364278452565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004926745182640996,
      "loss": 6.3396,
      "mean_token_accuracy": 0.13633881956338884,
      "num_tokens": 18471376.0,
      "step": 8530
    },
    {
      "entropy": 6.507524585723877,
      "epoch": 0.9133715019530205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004926648385770152,
      "loss": 6.4199,
      "mean_token_accuracy": 0.1259002670645714,
      "num_tokens": 18481950.0,
      "step": 8535
    },
    {
      "entropy": 6.492842006683349,
      "epoch": 0.9139065760607844,
      "grad_norm": 1.25,
      "learning_rate": 0.0004926551526048891,
      "loss": 6.3256,
      "mean_token_accuracy": 0.1331590600311756,
      "num_tokens": 18492436.0,
      "step": 8540
    },
    {
      "entropy": 6.432707691192627,
      "epoch": 0.9144416501685484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004926454603480009,
      "loss": 6.1622,
      "mean_token_accuracy": 0.13788460344076156,
      "num_tokens": 18504279.0,
      "step": 8545
    },
    {
      "entropy": 6.490119886398316,
      "epoch": 0.9149767242763123,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004926357618066307,
      "loss": 6.3103,
      "mean_token_accuracy": 0.13683086410164833,
      "num_tokens": 18515262.0,
      "step": 8550
    },
    {
      "entropy": 6.494428014755249,
      "epoch": 0.9155117983840761,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004926260569810586,
      "loss": 6.3657,
      "mean_token_accuracy": 0.13266137465834618,
      "num_tokens": 18525176.0,
      "step": 8555
    },
    {
      "entropy": 6.368859529495239,
      "epoch": 0.9160468724918401,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004926163458715645,
      "loss": 6.3497,
      "mean_token_accuracy": 0.1297522120177746,
      "num_tokens": 18536674.0,
      "step": 8560
    },
    {
      "entropy": 6.446297883987427,
      "epoch": 0.916581946599604,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004926066284784292,
      "loss": 6.3125,
      "mean_token_accuracy": 0.1322654105722904,
      "num_tokens": 18547398.0,
      "step": 8565
    },
    {
      "entropy": 6.519314384460449,
      "epoch": 0.917117020707368,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004925969048019329,
      "loss": 6.365,
      "mean_token_accuracy": 0.1379559777677059,
      "num_tokens": 18558838.0,
      "step": 8570
    },
    {
      "entropy": 6.458747339248657,
      "epoch": 0.9176520948151319,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004925871748423566,
      "loss": 6.3124,
      "mean_token_accuracy": 0.13683706372976304,
      "num_tokens": 18569881.0,
      "step": 8575
    },
    {
      "entropy": 6.403721332550049,
      "epoch": 0.9181871689228959,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004925774385999812,
      "loss": 6.2213,
      "mean_token_accuracy": 0.13998932242393494,
      "num_tokens": 18579556.0,
      "step": 8580
    },
    {
      "entropy": 6.307366132736206,
      "epoch": 0.9187222430306597,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004925676960750878,
      "loss": 6.2312,
      "mean_token_accuracy": 0.1437248632311821,
      "num_tokens": 18590580.0,
      "step": 8585
    },
    {
      "entropy": 6.5107158660888675,
      "epoch": 0.9192573171384236,
      "grad_norm": 1.125,
      "learning_rate": 0.0004925579472679578,
      "loss": 6.3397,
      "mean_token_accuracy": 0.13435586020350457,
      "num_tokens": 18600813.0,
      "step": 8590
    },
    {
      "entropy": 6.446278715133667,
      "epoch": 0.9197923912461876,
      "grad_norm": 1.125,
      "learning_rate": 0.0004925481921788726,
      "loss": 6.2616,
      "mean_token_accuracy": 0.14329060912132263,
      "num_tokens": 18610807.0,
      "step": 8595
    },
    {
      "entropy": 6.357829523086548,
      "epoch": 0.9203274653539515,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000492538430808114,
      "loss": 6.2373,
      "mean_token_accuracy": 0.1349279150366783,
      "num_tokens": 18622108.0,
      "step": 8600
    },
    {
      "entropy": 6.441023921966552,
      "epoch": 0.9208625394617155,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004925286631559636,
      "loss": 6.4048,
      "mean_token_accuracy": 0.1314627319574356,
      "num_tokens": 18633098.0,
      "step": 8605
    },
    {
      "entropy": 6.523507833480835,
      "epoch": 0.9213976135694794,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004925188892227038,
      "loss": 6.3048,
      "mean_token_accuracy": 0.13459505885839462,
      "num_tokens": 18643603.0,
      "step": 8610
    },
    {
      "entropy": 6.463659858703613,
      "epoch": 0.9219326876772433,
      "grad_norm": 1.125,
      "learning_rate": 0.0004925091090086166,
      "loss": 6.3165,
      "mean_token_accuracy": 0.138278029859066,
      "num_tokens": 18654354.0,
      "step": 8615
    },
    {
      "entropy": 6.460863971710205,
      "epoch": 0.9224677617850072,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004924993225139845,
      "loss": 6.3917,
      "mean_token_accuracy": 0.1363808736205101,
      "num_tokens": 18665349.0,
      "step": 8620
    },
    {
      "entropy": 6.537027740478516,
      "epoch": 0.9230028358927711,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004924895297390899,
      "loss": 6.3296,
      "mean_token_accuracy": 0.1294632613658905,
      "num_tokens": 18676575.0,
      "step": 8625
    },
    {
      "entropy": 6.467256307601929,
      "epoch": 0.9235379100005351,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004924797306842157,
      "loss": 6.274,
      "mean_token_accuracy": 0.1381203480064869,
      "num_tokens": 18686501.0,
      "step": 8630
    },
    {
      "entropy": 6.454415559768677,
      "epoch": 0.924072984108299,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004924699253496449,
      "loss": 6.3355,
      "mean_token_accuracy": 0.13682935982942582,
      "num_tokens": 18697246.0,
      "step": 8635
    },
    {
      "entropy": 6.413412141799927,
      "epoch": 0.924608058216063,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004924601137356605,
      "loss": 6.2729,
      "mean_token_accuracy": 0.14315944835543631,
      "num_tokens": 18708177.0,
      "step": 8640
    },
    {
      "entropy": 6.466000366210937,
      "epoch": 0.9251431323238268,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000492450295842546,
      "loss": 6.3235,
      "mean_token_accuracy": 0.13267922550439834,
      "num_tokens": 18719334.0,
      "step": 8645
    },
    {
      "entropy": 6.475500392913818,
      "epoch": 0.9256782064315908,
      "grad_norm": 2.5,
      "learning_rate": 0.0004924404716705847,
      "loss": 6.3028,
      "mean_token_accuracy": 0.13989831656217575,
      "num_tokens": 18729174.0,
      "step": 8650
    },
    {
      "entropy": 6.443165493011475,
      "epoch": 0.9262132805393547,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004924306412200603,
      "loss": 6.3853,
      "mean_token_accuracy": 0.12822362184524536,
      "num_tokens": 18740277.0,
      "step": 8655
    },
    {
      "entropy": 6.521066522598266,
      "epoch": 0.9267483546471186,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004924208044912567,
      "loss": 6.3434,
      "mean_token_accuracy": 0.13407693356275557,
      "num_tokens": 18751864.0,
      "step": 8660
    },
    {
      "entropy": 6.422997283935547,
      "epoch": 0.9272834287548826,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004924109614844579,
      "loss": 6.255,
      "mean_token_accuracy": 0.14270969852805138,
      "num_tokens": 18762786.0,
      "step": 8665
    },
    {
      "entropy": 6.358929061889649,
      "epoch": 0.9278185028626464,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004924011121999482,
      "loss": 6.3143,
      "mean_token_accuracy": 0.139602729678154,
      "num_tokens": 18772913.0,
      "step": 8670
    },
    {
      "entropy": 6.442178344726562,
      "epoch": 0.9283535769704104,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004923912566380118,
      "loss": 6.2791,
      "mean_token_accuracy": 0.13877590745687485,
      "num_tokens": 18784204.0,
      "step": 8675
    },
    {
      "entropy": 6.3998900890350345,
      "epoch": 0.9288886510781743,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004923813947989336,
      "loss": 6.3794,
      "mean_token_accuracy": 0.1320524051785469,
      "num_tokens": 18795425.0,
      "step": 8680
    },
    {
      "entropy": 6.521181583404541,
      "epoch": 0.9294237251859383,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004923715266829981,
      "loss": 6.3175,
      "mean_token_accuracy": 0.1337410591542721,
      "num_tokens": 18806850.0,
      "step": 8685
    },
    {
      "entropy": 6.427550506591797,
      "epoch": 0.9299587992937022,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004923616522904903,
      "loss": 6.2592,
      "mean_token_accuracy": 0.13981002643704415,
      "num_tokens": 18817272.0,
      "step": 8690
    },
    {
      "entropy": 6.4260178565979,
      "epoch": 0.930493873401466,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004923517716216953,
      "loss": 6.406,
      "mean_token_accuracy": 0.12655333951115608,
      "num_tokens": 18827075.0,
      "step": 8695
    },
    {
      "entropy": 6.500725936889649,
      "epoch": 0.93102894750923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004923418846768985,
      "loss": 6.3168,
      "mean_token_accuracy": 0.1307675339281559,
      "num_tokens": 18837635.0,
      "step": 8700
    },
    {
      "entropy": 6.451502656936645,
      "epoch": 0.9315640216169939,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004923319914563852,
      "loss": 6.3032,
      "mean_token_accuracy": 0.14600803107023239,
      "num_tokens": 18848559.0,
      "step": 8705
    },
    {
      "entropy": 6.494266748428345,
      "epoch": 0.9320990957247579,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004923220919604413,
      "loss": 6.3359,
      "mean_token_accuracy": 0.13247558549046518,
      "num_tokens": 18859327.0,
      "step": 8710
    },
    {
      "entropy": 6.430937147140503,
      "epoch": 0.9326341698325218,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004923121861893525,
      "loss": 6.2876,
      "mean_token_accuracy": 0.13356452882289888,
      "num_tokens": 18869403.0,
      "step": 8715
    },
    {
      "entropy": 6.45348744392395,
      "epoch": 0.9331692439402858,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004923022741434049,
      "loss": 6.3128,
      "mean_token_accuracy": 0.1300775960087776,
      "num_tokens": 18879890.0,
      "step": 8720
    },
    {
      "entropy": 6.454838037490845,
      "epoch": 0.9337043180480497,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004922923558228846,
      "loss": 6.3324,
      "mean_token_accuracy": 0.13287571668624878,
      "num_tokens": 18890361.0,
      "step": 8725
    },
    {
      "entropy": 6.413714027404785,
      "epoch": 0.9342393921558135,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004922824312280781,
      "loss": 6.2495,
      "mean_token_accuracy": 0.13993326127529143,
      "num_tokens": 18902344.0,
      "step": 8730
    },
    {
      "entropy": 6.373650074005127,
      "epoch": 0.9347744662635775,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000492272500359272,
      "loss": 6.2555,
      "mean_token_accuracy": 0.13796113207936286,
      "num_tokens": 18911364.0,
      "step": 8735
    },
    {
      "entropy": 6.418253326416016,
      "epoch": 0.9353095403713414,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004922625632167529,
      "loss": 6.3475,
      "mean_token_accuracy": 0.12562165707349776,
      "num_tokens": 18922254.0,
      "step": 8740
    },
    {
      "entropy": 6.512704133987427,
      "epoch": 0.9358446144791054,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004922526198008078,
      "loss": 6.3209,
      "mean_token_accuracy": 0.1317998453974724,
      "num_tokens": 18932943.0,
      "step": 8745
    },
    {
      "entropy": 6.4999556064605715,
      "epoch": 0.9363796885868693,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004922426701117239,
      "loss": 6.4034,
      "mean_token_accuracy": 0.12868692278861998,
      "num_tokens": 18943523.0,
      "step": 8750
    },
    {
      "entropy": 6.515548467636108,
      "epoch": 0.9369147626946333,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004922327141497884,
      "loss": 6.3705,
      "mean_token_accuracy": 0.12669930309057237,
      "num_tokens": 18955220.0,
      "step": 8755
    },
    {
      "entropy": 6.573355340957642,
      "epoch": 0.9374498368023971,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004922227519152888,
      "loss": 6.3934,
      "mean_token_accuracy": 0.131385987251997,
      "num_tokens": 18966714.0,
      "step": 8760
    },
    {
      "entropy": 6.417609405517578,
      "epoch": 0.937984910910161,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004922127834085128,
      "loss": 6.254,
      "mean_token_accuracy": 0.13833884298801422,
      "num_tokens": 18977998.0,
      "step": 8765
    },
    {
      "entropy": 6.386611223220825,
      "epoch": 0.938519985017925,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004922028086297481,
      "loss": 6.2507,
      "mean_token_accuracy": 0.13773412480950356,
      "num_tokens": 18990210.0,
      "step": 8770
    },
    {
      "entropy": 6.481419277191162,
      "epoch": 0.9390550591256889,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004921928275792828,
      "loss": 6.3594,
      "mean_token_accuracy": 0.13566964715719224,
      "num_tokens": 19000529.0,
      "step": 8775
    },
    {
      "entropy": 6.449409055709839,
      "epoch": 0.9395901332334529,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004921828402574052,
      "loss": 6.2865,
      "mean_token_accuracy": 0.13738374412059784,
      "num_tokens": 19011386.0,
      "step": 8780
    },
    {
      "entropy": 6.4130066394805905,
      "epoch": 0.9401252073412167,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004921728466644037,
      "loss": 6.2836,
      "mean_token_accuracy": 0.1371908076107502,
      "num_tokens": 19022196.0,
      "step": 8785
    },
    {
      "entropy": 6.388099002838135,
      "epoch": 0.9406602814489807,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004921628468005666,
      "loss": 6.3183,
      "mean_token_accuracy": 0.13676922619342805,
      "num_tokens": 19033221.0,
      "step": 8790
    },
    {
      "entropy": 6.4475117206573485,
      "epoch": 0.9411953555567446,
      "grad_norm": 1.171875,
      "learning_rate": 0.000492152840666183,
      "loss": 6.3557,
      "mean_token_accuracy": 0.1318049170076847,
      "num_tokens": 19044953.0,
      "step": 8795
    },
    {
      "entropy": 6.545235538482666,
      "epoch": 0.9417304296645085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004921428282615415,
      "loss": 6.3765,
      "mean_token_accuracy": 0.12960621118545532,
      "num_tokens": 19055512.0,
      "step": 8800
    },
    {
      "entropy": 6.476768255233765,
      "epoch": 0.9422655037722725,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004921328095869313,
      "loss": 6.3907,
      "mean_token_accuracy": 0.1289285570383072,
      "num_tokens": 19065659.0,
      "step": 8805
    },
    {
      "entropy": 6.425279521942139,
      "epoch": 0.9428005778800364,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004921227846426418,
      "loss": 6.291,
      "mean_token_accuracy": 0.14513665959239005,
      "num_tokens": 19076955.0,
      "step": 8810
    },
    {
      "entropy": 6.474784851074219,
      "epoch": 0.9433356519878003,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004921127534289624,
      "loss": 6.3874,
      "mean_token_accuracy": 0.12870006412267684,
      "num_tokens": 19088305.0,
      "step": 8815
    },
    {
      "entropy": 6.502227783203125,
      "epoch": 0.9438707260955642,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004921027159461829,
      "loss": 6.3742,
      "mean_token_accuracy": 0.12985786572098731,
      "num_tokens": 19098046.0,
      "step": 8820
    },
    {
      "entropy": 6.456517028808594,
      "epoch": 0.9444058002033282,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004920926721945927,
      "loss": 6.2888,
      "mean_token_accuracy": 0.13685355335474014,
      "num_tokens": 19109019.0,
      "step": 8825
    },
    {
      "entropy": 6.45530071258545,
      "epoch": 0.9449408743110921,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004920826221744823,
      "loss": 6.3062,
      "mean_token_accuracy": 0.13298919573426246,
      "num_tokens": 19119624.0,
      "step": 8830
    },
    {
      "entropy": 6.4864274024963375,
      "epoch": 0.945475948418856,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004920725658861417,
      "loss": 6.3499,
      "mean_token_accuracy": 0.1268065832555294,
      "num_tokens": 19130577.0,
      "step": 8835
    },
    {
      "entropy": 6.487700176239014,
      "epoch": 0.94601102252662,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004920625033298611,
      "loss": 6.2345,
      "mean_token_accuracy": 0.1382552169263363,
      "num_tokens": 19140254.0,
      "step": 8840
    },
    {
      "entropy": 6.397662210464477,
      "epoch": 0.9465460966343838,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004920524345059314,
      "loss": 6.2823,
      "mean_token_accuracy": 0.13530903160572053,
      "num_tokens": 19151357.0,
      "step": 8845
    },
    {
      "entropy": 6.472157955169678,
      "epoch": 0.9470811707421478,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004920423594146431,
      "loss": 6.2838,
      "mean_token_accuracy": 0.14035611003637313,
      "num_tokens": 19162069.0,
      "step": 8850
    },
    {
      "entropy": 6.375074100494385,
      "epoch": 0.9476162448499117,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004920322780562872,
      "loss": 6.1822,
      "mean_token_accuracy": 0.14188951328396798,
      "num_tokens": 19172515.0,
      "step": 8855
    },
    {
      "entropy": 6.3510942459106445,
      "epoch": 0.9481513189576757,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004920221904311546,
      "loss": 6.1998,
      "mean_token_accuracy": 0.13968603014945985,
      "num_tokens": 19182450.0,
      "step": 8860
    },
    {
      "entropy": 6.487783193588257,
      "epoch": 0.9486863930654396,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004920120965395369,
      "loss": 6.3956,
      "mean_token_accuracy": 0.1287924461066723,
      "num_tokens": 19193833.0,
      "step": 8865
    },
    {
      "entropy": 6.529372501373291,
      "epoch": 0.9492214671732034,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004920019963817253,
      "loss": 6.3463,
      "mean_token_accuracy": 0.1310620926320553,
      "num_tokens": 19204970.0,
      "step": 8870
    },
    {
      "entropy": 6.496806478500366,
      "epoch": 0.9497565412809674,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004919918899580117,
      "loss": 6.285,
      "mean_token_accuracy": 0.13851441517472268,
      "num_tokens": 19215795.0,
      "step": 8875
    },
    {
      "entropy": 6.419819736480713,
      "epoch": 0.9502916153887313,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004919817772686876,
      "loss": 6.3315,
      "mean_token_accuracy": 0.1281391352415085,
      "num_tokens": 19226375.0,
      "step": 8880
    },
    {
      "entropy": 6.486654472351074,
      "epoch": 0.9508266894964953,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004919716583140453,
      "loss": 6.3223,
      "mean_token_accuracy": 0.14177558049559594,
      "num_tokens": 19236066.0,
      "step": 8885
    },
    {
      "entropy": 6.377401447296142,
      "epoch": 0.9513617636042592,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004919615330943769,
      "loss": 6.2005,
      "mean_token_accuracy": 0.14210012480616568,
      "num_tokens": 19246757.0,
      "step": 8890
    },
    {
      "entropy": 6.4676275730133055,
      "epoch": 0.9518968377120232,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004919514016099746,
      "loss": 6.3604,
      "mean_token_accuracy": 0.13128288462758064,
      "num_tokens": 19257875.0,
      "step": 8895
    },
    {
      "entropy": 6.466962718963623,
      "epoch": 0.952431911819787,
      "grad_norm": 1.109375,
      "learning_rate": 0.000491941263861131,
      "loss": 6.3376,
      "mean_token_accuracy": 0.13478436544537545,
      "num_tokens": 19270236.0,
      "step": 8900
    },
    {
      "entropy": 6.418972063064575,
      "epoch": 0.9529669859275509,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000491931119848139,
      "loss": 6.3083,
      "mean_token_accuracy": 0.13070244267582892,
      "num_tokens": 19280877.0,
      "step": 8905
    },
    {
      "entropy": 6.4053953170776365,
      "epoch": 0.9535020600353149,
      "grad_norm": 1.25,
      "learning_rate": 0.0004919209695712915,
      "loss": 6.2248,
      "mean_token_accuracy": 0.13718185797333718,
      "num_tokens": 19292852.0,
      "step": 8910
    },
    {
      "entropy": 6.47879991531372,
      "epoch": 0.9540371341430788,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004919108130308814,
      "loss": 6.4263,
      "mean_token_accuracy": 0.12273015826940536,
      "num_tokens": 19305516.0,
      "step": 8915
    },
    {
      "entropy": 6.471704959869385,
      "epoch": 0.9545722082508428,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004919006502272019,
      "loss": 6.2429,
      "mean_token_accuracy": 0.1446073204278946,
      "num_tokens": 19315436.0,
      "step": 8920
    },
    {
      "entropy": 6.38612208366394,
      "epoch": 0.9551072823586066,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004918904811605468,
      "loss": 6.2859,
      "mean_token_accuracy": 0.136173015832901,
      "num_tokens": 19325393.0,
      "step": 8925
    },
    {
      "entropy": 6.392364454269409,
      "epoch": 0.9556423564663706,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004918803058312094,
      "loss": 6.2776,
      "mean_token_accuracy": 0.13736315220594406,
      "num_tokens": 19336722.0,
      "step": 8930
    },
    {
      "entropy": 6.5551470756530765,
      "epoch": 0.9561774305741345,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004918701242394837,
      "loss": 6.4449,
      "mean_token_accuracy": 0.1244717925786972,
      "num_tokens": 19348406.0,
      "step": 8935
    },
    {
      "entropy": 6.4295430183410645,
      "epoch": 0.9567125046818984,
      "grad_norm": 1.25,
      "learning_rate": 0.0004918599363856637,
      "loss": 6.2469,
      "mean_token_accuracy": 0.14535028114914894,
      "num_tokens": 19359714.0,
      "step": 8940
    },
    {
      "entropy": 6.425970792770386,
      "epoch": 0.9572475787896624,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004918497422700434,
      "loss": 6.2811,
      "mean_token_accuracy": 0.1344941198825836,
      "num_tokens": 19370833.0,
      "step": 8945
    },
    {
      "entropy": 6.4786412715911865,
      "epoch": 0.9577826528974263,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004918395418929174,
      "loss": 6.3818,
      "mean_token_accuracy": 0.12696493193507194,
      "num_tokens": 19382098.0,
      "step": 8950
    },
    {
      "entropy": 6.465514326095581,
      "epoch": 0.9583177270051902,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00049182933525458,
      "loss": 6.351,
      "mean_token_accuracy": 0.12936325147747993,
      "num_tokens": 19393595.0,
      "step": 8955
    },
    {
      "entropy": 6.465545988082885,
      "epoch": 0.9588528011129541,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000491819122355326,
      "loss": 6.2925,
      "mean_token_accuracy": 0.14031073376536368,
      "num_tokens": 19405296.0,
      "step": 8960
    },
    {
      "entropy": 6.464176940917969,
      "epoch": 0.9593878752207181,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004918089031954503,
      "loss": 6.3138,
      "mean_token_accuracy": 0.1319153942167759,
      "num_tokens": 19415693.0,
      "step": 8965
    },
    {
      "entropy": 6.440121650695801,
      "epoch": 0.959922949328482,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000491798677775248,
      "loss": 6.3507,
      "mean_token_accuracy": 0.13368432521820067,
      "num_tokens": 19426140.0,
      "step": 8970
    },
    {
      "entropy": 6.539132404327392,
      "epoch": 0.9604580234362459,
      "grad_norm": 1.125,
      "learning_rate": 0.0004917884460950144,
      "loss": 6.394,
      "mean_token_accuracy": 0.13047945350408555,
      "num_tokens": 19437543.0,
      "step": 8975
    },
    {
      "entropy": 6.447048616409302,
      "epoch": 0.9609930975440099,
      "grad_norm": 1.125,
      "learning_rate": 0.000491778208155045,
      "loss": 6.2929,
      "mean_token_accuracy": 0.13499191179871559,
      "num_tokens": 19447788.0,
      "step": 8980
    },
    {
      "entropy": 6.472111606597901,
      "epoch": 0.9615281716517737,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004917679639556351,
      "loss": 6.3143,
      "mean_token_accuracy": 0.13168717473745345,
      "num_tokens": 19457587.0,
      "step": 8985
    },
    {
      "entropy": 6.444050407409668,
      "epoch": 0.9620632457595377,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004917577134970808,
      "loss": 6.2378,
      "mean_token_accuracy": 0.1373984508216381,
      "num_tokens": 19467422.0,
      "step": 8990
    },
    {
      "entropy": 6.393634366989136,
      "epoch": 0.9625983198673016,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004917474567796778,
      "loss": 6.2883,
      "mean_token_accuracy": 0.13712013587355615,
      "num_tokens": 19479037.0,
      "step": 8995
    },
    {
      "entropy": 6.391675758361816,
      "epoch": 0.9631333939750656,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004917371938037226,
      "loss": 6.2177,
      "mean_token_accuracy": 0.15161346048116683,
      "num_tokens": 19489639.0,
      "step": 9000
    },
    {
      "epoch": 0.9631333939750656,
      "eval_entropy": 6.3108516717535466,
      "eval_loss": 6.369574546813965,
      "eval_mean_token_accuracy": 0.13738052544994858,
      "eval_num_tokens": 19489639.0,
      "eval_runtime": 22.6999,
      "eval_samples_per_second": 1307.45,
      "eval_steps_per_second": 163.437,
      "step": 9000
    },
    {
      "entropy": 6.476394319534302,
      "epoch": 0.9636684680828295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004917269245695114,
      "loss": 6.3574,
      "mean_token_accuracy": 0.1288781724870205,
      "num_tokens": 19500592.0,
      "step": 9005
    },
    {
      "entropy": 6.483854532241821,
      "epoch": 0.9642035421905933,
      "grad_norm": 1.25,
      "learning_rate": 0.0004917166490773406,
      "loss": 6.3546,
      "mean_token_accuracy": 0.1343536950647831,
      "num_tokens": 19510818.0,
      "step": 9010
    },
    {
      "entropy": 6.4669983863830565,
      "epoch": 0.9647386162983573,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004917063673275071,
      "loss": 6.3886,
      "mean_token_accuracy": 0.13162672594189645,
      "num_tokens": 19521283.0,
      "step": 9015
    },
    {
      "entropy": 6.4521952152252195,
      "epoch": 0.9652736904061212,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004916960793203077,
      "loss": 6.2182,
      "mean_token_accuracy": 0.13597461059689522,
      "num_tokens": 19531799.0,
      "step": 9020
    },
    {
      "entropy": 6.424303197860718,
      "epoch": 0.9658087645138852,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004916857850560394,
      "loss": 6.1958,
      "mean_token_accuracy": 0.15168601647019386,
      "num_tokens": 19542377.0,
      "step": 9025
    },
    {
      "entropy": 6.348746967315674,
      "epoch": 0.9663438386216491,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004916754845349996,
      "loss": 6.2151,
      "mean_token_accuracy": 0.1400285094976425,
      "num_tokens": 19552654.0,
      "step": 9030
    },
    {
      "entropy": 6.383569192886353,
      "epoch": 0.9668789127294131,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004916651777574856,
      "loss": 6.2966,
      "mean_token_accuracy": 0.1298234224319458,
      "num_tokens": 19563439.0,
      "step": 9035
    },
    {
      "entropy": 6.440616226196289,
      "epoch": 0.9674139868371769,
      "grad_norm": 1.171875,
      "learning_rate": 0.000491654864723795,
      "loss": 6.309,
      "mean_token_accuracy": 0.13405263051390648,
      "num_tokens": 19574264.0,
      "step": 9040
    },
    {
      "entropy": 6.302301359176636,
      "epoch": 0.9679490609449408,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004916445454342256,
      "loss": 6.1863,
      "mean_token_accuracy": 0.1487156257033348,
      "num_tokens": 19585293.0,
      "step": 9045
    },
    {
      "entropy": 6.4724022388458256,
      "epoch": 0.9684841350527048,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004916342198890754,
      "loss": 6.2253,
      "mean_token_accuracy": 0.13311864212155342,
      "num_tokens": 19594384.0,
      "step": 9050
    },
    {
      "entropy": 6.39474630355835,
      "epoch": 0.9690192091604687,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004916238880886426,
      "loss": 6.3363,
      "mean_token_accuracy": 0.13666247799992562,
      "num_tokens": 19605144.0,
      "step": 9055
    },
    {
      "entropy": 6.38230528831482,
      "epoch": 0.9695542832682327,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004916135500332255,
      "loss": 6.2221,
      "mean_token_accuracy": 0.14119862169027328,
      "num_tokens": 19615885.0,
      "step": 9060
    },
    {
      "entropy": 6.372761344909668,
      "epoch": 0.9700893573759966,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004916032057231225,
      "loss": 6.2458,
      "mean_token_accuracy": 0.13269392400979996,
      "num_tokens": 19625749.0,
      "step": 9065
    },
    {
      "entropy": 6.408434247970581,
      "epoch": 0.9706244314837605,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004915928551586324,
      "loss": 6.3232,
      "mean_token_accuracy": 0.12646848261356353,
      "num_tokens": 19638285.0,
      "step": 9070
    },
    {
      "entropy": 6.557410097122192,
      "epoch": 0.9711595055915244,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004915824983400541,
      "loss": 6.3625,
      "mean_token_accuracy": 0.13077090457081794,
      "num_tokens": 19649697.0,
      "step": 9075
    },
    {
      "entropy": 6.4512067317962645,
      "epoch": 0.9716945796992883,
      "grad_norm": 1.25,
      "learning_rate": 0.0004915721352676864,
      "loss": 6.3659,
      "mean_token_accuracy": 0.1305341102182865,
      "num_tokens": 19659668.0,
      "step": 9080
    },
    {
      "entropy": 6.402630615234375,
      "epoch": 0.9722296538070523,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004915617659418289,
      "loss": 6.2417,
      "mean_token_accuracy": 0.13872757852077483,
      "num_tokens": 19670301.0,
      "step": 9085
    },
    {
      "entropy": 6.440298843383789,
      "epoch": 0.9727647279148162,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004915513903627807,
      "loss": 6.2709,
      "mean_token_accuracy": 0.1327598512172699,
      "num_tokens": 19680926.0,
      "step": 9090
    },
    {
      "entropy": 6.3762918472290036,
      "epoch": 0.9732998020225802,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004915410085308417,
      "loss": 6.237,
      "mean_token_accuracy": 0.1411748304963112,
      "num_tokens": 19691467.0,
      "step": 9095
    },
    {
      "entropy": 6.453802967071534,
      "epoch": 0.973834876130344,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004915306204463115,
      "loss": 6.2655,
      "mean_token_accuracy": 0.13385210037231446,
      "num_tokens": 19701139.0,
      "step": 9100
    },
    {
      "entropy": 6.412873601913452,
      "epoch": 0.974369950238108,
      "grad_norm": 1.3125,
      "learning_rate": 0.00049152022610949,
      "loss": 6.2591,
      "mean_token_accuracy": 0.13067844286561012,
      "num_tokens": 19710628.0,
      "step": 9105
    },
    {
      "entropy": 6.447878360748291,
      "epoch": 0.9749050243458719,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004915098255206773,
      "loss": 6.2754,
      "mean_token_accuracy": 0.1310727097094059,
      "num_tokens": 19721824.0,
      "step": 9110
    },
    {
      "entropy": 6.469791984558105,
      "epoch": 0.9754400984536358,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000491499418680174,
      "loss": 6.2884,
      "mean_token_accuracy": 0.12671151086688043,
      "num_tokens": 19732557.0,
      "step": 9115
    },
    {
      "entropy": 6.410218620300293,
      "epoch": 0.9759751725613998,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004914890055882804,
      "loss": 6.2962,
      "mean_token_accuracy": 0.13039457872509957,
      "num_tokens": 19743436.0,
      "step": 9120
    },
    {
      "entropy": 6.3921653747558596,
      "epoch": 0.9765102466691636,
      "grad_norm": 1.3125,
      "learning_rate": 0.000491478586245297,
      "loss": 6.1787,
      "mean_token_accuracy": 0.14053898602724074,
      "num_tokens": 19755305.0,
      "step": 9125
    },
    {
      "entropy": 6.394151449203491,
      "epoch": 0.9770453207769276,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000491468160651525,
      "loss": 6.2129,
      "mean_token_accuracy": 0.14196542277932167,
      "num_tokens": 19766223.0,
      "step": 9130
    },
    {
      "entropy": 6.34791669845581,
      "epoch": 0.9775803948846915,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004914577288072652,
      "loss": 6.2876,
      "mean_token_accuracy": 0.13174567073583604,
      "num_tokens": 19777268.0,
      "step": 9135
    },
    {
      "entropy": 6.441683387756347,
      "epoch": 0.9781154689924555,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004914472907128189,
      "loss": 6.3762,
      "mean_token_accuracy": 0.1336386688053608,
      "num_tokens": 19787875.0,
      "step": 9140
    },
    {
      "entropy": 6.49074420928955,
      "epoch": 0.9786505431002194,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004914368463684876,
      "loss": 6.2804,
      "mean_token_accuracy": 0.1349836528301239,
      "num_tokens": 19798919.0,
      "step": 9145
    },
    {
      "entropy": 6.421940803527832,
      "epoch": 0.9791856172079834,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004914263957745726,
      "loss": 6.2436,
      "mean_token_accuracy": 0.14296989589929582,
      "num_tokens": 19809320.0,
      "step": 9150
    },
    {
      "entropy": 6.485382175445556,
      "epoch": 0.9797206913157472,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004914159389313759,
      "loss": 6.3353,
      "mean_token_accuracy": 0.1336093783378601,
      "num_tokens": 19819511.0,
      "step": 9155
    },
    {
      "entropy": 6.451630449295044,
      "epoch": 0.9802557654235111,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004914054758391994,
      "loss": 6.3735,
      "mean_token_accuracy": 0.13097211793065072,
      "num_tokens": 19830381.0,
      "step": 9160
    },
    {
      "entropy": 6.362272596359253,
      "epoch": 0.9807908395312751,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004913950064983452,
      "loss": 6.2184,
      "mean_token_accuracy": 0.13412745371460916,
      "num_tokens": 19840806.0,
      "step": 9165
    },
    {
      "entropy": 6.392962312698364,
      "epoch": 0.981325913639039,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004913845309091156,
      "loss": 6.2169,
      "mean_token_accuracy": 0.1440105304121971,
      "num_tokens": 19851095.0,
      "step": 9170
    },
    {
      "entropy": 6.358264446258545,
      "epoch": 0.981860987746803,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004913740490718131,
      "loss": 6.1786,
      "mean_token_accuracy": 0.14627249762415886,
      "num_tokens": 19860235.0,
      "step": 9175
    },
    {
      "entropy": 6.355158567428589,
      "epoch": 0.9823960618545668,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004913635609867403,
      "loss": 6.2887,
      "mean_token_accuracy": 0.1299964301288128,
      "num_tokens": 19872264.0,
      "step": 9180
    },
    {
      "entropy": 6.485144138336182,
      "epoch": 0.9829311359623308,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004913530666542001,
      "loss": 6.3905,
      "mean_token_accuracy": 0.13083416223526,
      "num_tokens": 19883628.0,
      "step": 9185
    },
    {
      "entropy": 6.451532697677612,
      "epoch": 0.9834662100700947,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004913425660744955,
      "loss": 6.2494,
      "mean_token_accuracy": 0.135823655128479,
      "num_tokens": 19894624.0,
      "step": 9190
    },
    {
      "entropy": 6.4656219482421875,
      "epoch": 0.9840012841778586,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004913320592479297,
      "loss": 6.2754,
      "mean_token_accuracy": 0.13404600843787193,
      "num_tokens": 19904882.0,
      "step": 9195
    },
    {
      "entropy": 6.368783330917358,
      "epoch": 0.9845363582856226,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004913215461748061,
      "loss": 6.2053,
      "mean_token_accuracy": 0.14201799854636193,
      "num_tokens": 19915344.0,
      "step": 9200
    },
    {
      "entropy": 6.385826301574707,
      "epoch": 0.9850714323933865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004913110268554282,
      "loss": 6.3476,
      "mean_token_accuracy": 0.12912106588482858,
      "num_tokens": 19927527.0,
      "step": 9205
    },
    {
      "entropy": 6.4332037448883055,
      "epoch": 0.9856065065011504,
      "grad_norm": 1.125,
      "learning_rate": 0.0004913005012900998,
      "loss": 6.2936,
      "mean_token_accuracy": 0.1338207244873047,
      "num_tokens": 19938893.0,
      "step": 9210
    },
    {
      "entropy": 6.434023475646972,
      "epoch": 0.9861415806089143,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004912899694791248,
      "loss": 6.2556,
      "mean_token_accuracy": 0.13635288253426553,
      "num_tokens": 19949297.0,
      "step": 9215
    },
    {
      "entropy": 6.3331304550170895,
      "epoch": 0.9866766547166783,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004912794314228074,
      "loss": 6.2507,
      "mean_token_accuracy": 0.13737955093383789,
      "num_tokens": 19959194.0,
      "step": 9220
    },
    {
      "entropy": 6.440803813934326,
      "epoch": 0.9872117288244422,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004912688871214518,
      "loss": 6.3068,
      "mean_token_accuracy": 0.13499441668391227,
      "num_tokens": 19969896.0,
      "step": 9225
    },
    {
      "entropy": 6.470992565155029,
      "epoch": 0.9877468029322061,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004912583365753625,
      "loss": 6.2767,
      "mean_token_accuracy": 0.1372372254729271,
      "num_tokens": 19981505.0,
      "step": 9230
    },
    {
      "entropy": 6.411077404022217,
      "epoch": 0.9882818770399701,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004912477797848441,
      "loss": 6.2491,
      "mean_token_accuracy": 0.13840068578720094,
      "num_tokens": 19992770.0,
      "step": 9235
    },
    {
      "entropy": 6.390095329284668,
      "epoch": 0.9888169511477339,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004912372167502014,
      "loss": 6.2644,
      "mean_token_accuracy": 0.1379077322781086,
      "num_tokens": 20003572.0,
      "step": 9240
    },
    {
      "entropy": 6.424465370178223,
      "epoch": 0.9893520252554979,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004912266474717394,
      "loss": 6.2906,
      "mean_token_accuracy": 0.13903327137231827,
      "num_tokens": 20014500.0,
      "step": 9245
    },
    {
      "entropy": 6.466884613037109,
      "epoch": 0.9898870993632618,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004912160719497635,
      "loss": 6.2925,
      "mean_token_accuracy": 0.13860635310411454,
      "num_tokens": 20025745.0,
      "step": 9250
    },
    {
      "entropy": 6.372235345840454,
      "epoch": 0.9904221734710258,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004912054901845788,
      "loss": 6.1589,
      "mean_token_accuracy": 0.13967020884156228,
      "num_tokens": 20036346.0,
      "step": 9255
    },
    {
      "entropy": 6.297544145584107,
      "epoch": 0.9909572475787897,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000491194902176491,
      "loss": 6.3138,
      "mean_token_accuracy": 0.13613977283239365,
      "num_tokens": 20048162.0,
      "step": 9260
    },
    {
      "entropy": 6.451009225845337,
      "epoch": 0.9914923216865535,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004911843079258057,
      "loss": 6.4173,
      "mean_token_accuracy": 0.1248807080090046,
      "num_tokens": 20058947.0,
      "step": 9265
    },
    {
      "entropy": 6.554871225357056,
      "epoch": 0.9920273957943175,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004911737074328289,
      "loss": 6.3275,
      "mean_token_accuracy": 0.1366781234741211,
      "num_tokens": 20069282.0,
      "step": 9270
    },
    {
      "entropy": 6.425151681900024,
      "epoch": 0.9925624699020814,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004911631006978667,
      "loss": 6.3018,
      "mean_token_accuracy": 0.1395340606570244,
      "num_tokens": 20078347.0,
      "step": 9275
    },
    {
      "entropy": 6.3900727272033695,
      "epoch": 0.9930975440098454,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004911524877212254,
      "loss": 6.3016,
      "mean_token_accuracy": 0.13120867535471917,
      "num_tokens": 20089129.0,
      "step": 9280
    },
    {
      "entropy": 6.456008958816528,
      "epoch": 0.9936326181176093,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004911418685032114,
      "loss": 6.3227,
      "mean_token_accuracy": 0.12962408363819122,
      "num_tokens": 20101008.0,
      "step": 9285
    },
    {
      "entropy": 6.3841489315032955,
      "epoch": 0.9941676922253733,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004911312430441312,
      "loss": 6.299,
      "mean_token_accuracy": 0.13915802761912346,
      "num_tokens": 20112272.0,
      "step": 9290
    },
    {
      "entropy": 6.392724132537841,
      "epoch": 0.9947027663331371,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004911206113442918,
      "loss": 6.2008,
      "mean_token_accuracy": 0.14379067569971085,
      "num_tokens": 20122814.0,
      "step": 9295
    },
    {
      "entropy": 6.337383794784546,
      "epoch": 0.995237840440901,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004911099734040001,
      "loss": 6.1561,
      "mean_token_accuracy": 0.1414629526436329,
      "num_tokens": 20133169.0,
      "step": 9300
    },
    {
      "entropy": 6.437785530090332,
      "epoch": 0.995772914548665,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004910993292235634,
      "loss": 6.3178,
      "mean_token_accuracy": 0.12474485263228416,
      "num_tokens": 20144355.0,
      "step": 9305
    },
    {
      "entropy": 6.497580051422119,
      "epoch": 0.9963079886564289,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004910886788032887,
      "loss": 6.3146,
      "mean_token_accuracy": 0.13189335390925408,
      "num_tokens": 20154486.0,
      "step": 9310
    },
    {
      "entropy": 6.444418430328369,
      "epoch": 0.9968430627641929,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004910780221434839,
      "loss": 6.2642,
      "mean_token_accuracy": 0.13965663835406303,
      "num_tokens": 20166359.0,
      "step": 9315
    },
    {
      "entropy": 6.395610523223877,
      "epoch": 0.9973781368719568,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004910673592444566,
      "loss": 6.2759,
      "mean_token_accuracy": 0.13586723580956458,
      "num_tokens": 20176469.0,
      "step": 9320
    },
    {
      "entropy": 6.45807294845581,
      "epoch": 0.9979132109797207,
      "grad_norm": 1.25,
      "learning_rate": 0.0004910566901065147,
      "loss": 6.3148,
      "mean_token_accuracy": 0.13653102815151213,
      "num_tokens": 20188334.0,
      "step": 9325
    },
    {
      "entropy": 6.41289119720459,
      "epoch": 0.9984482850874846,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004910460147299661,
      "loss": 6.205,
      "mean_token_accuracy": 0.14479871317744256,
      "num_tokens": 20198594.0,
      "step": 9330
    },
    {
      "entropy": 6.4022280216217045,
      "epoch": 0.9989833591952485,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004910353331151193,
      "loss": 6.3289,
      "mean_token_accuracy": 0.13693785220384597,
      "num_tokens": 20209456.0,
      "step": 9335
    },
    {
      "entropy": 6.43259859085083,
      "epoch": 0.9995184333030125,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004910246452622825,
      "loss": 6.304,
      "mean_token_accuracy": 0.13724788948893546,
      "num_tokens": 20219835.0,
      "step": 9340
    },
    {
      "entropy": 6.478713512420654,
      "epoch": 1.0,
      "grad_norm": 2.625,
      "learning_rate": 0.0004910139511717646,
      "loss": 6.3313,
      "mean_token_accuracy": 0.12968112611108357,
      "num_tokens": 20228258.0,
      "step": 9345
    },
    {
      "entropy": 6.434733629226685,
      "epoch": 1.0005350741077639,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004910032508438739,
      "loss": 6.1935,
      "mean_token_accuracy": 0.14255010932683945,
      "num_tokens": 20238875.0,
      "step": 9350
    },
    {
      "entropy": 6.41102876663208,
      "epoch": 1.0010701482155278,
      "grad_norm": 1.15625,
      "learning_rate": 0.00049099254427892,
      "loss": 6.2184,
      "mean_token_accuracy": 0.13098681196570397,
      "num_tokens": 20249750.0,
      "step": 9355
    },
    {
      "entropy": 6.460056352615356,
      "epoch": 1.0016052223232919,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004909818314772116,
      "loss": 6.2477,
      "mean_token_accuracy": 0.1302698940038681,
      "num_tokens": 20259772.0,
      "step": 9360
    },
    {
      "entropy": 6.3592267513275145,
      "epoch": 1.0021402964310557,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004909711124390582,
      "loss": 6.168,
      "mean_token_accuracy": 0.1427948519587517,
      "num_tokens": 20270230.0,
      "step": 9365
    },
    {
      "entropy": 6.283496999740601,
      "epoch": 1.0026753705388196,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004909603871647692,
      "loss": 6.0078,
      "mean_token_accuracy": 0.15665441155433654,
      "num_tokens": 20280317.0,
      "step": 9370
    },
    {
      "entropy": 6.341282367706299,
      "epoch": 1.0032104446465835,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004909496556546544,
      "loss": 6.1108,
      "mean_token_accuracy": 0.1458029180765152,
      "num_tokens": 20291147.0,
      "step": 9375
    },
    {
      "entropy": 6.299035120010376,
      "epoch": 1.0037455187543476,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004909389179090238,
      "loss": 6.172,
      "mean_token_accuracy": 0.14469484239816666,
      "num_tokens": 20301095.0,
      "step": 9380
    },
    {
      "entropy": 6.386391067504883,
      "epoch": 1.0042805928621115,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004909281739281871,
      "loss": 6.1767,
      "mean_token_accuracy": 0.14344390332698823,
      "num_tokens": 20311216.0,
      "step": 9385
    },
    {
      "entropy": 6.476996564865113,
      "epoch": 1.0048156669698753,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004909174237124548,
      "loss": 6.2413,
      "mean_token_accuracy": 0.12570007964968682,
      "num_tokens": 20321719.0,
      "step": 9390
    },
    {
      "entropy": 6.387945079803467,
      "epoch": 1.0053507410776392,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004909066672621372,
      "loss": 6.2369,
      "mean_token_accuracy": 0.14125329926609992,
      "num_tokens": 20331997.0,
      "step": 9395
    },
    {
      "entropy": 6.38417272567749,
      "epoch": 1.005885815185403,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004908959045775449,
      "loss": 6.2764,
      "mean_token_accuracy": 0.13274534717202185,
      "num_tokens": 20342973.0,
      "step": 9400
    },
    {
      "entropy": 6.3823747634887695,
      "epoch": 1.0064208892931672,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004908851356589887,
      "loss": 6.1471,
      "mean_token_accuracy": 0.14157598167657853,
      "num_tokens": 20352652.0,
      "step": 9405
    },
    {
      "entropy": 6.435864877700806,
      "epoch": 1.006955963400931,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004908743605067796,
      "loss": 6.2297,
      "mean_token_accuracy": 0.14013049155473709,
      "num_tokens": 20364577.0,
      "step": 9410
    },
    {
      "entropy": 6.38976092338562,
      "epoch": 1.007491037508695,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004908635791212287,
      "loss": 6.2857,
      "mean_token_accuracy": 0.13247171193361282,
      "num_tokens": 20375451.0,
      "step": 9415
    },
    {
      "entropy": 6.415505599975586,
      "epoch": 1.0080261116164588,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004908527915026472,
      "loss": 6.1813,
      "mean_token_accuracy": 0.13975390568375587,
      "num_tokens": 20385174.0,
      "step": 9420
    },
    {
      "entropy": 6.397196435928345,
      "epoch": 1.0085611857242227,
      "grad_norm": 1.125,
      "learning_rate": 0.0004908419976513467,
      "loss": 6.1097,
      "mean_token_accuracy": 0.14368658885359764,
      "num_tokens": 20395710.0,
      "step": 9425
    },
    {
      "entropy": 6.282300186157227,
      "epoch": 1.0090962598319868,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004908311975676388,
      "loss": 6.1363,
      "mean_token_accuracy": 0.13966590836644172,
      "num_tokens": 20406572.0,
      "step": 9430
    },
    {
      "entropy": 6.310158920288086,
      "epoch": 1.0096313339397507,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004908203912518355,
      "loss": 6.121,
      "mean_token_accuracy": 0.14767917096614838,
      "num_tokens": 20417786.0,
      "step": 9435
    },
    {
      "entropy": 6.4140305519104,
      "epoch": 1.0101664080475146,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004908095787042487,
      "loss": 6.1671,
      "mean_token_accuracy": 0.13980057686567307,
      "num_tokens": 20427680.0,
      "step": 9440
    },
    {
      "entropy": 6.309698629379272,
      "epoch": 1.0107014821552784,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004907987599251907,
      "loss": 6.1528,
      "mean_token_accuracy": 0.1391996294260025,
      "num_tokens": 20438640.0,
      "step": 9445
    },
    {
      "entropy": 6.360296106338501,
      "epoch": 1.0112365562630425,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004907879349149737,
      "loss": 6.1702,
      "mean_token_accuracy": 0.14285122305154802,
      "num_tokens": 20449939.0,
      "step": 9450
    },
    {
      "entropy": 6.337630796432495,
      "epoch": 1.0117716303708064,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004907771036739105,
      "loss": 6.1083,
      "mean_token_accuracy": 0.14407198876142502,
      "num_tokens": 20460845.0,
      "step": 9455
    },
    {
      "entropy": 6.396984195709228,
      "epoch": 1.0123067044785703,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004907662662023137,
      "loss": 6.2432,
      "mean_token_accuracy": 0.13334245085716248,
      "num_tokens": 20472407.0,
      "step": 9460
    },
    {
      "entropy": 6.40310115814209,
      "epoch": 1.0128417785863342,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004907554225004962,
      "loss": 6.2419,
      "mean_token_accuracy": 0.13837286904454232,
      "num_tokens": 20483192.0,
      "step": 9465
    },
    {
      "entropy": 6.472944021224976,
      "epoch": 1.013376852694098,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004907445725687714,
      "loss": 6.3054,
      "mean_token_accuracy": 0.13163694813847543,
      "num_tokens": 20493498.0,
      "step": 9470
    },
    {
      "entropy": 6.430975675582886,
      "epoch": 1.0139119268018622,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004907337164074523,
      "loss": 6.2173,
      "mean_token_accuracy": 0.13783009424805642,
      "num_tokens": 20503781.0,
      "step": 9475
    },
    {
      "entropy": 6.330007600784302,
      "epoch": 1.014447000909626,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004907228540168523,
      "loss": 6.2794,
      "mean_token_accuracy": 0.1342932716012001,
      "num_tokens": 20515142.0,
      "step": 9480
    },
    {
      "entropy": 6.436728811264038,
      "epoch": 1.01498207501739,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004907119853972854,
      "loss": 6.2351,
      "mean_token_accuracy": 0.1364154815673828,
      "num_tokens": 20527568.0,
      "step": 9485
    },
    {
      "entropy": 6.363270282745361,
      "epoch": 1.0155171491251538,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004907011105490651,
      "loss": 6.1458,
      "mean_token_accuracy": 0.142462557554245,
      "num_tokens": 20538198.0,
      "step": 9490
    },
    {
      "entropy": 6.429973173141479,
      "epoch": 1.0160522232329177,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004906902294725056,
      "loss": 6.1237,
      "mean_token_accuracy": 0.14695967733860016,
      "num_tokens": 20549355.0,
      "step": 9495
    },
    {
      "entropy": 6.3325855255126955,
      "epoch": 1.0165872973406818,
      "grad_norm": 1.21875,
      "learning_rate": 0.000490679342167921,
      "loss": 6.228,
      "mean_token_accuracy": 0.1414795108139515,
      "num_tokens": 20558984.0,
      "step": 9500
    },
    {
      "entropy": 6.354425621032715,
      "epoch": 1.0171223714484456,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004906684486356258,
      "loss": 6.247,
      "mean_token_accuracy": 0.13448161035776138,
      "num_tokens": 20570143.0,
      "step": 9505
    },
    {
      "entropy": 6.440261173248291,
      "epoch": 1.0176574455562095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004906575488759343,
      "loss": 6.2392,
      "mean_token_accuracy": 0.1359660729765892,
      "num_tokens": 20580973.0,
      "step": 9510
    },
    {
      "entropy": 6.435061740875244,
      "epoch": 1.0181925196639734,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004906466428891616,
      "loss": 6.202,
      "mean_token_accuracy": 0.13522051200270652,
      "num_tokens": 20591473.0,
      "step": 9515
    },
    {
      "entropy": 6.3393439769744875,
      "epoch": 1.0187275937717375,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004906357306756222,
      "loss": 6.1668,
      "mean_token_accuracy": 0.13699494451284408,
      "num_tokens": 20602347.0,
      "step": 9520
    },
    {
      "entropy": 6.413664817810059,
      "epoch": 1.0192626678795014,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004906248122356315,
      "loss": 6.2338,
      "mean_token_accuracy": 0.14229237884283066,
      "num_tokens": 20613168.0,
      "step": 9525
    },
    {
      "entropy": 6.40576753616333,
      "epoch": 1.0197977419872652,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004906138875695045,
      "loss": 6.1193,
      "mean_token_accuracy": 0.14158989638090133,
      "num_tokens": 20623523.0,
      "step": 9530
    },
    {
      "entropy": 6.304319953918457,
      "epoch": 1.0203328160950291,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004906029566775569,
      "loss": 6.1657,
      "mean_token_accuracy": 0.14247490763664244,
      "num_tokens": 20634085.0,
      "step": 9535
    },
    {
      "entropy": 6.367363739013672,
      "epoch": 1.020867890202793,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004905920195601042,
      "loss": 6.2349,
      "mean_token_accuracy": 0.13838397935032845,
      "num_tokens": 20644747.0,
      "step": 9540
    },
    {
      "entropy": 6.391586875915527,
      "epoch": 1.021402964310557,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004905810762174622,
      "loss": 6.159,
      "mean_token_accuracy": 0.12744452953338622,
      "num_tokens": 20655704.0,
      "step": 9545
    },
    {
      "entropy": 6.42578616142273,
      "epoch": 1.021938038418321,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004905701266499469,
      "loss": 6.1582,
      "mean_token_accuracy": 0.14301931038498877,
      "num_tokens": 20665844.0,
      "step": 9550
    },
    {
      "entropy": 6.428760385513305,
      "epoch": 1.0224731125260849,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004905591708578746,
      "loss": 6.2984,
      "mean_token_accuracy": 0.12952255681157113,
      "num_tokens": 20676921.0,
      "step": 9555
    },
    {
      "entropy": 6.4479869365692135,
      "epoch": 1.0230081866338487,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004905482088415615,
      "loss": 6.2235,
      "mean_token_accuracy": 0.13627402037382125,
      "num_tokens": 20688476.0,
      "step": 9560
    },
    {
      "entropy": 6.418657302856445,
      "epoch": 1.0235432607416126,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004905372406013241,
      "loss": 6.2311,
      "mean_token_accuracy": 0.1371014252305031,
      "num_tokens": 20698588.0,
      "step": 9565
    },
    {
      "entropy": 6.356596517562866,
      "epoch": 1.0240783348493767,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004905262661374792,
      "loss": 6.1996,
      "mean_token_accuracy": 0.14309823736548424,
      "num_tokens": 20709181.0,
      "step": 9570
    },
    {
      "entropy": 6.355239725112915,
      "epoch": 1.0246134089571406,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004905152854503436,
      "loss": 6.1305,
      "mean_token_accuracy": 0.14588012918829918,
      "num_tokens": 20719724.0,
      "step": 9575
    },
    {
      "entropy": 6.4224534034729,
      "epoch": 1.0251484830649045,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004905042985402345,
      "loss": 6.2375,
      "mean_token_accuracy": 0.13777188956737518,
      "num_tokens": 20730548.0,
      "step": 9580
    },
    {
      "entropy": 6.380663013458252,
      "epoch": 1.0256835571726683,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000490493305407469,
      "loss": 6.2157,
      "mean_token_accuracy": 0.13424318581819533,
      "num_tokens": 20742124.0,
      "step": 9585
    },
    {
      "entropy": 6.439380121231079,
      "epoch": 1.0262186312804324,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004904823060523646,
      "loss": 6.2917,
      "mean_token_accuracy": 0.1307416945695877,
      "num_tokens": 20753313.0,
      "step": 9590
    },
    {
      "entropy": 6.4981752872467045,
      "epoch": 1.0267537053881963,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004904713004752389,
      "loss": 6.2573,
      "mean_token_accuracy": 0.13681431859731674,
      "num_tokens": 20763539.0,
      "step": 9595
    },
    {
      "entropy": 6.428672027587891,
      "epoch": 1.0272887794959602,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004904602886764097,
      "loss": 6.3427,
      "mean_token_accuracy": 0.1298852026462555,
      "num_tokens": 20775849.0,
      "step": 9600
    },
    {
      "entropy": 6.422748327255249,
      "epoch": 1.027823853603724,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004904492706561948,
      "loss": 6.2067,
      "mean_token_accuracy": 0.133476722240448,
      "num_tokens": 20787460.0,
      "step": 9605
    },
    {
      "entropy": 6.428693437576294,
      "epoch": 1.028358927711488,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004904382464149127,
      "loss": 6.1257,
      "mean_token_accuracy": 0.14555379897356033,
      "num_tokens": 20798478.0,
      "step": 9610
    },
    {
      "entropy": 6.444856452941894,
      "epoch": 1.028894001819252,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004904272159528814,
      "loss": 6.2294,
      "mean_token_accuracy": 0.13855807334184647,
      "num_tokens": 20809730.0,
      "step": 9615
    },
    {
      "entropy": 6.379982852935791,
      "epoch": 1.029429075927016,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004904161792704195,
      "loss": 6.2131,
      "mean_token_accuracy": 0.14618081748485565,
      "num_tokens": 20821438.0,
      "step": 9620
    },
    {
      "entropy": 6.35428409576416,
      "epoch": 1.0299641500347798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004904051363678457,
      "loss": 6.1874,
      "mean_token_accuracy": 0.13236298710107802,
      "num_tokens": 20832328.0,
      "step": 9625
    },
    {
      "entropy": 6.3328474998474125,
      "epoch": 1.0304992241425437,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004903940872454789,
      "loss": 6.165,
      "mean_token_accuracy": 0.14299442619085312,
      "num_tokens": 20843956.0,
      "step": 9630
    },
    {
      "entropy": 6.466890335083008,
      "epoch": 1.0310342982503076,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004903830319036381,
      "loss": 6.21,
      "mean_token_accuracy": 0.13927969485521316,
      "num_tokens": 20854578.0,
      "step": 9635
    },
    {
      "entropy": 6.3322436809539795,
      "epoch": 1.0315693723580717,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004903719703426425,
      "loss": 6.14,
      "mean_token_accuracy": 0.13872639387845992,
      "num_tokens": 20865190.0,
      "step": 9640
    },
    {
      "entropy": 6.3720293045043945,
      "epoch": 1.0321044464658355,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004903609025628115,
      "loss": 6.1761,
      "mean_token_accuracy": 0.13710325807332993,
      "num_tokens": 20875733.0,
      "step": 9645
    },
    {
      "entropy": 6.343014287948608,
      "epoch": 1.0326395205735994,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004903498285644647,
      "loss": 6.1985,
      "mean_token_accuracy": 0.13315223529934883,
      "num_tokens": 20886900.0,
      "step": 9650
    },
    {
      "entropy": 6.423218679428101,
      "epoch": 1.0331745946813633,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004903387483479219,
      "loss": 6.2417,
      "mean_token_accuracy": 0.13559042438864707,
      "num_tokens": 20898125.0,
      "step": 9655
    },
    {
      "entropy": 6.488476419448853,
      "epoch": 1.0337096687891274,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004903276619135029,
      "loss": 6.2468,
      "mean_token_accuracy": 0.136351977288723,
      "num_tokens": 20909788.0,
      "step": 9660
    },
    {
      "entropy": 6.4295604705810545,
      "epoch": 1.0342447428968913,
      "grad_norm": 1.125,
      "learning_rate": 0.000490316569261528,
      "loss": 6.2338,
      "mean_token_accuracy": 0.13587179630994797,
      "num_tokens": 20921853.0,
      "step": 9665
    },
    {
      "entropy": 6.393593597412109,
      "epoch": 1.0347798170046552,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004903054703923175,
      "loss": 6.223,
      "mean_token_accuracy": 0.13195437863469123,
      "num_tokens": 20932574.0,
      "step": 9670
    },
    {
      "entropy": 6.386069631576538,
      "epoch": 1.035314891112419,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004902943653061916,
      "loss": 6.192,
      "mean_token_accuracy": 0.13450067043304442,
      "num_tokens": 20943458.0,
      "step": 9675
    },
    {
      "entropy": 6.3849098682403564,
      "epoch": 1.035849965220183,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004902832540034712,
      "loss": 6.1663,
      "mean_token_accuracy": 0.1430095337331295,
      "num_tokens": 20953213.0,
      "step": 9680
    },
    {
      "entropy": 6.347196626663208,
      "epoch": 1.036385039327947,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004902721364844773,
      "loss": 6.1902,
      "mean_token_accuracy": 0.13722362965345383,
      "num_tokens": 20963885.0,
      "step": 9685
    },
    {
      "entropy": 6.351762485504151,
      "epoch": 1.0369201134357109,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004902610127495305,
      "loss": 6.1933,
      "mean_token_accuracy": 0.1418081820011139,
      "num_tokens": 20975529.0,
      "step": 9690
    },
    {
      "entropy": 6.404678630828857,
      "epoch": 1.0374551875434748,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004902498827989523,
      "loss": 6.1557,
      "mean_token_accuracy": 0.1498997762799263,
      "num_tokens": 20985279.0,
      "step": 9695
    },
    {
      "entropy": 6.430759525299072,
      "epoch": 1.0379902616512386,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004902387466330639,
      "loss": 6.2129,
      "mean_token_accuracy": 0.1415492981672287,
      "num_tokens": 20995818.0,
      "step": 9700
    },
    {
      "entropy": 6.34493408203125,
      "epoch": 1.0385253357590025,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000490227604252187,
      "loss": 6.1355,
      "mean_token_accuracy": 0.13756826892495155,
      "num_tokens": 21006786.0,
      "step": 9705
    },
    {
      "entropy": 6.386413669586181,
      "epoch": 1.0390604098667666,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004902164556566433,
      "loss": 6.1889,
      "mean_token_accuracy": 0.14337150752544403,
      "num_tokens": 21017441.0,
      "step": 9710
    },
    {
      "entropy": 6.244846343994141,
      "epoch": 1.0395954839745305,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004902053008467546,
      "loss": 6.0644,
      "mean_token_accuracy": 0.1488602191209793,
      "num_tokens": 21027812.0,
      "step": 9715
    },
    {
      "entropy": 6.360077714920044,
      "epoch": 1.0401305580822944,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004901941398228431,
      "loss": 6.2463,
      "mean_token_accuracy": 0.14060127586126328,
      "num_tokens": 21038960.0,
      "step": 9720
    },
    {
      "entropy": 6.453442668914795,
      "epoch": 1.0406656321900583,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004901829725852311,
      "loss": 6.2054,
      "mean_token_accuracy": 0.14095733687281609,
      "num_tokens": 21050467.0,
      "step": 9725
    },
    {
      "entropy": 6.354308652877807,
      "epoch": 1.0412007062978224,
      "grad_norm": 1.109375,
      "learning_rate": 0.000490171799134241,
      "loss": 6.1306,
      "mean_token_accuracy": 0.13385181054472922,
      "num_tokens": 21061208.0,
      "step": 9730
    },
    {
      "entropy": 6.301920795440674,
      "epoch": 1.0417357804055862,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004901606194701954,
      "loss": 6.2185,
      "mean_token_accuracy": 0.14279760271310807,
      "num_tokens": 21072872.0,
      "step": 9735
    },
    {
      "entropy": 6.421150493621826,
      "epoch": 1.04227085451335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004901494335934172,
      "loss": 6.1952,
      "mean_token_accuracy": 0.14038705080747604,
      "num_tokens": 21084878.0,
      "step": 9740
    },
    {
      "entropy": 6.444381380081177,
      "epoch": 1.042805928621114,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004901382415042293,
      "loss": 6.2404,
      "mean_token_accuracy": 0.13579076677560806,
      "num_tokens": 21095289.0,
      "step": 9745
    },
    {
      "entropy": 6.404003286361695,
      "epoch": 1.0433410027288779,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004901270432029549,
      "loss": 6.1869,
      "mean_token_accuracy": 0.13918179795145988,
      "num_tokens": 21106077.0,
      "step": 9750
    },
    {
      "entropy": 6.2948565006256105,
      "epoch": 1.043876076836642,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004901158386899174,
      "loss": 6.1031,
      "mean_token_accuracy": 0.1473974458873272,
      "num_tokens": 21116548.0,
      "step": 9755
    },
    {
      "entropy": 6.33483476638794,
      "epoch": 1.0444111509444058,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004901046279654402,
      "loss": 6.1325,
      "mean_token_accuracy": 0.1511349469423294,
      "num_tokens": 21126163.0,
      "step": 9760
    },
    {
      "entropy": 6.400875473022461,
      "epoch": 1.0449462250521697,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004900934110298471,
      "loss": 6.241,
      "mean_token_accuracy": 0.13385264426469803,
      "num_tokens": 21136830.0,
      "step": 9765
    },
    {
      "entropy": 6.418988180160523,
      "epoch": 1.0454812991599336,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004900821878834621,
      "loss": 6.1799,
      "mean_token_accuracy": 0.1416347749531269,
      "num_tokens": 21146793.0,
      "step": 9770
    },
    {
      "entropy": 6.398195409774781,
      "epoch": 1.0460163732676975,
      "grad_norm": 1.34375,
      "learning_rate": 0.000490070958526609,
      "loss": 6.232,
      "mean_token_accuracy": 0.14153644517064096,
      "num_tokens": 21156799.0,
      "step": 9775
    },
    {
      "entropy": 6.349988412857056,
      "epoch": 1.0465514473754616,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004900597229596124,
      "loss": 6.2291,
      "mean_token_accuracy": 0.13786681443452836,
      "num_tokens": 21167757.0,
      "step": 9780
    },
    {
      "entropy": 6.4059953689575195,
      "epoch": 1.0470865214832255,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004900484811827963,
      "loss": 6.2192,
      "mean_token_accuracy": 0.14064437076449393,
      "num_tokens": 21178832.0,
      "step": 9785
    },
    {
      "entropy": 6.347813510894776,
      "epoch": 1.0476215955909893,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004900372331964856,
      "loss": 6.1276,
      "mean_token_accuracy": 0.1452965885400772,
      "num_tokens": 21189637.0,
      "step": 9790
    },
    {
      "entropy": 6.308510494232178,
      "epoch": 1.0481566696987532,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000490025979001005,
      "loss": 6.1611,
      "mean_token_accuracy": 0.1385588452219963,
      "num_tokens": 21200194.0,
      "step": 9795
    },
    {
      "entropy": 6.357607364654541,
      "epoch": 1.0486917438065173,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004900147185966795,
      "loss": 6.1168,
      "mean_token_accuracy": 0.1411648653447628,
      "num_tokens": 21211405.0,
      "step": 9800
    },
    {
      "entropy": 6.4042435646057125,
      "epoch": 1.0492268179142812,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004900034519838342,
      "loss": 6.1738,
      "mean_token_accuracy": 0.13165768459439278,
      "num_tokens": 21222722.0,
      "step": 9805
    },
    {
      "entropy": 6.35134973526001,
      "epoch": 1.049761892022045,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004899921791627945,
      "loss": 6.1301,
      "mean_token_accuracy": 0.14313431829214096,
      "num_tokens": 21233510.0,
      "step": 9810
    },
    {
      "entropy": 6.4295238018035885,
      "epoch": 1.050296966129809,
      "grad_norm": 1.25,
      "learning_rate": 0.0004899809001338857,
      "loss": 6.2726,
      "mean_token_accuracy": 0.1281655766069889,
      "num_tokens": 21243235.0,
      "step": 9815
    },
    {
      "entropy": 6.400166702270508,
      "epoch": 1.0508320402375728,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004899696148974338,
      "loss": 6.1684,
      "mean_token_accuracy": 0.14122225120663642,
      "num_tokens": 21254680.0,
      "step": 9820
    },
    {
      "entropy": 6.338776111602783,
      "epoch": 1.051367114345337,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004899583234537644,
      "loss": 6.1624,
      "mean_token_accuracy": 0.13677000254392624,
      "num_tokens": 21265116.0,
      "step": 9825
    },
    {
      "entropy": 6.370752239227295,
      "epoch": 1.0519021884531008,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004899470258032034,
      "loss": 6.1375,
      "mean_token_accuracy": 0.13780994936823845,
      "num_tokens": 21277066.0,
      "step": 9830
    },
    {
      "entropy": 6.4266783714294435,
      "epoch": 1.0524372625608647,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004899357219460775,
      "loss": 6.2252,
      "mean_token_accuracy": 0.13564271554350854,
      "num_tokens": 21287332.0,
      "step": 9835
    },
    {
      "entropy": 6.32109112739563,
      "epoch": 1.0529723366686286,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004899244118827128,
      "loss": 6.1997,
      "mean_token_accuracy": 0.14000920653343202,
      "num_tokens": 21298377.0,
      "step": 9840
    },
    {
      "entropy": 6.424011945724487,
      "epoch": 1.0535074107763924,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004899130956134358,
      "loss": 6.1746,
      "mean_token_accuracy": 0.1407710373401642,
      "num_tokens": 21309966.0,
      "step": 9845
    },
    {
      "entropy": 6.3457118511199955,
      "epoch": 1.0540424848841565,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004899017731385735,
      "loss": 6.0917,
      "mean_token_accuracy": 0.14841809794306754,
      "num_tokens": 21320399.0,
      "step": 9850
    },
    {
      "entropy": 6.381113767623901,
      "epoch": 1.0545775589919204,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004898904444584527,
      "loss": 6.2117,
      "mean_token_accuracy": 0.140783654153347,
      "num_tokens": 21331716.0,
      "step": 9855
    },
    {
      "entropy": 6.38757848739624,
      "epoch": 1.0551126330996843,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004898791095734004,
      "loss": 6.2246,
      "mean_token_accuracy": 0.1353653773665428,
      "num_tokens": 21341664.0,
      "step": 9860
    },
    {
      "entropy": 6.370021820068359,
      "epoch": 1.0556477072074482,
      "grad_norm": 1.125,
      "learning_rate": 0.000489867768483744,
      "loss": 6.1829,
      "mean_token_accuracy": 0.1387489691376686,
      "num_tokens": 21352220.0,
      "step": 9865
    },
    {
      "entropy": 6.451702451705932,
      "epoch": 1.0561827813152123,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004898564211898111,
      "loss": 6.2908,
      "mean_token_accuracy": 0.13319606631994246,
      "num_tokens": 21363947.0,
      "step": 9870
    },
    {
      "entropy": 6.417100143432617,
      "epoch": 1.0567178554229761,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004898450676919291,
      "loss": 6.1683,
      "mean_token_accuracy": 0.14398796260356903,
      "num_tokens": 21374011.0,
      "step": 9875
    },
    {
      "entropy": 6.352324533462524,
      "epoch": 1.05725292953074,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000489833707990426,
      "loss": 6.2341,
      "mean_token_accuracy": 0.14070496559143067,
      "num_tokens": 21384232.0,
      "step": 9880
    },
    {
      "entropy": 6.399838161468506,
      "epoch": 1.057788003638504,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004898223420856298,
      "loss": 6.1975,
      "mean_token_accuracy": 0.13823896646499634,
      "num_tokens": 21395928.0,
      "step": 9885
    },
    {
      "entropy": 6.421978855133057,
      "epoch": 1.0583230777462678,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004898109699778686,
      "loss": 6.2013,
      "mean_token_accuracy": 0.1374720647931099,
      "num_tokens": 21406391.0,
      "step": 9890
    },
    {
      "entropy": 6.366811466217041,
      "epoch": 1.0588581518540319,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000489799591667471,
      "loss": 6.2457,
      "mean_token_accuracy": 0.13486332446336746,
      "num_tokens": 21417859.0,
      "step": 9895
    },
    {
      "entropy": 6.285160827636719,
      "epoch": 1.0593932259617957,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004897882071547651,
      "loss": 6.1066,
      "mean_token_accuracy": 0.14253716319799423,
      "num_tokens": 21428284.0,
      "step": 9900
    },
    {
      "entropy": 6.3580926895141605,
      "epoch": 1.0599283000695596,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004897768164400801,
      "loss": 6.2523,
      "mean_token_accuracy": 0.13308200985193253,
      "num_tokens": 21439610.0,
      "step": 9905
    },
    {
      "entropy": 6.445244216918946,
      "epoch": 1.0604633741773235,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004897654195237446,
      "loss": 6.2546,
      "mean_token_accuracy": 0.13394749462604522,
      "num_tokens": 21449319.0,
      "step": 9910
    },
    {
      "entropy": 6.417538166046143,
      "epoch": 1.0609984482850874,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004897540164060879,
      "loss": 6.2568,
      "mean_token_accuracy": 0.14111557453870774,
      "num_tokens": 21459851.0,
      "step": 9915
    },
    {
      "entropy": 6.342381191253662,
      "epoch": 1.0615335223928515,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004897426070874392,
      "loss": 6.1882,
      "mean_token_accuracy": 0.1415725141763687,
      "num_tokens": 21469717.0,
      "step": 9920
    },
    {
      "entropy": 6.429379558563232,
      "epoch": 1.0620685965006154,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004897311915681278,
      "loss": 6.2996,
      "mean_token_accuracy": 0.13520136028528212,
      "num_tokens": 21480815.0,
      "step": 9925
    },
    {
      "entropy": 6.451131772994995,
      "epoch": 1.0626036706083792,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004897197698484834,
      "loss": 6.2002,
      "mean_token_accuracy": 0.13811461701989175,
      "num_tokens": 21492000.0,
      "step": 9930
    },
    {
      "entropy": 6.311298513412476,
      "epoch": 1.0631387447161431,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004897083419288358,
      "loss": 6.1793,
      "mean_token_accuracy": 0.13618579357862473,
      "num_tokens": 21502006.0,
      "step": 9935
    },
    {
      "entropy": 6.36651759147644,
      "epoch": 1.0636738188239072,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000489696907809515,
      "loss": 6.224,
      "mean_token_accuracy": 0.14429882168769836,
      "num_tokens": 21513066.0,
      "step": 9940
    },
    {
      "entropy": 6.370158720016479,
      "epoch": 1.064208892931671,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004896854674908512,
      "loss": 6.2058,
      "mean_token_accuracy": 0.13385366946458815,
      "num_tokens": 21524853.0,
      "step": 9945
    },
    {
      "entropy": 6.397624540328979,
      "epoch": 1.064743967039435,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004896740209731746,
      "loss": 6.1279,
      "mean_token_accuracy": 0.13694651499390603,
      "num_tokens": 21536497.0,
      "step": 9950
    },
    {
      "entropy": 6.335421895980835,
      "epoch": 1.0652790411471988,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004896625682568159,
      "loss": 6.1695,
      "mean_token_accuracy": 0.14211887270212173,
      "num_tokens": 21546213.0,
      "step": 9955
    },
    {
      "entropy": 6.342327451705932,
      "epoch": 1.0658141152549627,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004896511093421057,
      "loss": 6.1243,
      "mean_token_accuracy": 0.1506485767662525,
      "num_tokens": 21556990.0,
      "step": 9960
    },
    {
      "entropy": 6.31368842124939,
      "epoch": 1.0663491893627268,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004896396442293749,
      "loss": 6.1319,
      "mean_token_accuracy": 0.14206536412239074,
      "num_tokens": 21567567.0,
      "step": 9965
    },
    {
      "entropy": 6.35344386100769,
      "epoch": 1.0668842634704907,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004896281729189544,
      "loss": 6.1569,
      "mean_token_accuracy": 0.1394263654947281,
      "num_tokens": 21577457.0,
      "step": 9970
    },
    {
      "entropy": 6.328000497817993,
      "epoch": 1.0674193375782546,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004896166954111758,
      "loss": 6.1695,
      "mean_token_accuracy": 0.14929614067077637,
      "num_tokens": 21588173.0,
      "step": 9975
    },
    {
      "entropy": 6.347729063034057,
      "epoch": 1.0679544116860185,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048960521170637,
      "loss": 6.2038,
      "mean_token_accuracy": 0.1372479945421219,
      "num_tokens": 21599154.0,
      "step": 9980
    },
    {
      "entropy": 6.374818801879883,
      "epoch": 1.0684894857937826,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004895937218048692,
      "loss": 6.2421,
      "mean_token_accuracy": 0.1323709726333618,
      "num_tokens": 21610134.0,
      "step": 9985
    },
    {
      "entropy": 6.348971319198609,
      "epoch": 1.0690245599015464,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004895822257070046,
      "loss": 6.1124,
      "mean_token_accuracy": 0.14390253499150277,
      "num_tokens": 21619989.0,
      "step": 9990
    },
    {
      "entropy": 6.387176656723023,
      "epoch": 1.0695596340093103,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004895707234131084,
      "loss": 6.1727,
      "mean_token_accuracy": 0.14382659569382666,
      "num_tokens": 21632707.0,
      "step": 9995
    },
    {
      "entropy": 6.352945566177368,
      "epoch": 1.0700947081170742,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004895592149235128,
      "loss": 6.1391,
      "mean_token_accuracy": 0.14293183088302613,
      "num_tokens": 21643001.0,
      "step": 10000
    },
    {
      "entropy": 6.392285490036011,
      "epoch": 1.070629782224838,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048954770023855,
      "loss": 6.2209,
      "mean_token_accuracy": 0.13091161623597145,
      "num_tokens": 21653425.0,
      "step": 10005
    },
    {
      "entropy": 6.407132530212403,
      "epoch": 1.0711648563326022,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004895361793585525,
      "loss": 6.2457,
      "mean_token_accuracy": 0.1342737130820751,
      "num_tokens": 21664212.0,
      "step": 10010
    },
    {
      "entropy": 6.377192163467408,
      "epoch": 1.071699930440366,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004895246522838529,
      "loss": 6.221,
      "mean_token_accuracy": 0.14125536754727364,
      "num_tokens": 21675564.0,
      "step": 10015
    },
    {
      "entropy": 6.389852285385132,
      "epoch": 1.07223500454813,
      "grad_norm": 1.25,
      "learning_rate": 0.0004895131190147842,
      "loss": 6.2087,
      "mean_token_accuracy": 0.14097687974572182,
      "num_tokens": 21686225.0,
      "step": 10020
    },
    {
      "entropy": 6.35095796585083,
      "epoch": 1.0727700786558938,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004895015795516793,
      "loss": 6.1164,
      "mean_token_accuracy": 0.14123600497841834,
      "num_tokens": 21697479.0,
      "step": 10025
    },
    {
      "entropy": 6.383581018447876,
      "epoch": 1.0733051527636577,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004894900338948714,
      "loss": 6.2535,
      "mean_token_accuracy": 0.13488431945443152,
      "num_tokens": 21708496.0,
      "step": 10030
    },
    {
      "entropy": 6.403231906890869,
      "epoch": 1.0738402268714218,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004894784820446939,
      "loss": 6.1013,
      "mean_token_accuracy": 0.14549776092171668,
      "num_tokens": 21718290.0,
      "step": 10035
    },
    {
      "entropy": 6.334255504608154,
      "epoch": 1.0743753009791857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004894669240014804,
      "loss": 6.1237,
      "mean_token_accuracy": 0.14126087799668313,
      "num_tokens": 21729185.0,
      "step": 10040
    },
    {
      "entropy": 6.341536617279052,
      "epoch": 1.0749103750869495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004894553597655646,
      "loss": 6.2021,
      "mean_token_accuracy": 0.13568070307374,
      "num_tokens": 21740787.0,
      "step": 10045
    },
    {
      "entropy": 6.4046210765838625,
      "epoch": 1.0754454491947134,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004894437893372804,
      "loss": 6.2477,
      "mean_token_accuracy": 0.13528116419911385,
      "num_tokens": 21751804.0,
      "step": 10050
    },
    {
      "entropy": 6.425516891479492,
      "epoch": 1.0759805233024773,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000489432212716962,
      "loss": 6.1686,
      "mean_token_accuracy": 0.13829898089170456,
      "num_tokens": 21762730.0,
      "step": 10055
    },
    {
      "entropy": 6.353634595870972,
      "epoch": 1.0765155974102414,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004894206299049436,
      "loss": 6.2008,
      "mean_token_accuracy": 0.1432349555194378,
      "num_tokens": 21773470.0,
      "step": 10060
    },
    {
      "entropy": 6.4261232852935795,
      "epoch": 1.0770506715180053,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004894090409015595,
      "loss": 6.1969,
      "mean_token_accuracy": 0.1343509629368782,
      "num_tokens": 21783385.0,
      "step": 10065
    },
    {
      "entropy": 6.281267547607422,
      "epoch": 1.0775857456257691,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004893974457071445,
      "loss": 6.1999,
      "mean_token_accuracy": 0.13976912274956704,
      "num_tokens": 21794857.0,
      "step": 10070
    },
    {
      "entropy": 6.394709777832031,
      "epoch": 1.078120819733533,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004893858443220335,
      "loss": 6.2157,
      "mean_token_accuracy": 0.13416761085391044,
      "num_tokens": 21804296.0,
      "step": 10075
    },
    {
      "entropy": 6.403826713562012,
      "epoch": 1.0786558938412971,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004893742367465613,
      "loss": 6.1964,
      "mean_token_accuracy": 0.13898186087608339,
      "num_tokens": 21815899.0,
      "step": 10080
    },
    {
      "entropy": 6.376593065261841,
      "epoch": 1.079190967949061,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004893626229810631,
      "loss": 6.1865,
      "mean_token_accuracy": 0.14009021148085593,
      "num_tokens": 21825964.0,
      "step": 10085
    },
    {
      "entropy": 6.440527868270874,
      "epoch": 1.0797260420568249,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004893510030258743,
      "loss": 6.2778,
      "mean_token_accuracy": 0.13541262373328208,
      "num_tokens": 21836656.0,
      "step": 10090
    },
    {
      "entropy": 6.4241650104522705,
      "epoch": 1.0802611161645888,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004893393768813305,
      "loss": 6.2709,
      "mean_token_accuracy": 0.13260558396577835,
      "num_tokens": 21847658.0,
      "step": 10095
    },
    {
      "entropy": 6.425997495651245,
      "epoch": 1.0807961902723526,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004893277445477673,
      "loss": 6.2374,
      "mean_token_accuracy": 0.13200990930199624,
      "num_tokens": 21858338.0,
      "step": 10100
    },
    {
      "entropy": 6.373893117904663,
      "epoch": 1.0813312643801167,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004893161060255206,
      "loss": 6.1056,
      "mean_token_accuracy": 0.14343392625451087,
      "num_tokens": 21870656.0,
      "step": 10105
    },
    {
      "entropy": 6.400653600692749,
      "epoch": 1.0818663384878806,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004893044613149263,
      "loss": 6.2234,
      "mean_token_accuracy": 0.13905673176050187,
      "num_tokens": 21881806.0,
      "step": 10110
    },
    {
      "entropy": 6.266661739349365,
      "epoch": 1.0824014125956445,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004892928104163209,
      "loss": 6.129,
      "mean_token_accuracy": 0.14018828943371772,
      "num_tokens": 21892277.0,
      "step": 10115
    },
    {
      "entropy": 6.272743511199951,
      "epoch": 1.0829364867034084,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004892811533300407,
      "loss": 6.1277,
      "mean_token_accuracy": 0.14037966206669808,
      "num_tokens": 21903509.0,
      "step": 10120
    },
    {
      "entropy": 6.422238874435425,
      "epoch": 1.0834715608111725,
      "grad_norm": 1.25,
      "learning_rate": 0.0004892694900564223,
      "loss": 6.2395,
      "mean_token_accuracy": 0.13417447879910469,
      "num_tokens": 21914308.0,
      "step": 10125
    },
    {
      "entropy": 6.32716932296753,
      "epoch": 1.0840066349189363,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004892578205958025,
      "loss": 6.1357,
      "mean_token_accuracy": 0.14707659780979157,
      "num_tokens": 21924302.0,
      "step": 10130
    },
    {
      "entropy": 6.376426792144775,
      "epoch": 1.0845417090267002,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004892461449485182,
      "loss": 6.2129,
      "mean_token_accuracy": 0.13080279603600503,
      "num_tokens": 21935066.0,
      "step": 10135
    },
    {
      "entropy": 6.421293020248413,
      "epoch": 1.085076783134464,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004892344631149066,
      "loss": 6.1169,
      "mean_token_accuracy": 0.14315817952156068,
      "num_tokens": 21945798.0,
      "step": 10140
    },
    {
      "entropy": 6.341013669967651,
      "epoch": 1.085611857242228,
      "grad_norm": 1.203125,
      "learning_rate": 0.000489222775095305,
      "loss": 6.207,
      "mean_token_accuracy": 0.1371769927442074,
      "num_tokens": 21955667.0,
      "step": 10145
    },
    {
      "entropy": 6.346063232421875,
      "epoch": 1.086146931349992,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004892110808900509,
      "loss": 6.1792,
      "mean_token_accuracy": 0.13628155663609504,
      "num_tokens": 21966107.0,
      "step": 10150
    },
    {
      "entropy": 6.404660558700561,
      "epoch": 1.086682005457756,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004891993804994818,
      "loss": 6.1889,
      "mean_token_accuracy": 0.13703753426671028,
      "num_tokens": 21976408.0,
      "step": 10155
    },
    {
      "entropy": 6.484180116653443,
      "epoch": 1.0872170795655198,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004891876739239358,
      "loss": 6.1735,
      "mean_token_accuracy": 0.1416315495967865,
      "num_tokens": 21986481.0,
      "step": 10160
    },
    {
      "entropy": 6.388163089752197,
      "epoch": 1.0877521536732837,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004891759611637508,
      "loss": 6.2224,
      "mean_token_accuracy": 0.13773022666573526,
      "num_tokens": 21997554.0,
      "step": 10165
    },
    {
      "entropy": 6.424996471405029,
      "epoch": 1.0882872277810476,
      "grad_norm": 1.234375,
      "learning_rate": 0.000489164242219265,
      "loss": 6.2676,
      "mean_token_accuracy": 0.13574696481227874,
      "num_tokens": 22009048.0,
      "step": 10170
    },
    {
      "entropy": 6.431682825088501,
      "epoch": 1.0888223018888117,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004891525170908169,
      "loss": 6.2793,
      "mean_token_accuracy": 0.1367003306746483,
      "num_tokens": 22020625.0,
      "step": 10175
    },
    {
      "entropy": 6.416313743591308,
      "epoch": 1.0893573759965756,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004891407857787448,
      "loss": 6.1623,
      "mean_token_accuracy": 0.13500198274850844,
      "num_tokens": 22032298.0,
      "step": 10180
    },
    {
      "entropy": 6.308555650711059,
      "epoch": 1.0898924501043394,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004891290482833876,
      "loss": 6.1551,
      "mean_token_accuracy": 0.13973540365695952,
      "num_tokens": 22043693.0,
      "step": 10185
    },
    {
      "entropy": 6.35005784034729,
      "epoch": 1.0904275242121033,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004891173046050844,
      "loss": 6.2249,
      "mean_token_accuracy": 0.14575015977025033,
      "num_tokens": 22054373.0,
      "step": 10190
    },
    {
      "entropy": 6.367082834243774,
      "epoch": 1.0909625983198672,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000489105554744174,
      "loss": 6.1606,
      "mean_token_accuracy": 0.14085699766874313,
      "num_tokens": 22065039.0,
      "step": 10195
    },
    {
      "entropy": 6.387777853012085,
      "epoch": 1.0914976724276313,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004890937987009958,
      "loss": 6.2018,
      "mean_token_accuracy": 0.13782141134142875,
      "num_tokens": 22077003.0,
      "step": 10200
    },
    {
      "entropy": 6.414670610427857,
      "epoch": 1.0920327465353952,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004890820364758892,
      "loss": 6.2082,
      "mean_token_accuracy": 0.13570435345172882,
      "num_tokens": 22088558.0,
      "step": 10205
    },
    {
      "entropy": 6.440760755538941,
      "epoch": 1.092567820643159,
      "grad_norm": 1.28125,
      "learning_rate": 0.000489070268069194,
      "loss": 6.1847,
      "mean_token_accuracy": 0.13484859839081764,
      "num_tokens": 22097777.0,
      "step": 10210
    },
    {
      "entropy": 6.364702749252319,
      "epoch": 1.093102894750923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004890584934812498,
      "loss": 6.1045,
      "mean_token_accuracy": 0.14226649776101113,
      "num_tokens": 22107803.0,
      "step": 10215
    },
    {
      "entropy": 6.312130451202393,
      "epoch": 1.093637968858687,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004890467127123967,
      "loss": 6.1297,
      "mean_token_accuracy": 0.14537404477596283,
      "num_tokens": 22117965.0,
      "step": 10220
    },
    {
      "entropy": 6.386398553848267,
      "epoch": 1.094173042966451,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004890349257629749,
      "loss": 6.1691,
      "mean_token_accuracy": 0.14545267820358276,
      "num_tokens": 22127759.0,
      "step": 10225
    },
    {
      "entropy": 6.371078109741211,
      "epoch": 1.0947081170742148,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004890231326333246,
      "loss": 6.1968,
      "mean_token_accuracy": 0.13142700269818305,
      "num_tokens": 22138477.0,
      "step": 10230
    },
    {
      "entropy": 6.367229318618774,
      "epoch": 1.0952431911819787,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004890113333237865,
      "loss": 6.1861,
      "mean_token_accuracy": 0.137162484228611,
      "num_tokens": 22149568.0,
      "step": 10235
    },
    {
      "entropy": 6.374806261062622,
      "epoch": 1.0957782652897425,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004889995278347012,
      "loss": 6.1779,
      "mean_token_accuracy": 0.14030382931232452,
      "num_tokens": 22159949.0,
      "step": 10240
    },
    {
      "entropy": 6.372069215774536,
      "epoch": 1.0963133393975066,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004889877161664096,
      "loss": 6.1682,
      "mean_token_accuracy": 0.1367535598576069,
      "num_tokens": 22170115.0,
      "step": 10245
    },
    {
      "entropy": 6.399274826049805,
      "epoch": 1.0968484135052705,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004889758983192528,
      "loss": 6.2047,
      "mean_token_accuracy": 0.14183214604854583,
      "num_tokens": 22181323.0,
      "step": 10250
    },
    {
      "entropy": 6.309867000579834,
      "epoch": 1.0973834876130344,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004889640742935719,
      "loss": 6.1599,
      "mean_token_accuracy": 0.13460069224238397,
      "num_tokens": 22192078.0,
      "step": 10255
    },
    {
      "entropy": 6.373720407485962,
      "epoch": 1.0979185617207983,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004889522440897085,
      "loss": 6.2161,
      "mean_token_accuracy": 0.13826686143875122,
      "num_tokens": 22203211.0,
      "step": 10260
    },
    {
      "entropy": 6.401867628097534,
      "epoch": 1.0984536358285624,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004889404077080041,
      "loss": 6.1484,
      "mean_token_accuracy": 0.13527827635407447,
      "num_tokens": 22213487.0,
      "step": 10265
    },
    {
      "entropy": 6.409990358352661,
      "epoch": 1.0989887099363262,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004889285651488005,
      "loss": 6.1807,
      "mean_token_accuracy": 0.13432252705097197,
      "num_tokens": 22223717.0,
      "step": 10270
    },
    {
      "entropy": 6.323259782791138,
      "epoch": 1.0995237840440901,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004889167164124396,
      "loss": 6.1866,
      "mean_token_accuracy": 0.1422386184334755,
      "num_tokens": 22235141.0,
      "step": 10275
    },
    {
      "entropy": 6.373266172409058,
      "epoch": 1.100058858151854,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004889048614992636,
      "loss": 6.2441,
      "mean_token_accuracy": 0.13197602853178977,
      "num_tokens": 22246601.0,
      "step": 10280
    },
    {
      "entropy": 6.382775354385376,
      "epoch": 1.1005939322596179,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004888930004096148,
      "loss": 6.1814,
      "mean_token_accuracy": 0.1374364860355854,
      "num_tokens": 22257206.0,
      "step": 10285
    },
    {
      "entropy": 6.396820497512818,
      "epoch": 1.101129006367382,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004888811331438356,
      "loss": 6.25,
      "mean_token_accuracy": 0.13325197696685792,
      "num_tokens": 22267799.0,
      "step": 10290
    },
    {
      "entropy": 6.393010854721069,
      "epoch": 1.1016640804751459,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004888692597022689,
      "loss": 6.3044,
      "mean_token_accuracy": 0.13287948518991471,
      "num_tokens": 22279368.0,
      "step": 10295
    },
    {
      "entropy": 6.3397783756256105,
      "epoch": 1.1021991545829097,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004888573800852572,
      "loss": 6.0331,
      "mean_token_accuracy": 0.1460323452949524,
      "num_tokens": 22289998.0,
      "step": 10300
    },
    {
      "entropy": 6.3109893798828125,
      "epoch": 1.1027342286906736,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004888454942931438,
      "loss": 6.1639,
      "mean_token_accuracy": 0.14661996439099312,
      "num_tokens": 22300058.0,
      "step": 10305
    },
    {
      "entropy": 6.382072162628174,
      "epoch": 1.1032693027984375,
      "grad_norm": 1.375,
      "learning_rate": 0.0004888336023262718,
      "loss": 6.2647,
      "mean_token_accuracy": 0.13348389863967897,
      "num_tokens": 22310962.0,
      "step": 10310
    },
    {
      "entropy": 6.408809185028076,
      "epoch": 1.1038043769062016,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004888217041849846,
      "loss": 6.1371,
      "mean_token_accuracy": 0.14158515110611916,
      "num_tokens": 22321526.0,
      "step": 10315
    },
    {
      "entropy": 6.388979196548462,
      "epoch": 1.1043394510139655,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004888097998696256,
      "loss": 6.237,
      "mean_token_accuracy": 0.1399805538356304,
      "num_tokens": 22332558.0,
      "step": 10320
    },
    {
      "entropy": 6.379184818267822,
      "epoch": 1.1048745251217293,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004887978893805387,
      "loss": 6.1792,
      "mean_token_accuracy": 0.13994500935077667,
      "num_tokens": 22342392.0,
      "step": 10325
    },
    {
      "entropy": 6.375410890579223,
      "epoch": 1.1054095992294932,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004887859727180679,
      "loss": 6.2163,
      "mean_token_accuracy": 0.13442195802927018,
      "num_tokens": 22353089.0,
      "step": 10330
    },
    {
      "entropy": 6.335386848449707,
      "epoch": 1.105944673337257,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004887740498825572,
      "loss": 6.1445,
      "mean_token_accuracy": 0.14368956610560418,
      "num_tokens": 22364314.0,
      "step": 10335
    },
    {
      "entropy": 6.359664058685302,
      "epoch": 1.1064797474450212,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004887621208743507,
      "loss": 6.167,
      "mean_token_accuracy": 0.13805018216371537,
      "num_tokens": 22374985.0,
      "step": 10340
    },
    {
      "entropy": 6.347128677368164,
      "epoch": 1.107014821552785,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004887501856937932,
      "loss": 6.0662,
      "mean_token_accuracy": 0.15100962519645691,
      "num_tokens": 22385618.0,
      "step": 10345
    },
    {
      "entropy": 6.400017786026001,
      "epoch": 1.107549895660549,
      "grad_norm": 1.1875,
      "learning_rate": 0.000488738244341229,
      "loss": 6.1853,
      "mean_token_accuracy": 0.14385042637586593,
      "num_tokens": 22395871.0,
      "step": 10350
    },
    {
      "entropy": 6.3601720333099365,
      "epoch": 1.1080849697683128,
      "grad_norm": 1.125,
      "learning_rate": 0.000488726296817003,
      "loss": 6.1994,
      "mean_token_accuracy": 0.1419371522963047,
      "num_tokens": 22406707.0,
      "step": 10355
    },
    {
      "entropy": 6.371329355239868,
      "epoch": 1.108620043876077,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004887143431214602,
      "loss": 6.2033,
      "mean_token_accuracy": 0.13821865916252135,
      "num_tokens": 22417543.0,
      "step": 10360
    },
    {
      "entropy": 6.275408840179443,
      "epoch": 1.1091551179838408,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004887023832549459,
      "loss": 6.1548,
      "mean_token_accuracy": 0.14224184527993203,
      "num_tokens": 22428283.0,
      "step": 10365
    },
    {
      "entropy": 6.315039014816284,
      "epoch": 1.1096901920916047,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004886904172178052,
      "loss": 6.1302,
      "mean_token_accuracy": 0.14034454226493837,
      "num_tokens": 22439860.0,
      "step": 10370
    },
    {
      "entropy": 6.468148231506348,
      "epoch": 1.1102252661993686,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004886784450103838,
      "loss": 6.2007,
      "mean_token_accuracy": 0.14283432289958,
      "num_tokens": 22451250.0,
      "step": 10375
    },
    {
      "entropy": 6.365556764602661,
      "epoch": 1.1107603403071324,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004886664666330273,
      "loss": 6.141,
      "mean_token_accuracy": 0.14167733415961264,
      "num_tokens": 22461112.0,
      "step": 10380
    },
    {
      "entropy": 6.311558389663697,
      "epoch": 1.1112954144148965,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004886544820860816,
      "loss": 6.1707,
      "mean_token_accuracy": 0.1441367819905281,
      "num_tokens": 22471368.0,
      "step": 10385
    },
    {
      "entropy": 6.36181640625,
      "epoch": 1.1118304885226604,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004886424913698927,
      "loss": 6.1683,
      "mean_token_accuracy": 0.14117140769958497,
      "num_tokens": 22482011.0,
      "step": 10390
    },
    {
      "entropy": 6.379528427124024,
      "epoch": 1.1123655626304243,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004886304944848069,
      "loss": 6.1314,
      "mean_token_accuracy": 0.14957657605409622,
      "num_tokens": 22493648.0,
      "step": 10395
    },
    {
      "entropy": 6.4036760330200195,
      "epoch": 1.1129006367381882,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004886184914311706,
      "loss": 6.2438,
      "mean_token_accuracy": 0.13348260298371314,
      "num_tokens": 22505577.0,
      "step": 10400
    },
    {
      "entropy": 6.387641286849975,
      "epoch": 1.1134357108459523,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004886064822093305,
      "loss": 6.2057,
      "mean_token_accuracy": 0.13239317312836646,
      "num_tokens": 22515546.0,
      "step": 10405
    },
    {
      "entropy": 6.346754169464111,
      "epoch": 1.1139707849537162,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004885944668196332,
      "loss": 6.0957,
      "mean_token_accuracy": 0.14542415887117385,
      "num_tokens": 22526710.0,
      "step": 10410
    },
    {
      "entropy": 6.393271541595459,
      "epoch": 1.11450585906148,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004885824452624254,
      "loss": 6.2334,
      "mean_token_accuracy": 0.13556597679853438,
      "num_tokens": 22537809.0,
      "step": 10415
    },
    {
      "entropy": 6.453030395507812,
      "epoch": 1.115040933169244,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004885704175380548,
      "loss": 6.2048,
      "mean_token_accuracy": 0.13432461842894555,
      "num_tokens": 22549301.0,
      "step": 10420
    },
    {
      "entropy": 6.405787801742553,
      "epoch": 1.1155760072770078,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004885583836468683,
      "loss": 6.1083,
      "mean_token_accuracy": 0.1447925552725792,
      "num_tokens": 22560483.0,
      "step": 10425
    },
    {
      "entropy": 6.317955160140992,
      "epoch": 1.1161110813847719,
      "grad_norm": 1.25,
      "learning_rate": 0.0004885463435892136,
      "loss": 6.1892,
      "mean_token_accuracy": 0.13752674907445908,
      "num_tokens": 22570524.0,
      "step": 10430
    },
    {
      "entropy": 6.36132926940918,
      "epoch": 1.1166461554925358,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000488534297365438,
      "loss": 6.2249,
      "mean_token_accuracy": 0.13720738962292672,
      "num_tokens": 22581120.0,
      "step": 10435
    },
    {
      "entropy": 6.431585645675659,
      "epoch": 1.1171812296002996,
      "grad_norm": 1.25,
      "learning_rate": 0.0004885222449758899,
      "loss": 6.1931,
      "mean_token_accuracy": 0.1397352248430252,
      "num_tokens": 22592658.0,
      "step": 10440
    },
    {
      "entropy": 6.391028547286988,
      "epoch": 1.1177163037080635,
      "grad_norm": 1.25,
      "learning_rate": 0.0004885101864209167,
      "loss": 6.1752,
      "mean_token_accuracy": 0.13732331544160842,
      "num_tokens": 22603497.0,
      "step": 10445
    },
    {
      "entropy": 6.33019003868103,
      "epoch": 1.1182513778158274,
      "grad_norm": 1.265625,
      "learning_rate": 0.000488498121700867,
      "loss": 6.0841,
      "mean_token_accuracy": 0.14325874745845796,
      "num_tokens": 22612916.0,
      "step": 10450
    },
    {
      "entropy": 6.31720495223999,
      "epoch": 1.1187864519235915,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004884860508160891,
      "loss": 6.202,
      "mean_token_accuracy": 0.1370411865413189,
      "num_tokens": 22625345.0,
      "step": 10455
    },
    {
      "entropy": 6.3116796016693115,
      "epoch": 1.1193215260313554,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004884739737669314,
      "loss": 6.1177,
      "mean_token_accuracy": 0.1439550429582596,
      "num_tokens": 22637000.0,
      "step": 10460
    },
    {
      "entropy": 6.369991779327393,
      "epoch": 1.1198566001391193,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004884618905537427,
      "loss": 6.1221,
      "mean_token_accuracy": 0.14770162403583526,
      "num_tokens": 22647025.0,
      "step": 10465
    },
    {
      "entropy": 6.341109418869019,
      "epoch": 1.1203916742468831,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004884498011768719,
      "loss": 6.1301,
      "mean_token_accuracy": 0.14468767493963242,
      "num_tokens": 22657253.0,
      "step": 10470
    },
    {
      "entropy": 6.2405742645263675,
      "epoch": 1.120926748354647,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000488437705636668,
      "loss": 6.1439,
      "mean_token_accuracy": 0.14212991669774055,
      "num_tokens": 22668432.0,
      "step": 10475
    },
    {
      "entropy": 6.364749431610107,
      "epoch": 1.121461822462411,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004884256039334805,
      "loss": 6.19,
      "mean_token_accuracy": 0.13615185841917993,
      "num_tokens": 22678538.0,
      "step": 10480
    },
    {
      "entropy": 6.420324802398682,
      "epoch": 1.121996896570175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004884134960676586,
      "loss": 6.1707,
      "mean_token_accuracy": 0.1370847173035145,
      "num_tokens": 22688994.0,
      "step": 10485
    },
    {
      "entropy": 6.3969872951507565,
      "epoch": 1.1225319706779389,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000488401382039552,
      "loss": 6.2021,
      "mean_token_accuracy": 0.1388576991856098,
      "num_tokens": 22699584.0,
      "step": 10490
    },
    {
      "entropy": 6.323679399490357,
      "epoch": 1.1230670447857027,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004883892618495104,
      "loss": 6.1106,
      "mean_token_accuracy": 0.13969386443495752,
      "num_tokens": 22710093.0,
      "step": 10495
    },
    {
      "entropy": 6.3498148918151855,
      "epoch": 1.1236021188934668,
      "grad_norm": 1.125,
      "learning_rate": 0.000488377135497884,
      "loss": 6.1832,
      "mean_token_accuracy": 0.1415150910615921,
      "num_tokens": 22720533.0,
      "step": 10500
    },
    {
      "entropy": 6.317258882522583,
      "epoch": 1.1241371930012307,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004883650029850226,
      "loss": 6.0987,
      "mean_token_accuracy": 0.1412730872631073,
      "num_tokens": 22731010.0,
      "step": 10505
    },
    {
      "entropy": 6.28785891532898,
      "epoch": 1.1246722671089946,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004883528643112769,
      "loss": 6.1809,
      "mean_token_accuracy": 0.14027967751026155,
      "num_tokens": 22743045.0,
      "step": 10510
    },
    {
      "entropy": 6.3731495380401615,
      "epoch": 1.1252073412167585,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004883407194769972,
      "loss": 6.1713,
      "mean_token_accuracy": 0.1414507545530796,
      "num_tokens": 22753746.0,
      "step": 10515
    },
    {
      "entropy": 6.404212856292725,
      "epoch": 1.1257424153245223,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004883285684825342,
      "loss": 6.1944,
      "mean_token_accuracy": 0.14147845059633254,
      "num_tokens": 22765264.0,
      "step": 10520
    },
    {
      "entropy": 6.362782955169678,
      "epoch": 1.1262774894322864,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004883164113282386,
      "loss": 6.1816,
      "mean_token_accuracy": 0.14289152920246123,
      "num_tokens": 22775727.0,
      "step": 10525
    },
    {
      "entropy": 6.3755663394927975,
      "epoch": 1.1268125635400503,
      "grad_norm": 1.25,
      "learning_rate": 0.0004883042480144618,
      "loss": 6.1696,
      "mean_token_accuracy": 0.1390251860022545,
      "num_tokens": 22787228.0,
      "step": 10530
    },
    {
      "entropy": 6.489608430862427,
      "epoch": 1.1273476376478142,
      "grad_norm": 1.375,
      "learning_rate": 0.0004882920785415547,
      "loss": 6.2203,
      "mean_token_accuracy": 0.1374019704759121,
      "num_tokens": 22797467.0,
      "step": 10535
    },
    {
      "entropy": 6.406944513320923,
      "epoch": 1.127882711755578,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004882799029098689,
      "loss": 6.2064,
      "mean_token_accuracy": 0.13644992858171462,
      "num_tokens": 22808281.0,
      "step": 10540
    },
    {
      "entropy": 6.378452968597412,
      "epoch": 1.1284177858633422,
      "grad_norm": 1.5,
      "learning_rate": 0.00048826772111975583,
      "loss": 6.252,
      "mean_token_accuracy": 0.13718073591589927,
      "num_tokens": 22819387.0,
      "step": 10545
    },
    {
      "entropy": 6.330703258514404,
      "epoch": 1.128952859971106,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048825553317156717,
      "loss": 6.2114,
      "mean_token_accuracy": 0.13661018908023834,
      "num_tokens": 22829649.0,
      "step": 10550
    },
    {
      "entropy": 6.321175241470337,
      "epoch": 1.12948793407887,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004882433390656551,
      "loss": 6.1802,
      "mean_token_accuracy": 0.14422812312841415,
      "num_tokens": 22840667.0,
      "step": 10555
    },
    {
      "entropy": 6.348943853378296,
      "epoch": 1.1300230081866338,
      "grad_norm": 1.125,
      "learning_rate": 0.0004882311388023715,
      "loss": 6.1454,
      "mean_token_accuracy": 0.14260002374649047,
      "num_tokens": 22852461.0,
      "step": 10560
    },
    {
      "entropy": 6.404668140411377,
      "epoch": 1.1305580822943977,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004882189323820688,
      "loss": 6.2211,
      "mean_token_accuracy": 0.13300790563225745,
      "num_tokens": 22862781.0,
      "step": 10565
    },
    {
      "entropy": 6.454714250564575,
      "epoch": 1.1310931564021618,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048820671980509935,
      "loss": 6.2325,
      "mean_token_accuracy": 0.13897151872515678,
      "num_tokens": 22873393.0,
      "step": 10570
    },
    {
      "entropy": 6.447664785385132,
      "epoch": 1.1316282305099257,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048819450107181583,
      "loss": 6.2063,
      "mean_token_accuracy": 0.13727659061551095,
      "num_tokens": 22883931.0,
      "step": 10575
    },
    {
      "entropy": 6.379543876647949,
      "epoch": 1.1321633046176895,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004881822761825711,
      "loss": 6.1577,
      "mean_token_accuracy": 0.13911906704306604,
      "num_tokens": 22894829.0,
      "step": 10580
    },
    {
      "entropy": 6.311247110366821,
      "epoch": 1.1326983787254534,
      "grad_norm": 1.09375,
      "learning_rate": 0.000488170045137718,
      "loss": 6.1988,
      "mean_token_accuracy": 0.13939469531178475,
      "num_tokens": 22906059.0,
      "step": 10585
    },
    {
      "entropy": 6.395995712280273,
      "epoch": 1.1332334528332173,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004881578079376099,
      "loss": 6.187,
      "mean_token_accuracy": 0.13320620134472846,
      "num_tokens": 22917834.0,
      "step": 10590
    },
    {
      "entropy": 6.4128223896026615,
      "epoch": 1.1337685269409814,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048814556458259996,
      "loss": 6.0916,
      "mean_token_accuracy": 0.15180854499340057,
      "num_tokens": 22929157.0,
      "step": 10595
    },
    {
      "entropy": 6.438733291625977,
      "epoch": 1.1343036010487453,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004881333150730419,
      "loss": 6.2282,
      "mean_token_accuracy": 0.13689456060528754,
      "num_tokens": 22939396.0,
      "step": 10600
    },
    {
      "entropy": 6.338807535171509,
      "epoch": 1.1348386751565092,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048812105940928917,
      "loss": 6.2087,
      "mean_token_accuracy": 0.13827063664793968,
      "num_tokens": 22950253.0,
      "step": 10605
    },
    {
      "entropy": 6.386802864074707,
      "epoch": 1.135373749264273,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048810879759169593,
      "loss": 6.2051,
      "mean_token_accuracy": 0.1351565033197403,
      "num_tokens": 22960417.0,
      "step": 10610
    },
    {
      "entropy": 6.368847274780274,
      "epoch": 1.135908823372037,
      "grad_norm": 1.9140625,
      "learning_rate": 0.000488096529620616,
      "loss": 6.1892,
      "mean_token_accuracy": 0.13617856949567794,
      "num_tokens": 22971763.0,
      "step": 10615
    },
    {
      "entropy": 6.4445469856262205,
      "epoch": 1.136443897479801,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048808425549640383,
      "loss": 6.1934,
      "mean_token_accuracy": 0.14131694585084914,
      "num_tokens": 22981571.0,
      "step": 10620
    },
    {
      "entropy": 6.374335289001465,
      "epoch": 1.136978971587565,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048807197521941366,
      "loss": 6.1618,
      "mean_token_accuracy": 0.1432906024158001,
      "num_tokens": 22991644.0,
      "step": 10625
    },
    {
      "entropy": 6.323507499694824,
      "epoch": 1.1375140456953288,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004880596887900002,
      "loss": 6.0994,
      "mean_token_accuracy": 0.14439130648970605,
      "num_tokens": 23001557.0,
      "step": 10630
    },
    {
      "entropy": 6.399309968948364,
      "epoch": 1.1380491198030926,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048804739620851806,
      "loss": 6.1998,
      "mean_token_accuracy": 0.13571430668234824,
      "num_tokens": 23012715.0,
      "step": 10635
    },
    {
      "entropy": 6.405409431457519,
      "epoch": 1.1385841939108567,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048803509747532235,
      "loss": 6.1966,
      "mean_token_accuracy": 0.1414410263299942,
      "num_tokens": 23023016.0,
      "step": 10640
    },
    {
      "entropy": 6.344942474365235,
      "epoch": 1.1391192680186206,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004880227925907682,
      "loss": 6.1951,
      "mean_token_accuracy": 0.1372205026447773,
      "num_tokens": 23033511.0,
      "step": 10645
    },
    {
      "entropy": 6.417308902740478,
      "epoch": 1.1396543421263845,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004880104815552108,
      "loss": 6.2018,
      "mean_token_accuracy": 0.14261003509163855,
      "num_tokens": 23045149.0,
      "step": 10650
    },
    {
      "entropy": 6.337185525894165,
      "epoch": 1.1401894162341484,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004879981643690056,
      "loss": 6.146,
      "mean_token_accuracy": 0.14119565933942796,
      "num_tokens": 23054489.0,
      "step": 10655
    },
    {
      "entropy": 6.436928701400757,
      "epoch": 1.1407244903419125,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048798584103250847,
      "loss": 6.2863,
      "mean_token_accuracy": 0.1373275861144066,
      "num_tokens": 23064725.0,
      "step": 10660
    },
    {
      "entropy": 6.3601373672485355,
      "epoch": 1.1412595644496764,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000487973511546075,
      "loss": 6.2298,
      "mean_token_accuracy": 0.13637397661805153,
      "num_tokens": 23076473.0,
      "step": 10665
    },
    {
      "entropy": 6.402105712890625,
      "epoch": 1.1417946385574402,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004879611759100614,
      "loss": 6.1611,
      "mean_token_accuracy": 0.15056516230106354,
      "num_tokens": 23087449.0,
      "step": 10670
    },
    {
      "entropy": 6.436595630645752,
      "epoch": 1.1423297126652041,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004879488341248237,
      "loss": 6.1389,
      "mean_token_accuracy": 0.1409833699464798,
      "num_tokens": 23098284.0,
      "step": 10675
    },
    {
      "entropy": 6.326617670059204,
      "epoch": 1.142864786772968,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048793648619071834,
      "loss": 6.1371,
      "mean_token_accuracy": 0.14456309303641318,
      "num_tokens": 23108545.0,
      "step": 10680
    },
    {
      "entropy": 6.346481895446777,
      "epoch": 1.143399860880732,
      "grad_norm": 1.25,
      "learning_rate": 0.0004879241321081019,
      "loss": 6.1242,
      "mean_token_accuracy": 0.14047355428338051,
      "num_tokens": 23118568.0,
      "step": 10685
    },
    {
      "entropy": 6.404048681259155,
      "epoch": 1.143934934988496,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048791177187733104,
      "loss": 6.216,
      "mean_token_accuracy": 0.1373800665140152,
      "num_tokens": 23129189.0,
      "step": 10690
    },
    {
      "entropy": 6.4034381866455075,
      "epoch": 1.1444700090962598,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004878994054987627,
      "loss": 6.2213,
      "mean_token_accuracy": 0.13592005670070648,
      "num_tokens": 23140238.0,
      "step": 10695
    },
    {
      "entropy": 6.33728518486023,
      "epoch": 1.1450050832040237,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004878870329727539,
      "loss": 6.1161,
      "mean_token_accuracy": 0.14686858728528024,
      "num_tokens": 23149567.0,
      "step": 10700
    },
    {
      "entropy": 6.339387321472168,
      "epoch": 1.1455401573117876,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000487874654299662,
      "loss": 6.1067,
      "mean_token_accuracy": 0.14428818821907044,
      "num_tokens": 23159208.0,
      "step": 10705
    },
    {
      "entropy": 6.272930097579956,
      "epoch": 1.1460752314195517,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048786226947984435,
      "loss": 6.157,
      "mean_token_accuracy": 0.13887904733419418,
      "num_tokens": 23169092.0,
      "step": 10710
    },
    {
      "entropy": 6.355304718017578,
      "epoch": 1.1466103055273156,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004878498785136586,
      "loss": 6.1813,
      "mean_token_accuracy": 0.14076155573129653,
      "num_tokens": 23180385.0,
      "step": 10715
    },
    {
      "entropy": 6.428722381591797,
      "epoch": 1.1471453796350795,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048783748140146245,
      "loss": 6.1716,
      "mean_token_accuracy": 0.14089421555399895,
      "num_tokens": 23191180.0,
      "step": 10720
    },
    {
      "entropy": 6.392239904403686,
      "epoch": 1.1476804537428433,
      "grad_norm": 1.8125,
      "learning_rate": 0.000487825078143614,
      "loss": 6.1423,
      "mean_token_accuracy": 0.14577875509858132,
      "num_tokens": 23200920.0,
      "step": 10725
    },
    {
      "entropy": 6.344818162918091,
      "epoch": 1.1482155278506072,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004878126687404713,
      "loss": 6.1638,
      "mean_token_accuracy": 0.14055852517485617,
      "num_tokens": 23211140.0,
      "step": 10730
    },
    {
      "entropy": 6.3430173873901365,
      "epoch": 1.1487506019583713,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004878002531923927,
      "loss": 6.2087,
      "mean_token_accuracy": 0.1340080387890339,
      "num_tokens": 23222039.0,
      "step": 10735
    },
    {
      "entropy": 6.451751852035523,
      "epoch": 1.1492856760661352,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048778783149973674,
      "loss": 6.241,
      "mean_token_accuracy": 0.13416038751602172,
      "num_tokens": 23232502.0,
      "step": 10740
    },
    {
      "entropy": 6.425486660003662,
      "epoch": 1.149820750173899,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048777540366286195,
      "loss": 6.1641,
      "mean_token_accuracy": 0.1362004794180393,
      "num_tokens": 23243172.0,
      "step": 10745
    },
    {
      "entropy": 6.401501226425171,
      "epoch": 1.150355824281663,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004877629696821273,
      "loss": 6.2306,
      "mean_token_accuracy": 0.13752308562397958,
      "num_tokens": 23254558.0,
      "step": 10750
    },
    {
      "entropy": 6.428107786178589,
      "epoch": 1.1508908983894268,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048775052955789185,
      "loss": 6.2474,
      "mean_token_accuracy": 0.13754331693053246,
      "num_tokens": 23266283.0,
      "step": 10755
    },
    {
      "entropy": 6.32605209350586,
      "epoch": 1.151425972497191,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004877380832905147,
      "loss": 6.1164,
      "mean_token_accuracy": 0.14065871983766556,
      "num_tokens": 23277823.0,
      "step": 10760
    },
    {
      "entropy": 6.3388875961303714,
      "epoch": 1.1519610466049548,
      "grad_norm": 1.5,
      "learning_rate": 0.00048772563088035533,
      "loss": 6.1007,
      "mean_token_accuracy": 0.13963892459869384,
      "num_tokens": 23288159.0,
      "step": 10765
    },
    {
      "entropy": 6.326043319702149,
      "epoch": 1.1524961207127187,
      "grad_norm": 1.125,
      "learning_rate": 0.0004877131723277732,
      "loss": 6.132,
      "mean_token_accuracy": 0.13723269551992417,
      "num_tokens": 23299379.0,
      "step": 10770
    },
    {
      "entropy": 6.368733358383179,
      "epoch": 1.1530311948204826,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004877007076331282,
      "loss": 6.16,
      "mean_token_accuracy": 0.13965849354863166,
      "num_tokens": 23310572.0,
      "step": 10775
    },
    {
      "entropy": 6.360350275039673,
      "epoch": 1.1535662689282467,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004876882367967801,
      "loss": 6.158,
      "mean_token_accuracy": 0.14412407353520393,
      "num_tokens": 23320782.0,
      "step": 10780
    },
    {
      "entropy": 6.374213218688965,
      "epoch": 1.1541013430360105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048767575981908907,
      "loss": 6.2379,
      "mean_token_accuracy": 0.13344333544373513,
      "num_tokens": 23331427.0,
      "step": 10785
    },
    {
      "entropy": 6.42704668045044,
      "epoch": 1.1546364171437744,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048766327670041534,
      "loss": 6.3485,
      "mean_token_accuracy": 0.12839883118867873,
      "num_tokens": 23344250.0,
      "step": 10790
    },
    {
      "entropy": 6.437388610839844,
      "epoch": 1.1551714912515383,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004876507874411194,
      "loss": 6.19,
      "mean_token_accuracy": 0.14481368735432626,
      "num_tokens": 23354404.0,
      "step": 10795
    },
    {
      "entropy": 6.378191566467285,
      "epoch": 1.1557065653593024,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048763829204156187,
      "loss": 6.2007,
      "mean_token_accuracy": 0.13453842401504518,
      "num_tokens": 23364634.0,
      "step": 10800
    },
    {
      "entropy": 6.287075853347778,
      "epoch": 1.1562416394670663,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048762579050210344,
      "loss": 6.0716,
      "mean_token_accuracy": 0.145335054397583,
      "num_tokens": 23375816.0,
      "step": 10805
    },
    {
      "entropy": 6.398257064819336,
      "epoch": 1.1567767135748301,
      "grad_norm": 1.375,
      "learning_rate": 0.00048761328282310534,
      "loss": 6.2071,
      "mean_token_accuracy": 0.1408695861697197,
      "num_tokens": 23386583.0,
      "step": 10810
    },
    {
      "entropy": 6.338747644424439,
      "epoch": 1.157311787682594,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048760076900492846,
      "loss": 6.1367,
      "mean_token_accuracy": 0.14494462609291076,
      "num_tokens": 23396826.0,
      "step": 10815
    },
    {
      "entropy": 6.408685827255249,
      "epoch": 1.157846861790358,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004875882490479343,
      "loss": 6.2311,
      "mean_token_accuracy": 0.14126991704106331,
      "num_tokens": 23408661.0,
      "step": 10820
    },
    {
      "entropy": 6.358764171600342,
      "epoch": 1.158381935898122,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048757572295248425,
      "loss": 6.1254,
      "mean_token_accuracy": 0.14029526710510254,
      "num_tokens": 23419308.0,
      "step": 10825
    },
    {
      "entropy": 6.277323865890503,
      "epoch": 1.1589170100058859,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004875631907189402,
      "loss": 6.1169,
      "mean_token_accuracy": 0.1497894287109375,
      "num_tokens": 23430043.0,
      "step": 10830
    },
    {
      "entropy": 6.252225780487061,
      "epoch": 1.1594520841136498,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004875506523476638,
      "loss": 6.0619,
      "mean_token_accuracy": 0.14460479170084,
      "num_tokens": 23441246.0,
      "step": 10835
    },
    {
      "entropy": 6.3877557754516605,
      "epoch": 1.1599871582214136,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048753810783901716,
      "loss": 6.2046,
      "mean_token_accuracy": 0.13587984070181847,
      "num_tokens": 23452250.0,
      "step": 10840
    },
    {
      "entropy": 6.404697895050049,
      "epoch": 1.1605222323291775,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048752555719336257,
      "loss": 6.1798,
      "mean_token_accuracy": 0.13966346457600592,
      "num_tokens": 23464174.0,
      "step": 10845
    },
    {
      "entropy": 6.28532190322876,
      "epoch": 1.1610573064369416,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004875130004110623,
      "loss": 6.0517,
      "mean_token_accuracy": 0.14350549280643463,
      "num_tokens": 23475159.0,
      "step": 10850
    },
    {
      "entropy": 6.260467576980591,
      "epoch": 1.1615923805447055,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048750043749247907,
      "loss": 6.1262,
      "mean_token_accuracy": 0.14556582719087602,
      "num_tokens": 23486210.0,
      "step": 10855
    },
    {
      "entropy": 6.324217987060547,
      "epoch": 1.1621274546524694,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004874878684379756,
      "loss": 6.2344,
      "mean_token_accuracy": 0.13768360987305642,
      "num_tokens": 23497122.0,
      "step": 10860
    },
    {
      "entropy": 6.336201906204224,
      "epoch": 1.1626625287602332,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004874752932479148,
      "loss": 6.0709,
      "mean_token_accuracy": 0.14924161285161971,
      "num_tokens": 23507765.0,
      "step": 10865
    },
    {
      "entropy": 6.439267969131469,
      "epoch": 1.1631976028679971,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048746271192265974,
      "loss": 6.1703,
      "mean_token_accuracy": 0.1416929394006729,
      "num_tokens": 23517686.0,
      "step": 10870
    },
    {
      "entropy": 6.315063953399658,
      "epoch": 1.1637326769757612,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004874501244625737,
      "loss": 6.132,
      "mean_token_accuracy": 0.1390956722199917,
      "num_tokens": 23528325.0,
      "step": 10875
    },
    {
      "entropy": 6.338295936584473,
      "epoch": 1.164267751083525,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004874375308680202,
      "loss": 6.2015,
      "mean_token_accuracy": 0.13734544143080712,
      "num_tokens": 23540006.0,
      "step": 10880
    },
    {
      "entropy": 6.456075620651245,
      "epoch": 1.164802825191289,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048742493113936274,
      "loss": 6.1335,
      "mean_token_accuracy": 0.14085156917572023,
      "num_tokens": 23550419.0,
      "step": 10885
    },
    {
      "entropy": 6.408577251434326,
      "epoch": 1.1653378992990528,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004874123252769653,
      "loss": 6.2426,
      "mean_token_accuracy": 0.13463577032089233,
      "num_tokens": 23561227.0,
      "step": 10890
    },
    {
      "entropy": 6.353465795516968,
      "epoch": 1.1658729734068167,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048739971328119194,
      "loss": 6.1644,
      "mean_token_accuracy": 0.14023056775331497,
      "num_tokens": 23571357.0,
      "step": 10895
    },
    {
      "entropy": 6.386578035354614,
      "epoch": 1.1664080475145808,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004873870951524066,
      "loss": 6.1794,
      "mean_token_accuracy": 0.138648883998394,
      "num_tokens": 23581875.0,
      "step": 10900
    },
    {
      "entropy": 6.388548231124878,
      "epoch": 1.1669431216223447,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048737447089097373,
      "loss": 6.2553,
      "mean_token_accuracy": 0.13278844356536865,
      "num_tokens": 23594251.0,
      "step": 10905
    },
    {
      "entropy": 6.382857942581177,
      "epoch": 1.1674781957301086,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004873618404972579,
      "loss": 6.1901,
      "mean_token_accuracy": 0.13582184463739394,
      "num_tokens": 23605159.0,
      "step": 10910
    },
    {
      "entropy": 6.391416358947754,
      "epoch": 1.1680132698378725,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048734920397162376,
      "loss": 6.1328,
      "mean_token_accuracy": 0.14437243342399597,
      "num_tokens": 23615737.0,
      "step": 10915
    },
    {
      "entropy": 6.326221513748169,
      "epoch": 1.1685483439456366,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004873365613144361,
      "loss": 6.1579,
      "mean_token_accuracy": 0.13978007435798645,
      "num_tokens": 23626207.0,
      "step": 10920
    },
    {
      "entropy": 6.383629608154297,
      "epoch": 1.1690834180534004,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004873239125260602,
      "loss": 6.1951,
      "mean_token_accuracy": 0.1403285600244999,
      "num_tokens": 23637900.0,
      "step": 10925
    },
    {
      "entropy": 6.415868282318115,
      "epoch": 1.1696184921611643,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048731125760686106,
      "loss": 6.0972,
      "mean_token_accuracy": 0.14186388701200486,
      "num_tokens": 23647867.0,
      "step": 10930
    },
    {
      "entropy": 6.365462064743042,
      "epoch": 1.1701535662689282,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048729859655720424,
      "loss": 6.1596,
      "mean_token_accuracy": 0.14101602658629417,
      "num_tokens": 23658055.0,
      "step": 10935
    },
    {
      "entropy": 6.276657676696777,
      "epoch": 1.1706886403766923,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004872859293774553,
      "loss": 6.1607,
      "mean_token_accuracy": 0.13898070454597472,
      "num_tokens": 23669458.0,
      "step": 10940
    },
    {
      "entropy": 6.340875339508057,
      "epoch": 1.1712237144844562,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048727325606797995,
      "loss": 6.1657,
      "mean_token_accuracy": 0.1349417671561241,
      "num_tokens": 23679702.0,
      "step": 10945
    },
    {
      "entropy": 6.400650548934936,
      "epoch": 1.17175878859222,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004872605766291441,
      "loss": 6.1688,
      "mean_token_accuracy": 0.13870727643370628,
      "num_tokens": 23690844.0,
      "step": 10950
    },
    {
      "entropy": 6.3102741718292235,
      "epoch": 1.172293862699984,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000487247891061314,
      "loss": 6.138,
      "mean_token_accuracy": 0.13510648906230927,
      "num_tokens": 23701484.0,
      "step": 10955
    },
    {
      "entropy": 6.386873435974121,
      "epoch": 1.1728289368077478,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004872351993648559,
      "loss": 6.1711,
      "mean_token_accuracy": 0.14200300127267837,
      "num_tokens": 23711452.0,
      "step": 10960
    },
    {
      "entropy": 6.340666246414185,
      "epoch": 1.173364010915512,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048722250154013613,
      "loss": 6.1439,
      "mean_token_accuracy": 0.13472083434462548,
      "num_tokens": 23723514.0,
      "step": 10965
    },
    {
      "entropy": 6.386511135101318,
      "epoch": 1.1738990850232758,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004872097975875216,
      "loss": 6.1903,
      "mean_token_accuracy": 0.1397976741194725,
      "num_tokens": 23733722.0,
      "step": 10970
    },
    {
      "entropy": 6.409878921508789,
      "epoch": 1.1744341591310397,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004871970875073789,
      "loss": 6.1715,
      "mean_token_accuracy": 0.13753297701478004,
      "num_tokens": 23743403.0,
      "step": 10975
    },
    {
      "entropy": 6.359806919097901,
      "epoch": 1.1749692332388035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004871843713000751,
      "loss": 6.1341,
      "mean_token_accuracy": 0.14205780625343323,
      "num_tokens": 23754558.0,
      "step": 10980
    },
    {
      "entropy": 6.38535885810852,
      "epoch": 1.1755043073465674,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048717164896597737,
      "loss": 6.1962,
      "mean_token_accuracy": 0.13800126314163208,
      "num_tokens": 23766346.0,
      "step": 10985
    },
    {
      "entropy": 6.37688364982605,
      "epoch": 1.1760393814543315,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004871589205054532,
      "loss": 6.1794,
      "mean_token_accuracy": 0.1474433049559593,
      "num_tokens": 23776851.0,
      "step": 10990
    },
    {
      "entropy": 6.399209451675415,
      "epoch": 1.1765744555620954,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048714618591886997,
      "loss": 6.2305,
      "mean_token_accuracy": 0.13308942541480065,
      "num_tokens": 23787350.0,
      "step": 10995
    },
    {
      "entropy": 6.285842990875244,
      "epoch": 1.1771095296698593,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004871334452065954,
      "loss": 6.0938,
      "mean_token_accuracy": 0.1409935899078846,
      "num_tokens": 23797738.0,
      "step": 11000
    },
    {
      "entropy": 6.333844518661499,
      "epoch": 1.1776446037776231,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004871206983689974,
      "loss": 6.0939,
      "mean_token_accuracy": 0.1475593328475952,
      "num_tokens": 23809516.0,
      "step": 11005
    },
    {
      "entropy": 6.3549683570861815,
      "epoch": 1.178179677885387,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000487107945406444,
      "loss": 6.1674,
      "mean_token_accuracy": 0.14297233000397683,
      "num_tokens": 23819850.0,
      "step": 11010
    },
    {
      "entropy": 6.326850748062133,
      "epoch": 1.1787147519931511,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004870951863193036,
      "loss": 6.1623,
      "mean_token_accuracy": 0.14155297577381135,
      "num_tokens": 23829891.0,
      "step": 11015
    },
    {
      "entropy": 6.3073314189910885,
      "epoch": 1.179249826100915,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004870824211079444,
      "loss": 6.0938,
      "mean_token_accuracy": 0.14504248276352882,
      "num_tokens": 23840816.0,
      "step": 11020
    },
    {
      "entropy": 6.420108509063721,
      "epoch": 1.1797849002086789,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048706964977273515,
      "loss": 6.1978,
      "mean_token_accuracy": 0.1387620396912098,
      "num_tokens": 23852595.0,
      "step": 11025
    },
    {
      "entropy": 6.455548095703125,
      "epoch": 1.1803199743164428,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048705687231404455,
      "loss": 6.1951,
      "mean_token_accuracy": 0.13585883751511574,
      "num_tokens": 23863048.0,
      "step": 11030
    },
    {
      "entropy": 6.35600643157959,
      "epoch": 1.1808550484242066,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048704408873224156,
      "loss": 6.1619,
      "mean_token_accuracy": 0.1375263139605522,
      "num_tokens": 23873499.0,
      "step": 11035
    },
    {
      "entropy": 6.25940899848938,
      "epoch": 1.1813901225319707,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048703129902769526,
      "loss": 6.1029,
      "mean_token_accuracy": 0.14934418797492982,
      "num_tokens": 23885129.0,
      "step": 11040
    },
    {
      "entropy": 6.368331050872802,
      "epoch": 1.1819251966397346,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004870185032007751,
      "loss": 6.219,
      "mean_token_accuracy": 0.1341724745929241,
      "num_tokens": 23895538.0,
      "step": 11045
    },
    {
      "entropy": 6.379197692871093,
      "epoch": 1.1824602707474985,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004870057012518504,
      "loss": 6.2451,
      "mean_token_accuracy": 0.1327482908964157,
      "num_tokens": 23907341.0,
      "step": 11050
    },
    {
      "entropy": 6.335994005203247,
      "epoch": 1.1829953448552624,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048699289318129087,
      "loss": 6.0697,
      "mean_token_accuracy": 0.14457142129540443,
      "num_tokens": 23918501.0,
      "step": 11055
    },
    {
      "entropy": 6.380134153366089,
      "epoch": 1.1835304189630265,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004869800789894663,
      "loss": 6.1463,
      "mean_token_accuracy": 0.1378545306622982,
      "num_tokens": 23929204.0,
      "step": 11060
    },
    {
      "entropy": 6.333711338043213,
      "epoch": 1.1840654930707903,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004869672586767468,
      "loss": 6.186,
      "mean_token_accuracy": 0.14213462322950363,
      "num_tokens": 23939855.0,
      "step": 11065
    },
    {
      "entropy": 6.38549952507019,
      "epoch": 1.1846005671785542,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004869544322435024,
      "loss": 6.1262,
      "mean_token_accuracy": 0.1421860545873642,
      "num_tokens": 23949871.0,
      "step": 11070
    },
    {
      "entropy": 6.401708745956421,
      "epoch": 1.185135641286318,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004869415996901036,
      "loss": 6.2344,
      "mean_token_accuracy": 0.13710505664348602,
      "num_tokens": 23961650.0,
      "step": 11075
    },
    {
      "entropy": 6.406456041336059,
      "epoch": 1.1856707153940822,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004869287610169209,
      "loss": 6.2358,
      "mean_token_accuracy": 0.13767317086458206,
      "num_tokens": 23973103.0,
      "step": 11080
    },
    {
      "entropy": 6.3797437191009525,
      "epoch": 1.186205789501846,
      "grad_norm": 1.34375,
      "learning_rate": 0.000486915916224325,
      "loss": 6.1747,
      "mean_token_accuracy": 0.13768002316355704,
      "num_tokens": 23982972.0,
      "step": 11085
    },
    {
      "entropy": 6.27072491645813,
      "epoch": 1.18674086360961,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004869030653126868,
      "loss": 6.0268,
      "mean_token_accuracy": 0.14788830652832985,
      "num_tokens": 23993044.0,
      "step": 11090
    },
    {
      "entropy": 6.319503879547119,
      "epoch": 1.1872759377173738,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004868902082823774,
      "loss": 6.1275,
      "mean_token_accuracy": 0.1348693236708641,
      "num_tokens": 24003871.0,
      "step": 11095
    },
    {
      "entropy": 6.413490438461304,
      "epoch": 1.1878110118251377,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004868773451337679,
      "loss": 6.1826,
      "mean_token_accuracy": 0.13748824223876,
      "num_tokens": 24014613.0,
      "step": 11100
    },
    {
      "entropy": 6.36213059425354,
      "epoch": 1.1883460859329018,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048686447586722995,
      "loss": 6.1757,
      "mean_token_accuracy": 0.13895008340477943,
      "num_tokens": 24024172.0,
      "step": 11105
    },
    {
      "entropy": 6.33509669303894,
      "epoch": 1.1888811600406657,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048685160048313506,
      "loss": 6.2927,
      "mean_token_accuracy": 0.1348059020936489,
      "num_tokens": 24035301.0,
      "step": 11110
    },
    {
      "entropy": 6.415773057937622,
      "epoch": 1.1894162341484296,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048683871898185486,
      "loss": 6.1502,
      "mean_token_accuracy": 0.14853120297193528,
      "num_tokens": 24045661.0,
      "step": 11115
    },
    {
      "entropy": 6.339860630035401,
      "epoch": 1.1899513082561934,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048682583136376145,
      "loss": 6.1687,
      "mean_token_accuracy": 0.13792288824915885,
      "num_tokens": 24057027.0,
      "step": 11120
    },
    {
      "entropy": 6.381052112579345,
      "epoch": 1.1904863823639573,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004868129376292269,
      "loss": 6.2782,
      "mean_token_accuracy": 0.13411674723029138,
      "num_tokens": 24068322.0,
      "step": 11125
    },
    {
      "entropy": 6.412724161148072,
      "epoch": 1.1910214564717214,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048680003777862356,
      "loss": 6.09,
      "mean_token_accuracy": 0.1436728745698929,
      "num_tokens": 24078506.0,
      "step": 11130
    },
    {
      "entropy": 6.34953351020813,
      "epoch": 1.1915565305794853,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048678713181232394,
      "loss": 6.1266,
      "mean_token_accuracy": 0.1402457296848297,
      "num_tokens": 24090296.0,
      "step": 11135
    },
    {
      "entropy": 6.255198049545288,
      "epoch": 1.1920916046872492,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048677421973070063,
      "loss": 6.1481,
      "mean_token_accuracy": 0.13481734171509743,
      "num_tokens": 24101478.0,
      "step": 11140
    },
    {
      "entropy": 6.364823675155639,
      "epoch": 1.192626678795013,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048676130153412644,
      "loss": 6.1885,
      "mean_token_accuracy": 0.13936935663223265,
      "num_tokens": 24112552.0,
      "step": 11145
    },
    {
      "entropy": 6.341395139694214,
      "epoch": 1.193161752902777,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004867483772229745,
      "loss": 6.0682,
      "mean_token_accuracy": 0.1445258654654026,
      "num_tokens": 24124210.0,
      "step": 11150
    },
    {
      "entropy": 6.427348947525024,
      "epoch": 1.193696827010541,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004867354467976178,
      "loss": 6.1558,
      "mean_token_accuracy": 0.13990305960178376,
      "num_tokens": 24133525.0,
      "step": 11155
    },
    {
      "entropy": 6.34662766456604,
      "epoch": 1.194231901118305,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048672251025842994,
      "loss": 6.2138,
      "mean_token_accuracy": 0.13848276436328888,
      "num_tokens": 24145400.0,
      "step": 11160
    },
    {
      "entropy": 6.33854193687439,
      "epoch": 1.1947669752260688,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004867095676057843,
      "loss": 6.2106,
      "mean_token_accuracy": 0.13190727829933166,
      "num_tokens": 24155596.0,
      "step": 11165
    },
    {
      "entropy": 6.436871576309204,
      "epoch": 1.1953020493338327,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048669661884005467,
      "loss": 6.2125,
      "mean_token_accuracy": 0.13216416165232658,
      "num_tokens": 24166683.0,
      "step": 11170
    },
    {
      "entropy": 6.406686067581177,
      "epoch": 1.1958371234415965,
      "grad_norm": 1.375,
      "learning_rate": 0.00048668366396161503,
      "loss": 6.1368,
      "mean_token_accuracy": 0.14592012465000154,
      "num_tokens": 24177649.0,
      "step": 11175
    },
    {
      "entropy": 6.364505195617676,
      "epoch": 1.1963721975493606,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004866707029708392,
      "loss": 6.156,
      "mean_token_accuracy": 0.14543910771608354,
      "num_tokens": 24188522.0,
      "step": 11180
    },
    {
      "entropy": 6.374331331253051,
      "epoch": 1.1969072716571245,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004866577358681015,
      "loss": 6.1413,
      "mean_token_accuracy": 0.1412825718522072,
      "num_tokens": 24199285.0,
      "step": 11185
    },
    {
      "entropy": 6.311633634567261,
      "epoch": 1.1974423457648884,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004866447626537765,
      "loss": 6.1196,
      "mean_token_accuracy": 0.13971976637840272,
      "num_tokens": 24210929.0,
      "step": 11190
    },
    {
      "entropy": 6.308469486236572,
      "epoch": 1.1979774198726523,
      "grad_norm": 1.375,
      "learning_rate": 0.0004866317833282387,
      "loss": 6.1786,
      "mean_token_accuracy": 0.13834335282444954,
      "num_tokens": 24221394.0,
      "step": 11195
    },
    {
      "entropy": 6.388666009902954,
      "epoch": 1.1985124939804164,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048661879789186295,
      "loss": 6.1358,
      "mean_token_accuracy": 0.1437931962311268,
      "num_tokens": 24230297.0,
      "step": 11200
    },
    {
      "entropy": 6.324755764007568,
      "epoch": 1.1990475680881802,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048660580634502415,
      "loss": 6.284,
      "mean_token_accuracy": 0.1379542239010334,
      "num_tokens": 24241505.0,
      "step": 11205
    },
    {
      "entropy": 6.401583862304688,
      "epoch": 1.1995826421959441,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048659280868809744,
      "loss": 6.1447,
      "mean_token_accuracy": 0.1437310054898262,
      "num_tokens": 24252638.0,
      "step": 11210
    },
    {
      "entropy": 6.361992406845093,
      "epoch": 1.200117716303708,
      "grad_norm": 1.40625,
      "learning_rate": 0.000486579804921458,
      "loss": 6.1616,
      "mean_token_accuracy": 0.14515246748924254,
      "num_tokens": 24263156.0,
      "step": 11215
    },
    {
      "entropy": 6.274965429306031,
      "epoch": 1.200652790411472,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048656679504548145,
      "loss": 6.0478,
      "mean_token_accuracy": 0.14926579147577285,
      "num_tokens": 24274384.0,
      "step": 11220
    },
    {
      "entropy": 6.390253257751465,
      "epoch": 1.201187864519236,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004865537790605435,
      "loss": 6.2633,
      "mean_token_accuracy": 0.1277049206197262,
      "num_tokens": 24285159.0,
      "step": 11225
    },
    {
      "entropy": 6.4164785861969,
      "epoch": 1.2017229386269999,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004865407569670198,
      "loss": 6.2142,
      "mean_token_accuracy": 0.1367713987827301,
      "num_tokens": 24296202.0,
      "step": 11230
    },
    {
      "entropy": 6.387192916870117,
      "epoch": 1.2022580127347637,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004865277287652866,
      "loss": 6.1405,
      "mean_token_accuracy": 0.14048785269260405,
      "num_tokens": 24306691.0,
      "step": 11235
    },
    {
      "entropy": 6.336164760589599,
      "epoch": 1.2027930868425276,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004865146944557199,
      "loss": 6.1811,
      "mean_token_accuracy": 0.14225658327341079,
      "num_tokens": 24317499.0,
      "step": 11240
    },
    {
      "entropy": 6.378962993621826,
      "epoch": 1.2033281609502917,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004865016540386961,
      "loss": 6.1529,
      "mean_token_accuracy": 0.13606233894824982,
      "num_tokens": 24328096.0,
      "step": 11245
    },
    {
      "entropy": 6.376345825195313,
      "epoch": 1.2038632350580556,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048648860751459173,
      "loss": 6.1548,
      "mean_token_accuracy": 0.13841187357902526,
      "num_tokens": 24338188.0,
      "step": 11250
    },
    {
      "entropy": 6.38016562461853,
      "epoch": 1.2043983091658195,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048647555488378355,
      "loss": 6.1869,
      "mean_token_accuracy": 0.14600923359394075,
      "num_tokens": 24348578.0,
      "step": 11255
    },
    {
      "entropy": 6.2799492359161375,
      "epoch": 1.2049333832735833,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004864624961466485,
      "loss": 6.1553,
      "mean_token_accuracy": 0.1442287214100361,
      "num_tokens": 24358807.0,
      "step": 11260
    },
    {
      "entropy": 6.364462375640869,
      "epoch": 1.2054684573813472,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004864494313035635,
      "loss": 6.1596,
      "mean_token_accuracy": 0.13837054893374442,
      "num_tokens": 24369074.0,
      "step": 11265
    },
    {
      "entropy": 6.444828891754151,
      "epoch": 1.2060035314891113,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000486436360354906,
      "loss": 6.1441,
      "mean_token_accuracy": 0.1425137497484684,
      "num_tokens": 24380655.0,
      "step": 11270
    },
    {
      "entropy": 6.330973720550537,
      "epoch": 1.2065386055968752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048642328330105323,
      "loss": 6.1245,
      "mean_token_accuracy": 0.13961319625377655,
      "num_tokens": 24391701.0,
      "step": 11275
    },
    {
      "entropy": 6.26389102935791,
      "epoch": 1.207073679704639,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048641020014238286,
      "loss": 6.2089,
      "mean_token_accuracy": 0.1385771282017231,
      "num_tokens": 24403690.0,
      "step": 11280
    },
    {
      "entropy": 6.436042785644531,
      "epoch": 1.207608753812403,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004863971108792727,
      "loss": 6.1938,
      "mean_token_accuracy": 0.14043005034327508,
      "num_tokens": 24414670.0,
      "step": 11285
    },
    {
      "entropy": 6.414697694778442,
      "epoch": 1.2081438279201668,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048638401551210063,
      "loss": 6.2483,
      "mean_token_accuracy": 0.13419551476836206,
      "num_tokens": 24425237.0,
      "step": 11290
    },
    {
      "entropy": 6.397723865509033,
      "epoch": 1.208678902027931,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048637091404124484,
      "loss": 6.1949,
      "mean_token_accuracy": 0.13871749341487885,
      "num_tokens": 24437142.0,
      "step": 11295
    },
    {
      "entropy": 6.329138326644897,
      "epoch": 1.2092139761356948,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004863578064670837,
      "loss": 6.0996,
      "mean_token_accuracy": 0.1459878481924534,
      "num_tokens": 24448819.0,
      "step": 11300
    },
    {
      "entropy": 6.254271411895752,
      "epoch": 1.2097490502434587,
      "grad_norm": 1.125,
      "learning_rate": 0.00048634469278999545,
      "loss": 6.0501,
      "mean_token_accuracy": 0.15032576024532318,
      "num_tokens": 24459392.0,
      "step": 11305
    },
    {
      "entropy": 6.402129650115967,
      "epoch": 1.2102841243512226,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048633157301035895,
      "loss": 6.1325,
      "mean_token_accuracy": 0.13897140920162201,
      "num_tokens": 24470298.0,
      "step": 11310
    },
    {
      "entropy": 6.362146711349487,
      "epoch": 1.2108191984589864,
      "grad_norm": 1.234375,
      "learning_rate": 0.000486318447128553,
      "loss": 6.1257,
      "mean_token_accuracy": 0.14351900815963745,
      "num_tokens": 24480477.0,
      "step": 11315
    },
    {
      "entropy": 6.3257365226745605,
      "epoch": 1.2113542725667505,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004863053151449566,
      "loss": 6.2063,
      "mean_token_accuracy": 0.12840102463960648,
      "num_tokens": 24491890.0,
      "step": 11320
    },
    {
      "entropy": 6.38619475364685,
      "epoch": 1.2118893466745144,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048629217705994883,
      "loss": 6.1389,
      "mean_token_accuracy": 0.1378464877605438,
      "num_tokens": 24502583.0,
      "step": 11325
    },
    {
      "entropy": 6.428861951828003,
      "epoch": 1.2124244207822783,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004862790328739091,
      "loss": 6.245,
      "mean_token_accuracy": 0.13337404504418374,
      "num_tokens": 24514491.0,
      "step": 11330
    },
    {
      "entropy": 6.308004093170166,
      "epoch": 1.2129594948900422,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048626588258721717,
      "loss": 6.077,
      "mean_token_accuracy": 0.14916639849543573,
      "num_tokens": 24524788.0,
      "step": 11335
    },
    {
      "entropy": 6.284106159210205,
      "epoch": 1.2134945689978063,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004862527262002524,
      "loss": 6.1201,
      "mean_token_accuracy": 0.13238025084137917,
      "num_tokens": 24535699.0,
      "step": 11340
    },
    {
      "entropy": 6.323851156234741,
      "epoch": 1.2140296431055702,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048623956371339494,
      "loss": 6.1256,
      "mean_token_accuracy": 0.14018687307834626,
      "num_tokens": 24547339.0,
      "step": 11345
    },
    {
      "entropy": 6.375518846511841,
      "epoch": 1.214564717213334,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048622639512702477,
      "loss": 6.2417,
      "mean_token_accuracy": 0.13524217009544373,
      "num_tokens": 24558778.0,
      "step": 11350
    },
    {
      "entropy": 6.3299530982971195,
      "epoch": 1.215099791321098,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048621322044152207,
      "loss": 6.153,
      "mean_token_accuracy": 0.1406257189810276,
      "num_tokens": 24569032.0,
      "step": 11355
    },
    {
      "entropy": 6.367442035675049,
      "epoch": 1.215634865428862,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048620003965726727,
      "loss": 6.2583,
      "mean_token_accuracy": 0.1269288919866085,
      "num_tokens": 24581356.0,
      "step": 11360
    },
    {
      "entropy": 6.361095952987671,
      "epoch": 1.2161699395366259,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048618685277464116,
      "loss": 6.1161,
      "mean_token_accuracy": 0.14307422712445259,
      "num_tokens": 24591809.0,
      "step": 11365
    },
    {
      "entropy": 6.231331396102905,
      "epoch": 1.2167050136443898,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048617365979402417,
      "loss": 6.0387,
      "mean_token_accuracy": 0.14527350813150405,
      "num_tokens": 24603492.0,
      "step": 11370
    },
    {
      "entropy": 6.36796498298645,
      "epoch": 1.2172400877521536,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004861604607157975,
      "loss": 6.2073,
      "mean_token_accuracy": 0.137881401181221,
      "num_tokens": 24614837.0,
      "step": 11375
    },
    {
      "entropy": 6.409143543243408,
      "epoch": 1.2177751618599175,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004861472555403423,
      "loss": 6.1838,
      "mean_token_accuracy": 0.13644215390086173,
      "num_tokens": 24624714.0,
      "step": 11380
    },
    {
      "entropy": 6.387162780761718,
      "epoch": 1.2183102359676816,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048613404426803964,
      "loss": 6.2265,
      "mean_token_accuracy": 0.13210561498999596,
      "num_tokens": 24635758.0,
      "step": 11385
    },
    {
      "entropy": 6.4058421611785885,
      "epoch": 1.2188453100754455,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048612082689927116,
      "loss": 6.1928,
      "mean_token_accuracy": 0.14500827044248582,
      "num_tokens": 24645362.0,
      "step": 11390
    },
    {
      "entropy": 6.371025896072387,
      "epoch": 1.2193803841832094,
      "grad_norm": 1.125,
      "learning_rate": 0.0004861076034344185,
      "loss": 6.2105,
      "mean_token_accuracy": 0.1370255619287491,
      "num_tokens": 24656898.0,
      "step": 11395
    },
    {
      "entropy": 6.338381671905518,
      "epoch": 1.2199154582909733,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004860943738738634,
      "loss": 6.1436,
      "mean_token_accuracy": 0.14201639890670775,
      "num_tokens": 24668191.0,
      "step": 11400
    },
    {
      "entropy": 6.45799446105957,
      "epoch": 1.2204505323987371,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048608113821798786,
      "loss": 6.1561,
      "mean_token_accuracy": 0.14518523216247559,
      "num_tokens": 24678034.0,
      "step": 11405
    },
    {
      "entropy": 6.3682787895202635,
      "epoch": 1.2209856065065012,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004860678964671743,
      "loss": 6.1962,
      "mean_token_accuracy": 0.1416761003434658,
      "num_tokens": 24689062.0,
      "step": 11410
    },
    {
      "entropy": 6.338401985168457,
      "epoch": 1.221520680614265,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048605464862180474,
      "loss": 6.1725,
      "mean_token_accuracy": 0.13547325134277344,
      "num_tokens": 24699787.0,
      "step": 11415
    },
    {
      "entropy": 6.379491090774536,
      "epoch": 1.222055754722029,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004860413946822619,
      "loss": 6.1079,
      "mean_token_accuracy": 0.14030019342899322,
      "num_tokens": 24711233.0,
      "step": 11420
    },
    {
      "entropy": 6.336051940917969,
      "epoch": 1.2225908288297929,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004860281346489284,
      "loss": 6.1575,
      "mean_token_accuracy": 0.13204106837511062,
      "num_tokens": 24722414.0,
      "step": 11425
    },
    {
      "entropy": 6.38946533203125,
      "epoch": 1.2231259029375567,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004860148685221873,
      "loss": 6.1756,
      "mean_token_accuracy": 0.134771990776062,
      "num_tokens": 24733093.0,
      "step": 11430
    },
    {
      "entropy": 6.37145504951477,
      "epoch": 1.2236609770453208,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048600159630242135,
      "loss": 6.1292,
      "mean_token_accuracy": 0.14352304115891457,
      "num_tokens": 24743195.0,
      "step": 11435
    },
    {
      "entropy": 6.398423337936402,
      "epoch": 1.2241960511530847,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048598831799001406,
      "loss": 6.1572,
      "mean_token_accuracy": 0.1362826183438301,
      "num_tokens": 24753680.0,
      "step": 11440
    },
    {
      "entropy": 6.3395342350006105,
      "epoch": 1.2247311252608486,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004859750335853488,
      "loss": 6.1017,
      "mean_token_accuracy": 0.1428315296769142,
      "num_tokens": 24764135.0,
      "step": 11445
    },
    {
      "entropy": 6.23157958984375,
      "epoch": 1.2252661993686125,
      "grad_norm": 1.359375,
      "learning_rate": 0.000485961743088809,
      "loss": 6.0739,
      "mean_token_accuracy": 0.15708449184894563,
      "num_tokens": 24775740.0,
      "step": 11450
    },
    {
      "entropy": 6.392571067810058,
      "epoch": 1.2258012734763766,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004859484465007784,
      "loss": 6.211,
      "mean_token_accuracy": 0.13890906274318696,
      "num_tokens": 24786169.0,
      "step": 11455
    },
    {
      "entropy": 6.383392858505249,
      "epoch": 1.2263363475841405,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048593514382164123,
      "loss": 6.0936,
      "mean_token_accuracy": 0.14535240828990936,
      "num_tokens": 24796218.0,
      "step": 11460
    },
    {
      "entropy": 6.3617833137512205,
      "epoch": 1.2268714216919043,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004859218350517814,
      "loss": 6.2413,
      "mean_token_accuracy": 0.13647545650601386,
      "num_tokens": 24806957.0,
      "step": 11465
    },
    {
      "entropy": 6.380275630950928,
      "epoch": 1.2274064957996682,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004859085201915831,
      "loss": 6.1617,
      "mean_token_accuracy": 0.1416012167930603,
      "num_tokens": 24818761.0,
      "step": 11470
    },
    {
      "entropy": 6.312294960021973,
      "epoch": 1.227941569907432,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004858951992414311,
      "loss": 6.0856,
      "mean_token_accuracy": 0.14966220557689666,
      "num_tokens": 24829163.0,
      "step": 11475
    },
    {
      "entropy": 6.282326555252075,
      "epoch": 1.2284766440151962,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004858818722017097,
      "loss": 6.1073,
      "mean_token_accuracy": 0.14206337034702302,
      "num_tokens": 24839334.0,
      "step": 11480
    },
    {
      "entropy": 6.30363245010376,
      "epoch": 1.22901171812296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048586853907280395,
      "loss": 6.1322,
      "mean_token_accuracy": 0.14936811923980714,
      "num_tokens": 24850312.0,
      "step": 11485
    },
    {
      "entropy": 6.351392412185669,
      "epoch": 1.229546792230724,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004858551998550987,
      "loss": 6.1108,
      "mean_token_accuracy": 0.1508676141500473,
      "num_tokens": 24861835.0,
      "step": 11490
    },
    {
      "entropy": 6.396349668502808,
      "epoch": 1.2300818663384878,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048584185454897925,
      "loss": 6.1686,
      "mean_token_accuracy": 0.1430028885602951,
      "num_tokens": 24872489.0,
      "step": 11495
    },
    {
      "entropy": 6.318239021301269,
      "epoch": 1.230616940446252,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004858285031548309,
      "loss": 6.1624,
      "mean_token_accuracy": 0.14118192940950394,
      "num_tokens": 24882553.0,
      "step": 11500
    },
    {
      "entropy": 6.368698215484619,
      "epoch": 1.2311520145540158,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004858151456730391,
      "loss": 6.1855,
      "mean_token_accuracy": 0.1426655814051628,
      "num_tokens": 24894386.0,
      "step": 11505
    },
    {
      "entropy": 6.325570678710937,
      "epoch": 1.2316870886617797,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048580178210398954,
      "loss": 6.1589,
      "mean_token_accuracy": 0.13970115184783935,
      "num_tokens": 24905049.0,
      "step": 11510
    },
    {
      "entropy": 6.305087995529175,
      "epoch": 1.2322221627695435,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048578841244806823,
      "loss": 6.1252,
      "mean_token_accuracy": 0.14457620978355407,
      "num_tokens": 24914831.0,
      "step": 11515
    },
    {
      "entropy": 6.3696287155151365,
      "epoch": 1.2327572368773074,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048577503670566116,
      "loss": 6.1876,
      "mean_token_accuracy": 0.1407425045967102,
      "num_tokens": 24925668.0,
      "step": 11520
    },
    {
      "entropy": 6.398597288131714,
      "epoch": 1.2332923109850715,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048576165487715443,
      "loss": 6.1403,
      "mean_token_accuracy": 0.14060246720910072,
      "num_tokens": 24935983.0,
      "step": 11525
    },
    {
      "entropy": 6.310181140899658,
      "epoch": 1.2338273850928354,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048574826696293455,
      "loss": 6.1139,
      "mean_token_accuracy": 0.14657812491059302,
      "num_tokens": 24946673.0,
      "step": 11530
    },
    {
      "entropy": 6.304920959472656,
      "epoch": 1.2343624592005993,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048573487296338814,
      "loss": 6.0517,
      "mean_token_accuracy": 0.15180279985070227,
      "num_tokens": 24957075.0,
      "step": 11535
    },
    {
      "entropy": 6.378532123565674,
      "epoch": 1.2348975333083632,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048572147287890183,
      "loss": 6.2114,
      "mean_token_accuracy": 0.13791173473000526,
      "num_tokens": 24967261.0,
      "step": 11540
    },
    {
      "entropy": 6.3163422584533695,
      "epoch": 1.235432607416127,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048570806670986263,
      "loss": 6.1405,
      "mean_token_accuracy": 0.1422004833817482,
      "num_tokens": 24978097.0,
      "step": 11545
    },
    {
      "entropy": 6.368444776535034,
      "epoch": 1.2359676815238911,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048569465445665754,
      "loss": 6.1583,
      "mean_token_accuracy": 0.14133872613310813,
      "num_tokens": 24987932.0,
      "step": 11550
    },
    {
      "entropy": 6.353887414932251,
      "epoch": 1.236502755631655,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048568123611967396,
      "loss": 6.1485,
      "mean_token_accuracy": 0.13913731053471565,
      "num_tokens": 24998919.0,
      "step": 11555
    },
    {
      "entropy": 6.333116960525513,
      "epoch": 1.237037829739419,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004856678116992993,
      "loss": 6.1147,
      "mean_token_accuracy": 0.14224686920642854,
      "num_tokens": 25010545.0,
      "step": 11560
    },
    {
      "entropy": 6.337672567367553,
      "epoch": 1.2375729038471828,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004856543811959212,
      "loss": 6.0895,
      "mean_token_accuracy": 0.1547150045633316,
      "num_tokens": 25021133.0,
      "step": 11565
    },
    {
      "entropy": 6.290955972671509,
      "epoch": 1.2381079779549466,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004856409446099274,
      "loss": 6.1191,
      "mean_token_accuracy": 0.14032425731420517,
      "num_tokens": 25031838.0,
      "step": 11570
    },
    {
      "entropy": 6.329652309417725,
      "epoch": 1.2386430520627107,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048562750194170595,
      "loss": 6.1528,
      "mean_token_accuracy": 0.1390583410859108,
      "num_tokens": 25042737.0,
      "step": 11575
    },
    {
      "entropy": 6.420797395706177,
      "epoch": 1.2391781261704746,
      "grad_norm": 1.203125,
      "learning_rate": 0.000485614053191645,
      "loss": 6.2484,
      "mean_token_accuracy": 0.1368211105465889,
      "num_tokens": 25054442.0,
      "step": 11580
    },
    {
      "entropy": 6.281362533569336,
      "epoch": 1.2397132002782385,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004856005983601328,
      "loss": 5.9884,
      "mean_token_accuracy": 0.1529831327497959,
      "num_tokens": 25065509.0,
      "step": 11585
    },
    {
      "entropy": 6.388730764389038,
      "epoch": 1.2402482743860024,
      "grad_norm": 1.25,
      "learning_rate": 0.00048558713744755794,
      "loss": 6.2325,
      "mean_token_accuracy": 0.13230527117848395,
      "num_tokens": 25077306.0,
      "step": 11590
    },
    {
      "entropy": 6.369465923309326,
      "epoch": 1.2407833484937665,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004855736704543091,
      "loss": 6.1615,
      "mean_token_accuracy": 0.1445845365524292,
      "num_tokens": 25088230.0,
      "step": 11595
    },
    {
      "entropy": 6.447409963607788,
      "epoch": 1.2413184226015304,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004855601973807751,
      "loss": 6.2504,
      "mean_token_accuracy": 0.13314316421747208,
      "num_tokens": 25099658.0,
      "step": 11600
    },
    {
      "entropy": 6.3497272491455075,
      "epoch": 1.2418534967092942,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048554671822734495,
      "loss": 6.1378,
      "mean_token_accuracy": 0.1402084857225418,
      "num_tokens": 25110512.0,
      "step": 11605
    },
    {
      "entropy": 6.319833469390869,
      "epoch": 1.2423885708170581,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048553323299440807,
      "loss": 6.1474,
      "mean_token_accuracy": 0.14169399067759514,
      "num_tokens": 25120193.0,
      "step": 11610
    },
    {
      "entropy": 6.345179319381714,
      "epoch": 1.242923644924822,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048551974168235346,
      "loss": 6.1567,
      "mean_token_accuracy": 0.13806528598070145,
      "num_tokens": 25131752.0,
      "step": 11615
    },
    {
      "entropy": 6.403358840942383,
      "epoch": 1.243458719032586,
      "grad_norm": 1.25,
      "learning_rate": 0.000485506244291571,
      "loss": 6.1765,
      "mean_token_accuracy": 0.14041747599840165,
      "num_tokens": 25141541.0,
      "step": 11620
    },
    {
      "entropy": 6.320514965057373,
      "epoch": 1.24399379314035,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004854927408224503,
      "loss": 6.1483,
      "mean_token_accuracy": 0.1382579691708088,
      "num_tokens": 25152219.0,
      "step": 11625
    },
    {
      "entropy": 6.433775472640991,
      "epoch": 1.2445288672481138,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048547923127538126,
      "loss": 6.1669,
      "mean_token_accuracy": 0.14160683676600455,
      "num_tokens": 25163030.0,
      "step": 11630
    },
    {
      "entropy": 6.341416358947754,
      "epoch": 1.2450639413558777,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048546571565075396,
      "loss": 6.049,
      "mean_token_accuracy": 0.14477123320102692,
      "num_tokens": 25174637.0,
      "step": 11635
    },
    {
      "entropy": 6.270388650894165,
      "epoch": 1.2455990154636418,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004854521939489587,
      "loss": 6.0459,
      "mean_token_accuracy": 0.14917651638388635,
      "num_tokens": 25185067.0,
      "step": 11640
    },
    {
      "entropy": 6.316221618652344,
      "epoch": 1.2461340895714057,
      "grad_norm": 1.25,
      "learning_rate": 0.00048543866617038604,
      "loss": 6.0851,
      "mean_token_accuracy": 0.13966694995760917,
      "num_tokens": 25195046.0,
      "step": 11645
    },
    {
      "entropy": 6.325966310501099,
      "epoch": 1.2466691636791696,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048542513231542635,
      "loss": 6.1778,
      "mean_token_accuracy": 0.13401782736182213,
      "num_tokens": 25206265.0,
      "step": 11650
    },
    {
      "entropy": 6.400760650634766,
      "epoch": 1.2472042377869335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004854115923844705,
      "loss": 6.1803,
      "mean_token_accuracy": 0.13182243779301644,
      "num_tokens": 25217690.0,
      "step": 11655
    },
    {
      "entropy": 6.409377861022949,
      "epoch": 1.2477393118946973,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004853980463779096,
      "loss": 6.1623,
      "mean_token_accuracy": 0.13325203731656074,
      "num_tokens": 25228229.0,
      "step": 11660
    },
    {
      "entropy": 6.305456018447876,
      "epoch": 1.2482743860024614,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004853844942961346,
      "loss": 6.1029,
      "mean_token_accuracy": 0.14201263785362245,
      "num_tokens": 25240073.0,
      "step": 11665
    },
    {
      "entropy": 6.29432864189148,
      "epoch": 1.2488094601102253,
      "grad_norm": 1.203125,
      "learning_rate": 0.000485370936139537,
      "loss": 6.16,
      "mean_token_accuracy": 0.14866113737225534,
      "num_tokens": 25251915.0,
      "step": 11670
    },
    {
      "entropy": 6.320730924606323,
      "epoch": 1.2493445342179892,
      "grad_norm": 1.265625,
      "learning_rate": 0.000485357371908508,
      "loss": 6.1239,
      "mean_token_accuracy": 0.1431584894657135,
      "num_tokens": 25262932.0,
      "step": 11675
    },
    {
      "entropy": 6.349603128433228,
      "epoch": 1.249879608325753,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004853438016034396,
      "loss": 6.2056,
      "mean_token_accuracy": 0.1369818150997162,
      "num_tokens": 25274693.0,
      "step": 11680
    },
    {
      "entropy": 6.389609384536743,
      "epoch": 1.250414682433517,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048533022522472344,
      "loss": 6.0849,
      "mean_token_accuracy": 0.14332614243030548,
      "num_tokens": 25285359.0,
      "step": 11685
    },
    {
      "entropy": 6.328571605682373,
      "epoch": 1.250949756541281,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048531664277275154,
      "loss": 6.1122,
      "mean_token_accuracy": 0.14535362645983696,
      "num_tokens": 25296730.0,
      "step": 11690
    },
    {
      "entropy": 6.294487714767456,
      "epoch": 1.251484830649045,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004853030542479162,
      "loss": 6.1592,
      "mean_token_accuracy": 0.14992263168096542,
      "num_tokens": 25306422.0,
      "step": 11695
    },
    {
      "entropy": 6.338095617294312,
      "epoch": 1.2520199047568088,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004852894596506096,
      "loss": 6.1828,
      "mean_token_accuracy": 0.14128911644220352,
      "num_tokens": 25317980.0,
      "step": 11700
    },
    {
      "entropy": 6.334776878356934,
      "epoch": 1.2525549788645727,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004852758589812245,
      "loss": 6.2095,
      "mean_token_accuracy": 0.1504050999879837,
      "num_tokens": 25329905.0,
      "step": 11705
    },
    {
      "entropy": 6.4512287139892575,
      "epoch": 1.2530900529723366,
      "grad_norm": 1.25,
      "learning_rate": 0.00048526225224015355,
      "loss": 6.1718,
      "mean_token_accuracy": 0.1344292365014553,
      "num_tokens": 25340759.0,
      "step": 11710
    },
    {
      "entropy": 6.34198637008667,
      "epoch": 1.2536251270801007,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004852486394277896,
      "loss": 6.1968,
      "mean_token_accuracy": 0.13422559648752214,
      "num_tokens": 25351124.0,
      "step": 11715
    },
    {
      "entropy": 6.37648286819458,
      "epoch": 1.2541602011878645,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048523502054452565,
      "loss": 6.1304,
      "mean_token_accuracy": 0.1381153032183647,
      "num_tokens": 25361337.0,
      "step": 11720
    },
    {
      "entropy": 6.37690544128418,
      "epoch": 1.2546952752956284,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048522139559075507,
      "loss": 6.1694,
      "mean_token_accuracy": 0.14293650016188622,
      "num_tokens": 25371893.0,
      "step": 11725
    },
    {
      "entropy": 6.248219776153564,
      "epoch": 1.2552303494033923,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004852077645668712,
      "loss": 6.0524,
      "mean_token_accuracy": 0.1481030724942684,
      "num_tokens": 25383365.0,
      "step": 11730
    },
    {
      "entropy": 6.347385406494141,
      "epoch": 1.2557654235111562,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004851941274732678,
      "loss": 6.1371,
      "mean_token_accuracy": 0.1442984715104103,
      "num_tokens": 25393211.0,
      "step": 11735
    },
    {
      "entropy": 6.266662120819092,
      "epoch": 1.2563004976189203,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004851804843103384,
      "loss": 6.0194,
      "mean_token_accuracy": 0.13975931257009505,
      "num_tokens": 25404828.0,
      "step": 11740
    },
    {
      "entropy": 6.376990222930909,
      "epoch": 1.2568355717266841,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048516683507847705,
      "loss": 6.1582,
      "mean_token_accuracy": 0.14862019419670106,
      "num_tokens": 25416016.0,
      "step": 11745
    },
    {
      "entropy": 6.342615747451783,
      "epoch": 1.257370645834448,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004851531797780779,
      "loss": 6.1057,
      "mean_token_accuracy": 0.14726210236549378,
      "num_tokens": 25426173.0,
      "step": 11750
    },
    {
      "entropy": 6.322670221328735,
      "epoch": 1.2579057199422121,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004851395184095352,
      "loss": 6.0761,
      "mean_token_accuracy": 0.14972629994153977,
      "num_tokens": 25437507.0,
      "step": 11755
    },
    {
      "entropy": 6.364311695098877,
      "epoch": 1.258440794049976,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004851258509732434,
      "loss": 6.1684,
      "mean_token_accuracy": 0.13541021347045898,
      "num_tokens": 25448201.0,
      "step": 11760
    },
    {
      "entropy": 6.367018508911133,
      "epoch": 1.2589758681577399,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048511217746959723,
      "loss": 6.1664,
      "mean_token_accuracy": 0.14107578322291375,
      "num_tokens": 25458640.0,
      "step": 11765
    },
    {
      "entropy": 6.365227699279785,
      "epoch": 1.2595109422655038,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048509849789899146,
      "loss": 6.1756,
      "mean_token_accuracy": 0.1449488267302513,
      "num_tokens": 25469474.0,
      "step": 11770
    },
    {
      "entropy": 6.371990442276001,
      "epoch": 1.2600460163732676,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000485084812261821,
      "loss": 6.2538,
      "mean_token_accuracy": 0.1373679131269455,
      "num_tokens": 25479565.0,
      "step": 11775
    },
    {
      "entropy": 6.362201261520386,
      "epoch": 1.2605810904810317,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004850711205584812,
      "loss": 6.0964,
      "mean_token_accuracy": 0.14699497371912001,
      "num_tokens": 25490142.0,
      "step": 11780
    },
    {
      "entropy": 6.319552946090698,
      "epoch": 1.2611161645887956,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048505742278936726,
      "loss": 6.217,
      "mean_token_accuracy": 0.14117234349250793,
      "num_tokens": 25500824.0,
      "step": 11785
    },
    {
      "entropy": 6.366677474975586,
      "epoch": 1.2616512386965595,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048504371895487474,
      "loss": 6.1522,
      "mean_token_accuracy": 0.14228074103593827,
      "num_tokens": 25511251.0,
      "step": 11790
    },
    {
      "entropy": 6.381395483016968,
      "epoch": 1.2621863128043234,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048503000905539945,
      "loss": 6.144,
      "mean_token_accuracy": 0.1364334985613823,
      "num_tokens": 25521368.0,
      "step": 11795
    },
    {
      "entropy": 6.321301364898682,
      "epoch": 1.2627213869120872,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048501629309133707,
      "loss": 6.1326,
      "mean_token_accuracy": 0.13753886222839357,
      "num_tokens": 25532486.0,
      "step": 11800
    },
    {
      "entropy": 6.311598300933838,
      "epoch": 1.2632564610198513,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004850025710630838,
      "loss": 6.1242,
      "mean_token_accuracy": 0.14329409822821618,
      "num_tokens": 25542867.0,
      "step": 11805
    },
    {
      "entropy": 6.299130439758301,
      "epoch": 1.2637915351276152,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048498884297103575,
      "loss": 6.1089,
      "mean_token_accuracy": 0.1410810448229313,
      "num_tokens": 25553564.0,
      "step": 11810
    },
    {
      "entropy": 6.323140001296997,
      "epoch": 1.264326609235379,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004849751088155894,
      "loss": 6.1193,
      "mean_token_accuracy": 0.1419119268655777,
      "num_tokens": 25564017.0,
      "step": 11815
    },
    {
      "entropy": 6.312384986877442,
      "epoch": 1.264861683343143,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004849613685971413,
      "loss": 6.046,
      "mean_token_accuracy": 0.14667176604270935,
      "num_tokens": 25575125.0,
      "step": 11820
    },
    {
      "entropy": 6.33348240852356,
      "epoch": 1.2653967574509069,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004849476223160882,
      "loss": 6.1983,
      "mean_token_accuracy": 0.13863577470183372,
      "num_tokens": 25586405.0,
      "step": 11825
    },
    {
      "entropy": 6.336756849288941,
      "epoch": 1.265931831558671,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004849338699728269,
      "loss": 6.1556,
      "mean_token_accuracy": 0.14100272655487062,
      "num_tokens": 25597494.0,
      "step": 11830
    },
    {
      "entropy": 6.337206602096558,
      "epoch": 1.2664669056664348,
      "grad_norm": 1.25,
      "learning_rate": 0.00048492011156775467,
      "loss": 6.1193,
      "mean_token_accuracy": 0.1463923931121826,
      "num_tokens": 25608425.0,
      "step": 11835
    },
    {
      "entropy": 6.347609615325927,
      "epoch": 1.2670019797741987,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004849063471012688,
      "loss": 6.1436,
      "mean_token_accuracy": 0.13409415632486343,
      "num_tokens": 25619411.0,
      "step": 11840
    },
    {
      "entropy": 6.398722314834595,
      "epoch": 1.2675370538819626,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004848925765737665,
      "loss": 6.1924,
      "mean_token_accuracy": 0.134124518930912,
      "num_tokens": 25630477.0,
      "step": 11845
    },
    {
      "entropy": 6.384574127197266,
      "epoch": 1.2680721279897265,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048487879998564565,
      "loss": 6.091,
      "mean_token_accuracy": 0.14267901331186295,
      "num_tokens": 25640129.0,
      "step": 11850
    },
    {
      "entropy": 6.339227390289307,
      "epoch": 1.2686072020974906,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048486501733730386,
      "loss": 6.1687,
      "mean_token_accuracy": 0.1423572801053524,
      "num_tokens": 25650592.0,
      "step": 11855
    },
    {
      "entropy": 6.312867975234985,
      "epoch": 1.2691422762052544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048485122862913934,
      "loss": 6.0954,
      "mean_token_accuracy": 0.14822685420513154,
      "num_tokens": 25660408.0,
      "step": 11860
    },
    {
      "entropy": 6.332176208496094,
      "epoch": 1.2696773503130183,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048483743386154994,
      "loss": 6.2004,
      "mean_token_accuracy": 0.1429707810282707,
      "num_tokens": 25671563.0,
      "step": 11865
    },
    {
      "entropy": 6.3412879467010494,
      "epoch": 1.2702124244207824,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048482363303493415,
      "loss": 6.1094,
      "mean_token_accuracy": 0.1431986518204212,
      "num_tokens": 25682474.0,
      "step": 11870
    },
    {
      "entropy": 6.4208940982818605,
      "epoch": 1.270747498528546,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004848098261496905,
      "loss": 6.1291,
      "mean_token_accuracy": 0.1416504606604576,
      "num_tokens": 25692594.0,
      "step": 11875
    },
    {
      "entropy": 6.206357955932617,
      "epoch": 1.2712825726363102,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004847960132062175,
      "loss": 5.9551,
      "mean_token_accuracy": 0.15593411177396774,
      "num_tokens": 25703724.0,
      "step": 11880
    },
    {
      "entropy": 6.354020071029663,
      "epoch": 1.271817646744074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004847821942049142,
      "loss": 6.275,
      "mean_token_accuracy": 0.12894992753863335,
      "num_tokens": 25714911.0,
      "step": 11885
    },
    {
      "entropy": 6.420945167541504,
      "epoch": 1.272352720851838,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048476836914617945,
      "loss": 6.1443,
      "mean_token_accuracy": 0.14085684567689896,
      "num_tokens": 25725098.0,
      "step": 11890
    },
    {
      "entropy": 6.361540985107422,
      "epoch": 1.272887794959602,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048475453803041254,
      "loss": 6.101,
      "mean_token_accuracy": 0.13643536493182182,
      "num_tokens": 25735099.0,
      "step": 11895
    },
    {
      "entropy": 6.337470769882202,
      "epoch": 1.273422869067366,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004847407008580128,
      "loss": 6.1037,
      "mean_token_accuracy": 0.14218868911266327,
      "num_tokens": 25746852.0,
      "step": 11900
    },
    {
      "entropy": 6.362613296508789,
      "epoch": 1.2739579431751298,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048472685762937983,
      "loss": 6.1728,
      "mean_token_accuracy": 0.14067593216896057,
      "num_tokens": 25758579.0,
      "step": 11905
    },
    {
      "entropy": 6.323443126678467,
      "epoch": 1.2744930172828937,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048471300834491325,
      "loss": 6.1397,
      "mean_token_accuracy": 0.1384618729352951,
      "num_tokens": 25769965.0,
      "step": 11910
    },
    {
      "entropy": 6.3721636772155765,
      "epoch": 1.2750280913906575,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004846991530050132,
      "loss": 6.1484,
      "mean_token_accuracy": 0.14623067826032637,
      "num_tokens": 25780264.0,
      "step": 11915
    },
    {
      "entropy": 6.399501991271973,
      "epoch": 1.2755631654984216,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004846852916100794,
      "loss": 6.0882,
      "mean_token_accuracy": 0.14634262919425964,
      "num_tokens": 25791228.0,
      "step": 11920
    },
    {
      "entropy": 6.3572145938873295,
      "epoch": 1.2760982396061855,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004846714241605123,
      "loss": 6.0669,
      "mean_token_accuracy": 0.1476261168718338,
      "num_tokens": 25802428.0,
      "step": 11925
    },
    {
      "entropy": 6.33449969291687,
      "epoch": 1.2766333137139494,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048465755065671233,
      "loss": 6.1488,
      "mean_token_accuracy": 0.1439507246017456,
      "num_tokens": 25814220.0,
      "step": 11930
    },
    {
      "entropy": 6.3583086967468265,
      "epoch": 1.2771683878217133,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048464367109908005,
      "loss": 6.1501,
      "mean_token_accuracy": 0.1392370879650116,
      "num_tokens": 25824382.0,
      "step": 11935
    },
    {
      "entropy": 6.2743762016296385,
      "epoch": 1.2777034619294771,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004846297854880162,
      "loss": 6.1101,
      "mean_token_accuracy": 0.14475442916154863,
      "num_tokens": 25834896.0,
      "step": 11940
    },
    {
      "entropy": 6.385753583908081,
      "epoch": 1.2782385360372412,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048461589382392176,
      "loss": 6.2101,
      "mean_token_accuracy": 0.1378077410161495,
      "num_tokens": 25847300.0,
      "step": 11945
    },
    {
      "entropy": 6.415359544754028,
      "epoch": 1.2787736101450051,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004846019961071978,
      "loss": 6.1408,
      "mean_token_accuracy": 0.1443222776055336,
      "num_tokens": 25858599.0,
      "step": 11950
    },
    {
      "entropy": 6.298390674591064,
      "epoch": 1.279308684252769,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004845880923382457,
      "loss": 6.0908,
      "mean_token_accuracy": 0.14585315734148024,
      "num_tokens": 25869135.0,
      "step": 11955
    },
    {
      "entropy": 6.297159719467163,
      "epoch": 1.2798437583605329,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048457418251746695,
      "loss": 6.0924,
      "mean_token_accuracy": 0.14810308665037156,
      "num_tokens": 25879234.0,
      "step": 11960
    },
    {
      "entropy": 6.378995084762574,
      "epoch": 1.2803788324682968,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004845602666452629,
      "loss": 6.138,
      "mean_token_accuracy": 0.14431929886341094,
      "num_tokens": 25888230.0,
      "step": 11965
    },
    {
      "entropy": 6.351518726348877,
      "epoch": 1.2809139065760609,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004845463447220358,
      "loss": 6.1534,
      "mean_token_accuracy": 0.1512703627347946,
      "num_tokens": 25899429.0,
      "step": 11970
    },
    {
      "entropy": 6.29874963760376,
      "epoch": 1.2814489806838247,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004845324167481874,
      "loss": 6.0558,
      "mean_token_accuracy": 0.14683472216129304,
      "num_tokens": 25911042.0,
      "step": 11975
    },
    {
      "entropy": 6.374869728088379,
      "epoch": 1.2819840547915886,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004845184827241199,
      "loss": 6.099,
      "mean_token_accuracy": 0.1438555158674717,
      "num_tokens": 25921466.0,
      "step": 11980
    },
    {
      "entropy": 6.348728656768799,
      "epoch": 1.2825191288993525,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004845045426502357,
      "loss": 6.2209,
      "mean_token_accuracy": 0.13643710315227509,
      "num_tokens": 25931059.0,
      "step": 11985
    },
    {
      "entropy": 6.368209600448608,
      "epoch": 1.2830542030071164,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048449059652693724,
      "loss": 6.1211,
      "mean_token_accuracy": 0.14528272300958633,
      "num_tokens": 25941669.0,
      "step": 11990
    },
    {
      "entropy": 6.365387010574341,
      "epoch": 1.2835892771148805,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004844766443546272,
      "loss": 6.156,
      "mean_token_accuracy": 0.1462119273841381,
      "num_tokens": 25952066.0,
      "step": 11995
    },
    {
      "entropy": 6.377856159210205,
      "epoch": 1.2841243512226443,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004844626861337085,
      "loss": 6.0431,
      "mean_token_accuracy": 0.1487523227930069,
      "num_tokens": 25962388.0,
      "step": 12000
    },
    {
      "epoch": 1.2841243512226443,
      "eval_entropy": 6.166471233907735,
      "eval_loss": 6.27332878112793,
      "eval_mean_token_accuracy": 0.14317854465264074,
      "eval_num_tokens": 25962388.0,
      "eval_runtime": 22.8163,
      "eval_samples_per_second": 1300.781,
      "eval_steps_per_second": 162.603,
      "step": 12000
    },
    {
      "entropy": 6.234992265701294,
      "epoch": 1.2846594253304082,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004844487218645843,
      "loss": 6.1044,
      "mean_token_accuracy": 0.1421353615820408,
      "num_tokens": 25973865.0,
      "step": 12005
    },
    {
      "entropy": 6.272721624374389,
      "epoch": 1.2851944994381723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004844347515476575,
      "loss": 6.1204,
      "mean_token_accuracy": 0.1413175046443939,
      "num_tokens": 25985113.0,
      "step": 12010
    },
    {
      "entropy": 6.307812786102295,
      "epoch": 1.285729573545936,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048442077518333175,
      "loss": 6.0183,
      "mean_token_accuracy": 0.1522618815302849,
      "num_tokens": 25995980.0,
      "step": 12015
    },
    {
      "entropy": 6.27951979637146,
      "epoch": 1.2862646476537,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048440679277201063,
      "loss": 6.1326,
      "mean_token_accuracy": 0.1443421185016632,
      "num_tokens": 26006362.0,
      "step": 12020
    },
    {
      "entropy": 6.370836973190308,
      "epoch": 1.286799721761464,
      "grad_norm": 1.25,
      "learning_rate": 0.00048439280431409774,
      "loss": 6.1971,
      "mean_token_accuracy": 0.14199472814798356,
      "num_tokens": 26016623.0,
      "step": 12025
    },
    {
      "entropy": 6.252241420745849,
      "epoch": 1.2873347958692278,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048437880980999694,
      "loss": 6.0118,
      "mean_token_accuracy": 0.15094649642705918,
      "num_tokens": 26027202.0,
      "step": 12030
    },
    {
      "entropy": 6.261988687515259,
      "epoch": 1.287869869976992,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004843648092601125,
      "loss": 6.1162,
      "mean_token_accuracy": 0.13929837346076965,
      "num_tokens": 26037330.0,
      "step": 12035
    },
    {
      "entropy": 6.355669927597046,
      "epoch": 1.2884049440847558,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004843508026648487,
      "loss": 6.1258,
      "mean_token_accuracy": 0.1449706770479679,
      "num_tokens": 26047661.0,
      "step": 12040
    },
    {
      "entropy": 6.437519073486328,
      "epoch": 1.2889400181925197,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004843367900246098,
      "loss": 6.1847,
      "mean_token_accuracy": 0.13747022226452826,
      "num_tokens": 26058582.0,
      "step": 12045
    },
    {
      "entropy": 6.358186435699463,
      "epoch": 1.2894750923002836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048432277133980055,
      "loss": 6.17,
      "mean_token_accuracy": 0.13278811648488045,
      "num_tokens": 26069117.0,
      "step": 12050
    },
    {
      "entropy": 6.321809530258179,
      "epoch": 1.2900101664080474,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048430874661082565,
      "loss": 6.125,
      "mean_token_accuracy": 0.1461104415357113,
      "num_tokens": 26079327.0,
      "step": 12055
    },
    {
      "entropy": 6.3696681499481205,
      "epoch": 1.2905452405158115,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048429471583809016,
      "loss": 6.148,
      "mean_token_accuracy": 0.1354692466557026,
      "num_tokens": 26090709.0,
      "step": 12060
    },
    {
      "entropy": 6.415619039535523,
      "epoch": 1.2910803146235754,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004842806790219991,
      "loss": 6.144,
      "mean_token_accuracy": 0.13840869665145875,
      "num_tokens": 26101570.0,
      "step": 12065
    },
    {
      "entropy": 6.307449293136597,
      "epoch": 1.2916153887313393,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048426663616295783,
      "loss": 6.1138,
      "mean_token_accuracy": 0.14589108526706696,
      "num_tokens": 26112827.0,
      "step": 12070
    },
    {
      "entropy": 6.350975561141968,
      "epoch": 1.2921504628391032,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048425258726137187,
      "loss": 6.1778,
      "mean_token_accuracy": 0.13445790633559226,
      "num_tokens": 26124141.0,
      "step": 12075
    },
    {
      "entropy": 6.350972270965576,
      "epoch": 1.292685536946867,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048423853231764684,
      "loss": 6.1179,
      "mean_token_accuracy": 0.1429880067706108,
      "num_tokens": 26135372.0,
      "step": 12080
    },
    {
      "entropy": 6.324437952041626,
      "epoch": 1.2932206110546312,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048422447133218867,
      "loss": 6.0991,
      "mean_token_accuracy": 0.14110313802957536,
      "num_tokens": 26145933.0,
      "step": 12085
    },
    {
      "entropy": 6.373568677902222,
      "epoch": 1.293755685162395,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048421040430540323,
      "loss": 6.1869,
      "mean_token_accuracy": 0.1379242643713951,
      "num_tokens": 26155948.0,
      "step": 12090
    },
    {
      "entropy": 6.313356733322143,
      "epoch": 1.294290759270159,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004841963312376967,
      "loss": 6.0717,
      "mean_token_accuracy": 0.1489647552371025,
      "num_tokens": 26166410.0,
      "step": 12095
    },
    {
      "entropy": 6.2853649139404295,
      "epoch": 1.2948258333779228,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048418225212947566,
      "loss": 6.1029,
      "mean_token_accuracy": 0.14209917634725572,
      "num_tokens": 26177334.0,
      "step": 12100
    },
    {
      "entropy": 6.4342286586761475,
      "epoch": 1.2953609074856867,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004841681669811464,
      "loss": 6.169,
      "mean_token_accuracy": 0.134670490026474,
      "num_tokens": 26187987.0,
      "step": 12105
    },
    {
      "entropy": 6.349659538269043,
      "epoch": 1.2958959815934508,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004841540757931157,
      "loss": 6.1205,
      "mean_token_accuracy": 0.1350249208509922,
      "num_tokens": 26198600.0,
      "step": 12110
    },
    {
      "entropy": 6.296187877655029,
      "epoch": 1.2964310557012146,
      "grad_norm": 1.25,
      "learning_rate": 0.0004841399785657904,
      "loss": 6.1662,
      "mean_token_accuracy": 0.13952014967799187,
      "num_tokens": 26210977.0,
      "step": 12115
    },
    {
      "entropy": 6.365020656585694,
      "epoch": 1.2969661298089785,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048412587529957774,
      "loss": 6.1271,
      "mean_token_accuracy": 0.1409282088279724,
      "num_tokens": 26220813.0,
      "step": 12120
    },
    {
      "entropy": 6.3568277835845945,
      "epoch": 1.2975012039167424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004841117659948848,
      "loss": 6.1561,
      "mean_token_accuracy": 0.14278296381235123,
      "num_tokens": 26232042.0,
      "step": 12125
    },
    {
      "entropy": 6.397304105758667,
      "epoch": 1.2980362780245063,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048409765065211884,
      "loss": 6.1652,
      "mean_token_accuracy": 0.1406002052128315,
      "num_tokens": 26242968.0,
      "step": 12130
    },
    {
      "entropy": 6.428318452835083,
      "epoch": 1.2985713521322704,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048408352927168774,
      "loss": 6.1589,
      "mean_token_accuracy": 0.14254745244979858,
      "num_tokens": 26252751.0,
      "step": 12135
    },
    {
      "entropy": 6.3686905860900875,
      "epoch": 1.2991064262400343,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004840694018539991,
      "loss": 6.2127,
      "mean_token_accuracy": 0.13514733985066413,
      "num_tokens": 26263270.0,
      "step": 12140
    },
    {
      "entropy": 6.373553943634033,
      "epoch": 1.2996415003477981,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048405526839946086,
      "loss": 6.2197,
      "mean_token_accuracy": 0.13592303097248076,
      "num_tokens": 26274279.0,
      "step": 12145
    },
    {
      "entropy": 6.388205623626709,
      "epoch": 1.3001765744555622,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048404112890848105,
      "loss": 6.2276,
      "mean_token_accuracy": 0.13989843800663948,
      "num_tokens": 26286020.0,
      "step": 12150
    },
    {
      "entropy": 6.422605323791504,
      "epoch": 1.3007116485633259,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048402698338146803,
      "loss": 6.1361,
      "mean_token_accuracy": 0.14263813346624374,
      "num_tokens": 26296798.0,
      "step": 12155
    },
    {
      "entropy": 6.2984270572662355,
      "epoch": 1.30124672267109,
      "grad_norm": 1.125,
      "learning_rate": 0.0004840128318188303,
      "loss": 6.1182,
      "mean_token_accuracy": 0.14364108070731163,
      "num_tokens": 26306952.0,
      "step": 12160
    },
    {
      "entropy": 6.399915409088135,
      "epoch": 1.3017817967788539,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048399867422097644,
      "loss": 6.1916,
      "mean_token_accuracy": 0.14062896743416786,
      "num_tokens": 26317686.0,
      "step": 12165
    },
    {
      "entropy": 6.3822509288787845,
      "epoch": 1.3023168708866177,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048398451058831514,
      "loss": 6.1788,
      "mean_token_accuracy": 0.14291637986898423,
      "num_tokens": 26327997.0,
      "step": 12170
    },
    {
      "entropy": 6.3525773048400875,
      "epoch": 1.3028519449943818,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048397034092125554,
      "loss": 6.1536,
      "mean_token_accuracy": 0.1379798248410225,
      "num_tokens": 26340227.0,
      "step": 12175
    },
    {
      "entropy": 6.516258382797242,
      "epoch": 1.3033870191021457,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048395616522020667,
      "loss": 6.2626,
      "mean_token_accuracy": 0.13677845001220704,
      "num_tokens": 26350549.0,
      "step": 12180
    },
    {
      "entropy": 6.3615306377410885,
      "epoch": 1.3039220932099096,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004839419834855779,
      "loss": 6.1691,
      "mean_token_accuracy": 0.14084514304995538,
      "num_tokens": 26361822.0,
      "step": 12185
    },
    {
      "entropy": 6.244283962249756,
      "epoch": 1.3044571673176735,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004839277957177787,
      "loss": 6.1826,
      "mean_token_accuracy": 0.14139464125037193,
      "num_tokens": 26372720.0,
      "step": 12190
    },
    {
      "entropy": 6.305957937240601,
      "epoch": 1.3049922414254373,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004839136019172188,
      "loss": 6.1349,
      "mean_token_accuracy": 0.14600130617618562,
      "num_tokens": 26383092.0,
      "step": 12195
    },
    {
      "entropy": 6.3375396728515625,
      "epoch": 1.3055273155332014,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000483899402084308,
      "loss": 6.0947,
      "mean_token_accuracy": 0.14927577301859857,
      "num_tokens": 26395188.0,
      "step": 12200
    },
    {
      "entropy": 6.396602773666382,
      "epoch": 1.3060623896409653,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004838851962194563,
      "loss": 6.2123,
      "mean_token_accuracy": 0.1400100976228714,
      "num_tokens": 26406338.0,
      "step": 12205
    },
    {
      "entropy": 6.307674217224121,
      "epoch": 1.3065974637487292,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004838709843230739,
      "loss": 6.134,
      "mean_token_accuracy": 0.1432012252509594,
      "num_tokens": 26416983.0,
      "step": 12210
    },
    {
      "entropy": 6.316882085800171,
      "epoch": 1.307132537856493,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004838567663955712,
      "loss": 6.0851,
      "mean_token_accuracy": 0.14710725843906403,
      "num_tokens": 26426966.0,
      "step": 12215
    },
    {
      "entropy": 6.313268995285034,
      "epoch": 1.307667611964257,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004838425424373587,
      "loss": 6.0569,
      "mean_token_accuracy": 0.142745304107666,
      "num_tokens": 26437711.0,
      "step": 12220
    },
    {
      "entropy": 6.335655546188354,
      "epoch": 1.308202686072021,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004838283124488472,
      "loss": 6.207,
      "mean_token_accuracy": 0.13857814148068429,
      "num_tokens": 26448657.0,
      "step": 12225
    },
    {
      "entropy": 6.418257761001587,
      "epoch": 1.308737760179785,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048381407643044737,
      "loss": 6.1015,
      "mean_token_accuracy": 0.14642067849636078,
      "num_tokens": 26459363.0,
      "step": 12230
    },
    {
      "entropy": 6.334084177017212,
      "epoch": 1.3092728342875488,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004837998343825705,
      "loss": 6.1252,
      "mean_token_accuracy": 0.14405779615044595,
      "num_tokens": 26470149.0,
      "step": 12235
    },
    {
      "entropy": 6.306715488433838,
      "epoch": 1.3098079083953127,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048378558630562783,
      "loss": 6.0855,
      "mean_token_accuracy": 0.14852658212184905,
      "num_tokens": 26480424.0,
      "step": 12240
    },
    {
      "entropy": 6.325909805297852,
      "epoch": 1.3103429825030766,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004837713322000306,
      "loss": 6.1168,
      "mean_token_accuracy": 0.13921271860599518,
      "num_tokens": 26490312.0,
      "step": 12245
    },
    {
      "entropy": 6.331692314147949,
      "epoch": 1.3108780566108407,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004837570720661905,
      "loss": 6.1066,
      "mean_token_accuracy": 0.15162073820829391,
      "num_tokens": 26501119.0,
      "step": 12250
    },
    {
      "entropy": 6.371668815612793,
      "epoch": 1.3114131307186045,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048374280590451934,
      "loss": 6.1852,
      "mean_token_accuracy": 0.14228272438049316,
      "num_tokens": 26512438.0,
      "step": 12255
    },
    {
      "entropy": 6.337365531921387,
      "epoch": 1.3119482048263684,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048372853371542895,
      "loss": 6.1353,
      "mean_token_accuracy": 0.13777869045734406,
      "num_tokens": 26523930.0,
      "step": 12260
    },
    {
      "entropy": 6.373329782485962,
      "epoch": 1.3124832789341323,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004837142554993315,
      "loss": 6.2277,
      "mean_token_accuracy": 0.14057578071951865,
      "num_tokens": 26534937.0,
      "step": 12265
    },
    {
      "entropy": 6.288657474517822,
      "epoch": 1.3130183530418962,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004836999712566392,
      "loss": 6.0049,
      "mean_token_accuracy": 0.1459055595099926,
      "num_tokens": 26544555.0,
      "step": 12270
    },
    {
      "entropy": 6.342950010299683,
      "epoch": 1.3135534271496603,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048368568098776463,
      "loss": 6.1379,
      "mean_token_accuracy": 0.15345437526702882,
      "num_tokens": 26556144.0,
      "step": 12275
    },
    {
      "entropy": 6.391424512863159,
      "epoch": 1.3140885012574242,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004836713846931203,
      "loss": 6.2244,
      "mean_token_accuracy": 0.13335091844201089,
      "num_tokens": 26567350.0,
      "step": 12280
    },
    {
      "entropy": 6.362367248535156,
      "epoch": 1.314623575365188,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048365708237311904,
      "loss": 6.1102,
      "mean_token_accuracy": 0.14023372009396554,
      "num_tokens": 26579486.0,
      "step": 12285
    },
    {
      "entropy": 6.31515645980835,
      "epoch": 1.3151586494729521,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004836427740281739,
      "loss": 6.111,
      "mean_token_accuracy": 0.14438905715942382,
      "num_tokens": 26589665.0,
      "step": 12290
    },
    {
      "entropy": 6.343943166732788,
      "epoch": 1.3156937235807158,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048362845965869793,
      "loss": 6.1231,
      "mean_token_accuracy": 0.14355553239583968,
      "num_tokens": 26600187.0,
      "step": 12295
    },
    {
      "entropy": 6.356946659088135,
      "epoch": 1.31622879768848,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004836141392651046,
      "loss": 6.0942,
      "mean_token_accuracy": 0.1389931932091713,
      "num_tokens": 26612155.0,
      "step": 12300
    },
    {
      "entropy": 6.376056671142578,
      "epoch": 1.3167638717962438,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048359981284780727,
      "loss": 6.1264,
      "mean_token_accuracy": 0.13865249529480933,
      "num_tokens": 26623077.0,
      "step": 12305
    },
    {
      "entropy": 6.301869010925293,
      "epoch": 1.3172989459040076,
      "grad_norm": 1.375,
      "learning_rate": 0.0004835854804072196,
      "loss": 6.0571,
      "mean_token_accuracy": 0.15024217665195466,
      "num_tokens": 26632979.0,
      "step": 12310
    },
    {
      "entropy": 6.343993854522705,
      "epoch": 1.3178340200117717,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048357114194375555,
      "loss": 6.1788,
      "mean_token_accuracy": 0.1388743907213211,
      "num_tokens": 26644377.0,
      "step": 12315
    },
    {
      "entropy": 6.33191876411438,
      "epoch": 1.3183690941195356,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000483556797457829,
      "loss": 6.1615,
      "mean_token_accuracy": 0.13590002954006195,
      "num_tokens": 26655538.0,
      "step": 12320
    },
    {
      "entropy": 6.386764669418335,
      "epoch": 1.3189041682272995,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048354244694985435,
      "loss": 6.1074,
      "mean_token_accuracy": 0.13884449750185013,
      "num_tokens": 26668011.0,
      "step": 12325
    },
    {
      "entropy": 6.379239559173584,
      "epoch": 1.3194392423350634,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004835280904202457,
      "loss": 6.1382,
      "mean_token_accuracy": 0.14048136919736862,
      "num_tokens": 26678996.0,
      "step": 12330
    },
    {
      "entropy": 6.343082571029663,
      "epoch": 1.3199743164428273,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004835137278694178,
      "loss": 6.1066,
      "mean_token_accuracy": 0.1448737420141697,
      "num_tokens": 26689041.0,
      "step": 12335
    },
    {
      "entropy": 6.377063131332397,
      "epoch": 1.3205093905505914,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004834993592977854,
      "loss": 6.1049,
      "mean_token_accuracy": 0.14947788119316102,
      "num_tokens": 26699459.0,
      "step": 12340
    },
    {
      "entropy": 6.370327377319336,
      "epoch": 1.3210444646583552,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004834849847057632,
      "loss": 6.2287,
      "mean_token_accuracy": 0.13687031120061874,
      "num_tokens": 26711663.0,
      "step": 12345
    },
    {
      "entropy": 6.3510137557983395,
      "epoch": 1.321579538766119,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048347060409376644,
      "loss": 6.18,
      "mean_token_accuracy": 0.13505842462182044,
      "num_tokens": 26722960.0,
      "step": 12350
    },
    {
      "entropy": 6.3480370998382565,
      "epoch": 1.322114612873883,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048345621746221024,
      "loss": 6.1051,
      "mean_token_accuracy": 0.14648084789514543,
      "num_tokens": 26733274.0,
      "step": 12355
    },
    {
      "entropy": 6.352724075317383,
      "epoch": 1.3226496869816469,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048344182481151003,
      "loss": 6.1476,
      "mean_token_accuracy": 0.13709251806139947,
      "num_tokens": 26743906.0,
      "step": 12360
    },
    {
      "entropy": 6.300246524810791,
      "epoch": 1.323184761089411,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048342742614208145,
      "loss": 6.1595,
      "mean_token_accuracy": 0.1403728023171425,
      "num_tokens": 26755093.0,
      "step": 12365
    },
    {
      "entropy": 6.348132705688476,
      "epoch": 1.3237198351971748,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004834130214543402,
      "loss": 6.1208,
      "mean_token_accuracy": 0.1414766862988472,
      "num_tokens": 26765614.0,
      "step": 12370
    },
    {
      "entropy": 6.310495805740357,
      "epoch": 1.3242549093049387,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048339861074870227,
      "loss": 6.1207,
      "mean_token_accuracy": 0.1422338902950287,
      "num_tokens": 26778078.0,
      "step": 12375
    },
    {
      "entropy": 6.326334667205811,
      "epoch": 1.3247899834127026,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048338419402558375,
      "loss": 6.1463,
      "mean_token_accuracy": 0.14380858838558197,
      "num_tokens": 26788707.0,
      "step": 12380
    },
    {
      "entropy": 6.376149272918701,
      "epoch": 1.3253250575204665,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004833697712854009,
      "loss": 6.1134,
      "mean_token_accuracy": 0.14681189507246017,
      "num_tokens": 26798650.0,
      "step": 12385
    },
    {
      "entropy": 6.326212406158447,
      "epoch": 1.3258601316282306,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048335534252857016,
      "loss": 6.1511,
      "mean_token_accuracy": 0.14200389310717582,
      "num_tokens": 26809822.0,
      "step": 12390
    },
    {
      "entropy": 6.3200760841369625,
      "epoch": 1.3263952057359945,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004833409077555082,
      "loss": 6.1163,
      "mean_token_accuracy": 0.13777104914188384,
      "num_tokens": 26821061.0,
      "step": 12395
    },
    {
      "entropy": 6.435120439529419,
      "epoch": 1.3269302798437583,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048332646696663177,
      "loss": 6.1743,
      "mean_token_accuracy": 0.13782794252038003,
      "num_tokens": 26832020.0,
      "step": 12400
    },
    {
      "entropy": 6.414852523803711,
      "epoch": 1.3274653539515222,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004833120201623579,
      "loss": 6.1747,
      "mean_token_accuracy": 0.1416335955262184,
      "num_tokens": 26844287.0,
      "step": 12405
    },
    {
      "entropy": 6.215211248397827,
      "epoch": 1.328000428059286,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048329756734310375,
      "loss": 6.0223,
      "mean_token_accuracy": 0.15009742677211763,
      "num_tokens": 26855790.0,
      "step": 12410
    },
    {
      "entropy": 6.310049772262573,
      "epoch": 1.3285355021670502,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048328310850928656,
      "loss": 6.1545,
      "mean_token_accuracy": 0.1427854783833027,
      "num_tokens": 26867391.0,
      "step": 12415
    },
    {
      "entropy": 6.387913417816162,
      "epoch": 1.329070576274814,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048326864366132397,
      "loss": 6.159,
      "mean_token_accuracy": 0.14175219535827638,
      "num_tokens": 26879173.0,
      "step": 12420
    },
    {
      "entropy": 6.350165510177613,
      "epoch": 1.329605650382578,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004832541727996335,
      "loss": 6.0541,
      "mean_token_accuracy": 0.14805878922343255,
      "num_tokens": 26889839.0,
      "step": 12425
    },
    {
      "entropy": 6.23412675857544,
      "epoch": 1.330140724490342,
      "grad_norm": 1.234375,
      "learning_rate": 0.000483239695924633,
      "loss": 6.1312,
      "mean_token_accuracy": 0.1461761936545372,
      "num_tokens": 26900217.0,
      "step": 12430
    },
    {
      "entropy": 6.327548933029175,
      "epoch": 1.330675798598106,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048322521303674064,
      "loss": 6.1297,
      "mean_token_accuracy": 0.1465964987874031,
      "num_tokens": 26911366.0,
      "step": 12435
    },
    {
      "entropy": 6.41979603767395,
      "epoch": 1.3312108727058698,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048321072413637454,
      "loss": 6.1764,
      "mean_token_accuracy": 0.1361842028796673,
      "num_tokens": 26922135.0,
      "step": 12440
    },
    {
      "entropy": 6.307004833221436,
      "epoch": 1.3317459468136337,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004831962292239529,
      "loss": 6.106,
      "mean_token_accuracy": 0.14518800377845764,
      "num_tokens": 26932502.0,
      "step": 12445
    },
    {
      "entropy": 6.314956760406494,
      "epoch": 1.3322810209213976,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048318172829989447,
      "loss": 6.1506,
      "mean_token_accuracy": 0.14758610278367995,
      "num_tokens": 26944607.0,
      "step": 12450
    },
    {
      "entropy": 6.399128389358521,
      "epoch": 1.3328160950291617,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004831672213646179,
      "loss": 6.1806,
      "mean_token_accuracy": 0.14059799984097482,
      "num_tokens": 26955205.0,
      "step": 12455
    },
    {
      "entropy": 6.318999290466309,
      "epoch": 1.3333511691369255,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004831527084185421,
      "loss": 6.038,
      "mean_token_accuracy": 0.15502638816833497,
      "num_tokens": 26965558.0,
      "step": 12460
    },
    {
      "entropy": 6.292956399917602,
      "epoch": 1.3338862432446894,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000483138189462086,
      "loss": 6.1306,
      "mean_token_accuracy": 0.1385799340903759,
      "num_tokens": 26976176.0,
      "step": 12465
    },
    {
      "entropy": 6.373552942276001,
      "epoch": 1.3344213173524533,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048312366449566895,
      "loss": 6.1631,
      "mean_token_accuracy": 0.13987504169344903,
      "num_tokens": 26987099.0,
      "step": 12470
    },
    {
      "entropy": 6.332768774032592,
      "epoch": 1.3349563914602172,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004831091335197104,
      "loss": 6.1287,
      "mean_token_accuracy": 0.14131635054945946,
      "num_tokens": 26998122.0,
      "step": 12475
    },
    {
      "entropy": 6.296827411651611,
      "epoch": 1.3354914655679813,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004830945965346298,
      "loss": 6.0825,
      "mean_token_accuracy": 0.14246188551187516,
      "num_tokens": 27009427.0,
      "step": 12480
    },
    {
      "entropy": 6.3953358173370365,
      "epoch": 1.3360265396757451,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000483080053540847,
      "loss": 6.2178,
      "mean_token_accuracy": 0.1383180059492588,
      "num_tokens": 27020701.0,
      "step": 12485
    },
    {
      "entropy": 6.4030348777771,
      "epoch": 1.336561613783509,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004830655045387818,
      "loss": 6.1773,
      "mean_token_accuracy": 0.13219248428940772,
      "num_tokens": 27033215.0,
      "step": 12490
    },
    {
      "entropy": 6.359825325012207,
      "epoch": 1.337096687891273,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048305094952885453,
      "loss": 6.1609,
      "mean_token_accuracy": 0.13748872205615043,
      "num_tokens": 27043846.0,
      "step": 12495
    },
    {
      "entropy": 6.350378036499023,
      "epoch": 1.3376317619990368,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004830363885114853,
      "loss": 6.1434,
      "mean_token_accuracy": 0.14058151170611383,
      "num_tokens": 27054783.0,
      "step": 12500
    },
    {
      "entropy": 6.324601411819458,
      "epoch": 1.3381668361068009,
      "grad_norm": 1.125,
      "learning_rate": 0.0004830218214870946,
      "loss": 6.1389,
      "mean_token_accuracy": 0.14377840831875802,
      "num_tokens": 27066051.0,
      "step": 12505
    },
    {
      "entropy": 6.401280403137207,
      "epoch": 1.3387019102145647,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004830072484561029,
      "loss": 6.1719,
      "mean_token_accuracy": 0.14222983345389367,
      "num_tokens": 27078210.0,
      "step": 12510
    },
    {
      "entropy": 6.393268871307373,
      "epoch": 1.3392369843223286,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004829926694189312,
      "loss": 6.1846,
      "mean_token_accuracy": 0.13744294345378877,
      "num_tokens": 27089693.0,
      "step": 12515
    },
    {
      "entropy": 6.37256760597229,
      "epoch": 1.3397720584300925,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004829780843760005,
      "loss": 6.1674,
      "mean_token_accuracy": 0.1396560363471508,
      "num_tokens": 27099856.0,
      "step": 12520
    },
    {
      "entropy": 6.3640453815460205,
      "epoch": 1.3403071325378564,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004829634933277317,
      "loss": 6.1777,
      "mean_token_accuracy": 0.13947432711720467,
      "num_tokens": 27110218.0,
      "step": 12525
    },
    {
      "entropy": 6.388849592208862,
      "epoch": 1.3408422066456205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048294889627454636,
      "loss": 6.1183,
      "mean_token_accuracy": 0.1469648078083992,
      "num_tokens": 27120911.0,
      "step": 12530
    },
    {
      "entropy": 6.359098625183106,
      "epoch": 1.3413772807533844,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048293429321686585,
      "loss": 6.1105,
      "mean_token_accuracy": 0.1427801378071308,
      "num_tokens": 27130964.0,
      "step": 12535
    },
    {
      "entropy": 6.342700719833374,
      "epoch": 1.3419123548611482,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048291968415511174,
      "loss": 6.1289,
      "mean_token_accuracy": 0.14218351542949675,
      "num_tokens": 27141641.0,
      "step": 12540
    },
    {
      "entropy": 6.387347793579101,
      "epoch": 1.3424474289689121,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048290506908970607,
      "loss": 6.1936,
      "mean_token_accuracy": 0.13711366504430772,
      "num_tokens": 27152591.0,
      "step": 12545
    },
    {
      "entropy": 6.414228534698486,
      "epoch": 1.342982503076676,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004828904480210707,
      "loss": 6.2726,
      "mean_token_accuracy": 0.13119366988539696,
      "num_tokens": 27163839.0,
      "step": 12550
    },
    {
      "entropy": 6.409824466705322,
      "epoch": 1.34351757718444,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004828758209496278,
      "loss": 6.1399,
      "mean_token_accuracy": 0.13879499062895775,
      "num_tokens": 27174480.0,
      "step": 12555
    },
    {
      "entropy": 6.302602243423462,
      "epoch": 1.344052651292204,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048286118787579983,
      "loss": 6.1052,
      "mean_token_accuracy": 0.14200446233153344,
      "num_tokens": 27184546.0,
      "step": 12560
    },
    {
      "entropy": 6.273333787918091,
      "epoch": 1.3445877253999678,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048284654880000933,
      "loss": 5.9998,
      "mean_token_accuracy": 0.14740002453327178,
      "num_tokens": 27195093.0,
      "step": 12565
    },
    {
      "entropy": 6.371186256408691,
      "epoch": 1.345122799507732,
      "grad_norm": 1.25,
      "learning_rate": 0.00048283190372267886,
      "loss": 6.1561,
      "mean_token_accuracy": 0.14448166787624359,
      "num_tokens": 27206416.0,
      "step": 12570
    },
    {
      "entropy": 6.383977174758911,
      "epoch": 1.3456578736154958,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004828172526442314,
      "loss": 6.1022,
      "mean_token_accuracy": 0.14288169294595718,
      "num_tokens": 27215533.0,
      "step": 12575
    },
    {
      "entropy": 6.316321945190429,
      "epoch": 1.3461929477232597,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048280259556509,
      "loss": 6.1472,
      "mean_token_accuracy": 0.14412765949964523,
      "num_tokens": 27225985.0,
      "step": 12580
    },
    {
      "entropy": 6.31582202911377,
      "epoch": 1.3467280218310236,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004827879324856778,
      "loss": 6.1136,
      "mean_token_accuracy": 0.1440149188041687,
      "num_tokens": 27235967.0,
      "step": 12585
    },
    {
      "entropy": 6.333903360366821,
      "epoch": 1.3472630959387875,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048277326340641835,
      "loss": 6.0599,
      "mean_token_accuracy": 0.14773117005825043,
      "num_tokens": 27246854.0,
      "step": 12590
    },
    {
      "entropy": 6.3298054218292235,
      "epoch": 1.3477981700465516,
      "grad_norm": 1.25,
      "learning_rate": 0.000482758588327735,
      "loss": 6.1729,
      "mean_token_accuracy": 0.1434890255331993,
      "num_tokens": 27256640.0,
      "step": 12595
    },
    {
      "entropy": 6.3607079029083256,
      "epoch": 1.3483332441543154,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004827439072500516,
      "loss": 6.1466,
      "mean_token_accuracy": 0.13965646550059319,
      "num_tokens": 27266907.0,
      "step": 12600
    },
    {
      "entropy": 6.371982574462891,
      "epoch": 1.3488683182620793,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048272922017379217,
      "loss": 6.1202,
      "mean_token_accuracy": 0.14218313619494438,
      "num_tokens": 27277089.0,
      "step": 12605
    },
    {
      "entropy": 6.367926025390625,
      "epoch": 1.3494033923698432,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004827145270993807,
      "loss": 6.1024,
      "mean_token_accuracy": 0.14686949849128722,
      "num_tokens": 27286781.0,
      "step": 12610
    },
    {
      "entropy": 6.3627606391906735,
      "epoch": 1.349938466477607,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004826998280272414,
      "loss": 6.131,
      "mean_token_accuracy": 0.15075902491807938,
      "num_tokens": 27298011.0,
      "step": 12615
    },
    {
      "entropy": 6.350914525985718,
      "epoch": 1.3504735405853712,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048268512295779873,
      "loss": 6.1629,
      "mean_token_accuracy": 0.14318569153547286,
      "num_tokens": 27309052.0,
      "step": 12620
    },
    {
      "entropy": 6.31253571510315,
      "epoch": 1.351008614693135,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004826704118914773,
      "loss": 6.0715,
      "mean_token_accuracy": 0.1457437202334404,
      "num_tokens": 27319945.0,
      "step": 12625
    },
    {
      "entropy": 6.308938646316529,
      "epoch": 1.351543688800899,
      "grad_norm": 1.25,
      "learning_rate": 0.000482655694828702,
      "loss": 6.1134,
      "mean_token_accuracy": 0.1483929082751274,
      "num_tokens": 27330543.0,
      "step": 12630
    },
    {
      "entropy": 6.335883855819702,
      "epoch": 1.3520787629086628,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004826409717698976,
      "loss": 6.1744,
      "mean_token_accuracy": 0.1371154509484768,
      "num_tokens": 27341957.0,
      "step": 12635
    },
    {
      "entropy": 6.382826566696167,
      "epoch": 1.3526138370164267,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048262624271548936,
      "loss": 6.0965,
      "mean_token_accuracy": 0.13875237628817558,
      "num_tokens": 27351506.0,
      "step": 12640
    },
    {
      "entropy": 6.342896032333374,
      "epoch": 1.3531489111241908,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048261150766590257,
      "loss": 6.0735,
      "mean_token_accuracy": 0.13964856415987015,
      "num_tokens": 27362208.0,
      "step": 12645
    },
    {
      "entropy": 6.326386165618897,
      "epoch": 1.3536839852319547,
      "grad_norm": 1.25,
      "learning_rate": 0.00048259676662156263,
      "loss": 6.107,
      "mean_token_accuracy": 0.14293193817138672,
      "num_tokens": 27372890.0,
      "step": 12650
    },
    {
      "entropy": 6.263306617736816,
      "epoch": 1.3542190593397185,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004825820195828952,
      "loss": 6.0561,
      "mean_token_accuracy": 0.14772369265556334,
      "num_tokens": 27382879.0,
      "step": 12655
    },
    {
      "entropy": 6.301702070236206,
      "epoch": 1.3547541334474824,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004825672665503262,
      "loss": 6.0886,
      "mean_token_accuracy": 0.14625398963689804,
      "num_tokens": 27393878.0,
      "step": 12660
    },
    {
      "entropy": 6.372119379043579,
      "epoch": 1.3552892075552463,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004825525075242815,
      "loss": 6.1826,
      "mean_token_accuracy": 0.1390667237341404,
      "num_tokens": 27404928.0,
      "step": 12665
    },
    {
      "entropy": 6.375792932510376,
      "epoch": 1.3558242816630104,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004825377425051873,
      "loss": 6.1081,
      "mean_token_accuracy": 0.13962594643235207,
      "num_tokens": 27417473.0,
      "step": 12670
    },
    {
      "entropy": 6.402952718734741,
      "epoch": 1.3563593557707743,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048252297149346997,
      "loss": 6.1905,
      "mean_token_accuracy": 0.1376357562839985,
      "num_tokens": 27428378.0,
      "step": 12675
    },
    {
      "entropy": 6.2693507194519045,
      "epoch": 1.3568944298785381,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048250819448955595,
      "loss": 6.1327,
      "mean_token_accuracy": 0.14724907279014587,
      "num_tokens": 27438815.0,
      "step": 12680
    },
    {
      "entropy": 6.394313192367553,
      "epoch": 1.357429503986302,
      "grad_norm": 1.265625,
      "learning_rate": 0.000482493411493872,
      "loss": 6.1789,
      "mean_token_accuracy": 0.14177417978644372,
      "num_tokens": 27449295.0,
      "step": 12685
    },
    {
      "entropy": 6.326571321487426,
      "epoch": 1.357964578094066,
      "grad_norm": 1.265625,
      "learning_rate": 0.000482478622506845,
      "loss": 6.051,
      "mean_token_accuracy": 0.1486498162150383,
      "num_tokens": 27459357.0,
      "step": 12690
    },
    {
      "entropy": 6.3571165084838865,
      "epoch": 1.35849965220183,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004824638275289019,
      "loss": 6.0825,
      "mean_token_accuracy": 0.14484327286481857,
      "num_tokens": 27469405.0,
      "step": 12695
    },
    {
      "entropy": 6.348760747909546,
      "epoch": 1.3590347263095939,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004824490265604699,
      "loss": 6.1069,
      "mean_token_accuracy": 0.14078130051493645,
      "num_tokens": 27478818.0,
      "step": 12700
    },
    {
      "entropy": 6.280845355987549,
      "epoch": 1.3595698004173578,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004824342196019764,
      "loss": 6.0507,
      "mean_token_accuracy": 0.14911493584513663,
      "num_tokens": 27488631.0,
      "step": 12705
    },
    {
      "entropy": 6.326426839828491,
      "epoch": 1.3601048745251219,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048241940665384904,
      "loss": 6.1453,
      "mean_token_accuracy": 0.13811727315187455,
      "num_tokens": 27500895.0,
      "step": 12710
    },
    {
      "entropy": 6.347449922561646,
      "epoch": 1.3606399486328857,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004824045877165154,
      "loss": 6.1615,
      "mean_token_accuracy": 0.13618271350860595,
      "num_tokens": 27512578.0,
      "step": 12715
    },
    {
      "entropy": 6.363177537918091,
      "epoch": 1.3611750227406496,
      "grad_norm": 1.25,
      "learning_rate": 0.00048238976279040344,
      "loss": 6.1255,
      "mean_token_accuracy": 0.14161261320114135,
      "num_tokens": 27523448.0,
      "step": 12720
    },
    {
      "entropy": 6.331499195098877,
      "epoch": 1.3617100968484135,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004823749318759413,
      "loss": 6.1938,
      "mean_token_accuracy": 0.1380709171295166,
      "num_tokens": 27534919.0,
      "step": 12725
    },
    {
      "entropy": 6.28547854423523,
      "epoch": 1.3622451709561774,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000482360094973557,
      "loss": 6.0012,
      "mean_token_accuracy": 0.14526195004582404,
      "num_tokens": 27545603.0,
      "step": 12730
    },
    {
      "entropy": 6.389470863342285,
      "epoch": 1.3627802450639415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004823452520836792,
      "loss": 6.146,
      "mean_token_accuracy": 0.14062326923012733,
      "num_tokens": 27556868.0,
      "step": 12735
    },
    {
      "entropy": 6.478588104248047,
      "epoch": 1.3633153191717053,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048233040320673634,
      "loss": 6.2815,
      "mean_token_accuracy": 0.1329442009329796,
      "num_tokens": 27567200.0,
      "step": 12740
    },
    {
      "entropy": 6.323559951782227,
      "epoch": 1.3638503932794692,
      "grad_norm": 5.6875,
      "learning_rate": 0.00048231554834315716,
      "loss": 6.0737,
      "mean_token_accuracy": 0.14370636716485025,
      "num_tokens": 27577759.0,
      "step": 12745
    },
    {
      "entropy": 6.344102764129639,
      "epoch": 1.364385467387233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004823006874933708,
      "loss": 6.1811,
      "mean_token_accuracy": 0.1394178532063961,
      "num_tokens": 27588935.0,
      "step": 12750
    },
    {
      "entropy": 6.3698286533355715,
      "epoch": 1.364920541494997,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004822858206578061,
      "loss": 6.1124,
      "mean_token_accuracy": 0.1395154193043709,
      "num_tokens": 27600306.0,
      "step": 12755
    },
    {
      "entropy": 6.354205513000489,
      "epoch": 1.365455615602761,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048227094783689244,
      "loss": 6.1707,
      "mean_token_accuracy": 0.1449229322373867,
      "num_tokens": 27612322.0,
      "step": 12760
    },
    {
      "entropy": 6.3534095287323,
      "epoch": 1.365990689710525,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004822560690310594,
      "loss": 6.1098,
      "mean_token_accuracy": 0.1513633280992508,
      "num_tokens": 27623812.0,
      "step": 12765
    },
    {
      "entropy": 6.276389074325562,
      "epoch": 1.3665257638182888,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048224118424073644,
      "loss": 6.0115,
      "mean_token_accuracy": 0.14661350548267366,
      "num_tokens": 27633848.0,
      "step": 12770
    },
    {
      "entropy": 6.336257076263427,
      "epoch": 1.3670608379260527,
      "grad_norm": 1.71875,
      "learning_rate": 0.00048222629346635336,
      "loss": 6.133,
      "mean_token_accuracy": 0.14769816547632217,
      "num_tokens": 27644641.0,
      "step": 12775
    },
    {
      "entropy": 6.3297466278076175,
      "epoch": 1.3675959120338166,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048221139670834025,
      "loss": 6.1109,
      "mean_token_accuracy": 0.1479888990521431,
      "num_tokens": 27655552.0,
      "step": 12780
    },
    {
      "entropy": 6.294423961639405,
      "epoch": 1.3681309861415807,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004821964939671272,
      "loss": 6.1205,
      "mean_token_accuracy": 0.14775934368371962,
      "num_tokens": 27666396.0,
      "step": 12785
    },
    {
      "entropy": 6.366813516616821,
      "epoch": 1.3686660602493446,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004821815852431444,
      "loss": 6.0537,
      "mean_token_accuracy": 0.1459631234407425,
      "num_tokens": 27676522.0,
      "step": 12790
    },
    {
      "entropy": 6.312767410278321,
      "epoch": 1.3692011343571084,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048216667053682243,
      "loss": 6.0818,
      "mean_token_accuracy": 0.14403568655252458,
      "num_tokens": 27687088.0,
      "step": 12795
    },
    {
      "entropy": 6.3077850341796875,
      "epoch": 1.3697362084648723,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048215174984859207,
      "loss": 6.0929,
      "mean_token_accuracy": 0.14711561053991318,
      "num_tokens": 27697587.0,
      "step": 12800
    },
    {
      "entropy": 6.270805072784424,
      "epoch": 1.3702712825726362,
      "grad_norm": 1.078125,
      "learning_rate": 0.000482136823178884,
      "loss": 6.0209,
      "mean_token_accuracy": 0.1522081770002842,
      "num_tokens": 27708345.0,
      "step": 12805
    },
    {
      "entropy": 6.396086883544922,
      "epoch": 1.3708063566804003,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004821218905281292,
      "loss": 6.2428,
      "mean_token_accuracy": 0.13731568232178687,
      "num_tokens": 27719592.0,
      "step": 12810
    },
    {
      "entropy": 6.346749401092529,
      "epoch": 1.3713414307881642,
      "grad_norm": 1.25,
      "learning_rate": 0.000482106951896759,
      "loss": 6.1266,
      "mean_token_accuracy": 0.1436546340584755,
      "num_tokens": 27730071.0,
      "step": 12815
    },
    {
      "entropy": 6.312529563903809,
      "epoch": 1.371876504895928,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048209200728520466,
      "loss": 6.1434,
      "mean_token_accuracy": 0.13986873626708984,
      "num_tokens": 27740514.0,
      "step": 12820
    },
    {
      "entropy": 6.378898191452026,
      "epoch": 1.372411579003692,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048207705669389765,
      "loss": 6.1563,
      "mean_token_accuracy": 0.1392726831138134,
      "num_tokens": 27752088.0,
      "step": 12825
    },
    {
      "entropy": 6.42064847946167,
      "epoch": 1.3729466531114558,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048206210012326976,
      "loss": 6.1855,
      "mean_token_accuracy": 0.13632684126496314,
      "num_tokens": 27763401.0,
      "step": 12830
    },
    {
      "entropy": 6.371445798873902,
      "epoch": 1.37348172721922,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048204713757375283,
      "loss": 6.1132,
      "mean_token_accuracy": 0.14433660358190536,
      "num_tokens": 27773665.0,
      "step": 12835
    },
    {
      "entropy": 6.316506719589233,
      "epoch": 1.3740168013269838,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004820321690457789,
      "loss": 6.1177,
      "mean_token_accuracy": 0.14611696600914,
      "num_tokens": 27785474.0,
      "step": 12840
    },
    {
      "entropy": 6.307301187515259,
      "epoch": 1.3745518754347477,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048201719453978023,
      "loss": 6.1505,
      "mean_token_accuracy": 0.14330882132053374,
      "num_tokens": 27796520.0,
      "step": 12845
    },
    {
      "entropy": 6.403099298477173,
      "epoch": 1.3750869495425118,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004820022140561891,
      "loss": 6.2134,
      "mean_token_accuracy": 0.13672758489847184,
      "num_tokens": 27807559.0,
      "step": 12850
    },
    {
      "entropy": 6.36442813873291,
      "epoch": 1.3756220236502756,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004819872275954381,
      "loss": 6.1511,
      "mean_token_accuracy": 0.14052387177944184,
      "num_tokens": 27818208.0,
      "step": 12855
    },
    {
      "entropy": 6.343287563323974,
      "epoch": 1.3761570977580395,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048197223515796,
      "loss": 6.0488,
      "mean_token_accuracy": 0.14792466908693314,
      "num_tokens": 27828238.0,
      "step": 12860
    },
    {
      "entropy": 6.35619626045227,
      "epoch": 1.3766921718658034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004819572367441877,
      "loss": 6.147,
      "mean_token_accuracy": 0.1505136713385582,
      "num_tokens": 27837885.0,
      "step": 12865
    },
    {
      "entropy": 6.314773416519165,
      "epoch": 1.3772272459735673,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004819422323545543,
      "loss": 6.0492,
      "mean_token_accuracy": 0.1532662570476532,
      "num_tokens": 27847425.0,
      "step": 12870
    },
    {
      "entropy": 6.250513648986816,
      "epoch": 1.3777623200813314,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000481927221989493,
      "loss": 6.0936,
      "mean_token_accuracy": 0.13962285369634628,
      "num_tokens": 27859606.0,
      "step": 12875
    },
    {
      "entropy": 6.443209075927735,
      "epoch": 1.3782973941890952,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004819122056494373,
      "loss": 6.1652,
      "mean_token_accuracy": 0.144085530936718,
      "num_tokens": 27871651.0,
      "step": 12880
    },
    {
      "entropy": 6.450758934020996,
      "epoch": 1.3788324682968591,
      "grad_norm": 12.375,
      "learning_rate": 0.00048189718333482064,
      "loss": 6.0985,
      "mean_token_accuracy": 0.14231204390525817,
      "num_tokens": 27883138.0,
      "step": 12885
    },
    {
      "entropy": 6.4605823993682865,
      "epoch": 1.379367542404623,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000481882155046077,
      "loss": 6.1716,
      "mean_token_accuracy": 0.14429069608449935,
      "num_tokens": 27893597.0,
      "step": 12890
    },
    {
      "entropy": 6.358666038513183,
      "epoch": 1.3799026165123869,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048186712078364006,
      "loss": 6.0787,
      "mean_token_accuracy": 0.1497125156223774,
      "num_tokens": 27904249.0,
      "step": 12895
    },
    {
      "entropy": 6.2796509742736815,
      "epoch": 1.380437690620151,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048185208054794414,
      "loss": 6.1303,
      "mean_token_accuracy": 0.13967967927455902,
      "num_tokens": 27914713.0,
      "step": 12900
    },
    {
      "entropy": 6.288844394683838,
      "epoch": 1.3809727647279149,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004818370343394235,
      "loss": 6.1416,
      "mean_token_accuracy": 0.1500673472881317,
      "num_tokens": 27926257.0,
      "step": 12905
    },
    {
      "entropy": 6.333261156082154,
      "epoch": 1.3815078388356787,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004818219821585125,
      "loss": 6.1145,
      "mean_token_accuracy": 0.1445947989821434,
      "num_tokens": 27936839.0,
      "step": 12910
    },
    {
      "entropy": 6.422682809829712,
      "epoch": 1.3820429129434426,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048180692400564586,
      "loss": 6.2217,
      "mean_token_accuracy": 0.14021293073892593,
      "num_tokens": 27946318.0,
      "step": 12915
    },
    {
      "entropy": 6.424679946899414,
      "epoch": 1.3825779870512065,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048179185988125834,
      "loss": 6.2155,
      "mean_token_accuracy": 0.13530697524547577,
      "num_tokens": 27956990.0,
      "step": 12920
    },
    {
      "entropy": 6.423383378982544,
      "epoch": 1.3831130611589706,
      "grad_norm": 1.328125,
      "learning_rate": 0.000481776789785785,
      "loss": 6.148,
      "mean_token_accuracy": 0.13681581243872643,
      "num_tokens": 27968187.0,
      "step": 12925
    },
    {
      "entropy": 6.4501148700714115,
      "epoch": 1.3836481352667345,
      "grad_norm": 1.859375,
      "learning_rate": 0.00048176171371966083,
      "loss": 6.1846,
      "mean_token_accuracy": 0.143385910987854,
      "num_tokens": 27978337.0,
      "step": 12930
    },
    {
      "entropy": 6.329929447174072,
      "epoch": 1.3841832093744983,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004817466316833212,
      "loss": 6.1538,
      "mean_token_accuracy": 0.1365984782576561,
      "num_tokens": 27988681.0,
      "step": 12935
    },
    {
      "entropy": 6.259853029251099,
      "epoch": 1.3847182834822622,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004817315436772017,
      "loss": 6.0332,
      "mean_token_accuracy": 0.15356675833463668,
      "num_tokens": 27999233.0,
      "step": 12940
    },
    {
      "entropy": 6.410255002975464,
      "epoch": 1.385253357590026,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004817164497017379,
      "loss": 6.1775,
      "mean_token_accuracy": 0.14426740929484366,
      "num_tokens": 28010242.0,
      "step": 12945
    },
    {
      "entropy": 6.327193880081177,
      "epoch": 1.3857884316977902,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004817013497573656,
      "loss": 6.1598,
      "mean_token_accuracy": 0.1423150584101677,
      "num_tokens": 28020424.0,
      "step": 12950
    },
    {
      "entropy": 6.396161365509033,
      "epoch": 1.386323505805554,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000481686243844521,
      "loss": 6.1153,
      "mean_token_accuracy": 0.14187586680054665,
      "num_tokens": 28030287.0,
      "step": 12955
    },
    {
      "entropy": 6.405784940719604,
      "epoch": 1.386858579913318,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048167113196364,
      "loss": 6.0839,
      "mean_token_accuracy": 0.14374205619096755,
      "num_tokens": 28039997.0,
      "step": 12960
    },
    {
      "entropy": 6.331779670715332,
      "epoch": 1.387393654021082,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004816560141151593,
      "loss": 6.1156,
      "mean_token_accuracy": 0.13659428134560586,
      "num_tokens": 28051381.0,
      "step": 12965
    },
    {
      "entropy": 6.323930454254151,
      "epoch": 1.3879287281288457,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004816408902995151,
      "loss": 6.0818,
      "mean_token_accuracy": 0.142775297164917,
      "num_tokens": 28063603.0,
      "step": 12970
    },
    {
      "entropy": 6.302712440490723,
      "epoch": 1.3884638022366098,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004816257605171442,
      "loss": 6.0669,
      "mean_token_accuracy": 0.15081416070461273,
      "num_tokens": 28073942.0,
      "step": 12975
    },
    {
      "entropy": 6.323858213424683,
      "epoch": 1.3889988763443737,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048161062476848363,
      "loss": 6.1204,
      "mean_token_accuracy": 0.1460118055343628,
      "num_tokens": 28083837.0,
      "step": 12980
    },
    {
      "entropy": 6.331998300552368,
      "epoch": 1.3895339504521376,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004815954830539702,
      "loss": 6.0561,
      "mean_token_accuracy": 0.14729045182466508,
      "num_tokens": 28094150.0,
      "step": 12985
    },
    {
      "entropy": 6.30282883644104,
      "epoch": 1.3900690245599017,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048158033537404126,
      "loss": 6.1114,
      "mean_token_accuracy": 0.14873374849557877,
      "num_tokens": 28106345.0,
      "step": 12990
    },
    {
      "entropy": 6.32881908416748,
      "epoch": 1.3906040986676655,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048156518172913417,
      "loss": 6.0471,
      "mean_token_accuracy": 0.14546438828110694,
      "num_tokens": 28116858.0,
      "step": 12995
    },
    {
      "entropy": 6.354728937149048,
      "epoch": 1.3911391727754294,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004815500221196865,
      "loss": 6.2244,
      "mean_token_accuracy": 0.13930549174547197,
      "num_tokens": 28128390.0,
      "step": 13000
    },
    {
      "entropy": 6.2818399429321286,
      "epoch": 1.3916742468831933,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048153485654613587,
      "loss": 6.0723,
      "mean_token_accuracy": 0.14910908341407775,
      "num_tokens": 28139421.0,
      "step": 13005
    },
    {
      "entropy": 6.344704055786133,
      "epoch": 1.3922093209909572,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048151968500892043,
      "loss": 6.1605,
      "mean_token_accuracy": 0.1423213414847851,
      "num_tokens": 28151076.0,
      "step": 13010
    },
    {
      "entropy": 6.451549386978149,
      "epoch": 1.3927443950987213,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048150450750847795,
      "loss": 6.2703,
      "mean_token_accuracy": 0.13875743746757507,
      "num_tokens": 28162079.0,
      "step": 13015
    },
    {
      "entropy": 6.3479619979858395,
      "epoch": 1.3932794692064852,
      "grad_norm": 1.296875,
      "learning_rate": 0.000481489324045247,
      "loss": 6.1745,
      "mean_token_accuracy": 0.13940562233328818,
      "num_tokens": 28174442.0,
      "step": 13020
    },
    {
      "entropy": 6.338541460037232,
      "epoch": 1.393814543314249,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048147413461966564,
      "loss": 6.0411,
      "mean_token_accuracy": 0.14941960275173188,
      "num_tokens": 28184643.0,
      "step": 13025
    },
    {
      "entropy": 6.239337158203125,
      "epoch": 1.394349617422013,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048145893923217276,
      "loss": 6.0716,
      "mean_token_accuracy": 0.14679452478885652,
      "num_tokens": 28194021.0,
      "step": 13030
    },
    {
      "entropy": 6.380154132843018,
      "epoch": 1.3948846915297768,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004814437378832071,
      "loss": 6.1293,
      "mean_token_accuracy": 0.13931864127516747,
      "num_tokens": 28204750.0,
      "step": 13035
    },
    {
      "entropy": 6.446542549133301,
      "epoch": 1.3954197656375409,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004814285305732074,
      "loss": 6.205,
      "mean_token_accuracy": 0.140470639616251,
      "num_tokens": 28216499.0,
      "step": 13040
    },
    {
      "entropy": 6.384186506271362,
      "epoch": 1.3959548397453048,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048141331730261293,
      "loss": 6.1089,
      "mean_token_accuracy": 0.14840888231992722,
      "num_tokens": 28226573.0,
      "step": 13045
    },
    {
      "entropy": 6.320884656906128,
      "epoch": 1.3964899138530686,
      "grad_norm": 1.390625,
      "learning_rate": 0.000481398098071863,
      "loss": 6.0773,
      "mean_token_accuracy": 0.1507680431008339,
      "num_tokens": 28236688.0,
      "step": 13050
    },
    {
      "entropy": 6.310642337799072,
      "epoch": 1.3970249879608325,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048138287288139686,
      "loss": 6.1567,
      "mean_token_accuracy": 0.14499348253011704,
      "num_tokens": 28247477.0,
      "step": 13055
    },
    {
      "entropy": 6.3269823551177975,
      "epoch": 1.3975600620685964,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004813676417316543,
      "loss": 6.151,
      "mean_token_accuracy": 0.14014082551002502,
      "num_tokens": 28258286.0,
      "step": 13060
    },
    {
      "entropy": 6.489237880706787,
      "epoch": 1.3980951361763605,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004813524046230751,
      "loss": 6.2922,
      "mean_token_accuracy": 0.13369591534137726,
      "num_tokens": 28268587.0,
      "step": 13065
    },
    {
      "entropy": 6.450521326065063,
      "epoch": 1.3986302102841244,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048133716155609926,
      "loss": 6.1716,
      "mean_token_accuracy": 0.13507251515984536,
      "num_tokens": 28279352.0,
      "step": 13070
    },
    {
      "entropy": 6.325395727157593,
      "epoch": 1.3991652843918883,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004813219125311668,
      "loss": 6.0935,
      "mean_token_accuracy": 0.14247297048568724,
      "num_tokens": 28290114.0,
      "step": 13075
    },
    {
      "entropy": 6.30522894859314,
      "epoch": 1.3997003584996521,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048130665754871814,
      "loss": 6.185,
      "mean_token_accuracy": 0.1400682345032692,
      "num_tokens": 28301845.0,
      "step": 13080
    },
    {
      "entropy": 6.419568824768066,
      "epoch": 1.400235432607416,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048129139660919374,
      "loss": 6.1583,
      "mean_token_accuracy": 0.13759351149201393,
      "num_tokens": 28313162.0,
      "step": 13085
    },
    {
      "entropy": 6.355233287811279,
      "epoch": 1.40077050671518,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048127612971303425,
      "loss": 6.1023,
      "mean_token_accuracy": 0.1442082144320011,
      "num_tokens": 28323888.0,
      "step": 13090
    },
    {
      "entropy": 6.362844085693359,
      "epoch": 1.401305580822944,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004812608568606805,
      "loss": 6.1403,
      "mean_token_accuracy": 0.13905975893139838,
      "num_tokens": 28333735.0,
      "step": 13095
    },
    {
      "entropy": 6.353046464920044,
      "epoch": 1.4018406549307079,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004812455780525735,
      "loss": 6.1301,
      "mean_token_accuracy": 0.13770420774817466,
      "num_tokens": 28343947.0,
      "step": 13100
    },
    {
      "entropy": 6.379090023040772,
      "epoch": 1.402375729038472,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004812302932891544,
      "loss": 6.2149,
      "mean_token_accuracy": 0.13213214948773383,
      "num_tokens": 28354635.0,
      "step": 13105
    },
    {
      "entropy": 6.366495847702026,
      "epoch": 1.4029108031462356,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004812150025708646,
      "loss": 6.1337,
      "mean_token_accuracy": 0.13718286976218225,
      "num_tokens": 28365392.0,
      "step": 13110
    },
    {
      "entropy": 6.3908545017242435,
      "epoch": 1.4034458772539997,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048119970589814557,
      "loss": 6.1226,
      "mean_token_accuracy": 0.13827935829758645,
      "num_tokens": 28376527.0,
      "step": 13115
    },
    {
      "entropy": 6.360084056854248,
      "epoch": 1.4039809513617636,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004811844032714389,
      "loss": 6.1287,
      "mean_token_accuracy": 0.144601196795702,
      "num_tokens": 28387178.0,
      "step": 13120
    },
    {
      "entropy": 6.3346864700317385,
      "epoch": 1.4045160254695275,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048116909469118663,
      "loss": 6.0341,
      "mean_token_accuracy": 0.14410272389650344,
      "num_tokens": 28397578.0,
      "step": 13125
    },
    {
      "entropy": 6.320469903945923,
      "epoch": 1.4050510995772916,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004811537801578308,
      "loss": 6.1831,
      "mean_token_accuracy": 0.13533378541469573,
      "num_tokens": 28408933.0,
      "step": 13130
    },
    {
      "entropy": 6.343297910690308,
      "epoch": 1.4055861736850555,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048113845967181346,
      "loss": 6.145,
      "mean_token_accuracy": 0.13573747873306274,
      "num_tokens": 28421906.0,
      "step": 13135
    },
    {
      "entropy": 6.364819526672363,
      "epoch": 1.4061212477928193,
      "grad_norm": 1.390625,
      "learning_rate": 0.000481123133233577,
      "loss": 6.0903,
      "mean_token_accuracy": 0.14649079516530036,
      "num_tokens": 28432677.0,
      "step": 13140
    },
    {
      "entropy": 6.3343177318573,
      "epoch": 1.4066563219005832,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048110780084356414,
      "loss": 6.1881,
      "mean_token_accuracy": 0.14125317335128784,
      "num_tokens": 28443795.0,
      "step": 13145
    },
    {
      "entropy": 6.337538623809815,
      "epoch": 1.407191396008347,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004810924625022174,
      "loss": 6.1633,
      "mean_token_accuracy": 0.1392707794904709,
      "num_tokens": 28455727.0,
      "step": 13150
    },
    {
      "entropy": 6.44963755607605,
      "epoch": 1.4077264701161112,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004810771182099799,
      "loss": 6.1774,
      "mean_token_accuracy": 0.13331684544682504,
      "num_tokens": 28467554.0,
      "step": 13155
    },
    {
      "entropy": 6.459814548492432,
      "epoch": 1.408261544223875,
      "grad_norm": 1.125,
      "learning_rate": 0.0004810617679672945,
      "loss": 6.1962,
      "mean_token_accuracy": 0.14775989055633545,
      "num_tokens": 28479268.0,
      "step": 13160
    },
    {
      "entropy": 6.41382360458374,
      "epoch": 1.408796618331639,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048104641177460443,
      "loss": 6.1831,
      "mean_token_accuracy": 0.13741683214902878,
      "num_tokens": 28489887.0,
      "step": 13165
    },
    {
      "entropy": 6.2944518566131595,
      "epoch": 1.4093316924394028,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004810310496323533,
      "loss": 6.0687,
      "mean_token_accuracy": 0.14717003852128982,
      "num_tokens": 28500736.0,
      "step": 13170
    },
    {
      "entropy": 6.27557053565979,
      "epoch": 1.4098667665471667,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048101568154098446,
      "loss": 6.0479,
      "mean_token_accuracy": 0.15149986892938613,
      "num_tokens": 28511458.0,
      "step": 13175
    },
    {
      "entropy": 6.406965541839599,
      "epoch": 1.4104018406549308,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048100030750094185,
      "loss": 6.1616,
      "mean_token_accuracy": 0.13601041883230208,
      "num_tokens": 28522354.0,
      "step": 13180
    },
    {
      "entropy": 6.3143298625946045,
      "epoch": 1.4109369147626947,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004809849275126693,
      "loss": 6.0145,
      "mean_token_accuracy": 0.15260105729103088,
      "num_tokens": 28533061.0,
      "step": 13185
    },
    {
      "entropy": 6.345221853256225,
      "epoch": 1.4114719888704585,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048096954157661085,
      "loss": 6.1544,
      "mean_token_accuracy": 0.1399749018251896,
      "num_tokens": 28544230.0,
      "step": 13190
    },
    {
      "entropy": 6.3634875297546385,
      "epoch": 1.4120070629782224,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048095414969321086,
      "loss": 6.1638,
      "mean_token_accuracy": 0.14462606012821197,
      "num_tokens": 28555461.0,
      "step": 13195
    },
    {
      "entropy": 6.366601181030274,
      "epoch": 1.4125421370859863,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048093875186291376,
      "loss": 6.0618,
      "mean_token_accuracy": 0.14818851351737977,
      "num_tokens": 28565992.0,
      "step": 13200
    },
    {
      "entropy": 6.365614366531372,
      "epoch": 1.4130772111937504,
      "grad_norm": 1.5,
      "learning_rate": 0.00048092334808616413,
      "loss": 6.0967,
      "mean_token_accuracy": 0.14526073187589644,
      "num_tokens": 28575802.0,
      "step": 13205
    },
    {
      "entropy": 6.296118307113647,
      "epoch": 1.4136122853015143,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004809079383634067,
      "loss": 6.0703,
      "mean_token_accuracy": 0.146048341691494,
      "num_tokens": 28586982.0,
      "step": 13210
    },
    {
      "entropy": 6.340404224395752,
      "epoch": 1.4141473594092782,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048089252269508656,
      "loss": 6.1685,
      "mean_token_accuracy": 0.14238498657941817,
      "num_tokens": 28598267.0,
      "step": 13215
    },
    {
      "entropy": 6.410357284545898,
      "epoch": 1.414682433517042,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048087710108164874,
      "loss": 6.1464,
      "mean_token_accuracy": 0.1398368440568447,
      "num_tokens": 28608213.0,
      "step": 13220
    },
    {
      "entropy": 6.309383535385132,
      "epoch": 1.415217507624806,
      "grad_norm": 1.25,
      "learning_rate": 0.0004808616735235385,
      "loss": 6.1113,
      "mean_token_accuracy": 0.13841313123703003,
      "num_tokens": 28619674.0,
      "step": 13225
    },
    {
      "entropy": 6.295738554000854,
      "epoch": 1.41575258173257,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004808462400212014,
      "loss": 6.0444,
      "mean_token_accuracy": 0.14815330803394317,
      "num_tokens": 28630824.0,
      "step": 13230
    },
    {
      "entropy": 6.31119475364685,
      "epoch": 1.416287655840334,
      "grad_norm": 2.484375,
      "learning_rate": 0.00048083080057508313,
      "loss": 6.1635,
      "mean_token_accuracy": 0.14462573900818826,
      "num_tokens": 28642270.0,
      "step": 13235
    },
    {
      "entropy": 6.364747428894043,
      "epoch": 1.4168227299480978,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048081535518562933,
      "loss": 6.1698,
      "mean_token_accuracy": 0.13775825053453444,
      "num_tokens": 28653649.0,
      "step": 13240
    },
    {
      "entropy": 6.347034215927124,
      "epoch": 1.4173578040558619,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000480799903853286,
      "loss": 6.0193,
      "mean_token_accuracy": 0.14273491725325585,
      "num_tokens": 28664044.0,
      "step": 13245
    },
    {
      "entropy": 6.356173038482666,
      "epoch": 1.4178928781636255,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004807844465784995,
      "loss": 6.1277,
      "mean_token_accuracy": 0.14577305614948272,
      "num_tokens": 28675394.0,
      "step": 13250
    },
    {
      "entropy": 6.305230760574341,
      "epoch": 1.4184279522713896,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048076898336171596,
      "loss": 6.0899,
      "mean_token_accuracy": 0.14937041476368904,
      "num_tokens": 28686292.0,
      "step": 13255
    },
    {
      "entropy": 6.350478172302246,
      "epoch": 1.4189630263791535,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048075351420338196,
      "loss": 6.1408,
      "mean_token_accuracy": 0.14343406707048417,
      "num_tokens": 28696928.0,
      "step": 13260
    },
    {
      "entropy": 6.366587686538696,
      "epoch": 1.4194981004869174,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004807380391039441,
      "loss": 6.1525,
      "mean_token_accuracy": 0.1384634904563427,
      "num_tokens": 28707986.0,
      "step": 13265
    },
    {
      "entropy": 6.432022333145142,
      "epoch": 1.4200331745946815,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048072255806384936,
      "loss": 6.164,
      "mean_token_accuracy": 0.14104801416397095,
      "num_tokens": 28719296.0,
      "step": 13270
    },
    {
      "entropy": 6.2813694953918455,
      "epoch": 1.4205682487024454,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004807070710835447,
      "loss": 6.0246,
      "mean_token_accuracy": 0.1526571676135063,
      "num_tokens": 28729523.0,
      "step": 13275
    },
    {
      "entropy": 6.366316127777099,
      "epoch": 1.4211033228102092,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004806915781634771,
      "loss": 6.1403,
      "mean_token_accuracy": 0.1422620452940464,
      "num_tokens": 28739026.0,
      "step": 13280
    },
    {
      "entropy": 6.370488929748535,
      "epoch": 1.4216383969179731,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004806760793040942,
      "loss": 6.2112,
      "mean_token_accuracy": 0.14133259057998657,
      "num_tokens": 28749513.0,
      "step": 13285
    },
    {
      "entropy": 6.403868198394775,
      "epoch": 1.422173471025737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048066057450584334,
      "loss": 6.0992,
      "mean_token_accuracy": 0.1418754369020462,
      "num_tokens": 28760722.0,
      "step": 13290
    },
    {
      "entropy": 6.382450437545776,
      "epoch": 1.422708545133501,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004806450637691723,
      "loss": 6.0754,
      "mean_token_accuracy": 0.15397344678640365,
      "num_tokens": 28771235.0,
      "step": 13295
    },
    {
      "entropy": 6.321266078948975,
      "epoch": 1.423243619241265,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048062954709452907,
      "loss": 6.0825,
      "mean_token_accuracy": 0.14391638487577438,
      "num_tokens": 28782642.0,
      "step": 13300
    },
    {
      "entropy": 6.322711706161499,
      "epoch": 1.4237786933490288,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004806140244823615,
      "loss": 6.0881,
      "mean_token_accuracy": 0.14307290315628052,
      "num_tokens": 28793628.0,
      "step": 13305
    },
    {
      "entropy": 6.3728859424591064,
      "epoch": 1.4243137674567927,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048059849593311776,
      "loss": 6.16,
      "mean_token_accuracy": 0.14480257406830788,
      "num_tokens": 28804034.0,
      "step": 13310
    },
    {
      "entropy": 6.353333902359009,
      "epoch": 1.4248488415645566,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004805829614472464,
      "loss": 6.1063,
      "mean_token_accuracy": 0.1414711058139801,
      "num_tokens": 28814679.0,
      "step": 13315
    },
    {
      "entropy": 6.328074932098389,
      "epoch": 1.4253839156723207,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048056742102519593,
      "loss": 6.1293,
      "mean_token_accuracy": 0.14373231530189515,
      "num_tokens": 28825130.0,
      "step": 13320
    },
    {
      "entropy": 6.34884877204895,
      "epoch": 1.4259189897800846,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048055187466741506,
      "loss": 6.158,
      "mean_token_accuracy": 0.13860179707407952,
      "num_tokens": 28835500.0,
      "step": 13325
    },
    {
      "entropy": 6.3502074718475345,
      "epoch": 1.4264540638878485,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004805363223743527,
      "loss": 6.0994,
      "mean_token_accuracy": 0.14385845363140107,
      "num_tokens": 28845794.0,
      "step": 13330
    },
    {
      "entropy": 6.4151159763336185,
      "epoch": 1.4269891379956123,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004805207641464579,
      "loss": 6.1218,
      "mean_token_accuracy": 0.1367532953619957,
      "num_tokens": 28856326.0,
      "step": 13335
    },
    {
      "entropy": 6.285000658035278,
      "epoch": 1.4275242121033762,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004805051999841799,
      "loss": 6.1044,
      "mean_token_accuracy": 0.1405942842364311,
      "num_tokens": 28867592.0,
      "step": 13340
    },
    {
      "entropy": 6.412256145477295,
      "epoch": 1.4280592862111403,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004804896298879682,
      "loss": 6.2381,
      "mean_token_accuracy": 0.13093710467219352,
      "num_tokens": 28880352.0,
      "step": 13345
    },
    {
      "entropy": 6.442687702178955,
      "epoch": 1.4285943603189042,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004804740538582723,
      "loss": 6.1135,
      "mean_token_accuracy": 0.14259385466575622,
      "num_tokens": 28892012.0,
      "step": 13350
    },
    {
      "entropy": 6.383967447280884,
      "epoch": 1.429129434426668,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004804584718955419,
      "loss": 6.08,
      "mean_token_accuracy": 0.1458110511302948,
      "num_tokens": 28902247.0,
      "step": 13355
    },
    {
      "entropy": 6.290881109237671,
      "epoch": 1.429664508534432,
      "grad_norm": 1.46875,
      "learning_rate": 0.000480442884000227,
      "loss": 6.2035,
      "mean_token_accuracy": 0.1396358996629715,
      "num_tokens": 28913222.0,
      "step": 13360
    },
    {
      "entropy": 6.327722883224487,
      "epoch": 1.4301995826421958,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004804272901727778,
      "loss": 6.1058,
      "mean_token_accuracy": 0.14039731174707412,
      "num_tokens": 28924038.0,
      "step": 13365
    },
    {
      "entropy": 6.378175210952759,
      "epoch": 1.43073465674996,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048041169041364437,
      "loss": 6.1408,
      "mean_token_accuracy": 0.14304384142160415,
      "num_tokens": 28934448.0,
      "step": 13370
    },
    {
      "entropy": 6.33401198387146,
      "epoch": 1.4312697308577238,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048039608472327724,
      "loss": 6.1008,
      "mean_token_accuracy": 0.14583365693688394,
      "num_tokens": 28945175.0,
      "step": 13375
    },
    {
      "entropy": 6.33856234550476,
      "epoch": 1.4318048049654877,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048038047310212706,
      "loss": 6.0647,
      "mean_token_accuracy": 0.14740571677684783,
      "num_tokens": 28955116.0,
      "step": 13380
    },
    {
      "entropy": 6.348536539077759,
      "epoch": 1.4323398790732518,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048036485555064454,
      "loss": 6.1511,
      "mean_token_accuracy": 0.14285880476236343,
      "num_tokens": 28966999.0,
      "step": 13385
    },
    {
      "entropy": 6.339142799377441,
      "epoch": 1.4328749531810157,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048034923206928065,
      "loss": 6.1571,
      "mean_token_accuracy": 0.13627411648631096,
      "num_tokens": 28977388.0,
      "step": 13390
    },
    {
      "entropy": 6.312268447875977,
      "epoch": 1.4334100272887795,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048033360265848664,
      "loss": 6.0481,
      "mean_token_accuracy": 0.14278963953256607,
      "num_tokens": 28987792.0,
      "step": 13395
    },
    {
      "entropy": 6.450326442718506,
      "epoch": 1.4339451013965434,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048031796731871364,
      "loss": 6.1821,
      "mean_token_accuracy": 0.14053000509738922,
      "num_tokens": 28998319.0,
      "step": 13400
    },
    {
      "entropy": 6.358760356903076,
      "epoch": 1.4344801755043073,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048030232605041323,
      "loss": 6.152,
      "mean_token_accuracy": 0.13976763486862182,
      "num_tokens": 29007942.0,
      "step": 13405
    },
    {
      "entropy": 6.3405077934265135,
      "epoch": 1.4350152496120714,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004802866788540369,
      "loss": 6.1124,
      "mean_token_accuracy": 0.1470702365040779,
      "num_tokens": 29018217.0,
      "step": 13410
    },
    {
      "entropy": 6.317957973480224,
      "epoch": 1.4355503237198353,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004802710257300367,
      "loss": 6.1394,
      "mean_token_accuracy": 0.1432683140039444,
      "num_tokens": 29028472.0,
      "step": 13415
    },
    {
      "entropy": 6.315882873535156,
      "epoch": 1.4360853978275991,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004802553666788644,
      "loss": 6.1056,
      "mean_token_accuracy": 0.1435897946357727,
      "num_tokens": 29040553.0,
      "step": 13420
    },
    {
      "entropy": 6.387766790390015,
      "epoch": 1.436620471935363,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048023970170097226,
      "loss": 6.1371,
      "mean_token_accuracy": 0.13718340769410134,
      "num_tokens": 29050142.0,
      "step": 13425
    },
    {
      "entropy": 6.314556455612182,
      "epoch": 1.437155546043127,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048022403079681254,
      "loss": 6.0765,
      "mean_token_accuracy": 0.13892532587051393,
      "num_tokens": 29061762.0,
      "step": 13430
    },
    {
      "entropy": 6.382308626174927,
      "epoch": 1.437690620150891,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048020835396683775,
      "loss": 6.1423,
      "mean_token_accuracy": 0.14013155698776245,
      "num_tokens": 29072095.0,
      "step": 13435
    },
    {
      "entropy": 6.42381181716919,
      "epoch": 1.4382256942586549,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048019267121150066,
      "loss": 6.17,
      "mean_token_accuracy": 0.13957416340708734,
      "num_tokens": 29082561.0,
      "step": 13440
    },
    {
      "entropy": 6.336284255981445,
      "epoch": 1.4387607683664188,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048017698253125397,
      "loss": 6.1168,
      "mean_token_accuracy": 0.14208776950836183,
      "num_tokens": 29093813.0,
      "step": 13445
    },
    {
      "entropy": 6.304681253433228,
      "epoch": 1.4392958424741826,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004801612879265507,
      "loss": 6.1186,
      "mean_token_accuracy": 0.14173825159668924,
      "num_tokens": 29104064.0,
      "step": 13450
    },
    {
      "entropy": 6.36153359413147,
      "epoch": 1.4398309165819465,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004801455873978442,
      "loss": 6.1285,
      "mean_token_accuracy": 0.14259218722581862,
      "num_tokens": 29115086.0,
      "step": 13455
    },
    {
      "entropy": 6.364845609664917,
      "epoch": 1.4403659906897106,
      "grad_norm": 1.375,
      "learning_rate": 0.0004801298809455876,
      "loss": 6.0973,
      "mean_token_accuracy": 0.15521452128887175,
      "num_tokens": 29125536.0,
      "step": 13460
    },
    {
      "entropy": 6.379434251785279,
      "epoch": 1.4409010647974745,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004801141685702345,
      "loss": 6.1778,
      "mean_token_accuracy": 0.14300027638673782,
      "num_tokens": 29137007.0,
      "step": 13465
    },
    {
      "entropy": 6.333104991912842,
      "epoch": 1.4414361389052384,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048009845027223864,
      "loss": 6.1643,
      "mean_token_accuracy": 0.1330385409295559,
      "num_tokens": 29147072.0,
      "step": 13470
    },
    {
      "entropy": 6.337000846862793,
      "epoch": 1.4419712130130022,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048008272605205377,
      "loss": 6.1202,
      "mean_token_accuracy": 0.14398149028420448,
      "num_tokens": 29157803.0,
      "step": 13475
    },
    {
      "entropy": 6.4085485458374025,
      "epoch": 1.4425062871207661,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048006699591013417,
      "loss": 6.0882,
      "mean_token_accuracy": 0.14155226051807404,
      "num_tokens": 29168173.0,
      "step": 13480
    },
    {
      "entropy": 6.413027381896972,
      "epoch": 1.4430413612285302,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004800512598469337,
      "loss": 6.1779,
      "mean_token_accuracy": 0.13701770976185798,
      "num_tokens": 29179377.0,
      "step": 13485
    },
    {
      "entropy": 6.344700527191162,
      "epoch": 1.443576435336294,
      "grad_norm": 1.21875,
      "learning_rate": 0.000480035517862907,
      "loss": 6.1667,
      "mean_token_accuracy": 0.1403437152504921,
      "num_tokens": 29190909.0,
      "step": 13490
    },
    {
      "entropy": 6.3427571773529055,
      "epoch": 1.444111509444058,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048001976995850856,
      "loss": 6.071,
      "mean_token_accuracy": 0.14756202548742295,
      "num_tokens": 29200977.0,
      "step": 13495
    },
    {
      "entropy": 6.356798601150513,
      "epoch": 1.4446465835518219,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048000401613419296,
      "loss": 6.1156,
      "mean_token_accuracy": 0.13602641299366952,
      "num_tokens": 29210702.0,
      "step": 13500
    },
    {
      "entropy": 6.358712387084961,
      "epoch": 1.4451816576595857,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047998825639041534,
      "loss": 6.1973,
      "mean_token_accuracy": 0.1366298571228981,
      "num_tokens": 29221338.0,
      "step": 13505
    },
    {
      "entropy": 6.412976264953613,
      "epoch": 1.4457167317673498,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004799724907276306,
      "loss": 6.0539,
      "mean_token_accuracy": 0.1525439664721489,
      "num_tokens": 29232091.0,
      "step": 13510
    },
    {
      "entropy": 6.2935380935668945,
      "epoch": 1.4462518058751137,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004799567191462939,
      "loss": 6.0543,
      "mean_token_accuracy": 0.14835237562656403,
      "num_tokens": 29241903.0,
      "step": 13515
    },
    {
      "entropy": 6.330442333221436,
      "epoch": 1.4467868799828776,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047994094164686074,
      "loss": 6.1738,
      "mean_token_accuracy": 0.14087651222944259,
      "num_tokens": 29253414.0,
      "step": 13520
    },
    {
      "entropy": 6.4127284526824955,
      "epoch": 1.4473219540906417,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004799251582297868,
      "loss": 6.0477,
      "mean_token_accuracy": 0.15497565716505052,
      "num_tokens": 29264066.0,
      "step": 13525
    },
    {
      "entropy": 6.342731046676636,
      "epoch": 1.4478570281984056,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047990936889552767,
      "loss": 6.11,
      "mean_token_accuracy": 0.1457957960665226,
      "num_tokens": 29274915.0,
      "step": 13530
    },
    {
      "entropy": 6.301246976852417,
      "epoch": 1.4483921023061694,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004798935736445392,
      "loss": 6.1124,
      "mean_token_accuracy": 0.1376790277659893,
      "num_tokens": 29287345.0,
      "step": 13535
    },
    {
      "entropy": 6.305440092086792,
      "epoch": 1.4489271764139333,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004798777724772777,
      "loss": 6.0876,
      "mean_token_accuracy": 0.15104828774929047,
      "num_tokens": 29297432.0,
      "step": 13540
    },
    {
      "entropy": 6.327405595779419,
      "epoch": 1.4494622505216972,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047986196539419936,
      "loss": 6.019,
      "mean_token_accuracy": 0.1488790363073349,
      "num_tokens": 29307534.0,
      "step": 13545
    },
    {
      "entropy": 6.3356616497039795,
      "epoch": 1.4499973246294613,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004798461523957604,
      "loss": 6.0776,
      "mean_token_accuracy": 0.14556969106197357,
      "num_tokens": 29319412.0,
      "step": 13550
    },
    {
      "entropy": 6.2869456768035885,
      "epoch": 1.4505323987372252,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047983033348241767,
      "loss": 6.0619,
      "mean_token_accuracy": 0.15304907262325287,
      "num_tokens": 29330477.0,
      "step": 13555
    },
    {
      "entropy": 6.2793616771698,
      "epoch": 1.451067472844989,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004798145086546278,
      "loss": 6.1199,
      "mean_token_accuracy": 0.1443927250802517,
      "num_tokens": 29341481.0,
      "step": 13560
    },
    {
      "entropy": 6.315534067153931,
      "epoch": 1.451602546952753,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004797986779128478,
      "loss": 6.0908,
      "mean_token_accuracy": 0.14434681534767152,
      "num_tokens": 29353273.0,
      "step": 13565
    },
    {
      "entropy": 6.380755090713501,
      "epoch": 1.4521376210605168,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004797828412575348,
      "loss": 6.108,
      "mean_token_accuracy": 0.14612562209367752,
      "num_tokens": 29364108.0,
      "step": 13570
    },
    {
      "entropy": 6.241926956176758,
      "epoch": 1.452672695168281,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047976699868914594,
      "loss": 6.0749,
      "mean_token_accuracy": 0.15299162566661834,
      "num_tokens": 29375272.0,
      "step": 13575
    },
    {
      "entropy": 6.265605926513672,
      "epoch": 1.4532077692760448,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004797511502081388,
      "loss": 6.1068,
      "mean_token_accuracy": 0.14344103932380675,
      "num_tokens": 29385749.0,
      "step": 13580
    },
    {
      "entropy": 6.2807090282440186,
      "epoch": 1.4537428433838087,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000479735295814971,
      "loss": 6.0394,
      "mean_token_accuracy": 0.15001987665891647,
      "num_tokens": 29396430.0,
      "step": 13585
    },
    {
      "entropy": 6.38220911026001,
      "epoch": 1.4542779174915725,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047971943551010035,
      "loss": 6.0984,
      "mean_token_accuracy": 0.14052857384085654,
      "num_tokens": 29407020.0,
      "step": 13590
    },
    {
      "entropy": 6.36534743309021,
      "epoch": 1.4548129915993364,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047970356929398467,
      "loss": 6.1262,
      "mean_token_accuracy": 0.1448627457022667,
      "num_tokens": 29418428.0,
      "step": 13595
    },
    {
      "entropy": 6.3651588439941404,
      "epoch": 1.4553480657071005,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004796876971670822,
      "loss": 6.1987,
      "mean_token_accuracy": 0.14081755504012108,
      "num_tokens": 29430611.0,
      "step": 13600
    },
    {
      "entropy": 6.39989161491394,
      "epoch": 1.4558831398148644,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004796718191298512,
      "loss": 6.0968,
      "mean_token_accuracy": 0.1363537833094597,
      "num_tokens": 29441545.0,
      "step": 13605
    },
    {
      "entropy": 6.423513412475586,
      "epoch": 1.4564182139226283,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004796559351827503,
      "loss": 6.1954,
      "mean_token_accuracy": 0.14107317179441453,
      "num_tokens": 29452529.0,
      "step": 13610
    },
    {
      "entropy": 6.315101671218872,
      "epoch": 1.4569532880303921,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000479640045326238,
      "loss": 6.0862,
      "mean_token_accuracy": 0.14820460006594657,
      "num_tokens": 29463860.0,
      "step": 13615
    },
    {
      "entropy": 6.43165283203125,
      "epoch": 1.457488362138156,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004796241495607731,
      "loss": 6.1636,
      "mean_token_accuracy": 0.14036834761500358,
      "num_tokens": 29475004.0,
      "step": 13620
    },
    {
      "entropy": 6.3684995651245115,
      "epoch": 1.4580234362459201,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004796082478868146,
      "loss": 6.231,
      "mean_token_accuracy": 0.1333060532808304,
      "num_tokens": 29485921.0,
      "step": 13625
    },
    {
      "entropy": 6.372405052185059,
      "epoch": 1.458558510353684,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047959234030482185,
      "loss": 6.0849,
      "mean_token_accuracy": 0.15278844386339188,
      "num_tokens": 29495729.0,
      "step": 13630
    },
    {
      "entropy": 6.392063140869141,
      "epoch": 1.4590935844614479,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004795764268152538,
      "loss": 6.1496,
      "mean_token_accuracy": 0.14221642315387725,
      "num_tokens": 29506051.0,
      "step": 13635
    },
    {
      "entropy": 6.353977680206299,
      "epoch": 1.4596286585692118,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004795605074185704,
      "loss": 6.0926,
      "mean_token_accuracy": 0.14395012110471725,
      "num_tokens": 29516234.0,
      "step": 13640
    },
    {
      "entropy": 6.356445932388306,
      "epoch": 1.4601637326769756,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047954458211523095,
      "loss": 6.1046,
      "mean_token_accuracy": 0.14281121417880058,
      "num_tokens": 29527649.0,
      "step": 13645
    },
    {
      "entropy": 6.373775959014893,
      "epoch": 1.4606988067847397,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047952865090569545,
      "loss": 6.1557,
      "mean_token_accuracy": 0.13532271534204482,
      "num_tokens": 29539150.0,
      "step": 13650
    },
    {
      "entropy": 6.317780017852783,
      "epoch": 1.4612338808925036,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047951271379042393,
      "loss": 6.0348,
      "mean_token_accuracy": 0.14665995091199874,
      "num_tokens": 29550058.0,
      "step": 13655
    },
    {
      "entropy": 6.257121562957764,
      "epoch": 1.4617689550002675,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004794967707698765,
      "loss": 6.0529,
      "mean_token_accuracy": 0.155406591296196,
      "num_tokens": 29560377.0,
      "step": 13660
    },
    {
      "entropy": 6.265567874908447,
      "epoch": 1.4623040291080316,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004794808218445136,
      "loss": 5.984,
      "mean_token_accuracy": 0.15268293991684914,
      "num_tokens": 29571378.0,
      "step": 13665
    },
    {
      "entropy": 6.372422361373902,
      "epoch": 1.4628391032157955,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047946486701479576,
      "loss": 6.143,
      "mean_token_accuracy": 0.14150616824626921,
      "num_tokens": 29582187.0,
      "step": 13670
    },
    {
      "entropy": 6.294898748397827,
      "epoch": 1.4633741773235593,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004794489062811835,
      "loss": 6.1288,
      "mean_token_accuracy": 0.14087508842349053,
      "num_tokens": 29592982.0,
      "step": 13675
    },
    {
      "entropy": 6.373124647140503,
      "epoch": 1.4639092514313232,
      "grad_norm": 1.375,
      "learning_rate": 0.0004794329396441378,
      "loss": 6.0991,
      "mean_token_accuracy": 0.1427845224738121,
      "num_tokens": 29603639.0,
      "step": 13680
    },
    {
      "entropy": 6.356724977493286,
      "epoch": 1.464444325539087,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047941696710411975,
      "loss": 6.0325,
      "mean_token_accuracy": 0.1557264119386673,
      "num_tokens": 29615103.0,
      "step": 13685
    },
    {
      "entropy": 6.3549604415893555,
      "epoch": 1.4649793996468512,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004794009886615904,
      "loss": 6.1256,
      "mean_token_accuracy": 0.14616781771183013,
      "num_tokens": 29626184.0,
      "step": 13690
    },
    {
      "entropy": 6.331011199951172,
      "epoch": 1.465514473754615,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047938500431701135,
      "loss": 6.1053,
      "mean_token_accuracy": 0.15402669757604598,
      "num_tokens": 29637615.0,
      "step": 13695
    },
    {
      "entropy": 6.362054920196533,
      "epoch": 1.466049547862379,
      "grad_norm": 1.296875,
      "learning_rate": 0.000479369014070844,
      "loss": 6.1484,
      "mean_token_accuracy": 0.14322591871023177,
      "num_tokens": 29648648.0,
      "step": 13700
    },
    {
      "entropy": 6.368297863006592,
      "epoch": 1.4665846219701428,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047935301792355003,
      "loss": 6.093,
      "mean_token_accuracy": 0.14833443909883498,
      "num_tokens": 29660542.0,
      "step": 13705
    },
    {
      "entropy": 6.3694963455200195,
      "epoch": 1.4671196960779067,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004793370158755914,
      "loss": 6.1347,
      "mean_token_accuracy": 0.1433333456516266,
      "num_tokens": 29670342.0,
      "step": 13710
    },
    {
      "entropy": 6.272875595092773,
      "epoch": 1.4676547701856708,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047932100792743014,
      "loss": 6.0992,
      "mean_token_accuracy": 0.14908762723207475,
      "num_tokens": 29681191.0,
      "step": 13715
    },
    {
      "entropy": 6.319122552871704,
      "epoch": 1.4681898442934347,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047930499407952855,
      "loss": 6.0389,
      "mean_token_accuracy": 0.1511463515460491,
      "num_tokens": 29691299.0,
      "step": 13720
    },
    {
      "entropy": 6.319364452362061,
      "epoch": 1.4687249184011986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047928897433234893,
      "loss": 6.0951,
      "mean_token_accuracy": 0.14009829387068748,
      "num_tokens": 29702125.0,
      "step": 13725
    },
    {
      "entropy": 6.357442045211792,
      "epoch": 1.4692599925089624,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004792729486863539,
      "loss": 6.1105,
      "mean_token_accuracy": 0.14203980565071106,
      "num_tokens": 29712679.0,
      "step": 13730
    },
    {
      "entropy": 6.346800327301025,
      "epoch": 1.4697950666167263,
      "grad_norm": 1.25,
      "learning_rate": 0.0004792569171420061,
      "loss": 6.0483,
      "mean_token_accuracy": 0.15139901041984558,
      "num_tokens": 29723833.0,
      "step": 13735
    },
    {
      "entropy": 6.3696918964385985,
      "epoch": 1.4703301407244904,
      "grad_norm": 1.375,
      "learning_rate": 0.0004792408796997687,
      "loss": 6.1639,
      "mean_token_accuracy": 0.14215011522173882,
      "num_tokens": 29734646.0,
      "step": 13740
    },
    {
      "entropy": 6.336226177215576,
      "epoch": 1.4708652148322543,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004792248363601044,
      "loss": 6.1144,
      "mean_token_accuracy": 0.1452825590968132,
      "num_tokens": 29745129.0,
      "step": 13745
    },
    {
      "entropy": 6.3548754215240475,
      "epoch": 1.4714002889400182,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004792087871234767,
      "loss": 6.1177,
      "mean_token_accuracy": 0.14167182371020318,
      "num_tokens": 29756308.0,
      "step": 13750
    },
    {
      "entropy": 6.334949731826782,
      "epoch": 1.471935363047782,
      "grad_norm": 1.40625,
      "learning_rate": 0.000479192731990349,
      "loss": 6.1148,
      "mean_token_accuracy": 0.13842023834586142,
      "num_tokens": 29767835.0,
      "step": 13755
    },
    {
      "entropy": 6.3524439334869385,
      "epoch": 1.472470437155546,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004791766709611849,
      "loss": 6.148,
      "mean_token_accuracy": 0.14676832556724548,
      "num_tokens": 29778776.0,
      "step": 13760
    },
    {
      "entropy": 6.452246379852295,
      "epoch": 1.47300551126331,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000479160604036448,
      "loss": 6.173,
      "mean_token_accuracy": 0.13587491512298583,
      "num_tokens": 29788800.0,
      "step": 13765
    },
    {
      "entropy": 6.338260507583618,
      "epoch": 1.473540585371074,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004791445312166025,
      "loss": 6.0748,
      "mean_token_accuracy": 0.15130684450268744,
      "num_tokens": 29799375.0,
      "step": 13770
    },
    {
      "entropy": 6.349206066131591,
      "epoch": 1.4740756594788378,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004791284525021122,
      "loss": 6.1899,
      "mean_token_accuracy": 0.13416427001357079,
      "num_tokens": 29811267.0,
      "step": 13775
    },
    {
      "entropy": 6.323820781707764,
      "epoch": 1.4746107335866017,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004791123678934416,
      "loss": 6.1282,
      "mean_token_accuracy": 0.14519483745098113,
      "num_tokens": 29821616.0,
      "step": 13780
    },
    {
      "entropy": 6.318466472625732,
      "epoch": 1.4751458076943655,
      "grad_norm": 1.25,
      "learning_rate": 0.0004790962773910551,
      "loss": 6.1123,
      "mean_token_accuracy": 0.14671307504177095,
      "num_tokens": 29832233.0,
      "step": 13785
    },
    {
      "entropy": 6.353925466537476,
      "epoch": 1.4756808818021296,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047908018099541717,
      "loss": 6.0859,
      "mean_token_accuracy": 0.14370028227567672,
      "num_tokens": 29843465.0,
      "step": 13790
    },
    {
      "entropy": 6.300717782974243,
      "epoch": 1.4762159559098935,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004790640787069927,
      "loss": 5.9731,
      "mean_token_accuracy": 0.14985278025269508,
      "num_tokens": 29854123.0,
      "step": 13795
    },
    {
      "entropy": 6.384717321395874,
      "epoch": 1.4767510300176574,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004790479705262467,
      "loss": 6.192,
      "mean_token_accuracy": 0.13484957218170165,
      "num_tokens": 29864831.0,
      "step": 13800
    },
    {
      "entropy": 6.289125680923462,
      "epoch": 1.4772861041254215,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004790318564536441,
      "loss": 6.0595,
      "mean_token_accuracy": 0.13973388820886612,
      "num_tokens": 29876273.0,
      "step": 13805
    },
    {
      "entropy": 6.322299242019653,
      "epoch": 1.4778211782331854,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047901573648965047,
      "loss": 6.0688,
      "mean_token_accuracy": 0.14085831567645074,
      "num_tokens": 29886560.0,
      "step": 13810
    },
    {
      "entropy": 6.3450947284698485,
      "epoch": 1.4783562523409493,
      "grad_norm": 1.296875,
      "learning_rate": 0.000478999610634731,
      "loss": 6.0713,
      "mean_token_accuracy": 0.14663850516080856,
      "num_tokens": 29896916.0,
      "step": 13815
    },
    {
      "entropy": 6.3019537925720215,
      "epoch": 1.4788913264487131,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004789834788893515,
      "loss": 6.1261,
      "mean_token_accuracy": 0.13751569837331773,
      "num_tokens": 29907288.0,
      "step": 13820
    },
    {
      "entropy": 6.263100290298462,
      "epoch": 1.479426400556477,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004789673412539777,
      "loss": 6.0604,
      "mean_token_accuracy": 0.14319539591670036,
      "num_tokens": 29918194.0,
      "step": 13825
    },
    {
      "entropy": 6.341124629974365,
      "epoch": 1.479961474664241,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047895119772907566,
      "loss": 6.0924,
      "mean_token_accuracy": 0.1442139446735382,
      "num_tokens": 29928521.0,
      "step": 13830
    },
    {
      "entropy": 6.322500514984131,
      "epoch": 1.480496548772005,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004789350483151113,
      "loss": 6.0607,
      "mean_token_accuracy": 0.14264947026968003,
      "num_tokens": 29938885.0,
      "step": 13835
    },
    {
      "entropy": 6.321271753311157,
      "epoch": 1.4810316228797689,
      "grad_norm": 1.25,
      "learning_rate": 0.0004789188930125512,
      "loss": 6.1179,
      "mean_token_accuracy": 0.15309633761644365,
      "num_tokens": 29948921.0,
      "step": 13840
    },
    {
      "entropy": 6.28307843208313,
      "epoch": 1.4815666969875327,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004789027318218616,
      "loss": 6.0386,
      "mean_token_accuracy": 0.1553712233901024,
      "num_tokens": 29959633.0,
      "step": 13845
    },
    {
      "entropy": 6.298843574523926,
      "epoch": 1.4821017710952966,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004788865647435093,
      "loss": 6.0738,
      "mean_token_accuracy": 0.14177014231681823,
      "num_tokens": 29970639.0,
      "step": 13850
    },
    {
      "entropy": 6.354288625717163,
      "epoch": 1.4826368452030607,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004788703917779612,
      "loss": 6.1762,
      "mean_token_accuracy": 0.1442151226103306,
      "num_tokens": 29981752.0,
      "step": 13855
    },
    {
      "entropy": 6.353398418426513,
      "epoch": 1.4831719193108246,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004788542129256841,
      "loss": 6.0306,
      "mean_token_accuracy": 0.14744187891483307,
      "num_tokens": 29991037.0,
      "step": 13860
    },
    {
      "entropy": 6.329816436767578,
      "epoch": 1.4837069934185885,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047883802818714526,
      "loss": 6.1414,
      "mean_token_accuracy": 0.1330692760646343,
      "num_tokens": 30001980.0,
      "step": 13865
    },
    {
      "entropy": 6.3736635684967045,
      "epoch": 1.4842420675263523,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000478821837562812,
      "loss": 6.1245,
      "mean_token_accuracy": 0.14060747846961022,
      "num_tokens": 30013377.0,
      "step": 13870
    },
    {
      "entropy": 6.3984130859375,
      "epoch": 1.4847771416341162,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004788056410531518,
      "loss": 6.1723,
      "mean_token_accuracy": 0.1465814918279648,
      "num_tokens": 30023909.0,
      "step": 13875
    },
    {
      "entropy": 6.329588222503662,
      "epoch": 1.4853122157418803,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004787894386586324,
      "loss": 6.0423,
      "mean_token_accuracy": 0.14698239266872407,
      "num_tokens": 30034069.0,
      "step": 13880
    },
    {
      "entropy": 6.279343128204346,
      "epoch": 1.4858472898496442,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047877323037972153,
      "loss": 6.0659,
      "mean_token_accuracy": 0.14743561446666717,
      "num_tokens": 30043940.0,
      "step": 13885
    },
    {
      "entropy": 6.261647033691406,
      "epoch": 1.486382363957408,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004787570162168874,
      "loss": 6.0461,
      "mean_token_accuracy": 0.1519550561904907,
      "num_tokens": 30054985.0,
      "step": 13890
    },
    {
      "entropy": 6.3471925258636475,
      "epoch": 1.486917438065172,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047874079617059783,
      "loss": 6.0754,
      "mean_token_accuracy": 0.14557663276791571,
      "num_tokens": 30065675.0,
      "step": 13895
    },
    {
      "entropy": 6.378845930099487,
      "epoch": 1.4874525121729358,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004787245702413216,
      "loss": 6.0351,
      "mean_token_accuracy": 0.14570895582437515,
      "num_tokens": 30075305.0,
      "step": 13900
    },
    {
      "entropy": 6.340493154525757,
      "epoch": 1.4879875862807,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047870833842952697,
      "loss": 6.1223,
      "mean_token_accuracy": 0.14515430107712746,
      "num_tokens": 30087309.0,
      "step": 13905
    },
    {
      "entropy": 6.324948501586914,
      "epoch": 1.4885226603884638,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004786921007356827,
      "loss": 6.0548,
      "mean_token_accuracy": 0.14190238565206528,
      "num_tokens": 30097262.0,
      "step": 13910
    },
    {
      "entropy": 6.395900774002075,
      "epoch": 1.4890577344962277,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047867585716025767,
      "loss": 6.1504,
      "mean_token_accuracy": 0.14323515743017196,
      "num_tokens": 30107067.0,
      "step": 13915
    },
    {
      "entropy": 6.384770679473877,
      "epoch": 1.4895928086039918,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004786596077037209,
      "loss": 6.1581,
      "mean_token_accuracy": 0.1373176544904709,
      "num_tokens": 30117115.0,
      "step": 13920
    },
    {
      "entropy": 6.35680456161499,
      "epoch": 1.4901278827117554,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047864335236654144,
      "loss": 6.0776,
      "mean_token_accuracy": 0.14542236030101777,
      "num_tokens": 30127202.0,
      "step": 13925
    },
    {
      "entropy": 6.27435827255249,
      "epoch": 1.4906629568195195,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004786270911491889,
      "loss": 6.087,
      "mean_token_accuracy": 0.14455239698290825,
      "num_tokens": 30138608.0,
      "step": 13930
    },
    {
      "entropy": 6.399806022644043,
      "epoch": 1.4911980309272834,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004786108240521327,
      "loss": 6.115,
      "mean_token_accuracy": 0.14914220720529556,
      "num_tokens": 30148500.0,
      "step": 13935
    },
    {
      "entropy": 6.286115312576294,
      "epoch": 1.4917331050350473,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047859455107584256,
      "loss": 6.05,
      "mean_token_accuracy": 0.1448947884142399,
      "num_tokens": 30159062.0,
      "step": 13940
    },
    {
      "entropy": 6.277351808547974,
      "epoch": 1.4922681791428114,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047857827222078835,
      "loss": 6.0959,
      "mean_token_accuracy": 0.1453048586845398,
      "num_tokens": 30170468.0,
      "step": 13945
    },
    {
      "entropy": 6.261321020126343,
      "epoch": 1.4928032532505753,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004785619874874402,
      "loss": 5.989,
      "mean_token_accuracy": 0.1512428715825081,
      "num_tokens": 30180944.0,
      "step": 13950
    },
    {
      "entropy": 6.258210897445679,
      "epoch": 1.4933383273583392,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004785456968762682,
      "loss": 5.995,
      "mean_token_accuracy": 0.15131490603089331,
      "num_tokens": 30191614.0,
      "step": 13955
    },
    {
      "entropy": 6.310832071304321,
      "epoch": 1.493873401466103,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047852940038774276,
      "loss": 6.0258,
      "mean_token_accuracy": 0.1542426347732544,
      "num_tokens": 30202035.0,
      "step": 13960
    },
    {
      "entropy": 6.290114116668701,
      "epoch": 1.494408475573867,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047851309802233457,
      "loss": 6.1082,
      "mean_token_accuracy": 0.1489107832312584,
      "num_tokens": 30213205.0,
      "step": 13965
    },
    {
      "entropy": 6.364186239242554,
      "epoch": 1.494943549681631,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004784967897805143,
      "loss": 6.12,
      "mean_token_accuracy": 0.14308732599020005,
      "num_tokens": 30223971.0,
      "step": 13970
    },
    {
      "entropy": 6.360693740844726,
      "epoch": 1.495478623789395,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004784804756627528,
      "loss": 6.1217,
      "mean_token_accuracy": 0.14283821731805801,
      "num_tokens": 30235346.0,
      "step": 13975
    },
    {
      "entropy": 6.303307247161865,
      "epoch": 1.4960136978971588,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047846415566952106,
      "loss": 6.0604,
      "mean_token_accuracy": 0.14785521626472473,
      "num_tokens": 30246033.0,
      "step": 13980
    },
    {
      "entropy": 6.3066637992858885,
      "epoch": 1.4965487720049226,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047844782980129057,
      "loss": 6.1255,
      "mean_token_accuracy": 0.14006659686565398,
      "num_tokens": 30258208.0,
      "step": 13985
    },
    {
      "entropy": 6.350608777999878,
      "epoch": 1.4970838461126865,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047843149805853254,
      "loss": 6.1298,
      "mean_token_accuracy": 0.14496227726340294,
      "num_tokens": 30268824.0,
      "step": 13990
    },
    {
      "entropy": 6.432400751113891,
      "epoch": 1.4976189202204506,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004784151604417185,
      "loss": 6.1663,
      "mean_token_accuracy": 0.13490559607744218,
      "num_tokens": 30280078.0,
      "step": 13995
    },
    {
      "entropy": 6.328733444213867,
      "epoch": 1.4981539943282145,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047839881695132034,
      "loss": 6.0485,
      "mean_token_accuracy": 0.15326653569936752,
      "num_tokens": 30290939.0,
      "step": 14000
    },
    {
      "entropy": 6.259870815277099,
      "epoch": 1.4986890684359784,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047838246758780996,
      "loss": 6.1039,
      "mean_token_accuracy": 0.14923280030488967,
      "num_tokens": 30301198.0,
      "step": 14005
    },
    {
      "entropy": 6.274726057052613,
      "epoch": 1.4992241425437423,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004783661123516593,
      "loss": 6.0521,
      "mean_token_accuracy": 0.14646771252155305,
      "num_tokens": 30312021.0,
      "step": 14010
    },
    {
      "entropy": 6.3456458568573,
      "epoch": 1.4997592166515061,
      "grad_norm": 1.25,
      "learning_rate": 0.0004783497512433408,
      "loss": 6.077,
      "mean_token_accuracy": 0.1471131831407547,
      "num_tokens": 30323335.0,
      "step": 14015
    },
    {
      "entropy": 6.328543424606323,
      "epoch": 1.5002942907592702,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004783333842633268,
      "loss": 6.1176,
      "mean_token_accuracy": 0.14810129404067993,
      "num_tokens": 30334389.0,
      "step": 14020
    },
    {
      "entropy": 6.266499853134155,
      "epoch": 1.500829364867034,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047831701141208994,
      "loss": 6.0619,
      "mean_token_accuracy": 0.13776419684290886,
      "num_tokens": 30345407.0,
      "step": 14025
    },
    {
      "entropy": 6.350656175613404,
      "epoch": 1.501364438974798,
      "grad_norm": 1.65625,
      "learning_rate": 0.00047830063269010286,
      "loss": 6.0844,
      "mean_token_accuracy": 0.14303870275616645,
      "num_tokens": 30356718.0,
      "step": 14030
    },
    {
      "entropy": 6.290414905548095,
      "epoch": 1.501899513082562,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047828424809783865,
      "loss": 6.0413,
      "mean_token_accuracy": 0.1465670645236969,
      "num_tokens": 30367617.0,
      "step": 14035
    },
    {
      "entropy": 6.294115114212036,
      "epoch": 1.5024345871903257,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004782678576357703,
      "loss": 6.0832,
      "mean_token_accuracy": 0.1455468326807022,
      "num_tokens": 30377923.0,
      "step": 14040
    },
    {
      "entropy": 6.336675071716309,
      "epoch": 1.5029696612980898,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004782514613043711,
      "loss": 6.0907,
      "mean_token_accuracy": 0.14969113618135452,
      "num_tokens": 30388719.0,
      "step": 14045
    },
    {
      "entropy": 6.308081245422363,
      "epoch": 1.5035047354058537,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047823505910411465,
      "loss": 6.0244,
      "mean_token_accuracy": 0.1478879362344742,
      "num_tokens": 30400191.0,
      "step": 14050
    },
    {
      "entropy": 6.309054565429688,
      "epoch": 1.5040398095136176,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047821865103547437,
      "loss": 6.1019,
      "mean_token_accuracy": 0.14360111355781555,
      "num_tokens": 30410267.0,
      "step": 14055
    },
    {
      "entropy": 6.307495546340943,
      "epoch": 1.5045748836213817,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000478202237098924,
      "loss": 6.1194,
      "mean_token_accuracy": 0.14021419882774352,
      "num_tokens": 30420852.0,
      "step": 14060
    },
    {
      "entropy": 6.382779598236084,
      "epoch": 1.5051099577291454,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004781858172949376,
      "loss": 6.1648,
      "mean_token_accuracy": 0.14720651432871817,
      "num_tokens": 30432214.0,
      "step": 14065
    },
    {
      "entropy": 6.4346067905426025,
      "epoch": 1.5056450318369095,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004781693916239894,
      "loss": 6.1767,
      "mean_token_accuracy": 0.13551431894302368,
      "num_tokens": 30443919.0,
      "step": 14070
    },
    {
      "entropy": 6.383563041687012,
      "epoch": 1.5061801059446733,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047815296008655347,
      "loss": 6.2321,
      "mean_token_accuracy": 0.13739842250943185,
      "num_tokens": 30453930.0,
      "step": 14075
    },
    {
      "entropy": 6.338282680511474,
      "epoch": 1.5067151800524372,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047813652268310436,
      "loss": 6.1066,
      "mean_token_accuracy": 0.14010724574327468,
      "num_tokens": 30464612.0,
      "step": 14080
    },
    {
      "entropy": 6.376251363754273,
      "epoch": 1.5072502541602013,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047812007941411673,
      "loss": 6.061,
      "mean_token_accuracy": 0.1443423479795456,
      "num_tokens": 30474819.0,
      "step": 14085
    },
    {
      "entropy": 6.283281564712524,
      "epoch": 1.507785328267965,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047810363028006535,
      "loss": 6.0437,
      "mean_token_accuracy": 0.15429361760616303,
      "num_tokens": 30485481.0,
      "step": 14090
    },
    {
      "entropy": 6.279097843170166,
      "epoch": 1.508320402375729,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004780871752814252,
      "loss": 6.0936,
      "mean_token_accuracy": 0.1503777399659157,
      "num_tokens": 30496321.0,
      "step": 14095
    },
    {
      "entropy": 6.293540668487549,
      "epoch": 1.508855476483493,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004780707144186714,
      "loss": 6.1799,
      "mean_token_accuracy": 0.14072953760623932,
      "num_tokens": 30507578.0,
      "step": 14100
    },
    {
      "entropy": 6.417983818054199,
      "epoch": 1.5093905505912568,
      "grad_norm": 1.25,
      "learning_rate": 0.00047805424769227923,
      "loss": 6.1159,
      "mean_token_accuracy": 0.1410513386130333,
      "num_tokens": 30518328.0,
      "step": 14105
    },
    {
      "entropy": 6.321520709991455,
      "epoch": 1.509925624699021,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047803777510272425,
      "loss": 6.0684,
      "mean_token_accuracy": 0.14671278595924378,
      "num_tokens": 30528555.0,
      "step": 14110
    },
    {
      "entropy": 6.24947419166565,
      "epoch": 1.5104606988067846,
      "grad_norm": 1.265625,
      "learning_rate": 0.000478021296650482,
      "loss": 6.0018,
      "mean_token_accuracy": 0.15414322316646575,
      "num_tokens": 30540217.0,
      "step": 14115
    },
    {
      "entropy": 6.385993337631225,
      "epoch": 1.5109957729145487,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004780048123360283,
      "loss": 6.1381,
      "mean_token_accuracy": 0.14674227312207222,
      "num_tokens": 30550937.0,
      "step": 14120
    },
    {
      "entropy": 6.450168037414551,
      "epoch": 1.5115308470223126,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047798832215983916,
      "loss": 6.1316,
      "mean_token_accuracy": 0.14239492416381835,
      "num_tokens": 30562248.0,
      "step": 14125
    },
    {
      "entropy": 6.4028575897216795,
      "epoch": 1.5120659211300764,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004779718261223908,
      "loss": 6.1104,
      "mean_token_accuracy": 0.1503224715590477,
      "num_tokens": 30573362.0,
      "step": 14130
    },
    {
      "entropy": 6.300021362304688,
      "epoch": 1.5126009952378405,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047795532422415947,
      "loss": 6.1102,
      "mean_token_accuracy": 0.14241843596100806,
      "num_tokens": 30584052.0,
      "step": 14135
    },
    {
      "entropy": 6.307920122146607,
      "epoch": 1.5131360693456044,
      "grad_norm": 1.375,
      "learning_rate": 0.0004779388164656217,
      "loss": 6.1284,
      "mean_token_accuracy": 0.14281389117240906,
      "num_tokens": 30594630.0,
      "step": 14140
    },
    {
      "entropy": 6.355758762359619,
      "epoch": 1.5136711434533683,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047792230284725415,
      "loss": 6.0988,
      "mean_token_accuracy": 0.1405882380902767,
      "num_tokens": 30605625.0,
      "step": 14145
    },
    {
      "entropy": 6.319501781463623,
      "epoch": 1.5142062175611322,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047790578336953356,
      "loss": 6.0861,
      "mean_token_accuracy": 0.1479765497148037,
      "num_tokens": 30616520.0,
      "step": 14150
    },
    {
      "entropy": 6.355174350738525,
      "epoch": 1.514741291668896,
      "grad_norm": 1.390625,
      "learning_rate": 0.000477889258032937,
      "loss": 6.0649,
      "mean_token_accuracy": 0.14791652858257293,
      "num_tokens": 30627372.0,
      "step": 14155
    },
    {
      "entropy": 6.357487583160401,
      "epoch": 1.5152763657766601,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047787272683794157,
      "loss": 6.0454,
      "mean_token_accuracy": 0.15269929468631743,
      "num_tokens": 30636974.0,
      "step": 14160
    },
    {
      "entropy": 6.346766710281372,
      "epoch": 1.515811439884424,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004778561897850247,
      "loss": 6.1122,
      "mean_token_accuracy": 0.14738047048449515,
      "num_tokens": 30648202.0,
      "step": 14165
    },
    {
      "entropy": 6.323415851593017,
      "epoch": 1.516346513992188,
      "grad_norm": 1.9375,
      "learning_rate": 0.00047783964687466384,
      "loss": 6.1136,
      "mean_token_accuracy": 0.14402420967817306,
      "num_tokens": 30657981.0,
      "step": 14170
    },
    {
      "entropy": 6.3754730224609375,
      "epoch": 1.516881588099952,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047782309810733676,
      "loss": 6.1248,
      "mean_token_accuracy": 0.13893367499113082,
      "num_tokens": 30668496.0,
      "step": 14175
    },
    {
      "entropy": 6.323855972290039,
      "epoch": 1.5174166622077156,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047780654348352115,
      "loss": 6.0069,
      "mean_token_accuracy": 0.14946904480457307,
      "num_tokens": 30678868.0,
      "step": 14180
    },
    {
      "entropy": 6.354461908340454,
      "epoch": 1.5179517363154797,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047778998300369516,
      "loss": 6.1447,
      "mean_token_accuracy": 0.14793603569269181,
      "num_tokens": 30690066.0,
      "step": 14185
    },
    {
      "entropy": 6.29534764289856,
      "epoch": 1.5184868104232436,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004777734166683368,
      "loss": 6.022,
      "mean_token_accuracy": 0.14419885352253914,
      "num_tokens": 30700624.0,
      "step": 14190
    },
    {
      "entropy": 6.275360250473023,
      "epoch": 1.5190218845310075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047775684447792457,
      "loss": 6.0367,
      "mean_token_accuracy": 0.15461383014917374,
      "num_tokens": 30711183.0,
      "step": 14195
    },
    {
      "entropy": 6.312857246398925,
      "epoch": 1.5195569586387716,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000477740266432937,
      "loss": 6.0689,
      "mean_token_accuracy": 0.15200629979372024,
      "num_tokens": 30721446.0,
      "step": 14200
    },
    {
      "entropy": 6.363908290863037,
      "epoch": 1.5200920327465353,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047772368253385267,
      "loss": 6.1312,
      "mean_token_accuracy": 0.14002955257892608,
      "num_tokens": 30731901.0,
      "step": 14205
    },
    {
      "entropy": 6.400300216674805,
      "epoch": 1.5206271068542994,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004777070927811505,
      "loss": 6.1041,
      "mean_token_accuracy": 0.13925700038671493,
      "num_tokens": 30741701.0,
      "step": 14210
    },
    {
      "entropy": 6.3462238788604735,
      "epoch": 1.5211621809620632,
      "grad_norm": 1.734375,
      "learning_rate": 0.00047769049717530956,
      "loss": 6.0931,
      "mean_token_accuracy": 0.1435927391052246,
      "num_tokens": 30751594.0,
      "step": 14215
    },
    {
      "entropy": 6.317065095901489,
      "epoch": 1.5216972550698271,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047767389571680896,
      "loss": 6.1764,
      "mean_token_accuracy": 0.1335340164601803,
      "num_tokens": 30764418.0,
      "step": 14220
    },
    {
      "entropy": 6.372600078582764,
      "epoch": 1.5222323291775912,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004776572884061281,
      "loss": 6.0844,
      "mean_token_accuracy": 0.1477690428495407,
      "num_tokens": 30773976.0,
      "step": 14225
    },
    {
      "entropy": 6.3617844581604,
      "epoch": 1.5227674032853549,
      "grad_norm": 1.25,
      "learning_rate": 0.0004776406752437465,
      "loss": 6.1057,
      "mean_token_accuracy": 0.1507497951388359,
      "num_tokens": 30784210.0,
      "step": 14230
    },
    {
      "entropy": 6.347768354415893,
      "epoch": 1.523302477393119,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004776240562301438,
      "loss": 6.1658,
      "mean_token_accuracy": 0.13491279855370522,
      "num_tokens": 30794322.0,
      "step": 14235
    },
    {
      "entropy": 6.378882312774659,
      "epoch": 1.5238375515008828,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004776074313658001,
      "loss": 6.1093,
      "mean_token_accuracy": 0.14816105365753174,
      "num_tokens": 30805264.0,
      "step": 14240
    },
    {
      "entropy": 6.294718503952026,
      "epoch": 1.5243726256086467,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004775908006511952,
      "loss": 6.0867,
      "mean_token_accuracy": 0.1425347924232483,
      "num_tokens": 30815909.0,
      "step": 14245
    },
    {
      "entropy": 6.361842012405395,
      "epoch": 1.5249076997164108,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004775741640868095,
      "loss": 6.0825,
      "mean_token_accuracy": 0.15103104710578918,
      "num_tokens": 30826867.0,
      "step": 14250
    },
    {
      "entropy": 6.360204601287842,
      "epoch": 1.5254427738241747,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047755752167312314,
      "loss": 6.0569,
      "mean_token_accuracy": 0.13976811692118646,
      "num_tokens": 30837696.0,
      "step": 14255
    },
    {
      "entropy": 6.296175765991211,
      "epoch": 1.5259778479319386,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004775408734106169,
      "loss": 6.0821,
      "mean_token_accuracy": 0.14574447721242906,
      "num_tokens": 30848155.0,
      "step": 14260
    },
    {
      "entropy": 6.24638524055481,
      "epoch": 1.5265129220397025,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047752421929977143,
      "loss": 5.9582,
      "mean_token_accuracy": 0.15410574972629548,
      "num_tokens": 30857804.0,
      "step": 14265
    },
    {
      "entropy": 6.3576806545257565,
      "epoch": 1.5270479961474663,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004775075593410675,
      "loss": 6.1069,
      "mean_token_accuracy": 0.1490521937608719,
      "num_tokens": 30867290.0,
      "step": 14270
    },
    {
      "entropy": 6.347940444946289,
      "epoch": 1.5275830702552304,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047749089353498625,
      "loss": 6.1804,
      "mean_token_accuracy": 0.1459521546959877,
      "num_tokens": 30877634.0,
      "step": 14275
    },
    {
      "entropy": 6.407077360153198,
      "epoch": 1.5281181443629943,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000477474221882009,
      "loss": 6.1483,
      "mean_token_accuracy": 0.1397402800619602,
      "num_tokens": 30889176.0,
      "step": 14280
    },
    {
      "entropy": 6.382671928405761,
      "epoch": 1.5286532184707582,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004774575443826169,
      "loss": 6.0427,
      "mean_token_accuracy": 0.14514234960079192,
      "num_tokens": 30899516.0,
      "step": 14285
    },
    {
      "entropy": 6.343927574157715,
      "epoch": 1.5291882925785223,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004774408610372918,
      "loss": 6.1412,
      "mean_token_accuracy": 0.1445561297237873,
      "num_tokens": 30910155.0,
      "step": 14290
    },
    {
      "entropy": 6.3796735286712645,
      "epoch": 1.529723366686286,
      "grad_norm": 1.25,
      "learning_rate": 0.0004774241718465152,
      "loss": 6.166,
      "mean_token_accuracy": 0.13613315373659135,
      "num_tokens": 30921486.0,
      "step": 14295
    },
    {
      "entropy": 6.4204775333404545,
      "epoch": 1.53025844079405,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004774074768107691,
      "loss": 6.0892,
      "mean_token_accuracy": 0.1428506478667259,
      "num_tokens": 30932361.0,
      "step": 14300
    },
    {
      "entropy": 6.343177556991577,
      "epoch": 1.530793514901814,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047739077593053554,
      "loss": 6.0715,
      "mean_token_accuracy": 0.14367761537432672,
      "num_tokens": 30943033.0,
      "step": 14305
    },
    {
      "entropy": 6.285774374008179,
      "epoch": 1.5313285890095778,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047737406920629675,
      "loss": 6.0885,
      "mean_token_accuracy": 0.14528725072741508,
      "num_tokens": 30953094.0,
      "step": 14310
    },
    {
      "entropy": 6.285615396499634,
      "epoch": 1.531863663117342,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047735735663853514,
      "loss": 6.1027,
      "mean_token_accuracy": 0.14244612902402878,
      "num_tokens": 30964604.0,
      "step": 14315
    },
    {
      "entropy": 6.342643594741821,
      "epoch": 1.5323987372251056,
      "grad_norm": 1.25,
      "learning_rate": 0.0004773406382277334,
      "loss": 6.17,
      "mean_token_accuracy": 0.14099033400416375,
      "num_tokens": 30975681.0,
      "step": 14320
    },
    {
      "entropy": 6.447955226898193,
      "epoch": 1.5329338113328697,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004773239139743741,
      "loss": 6.1338,
      "mean_token_accuracy": 0.1414648860692978,
      "num_tokens": 30986319.0,
      "step": 14325
    },
    {
      "entropy": 6.383595275878906,
      "epoch": 1.5334688854406335,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004773071838789402,
      "loss": 6.0954,
      "mean_token_accuracy": 0.1445777490735054,
      "num_tokens": 30997077.0,
      "step": 14330
    },
    {
      "entropy": 6.272831439971924,
      "epoch": 1.5340039595483974,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004772904479419148,
      "loss": 6.0085,
      "mean_token_accuracy": 0.1501988649368286,
      "num_tokens": 31007485.0,
      "step": 14335
    },
    {
      "entropy": 6.307893705368042,
      "epoch": 1.5345390336561615,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004772737061637811,
      "loss": 6.1034,
      "mean_token_accuracy": 0.14843927025794984,
      "num_tokens": 31017827.0,
      "step": 14340
    },
    {
      "entropy": 6.353067064285279,
      "epoch": 1.5350741077639252,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047725695854502267,
      "loss": 6.1267,
      "mean_token_accuracy": 0.14730988144874574,
      "num_tokens": 31028167.0,
      "step": 14345
    },
    {
      "entropy": 6.33044114112854,
      "epoch": 1.5356091818716893,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047724020508612287,
      "loss": 6.0626,
      "mean_token_accuracy": 0.15088969916105271,
      "num_tokens": 31038941.0,
      "step": 14350
    },
    {
      "entropy": 6.293562507629394,
      "epoch": 1.5361442559794531,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047722344578756564,
      "loss": 6.0648,
      "mean_token_accuracy": 0.1425795339047909,
      "num_tokens": 31049949.0,
      "step": 14355
    },
    {
      "entropy": 6.330544471740723,
      "epoch": 1.536679330087217,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047720668064983476,
      "loss": 6.092,
      "mean_token_accuracy": 0.14448975026607513,
      "num_tokens": 31059299.0,
      "step": 14360
    },
    {
      "entropy": 6.317178344726562,
      "epoch": 1.5372144041949811,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004771899096734145,
      "loss": 6.1836,
      "mean_token_accuracy": 0.13760803639888763,
      "num_tokens": 31070508.0,
      "step": 14365
    },
    {
      "entropy": 6.289751577377319,
      "epoch": 1.5377494783027448,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047717313285878886,
      "loss": 6.0592,
      "mean_token_accuracy": 0.14989213198423385,
      "num_tokens": 31082354.0,
      "step": 14370
    },
    {
      "entropy": 6.506394147872925,
      "epoch": 1.5382845524105089,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047715635020644247,
      "loss": 6.2143,
      "mean_token_accuracy": 0.13742856830358505,
      "num_tokens": 31093496.0,
      "step": 14375
    },
    {
      "entropy": 6.366935777664184,
      "epoch": 1.5388196265182728,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047713956171685994,
      "loss": 6.0515,
      "mean_token_accuracy": 0.14856111705303193,
      "num_tokens": 31104289.0,
      "step": 14380
    },
    {
      "entropy": 6.201303243637085,
      "epoch": 1.5393547006260366,
      "grad_norm": 1.234375,
      "learning_rate": 0.000477122767390526,
      "loss": 6.011,
      "mean_token_accuracy": 0.154781836271286,
      "num_tokens": 31114998.0,
      "step": 14385
    },
    {
      "entropy": 6.270689964294434,
      "epoch": 1.5398897747338007,
      "grad_norm": 1.25,
      "learning_rate": 0.00047710596722792543,
      "loss": 6.0724,
      "mean_token_accuracy": 0.1416308768093586,
      "num_tokens": 31127393.0,
      "step": 14390
    },
    {
      "entropy": 6.2720390319824215,
      "epoch": 1.5404248488415646,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004770891612295435,
      "loss": 5.9202,
      "mean_token_accuracy": 0.16328924894332886,
      "num_tokens": 31137484.0,
      "step": 14395
    },
    {
      "entropy": 6.345932245254517,
      "epoch": 1.5409599229493285,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004770723493958654,
      "loss": 6.1551,
      "mean_token_accuracy": 0.13954937532544137,
      "num_tokens": 31147863.0,
      "step": 14400
    },
    {
      "entropy": 6.290973234176636,
      "epoch": 1.5414949970570924,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004770555317273767,
      "loss": 6.0173,
      "mean_token_accuracy": 0.14998056441545488,
      "num_tokens": 31159018.0,
      "step": 14405
    },
    {
      "entropy": 6.353177499771118,
      "epoch": 1.5420300711648562,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047703870822456287,
      "loss": 6.042,
      "mean_token_accuracy": 0.14606124311685562,
      "num_tokens": 31170673.0,
      "step": 14410
    },
    {
      "entropy": 6.174723958969116,
      "epoch": 1.5425651452726203,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004770218788879097,
      "loss": 5.9205,
      "mean_token_accuracy": 0.16160815060138703,
      "num_tokens": 31181165.0,
      "step": 14415
    },
    {
      "entropy": 6.214414310455322,
      "epoch": 1.5431002193803842,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004770050437179033,
      "loss": 6.0734,
      "mean_token_accuracy": 0.14637484177947044,
      "num_tokens": 31192976.0,
      "step": 14420
    },
    {
      "entropy": 6.3880236625671385,
      "epoch": 1.543635293488148,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004769882027150295,
      "loss": 6.1451,
      "mean_token_accuracy": 0.1476966083049774,
      "num_tokens": 31203979.0,
      "step": 14425
    },
    {
      "entropy": 6.401928949356079,
      "epoch": 1.5441703675959122,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047697135587977473,
      "loss": 6.0213,
      "mean_token_accuracy": 0.14770625680685043,
      "num_tokens": 31214972.0,
      "step": 14430
    },
    {
      "entropy": 6.366449928283691,
      "epoch": 1.5447054417036759,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047695450321262546,
      "loss": 6.1438,
      "mean_token_accuracy": 0.14479298889636993,
      "num_tokens": 31224726.0,
      "step": 14435
    },
    {
      "entropy": 6.327470445632935,
      "epoch": 1.54524051581144,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004769376447140683,
      "loss": 6.0202,
      "mean_token_accuracy": 0.14906784296035766,
      "num_tokens": 31235761.0,
      "step": 14440
    },
    {
      "entropy": 6.356503105163574,
      "epoch": 1.5457755899192038,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047692078038459015,
      "loss": 6.1118,
      "mean_token_accuracy": 0.14897409528493882,
      "num_tokens": 31246221.0,
      "step": 14445
    },
    {
      "entropy": 6.31176347732544,
      "epoch": 1.5463106640269677,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004769039102246777,
      "loss": 6.0788,
      "mean_token_accuracy": 0.14313893318176268,
      "num_tokens": 31257364.0,
      "step": 14450
    },
    {
      "entropy": 6.333934259414673,
      "epoch": 1.5468457381347318,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047688703423481827,
      "loss": 6.084,
      "mean_token_accuracy": 0.1431208334863186,
      "num_tokens": 31268343.0,
      "step": 14455
    },
    {
      "entropy": 6.33518762588501,
      "epoch": 1.5473808122424955,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047687015241549896,
      "loss": 6.1307,
      "mean_token_accuracy": 0.14275304973125458,
      "num_tokens": 31278335.0,
      "step": 14460
    },
    {
      "entropy": 6.316238355636597,
      "epoch": 1.5479158863502596,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004768532647672075,
      "loss": 6.0335,
      "mean_token_accuracy": 0.13962429985404015,
      "num_tokens": 31288537.0,
      "step": 14465
    },
    {
      "entropy": 6.251456546783447,
      "epoch": 1.5484509604580234,
      "grad_norm": 1.375,
      "learning_rate": 0.0004768363712904314,
      "loss": 6.0755,
      "mean_token_accuracy": 0.14751550629734994,
      "num_tokens": 31299353.0,
      "step": 14470
    },
    {
      "entropy": 6.4102270126342775,
      "epoch": 1.5489860345657873,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004768194719856584,
      "loss": 6.0936,
      "mean_token_accuracy": 0.1434257909655571,
      "num_tokens": 31309655.0,
      "step": 14475
    },
    {
      "entropy": 6.353145933151245,
      "epoch": 1.5495211086735514,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004768025668533765,
      "loss": 6.1347,
      "mean_token_accuracy": 0.15157814472913742,
      "num_tokens": 31321008.0,
      "step": 14480
    },
    {
      "entropy": 6.231277418136597,
      "epoch": 1.550056182781315,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047678565589407383,
      "loss": 6.0048,
      "mean_token_accuracy": 0.15123332738876344,
      "num_tokens": 31331978.0,
      "step": 14485
    },
    {
      "entropy": 6.301927518844605,
      "epoch": 1.5505912568890792,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004767687391082387,
      "loss": 6.0768,
      "mean_token_accuracy": 0.15358749777078629,
      "num_tokens": 31341703.0,
      "step": 14490
    },
    {
      "entropy": 6.34998254776001,
      "epoch": 1.551126330996843,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047675181649635965,
      "loss": 6.1145,
      "mean_token_accuracy": 0.1483453929424286,
      "num_tokens": 31352787.0,
      "step": 14495
    },
    {
      "entropy": 6.275788640975952,
      "epoch": 1.551661405104607,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047673488805892515,
      "loss": 6.0783,
      "mean_token_accuracy": 0.14663477540016173,
      "num_tokens": 31363702.0,
      "step": 14500
    },
    {
      "entropy": 6.3356890201568605,
      "epoch": 1.552196479212371,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047671795379642415,
      "loss": 6.0429,
      "mean_token_accuracy": 0.13820092231035233,
      "num_tokens": 31374719.0,
      "step": 14505
    },
    {
      "entropy": 6.405061817169189,
      "epoch": 1.5527315533201347,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004767010137093456,
      "loss": 6.1904,
      "mean_token_accuracy": 0.13668452203273773,
      "num_tokens": 31386289.0,
      "step": 14510
    },
    {
      "entropy": 6.28100471496582,
      "epoch": 1.5532666274278988,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047668406779817864,
      "loss": 6.0089,
      "mean_token_accuracy": 0.14945143610239028,
      "num_tokens": 31397103.0,
      "step": 14515
    },
    {
      "entropy": 6.308429288864136,
      "epoch": 1.5538017015356627,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004766671160634125,
      "loss": 6.1142,
      "mean_token_accuracy": 0.14661767333745956,
      "num_tokens": 31408715.0,
      "step": 14520
    },
    {
      "entropy": 6.406673526763916,
      "epoch": 1.5543367756434265,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004766501585055367,
      "loss": 6.1348,
      "mean_token_accuracy": 0.1368098095059395,
      "num_tokens": 31420365.0,
      "step": 14525
    },
    {
      "entropy": 6.3008934497833256,
      "epoch": 1.5548718497511906,
      "grad_norm": 1.46875,
      "learning_rate": 0.000476633195125041,
      "loss": 6.052,
      "mean_token_accuracy": 0.1440028354525566,
      "num_tokens": 31431635.0,
      "step": 14530
    },
    {
      "entropy": 6.336616849899292,
      "epoch": 1.5554069238589545,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047661622592241507,
      "loss": 6.0874,
      "mean_token_accuracy": 0.146408099681139,
      "num_tokens": 31442513.0,
      "step": 14535
    },
    {
      "entropy": 6.2940465927124025,
      "epoch": 1.5559419979667184,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004765992508981489,
      "loss": 6.0953,
      "mean_token_accuracy": 0.1374914191663265,
      "num_tokens": 31453726.0,
      "step": 14540
    },
    {
      "entropy": 6.314603805541992,
      "epoch": 1.5564770720744823,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004765822700527328,
      "loss": 6.0247,
      "mean_token_accuracy": 0.1485890790820122,
      "num_tokens": 31463980.0,
      "step": 14545
    },
    {
      "entropy": 6.3265117645263675,
      "epoch": 1.5570121461822461,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004765652833866569,
      "loss": 6.1787,
      "mean_token_accuracy": 0.13598474711179734,
      "num_tokens": 31474615.0,
      "step": 14550
    },
    {
      "entropy": 6.320029783248901,
      "epoch": 1.5575472202900102,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004765482909004118,
      "loss": 6.0245,
      "mean_token_accuracy": 0.15409590303897858,
      "num_tokens": 31485283.0,
      "step": 14555
    },
    {
      "entropy": 6.344710779190064,
      "epoch": 1.5580822943977741,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004765312925944881,
      "loss": 6.0763,
      "mean_token_accuracy": 0.1405719332396984,
      "num_tokens": 31496997.0,
      "step": 14560
    },
    {
      "entropy": 6.380016183853149,
      "epoch": 1.558617368505538,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047651428846937673,
      "loss": 6.0879,
      "mean_token_accuracy": 0.15161072760820388,
      "num_tokens": 31507585.0,
      "step": 14565
    },
    {
      "entropy": 6.31249771118164,
      "epoch": 1.559152442613302,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047649727852556854,
      "loss": 6.0127,
      "mean_token_accuracy": 0.1477291017770767,
      "num_tokens": 31517078.0,
      "step": 14570
    },
    {
      "entropy": 6.2198230743408205,
      "epoch": 1.5596875167210658,
      "grad_norm": 1.25,
      "learning_rate": 0.0004764802627635547,
      "loss": 6.0222,
      "mean_token_accuracy": 0.15557388663291932,
      "num_tokens": 31528576.0,
      "step": 14575
    },
    {
      "entropy": 6.414737129211426,
      "epoch": 1.5602225908288299,
      "grad_norm": 1.25,
      "learning_rate": 0.0004764632411838266,
      "loss": 6.134,
      "mean_token_accuracy": 0.14385116547346116,
      "num_tokens": 31540095.0,
      "step": 14580
    },
    {
      "entropy": 6.341256761550904,
      "epoch": 1.5607576649365937,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047644621378687573,
      "loss": 6.0225,
      "mean_token_accuracy": 0.14182841926813125,
      "num_tokens": 31550980.0,
      "step": 14585
    },
    {
      "entropy": 6.114298295974732,
      "epoch": 1.5612927390443576,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047642918057319377,
      "loss": 5.8368,
      "mean_token_accuracy": 0.16773824095726014,
      "num_tokens": 31561842.0,
      "step": 14590
    },
    {
      "entropy": 6.3106592178344725,
      "epoch": 1.5618278131521217,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004764121415432725,
      "loss": 6.1798,
      "mean_token_accuracy": 0.13977380841970444,
      "num_tokens": 31572798.0,
      "step": 14595
    },
    {
      "entropy": 6.326402997970581,
      "epoch": 1.5623628872598854,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004763950966976039,
      "loss": 6.1585,
      "mean_token_accuracy": 0.14347514510154724,
      "num_tokens": 31584774.0,
      "step": 14600
    },
    {
      "entropy": 6.324334812164307,
      "epoch": 1.5628979613676495,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004763780460366803,
      "loss": 6.0579,
      "mean_token_accuracy": 0.1415022112429142,
      "num_tokens": 31596197.0,
      "step": 14605
    },
    {
      "entropy": 6.3284495830535885,
      "epoch": 1.5634330354754133,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004763609895609939,
      "loss": 6.0887,
      "mean_token_accuracy": 0.1433653250336647,
      "num_tokens": 31607455.0,
      "step": 14610
    },
    {
      "entropy": 6.3163309574127195,
      "epoch": 1.5639681095831772,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047634392727103714,
      "loss": 6.1327,
      "mean_token_accuracy": 0.14328678473830223,
      "num_tokens": 31618995.0,
      "step": 14615
    },
    {
      "entropy": 6.2557422637939455,
      "epoch": 1.5645031836909413,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004763268591673028,
      "loss": 5.9922,
      "mean_token_accuracy": 0.14590707421302795,
      "num_tokens": 31629984.0,
      "step": 14620
    },
    {
      "entropy": 6.348090553283692,
      "epoch": 1.565038257798705,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004763097852502837,
      "loss": 6.0167,
      "mean_token_accuracy": 0.14814187437295914,
      "num_tokens": 31641362.0,
      "step": 14625
    },
    {
      "entropy": 6.3285492897033695,
      "epoch": 1.565573331906469,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047629270552047284,
      "loss": 6.1956,
      "mean_token_accuracy": 0.13666235134005547,
      "num_tokens": 31652650.0,
      "step": 14630
    },
    {
      "entropy": 6.309994268417358,
      "epoch": 1.566108406014233,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047627561997836337,
      "loss": 6.1002,
      "mean_token_accuracy": 0.13629725947976112,
      "num_tokens": 31664201.0,
      "step": 14635
    },
    {
      "entropy": 6.352754688262939,
      "epoch": 1.5666434801219968,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004762585286244487,
      "loss": 6.0575,
      "mean_token_accuracy": 0.14476294443011284,
      "num_tokens": 31674480.0,
      "step": 14640
    },
    {
      "entropy": 6.313599920272827,
      "epoch": 1.567178554229761,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047624143145922227,
      "loss": 6.0836,
      "mean_token_accuracy": 0.14541919976472856,
      "num_tokens": 31685377.0,
      "step": 14645
    },
    {
      "entropy": 6.347984743118286,
      "epoch": 1.5677136283375246,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047622432848317775,
      "loss": 6.0669,
      "mean_token_accuracy": 0.1431708037853241,
      "num_tokens": 31695199.0,
      "step": 14650
    },
    {
      "entropy": 6.288880014419556,
      "epoch": 1.5682487024452887,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004762072196968091,
      "loss": 6.1052,
      "mean_token_accuracy": 0.14553255140781401,
      "num_tokens": 31705528.0,
      "step": 14655
    },
    {
      "entropy": 6.218687295913696,
      "epoch": 1.5687837765530526,
      "grad_norm": 1.25,
      "learning_rate": 0.0004761901051006102,
      "loss": 6.0187,
      "mean_token_accuracy": 0.14986263066530228,
      "num_tokens": 31718156.0,
      "step": 14660
    },
    {
      "entropy": 6.322576951980591,
      "epoch": 1.5693188506608164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047617298469507534,
      "loss": 6.0439,
      "mean_token_accuracy": 0.1496296629309654,
      "num_tokens": 31729783.0,
      "step": 14665
    },
    {
      "entropy": 6.373491001129151,
      "epoch": 1.5698539247685805,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004761558584806988,
      "loss": 6.0375,
      "mean_token_accuracy": 0.1477369710803032,
      "num_tokens": 31740796.0,
      "step": 14670
    },
    {
      "entropy": 6.416421747207641,
      "epoch": 1.5703889988763444,
      "grad_norm": 1.375,
      "learning_rate": 0.0004761387264579751,
      "loss": 6.1404,
      "mean_token_accuracy": 0.1396790139377117,
      "num_tokens": 31750559.0,
      "step": 14675
    },
    {
      "entropy": 6.3035050392150875,
      "epoch": 1.5709240729841083,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004761215886273989,
      "loss": 6.033,
      "mean_token_accuracy": 0.1437011532485485,
      "num_tokens": 31761559.0,
      "step": 14680
    },
    {
      "entropy": 6.290442419052124,
      "epoch": 1.5714591470918722,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047610444498946507,
      "loss": 6.0134,
      "mean_token_accuracy": 0.149689469486475,
      "num_tokens": 31771677.0,
      "step": 14685
    },
    {
      "entropy": 6.342159986495972,
      "epoch": 1.571994221199636,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004760872955446688,
      "loss": 6.1496,
      "mean_token_accuracy": 0.13930685371160506,
      "num_tokens": 31783028.0,
      "step": 14690
    },
    {
      "entropy": 6.352302122116089,
      "epoch": 1.5725292953074002,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047607014029350504,
      "loss": 6.112,
      "mean_token_accuracy": 0.14765870720148086,
      "num_tokens": 31793110.0,
      "step": 14695
    },
    {
      "entropy": 6.286095380783081,
      "epoch": 1.573064369415164,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047605297923646923,
      "loss": 6.0802,
      "mean_token_accuracy": 0.1427620194852352,
      "num_tokens": 31804193.0,
      "step": 14700
    },
    {
      "entropy": 6.327849960327148,
      "epoch": 1.573599443522928,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047603581237405684,
      "loss": 6.0832,
      "mean_token_accuracy": 0.14877094328403473,
      "num_tokens": 31815752.0,
      "step": 14705
    },
    {
      "entropy": 6.380163335800171,
      "epoch": 1.574134517630692,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047601863970676367,
      "loss": 6.1459,
      "mean_token_accuracy": 0.14161180555820466,
      "num_tokens": 31825936.0,
      "step": 14710
    },
    {
      "entropy": 6.341240930557251,
      "epoch": 1.5746695917384557,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047600146123508554,
      "loss": 6.0527,
      "mean_token_accuracy": 0.14389993026852607,
      "num_tokens": 31836601.0,
      "step": 14715
    },
    {
      "entropy": 6.203498125076294,
      "epoch": 1.5752046658462198,
      "grad_norm": 2.78125,
      "learning_rate": 0.00047598427695951846,
      "loss": 5.9068,
      "mean_token_accuracy": 0.15965503603219985,
      "num_tokens": 31846476.0,
      "step": 14720
    },
    {
      "entropy": 6.205095052719116,
      "epoch": 1.5757397399539836,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047596708688055857,
      "loss": 5.9229,
      "mean_token_accuracy": 0.1519387111067772,
      "num_tokens": 31856151.0,
      "step": 14725
    },
    {
      "entropy": 6.380023050308227,
      "epoch": 1.5762748140617475,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004759498909987023,
      "loss": 6.1231,
      "mean_token_accuracy": 0.1397507220506668,
      "num_tokens": 31867419.0,
      "step": 14730
    },
    {
      "entropy": 6.406047725677491,
      "epoch": 1.5768098881695116,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004759326893144462,
      "loss": 6.1437,
      "mean_token_accuracy": 0.14064666330814363,
      "num_tokens": 31878036.0,
      "step": 14735
    },
    {
      "entropy": 6.339225769042969,
      "epoch": 1.5773449622772753,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000475915481828287,
      "loss": 6.0703,
      "mean_token_accuracy": 0.1381204254925251,
      "num_tokens": 31887919.0,
      "step": 14740
    },
    {
      "entropy": 6.25793046951294,
      "epoch": 1.5778800363850394,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047589826854072133,
      "loss": 6.0715,
      "mean_token_accuracy": 0.14568737968802453,
      "num_tokens": 31898962.0,
      "step": 14745
    },
    {
      "entropy": 6.349749755859375,
      "epoch": 1.5784151104928033,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047588104945224654,
      "loss": 6.1931,
      "mean_token_accuracy": 0.1377079702913761,
      "num_tokens": 31911027.0,
      "step": 14750
    },
    {
      "entropy": 6.4052557945251465,
      "epoch": 1.5789501846005671,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004758638245633595,
      "loss": 6.1055,
      "mean_token_accuracy": 0.14695342108607293,
      "num_tokens": 31921226.0,
      "step": 14755
    },
    {
      "entropy": 6.302876043319702,
      "epoch": 1.5794852587083312,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004758465938745579,
      "loss": 6.0426,
      "mean_token_accuracy": 0.14637623280286788,
      "num_tokens": 31932848.0,
      "step": 14760
    },
    {
      "entropy": 6.229189729690551,
      "epoch": 1.5800203328160949,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047582935738633915,
      "loss": 5.9563,
      "mean_token_accuracy": 0.14589882642030716,
      "num_tokens": 31943757.0,
      "step": 14765
    },
    {
      "entropy": 6.345972681045533,
      "epoch": 1.580555406923859,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004758121150992009,
      "loss": 6.1343,
      "mean_token_accuracy": 0.1413264110684395,
      "num_tokens": 31954391.0,
      "step": 14770
    },
    {
      "entropy": 6.364095306396484,
      "epoch": 1.5810904810316229,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047579486701364093,
      "loss": 6.1292,
      "mean_token_accuracy": 0.14576494842767715,
      "num_tokens": 31964497.0,
      "step": 14775
    },
    {
      "entropy": 6.376286268234253,
      "epoch": 1.5816255551393867,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004757776131301575,
      "loss": 6.1179,
      "mean_token_accuracy": 0.13755100220441818,
      "num_tokens": 31974466.0,
      "step": 14780
    },
    {
      "entropy": 6.389003133773803,
      "epoch": 1.5821606292471508,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004757603534492487,
      "loss": 6.1209,
      "mean_token_accuracy": 0.14735435843467712,
      "num_tokens": 31985890.0,
      "step": 14785
    },
    {
      "entropy": 6.339353418350219,
      "epoch": 1.5826957033549145,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004757430879714128,
      "loss": 6.2139,
      "mean_token_accuracy": 0.14752443581819535,
      "num_tokens": 31996865.0,
      "step": 14790
    },
    {
      "entropy": 6.259319067001343,
      "epoch": 1.5832307774626786,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004757258166971485,
      "loss": 5.9544,
      "mean_token_accuracy": 0.1527875617146492,
      "num_tokens": 32008121.0,
      "step": 14795
    },
    {
      "entropy": 6.346803903579712,
      "epoch": 1.5837658515704425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004757085396269545,
      "loss": 6.1706,
      "mean_token_accuracy": 0.14060566276311875,
      "num_tokens": 32020552.0,
      "step": 14800
    },
    {
      "entropy": 6.337247848510742,
      "epoch": 1.5843009256782064,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047569125676132963,
      "loss": 6.1101,
      "mean_token_accuracy": 0.1461892232298851,
      "num_tokens": 32032361.0,
      "step": 14805
    },
    {
      "entropy": 6.376564359664917,
      "epoch": 1.5848359997859705,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047567396810077286,
      "loss": 6.098,
      "mean_token_accuracy": 0.1389150232076645,
      "num_tokens": 32043233.0,
      "step": 14810
    },
    {
      "entropy": 6.356270599365234,
      "epoch": 1.5853710738937343,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047565667364578344,
      "loss": 6.0546,
      "mean_token_accuracy": 0.1503261424601078,
      "num_tokens": 32054273.0,
      "step": 14815
    },
    {
      "entropy": 6.311000919342041,
      "epoch": 1.5859061480014982,
      "grad_norm": 1.25,
      "learning_rate": 0.0004756393733968608,
      "loss": 6.0538,
      "mean_token_accuracy": 0.1526677742600441,
      "num_tokens": 32064200.0,
      "step": 14820
    },
    {
      "entropy": 6.229442834854126,
      "epoch": 1.586441222109262,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004756220673545045,
      "loss": 6.0488,
      "mean_token_accuracy": 0.14899467825889587,
      "num_tokens": 32074823.0,
      "step": 14825
    },
    {
      "entropy": 6.360239696502686,
      "epoch": 1.586976296217026,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047560475551921416,
      "loss": 6.1372,
      "mean_token_accuracy": 0.14487430155277253,
      "num_tokens": 32085333.0,
      "step": 14830
    },
    {
      "entropy": 6.420678329467774,
      "epoch": 1.58751137032479,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004755874378914897,
      "loss": 6.1169,
      "mean_token_accuracy": 0.1399102970957756,
      "num_tokens": 32095827.0,
      "step": 14835
    },
    {
      "entropy": 6.327107667922974,
      "epoch": 1.588046444432554,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004755701144718311,
      "loss": 6.0979,
      "mean_token_accuracy": 0.14499910324811935,
      "num_tokens": 32106949.0,
      "step": 14840
    },
    {
      "entropy": 6.28909821510315,
      "epoch": 1.5885815185403178,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047555278526073867,
      "loss": 6.0437,
      "mean_token_accuracy": 0.15142154842615127,
      "num_tokens": 32118269.0,
      "step": 14845
    },
    {
      "entropy": 6.291136455535889,
      "epoch": 1.589116592648082,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004755354502587127,
      "loss": 6.0865,
      "mean_token_accuracy": 0.1443430095911026,
      "num_tokens": 32128406.0,
      "step": 14850
    },
    {
      "entropy": 6.295015048980713,
      "epoch": 1.5896516667558456,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047551810946625385,
      "loss": 6.1023,
      "mean_token_accuracy": 0.14704330861568451,
      "num_tokens": 32140112.0,
      "step": 14855
    },
    {
      "entropy": 6.334457731246948,
      "epoch": 1.5901867408636097,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047550076288386276,
      "loss": 6.0305,
      "mean_token_accuracy": 0.15124760419130326,
      "num_tokens": 32150010.0,
      "step": 14860
    },
    {
      "entropy": 6.356798982620239,
      "epoch": 1.5907218149713735,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004754834105120403,
      "loss": 6.1327,
      "mean_token_accuracy": 0.14115067720413207,
      "num_tokens": 32160694.0,
      "step": 14865
    },
    {
      "entropy": 6.352393960952758,
      "epoch": 1.5912568890791374,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047546605235128755,
      "loss": 6.1488,
      "mean_token_accuracy": 0.14542570933699608,
      "num_tokens": 32171947.0,
      "step": 14870
    },
    {
      "entropy": 6.3841346263885494,
      "epoch": 1.5917919631869015,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004754486884021057,
      "loss": 6.1136,
      "mean_token_accuracy": 0.1433247335255146,
      "num_tokens": 32182849.0,
      "step": 14875
    },
    {
      "entropy": 6.3551420211792,
      "epoch": 1.5923270372946652,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004754313186649961,
      "loss": 6.1136,
      "mean_token_accuracy": 0.143054236471653,
      "num_tokens": 32192871.0,
      "step": 14880
    },
    {
      "entropy": 6.375740718841553,
      "epoch": 1.5928621114024293,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047541394314046037,
      "loss": 6.1521,
      "mean_token_accuracy": 0.13562124222517014,
      "num_tokens": 32203289.0,
      "step": 14885
    },
    {
      "entropy": 6.364854955673218,
      "epoch": 1.5933971855101932,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004753965618290002,
      "loss": 6.0998,
      "mean_token_accuracy": 0.14458197802305223,
      "num_tokens": 32213238.0,
      "step": 14890
    },
    {
      "entropy": 6.276906490325928,
      "epoch": 1.593932259617957,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004753791747311175,
      "loss": 6.0135,
      "mean_token_accuracy": 0.1524654358625412,
      "num_tokens": 32223384.0,
      "step": 14895
    },
    {
      "entropy": 6.330320024490357,
      "epoch": 1.5944673337257211,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047536178184731426,
      "loss": 6.0918,
      "mean_token_accuracy": 0.1461148366332054,
      "num_tokens": 32235358.0,
      "step": 14900
    },
    {
      "entropy": 6.430008268356323,
      "epoch": 1.5950024078334848,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004753443831780927,
      "loss": 6.1454,
      "mean_token_accuracy": 0.13853377997875213,
      "num_tokens": 32246459.0,
      "step": 14905
    },
    {
      "entropy": 6.366577911376953,
      "epoch": 1.595537481941249,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004753269787239552,
      "loss": 6.0409,
      "mean_token_accuracy": 0.1458422914147377,
      "num_tokens": 32257798.0,
      "step": 14910
    },
    {
      "entropy": 6.2818849086761475,
      "epoch": 1.5960725560490128,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047530956848540435,
      "loss": 6.0749,
      "mean_token_accuracy": 0.14414617642760277,
      "num_tokens": 32268118.0,
      "step": 14915
    },
    {
      "entropy": 6.226303815841675,
      "epoch": 1.5966076301567766,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004752921524629429,
      "loss": 6.0077,
      "mean_token_accuracy": 0.14842186570167543,
      "num_tokens": 32279676.0,
      "step": 14920
    },
    {
      "entropy": 6.326802158355713,
      "epoch": 1.5971427042645407,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004752747306570737,
      "loss": 6.0638,
      "mean_token_accuracy": 0.14963992089033126,
      "num_tokens": 32290613.0,
      "step": 14925
    },
    {
      "entropy": 6.39419584274292,
      "epoch": 1.5976777783723044,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047525730306829977,
      "loss": 6.0884,
      "mean_token_accuracy": 0.14174655377864837,
      "num_tokens": 32300873.0,
      "step": 14930
    },
    {
      "entropy": 6.342017364501953,
      "epoch": 1.5982128524800685,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004752398696971244,
      "loss": 6.0922,
      "mean_token_accuracy": 0.14401648417115212,
      "num_tokens": 32312326.0,
      "step": 14935
    },
    {
      "entropy": 6.3380406379699705,
      "epoch": 1.5987479265878324,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047522243054405085,
      "loss": 6.0711,
      "mean_token_accuracy": 0.13908719569444655,
      "num_tokens": 32324322.0,
      "step": 14940
    },
    {
      "entropy": 6.332473707199097,
      "epoch": 1.5992830006955963,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004752049856095828,
      "loss": 6.0543,
      "mean_token_accuracy": 0.14296673461794854,
      "num_tokens": 32334397.0,
      "step": 14945
    },
    {
      "entropy": 6.284391403198242,
      "epoch": 1.5998180748033604,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047518753489422393,
      "loss": 6.0632,
      "mean_token_accuracy": 0.14778072237968445,
      "num_tokens": 32345131.0,
      "step": 14950
    },
    {
      "entropy": 6.284153079986572,
      "epoch": 1.6003531489111242,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047517007839847816,
      "loss": 6.0133,
      "mean_token_accuracy": 0.14540596753358842,
      "num_tokens": 32355520.0,
      "step": 14955
    },
    {
      "entropy": 6.33124475479126,
      "epoch": 1.6008882230188881,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004751526161228494,
      "loss": 6.1289,
      "mean_token_accuracy": 0.14558010324835777,
      "num_tokens": 32366695.0,
      "step": 14960
    },
    {
      "entropy": 6.245315599441528,
      "epoch": 1.601423297126652,
      "grad_norm": 1.234375,
      "learning_rate": 0.000475135148067842,
      "loss": 6.0648,
      "mean_token_accuracy": 0.14385106787085533,
      "num_tokens": 32377467.0,
      "step": 14965
    },
    {
      "entropy": 6.378677225112915,
      "epoch": 1.6019583712344159,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047511767423396033,
      "loss": 6.0675,
      "mean_token_accuracy": 0.14889362826943398,
      "num_tokens": 32388021.0,
      "step": 14970
    },
    {
      "entropy": 6.381234931945801,
      "epoch": 1.60249344534218,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000475100194621709,
      "loss": 6.043,
      "mean_token_accuracy": 0.1467800423502922,
      "num_tokens": 32398105.0,
      "step": 14975
    },
    {
      "entropy": 6.322394609451294,
      "epoch": 1.6030285194499438,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004750827092315926,
      "loss": 6.1866,
      "mean_token_accuracy": 0.1403606876730919,
      "num_tokens": 32409876.0,
      "step": 14980
    },
    {
      "entropy": 6.342028713226318,
      "epoch": 1.6035635935577077,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047506521806411613,
      "loss": 6.0372,
      "mean_token_accuracy": 0.15578233003616332,
      "num_tokens": 32420283.0,
      "step": 14985
    },
    {
      "entropy": 6.334725475311279,
      "epoch": 1.6040986676654718,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047504772111978463,
      "loss": 6.1541,
      "mean_token_accuracy": 0.13505211770534514,
      "num_tokens": 32431164.0,
      "step": 14990
    },
    {
      "entropy": 6.361402750015259,
      "epoch": 1.6046337417732355,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004750302183991032,
      "loss": 6.0538,
      "mean_token_accuracy": 0.14784753024578096,
      "num_tokens": 32442253.0,
      "step": 14995
    },
    {
      "entropy": 6.3861876964569095,
      "epoch": 1.6051688158809996,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004750127099025774,
      "loss": 6.1598,
      "mean_token_accuracy": 0.1469942808151245,
      "num_tokens": 32452800.0,
      "step": 15000
    },
    {
      "epoch": 1.6051688158809996,
      "eval_entropy": 6.255781119210379,
      "eval_loss": 6.195542812347412,
      "eval_mean_token_accuracy": 0.14796212170539158,
      "eval_num_tokens": 32452800.0,
      "eval_runtime": 22.9529,
      "eval_samples_per_second": 1293.039,
      "eval_steps_per_second": 161.635,
      "step": 15000
    },
    {
      "entropy": 6.407533025741577,
      "epoch": 1.6057038899887635,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004749951956307126,
      "loss": 6.1085,
      "mean_token_accuracy": 0.141890037804842,
      "num_tokens": 32462890.0,
      "step": 15005
    },
    {
      "entropy": 6.278880071640015,
      "epoch": 1.6062389640965273,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004749776755840148,
      "loss": 6.0645,
      "mean_token_accuracy": 0.15134830474853517,
      "num_tokens": 32473422.0,
      "step": 15010
    },
    {
      "entropy": 6.309026384353638,
      "epoch": 1.6067740382042914,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047496014976298957,
      "loss": 6.1012,
      "mean_token_accuracy": 0.1456957995891571,
      "num_tokens": 32483980.0,
      "step": 15015
    },
    {
      "entropy": 6.419345474243164,
      "epoch": 1.607309112312055,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004749426181681432,
      "loss": 6.1298,
      "mean_token_accuracy": 0.1384550742805004,
      "num_tokens": 32494905.0,
      "step": 15020
    },
    {
      "entropy": 6.3931128025054935,
      "epoch": 1.6078441864198192,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047492508079998173,
      "loss": 6.1515,
      "mean_token_accuracy": 0.13876751959323883,
      "num_tokens": 32505799.0,
      "step": 15025
    },
    {
      "entropy": 6.3200477123260494,
      "epoch": 1.608379260527583,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004749075376590117,
      "loss": 6.1296,
      "mean_token_accuracy": 0.14129856526851653,
      "num_tokens": 32516328.0,
      "step": 15030
    },
    {
      "entropy": 6.375245189666748,
      "epoch": 1.608914334635347,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004748899887457396,
      "loss": 6.0931,
      "mean_token_accuracy": 0.14479369074106216,
      "num_tokens": 32526335.0,
      "step": 15035
    },
    {
      "entropy": 6.329014682769776,
      "epoch": 1.609449408743111,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047487243406067227,
      "loss": 6.0772,
      "mean_token_accuracy": 0.148054601252079,
      "num_tokens": 32536896.0,
      "step": 15040
    },
    {
      "entropy": 6.297329711914062,
      "epoch": 1.6099844828508747,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004748548736043164,
      "loss": 6.0016,
      "mean_token_accuracy": 0.14862256199121476,
      "num_tokens": 32547604.0,
      "step": 15045
    },
    {
      "entropy": 6.36840591430664,
      "epoch": 1.6105195569586388,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047483730737717914,
      "loss": 6.117,
      "mean_token_accuracy": 0.14068032503128053,
      "num_tokens": 32558539.0,
      "step": 15050
    },
    {
      "entropy": 6.352081203460694,
      "epoch": 1.6110546310664027,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004748197353797677,
      "loss": 6.1869,
      "mean_token_accuracy": 0.1405450336635113,
      "num_tokens": 32569787.0,
      "step": 15055
    },
    {
      "entropy": 6.349471473693848,
      "epoch": 1.6115897051741666,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004748021576125895,
      "loss": 6.1254,
      "mean_token_accuracy": 0.14747645035386087,
      "num_tokens": 32580146.0,
      "step": 15060
    },
    {
      "entropy": 6.310051393508911,
      "epoch": 1.6121247792819307,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004747845740761521,
      "loss": 6.0881,
      "mean_token_accuracy": 0.14300649613142014,
      "num_tokens": 32591243.0,
      "step": 15065
    },
    {
      "entropy": 6.370094108581543,
      "epoch": 1.6126598533896943,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004747669847709631,
      "loss": 6.1445,
      "mean_token_accuracy": 0.1428090050816536,
      "num_tokens": 32602424.0,
      "step": 15070
    },
    {
      "entropy": 6.3976428508758545,
      "epoch": 1.6131949274974584,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004747493896975306,
      "loss": 6.1957,
      "mean_token_accuracy": 0.13671911805868148,
      "num_tokens": 32613703.0,
      "step": 15075
    },
    {
      "entropy": 6.338956451416015,
      "epoch": 1.6137300016052223,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004747317888563625,
      "loss": 6.1415,
      "mean_token_accuracy": 0.13831704929471017,
      "num_tokens": 32624371.0,
      "step": 15080
    },
    {
      "entropy": 6.3603373050689695,
      "epoch": 1.6142650757129862,
      "grad_norm": 1.25,
      "learning_rate": 0.00047471418224796704,
      "loss": 6.0795,
      "mean_token_accuracy": 0.14491643905639648,
      "num_tokens": 32634039.0,
      "step": 15085
    },
    {
      "entropy": 6.31891975402832,
      "epoch": 1.6148001498207503,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004746965698728526,
      "loss": 6.0603,
      "mean_token_accuracy": 0.14881459325551988,
      "num_tokens": 32644644.0,
      "step": 15090
    },
    {
      "entropy": 6.406977748870849,
      "epoch": 1.6153352239285141,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047467895173152787,
      "loss": 6.1298,
      "mean_token_accuracy": 0.14806411489844323,
      "num_tokens": 32655961.0,
      "step": 15095
    },
    {
      "entropy": 6.296580457687378,
      "epoch": 1.615870298036278,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047466132782450146,
      "loss": 6.0025,
      "mean_token_accuracy": 0.15788218826055528,
      "num_tokens": 32665506.0,
      "step": 15100
    },
    {
      "entropy": 6.29225435256958,
      "epoch": 1.616405372144042,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047464369815228215,
      "loss": 6.0433,
      "mean_token_accuracy": 0.14878323525190354,
      "num_tokens": 32676278.0,
      "step": 15105
    },
    {
      "entropy": 6.351658678054809,
      "epoch": 1.6169404462518058,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004746260627153792,
      "loss": 6.0265,
      "mean_token_accuracy": 0.1450519159436226,
      "num_tokens": 32686835.0,
      "step": 15110
    },
    {
      "entropy": 6.385546016693115,
      "epoch": 1.6174755203595699,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047460842151430166,
      "loss": 6.1782,
      "mean_token_accuracy": 0.14035655185580254,
      "num_tokens": 32696856.0,
      "step": 15115
    },
    {
      "entropy": 6.335049200057983,
      "epoch": 1.6180105944673338,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047459077454955914,
      "loss": 6.1041,
      "mean_token_accuracy": 0.1448364481329918,
      "num_tokens": 32707514.0,
      "step": 15120
    },
    {
      "entropy": 6.367025756835938,
      "epoch": 1.6185456685750976,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047457312182166094,
      "loss": 6.1821,
      "mean_token_accuracy": 0.14333444684743882,
      "num_tokens": 32718934.0,
      "step": 15125
    },
    {
      "entropy": 6.374139213562012,
      "epoch": 1.6190807426828617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004745554633311169,
      "loss": 6.0622,
      "mean_token_accuracy": 0.14312638640403746,
      "num_tokens": 32729249.0,
      "step": 15130
    },
    {
      "entropy": 6.339042520523071,
      "epoch": 1.6196158167906254,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047453779907843686,
      "loss": 6.141,
      "mean_token_accuracy": 0.13900220468640329,
      "num_tokens": 32739948.0,
      "step": 15135
    },
    {
      "entropy": 6.351451301574707,
      "epoch": 1.6201508908983895,
      "grad_norm": 1.125,
      "learning_rate": 0.00047452012906413105,
      "loss": 6.1343,
      "mean_token_accuracy": 0.14350407123565673,
      "num_tokens": 32751482.0,
      "step": 15140
    },
    {
      "entropy": 6.303815937042236,
      "epoch": 1.6206859650061534,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047450245328870943,
      "loss": 5.9799,
      "mean_token_accuracy": 0.15498422980308532,
      "num_tokens": 32761260.0,
      "step": 15145
    },
    {
      "entropy": 6.33647985458374,
      "epoch": 1.6212210391139172,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047448477175268255,
      "loss": 6.0976,
      "mean_token_accuracy": 0.14451224133372306,
      "num_tokens": 32772684.0,
      "step": 15150
    },
    {
      "entropy": 6.203201246261597,
      "epoch": 1.6217561132216813,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047446708445656095,
      "loss": 5.9098,
      "mean_token_accuracy": 0.15870766043663026,
      "num_tokens": 32784022.0,
      "step": 15155
    },
    {
      "entropy": 6.275664854049682,
      "epoch": 1.622291187329445,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004744493914008552,
      "loss": 6.0344,
      "mean_token_accuracy": 0.14753157570958136,
      "num_tokens": 32794104.0,
      "step": 15160
    },
    {
      "entropy": 6.3281909942626955,
      "epoch": 1.622826261437209,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004744316925860764,
      "loss": 6.0286,
      "mean_token_accuracy": 0.1498705193400383,
      "num_tokens": 32804038.0,
      "step": 15165
    },
    {
      "entropy": 6.32404260635376,
      "epoch": 1.623361335544973,
      "grad_norm": 1.25,
      "learning_rate": 0.0004744139880127355,
      "loss": 6.0995,
      "mean_token_accuracy": 0.1444015011191368,
      "num_tokens": 32815620.0,
      "step": 15170
    },
    {
      "entropy": 6.327903509140015,
      "epoch": 1.6238964096527368,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047439627768134366,
      "loss": 6.0123,
      "mean_token_accuracy": 0.15537536069750785,
      "num_tokens": 32826353.0,
      "step": 15175
    },
    {
      "entropy": 6.330361843109131,
      "epoch": 1.624431483760501,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004743785615924124,
      "loss": 6.1383,
      "mean_token_accuracy": 0.14190190136432648,
      "num_tokens": 32836582.0,
      "step": 15180
    },
    {
      "entropy": 6.272754859924317,
      "epoch": 1.6249665578682646,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004743608397464531,
      "loss": 6.0853,
      "mean_token_accuracy": 0.14377259388566016,
      "num_tokens": 32848073.0,
      "step": 15185
    },
    {
      "entropy": 6.342014455795288,
      "epoch": 1.6255016319760287,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047434311214397765,
      "loss": 6.1215,
      "mean_token_accuracy": 0.14226677268743515,
      "num_tokens": 32857839.0,
      "step": 15190
    },
    {
      "entropy": 6.323078870773315,
      "epoch": 1.6260367060837926,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047432537878549786,
      "loss": 6.0396,
      "mean_token_accuracy": 0.15850775092840194,
      "num_tokens": 32868087.0,
      "step": 15195
    },
    {
      "entropy": 6.3330803394317625,
      "epoch": 1.6265717801915565,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047430763967152564,
      "loss": 6.0947,
      "mean_token_accuracy": 0.1439040407538414,
      "num_tokens": 32880622.0,
      "step": 15200
    },
    {
      "entropy": 6.380010890960693,
      "epoch": 1.6271068542993206,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047428989480257337,
      "loss": 6.0964,
      "mean_token_accuracy": 0.1428353652358055,
      "num_tokens": 32891405.0,
      "step": 15205
    },
    {
      "entropy": 6.353314447402954,
      "epoch": 1.6276419284070842,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004742721441791535,
      "loss": 6.0342,
      "mean_token_accuracy": 0.1476680651307106,
      "num_tokens": 32900487.0,
      "step": 15210
    },
    {
      "entropy": 6.292839574813843,
      "epoch": 1.6281770025148483,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004742543878017784,
      "loss": 6.0059,
      "mean_token_accuracy": 0.1558583162724972,
      "num_tokens": 32912139.0,
      "step": 15215
    },
    {
      "entropy": 6.307471752166748,
      "epoch": 1.6287120766226122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047423662567096085,
      "loss": 6.0887,
      "mean_token_accuracy": 0.14408653900027274,
      "num_tokens": 32922972.0,
      "step": 15220
    },
    {
      "entropy": 6.298241376876831,
      "epoch": 1.629247150730376,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047421885778721376,
      "loss": 6.02,
      "mean_token_accuracy": 0.14925383627414704,
      "num_tokens": 32932609.0,
      "step": 15225
    },
    {
      "entropy": 6.2436936378479,
      "epoch": 1.6297822248381402,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047420108415105006,
      "loss": 6.0498,
      "mean_token_accuracy": 0.15616401731967927,
      "num_tokens": 32942706.0,
      "step": 15230
    },
    {
      "entropy": 6.3183318138122555,
      "epoch": 1.630317298945904,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047418330476298317,
      "loss": 6.054,
      "mean_token_accuracy": 0.14693014472723007,
      "num_tokens": 32952240.0,
      "step": 15235
    },
    {
      "entropy": 6.313573360443115,
      "epoch": 1.630852373053668,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004741655196235262,
      "loss": 6.1126,
      "mean_token_accuracy": 0.14152360931038857,
      "num_tokens": 32963249.0,
      "step": 15240
    },
    {
      "entropy": 6.341171169281006,
      "epoch": 1.6313874471614318,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047414772873319294,
      "loss": 6.0947,
      "mean_token_accuracy": 0.1525577649474144,
      "num_tokens": 32974004.0,
      "step": 15245
    },
    {
      "entropy": 6.3614380836486815,
      "epoch": 1.6319225212691957,
      "grad_norm": 1.296875,
      "learning_rate": 0.000474129932092497,
      "loss": 6.0709,
      "mean_token_accuracy": 0.14759385213255882,
      "num_tokens": 32985236.0,
      "step": 15250
    },
    {
      "entropy": 6.384084796905517,
      "epoch": 1.6324575953769598,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004741121297019522,
      "loss": 6.123,
      "mean_token_accuracy": 0.14208113625645638,
      "num_tokens": 32995155.0,
      "step": 15255
    },
    {
      "entropy": 6.365670394897461,
      "epoch": 1.6329926694847237,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047409432156207264,
      "loss": 6.1007,
      "mean_token_accuracy": 0.15096320062875748,
      "num_tokens": 33007562.0,
      "step": 15260
    },
    {
      "entropy": 6.293545818328857,
      "epoch": 1.6335277435924875,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047407650767337256,
      "loss": 6.0713,
      "mean_token_accuracy": 0.14037879183888435,
      "num_tokens": 33018653.0,
      "step": 15265
    },
    {
      "entropy": 6.4050110340118405,
      "epoch": 1.6340628177002516,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004740586880363663,
      "loss": 6.1785,
      "mean_token_accuracy": 0.13667976409196853,
      "num_tokens": 33029470.0,
      "step": 15270
    },
    {
      "entropy": 6.351898193359375,
      "epoch": 1.6345978918080153,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047404086265156835,
      "loss": 6.0372,
      "mean_token_accuracy": 0.15067508965730667,
      "num_tokens": 33039597.0,
      "step": 15275
    },
    {
      "entropy": 6.311202049255371,
      "epoch": 1.6351329659157794,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004740230315194935,
      "loss": 6.0682,
      "mean_token_accuracy": 0.14559220969676973,
      "num_tokens": 33050984.0,
      "step": 15280
    },
    {
      "entropy": 6.326210594177246,
      "epoch": 1.6356680400235433,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047400519464065656,
      "loss": 6.147,
      "mean_token_accuracy": 0.14421052634716033,
      "num_tokens": 33062250.0,
      "step": 15285
    },
    {
      "entropy": 6.372294664382935,
      "epoch": 1.6362031141313071,
      "grad_norm": 1.125,
      "learning_rate": 0.0004739873520155726,
      "loss": 6.1603,
      "mean_token_accuracy": 0.1429736942052841,
      "num_tokens": 33072129.0,
      "step": 15290
    },
    {
      "entropy": 6.3786357879638675,
      "epoch": 1.6367381882390712,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004739695036447568,
      "loss": 6.018,
      "mean_token_accuracy": 0.14188072606921195,
      "num_tokens": 33082475.0,
      "step": 15295
    },
    {
      "entropy": 6.273272371292114,
      "epoch": 1.637273262346835,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047395164952872455,
      "loss": 6.0511,
      "mean_token_accuracy": 0.15629055798053743,
      "num_tokens": 33093791.0,
      "step": 15300
    },
    {
      "entropy": 6.297030067443847,
      "epoch": 1.637808336454599,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004739337896679914,
      "loss": 6.0282,
      "mean_token_accuracy": 0.15334292650222778,
      "num_tokens": 33104231.0,
      "step": 15305
    },
    {
      "entropy": 6.263558006286621,
      "epoch": 1.6383434105623629,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047391592406307296,
      "loss": 6.0677,
      "mean_token_accuracy": 0.14128430485725402,
      "num_tokens": 33115118.0,
      "step": 15310
    },
    {
      "entropy": 6.336869144439698,
      "epoch": 1.6388784846701268,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004738980527144852,
      "loss": 6.0254,
      "mean_token_accuracy": 0.1534215748310089,
      "num_tokens": 33126154.0,
      "step": 15315
    },
    {
      "entropy": 6.379892015457154,
      "epoch": 1.6394135587778909,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004738801756227441,
      "loss": 6.1174,
      "mean_token_accuracy": 0.14505281299352646,
      "num_tokens": 33137342.0,
      "step": 15320
    },
    {
      "entropy": 6.25922966003418,
      "epoch": 1.6399486328856545,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047386229278836595,
      "loss": 6.0722,
      "mean_token_accuracy": 0.15355543345212935,
      "num_tokens": 33147661.0,
      "step": 15325
    },
    {
      "entropy": 6.334693145751953,
      "epoch": 1.6404837069934186,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047384440421186695,
      "loss": 6.1165,
      "mean_token_accuracy": 0.1433240368962288,
      "num_tokens": 33157106.0,
      "step": 15330
    },
    {
      "entropy": 6.374489164352417,
      "epoch": 1.6410187811011825,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004738265098937638,
      "loss": 6.1172,
      "mean_token_accuracy": 0.14548313915729522,
      "num_tokens": 33169451.0,
      "step": 15335
    },
    {
      "entropy": 6.389806747436523,
      "epoch": 1.6415538552089464,
      "grad_norm": 1.140625,
      "learning_rate": 0.000473808609834573,
      "loss": 6.1154,
      "mean_token_accuracy": 0.14536041021347046,
      "num_tokens": 33180799.0,
      "step": 15340
    },
    {
      "entropy": 6.409138965606689,
      "epoch": 1.6420889293167105,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047379070403481165,
      "loss": 6.1868,
      "mean_token_accuracy": 0.1393803261220455,
      "num_tokens": 33191948.0,
      "step": 15345
    },
    {
      "entropy": 6.326769828796387,
      "epoch": 1.6426240034244743,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047377279249499656,
      "loss": 6.0818,
      "mean_token_accuracy": 0.14053848683834075,
      "num_tokens": 33203536.0,
      "step": 15350
    },
    {
      "entropy": 6.324829387664795,
      "epoch": 1.6431590775322382,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004737548752156451,
      "loss": 6.0876,
      "mean_token_accuracy": 0.14610656052827836,
      "num_tokens": 33214526.0,
      "step": 15355
    },
    {
      "entropy": 6.281860828399658,
      "epoch": 1.643694151640002,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004737369521972745,
      "loss": 6.0496,
      "mean_token_accuracy": 0.14881749004125594,
      "num_tokens": 33226560.0,
      "step": 15360
    },
    {
      "entropy": 6.327933120727539,
      "epoch": 1.644229225747766,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047371902344040234,
      "loss": 6.0726,
      "mean_token_accuracy": 0.1475036010146141,
      "num_tokens": 33237441.0,
      "step": 15365
    },
    {
      "entropy": 6.350115728378296,
      "epoch": 1.64476429985553,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047370108894554633,
      "loss": 6.0349,
      "mean_token_accuracy": 0.14142273887991905,
      "num_tokens": 33248229.0,
      "step": 15370
    },
    {
      "entropy": 6.2672892093658445,
      "epoch": 1.645299373963294,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004736831487132243,
      "loss": 5.9826,
      "mean_token_accuracy": 0.1569042645394802,
      "num_tokens": 33258992.0,
      "step": 15375
    },
    {
      "entropy": 6.254970550537109,
      "epoch": 1.6458344480710578,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004736652027439542,
      "loss": 6.0094,
      "mean_token_accuracy": 0.1490224376320839,
      "num_tokens": 33269583.0,
      "step": 15380
    },
    {
      "entropy": 6.340966272354126,
      "epoch": 1.646369522178822,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047364725103825437,
      "loss": 6.1363,
      "mean_token_accuracy": 0.13998061046004295,
      "num_tokens": 33280236.0,
      "step": 15385
    },
    {
      "entropy": 6.330015087127686,
      "epoch": 1.6469045962865856,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000473629293596643,
      "loss": 6.0481,
      "mean_token_accuracy": 0.14414651393890382,
      "num_tokens": 33291327.0,
      "step": 15390
    },
    {
      "entropy": 6.378418064117431,
      "epoch": 1.6474396703943497,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047361133041963873,
      "loss": 6.1334,
      "mean_token_accuracy": 0.14076314568519593,
      "num_tokens": 33302125.0,
      "step": 15395
    },
    {
      "entropy": 6.357411193847656,
      "epoch": 1.6479747445021136,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004735933615077602,
      "loss": 6.0766,
      "mean_token_accuracy": 0.143499056994915,
      "num_tokens": 33312576.0,
      "step": 15400
    },
    {
      "entropy": 6.366574621200561,
      "epoch": 1.6485098186098774,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047357538686152627,
      "loss": 6.0418,
      "mean_token_accuracy": 0.14490774124860764,
      "num_tokens": 33322821.0,
      "step": 15405
    },
    {
      "entropy": 6.36388783454895,
      "epoch": 1.6490448927176415,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047355740648145596,
      "loss": 6.1242,
      "mean_token_accuracy": 0.1415700152516365,
      "num_tokens": 33333951.0,
      "step": 15410
    },
    {
      "entropy": 6.335728502273559,
      "epoch": 1.6495799668254052,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004735394203680684,
      "loss": 6.0431,
      "mean_token_accuracy": 0.14931980967521669,
      "num_tokens": 33344739.0,
      "step": 15415
    },
    {
      "entropy": 6.329457712173462,
      "epoch": 1.6501150409331693,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047352142852188304,
      "loss": 6.0213,
      "mean_token_accuracy": 0.1471880093216896,
      "num_tokens": 33354956.0,
      "step": 15420
    },
    {
      "entropy": 6.241232252120971,
      "epoch": 1.6506501150409332,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004735034309434193,
      "loss": 5.9868,
      "mean_token_accuracy": 0.15246089547872543,
      "num_tokens": 33364809.0,
      "step": 15425
    },
    {
      "entropy": 6.371157789230347,
      "epoch": 1.651185189148697,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004734854276331968,
      "loss": 6.1045,
      "mean_token_accuracy": 0.1490280121564865,
      "num_tokens": 33375324.0,
      "step": 15430
    },
    {
      "entropy": 6.393693161010742,
      "epoch": 1.6517202632564612,
      "grad_norm": 1.25,
      "learning_rate": 0.00047346741859173564,
      "loss": 6.1667,
      "mean_token_accuracy": 0.14081336930394173,
      "num_tokens": 33385963.0,
      "step": 15435
    },
    {
      "entropy": 6.388902616500855,
      "epoch": 1.6522553373642248,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004734494038195555,
      "loss": 6.1713,
      "mean_token_accuracy": 0.143441142141819,
      "num_tokens": 33395927.0,
      "step": 15440
    },
    {
      "entropy": 6.325886917114258,
      "epoch": 1.652790411471989,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004734313833171768,
      "loss": 6.0718,
      "mean_token_accuracy": 0.1456935867667198,
      "num_tokens": 33406294.0,
      "step": 15445
    },
    {
      "entropy": 6.304380750656128,
      "epoch": 1.6533254855797528,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004734133570851198,
      "loss": 6.0374,
      "mean_token_accuracy": 0.1539628028869629,
      "num_tokens": 33416995.0,
      "step": 15450
    },
    {
      "entropy": 6.249418830871582,
      "epoch": 1.6538605596875167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004733953251239049,
      "loss": 5.9842,
      "mean_token_accuracy": 0.14805758148431777,
      "num_tokens": 33428087.0,
      "step": 15455
    },
    {
      "entropy": 6.357247829437256,
      "epoch": 1.6543956337952808,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004733772874340529,
      "loss": 6.1805,
      "mean_token_accuracy": 0.14469256475567818,
      "num_tokens": 33439151.0,
      "step": 15460
    },
    {
      "entropy": 6.384392118453979,
      "epoch": 1.6549307079030444,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004733592440160846,
      "loss": 6.0611,
      "mean_token_accuracy": 0.14940285831689834,
      "num_tokens": 33449257.0,
      "step": 15465
    },
    {
      "entropy": 6.292202568054199,
      "epoch": 1.6554657820108085,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047334119487052104,
      "loss": 6.0321,
      "mean_token_accuracy": 0.14963673055171967,
      "num_tokens": 33460472.0,
      "step": 15470
    },
    {
      "entropy": 6.336902046203614,
      "epoch": 1.6560008561185724,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047332313999788334,
      "loss": 6.0215,
      "mean_token_accuracy": 0.14364624693989753,
      "num_tokens": 33470193.0,
      "step": 15475
    },
    {
      "entropy": 6.215398597717285,
      "epoch": 1.6565359302263363,
      "grad_norm": 1.25,
      "learning_rate": 0.00047330507939869284,
      "loss": 5.9719,
      "mean_token_accuracy": 0.14714485555887222,
      "num_tokens": 33481098.0,
      "step": 15480
    },
    {
      "entropy": 6.355716037750244,
      "epoch": 1.6570710043341004,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000473287013073471,
      "loss": 6.0555,
      "mean_token_accuracy": 0.1454748660326004,
      "num_tokens": 33492514.0,
      "step": 15485
    },
    {
      "entropy": 6.358044672012329,
      "epoch": 1.6576060784418643,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004732689410227396,
      "loss": 6.045,
      "mean_token_accuracy": 0.14679111838340758,
      "num_tokens": 33504254.0,
      "step": 15490
    },
    {
      "entropy": 6.2881083488464355,
      "epoch": 1.6581411525496281,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004732508632470203,
      "loss": 6.0676,
      "mean_token_accuracy": 0.14167034402489662,
      "num_tokens": 33515101.0,
      "step": 15495
    },
    {
      "entropy": 6.340255880355835,
      "epoch": 1.658676226657392,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004732327797468352,
      "loss": 6.0626,
      "mean_token_accuracy": 0.14794110208749772,
      "num_tokens": 33525052.0,
      "step": 15500
    },
    {
      "entropy": 6.408097982406616,
      "epoch": 1.6592113007651559,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004732146905227065,
      "loss": 6.2181,
      "mean_token_accuracy": 0.13463521748781204,
      "num_tokens": 33536841.0,
      "step": 15505
    },
    {
      "entropy": 6.385745668411255,
      "epoch": 1.65974637487292,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004731965955751564,
      "loss": 6.1009,
      "mean_token_accuracy": 0.1457211598753929,
      "num_tokens": 33547292.0,
      "step": 15510
    },
    {
      "entropy": 6.3324782371521,
      "epoch": 1.6602814489806839,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004731784949047075,
      "loss": 6.024,
      "mean_token_accuracy": 0.1451369933784008,
      "num_tokens": 33557674.0,
      "step": 15515
    },
    {
      "entropy": 6.324947690963745,
      "epoch": 1.6608165230884477,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004731603885118824,
      "loss": 6.1308,
      "mean_token_accuracy": 0.1391010195016861,
      "num_tokens": 33568396.0,
      "step": 15520
    },
    {
      "entropy": 6.351860761642456,
      "epoch": 1.6613515971962118,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004731422763972039,
      "loss": 6.1556,
      "mean_token_accuracy": 0.1424059599637985,
      "num_tokens": 33579688.0,
      "step": 15525
    },
    {
      "entropy": 6.370872974395752,
      "epoch": 1.6618866713039755,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047312415856119505,
      "loss": 6.1736,
      "mean_token_accuracy": 0.13958622962236406,
      "num_tokens": 33591550.0,
      "step": 15530
    },
    {
      "entropy": 6.427923154830933,
      "epoch": 1.6624217454117396,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047310603500437895,
      "loss": 6.1151,
      "mean_token_accuracy": 0.14738670140504836,
      "num_tokens": 33600792.0,
      "step": 15535
    },
    {
      "entropy": 6.270657300949097,
      "epoch": 1.6629568195195035,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047308790572727895,
      "loss": 6.0417,
      "mean_token_accuracy": 0.14855166226625444,
      "num_tokens": 33611873.0,
      "step": 15540
    },
    {
      "entropy": 6.277184867858887,
      "epoch": 1.6634918936272673,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004730697707304184,
      "loss": 6.0691,
      "mean_token_accuracy": 0.14565051048994065,
      "num_tokens": 33622960.0,
      "step": 15545
    },
    {
      "entropy": 6.3507942199707035,
      "epoch": 1.6640269677350314,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004730516300143211,
      "loss": 6.1624,
      "mean_token_accuracy": 0.14096125662326814,
      "num_tokens": 33635391.0,
      "step": 15550
    },
    {
      "entropy": 6.393361043930054,
      "epoch": 1.664562041842795,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004730334835795109,
      "loss": 6.0866,
      "mean_token_accuracy": 0.14057242050766944,
      "num_tokens": 33647942.0,
      "step": 15555
    },
    {
      "entropy": 6.386082363128662,
      "epoch": 1.6650971159505592,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004730153314265116,
      "loss": 6.0753,
      "mean_token_accuracy": 0.14181498289108277,
      "num_tokens": 33658425.0,
      "step": 15560
    },
    {
      "entropy": 6.2878796577453615,
      "epoch": 1.665632190058323,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004729971735558475,
      "loss": 6.0448,
      "mean_token_accuracy": 0.1513732761144638,
      "num_tokens": 33669645.0,
      "step": 15565
    },
    {
      "entropy": 6.353420829772949,
      "epoch": 1.666167264166087,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004729790099680428,
      "loss": 6.1325,
      "mean_token_accuracy": 0.1426586203277111,
      "num_tokens": 33680415.0,
      "step": 15570
    },
    {
      "entropy": 6.343821287155151,
      "epoch": 1.666702338273851,
      "grad_norm": 1.421875,
      "learning_rate": 0.000472960840663622,
      "loss": 6.1572,
      "mean_token_accuracy": 0.13859347403049468,
      "num_tokens": 33691425.0,
      "step": 15575
    },
    {
      "entropy": 6.360053634643554,
      "epoch": 1.6672374123816147,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004729426656431097,
      "loss": 6.0825,
      "mean_token_accuracy": 0.150442636013031,
      "num_tokens": 33703845.0,
      "step": 15580
    },
    {
      "entropy": 6.425518941879273,
      "epoch": 1.6677724864893788,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004729244849070308,
      "loss": 6.06,
      "mean_token_accuracy": 0.14594260677695275,
      "num_tokens": 33714737.0,
      "step": 15585
    },
    {
      "entropy": 6.299434471130371,
      "epoch": 1.6683075605971427,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047290629845591026,
      "loss": 6.0185,
      "mean_token_accuracy": 0.14240612983703613,
      "num_tokens": 33725042.0,
      "step": 15590
    },
    {
      "entropy": 6.2546264171600345,
      "epoch": 1.6688426347049066,
      "grad_norm": 1.15625,
      "learning_rate": 0.000472888106290273,
      "loss": 6.0581,
      "mean_token_accuracy": 0.14705297872424125,
      "num_tokens": 33736306.0,
      "step": 15595
    },
    {
      "entropy": 6.423629903793335,
      "epoch": 1.6693777088126707,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047286990841064467,
      "loss": 6.1373,
      "mean_token_accuracy": 0.13712395653128623,
      "num_tokens": 33747327.0,
      "step": 15600
    },
    {
      "entropy": 6.3581384181976315,
      "epoch": 1.6699127829204343,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004728517048175504,
      "loss": 6.0591,
      "mean_token_accuracy": 0.1436133824288845,
      "num_tokens": 33757723.0,
      "step": 15605
    },
    {
      "entropy": 6.225681352615356,
      "epoch": 1.6704478570281984,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000472833495511516,
      "loss": 6.027,
      "mean_token_accuracy": 0.14908606335520744,
      "num_tokens": 33769958.0,
      "step": 15610
    },
    {
      "entropy": 6.290177202224731,
      "epoch": 1.6709829311359623,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047281528049306727,
      "loss": 6.0516,
      "mean_token_accuracy": 0.14777415841817856,
      "num_tokens": 33780107.0,
      "step": 15615
    },
    {
      "entropy": 6.302045488357544,
      "epoch": 1.6715180052437262,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004727970597627301,
      "loss": 6.0734,
      "mean_token_accuracy": 0.14625557363033295,
      "num_tokens": 33791023.0,
      "step": 15620
    },
    {
      "entropy": 6.287527465820313,
      "epoch": 1.6720530793514903,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004727788333210305,
      "loss": 6.0881,
      "mean_token_accuracy": 0.1519307501614094,
      "num_tokens": 33801195.0,
      "step": 15625
    },
    {
      "entropy": 6.327879428863525,
      "epoch": 1.6725881534592542,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000472760601168495,
      "loss": 6.047,
      "mean_token_accuracy": 0.15066975355148315,
      "num_tokens": 33812957.0,
      "step": 15630
    },
    {
      "entropy": 6.33943395614624,
      "epoch": 1.673123227567018,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047274236330564995,
      "loss": 6.1035,
      "mean_token_accuracy": 0.1419738471508026,
      "num_tokens": 33823874.0,
      "step": 15635
    },
    {
      "entropy": 6.33526611328125,
      "epoch": 1.673658301674782,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004727241197330219,
      "loss": 6.0656,
      "mean_token_accuracy": 0.15061481148004532,
      "num_tokens": 33834284.0,
      "step": 15640
    },
    {
      "entropy": 6.376033544540405,
      "epoch": 1.6741933757825458,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004727058704511376,
      "loss": 6.0922,
      "mean_token_accuracy": 0.14600834250450134,
      "num_tokens": 33845544.0,
      "step": 15645
    },
    {
      "entropy": 6.39588189125061,
      "epoch": 1.67472844989031,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004726876154605242,
      "loss": 6.0796,
      "mean_token_accuracy": 0.14421869069337845,
      "num_tokens": 33855420.0,
      "step": 15650
    },
    {
      "entropy": 6.313169956207275,
      "epoch": 1.6752635239980738,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004726693547617086,
      "loss": 5.9646,
      "mean_token_accuracy": 0.15006972402334212,
      "num_tokens": 33864904.0,
      "step": 15655
    },
    {
      "entropy": 6.2635383129119875,
      "epoch": 1.6757985981058376,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047265108835521825,
      "loss": 6.0372,
      "mean_token_accuracy": 0.1458727613091469,
      "num_tokens": 33875139.0,
      "step": 15660
    },
    {
      "entropy": 6.310704946517944,
      "epoch": 1.6763336722136017,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004726328162415804,
      "loss": 6.05,
      "mean_token_accuracy": 0.1479961708188057,
      "num_tokens": 33886401.0,
      "step": 15665
    },
    {
      "entropy": 6.340587520599366,
      "epoch": 1.6768687463213654,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047261453842132277,
      "loss": 6.0956,
      "mean_token_accuracy": 0.1517249435186386,
      "num_tokens": 33896382.0,
      "step": 15670
    },
    {
      "entropy": 6.3356701850891115,
      "epoch": 1.6774038204291295,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047259625489497317,
      "loss": 6.0744,
      "mean_token_accuracy": 0.1475844591856003,
      "num_tokens": 33906403.0,
      "step": 15675
    },
    {
      "entropy": 6.212588882446289,
      "epoch": 1.6779388945368934,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047257796566305946,
      "loss": 5.9602,
      "mean_token_accuracy": 0.1500474914908409,
      "num_tokens": 33916992.0,
      "step": 15680
    },
    {
      "entropy": 6.341167449951172,
      "epoch": 1.6784739686446573,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004725596707261097,
      "loss": 6.0877,
      "mean_token_accuracy": 0.14545392096042634,
      "num_tokens": 33927599.0,
      "step": 15685
    },
    {
      "entropy": 6.394090986251831,
      "epoch": 1.6790090427524214,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004725413700846522,
      "loss": 6.185,
      "mean_token_accuracy": 0.1378534257411957,
      "num_tokens": 33938957.0,
      "step": 15690
    },
    {
      "entropy": 6.3581148147583,
      "epoch": 1.679544116860185,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047252306373921543,
      "loss": 6.0784,
      "mean_token_accuracy": 0.1506371945142746,
      "num_tokens": 33950169.0,
      "step": 15695
    },
    {
      "entropy": 6.279310894012451,
      "epoch": 1.680079190967949,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047250475169032804,
      "loss": 6.0762,
      "mean_token_accuracy": 0.14121356308460237,
      "num_tokens": 33960800.0,
      "step": 15700
    },
    {
      "entropy": 6.309115076065064,
      "epoch": 1.680614265075713,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004724864339385185,
      "loss": 6.1275,
      "mean_token_accuracy": 0.13843308985233307,
      "num_tokens": 33973060.0,
      "step": 15705
    },
    {
      "entropy": 6.401444673538208,
      "epoch": 1.6811493391834769,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004724681104843161,
      "loss": 6.1599,
      "mean_token_accuracy": 0.14316844344139099,
      "num_tokens": 33983185.0,
      "step": 15710
    },
    {
      "entropy": 6.393588304519653,
      "epoch": 1.681684413291241,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004724497813282497,
      "loss": 6.1045,
      "mean_token_accuracy": 0.1441698394715786,
      "num_tokens": 33993373.0,
      "step": 15715
    },
    {
      "entropy": 6.274402379989624,
      "epoch": 1.6822194873990046,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004724314464708486,
      "loss": 6.0089,
      "mean_token_accuracy": 0.15651881843805313,
      "num_tokens": 34005793.0,
      "step": 15720
    },
    {
      "entropy": 6.340911865234375,
      "epoch": 1.6827545615067687,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004724131059126423,
      "loss": 6.1623,
      "mean_token_accuracy": 0.1373403824865818,
      "num_tokens": 34016415.0,
      "step": 15725
    },
    {
      "entropy": 6.300334644317627,
      "epoch": 1.6832896356145326,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047239475965416016,
      "loss": 6.0316,
      "mean_token_accuracy": 0.14693651497364044,
      "num_tokens": 34026859.0,
      "step": 15730
    },
    {
      "entropy": 6.344279146194458,
      "epoch": 1.6838247097222965,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047237640769593224,
      "loss": 5.9264,
      "mean_token_accuracy": 0.16292384266853333,
      "num_tokens": 34037942.0,
      "step": 15735
    },
    {
      "entropy": 6.296831893920898,
      "epoch": 1.6843597838300606,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004723580500384882,
      "loss": 6.0267,
      "mean_token_accuracy": 0.1493395671248436,
      "num_tokens": 34048659.0,
      "step": 15740
    },
    {
      "entropy": 6.22990231513977,
      "epoch": 1.6848948579378242,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047233968668235817,
      "loss": 6.0357,
      "mean_token_accuracy": 0.14771153032779694,
      "num_tokens": 34059961.0,
      "step": 15745
    },
    {
      "entropy": 6.335367345809937,
      "epoch": 1.6854299320455883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047232131762807244,
      "loss": 6.0716,
      "mean_token_accuracy": 0.15087409019470216,
      "num_tokens": 34069978.0,
      "step": 15750
    },
    {
      "entropy": 6.366266202926636,
      "epoch": 1.6859650061533522,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004723029428761614,
      "loss": 6.0899,
      "mean_token_accuracy": 0.14691049233078957,
      "num_tokens": 34080301.0,
      "step": 15755
    },
    {
      "entropy": 6.291792345046997,
      "epoch": 1.686500080261116,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047228456242715564,
      "loss": 6.0908,
      "mean_token_accuracy": 0.14770060777664185,
      "num_tokens": 34092376.0,
      "step": 15760
    },
    {
      "entropy": 6.337923669815064,
      "epoch": 1.6870351543688802,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004722661762815858,
      "loss": 6.0477,
      "mean_token_accuracy": 0.14752384424209594,
      "num_tokens": 34102761.0,
      "step": 15765
    },
    {
      "entropy": 6.272897672653198,
      "epoch": 1.687570228476644,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004722477844399829,
      "loss": 6.0583,
      "mean_token_accuracy": 0.1408931076526642,
      "num_tokens": 34113370.0,
      "step": 15770
    },
    {
      "entropy": 6.361356544494629,
      "epoch": 1.688105302584408,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004722293869028779,
      "loss": 6.0687,
      "mean_token_accuracy": 0.15379619002342224,
      "num_tokens": 34125038.0,
      "step": 15775
    },
    {
      "entropy": 6.407261610031128,
      "epoch": 1.6886403766921718,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004722109836708021,
      "loss": 6.1061,
      "mean_token_accuracy": 0.14233603477478027,
      "num_tokens": 34135595.0,
      "step": 15780
    },
    {
      "entropy": 6.3187422275543215,
      "epoch": 1.6891754507999357,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047219257474428685,
      "loss": 6.0963,
      "mean_token_accuracy": 0.141315259039402,
      "num_tokens": 34146755.0,
      "step": 15785
    },
    {
      "entropy": 6.21128363609314,
      "epoch": 1.6897105249076998,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047217416012386367,
      "loss": 5.9402,
      "mean_token_accuracy": 0.15987073481082917,
      "num_tokens": 34158465.0,
      "step": 15790
    },
    {
      "entropy": 6.301998329162598,
      "epoch": 1.6902455990154637,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004721557398100644,
      "loss": 6.0401,
      "mean_token_accuracy": 0.1498855471611023,
      "num_tokens": 34170007.0,
      "step": 15795
    },
    {
      "entropy": 6.378392124176026,
      "epoch": 1.6907806731232276,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047213731380342077,
      "loss": 6.1351,
      "mean_token_accuracy": 0.14153124615550042,
      "num_tokens": 34181595.0,
      "step": 15800
    },
    {
      "entropy": 6.3542078018188475,
      "epoch": 1.6913157472309917,
      "grad_norm": 2.046875,
      "learning_rate": 0.000472118882104465,
      "loss": 6.086,
      "mean_token_accuracy": 0.14671344235539435,
      "num_tokens": 34191791.0,
      "step": 15805
    },
    {
      "entropy": 6.355745029449463,
      "epoch": 1.6918508213387553,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047210044471372915,
      "loss": 6.0044,
      "mean_token_accuracy": 0.14850385040044783,
      "num_tokens": 34202636.0,
      "step": 15810
    },
    {
      "entropy": 6.397041702270508,
      "epoch": 1.6923858954465194,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004720820016317456,
      "loss": 6.1407,
      "mean_token_accuracy": 0.1378217250108719,
      "num_tokens": 34213333.0,
      "step": 15815
    },
    {
      "entropy": 6.391458320617676,
      "epoch": 1.6929209695542833,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000472063552859047,
      "loss": 6.0824,
      "mean_token_accuracy": 0.14441529363393785,
      "num_tokens": 34223679.0,
      "step": 15820
    },
    {
      "entropy": 6.343783521652222,
      "epoch": 1.6934560436620472,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047204509839616604,
      "loss": 6.2105,
      "mean_token_accuracy": 0.13633554726839064,
      "num_tokens": 34234528.0,
      "step": 15825
    },
    {
      "entropy": 6.317445278167725,
      "epoch": 1.6939911177698113,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004720266382436355,
      "loss": 6.0237,
      "mean_token_accuracy": 0.14657506048679353,
      "num_tokens": 34246203.0,
      "step": 15830
    },
    {
      "entropy": 6.318903923034668,
      "epoch": 1.694526191877575,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047200817240198846,
      "loss": 6.0201,
      "mean_token_accuracy": 0.1465979151427746,
      "num_tokens": 34257231.0,
      "step": 15835
    },
    {
      "entropy": 6.314963722229004,
      "epoch": 1.695061265985339,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047198970087175817,
      "loss": 6.0427,
      "mean_token_accuracy": 0.14842984080314636,
      "num_tokens": 34268900.0,
      "step": 15840
    },
    {
      "entropy": 6.310456895828247,
      "epoch": 1.695596340093103,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004719712236534779,
      "loss": 6.1289,
      "mean_token_accuracy": 0.14229328706860542,
      "num_tokens": 34279343.0,
      "step": 15845
    },
    {
      "entropy": 6.332660675048828,
      "epoch": 1.6961314142008668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004719527407476812,
      "loss": 6.0687,
      "mean_token_accuracy": 0.15176492035388947,
      "num_tokens": 34291952.0,
      "step": 15850
    },
    {
      "entropy": 6.389016103744507,
      "epoch": 1.6966664883086309,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047193425215490186,
      "loss": 6.0941,
      "mean_token_accuracy": 0.1357884883880615,
      "num_tokens": 34302516.0,
      "step": 15855
    },
    {
      "entropy": 6.336066389083863,
      "epoch": 1.6972015624163945,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047191575787567373,
      "loss": 6.0191,
      "mean_token_accuracy": 0.14880041182041168,
      "num_tokens": 34313094.0,
      "step": 15860
    },
    {
      "entropy": 6.3467083930969235,
      "epoch": 1.6977366365241586,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004718972579105306,
      "loss": 6.1494,
      "mean_token_accuracy": 0.14323625713586807,
      "num_tokens": 34324803.0,
      "step": 15865
    },
    {
      "entropy": 6.3344385623931885,
      "epoch": 1.6982717106319225,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004718787522600069,
      "loss": 6.102,
      "mean_token_accuracy": 0.13894970789551736,
      "num_tokens": 34336386.0,
      "step": 15870
    },
    {
      "entropy": 6.399340963363647,
      "epoch": 1.6988067847396864,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004718602409246369,
      "loss": 6.1183,
      "mean_token_accuracy": 0.1405338652431965,
      "num_tokens": 34347085.0,
      "step": 15875
    },
    {
      "entropy": 6.333723783493042,
      "epoch": 1.6993418588474505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047184172390495505,
      "loss": 6.023,
      "mean_token_accuracy": 0.15314479172229767,
      "num_tokens": 34358146.0,
      "step": 15880
    },
    {
      "entropy": 6.2894124507904055,
      "epoch": 1.6998769329552141,
      "grad_norm": 1.125,
      "learning_rate": 0.0004718232012014961,
      "loss": 6.0743,
      "mean_token_accuracy": 0.14483475014567376,
      "num_tokens": 34370403.0,
      "step": 15885
    },
    {
      "entropy": 6.36286096572876,
      "epoch": 1.7004120070629782,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047180467281479496,
      "loss": 6.0679,
      "mean_token_accuracy": 0.15029091835021974,
      "num_tokens": 34381130.0,
      "step": 15890
    },
    {
      "entropy": 6.275781679153442,
      "epoch": 1.7009470811707421,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004717861387453865,
      "loss": 6.1121,
      "mean_token_accuracy": 0.1421474903821945,
      "num_tokens": 34391610.0,
      "step": 15895
    },
    {
      "entropy": 6.303278160095215,
      "epoch": 1.701482155278506,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004717675989938059,
      "loss": 6.075,
      "mean_token_accuracy": 0.14379487931728363,
      "num_tokens": 34402262.0,
      "step": 15900
    },
    {
      "entropy": 6.356523656845093,
      "epoch": 1.70201722938627,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047174905356058857,
      "loss": 6.0831,
      "mean_token_accuracy": 0.14323097020387648,
      "num_tokens": 34412707.0,
      "step": 15905
    },
    {
      "entropy": 6.297907304763794,
      "epoch": 1.702552303494034,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047173050244626993,
      "loss": 6.0134,
      "mean_token_accuracy": 0.15493876338005066,
      "num_tokens": 34423203.0,
      "step": 15910
    },
    {
      "entropy": 6.306429004669189,
      "epoch": 1.7030873776017978,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004717119456513857,
      "loss": 6.1159,
      "mean_token_accuracy": 0.14207639470696448,
      "num_tokens": 34433279.0,
      "step": 15915
    },
    {
      "entropy": 6.356398630142212,
      "epoch": 1.7036224517095617,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047169338317647164,
      "loss": 6.1345,
      "mean_token_accuracy": 0.14349216222763062,
      "num_tokens": 34444601.0,
      "step": 15920
    },
    {
      "entropy": 6.385204267501831,
      "epoch": 1.7041575258173256,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047167481502206383,
      "loss": 6.0859,
      "mean_token_accuracy": 0.1485481083393097,
      "num_tokens": 34455103.0,
      "step": 15925
    },
    {
      "entropy": 6.306486988067627,
      "epoch": 1.7046925999250897,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047165624118869834,
      "loss": 6.1182,
      "mean_token_accuracy": 0.139141433686018,
      "num_tokens": 34465835.0,
      "step": 15930
    },
    {
      "entropy": 6.413507175445557,
      "epoch": 1.7052276740328536,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004716376616769115,
      "loss": 6.1404,
      "mean_token_accuracy": 0.14067093804478645,
      "num_tokens": 34476602.0,
      "step": 15935
    },
    {
      "entropy": 6.385562229156494,
      "epoch": 1.7057627481406175,
      "grad_norm": 1.375,
      "learning_rate": 0.0004716190764872399,
      "loss": 6.0255,
      "mean_token_accuracy": 0.14416321218013764,
      "num_tokens": 34488026.0,
      "step": 15940
    },
    {
      "entropy": 6.346787357330323,
      "epoch": 1.7062978222483816,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047160048562022006,
      "loss": 6.1318,
      "mean_token_accuracy": 0.14960070103406906,
      "num_tokens": 34497837.0,
      "step": 15945
    },
    {
      "entropy": 6.321001482009888,
      "epoch": 1.7068328963561452,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004715818890763888,
      "loss": 6.0687,
      "mean_token_accuracy": 0.1490338370203972,
      "num_tokens": 34507727.0,
      "step": 15950
    },
    {
      "entropy": 6.385450553894043,
      "epoch": 1.7073679704639093,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047156328685628307,
      "loss": 6.072,
      "mean_token_accuracy": 0.1403654083609581,
      "num_tokens": 34518595.0,
      "step": 15955
    },
    {
      "entropy": 6.286150646209717,
      "epoch": 1.7079030445716732,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004715446789604401,
      "loss": 6.0126,
      "mean_token_accuracy": 0.15153414905071258,
      "num_tokens": 34529415.0,
      "step": 15960
    },
    {
      "entropy": 6.32280330657959,
      "epoch": 1.708438118679437,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004715260653893972,
      "loss": 6.0588,
      "mean_token_accuracy": 0.1525798484683037,
      "num_tokens": 34539593.0,
      "step": 15965
    },
    {
      "entropy": 6.362377882003784,
      "epoch": 1.7089731927872012,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004715074461436917,
      "loss": 6.0502,
      "mean_token_accuracy": 0.14747428447008132,
      "num_tokens": 34549441.0,
      "step": 15970
    },
    {
      "entropy": 6.296057987213135,
      "epoch": 1.7095082668949648,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047148882122386126,
      "loss": 6.007,
      "mean_token_accuracy": 0.14621836617588996,
      "num_tokens": 34560211.0,
      "step": 15975
    },
    {
      "entropy": 6.363714265823364,
      "epoch": 1.710043341002729,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047147019063044387,
      "loss": 6.1204,
      "mean_token_accuracy": 0.14531927555799484,
      "num_tokens": 34570479.0,
      "step": 15980
    },
    {
      "entropy": 6.343081855773926,
      "epoch": 1.7105784151104928,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047145155436397726,
      "loss": 6.1021,
      "mean_token_accuracy": 0.14623943269252776,
      "num_tokens": 34581382.0,
      "step": 15985
    },
    {
      "entropy": 6.30502724647522,
      "epoch": 1.7111134892182567,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004714329124249996,
      "loss": 6.0365,
      "mean_token_accuracy": 0.15194238796830178,
      "num_tokens": 34592131.0,
      "step": 15990
    },
    {
      "entropy": 6.381475448608398,
      "epoch": 1.7116485633260208,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004714142648140492,
      "loss": 6.1108,
      "mean_token_accuracy": 0.14534670412540435,
      "num_tokens": 34604407.0,
      "step": 15995
    },
    {
      "entropy": 6.327456665039063,
      "epoch": 1.7121836374337844,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004713956115316645,
      "loss": 5.9947,
      "mean_token_accuracy": 0.144296994805336,
      "num_tokens": 34614842.0,
      "step": 16000
    },
    {
      "entropy": 6.173287343978882,
      "epoch": 1.7127187115415485,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004713769525783841,
      "loss": 5.9508,
      "mean_token_accuracy": 0.15218990594148635,
      "num_tokens": 34625991.0,
      "step": 16005
    },
    {
      "entropy": 6.27839994430542,
      "epoch": 1.7132537856493124,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004713582879547469,
      "loss": 6.1219,
      "mean_token_accuracy": 0.14842924773693084,
      "num_tokens": 34637317.0,
      "step": 16010
    },
    {
      "entropy": 6.386443948745727,
      "epoch": 1.7137888597570763,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004713396176612916,
      "loss": 6.0904,
      "mean_token_accuracy": 0.1413798987865448,
      "num_tokens": 34648282.0,
      "step": 16015
    },
    {
      "entropy": 6.344794702529907,
      "epoch": 1.7143239338648404,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004713209416985575,
      "loss": 6.0688,
      "mean_token_accuracy": 0.14315328374505043,
      "num_tokens": 34658638.0,
      "step": 16020
    },
    {
      "entropy": 6.307915353775025,
      "epoch": 1.714859007972604,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047130226006708375,
      "loss": 6.0967,
      "mean_token_accuracy": 0.14135226905345916,
      "num_tokens": 34669308.0,
      "step": 16025
    },
    {
      "entropy": 6.329539489746094,
      "epoch": 1.7153940820803681,
      "grad_norm": 1.125,
      "learning_rate": 0.00047128357276740986,
      "loss": 6.0688,
      "mean_token_accuracy": 0.14391349107027054,
      "num_tokens": 34679513.0,
      "step": 16030
    },
    {
      "entropy": 6.360099458694458,
      "epoch": 1.715929156188132,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004712648798000754,
      "loss": 6.0678,
      "mean_token_accuracy": 0.14611221998929977,
      "num_tokens": 34690736.0,
      "step": 16035
    },
    {
      "entropy": 6.356008672714234,
      "epoch": 1.716464230295896,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047124618116562003,
      "loss": 6.0554,
      "mean_token_accuracy": 0.14317544549703598,
      "num_tokens": 34701272.0,
      "step": 16040
    },
    {
      "entropy": 6.247581624984742,
      "epoch": 1.71699930440366,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004712274768645839,
      "loss": 5.9812,
      "mean_token_accuracy": 0.1547914370894432,
      "num_tokens": 34712486.0,
      "step": 16045
    },
    {
      "entropy": 6.399283170700073,
      "epoch": 1.7175343785114239,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047120876689750685,
      "loss": 6.0609,
      "mean_token_accuracy": 0.14695574939250947,
      "num_tokens": 34721979.0,
      "step": 16050
    },
    {
      "entropy": 6.321034288406372,
      "epoch": 1.7180694526191878,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004711900512649292,
      "loss": 6.0584,
      "mean_token_accuracy": 0.14245199486613275,
      "num_tokens": 34733496.0,
      "step": 16055
    },
    {
      "entropy": 6.330378437042237,
      "epoch": 1.7186045267269516,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004711713299673914,
      "loss": 6.0668,
      "mean_token_accuracy": 0.15036619901657106,
      "num_tokens": 34744223.0,
      "step": 16060
    },
    {
      "entropy": 6.2455933570861815,
      "epoch": 1.7191396008347155,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047115260300543407,
      "loss": 6.1372,
      "mean_token_accuracy": 0.14558727219700812,
      "num_tokens": 34754931.0,
      "step": 16065
    },
    {
      "entropy": 6.260196161270142,
      "epoch": 1.7196746749424796,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004711338703795978,
      "loss": 5.9649,
      "mean_token_accuracy": 0.14685256257653237,
      "num_tokens": 34766285.0,
      "step": 16070
    },
    {
      "entropy": 6.342743682861328,
      "epoch": 1.7202097490502435,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004711151320904237,
      "loss": 6.047,
      "mean_token_accuracy": 0.15122140794992447,
      "num_tokens": 34776979.0,
      "step": 16075
    },
    {
      "entropy": 6.3693536758422855,
      "epoch": 1.7207448231580074,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004710963881384527,
      "loss": 6.1395,
      "mean_token_accuracy": 0.13906437829136847,
      "num_tokens": 34787466.0,
      "step": 16080
    },
    {
      "entropy": 6.4205950736999515,
      "epoch": 1.7212798972657715,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000471077638524226,
      "loss": 6.0472,
      "mean_token_accuracy": 0.148925518989563,
      "num_tokens": 34797664.0,
      "step": 16085
    },
    {
      "entropy": 6.25348162651062,
      "epoch": 1.7218149713735351,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047105888324828506,
      "loss": 6.0382,
      "mean_token_accuracy": 0.15807190537452698,
      "num_tokens": 34809082.0,
      "step": 16090
    },
    {
      "entropy": 6.296187925338745,
      "epoch": 1.7223500454812992,
      "grad_norm": 1.765625,
      "learning_rate": 0.00047104012231117143,
      "loss": 6.0576,
      "mean_token_accuracy": 0.14629951119422913,
      "num_tokens": 34820223.0,
      "step": 16095
    },
    {
      "entropy": 6.219951820373535,
      "epoch": 1.722885119589063,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047102135571342683,
      "loss": 5.9635,
      "mean_token_accuracy": 0.14864859506487846,
      "num_tokens": 34830428.0,
      "step": 16100
    },
    {
      "entropy": 6.2989856719970705,
      "epoch": 1.723420193696827,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047100258345559307,
      "loss": 6.1012,
      "mean_token_accuracy": 0.14136408120393754,
      "num_tokens": 34842787.0,
      "step": 16105
    },
    {
      "entropy": 6.384682846069336,
      "epoch": 1.723955267804591,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004709838055382123,
      "loss": 6.0405,
      "mean_token_accuracy": 0.14370236322283744,
      "num_tokens": 34854323.0,
      "step": 16110
    },
    {
      "entropy": 6.326080703735352,
      "epoch": 1.7244903419123547,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004709650219618266,
      "loss": 6.0223,
      "mean_token_accuracy": 0.14899590015411376,
      "num_tokens": 34865279.0,
      "step": 16115
    },
    {
      "entropy": 6.282795763015747,
      "epoch": 1.7250254160201188,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004709462327269785,
      "loss": 6.0483,
      "mean_token_accuracy": 0.14657021984457969,
      "num_tokens": 34875951.0,
      "step": 16120
    },
    {
      "entropy": 6.2755876064300535,
      "epoch": 1.7255604901278827,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004709274378342105,
      "loss": 6.0692,
      "mean_token_accuracy": 0.15216244757175446,
      "num_tokens": 34885815.0,
      "step": 16125
    },
    {
      "entropy": 6.37507700920105,
      "epoch": 1.7260955642356466,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047090863728406517,
      "loss": 6.0919,
      "mean_token_accuracy": 0.14527183771133423,
      "num_tokens": 34897158.0,
      "step": 16130
    },
    {
      "entropy": 6.3776219367980955,
      "epoch": 1.7266306383434107,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004708898310770856,
      "loss": 6.108,
      "mean_token_accuracy": 0.14382792189717292,
      "num_tokens": 34907815.0,
      "step": 16135
    },
    {
      "entropy": 6.259689569473267,
      "epoch": 1.7271657124511743,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004708710192138146,
      "loss": 6.0592,
      "mean_token_accuracy": 0.15149239003658294,
      "num_tokens": 34918820.0,
      "step": 16140
    },
    {
      "entropy": 6.371396064758301,
      "epoch": 1.7277007865589384,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004708522016947954,
      "loss": 6.1113,
      "mean_token_accuracy": 0.14508823305368423,
      "num_tokens": 34929998.0,
      "step": 16145
    },
    {
      "entropy": 6.320157432556153,
      "epoch": 1.7282358606667023,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004708333785205715,
      "loss": 6.0191,
      "mean_token_accuracy": 0.1565701559185982,
      "num_tokens": 34940989.0,
      "step": 16150
    },
    {
      "entropy": 6.3278289318084715,
      "epoch": 1.7287709347744662,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004708145496916862,
      "loss": 6.0438,
      "mean_token_accuracy": 0.14891254603862764,
      "num_tokens": 34952433.0,
      "step": 16155
    },
    {
      "entropy": 6.3724617004394535,
      "epoch": 1.7293060088822303,
      "grad_norm": 1.25,
      "learning_rate": 0.00047079571520868344,
      "loss": 6.0927,
      "mean_token_accuracy": 0.14175234884023666,
      "num_tokens": 34963074.0,
      "step": 16160
    },
    {
      "entropy": 6.300539588928222,
      "epoch": 1.729841082989994,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047077687507210674,
      "loss": 6.0102,
      "mean_token_accuracy": 0.15126861929893493,
      "num_tokens": 34974315.0,
      "step": 16165
    },
    {
      "entropy": 6.3259190082550045,
      "epoch": 1.730376157097758,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004707580292825004,
      "loss": 5.9913,
      "mean_token_accuracy": 0.16079479679465294,
      "num_tokens": 34985059.0,
      "step": 16170
    },
    {
      "entropy": 6.325475978851318,
      "epoch": 1.730911231205522,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004707391778404085,
      "loss": 6.1617,
      "mean_token_accuracy": 0.13543324768543244,
      "num_tokens": 34996156.0,
      "step": 16175
    },
    {
      "entropy": 6.368440437316894,
      "epoch": 1.7314463053132858,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004707203207463752,
      "loss": 6.0693,
      "mean_token_accuracy": 0.14433252215385436,
      "num_tokens": 35007410.0,
      "step": 16180
    },
    {
      "entropy": 6.350421619415283,
      "epoch": 1.73198137942105,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004707014580009453,
      "loss": 6.0469,
      "mean_token_accuracy": 0.1456167921423912,
      "num_tokens": 35018492.0,
      "step": 16185
    },
    {
      "entropy": 6.346235370635986,
      "epoch": 1.7325164535288138,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004706825896046632,
      "loss": 6.0915,
      "mean_token_accuracy": 0.14456754177808762,
      "num_tokens": 35029923.0,
      "step": 16190
    },
    {
      "entropy": 6.294757413864136,
      "epoch": 1.7330515276365777,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047066371555807386,
      "loss": 6.0259,
      "mean_token_accuracy": 0.14315504655241967,
      "num_tokens": 35040370.0,
      "step": 16195
    },
    {
      "entropy": 6.336321020126343,
      "epoch": 1.7335866017443415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004706448358617222,
      "loss": 6.0981,
      "mean_token_accuracy": 0.13890049681067468,
      "num_tokens": 35051414.0,
      "step": 16200
    },
    {
      "entropy": 6.411770439147949,
      "epoch": 1.7341216758521054,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047062595051615343,
      "loss": 6.0305,
      "mean_token_accuracy": 0.15238111168146135,
      "num_tokens": 35061015.0,
      "step": 16205
    },
    {
      "entropy": 6.342853260040283,
      "epoch": 1.7346567499598695,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004706070595219128,
      "loss": 6.1248,
      "mean_token_accuracy": 0.1468021884560585,
      "num_tokens": 35071757.0,
      "step": 16210
    },
    {
      "entropy": 6.325554132461548,
      "epoch": 1.7351918240676334,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047058816287954587,
      "loss": 6.0913,
      "mean_token_accuracy": 0.1443137988448143,
      "num_tokens": 35082912.0,
      "step": 16215
    },
    {
      "entropy": 6.34667854309082,
      "epoch": 1.7357268981753973,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047056926058959813,
      "loss": 6.0482,
      "mean_token_accuracy": 0.1581280380487442,
      "num_tokens": 35094103.0,
      "step": 16220
    },
    {
      "entropy": 6.306059694290161,
      "epoch": 1.7362619722831614,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004705503526526155,
      "loss": 5.9946,
      "mean_token_accuracy": 0.15630710124969482,
      "num_tokens": 35105960.0,
      "step": 16225
    },
    {
      "entropy": 6.265002918243408,
      "epoch": 1.736797046390925,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047053143906914405,
      "loss": 6.0047,
      "mean_token_accuracy": 0.15657003670930864,
      "num_tokens": 35116482.0,
      "step": 16230
    },
    {
      "entropy": 6.365628242492676,
      "epoch": 1.7373321204986891,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047051251983972966,
      "loss": 6.0939,
      "mean_token_accuracy": 0.14361643195152282,
      "num_tokens": 35127976.0,
      "step": 16235
    },
    {
      "entropy": 6.324440860748291,
      "epoch": 1.737867194606453,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004704935949649187,
      "loss": 6.0818,
      "mean_token_accuracy": 0.14946580529212952,
      "num_tokens": 35137712.0,
      "step": 16240
    },
    {
      "entropy": 6.251816940307617,
      "epoch": 1.7384022687142169,
      "grad_norm": 1.125,
      "learning_rate": 0.0004704746644452577,
      "loss": 5.989,
      "mean_token_accuracy": 0.1528105229139328,
      "num_tokens": 35148524.0,
      "step": 16245
    },
    {
      "entropy": 6.326816177368164,
      "epoch": 1.738937342821981,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047045572828129323,
      "loss": 6.0708,
      "mean_token_accuracy": 0.14791148006916047,
      "num_tokens": 35159080.0,
      "step": 16250
    },
    {
      "entropy": 6.2343793392181395,
      "epoch": 1.7394724169297446,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000470436786473572,
      "loss": 5.9797,
      "mean_token_accuracy": 0.15180525407195092,
      "num_tokens": 35169993.0,
      "step": 16255
    },
    {
      "entropy": 6.290544223785401,
      "epoch": 1.7400074910375087,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004704178390226411,
      "loss": 6.0062,
      "mean_token_accuracy": 0.14562134593725204,
      "num_tokens": 35181413.0,
      "step": 16260
    },
    {
      "entropy": 6.2540717124938965,
      "epoch": 1.7405425651452726,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047039888592904753,
      "loss": 6.02,
      "mean_token_accuracy": 0.14805859178304673,
      "num_tokens": 35191600.0,
      "step": 16265
    },
    {
      "entropy": 6.283225393295288,
      "epoch": 1.7410776392530365,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047037992719333864,
      "loss": 6.0445,
      "mean_token_accuracy": 0.1468050166964531,
      "num_tokens": 35202535.0,
      "step": 16270
    },
    {
      "entropy": 6.334268856048584,
      "epoch": 1.7416127133608006,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047036096281606176,
      "loss": 6.0301,
      "mean_token_accuracy": 0.14224215894937514,
      "num_tokens": 35214798.0,
      "step": 16275
    },
    {
      "entropy": 6.3207920551300045,
      "epoch": 1.7421477874685642,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047034199279776456,
      "loss": 5.9997,
      "mean_token_accuracy": 0.1522619366645813,
      "num_tokens": 35224870.0,
      "step": 16280
    },
    {
      "entropy": 6.294228601455688,
      "epoch": 1.7426828615763283,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004703230171389948,
      "loss": 6.024,
      "mean_token_accuracy": 0.1479485347867012,
      "num_tokens": 35235408.0,
      "step": 16285
    },
    {
      "entropy": 6.271020841598511,
      "epoch": 1.7432179356840922,
      "grad_norm": 1.25,
      "learning_rate": 0.00047030403584030024,
      "loss": 6.0433,
      "mean_token_accuracy": 0.15014452263712882,
      "num_tokens": 35246310.0,
      "step": 16290
    },
    {
      "entropy": 6.313498687744141,
      "epoch": 1.743753009791856,
      "grad_norm": 1.125,
      "learning_rate": 0.0004702850489022292,
      "loss": 6.0729,
      "mean_token_accuracy": 0.14587150365114213,
      "num_tokens": 35257027.0,
      "step": 16295
    },
    {
      "entropy": 6.398423767089843,
      "epoch": 1.7442880838996202,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047026605632532976,
      "loss": 6.0483,
      "mean_token_accuracy": 0.1472041055560112,
      "num_tokens": 35267397.0,
      "step": 16300
    },
    {
      "entropy": 6.224659538269043,
      "epoch": 1.7448231580073839,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047024705811015045,
      "loss": 5.9922,
      "mean_token_accuracy": 0.15108728408813477,
      "num_tokens": 35278775.0,
      "step": 16305
    },
    {
      "entropy": 6.265177774429321,
      "epoch": 1.745358232115148,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004702280542572397,
      "loss": 5.9332,
      "mean_token_accuracy": 0.1576532632112503,
      "num_tokens": 35288818.0,
      "step": 16310
    },
    {
      "entropy": 6.328145503997803,
      "epoch": 1.7458933062229118,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004702090447671464,
      "loss": 6.1449,
      "mean_token_accuracy": 0.13946997225284577,
      "num_tokens": 35299385.0,
      "step": 16315
    },
    {
      "entropy": 6.330541706085205,
      "epoch": 1.7464283803306757,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004701900296404193,
      "loss": 6.012,
      "mean_token_accuracy": 0.1516396924853325,
      "num_tokens": 35310626.0,
      "step": 16320
    },
    {
      "entropy": 6.3268335342407225,
      "epoch": 1.7469634544384398,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004701710088776075,
      "loss": 6.1098,
      "mean_token_accuracy": 0.1416493348777294,
      "num_tokens": 35321615.0,
      "step": 16325
    },
    {
      "entropy": 6.361673212051391,
      "epoch": 1.7474985285462037,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047015198247926025,
      "loss": 6.103,
      "mean_token_accuracy": 0.14403798133134843,
      "num_tokens": 35332012.0,
      "step": 16330
    },
    {
      "entropy": 6.37499303817749,
      "epoch": 1.7480336026539676,
      "grad_norm": 1.21875,
      "learning_rate": 0.000470132950445927,
      "loss": 6.104,
      "mean_token_accuracy": 0.1380776956677437,
      "num_tokens": 35342153.0,
      "step": 16335
    },
    {
      "entropy": 6.3214499950408936,
      "epoch": 1.7485686767617317,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004701139127781571,
      "loss": 6.0453,
      "mean_token_accuracy": 0.15050827711820602,
      "num_tokens": 35352858.0,
      "step": 16340
    },
    {
      "entropy": 6.278564214706421,
      "epoch": 1.7491037508694953,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004700948694765005,
      "loss": 5.9877,
      "mean_token_accuracy": 0.15004172921180725,
      "num_tokens": 35363489.0,
      "step": 16345
    },
    {
      "entropy": 6.388910484313965,
      "epoch": 1.7496388249772594,
      "grad_norm": 1.375,
      "learning_rate": 0.0004700758205415069,
      "loss": 6.16,
      "mean_token_accuracy": 0.1427878201007843,
      "num_tokens": 35375816.0,
      "step": 16350
    },
    {
      "entropy": 6.358825016021728,
      "epoch": 1.7501738990850233,
      "grad_norm": 1.25,
      "learning_rate": 0.00047005676597372645,
      "loss": 6.0582,
      "mean_token_accuracy": 0.14049867987632753,
      "num_tokens": 35386708.0,
      "step": 16355
    },
    {
      "entropy": 6.289164018630982,
      "epoch": 1.7507089731927872,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047003770577370925,
      "loss": 6.1165,
      "mean_token_accuracy": 0.14247289672493935,
      "num_tokens": 35397734.0,
      "step": 16360
    },
    {
      "entropy": 6.3904094219207765,
      "epoch": 1.7512440473005513,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047001863994200574,
      "loss": 6.0513,
      "mean_token_accuracy": 0.15281543359160424,
      "num_tokens": 35408870.0,
      "step": 16365
    },
    {
      "entropy": 6.287455368041992,
      "epoch": 1.751779121408315,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004699995684791664,
      "loss": 6.004,
      "mean_token_accuracy": 0.1497369885444641,
      "num_tokens": 35419498.0,
      "step": 16370
    },
    {
      "entropy": 6.293266153335571,
      "epoch": 1.752314195516079,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046998049138574177,
      "loss": 6.0335,
      "mean_token_accuracy": 0.14996637552976608,
      "num_tokens": 35430690.0,
      "step": 16375
    },
    {
      "entropy": 6.2703132152557375,
      "epoch": 1.752849269623843,
      "grad_norm": 1.1875,
      "learning_rate": 0.000469961408662283,
      "loss": 5.9669,
      "mean_token_accuracy": 0.1548752248287201,
      "num_tokens": 35441007.0,
      "step": 16380
    },
    {
      "entropy": 6.3006064891815186,
      "epoch": 1.7533843437316068,
      "grad_norm": 1.15625,
      "learning_rate": 0.000469942320309341,
      "loss": 6.0462,
      "mean_token_accuracy": 0.1585925817489624,
      "num_tokens": 35453202.0,
      "step": 16385
    },
    {
      "entropy": 6.328655147552491,
      "epoch": 1.7539194178393709,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046992322632746677,
      "loss": 6.0562,
      "mean_token_accuracy": 0.15081461519002914,
      "num_tokens": 35463303.0,
      "step": 16390
    },
    {
      "entropy": 6.3154651641845705,
      "epoch": 1.7544544919471345,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046990412671721186,
      "loss": 6.0596,
      "mean_token_accuracy": 0.15097276717424393,
      "num_tokens": 35474104.0,
      "step": 16395
    },
    {
      "entropy": 6.283804130554199,
      "epoch": 1.7549895660548986,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004698850214791276,
      "loss": 6.0599,
      "mean_token_accuracy": 0.14897584468126296,
      "num_tokens": 35486412.0,
      "step": 16400
    },
    {
      "entropy": 6.326126146316528,
      "epoch": 1.7555246401626625,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004698659106137658,
      "loss": 6.0319,
      "mean_token_accuracy": 0.14529660791158677,
      "num_tokens": 35497505.0,
      "step": 16405
    },
    {
      "entropy": 6.399095869064331,
      "epoch": 1.7560597142704264,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004698467941216782,
      "loss": 6.0986,
      "mean_token_accuracy": 0.15053761154413223,
      "num_tokens": 35508397.0,
      "step": 16410
    },
    {
      "entropy": 6.366864252090454,
      "epoch": 1.7565947883781905,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046982767200341675,
      "loss": 6.13,
      "mean_token_accuracy": 0.1390926033258438,
      "num_tokens": 35520261.0,
      "step": 16415
    },
    {
      "entropy": 6.31164288520813,
      "epoch": 1.7571298624859542,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046980854425953365,
      "loss": 6.0165,
      "mean_token_accuracy": 0.1454442948102951,
      "num_tokens": 35531816.0,
      "step": 16420
    },
    {
      "entropy": 6.391896438598633,
      "epoch": 1.7576649365937183,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004697894108905813,
      "loss": 6.1137,
      "mean_token_accuracy": 0.14583204537630082,
      "num_tokens": 35542233.0,
      "step": 16425
    },
    {
      "entropy": 6.34067964553833,
      "epoch": 1.7582000107014821,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046977027189711197,
      "loss": 6.0407,
      "mean_token_accuracy": 0.14762033224105836,
      "num_tokens": 35552881.0,
      "step": 16430
    },
    {
      "entropy": 6.34464659690857,
      "epoch": 1.758735084809246,
      "grad_norm": 1.375,
      "learning_rate": 0.0004697511272796785,
      "loss": 5.9962,
      "mean_token_accuracy": 0.1505495563149452,
      "num_tokens": 35562392.0,
      "step": 16435
    },
    {
      "entropy": 6.303057384490967,
      "epoch": 1.75927015891701,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004697319770388335,
      "loss": 6.0715,
      "mean_token_accuracy": 0.1403426192700863,
      "num_tokens": 35574587.0,
      "step": 16440
    },
    {
      "entropy": 6.280681419372558,
      "epoch": 1.759805233024774,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004697128211751301,
      "loss": 5.9399,
      "mean_token_accuracy": 0.15116320103406905,
      "num_tokens": 35585178.0,
      "step": 16445
    },
    {
      "entropy": 6.270676231384277,
      "epoch": 1.7603403071325379,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004696936596891213,
      "loss": 6.0102,
      "mean_token_accuracy": 0.14633894860744476,
      "num_tokens": 35595501.0,
      "step": 16450
    },
    {
      "entropy": 6.3159901142120365,
      "epoch": 1.7608753812403017,
      "grad_norm": 1.25,
      "learning_rate": 0.0004696744925813605,
      "loss": 6.0228,
      "mean_token_accuracy": 0.14936779141426088,
      "num_tokens": 35606794.0,
      "step": 16455
    },
    {
      "entropy": 6.304333686828613,
      "epoch": 1.7614104553480656,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004696553198524011,
      "loss": 6.053,
      "mean_token_accuracy": 0.14798649549484252,
      "num_tokens": 35618732.0,
      "step": 16460
    },
    {
      "entropy": 6.303976058959961,
      "epoch": 1.7619455294558297,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004696361415027966,
      "loss": 5.9859,
      "mean_token_accuracy": 0.14916924089193345,
      "num_tokens": 35629802.0,
      "step": 16465
    },
    {
      "entropy": 6.270685243606567,
      "epoch": 1.7624806035635936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004696169575331009,
      "loss": 6.1074,
      "mean_token_accuracy": 0.14073484688997268,
      "num_tokens": 35640826.0,
      "step": 16470
    },
    {
      "entropy": 6.345275354385376,
      "epoch": 1.7630156776713575,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000469597767943868,
      "loss": 6.0925,
      "mean_token_accuracy": 0.14459955617785453,
      "num_tokens": 35651613.0,
      "step": 16475
    },
    {
      "entropy": 6.330532550811768,
      "epoch": 1.7635507517791216,
      "grad_norm": 1.125,
      "learning_rate": 0.0004695785727356517,
      "loss": 5.9506,
      "mean_token_accuracy": 0.14924613684415816,
      "num_tokens": 35662334.0,
      "step": 16480
    },
    {
      "entropy": 6.345506048202514,
      "epoch": 1.7640858258868852,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004695593719090066,
      "loss": 6.1189,
      "mean_token_accuracy": 0.14219709262251853,
      "num_tokens": 35673466.0,
      "step": 16485
    },
    {
      "entropy": 6.293130874633789,
      "epoch": 1.7646208999946493,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004695401654644869,
      "loss": 6.0596,
      "mean_token_accuracy": 0.1501174107193947,
      "num_tokens": 35684708.0,
      "step": 16490
    },
    {
      "entropy": 6.258680152893066,
      "epoch": 1.7651559741024132,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046952095340264727,
      "loss": 5.9076,
      "mean_token_accuracy": 0.1513049431145191,
      "num_tokens": 35694888.0,
      "step": 16495
    },
    {
      "entropy": 6.309222650527954,
      "epoch": 1.765691048210177,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046950173572404255,
      "loss": 5.9786,
      "mean_token_accuracy": 0.1439794510602951,
      "num_tokens": 35704844.0,
      "step": 16500
    },
    {
      "entropy": 6.264012432098388,
      "epoch": 1.7662261223179412,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004694825124292275,
      "loss": 6.0687,
      "mean_token_accuracy": 0.1483401656150818,
      "num_tokens": 35715468.0,
      "step": 16505
    },
    {
      "entropy": 6.323961448669434,
      "epoch": 1.7667611964257048,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004694632835187572,
      "loss": 6.1215,
      "mean_token_accuracy": 0.14364655464887618,
      "num_tokens": 35725542.0,
      "step": 16510
    },
    {
      "entropy": 6.353782224655151,
      "epoch": 1.767296270533469,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046944404899318685,
      "loss": 6.0179,
      "mean_token_accuracy": 0.14651695042848586,
      "num_tokens": 35737032.0,
      "step": 16515
    },
    {
      "entropy": 6.334203624725342,
      "epoch": 1.7678313446412328,
      "grad_norm": 1.109375,
      "learning_rate": 0.000469424808853072,
      "loss": 6.0304,
      "mean_token_accuracy": 0.15022097527980804,
      "num_tokens": 35747742.0,
      "step": 16520
    },
    {
      "entropy": 6.27889928817749,
      "epoch": 1.7683664187489967,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004694055630989681,
      "loss": 6.0178,
      "mean_token_accuracy": 0.147498482465744,
      "num_tokens": 35757795.0,
      "step": 16525
    },
    {
      "entropy": 6.291592264175415,
      "epoch": 1.7689014928567608,
      "grad_norm": 1.125,
      "learning_rate": 0.0004693863117314308,
      "loss": 6.0001,
      "mean_token_accuracy": 0.15652501583099365,
      "num_tokens": 35767962.0,
      "step": 16530
    },
    {
      "entropy": 6.28884801864624,
      "epoch": 1.7694365669645244,
      "grad_norm": 5.53125,
      "learning_rate": 0.0004693670547510161,
      "loss": 6.0922,
      "mean_token_accuracy": 0.1434783570468426,
      "num_tokens": 35778564.0,
      "step": 16535
    },
    {
      "entropy": 6.342926406860352,
      "epoch": 1.7699716410722885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046934779215828006,
      "loss": 6.034,
      "mean_token_accuracy": 0.15205717980861663,
      "num_tokens": 35788417.0,
      "step": 16540
    },
    {
      "entropy": 6.278888273239136,
      "epoch": 1.7705067151800524,
      "grad_norm": 1.125,
      "learning_rate": 0.0004693285239537788,
      "loss": 6.0626,
      "mean_token_accuracy": 0.14607202634215355,
      "num_tokens": 35799724.0,
      "step": 16545
    },
    {
      "entropy": 6.354902076721191,
      "epoch": 1.7710417892878163,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046930925013806873,
      "loss": 5.9838,
      "mean_token_accuracy": 0.15139147341251374,
      "num_tokens": 35810306.0,
      "step": 16550
    },
    {
      "entropy": 6.351597023010254,
      "epoch": 1.7715768633955804,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004692899707117064,
      "loss": 6.0278,
      "mean_token_accuracy": 0.15211757868528367,
      "num_tokens": 35821103.0,
      "step": 16555
    },
    {
      "entropy": 6.249758863449097,
      "epoch": 1.772111937503344,
      "grad_norm": 1.125,
      "learning_rate": 0.0004692706856752484,
      "loss": 6.0573,
      "mean_token_accuracy": 0.15147810876369477,
      "num_tokens": 35832271.0,
      "step": 16560
    },
    {
      "entropy": 6.296318912506104,
      "epoch": 1.7726470116111082,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046925139502925167,
      "loss": 6.067,
      "mean_token_accuracy": 0.1426429904997349,
      "num_tokens": 35843875.0,
      "step": 16565
    },
    {
      "entropy": 6.370083951950074,
      "epoch": 1.773182085718872,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004692320987742732,
      "loss": 6.0683,
      "mean_token_accuracy": 0.1485453501343727,
      "num_tokens": 35854085.0,
      "step": 16570
    },
    {
      "entropy": 6.366630601882934,
      "epoch": 1.773717159826636,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004692127969108702,
      "loss": 6.0976,
      "mean_token_accuracy": 0.1444030724465847,
      "num_tokens": 35864777.0,
      "step": 16575
    },
    {
      "entropy": 6.290893936157227,
      "epoch": 1.7742522339344,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004691934894396,
      "loss": 6.1081,
      "mean_token_accuracy": 0.14207665771245956,
      "num_tokens": 35876822.0,
      "step": 16580
    },
    {
      "entropy": 6.306673192977906,
      "epoch": 1.774787308042164,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046917417636102,
      "loss": 6.0154,
      "mean_token_accuracy": 0.14802766293287278,
      "num_tokens": 35886997.0,
      "step": 16585
    },
    {
      "entropy": 6.329456806182861,
      "epoch": 1.7753223821499278,
      "grad_norm": 1.34375,
      "learning_rate": 0.000469154857675688,
      "loss": 6.0827,
      "mean_token_accuracy": 0.14750705808401107,
      "num_tokens": 35897871.0,
      "step": 16590
    },
    {
      "entropy": 6.283900022506714,
      "epoch": 1.7758574562576916,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004691355333841617,
      "loss": 6.0657,
      "mean_token_accuracy": 0.1571588173508644,
      "num_tokens": 35908791.0,
      "step": 16595
    },
    {
      "entropy": 6.371531915664673,
      "epoch": 1.7763925303654555,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004691162034869993,
      "loss": 6.0921,
      "mean_token_accuracy": 0.14610493332147598,
      "num_tokens": 35919436.0,
      "step": 16600
    },
    {
      "entropy": 6.366421699523926,
      "epoch": 1.7769276044732196,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004690968679847586,
      "loss": 5.986,
      "mean_token_accuracy": 0.15042618066072463,
      "num_tokens": 35929851.0,
      "step": 16605
    },
    {
      "entropy": 6.292019176483154,
      "epoch": 1.7774626785809835,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004690775268779983,
      "loss": 6.0152,
      "mean_token_accuracy": 0.14911804497241973,
      "num_tokens": 35940271.0,
      "step": 16610
    },
    {
      "entropy": 6.306169700622559,
      "epoch": 1.7779977526887474,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004690581801672765,
      "loss": 6.129,
      "mean_token_accuracy": 0.1419995665550232,
      "num_tokens": 35951460.0,
      "step": 16615
    },
    {
      "entropy": 6.2940164566040036,
      "epoch": 1.7785328267965115,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046903882785315215,
      "loss": 6.0172,
      "mean_token_accuracy": 0.14840455800294877,
      "num_tokens": 35961530.0,
      "step": 16620
    },
    {
      "entropy": 6.27038803100586,
      "epoch": 1.7790679009042751,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004690194699361839,
      "loss": 5.9864,
      "mean_token_accuracy": 0.15904172658920288,
      "num_tokens": 35971085.0,
      "step": 16625
    },
    {
      "entropy": 6.2459009170532225,
      "epoch": 1.7796029750120392,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046900010641693073,
      "loss": 6.0577,
      "mean_token_accuracy": 0.1438824824988842,
      "num_tokens": 35981393.0,
      "step": 16630
    },
    {
      "entropy": 6.290297508239746,
      "epoch": 1.7801380491198031,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004689807372959517,
      "loss": 6.0001,
      "mean_token_accuracy": 0.15173507034778594,
      "num_tokens": 35992847.0,
      "step": 16635
    },
    {
      "entropy": 6.291001272201538,
      "epoch": 1.780673123227567,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046896136257380615,
      "loss": 6.0507,
      "mean_token_accuracy": 0.1454372376203537,
      "num_tokens": 36002991.0,
      "step": 16640
    },
    {
      "entropy": 6.305175161361694,
      "epoch": 1.781208197335331,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046894198225105356,
      "loss": 6.079,
      "mean_token_accuracy": 0.15207182765007018,
      "num_tokens": 36015140.0,
      "step": 16645
    },
    {
      "entropy": 6.295879888534546,
      "epoch": 1.7817432714430947,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004689225963282534,
      "loss": 5.9683,
      "mean_token_accuracy": 0.15212046056985856,
      "num_tokens": 36026014.0,
      "step": 16650
    },
    {
      "entropy": 6.389534330368042,
      "epoch": 1.7822783455508588,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004689032048059656,
      "loss": 6.1406,
      "mean_token_accuracy": 0.1425458773970604,
      "num_tokens": 36035926.0,
      "step": 16655
    },
    {
      "entropy": 6.314208459854126,
      "epoch": 1.7828134196586227,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046888380768475,
      "loss": 6.0491,
      "mean_token_accuracy": 0.1471663936972618,
      "num_tokens": 36046406.0,
      "step": 16660
    },
    {
      "entropy": 6.2664836883544925,
      "epoch": 1.7833484937663866,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004688644049651667,
      "loss": 5.9728,
      "mean_token_accuracy": 0.15177134424448013,
      "num_tokens": 36056440.0,
      "step": 16665
    },
    {
      "entropy": 6.246055603027344,
      "epoch": 1.7838835678741507,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046884499664777586,
      "loss": 6.0736,
      "mean_token_accuracy": 0.14520161002874374,
      "num_tokens": 36067244.0,
      "step": 16670
    },
    {
      "entropy": 6.282402515411377,
      "epoch": 1.7844186419819144,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004688255827331381,
      "loss": 5.9793,
      "mean_token_accuracy": 0.15468614995479585,
      "num_tokens": 36078110.0,
      "step": 16675
    },
    {
      "entropy": 6.356593894958496,
      "epoch": 1.7849537160896785,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004688061632218137,
      "loss": 6.059,
      "mean_token_accuracy": 0.14635979533195495,
      "num_tokens": 36088805.0,
      "step": 16680
    },
    {
      "entropy": 6.311991500854492,
      "epoch": 1.7854887901974423,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046878673811436375,
      "loss": 6.0371,
      "mean_token_accuracy": 0.14942564815282822,
      "num_tokens": 36098619.0,
      "step": 16685
    },
    {
      "entropy": 6.341496133804322,
      "epoch": 1.7860238643052062,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004687673074113489,
      "loss": 6.1013,
      "mean_token_accuracy": 0.14567922577261924,
      "num_tokens": 36110604.0,
      "step": 16690
    },
    {
      "entropy": 6.421867227554321,
      "epoch": 1.7865589384129703,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004687478711133303,
      "loss": 6.0765,
      "mean_token_accuracy": 0.14369807839393617,
      "num_tokens": 36121669.0,
      "step": 16695
    },
    {
      "entropy": 6.185075664520264,
      "epoch": 1.787094012520734,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004687284292208691,
      "loss": 5.8795,
      "mean_token_accuracy": 0.16030368953943253,
      "num_tokens": 36132165.0,
      "step": 16700
    },
    {
      "entropy": 6.2538793087005615,
      "epoch": 1.787629086628498,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046870898173452676,
      "loss": 6.0107,
      "mean_token_accuracy": 0.14922809451818467,
      "num_tokens": 36142830.0,
      "step": 16705
    },
    {
      "entropy": 6.278341293334961,
      "epoch": 1.788164160736262,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004686895286548648,
      "loss": 6.0059,
      "mean_token_accuracy": 0.15039270967245102,
      "num_tokens": 36153812.0,
      "step": 16710
    },
    {
      "entropy": 6.277557373046875,
      "epoch": 1.7886992348440258,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046867006998244494,
      "loss": 6.0393,
      "mean_token_accuracy": 0.14990729838609695,
      "num_tokens": 36163599.0,
      "step": 16715
    },
    {
      "entropy": 6.27621021270752,
      "epoch": 1.78923430895179,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046865060571782896,
      "loss": 6.0726,
      "mean_token_accuracy": 0.14328675121068954,
      "num_tokens": 36173815.0,
      "step": 16720
    },
    {
      "entropy": 6.418244361877441,
      "epoch": 1.7897693830595538,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046863113586157894,
      "loss": 6.1442,
      "mean_token_accuracy": 0.14303762912750245,
      "num_tokens": 36186131.0,
      "step": 16725
    },
    {
      "entropy": 6.341207265853882,
      "epoch": 1.7903044571673177,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004686116604142572,
      "loss": 6.0096,
      "mean_token_accuracy": 0.15259127467870712,
      "num_tokens": 36196084.0,
      "step": 16730
    },
    {
      "entropy": 6.230683517456055,
      "epoch": 1.7908395312750816,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004685921793764259,
      "loss": 5.9712,
      "mean_token_accuracy": 0.15643182322382926,
      "num_tokens": 36206948.0,
      "step": 16735
    },
    {
      "entropy": 6.336044788360596,
      "epoch": 1.7913746053828454,
      "grad_norm": 1.25,
      "learning_rate": 0.0004685726927486476,
      "loss": 6.1122,
      "mean_token_accuracy": 0.1414343848824501,
      "num_tokens": 36217202.0,
      "step": 16740
    },
    {
      "entropy": 6.3907835483551025,
      "epoch": 1.7919096794906095,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046855320053148505,
      "loss": 6.1386,
      "mean_token_accuracy": 0.13865508735179902,
      "num_tokens": 36227914.0,
      "step": 16745
    },
    {
      "entropy": 6.393190002441406,
      "epoch": 1.7924447535983734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000468533702725501,
      "loss": 6.0805,
      "mean_token_accuracy": 0.14359190911054612,
      "num_tokens": 36239344.0,
      "step": 16750
    },
    {
      "entropy": 6.338386011123657,
      "epoch": 1.7929798277061373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046851419933125843,
      "loss": 6.1054,
      "mean_token_accuracy": 0.1462898939847946,
      "num_tokens": 36250438.0,
      "step": 16755
    },
    {
      "entropy": 6.330519580841065,
      "epoch": 1.7935149018139014,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004684946903493206,
      "loss": 6.0512,
      "mean_token_accuracy": 0.1429305411875248,
      "num_tokens": 36260545.0,
      "step": 16760
    },
    {
      "entropy": 6.195946788787841,
      "epoch": 1.794049975921665,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004684751757802508,
      "loss": 5.967,
      "mean_token_accuracy": 0.14726777225732804,
      "num_tokens": 36272577.0,
      "step": 16765
    },
    {
      "entropy": 6.302812242507935,
      "epoch": 1.7945850500294291,
      "grad_norm": 1.0,
      "learning_rate": 0.00046845565562461243,
      "loss": 6.0129,
      "mean_token_accuracy": 0.1490445002913475,
      "num_tokens": 36284161.0,
      "step": 16770
    },
    {
      "entropy": 6.275001287460327,
      "epoch": 1.795120124137193,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046843612988296926,
      "loss": 6.018,
      "mean_token_accuracy": 0.14679028540849687,
      "num_tokens": 36294910.0,
      "step": 16775
    },
    {
      "entropy": 6.322486305236817,
      "epoch": 1.795655198244957,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004684165985558849,
      "loss": 5.9847,
      "mean_token_accuracy": 0.15131528526544571,
      "num_tokens": 36305388.0,
      "step": 16780
    },
    {
      "entropy": 6.302372455596924,
      "epoch": 1.796190272352721,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004683970616439236,
      "loss": 6.071,
      "mean_token_accuracy": 0.15472257286310195,
      "num_tokens": 36316696.0,
      "step": 16785
    },
    {
      "entropy": 6.260030078887939,
      "epoch": 1.7967253464604847,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004683775191476492,
      "loss": 5.9698,
      "mean_token_accuracy": 0.15940692126750947,
      "num_tokens": 36327246.0,
      "step": 16790
    },
    {
      "entropy": 6.3624152660369875,
      "epoch": 1.7972604205682488,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004683579710676262,
      "loss": 6.145,
      "mean_token_accuracy": 0.13815257400274278,
      "num_tokens": 36338089.0,
      "step": 16795
    },
    {
      "entropy": 6.328536891937256,
      "epoch": 1.7977954946760126,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046833841740441886,
      "loss": 6.0553,
      "mean_token_accuracy": 0.14595745727419854,
      "num_tokens": 36350285.0,
      "step": 16800
    },
    {
      "entropy": 6.33431339263916,
      "epoch": 1.7983305687837765,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004683188581585919,
      "loss": 6.0398,
      "mean_token_accuracy": 0.1432052507996559,
      "num_tokens": 36360431.0,
      "step": 16805
    },
    {
      "entropy": 6.294195508956909,
      "epoch": 1.7988656428915406,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004682992933307101,
      "loss": 5.9927,
      "mean_token_accuracy": 0.15207217633724213,
      "num_tokens": 36371257.0,
      "step": 16810
    },
    {
      "entropy": 6.211703157424926,
      "epoch": 1.7994007169993043,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046827972292133835,
      "loss": 5.9734,
      "mean_token_accuracy": 0.1521460846066475,
      "num_tokens": 36382613.0,
      "step": 16815
    },
    {
      "entropy": 6.361332702636719,
      "epoch": 1.7999357911070684,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004682601469310418,
      "loss": 6.05,
      "mean_token_accuracy": 0.1433391347527504,
      "num_tokens": 36393799.0,
      "step": 16820
    },
    {
      "entropy": 6.351719379425049,
      "epoch": 1.8004708652148322,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046824056536038565,
      "loss": 6.0579,
      "mean_token_accuracy": 0.14433461800217628,
      "num_tokens": 36403914.0,
      "step": 16825
    },
    {
      "entropy": 6.201854610443116,
      "epoch": 1.8010059393225961,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004682209782099353,
      "loss": 6.0075,
      "mean_token_accuracy": 0.14714679569005967,
      "num_tokens": 36415022.0,
      "step": 16830
    },
    {
      "entropy": 6.241609716415406,
      "epoch": 1.8015410134303602,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046820138548025634,
      "loss": 5.967,
      "mean_token_accuracy": 0.15559637397527695,
      "num_tokens": 36426155.0,
      "step": 16835
    },
    {
      "entropy": 6.291421794891358,
      "epoch": 1.8020760875381239,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046818178717191464,
      "loss": 6.0285,
      "mean_token_accuracy": 0.1441032961010933,
      "num_tokens": 36437345.0,
      "step": 16840
    },
    {
      "entropy": 6.30050220489502,
      "epoch": 1.802611161645888,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004681621832854759,
      "loss": 6.0634,
      "mean_token_accuracy": 0.1561509773135185,
      "num_tokens": 36448634.0,
      "step": 16845
    },
    {
      "entropy": 6.3130943775177,
      "epoch": 1.8031462357536518,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046814257382150627,
      "loss": 6.0066,
      "mean_token_accuracy": 0.15918610990047455,
      "num_tokens": 36458870.0,
      "step": 16850
    },
    {
      "entropy": 6.318803787231445,
      "epoch": 1.8036813098614157,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046812295878057193,
      "loss": 6.1567,
      "mean_token_accuracy": 0.14472266063094139,
      "num_tokens": 36469865.0,
      "step": 16855
    },
    {
      "entropy": 6.306317615509033,
      "epoch": 1.8042163839691798,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004681033381632393,
      "loss": 6.0354,
      "mean_token_accuracy": 0.14690264016389848,
      "num_tokens": 36480502.0,
      "step": 16860
    },
    {
      "entropy": 6.37965030670166,
      "epoch": 1.8047514580769437,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004680837119700749,
      "loss": 6.0516,
      "mean_token_accuracy": 0.1477011561393738,
      "num_tokens": 36492356.0,
      "step": 16865
    },
    {
      "entropy": 6.19101185798645,
      "epoch": 1.8052865321847076,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046806408020164553,
      "loss": 6.0318,
      "mean_token_accuracy": 0.1480714909732342,
      "num_tokens": 36504659.0,
      "step": 16870
    },
    {
      "entropy": 6.292754888534546,
      "epoch": 1.8058216062924715,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046804444285851796,
      "loss": 6.0225,
      "mean_token_accuracy": 0.14760567843914033,
      "num_tokens": 36516042.0,
      "step": 16875
    },
    {
      "entropy": 6.377509260177613,
      "epoch": 1.8063566804002353,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046802479994125916,
      "loss": 6.1281,
      "mean_token_accuracy": 0.14353615790605545,
      "num_tokens": 36527417.0,
      "step": 16880
    },
    {
      "entropy": 6.257154893875122,
      "epoch": 1.8068917545079994,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004680051514504364,
      "loss": 6.0296,
      "mean_token_accuracy": 0.15690145939588546,
      "num_tokens": 36538182.0,
      "step": 16885
    },
    {
      "entropy": 6.309271192550659,
      "epoch": 1.8074268286157633,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004679854973866171,
      "loss": 6.1064,
      "mean_token_accuracy": 0.14593173712491989,
      "num_tokens": 36549835.0,
      "step": 16890
    },
    {
      "entropy": 6.302314376831054,
      "epoch": 1.8079619027235272,
      "grad_norm": 1.75,
      "learning_rate": 0.00046796583775036857,
      "loss": 6.0186,
      "mean_token_accuracy": 0.1555558517575264,
      "num_tokens": 36560238.0,
      "step": 16895
    },
    {
      "entropy": 6.300615072250366,
      "epoch": 1.8084969768312913,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046794617254225865,
      "loss": 6.0211,
      "mean_token_accuracy": 0.14985527992248535,
      "num_tokens": 36571267.0,
      "step": 16900
    },
    {
      "entropy": 6.305903196334839,
      "epoch": 1.809032050939055,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004679265017628551,
      "loss": 6.0954,
      "mean_token_accuracy": 0.14890325218439102,
      "num_tokens": 36582174.0,
      "step": 16905
    },
    {
      "entropy": 6.373585891723633,
      "epoch": 1.809567125046819,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004679068254127259,
      "loss": 6.0608,
      "mean_token_accuracy": 0.1488943099975586,
      "num_tokens": 36593054.0,
      "step": 16910
    },
    {
      "entropy": 6.360124588012695,
      "epoch": 1.810102199154583,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046788714349243923,
      "loss": 6.0608,
      "mean_token_accuracy": 0.14901149719953538,
      "num_tokens": 36603962.0,
      "step": 16915
    },
    {
      "entropy": 6.323988342285157,
      "epoch": 1.8106372732623468,
      "grad_norm": 1.125,
      "learning_rate": 0.0004678674560025634,
      "loss": 6.028,
      "mean_token_accuracy": 0.15194036662578583,
      "num_tokens": 36616003.0,
      "step": 16920
    },
    {
      "entropy": 6.3715568542480465,
      "epoch": 1.811172347370111,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004678477629436669,
      "loss": 6.1304,
      "mean_token_accuracy": 0.13937999531626702,
      "num_tokens": 36627189.0,
      "step": 16925
    },
    {
      "entropy": 6.349302434921265,
      "epoch": 1.8117074214778746,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004678280643163184,
      "loss": 6.1399,
      "mean_token_accuracy": 0.14371826350688935,
      "num_tokens": 36637756.0,
      "step": 16930
    },
    {
      "entropy": 6.335540246963501,
      "epoch": 1.8122424955856387,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004678083601210865,
      "loss": 6.0782,
      "mean_token_accuracy": 0.14996321499347687,
      "num_tokens": 36648946.0,
      "step": 16935
    },
    {
      "entropy": 6.376877737045288,
      "epoch": 1.8127775696934025,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004677886503585404,
      "loss": 6.0879,
      "mean_token_accuracy": 0.14865182489156722,
      "num_tokens": 36660184.0,
      "step": 16940
    },
    {
      "entropy": 6.3987939834594725,
      "epoch": 1.8133126438011664,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046776893502924904,
      "loss": 6.1418,
      "mean_token_accuracy": 0.1420721024274826,
      "num_tokens": 36671346.0,
      "step": 16945
    },
    {
      "entropy": 6.364566707611084,
      "epoch": 1.8138477179089305,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004677492141337818,
      "loss": 6.0365,
      "mean_token_accuracy": 0.1474158525466919,
      "num_tokens": 36682917.0,
      "step": 16950
    },
    {
      "entropy": 6.297361230850219,
      "epoch": 1.8143827920166942,
      "grad_norm": 1.25,
      "learning_rate": 0.0004677294876727081,
      "loss": 6.1223,
      "mean_token_accuracy": 0.13392944782972335,
      "num_tokens": 36693577.0,
      "step": 16955
    },
    {
      "entropy": 6.383313131332398,
      "epoch": 1.8149178661244583,
      "grad_norm": 1.25,
      "learning_rate": 0.0004677097556465975,
      "loss": 6.0676,
      "mean_token_accuracy": 0.14632795229554177,
      "num_tokens": 36704915.0,
      "step": 16960
    },
    {
      "entropy": 6.384161138534546,
      "epoch": 1.8154529402322221,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046769001805601983,
      "loss": 6.1637,
      "mean_token_accuracy": 0.13672029599547386,
      "num_tokens": 36716086.0,
      "step": 16965
    },
    {
      "entropy": 6.3045008182525635,
      "epoch": 1.815988014339986,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046767027490154494,
      "loss": 6.023,
      "mean_token_accuracy": 0.14078457802534103,
      "num_tokens": 36726110.0,
      "step": 16970
    },
    {
      "entropy": 6.368986797332764,
      "epoch": 1.8165230884477501,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046765052618374296,
      "loss": 5.9935,
      "mean_token_accuracy": 0.15383180975914001,
      "num_tokens": 36735846.0,
      "step": 16975
    },
    {
      "entropy": 6.292939615249634,
      "epoch": 1.8170581625555138,
      "grad_norm": 1.21875,
      "learning_rate": 0.000467630771903184,
      "loss": 6.0777,
      "mean_token_accuracy": 0.15130159854888917,
      "num_tokens": 36746851.0,
      "step": 16980
    },
    {
      "entropy": 6.327389287948608,
      "epoch": 1.8175932366632779,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004676110120604387,
      "loss": 6.051,
      "mean_token_accuracy": 0.14787747114896774,
      "num_tokens": 36758032.0,
      "step": 16985
    },
    {
      "entropy": 6.354951429367065,
      "epoch": 1.8181283107710418,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004675912466560775,
      "loss": 6.0729,
      "mean_token_accuracy": 0.14845812171697617,
      "num_tokens": 36768939.0,
      "step": 16990
    },
    {
      "entropy": 6.339537286758423,
      "epoch": 1.8186633848788056,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046757147569067106,
      "loss": 6.1136,
      "mean_token_accuracy": 0.14232637509703636,
      "num_tokens": 36780338.0,
      "step": 16995
    },
    {
      "entropy": 6.312480354309082,
      "epoch": 1.8191984589865697,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004675516991647903,
      "loss": 5.9781,
      "mean_token_accuracy": 0.1421303890645504,
      "num_tokens": 36791384.0,
      "step": 17000
    },
    {
      "entropy": 6.2985029220581055,
      "epoch": 1.8197335330943336,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046753191707900636,
      "loss": 6.0159,
      "mean_token_accuracy": 0.14502909481525422,
      "num_tokens": 36802460.0,
      "step": 17005
    },
    {
      "entropy": 6.345726060867309,
      "epoch": 1.8202686072020975,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046751212943389026,
      "loss": 6.1001,
      "mean_token_accuracy": 0.14315566122531892,
      "num_tokens": 36813599.0,
      "step": 17010
    },
    {
      "entropy": 6.342699813842773,
      "epoch": 1.8208036813098614,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004674923362300135,
      "loss": 5.9877,
      "mean_token_accuracy": 0.14348411858081817,
      "num_tokens": 36823100.0,
      "step": 17015
    },
    {
      "entropy": 6.321341848373413,
      "epoch": 1.8213387554176252,
      "grad_norm": 1.375,
      "learning_rate": 0.00046747253746794767,
      "loss": 6.1285,
      "mean_token_accuracy": 0.14412996768951417,
      "num_tokens": 36834245.0,
      "step": 17020
    },
    {
      "entropy": 6.35323748588562,
      "epoch": 1.8218738295253893,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046745273314826425,
      "loss": 5.9868,
      "mean_token_accuracy": 0.15694578289985656,
      "num_tokens": 36844126.0,
      "step": 17025
    },
    {
      "entropy": 6.31890287399292,
      "epoch": 1.8224089036331532,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004674329232715353,
      "loss": 5.9833,
      "mean_token_accuracy": 0.15810038298368453,
      "num_tokens": 36855451.0,
      "step": 17030
    },
    {
      "entropy": 6.329202508926391,
      "epoch": 1.822943977740917,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004674131078383327,
      "loss": 6.0649,
      "mean_token_accuracy": 0.1482343539595604,
      "num_tokens": 36864994.0,
      "step": 17035
    },
    {
      "entropy": 6.276311874389648,
      "epoch": 1.8234790518486812,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004673932868492286,
      "loss": 6.0069,
      "mean_token_accuracy": 0.1463366076350212,
      "num_tokens": 36875131.0,
      "step": 17040
    },
    {
      "entropy": 6.3454710960388185,
      "epoch": 1.8240141259564449,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046737346030479546,
      "loss": 6.0698,
      "mean_token_accuracy": 0.14504208117723466,
      "num_tokens": 36884789.0,
      "step": 17045
    },
    {
      "entropy": 6.312895107269287,
      "epoch": 1.824549200064209,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046735362820560564,
      "loss": 5.9895,
      "mean_token_accuracy": 0.14372953325510024,
      "num_tokens": 36895199.0,
      "step": 17050
    },
    {
      "entropy": 6.333379030227661,
      "epoch": 1.8250842741719728,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004673337905522318,
      "loss": 6.0933,
      "mean_token_accuracy": 0.14372849613428115,
      "num_tokens": 36906495.0,
      "step": 17055
    },
    {
      "entropy": 6.28471155166626,
      "epoch": 1.8256193482797367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046731394734524686,
      "loss": 6.0481,
      "mean_token_accuracy": 0.14240381196141244,
      "num_tokens": 36916907.0,
      "step": 17060
    },
    {
      "entropy": 6.283461856842041,
      "epoch": 1.8261544223875008,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046729409858522363,
      "loss": 6.0104,
      "mean_token_accuracy": 0.14673244804143906,
      "num_tokens": 36927872.0,
      "step": 17065
    },
    {
      "entropy": 6.30297737121582,
      "epoch": 1.8266894964952645,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004672742442727354,
      "loss": 6.0513,
      "mean_token_accuracy": 0.14671421200037002,
      "num_tokens": 36938575.0,
      "step": 17070
    },
    {
      "entropy": 6.298050165176392,
      "epoch": 1.8272245706030286,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046725438440835536,
      "loss": 5.9204,
      "mean_token_accuracy": 0.1612280011177063,
      "num_tokens": 36949179.0,
      "step": 17075
    },
    {
      "entropy": 6.24626784324646,
      "epoch": 1.8277596447107924,
      "grad_norm": 1.25,
      "learning_rate": 0.000467234518992657,
      "loss": 5.9793,
      "mean_token_accuracy": 0.15983165800571442,
      "num_tokens": 36959164.0,
      "step": 17080
    },
    {
      "entropy": 6.324422311782837,
      "epoch": 1.8282947188185563,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046721464802621383,
      "loss": 6.1243,
      "mean_token_accuracy": 0.14838092029094696,
      "num_tokens": 36969091.0,
      "step": 17085
    },
    {
      "entropy": 6.347911024093628,
      "epoch": 1.8288297929263204,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046719477150959984,
      "loss": 6.0666,
      "mean_token_accuracy": 0.14446090236306192,
      "num_tokens": 36980481.0,
      "step": 17090
    },
    {
      "entropy": 6.3484354496002195,
      "epoch": 1.829364867034084,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046717488944338873,
      "loss": 6.042,
      "mean_token_accuracy": 0.14804359078407286,
      "num_tokens": 36991349.0,
      "step": 17095
    },
    {
      "entropy": 6.301275253295898,
      "epoch": 1.8298999411418482,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046715500182815473,
      "loss": 6.1188,
      "mean_token_accuracy": 0.1486465096473694,
      "num_tokens": 37002803.0,
      "step": 17100
    },
    {
      "entropy": 6.352608585357666,
      "epoch": 1.830435015249612,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046713510866447204,
      "loss": 6.0709,
      "mean_token_accuracy": 0.1492726743221283,
      "num_tokens": 37013578.0,
      "step": 17105
    },
    {
      "entropy": 6.284890127182007,
      "epoch": 1.830970089357376,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004671152099529152,
      "loss": 5.9733,
      "mean_token_accuracy": 0.1528842031955719,
      "num_tokens": 37024012.0,
      "step": 17110
    },
    {
      "entropy": 6.245310592651367,
      "epoch": 1.83150516346514,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046709530569405855,
      "loss": 5.9645,
      "mean_token_accuracy": 0.15010854303836824,
      "num_tokens": 37034149.0,
      "step": 17115
    },
    {
      "entropy": 6.260279750823974,
      "epoch": 1.8320402375729037,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004670753958884769,
      "loss": 5.9683,
      "mean_token_accuracy": 0.15384056568145751,
      "num_tokens": 37045669.0,
      "step": 17120
    },
    {
      "entropy": 6.30774450302124,
      "epoch": 1.8325753116806678,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046705548053674524,
      "loss": 6.0626,
      "mean_token_accuracy": 0.14800645411014557,
      "num_tokens": 37056611.0,
      "step": 17125
    },
    {
      "entropy": 6.288930177688599,
      "epoch": 1.8331103857884317,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046703555963943855,
      "loss": 6.0349,
      "mean_token_accuracy": 0.14536865949630737,
      "num_tokens": 37067387.0,
      "step": 17130
    },
    {
      "entropy": 6.436623048782349,
      "epoch": 1.8336454598961955,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046701563319713214,
      "loss": 6.1147,
      "mean_token_accuracy": 0.1425323911011219,
      "num_tokens": 37078151.0,
      "step": 17135
    },
    {
      "entropy": 6.367936801910401,
      "epoch": 1.8341805340039596,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046699570121040126,
      "loss": 6.0866,
      "mean_token_accuracy": 0.14876790940761567,
      "num_tokens": 37088219.0,
      "step": 17140
    },
    {
      "entropy": 6.225270462036133,
      "epoch": 1.8347156081117235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004669757636798215,
      "loss": 5.9459,
      "mean_token_accuracy": 0.15095793455839157,
      "num_tokens": 37098769.0,
      "step": 17145
    },
    {
      "entropy": 6.341804647445679,
      "epoch": 1.8352506822194874,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004669558206059684,
      "loss": 6.1086,
      "mean_token_accuracy": 0.13835907503962516,
      "num_tokens": 37110716.0,
      "step": 17150
    },
    {
      "entropy": 6.39202995300293,
      "epoch": 1.8357857563272513,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004669358719894181,
      "loss": 6.0076,
      "mean_token_accuracy": 0.14997104108333587,
      "num_tokens": 37122785.0,
      "step": 17155
    },
    {
      "entropy": 6.2731812477111815,
      "epoch": 1.8363208304350152,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004669159178307465,
      "loss": 6.0181,
      "mean_token_accuracy": 0.1491830587387085,
      "num_tokens": 37133308.0,
      "step": 17160
    },
    {
      "entropy": 6.193302583694458,
      "epoch": 1.8368559045427792,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004668959581305296,
      "loss": 5.9798,
      "mean_token_accuracy": 0.15169450342655183,
      "num_tokens": 37143916.0,
      "step": 17165
    },
    {
      "entropy": 6.297737264633179,
      "epoch": 1.8373909786505431,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046687599288934394,
      "loss": 5.9833,
      "mean_token_accuracy": 0.14813824295997619,
      "num_tokens": 37154335.0,
      "step": 17170
    },
    {
      "entropy": 6.332500553131103,
      "epoch": 1.837926052758307,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000466856022107766,
      "loss": 6.0432,
      "mean_token_accuracy": 0.14909263998270034,
      "num_tokens": 37164116.0,
      "step": 17175
    },
    {
      "entropy": 6.319372463226318,
      "epoch": 1.838461126866071,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046683604578637235,
      "loss": 6.114,
      "mean_token_accuracy": 0.14539389610290526,
      "num_tokens": 37175407.0,
      "step": 17180
    },
    {
      "entropy": 6.257063674926758,
      "epoch": 1.8389962009738348,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046681606392573975,
      "loss": 5.9861,
      "mean_token_accuracy": 0.14948231726884842,
      "num_tokens": 37185339.0,
      "step": 17185
    },
    {
      "entropy": 6.367322492599487,
      "epoch": 1.8395312750815989,
      "grad_norm": 2.25,
      "learning_rate": 0.0004667960765264454,
      "loss": 6.0869,
      "mean_token_accuracy": 0.14474356025457383,
      "num_tokens": 37195145.0,
      "step": 17190
    },
    {
      "entropy": 6.34392409324646,
      "epoch": 1.8400663491893627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004667760835890662,
      "loss": 6.1112,
      "mean_token_accuracy": 0.14479876607656478,
      "num_tokens": 37206454.0,
      "step": 17195
    },
    {
      "entropy": 6.279236268997193,
      "epoch": 1.8406014232971266,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046675608511417957,
      "loss": 5.988,
      "mean_token_accuracy": 0.15338556617498397,
      "num_tokens": 37216646.0,
      "step": 17200
    },
    {
      "entropy": 6.262100458145142,
      "epoch": 1.8411364974048907,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046673608110236293,
      "loss": 5.9785,
      "mean_token_accuracy": 0.15597550421953202,
      "num_tokens": 37227368.0,
      "step": 17205
    },
    {
      "entropy": 6.345663070678711,
      "epoch": 1.8416715715126544,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004667160715541939,
      "loss": 6.0858,
      "mean_token_accuracy": 0.14623727425932884,
      "num_tokens": 37238150.0,
      "step": 17210
    },
    {
      "entropy": 6.289287853240967,
      "epoch": 1.8422066456204185,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046669605647025014,
      "loss": 6.0359,
      "mean_token_accuracy": 0.1459670767188072,
      "num_tokens": 37249534.0,
      "step": 17215
    },
    {
      "entropy": 6.405678415298462,
      "epoch": 1.8427417197281823,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004666760358511098,
      "loss": 6.1053,
      "mean_token_accuracy": 0.15107614770531655,
      "num_tokens": 37260633.0,
      "step": 17220
    },
    {
      "entropy": 6.382426738739014,
      "epoch": 1.8432767938359462,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046665600969735086,
      "loss": 6.0677,
      "mean_token_accuracy": 0.14446606785058974,
      "num_tokens": 37271978.0,
      "step": 17225
    },
    {
      "entropy": 6.226320266723633,
      "epoch": 1.8438118679437103,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004666359780095515,
      "loss": 5.9195,
      "mean_token_accuracy": 0.14825134351849556,
      "num_tokens": 37284130.0,
      "step": 17230
    },
    {
      "entropy": 6.337172174453736,
      "epoch": 1.844346942051474,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004666159407882902,
      "loss": 6.0981,
      "mean_token_accuracy": 0.1414434254169464,
      "num_tokens": 37295648.0,
      "step": 17235
    },
    {
      "entropy": 6.357077693939209,
      "epoch": 1.844882016159238,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004665958980341456,
      "loss": 6.0851,
      "mean_token_accuracy": 0.14624134153127671,
      "num_tokens": 37306715.0,
      "step": 17240
    },
    {
      "entropy": 6.301657867431641,
      "epoch": 1.845417090267002,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046657584974769636,
      "loss": 6.0962,
      "mean_token_accuracy": 0.1478027120232582,
      "num_tokens": 37317605.0,
      "step": 17245
    },
    {
      "entropy": 6.284667825698852,
      "epoch": 1.8459521643747658,
      "grad_norm": 1.125,
      "learning_rate": 0.0004665557959295213,
      "loss": 6.0254,
      "mean_token_accuracy": 0.1474469408392906,
      "num_tokens": 37328898.0,
      "step": 17250
    },
    {
      "entropy": 6.282193851470947,
      "epoch": 1.84648723848253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046653573658019963,
      "loss": 5.9806,
      "mean_token_accuracy": 0.1518290489912033,
      "num_tokens": 37340108.0,
      "step": 17255
    },
    {
      "entropy": 6.343241262435913,
      "epoch": 1.8470223125902936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046651567170031053,
      "loss": 6.1146,
      "mean_token_accuracy": 0.14469715654850007,
      "num_tokens": 37351773.0,
      "step": 17260
    },
    {
      "entropy": 6.302702903747559,
      "epoch": 1.8475573866980577,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004664956012904332,
      "loss": 6.0215,
      "mean_token_accuracy": 0.15040826201438903,
      "num_tokens": 37362748.0,
      "step": 17265
    },
    {
      "entropy": 6.281106805801391,
      "epoch": 1.8480924608058216,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004664755253511474,
      "loss": 6.1044,
      "mean_token_accuracy": 0.1479063995182514,
      "num_tokens": 37373487.0,
      "step": 17270
    },
    {
      "entropy": 6.32073130607605,
      "epoch": 1.8486275349135854,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004664554438830326,
      "loss": 5.9573,
      "mean_token_accuracy": 0.15052159279584884,
      "num_tokens": 37383302.0,
      "step": 17275
    },
    {
      "entropy": 6.328805303573608,
      "epoch": 1.8491626090213495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004664353568866689,
      "loss": 6.0071,
      "mean_token_accuracy": 0.14974130243062972,
      "num_tokens": 37393085.0,
      "step": 17280
    },
    {
      "entropy": 6.364643383026123,
      "epoch": 1.8496976831291134,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000466415264362636,
      "loss": 6.0822,
      "mean_token_accuracy": 0.14666857421398163,
      "num_tokens": 37403774.0,
      "step": 17285
    },
    {
      "entropy": 6.326553583145142,
      "epoch": 1.8502327572368773,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046639516631151435,
      "loss": 6.0348,
      "mean_token_accuracy": 0.15452749654650688,
      "num_tokens": 37414356.0,
      "step": 17290
    },
    {
      "entropy": 6.274716424942016,
      "epoch": 1.8507678313446412,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046637506273388416,
      "loss": 6.0883,
      "mean_token_accuracy": 0.14287425354123115,
      "num_tokens": 37426381.0,
      "step": 17295
    },
    {
      "entropy": 6.3318507194519045,
      "epoch": 1.851302905452405,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046635495363032595,
      "loss": 6.0236,
      "mean_token_accuracy": 0.1496220611035824,
      "num_tokens": 37436259.0,
      "step": 17300
    },
    {
      "entropy": 6.301131868362427,
      "epoch": 1.8518379795601692,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046633483900142026,
      "loss": 6.0062,
      "mean_token_accuracy": 0.1530321329832077,
      "num_tokens": 37447498.0,
      "step": 17305
    },
    {
      "entropy": 6.2978685855865475,
      "epoch": 1.852373053667933,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000466314718847748,
      "loss": 6.0062,
      "mean_token_accuracy": 0.14875539690256118,
      "num_tokens": 37457323.0,
      "step": 17310
    },
    {
      "entropy": 6.314085197448731,
      "epoch": 1.852908127775697,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004662945931698901,
      "loss": 6.0282,
      "mean_token_accuracy": 0.14514394402503966,
      "num_tokens": 37468121.0,
      "step": 17315
    },
    {
      "entropy": 6.342468738555908,
      "epoch": 1.853443201883461,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046627446196842784,
      "loss": 6.0669,
      "mean_token_accuracy": 0.1443673774600029,
      "num_tokens": 37479205.0,
      "step": 17320
    },
    {
      "entropy": 6.369778728485107,
      "epoch": 1.8539782759912247,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004662543252439422,
      "loss": 6.1653,
      "mean_token_accuracy": 0.1410879135131836,
      "num_tokens": 37491015.0,
      "step": 17325
    },
    {
      "entropy": 6.376108646392822,
      "epoch": 1.8545133500989888,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046623418299701494,
      "loss": 6.0914,
      "mean_token_accuracy": 0.1438198134303093,
      "num_tokens": 37503078.0,
      "step": 17330
    },
    {
      "entropy": 6.36142930984497,
      "epoch": 1.8550484242067526,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004662140352282274,
      "loss": 6.0569,
      "mean_token_accuracy": 0.15251701176166535,
      "num_tokens": 37514623.0,
      "step": 17335
    },
    {
      "entropy": 6.262423276901245,
      "epoch": 1.8555834983145165,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046619388193816155,
      "loss": 6.0144,
      "mean_token_accuracy": 0.14633769690990447,
      "num_tokens": 37525618.0,
      "step": 17340
    },
    {
      "entropy": 6.337553596496582,
      "epoch": 1.8561185724222806,
      "grad_norm": 1.125,
      "learning_rate": 0.00046617372312739917,
      "loss": 6.0818,
      "mean_token_accuracy": 0.14901894330978394,
      "num_tokens": 37536023.0,
      "step": 17345
    },
    {
      "entropy": 6.395966291427612,
      "epoch": 1.8566536465300443,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004661535587965224,
      "loss": 6.0802,
      "mean_token_accuracy": 0.14241707175970078,
      "num_tokens": 37545886.0,
      "step": 17350
    },
    {
      "entropy": 6.261256837844849,
      "epoch": 1.8571887206378084,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046613338894611347,
      "loss": 6.0155,
      "mean_token_accuracy": 0.14702889919281006,
      "num_tokens": 37557587.0,
      "step": 17355
    },
    {
      "entropy": 6.175627517700195,
      "epoch": 1.8577237947455723,
      "grad_norm": 1.25,
      "learning_rate": 0.0004661132135767548,
      "loss": 5.9035,
      "mean_token_accuracy": 0.16342216581106186,
      "num_tokens": 37569379.0,
      "step": 17360
    },
    {
      "entropy": 6.355815076828003,
      "epoch": 1.8582588688533361,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000466093032689029,
      "loss": 6.1036,
      "mean_token_accuracy": 0.14511343389749526,
      "num_tokens": 37580478.0,
      "step": 17365
    },
    {
      "entropy": 6.304299974441529,
      "epoch": 1.8587939429611002,
      "grad_norm": 1.25,
      "learning_rate": 0.0004660728462835187,
      "loss": 6.03,
      "mean_token_accuracy": 0.15120670273900033,
      "num_tokens": 37591506.0,
      "step": 17370
    },
    {
      "entropy": 6.2582886695861815,
      "epoch": 1.8593290170688639,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004660526543608068,
      "loss": 5.8722,
      "mean_token_accuracy": 0.15693795830011367,
      "num_tokens": 37602074.0,
      "step": 17375
    },
    {
      "entropy": 6.289897060394287,
      "epoch": 1.859864091176628,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004660324569214763,
      "loss": 6.0702,
      "mean_token_accuracy": 0.14545617997646332,
      "num_tokens": 37612180.0,
      "step": 17380
    },
    {
      "entropy": 6.287088203430176,
      "epoch": 1.8603991652843919,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004660122539661106,
      "loss": 6.0279,
      "mean_token_accuracy": 0.14495737180113794,
      "num_tokens": 37622588.0,
      "step": 17385
    },
    {
      "entropy": 6.311354446411133,
      "epoch": 1.8609342393921557,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004659920454952928,
      "loss": 6.0298,
      "mean_token_accuracy": 0.15115011855959892,
      "num_tokens": 37634323.0,
      "step": 17390
    },
    {
      "entropy": 6.33291597366333,
      "epoch": 1.8614693134999198,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004659718315096065,
      "loss": 6.081,
      "mean_token_accuracy": 0.14115844294428825,
      "num_tokens": 37645816.0,
      "step": 17395
    },
    {
      "entropy": 6.383894062042236,
      "epoch": 1.8620043876076835,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046595161200963545,
      "loss": 6.0977,
      "mean_token_accuracy": 0.14632151797413825,
      "num_tokens": 37656143.0,
      "step": 17400
    },
    {
      "entropy": 6.299978351593017,
      "epoch": 1.8625394617154476,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046593138699596343,
      "loss": 5.984,
      "mean_token_accuracy": 0.14558647572994232,
      "num_tokens": 37665959.0,
      "step": 17405
    },
    {
      "entropy": 6.297830057144165,
      "epoch": 1.8630745358232115,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046591115646917443,
      "loss": 6.0538,
      "mean_token_accuracy": 0.14295567721128463,
      "num_tokens": 37677176.0,
      "step": 17410
    },
    {
      "entropy": 6.333182334899902,
      "epoch": 1.8636096099309754,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004658909204298526,
      "loss": 6.0728,
      "mean_token_accuracy": 0.14910986870527268,
      "num_tokens": 37687740.0,
      "step": 17415
    },
    {
      "entropy": 6.2625665187835695,
      "epoch": 1.8641446840387395,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004658706788785823,
      "loss": 6.0106,
      "mean_token_accuracy": 0.15245715752243996,
      "num_tokens": 37698536.0,
      "step": 17420
    },
    {
      "entropy": 6.332885265350342,
      "epoch": 1.8646797581465033,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046585043181594793,
      "loss": 6.0436,
      "mean_token_accuracy": 0.15239830166101456,
      "num_tokens": 37708795.0,
      "step": 17425
    },
    {
      "entropy": 6.353929615020752,
      "epoch": 1.8652148322542672,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046583017924253426,
      "loss": 6.0502,
      "mean_token_accuracy": 0.15134222954511642,
      "num_tokens": 37719136.0,
      "step": 17430
    },
    {
      "entropy": 6.440292024612427,
      "epoch": 1.8657499063620313,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004658099211589259,
      "loss": 6.1347,
      "mean_token_accuracy": 0.1404387392103672,
      "num_tokens": 37730513.0,
      "step": 17435
    },
    {
      "entropy": 6.274427795410157,
      "epoch": 1.866284980469795,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046578965756570787,
      "loss": 6.0636,
      "mean_token_accuracy": 0.1477853387594223,
      "num_tokens": 37741811.0,
      "step": 17440
    },
    {
      "entropy": 6.238445663452149,
      "epoch": 1.866820054577559,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046576938846346527,
      "loss": 5.8936,
      "mean_token_accuracy": 0.15747265368700028,
      "num_tokens": 37752213.0,
      "step": 17445
    },
    {
      "entropy": 6.218270778656006,
      "epoch": 1.867355128685323,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046574911385278343,
      "loss": 5.8844,
      "mean_token_accuracy": 0.1623547777533531,
      "num_tokens": 37763253.0,
      "step": 17450
    },
    {
      "entropy": 6.307385540008545,
      "epoch": 1.8678902027930868,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046572883373424776,
      "loss": 6.0745,
      "mean_token_accuracy": 0.14409469813108444,
      "num_tokens": 37775087.0,
      "step": 17455
    },
    {
      "entropy": 6.3065132141113285,
      "epoch": 1.868425276900851,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046570854810844373,
      "loss": 6.0842,
      "mean_token_accuracy": 0.14057199358940126,
      "num_tokens": 37786119.0,
      "step": 17460
    },
    {
      "entropy": 6.351084995269775,
      "epoch": 1.8689603510086146,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046568825697595724,
      "loss": 6.0831,
      "mean_token_accuracy": 0.14282397255301477,
      "num_tokens": 37797482.0,
      "step": 17465
    },
    {
      "entropy": 6.310936403274536,
      "epoch": 1.8694954251163787,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004656679603373741,
      "loss": 6.011,
      "mean_token_accuracy": 0.14718472808599473,
      "num_tokens": 37808067.0,
      "step": 17470
    },
    {
      "entropy": 6.348761177062988,
      "epoch": 1.8700304992241426,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004656476581932804,
      "loss": 6.1609,
      "mean_token_accuracy": 0.13783096820116042,
      "num_tokens": 37819762.0,
      "step": 17475
    },
    {
      "entropy": 6.303392553329468,
      "epoch": 1.8705655733319064,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004656273505442624,
      "loss": 6.0256,
      "mean_token_accuracy": 0.14932544231414796,
      "num_tokens": 37829510.0,
      "step": 17480
    },
    {
      "entropy": 6.343889904022217,
      "epoch": 1.8711006474396705,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046560703739090633,
      "loss": 6.0966,
      "mean_token_accuracy": 0.14320994466543197,
      "num_tokens": 37839431.0,
      "step": 17485
    },
    {
      "entropy": 6.380897808074951,
      "epoch": 1.8716357215474342,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046558671873379886,
      "loss": 6.099,
      "mean_token_accuracy": 0.1362930029630661,
      "num_tokens": 37850531.0,
      "step": 17490
    },
    {
      "entropy": 6.3672568798065186,
      "epoch": 1.8721707956551983,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004655663945735268,
      "loss": 6.1207,
      "mean_token_accuracy": 0.14048374071717262,
      "num_tokens": 37861964.0,
      "step": 17495
    },
    {
      "entropy": 6.1422323226928714,
      "epoch": 1.8727058697629622,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004655460649106768,
      "loss": 5.8248,
      "mean_token_accuracy": 0.16607989817857743,
      "num_tokens": 37873598.0,
      "step": 17500
    },
    {
      "entropy": 6.37021541595459,
      "epoch": 1.873240943870726,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046552572974583597,
      "loss": 6.0873,
      "mean_token_accuracy": 0.1414860188961029,
      "num_tokens": 37884643.0,
      "step": 17505
    },
    {
      "entropy": 6.326335620880127,
      "epoch": 1.8737760179784901,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004655053890795914,
      "loss": 6.0861,
      "mean_token_accuracy": 0.14260366708040237,
      "num_tokens": 37896566.0,
      "step": 17510
    },
    {
      "entropy": 6.36624960899353,
      "epoch": 1.8743110920862538,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046548504291253054,
      "loss": 6.113,
      "mean_token_accuracy": 0.1451176829636097,
      "num_tokens": 37907424.0,
      "step": 17515
    },
    {
      "entropy": 6.350479507446289,
      "epoch": 1.874846166194018,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004654646912452408,
      "loss": 6.0592,
      "mean_token_accuracy": 0.15171001255512237,
      "num_tokens": 37917318.0,
      "step": 17520
    },
    {
      "entropy": 6.283895111083984,
      "epoch": 1.8753812403017818,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004654443340783099,
      "loss": 6.0067,
      "mean_token_accuracy": 0.14894810318946838,
      "num_tokens": 37928521.0,
      "step": 17525
    },
    {
      "entropy": 6.306560897827149,
      "epoch": 1.8759163144095456,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046542397141232554,
      "loss": 6.0783,
      "mean_token_accuracy": 0.1489613100886345,
      "num_tokens": 37939930.0,
      "step": 17530
    },
    {
      "entropy": 6.280333042144775,
      "epoch": 1.8764513885173097,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004654036032478759,
      "loss": 5.9951,
      "mean_token_accuracy": 0.1482404351234436,
      "num_tokens": 37949690.0,
      "step": 17535
    },
    {
      "entropy": 6.329837322235107,
      "epoch": 1.8769864626250736,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046538322958554885,
      "loss": 6.0101,
      "mean_token_accuracy": 0.15361273735761644,
      "num_tokens": 37962162.0,
      "step": 17540
    },
    {
      "entropy": 6.449858093261719,
      "epoch": 1.8775215367328375,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004653628504259329,
      "loss": 6.0622,
      "mean_token_accuracy": 0.14640085399150848,
      "num_tokens": 37972707.0,
      "step": 17545
    },
    {
      "entropy": 6.259990406036377,
      "epoch": 1.8780566108406014,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046534246576961633,
      "loss": 5.9769,
      "mean_token_accuracy": 0.1556459903717041,
      "num_tokens": 37983101.0,
      "step": 17550
    },
    {
      "entropy": 6.168662118911743,
      "epoch": 1.8785916849483653,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046532207561718775,
      "loss": 5.9478,
      "mean_token_accuracy": 0.15862448364496232,
      "num_tokens": 37994025.0,
      "step": 17555
    },
    {
      "entropy": 6.32290506362915,
      "epoch": 1.8791267590561294,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046530167996923613,
      "loss": 6.0495,
      "mean_token_accuracy": 0.1394786097109318,
      "num_tokens": 38004761.0,
      "step": 17560
    },
    {
      "entropy": 6.415972471237183,
      "epoch": 1.8796618331638932,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004652812788263501,
      "loss": 6.054,
      "mean_token_accuracy": 0.14856105893850327,
      "num_tokens": 38015129.0,
      "step": 17565
    },
    {
      "entropy": 6.301147365570069,
      "epoch": 1.8801969072716571,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046526087218911894,
      "loss": 6.0604,
      "mean_token_accuracy": 0.15063174217939376,
      "num_tokens": 38025501.0,
      "step": 17570
    },
    {
      "entropy": 6.2872171878814695,
      "epoch": 1.8807319813794212,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046524046005813185,
      "loss": 6.04,
      "mean_token_accuracy": 0.1467640705406666,
      "num_tokens": 38034912.0,
      "step": 17575
    },
    {
      "entropy": 6.2926897525787355,
      "epoch": 1.8812670554871849,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004652200424339782,
      "loss": 5.9742,
      "mean_token_accuracy": 0.14987795650959015,
      "num_tokens": 38045417.0,
      "step": 17580
    },
    {
      "entropy": 6.342170715332031,
      "epoch": 1.881802129594949,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004651996193172475,
      "loss": 6.1213,
      "mean_token_accuracy": 0.14215622022747992,
      "num_tokens": 38057033.0,
      "step": 17585
    },
    {
      "entropy": 6.361774969100952,
      "epoch": 1.8823372037027128,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004651791907085296,
      "loss": 6.039,
      "mean_token_accuracy": 0.14554664492607117,
      "num_tokens": 38067804.0,
      "step": 17590
    },
    {
      "entropy": 6.325141143798828,
      "epoch": 1.8828722778104767,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004651587566084143,
      "loss": 6.0241,
      "mean_token_accuracy": 0.15701245218515397,
      "num_tokens": 38079246.0,
      "step": 17595
    },
    {
      "entropy": 6.341245079040528,
      "epoch": 1.8834073519182408,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046513831701749155,
      "loss": 6.179,
      "mean_token_accuracy": 0.14081829488277436,
      "num_tokens": 38090628.0,
      "step": 17600
    },
    {
      "entropy": 6.30099458694458,
      "epoch": 1.8839424260260045,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046511787193635165,
      "loss": 5.9517,
      "mean_token_accuracy": 0.15635764375329017,
      "num_tokens": 38100848.0,
      "step": 17605
    },
    {
      "entropy": 6.343262338638306,
      "epoch": 1.8844775001337686,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046509742136558493,
      "loss": 6.0131,
      "mean_token_accuracy": 0.14849376380443574,
      "num_tokens": 38110732.0,
      "step": 17610
    },
    {
      "entropy": 6.2266546249389645,
      "epoch": 1.8850125742415325,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046507696530578185,
      "loss": 6.0065,
      "mean_token_accuracy": 0.15698181986808776,
      "num_tokens": 38121002.0,
      "step": 17615
    },
    {
      "entropy": 6.317318105697632,
      "epoch": 1.8855476483492963,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004650565037575331,
      "loss": 6.0669,
      "mean_token_accuracy": 0.1473769649863243,
      "num_tokens": 38132294.0,
      "step": 17620
    },
    {
      "entropy": 6.298158931732178,
      "epoch": 1.8860827224570604,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046503603672142955,
      "loss": 5.9515,
      "mean_token_accuracy": 0.15036978274583818,
      "num_tokens": 38142962.0,
      "step": 17625
    },
    {
      "entropy": 6.3926512718200685,
      "epoch": 1.886617796564824,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004650155641980621,
      "loss": 6.0663,
      "mean_token_accuracy": 0.1506273165345192,
      "num_tokens": 38153737.0,
      "step": 17630
    },
    {
      "entropy": 6.345070314407349,
      "epoch": 1.8871528706725882,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046499508618802195,
      "loss": 6.0461,
      "mean_token_accuracy": 0.14506750255823136,
      "num_tokens": 38163897.0,
      "step": 17635
    },
    {
      "entropy": 6.240292549133301,
      "epoch": 1.887687944780352,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004649746026919004,
      "loss": 6.0138,
      "mean_token_accuracy": 0.15763003528118133,
      "num_tokens": 38175123.0,
      "step": 17640
    },
    {
      "entropy": 6.285467004776001,
      "epoch": 1.888223018888116,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00046495411371028886,
      "loss": 6.0395,
      "mean_token_accuracy": 0.15418533533811568,
      "num_tokens": 38186100.0,
      "step": 17645
    },
    {
      "entropy": 6.371870040893555,
      "epoch": 1.88875809299588,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046493361924377903,
      "loss": 6.0621,
      "mean_token_accuracy": 0.1501191645860672,
      "num_tokens": 38196385.0,
      "step": 17650
    },
    {
      "entropy": 6.279502868652344,
      "epoch": 1.8892931671036437,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004649131192929626,
      "loss": 5.9933,
      "mean_token_accuracy": 0.15192514657974243,
      "num_tokens": 38207260.0,
      "step": 17655
    },
    {
      "entropy": 6.245561599731445,
      "epoch": 1.8898282412114078,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004648926138584315,
      "loss": 5.9003,
      "mean_token_accuracy": 0.15646348744630814,
      "num_tokens": 38217537.0,
      "step": 17660
    },
    {
      "entropy": 6.343149900436401,
      "epoch": 1.8903633153191717,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046487210294077794,
      "loss": 6.0175,
      "mean_token_accuracy": 0.15129930526018143,
      "num_tokens": 38228956.0,
      "step": 17665
    },
    {
      "entropy": 6.223767328262329,
      "epoch": 1.8908983894269356,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046485158654059405,
      "loss": 5.9557,
      "mean_token_accuracy": 0.15658189132809638,
      "num_tokens": 38240236.0,
      "step": 17670
    },
    {
      "entropy": 6.350696849822998,
      "epoch": 1.8914334635346997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004648310646584723,
      "loss": 6.0932,
      "mean_token_accuracy": 0.14926495403051376,
      "num_tokens": 38251739.0,
      "step": 17675
    },
    {
      "entropy": 6.300583076477051,
      "epoch": 1.8919685376424635,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046481053729500516,
      "loss": 6.02,
      "mean_token_accuracy": 0.13941954895853997,
      "num_tokens": 38262090.0,
      "step": 17680
    },
    {
      "entropy": 6.317531967163086,
      "epoch": 1.8925036117502274,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004647900044507855,
      "loss": 6.0417,
      "mean_token_accuracy": 0.1480513721704483,
      "num_tokens": 38272770.0,
      "step": 17685
    },
    {
      "entropy": 6.2865033626556395,
      "epoch": 1.8930386858579913,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004647694661264061,
      "loss": 5.9966,
      "mean_token_accuracy": 0.14645040184259414,
      "num_tokens": 38283449.0,
      "step": 17690
    },
    {
      "entropy": 6.29221830368042,
      "epoch": 1.8935737599657552,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046474892232246,
      "loss": 6.0329,
      "mean_token_accuracy": 0.15198683440685273,
      "num_tokens": 38293831.0,
      "step": 17695
    },
    {
      "entropy": 6.305269765853882,
      "epoch": 1.8941088340735193,
      "grad_norm": 1.25,
      "learning_rate": 0.00046472837303954053,
      "loss": 6.0016,
      "mean_token_accuracy": 0.14750904142856597,
      "num_tokens": 38304782.0,
      "step": 17700
    },
    {
      "entropy": 6.296178245544434,
      "epoch": 1.8946439081812831,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046470781827824087,
      "loss": 6.0365,
      "mean_token_accuracy": 0.15468640774488449,
      "num_tokens": 38316547.0,
      "step": 17705
    },
    {
      "entropy": 6.387094259262085,
      "epoch": 1.895178982289047,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046468725803915464,
      "loss": 6.071,
      "mean_token_accuracy": 0.1527080163359642,
      "num_tokens": 38329488.0,
      "step": 17710
    },
    {
      "entropy": 6.200169277191162,
      "epoch": 1.8957140563968111,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046466669232287543,
      "loss": 5.9105,
      "mean_token_accuracy": 0.15840624272823334,
      "num_tokens": 38341290.0,
      "step": 17715
    },
    {
      "entropy": 6.308721542358398,
      "epoch": 1.8962491305045748,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004646461211299973,
      "loss": 6.0603,
      "mean_token_accuracy": 0.14425690099596977,
      "num_tokens": 38352445.0,
      "step": 17720
    },
    {
      "entropy": 6.303908205032348,
      "epoch": 1.8967842046123389,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004646255444611139,
      "loss": 6.0662,
      "mean_token_accuracy": 0.1480732724070549,
      "num_tokens": 38363825.0,
      "step": 17725
    },
    {
      "entropy": 6.28974027633667,
      "epoch": 1.8973192787201028,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046460496231681957,
      "loss": 6.1197,
      "mean_token_accuracy": 0.1497628442943096,
      "num_tokens": 38375454.0,
      "step": 17730
    },
    {
      "entropy": 6.281067609786987,
      "epoch": 1.8978543528278666,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046458437469770866,
      "loss": 6.0604,
      "mean_token_accuracy": 0.14820998013019562,
      "num_tokens": 38385938.0,
      "step": 17735
    },
    {
      "entropy": 6.345644140243531,
      "epoch": 1.8983894269356307,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046456378160437555,
      "loss": 5.9968,
      "mean_token_accuracy": 0.1523777149617672,
      "num_tokens": 38396143.0,
      "step": 17740
    },
    {
      "entropy": 6.253451538085938,
      "epoch": 1.8989245010433944,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004645431830374149,
      "loss": 6.0691,
      "mean_token_accuracy": 0.14006222411990166,
      "num_tokens": 38407445.0,
      "step": 17745
    },
    {
      "entropy": 6.339150094985962,
      "epoch": 1.8994595751511585,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046452257899742144,
      "loss": 6.103,
      "mean_token_accuracy": 0.13872334435582162,
      "num_tokens": 38418438.0,
      "step": 17750
    },
    {
      "entropy": 6.320700311660767,
      "epoch": 1.8999946492589224,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046450196948499015,
      "loss": 5.9822,
      "mean_token_accuracy": 0.14646454229950906,
      "num_tokens": 38428821.0,
      "step": 17755
    },
    {
      "entropy": 6.309258270263672,
      "epoch": 1.9005297233666862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004644813545007161,
      "loss": 5.9537,
      "mean_token_accuracy": 0.14943273663520812,
      "num_tokens": 38439560.0,
      "step": 17760
    },
    {
      "entropy": 6.297953319549561,
      "epoch": 1.9010647974744503,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004644607340451946,
      "loss": 6.0594,
      "mean_token_accuracy": 0.1509922757744789,
      "num_tokens": 38449724.0,
      "step": 17765
    },
    {
      "entropy": 6.281881427764892,
      "epoch": 1.901599871582214,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046444010811902105,
      "loss": 5.9868,
      "mean_token_accuracy": 0.15119580328464508,
      "num_tokens": 38459833.0,
      "step": 17770
    },
    {
      "entropy": 6.248759460449219,
      "epoch": 1.902134945689978,
      "grad_norm": 1.25,
      "learning_rate": 0.00046441947672279095,
      "loss": 5.9989,
      "mean_token_accuracy": 0.15790337771177293,
      "num_tokens": 38470682.0,
      "step": 17775
    },
    {
      "entropy": 6.325793933868408,
      "epoch": 1.902670019797742,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004643988398571001,
      "loss": 6.0285,
      "mean_token_accuracy": 0.1470477595925331,
      "num_tokens": 38481362.0,
      "step": 17780
    },
    {
      "entropy": 6.2854132652282715,
      "epoch": 1.9032050939055059,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004643781975225444,
      "loss": 6.0053,
      "mean_token_accuracy": 0.15276707857847213,
      "num_tokens": 38492941.0,
      "step": 17785
    },
    {
      "entropy": 6.333832454681397,
      "epoch": 1.90374016801327,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046435754971971976,
      "loss": 6.0445,
      "mean_token_accuracy": 0.14332455843687059,
      "num_tokens": 38503823.0,
      "step": 17790
    },
    {
      "entropy": 6.301437282562256,
      "epoch": 1.9042752421210336,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046433689644922256,
      "loss": 5.9726,
      "mean_token_accuracy": 0.1584117367863655,
      "num_tokens": 38514482.0,
      "step": 17795
    },
    {
      "entropy": 6.29148302078247,
      "epoch": 1.9048103162287977,
      "grad_norm": 1.125,
      "learning_rate": 0.00046431623771164897,
      "loss": 6.0391,
      "mean_token_accuracy": 0.14927564710378646,
      "num_tokens": 38525580.0,
      "step": 17800
    },
    {
      "entropy": 6.292261791229248,
      "epoch": 1.9053453903365616,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004642955735075958,
      "loss": 6.0307,
      "mean_token_accuracy": 0.15435453578829766,
      "num_tokens": 38537025.0,
      "step": 17805
    },
    {
      "entropy": 6.297820234298706,
      "epoch": 1.9058804644443255,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046427490383765943,
      "loss": 5.9968,
      "mean_token_accuracy": 0.15531120002269744,
      "num_tokens": 38548974.0,
      "step": 17810
    },
    {
      "entropy": 6.314402914047241,
      "epoch": 1.9064155385520896,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004642542287024368,
      "loss": 5.9266,
      "mean_token_accuracy": 0.15357589051127435,
      "num_tokens": 38559971.0,
      "step": 17815
    },
    {
      "entropy": 6.285403251647949,
      "epoch": 1.9069506126598534,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000464233548102525,
      "loss": 5.9952,
      "mean_token_accuracy": 0.14812205135822296,
      "num_tokens": 38570326.0,
      "step": 17820
    },
    {
      "entropy": 6.21509485244751,
      "epoch": 1.9074856867676173,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000464212862038521,
      "loss": 6.0042,
      "mean_token_accuracy": 0.1517790824174881,
      "num_tokens": 38580799.0,
      "step": 17825
    },
    {
      "entropy": 6.227622365951538,
      "epoch": 1.9080207608753812,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046419217051102223,
      "loss": 6.0863,
      "mean_token_accuracy": 0.14219640493392943,
      "num_tokens": 38591136.0,
      "step": 17830
    },
    {
      "entropy": 6.3695900440216064,
      "epoch": 1.908555834983145,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004641714735206261,
      "loss": 5.9781,
      "mean_token_accuracy": 0.15012947618961334,
      "num_tokens": 38602600.0,
      "step": 17835
    },
    {
      "entropy": 6.305353164672852,
      "epoch": 1.9090909090909092,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046415077106793024,
      "loss": 5.9974,
      "mean_token_accuracy": 0.14993587732315064,
      "num_tokens": 38612641.0,
      "step": 17840
    },
    {
      "entropy": 6.267118120193482,
      "epoch": 1.909625983198673,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004641300631535325,
      "loss": 6.0128,
      "mean_token_accuracy": 0.148307603597641,
      "num_tokens": 38622873.0,
      "step": 17845
    },
    {
      "entropy": 6.258100795745849,
      "epoch": 1.910161057306437,
      "grad_norm": 1.375,
      "learning_rate": 0.0004641093497780307,
      "loss": 6.0094,
      "mean_token_accuracy": 0.15239158123731614,
      "num_tokens": 38633523.0,
      "step": 17850
    },
    {
      "entropy": 6.326640319824219,
      "epoch": 1.910696131414201,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004640886309420231,
      "loss": 6.038,
      "mean_token_accuracy": 0.14205127954483032,
      "num_tokens": 38645635.0,
      "step": 17855
    },
    {
      "entropy": 6.379506063461304,
      "epoch": 1.9112312055219647,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004640679066461078,
      "loss": 5.9903,
      "mean_token_accuracy": 0.1521977350115776,
      "num_tokens": 38656293.0,
      "step": 17860
    },
    {
      "entropy": 6.285054445266724,
      "epoch": 1.9117662796297288,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004640471768908832,
      "loss": 6.1156,
      "mean_token_accuracy": 0.14167936220765115,
      "num_tokens": 38667307.0,
      "step": 17865
    },
    {
      "entropy": 6.291289663314819,
      "epoch": 1.9123013537374927,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004640264416769481,
      "loss": 5.9904,
      "mean_token_accuracy": 0.14595004469156264,
      "num_tokens": 38677775.0,
      "step": 17870
    },
    {
      "entropy": 6.2913422107696535,
      "epoch": 1.9128364278452565,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046400570100490086,
      "loss": 6.0854,
      "mean_token_accuracy": 0.14743582904338837,
      "num_tokens": 38688652.0,
      "step": 17875
    },
    {
      "entropy": 6.348163747787476,
      "epoch": 1.9133715019530206,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046398495487534066,
      "loss": 6.0443,
      "mean_token_accuracy": 0.14716565683484079,
      "num_tokens": 38698694.0,
      "step": 17880
    },
    {
      "entropy": 6.364686393737793,
      "epoch": 1.9139065760607843,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004639642032888664,
      "loss": 6.0563,
      "mean_token_accuracy": 0.14937563464045525,
      "num_tokens": 38709864.0,
      "step": 17885
    },
    {
      "entropy": 6.305392646789551,
      "epoch": 1.9144416501685484,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004639434462460774,
      "loss": 6.0503,
      "mean_token_accuracy": 0.14635968655347825,
      "num_tokens": 38721034.0,
      "step": 17890
    },
    {
      "entropy": 6.3241981029510494,
      "epoch": 1.9149767242763123,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046392268374757285,
      "loss": 6.0494,
      "mean_token_accuracy": 0.1502541109919548,
      "num_tokens": 38731874.0,
      "step": 17895
    },
    {
      "entropy": 6.2656043529510494,
      "epoch": 1.9155117983840761,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004639019157939524,
      "loss": 5.9951,
      "mean_token_accuracy": 0.15119791328907012,
      "num_tokens": 38742447.0,
      "step": 17900
    },
    {
      "entropy": 6.291540479660034,
      "epoch": 1.9160468724918402,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004638811423858157,
      "loss": 5.9302,
      "mean_token_accuracy": 0.14806411415338516,
      "num_tokens": 38754117.0,
      "step": 17905
    },
    {
      "entropy": 6.3333312511444095,
      "epoch": 1.916581946599604,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046386036352376255,
      "loss": 6.0266,
      "mean_token_accuracy": 0.13814645186066626,
      "num_tokens": 38765057.0,
      "step": 17910
    },
    {
      "entropy": 6.296029901504516,
      "epoch": 1.917117020707368,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046383957920839285,
      "loss": 5.9881,
      "mean_token_accuracy": 0.1498717799782753,
      "num_tokens": 38775938.0,
      "step": 17915
    },
    {
      "entropy": 6.211456155776977,
      "epoch": 1.9176520948151319,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046381878944030697,
      "loss": 5.9053,
      "mean_token_accuracy": 0.15291634127497672,
      "num_tokens": 38787013.0,
      "step": 17920
    },
    {
      "entropy": 6.2924071788787845,
      "epoch": 1.9181871689228958,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000463797994220105,
      "loss": 5.9559,
      "mean_token_accuracy": 0.1530446156859398,
      "num_tokens": 38798412.0,
      "step": 17925
    },
    {
      "entropy": 6.369398593902588,
      "epoch": 1.9187222430306599,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046377719354838756,
      "loss": 6.0528,
      "mean_token_accuracy": 0.14815336391329764,
      "num_tokens": 38809983.0,
      "step": 17930
    },
    {
      "entropy": 6.328118467330933,
      "epoch": 1.9192573171384235,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046375638742575506,
      "loss": 6.0405,
      "mean_token_accuracy": 0.15097953751683235,
      "num_tokens": 38820418.0,
      "step": 17935
    },
    {
      "entropy": 6.316989469528198,
      "epoch": 1.9197923912461876,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004637355758528085,
      "loss": 6.0402,
      "mean_token_accuracy": 0.15056614577770233,
      "num_tokens": 38831482.0,
      "step": 17940
    },
    {
      "entropy": 6.3982720375061035,
      "epoch": 1.9203274653539515,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004637147588301486,
      "loss": 6.0594,
      "mean_token_accuracy": 0.14227498918771744,
      "num_tokens": 38842110.0,
      "step": 17945
    },
    {
      "entropy": 6.236125516891479,
      "epoch": 1.9208625394617154,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046369393635837657,
      "loss": 5.9149,
      "mean_token_accuracy": 0.15191808044910432,
      "num_tokens": 38852861.0,
      "step": 17950
    },
    {
      "entropy": 6.258435869216919,
      "epoch": 1.9213976135694795,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046367310843809374,
      "loss": 6.0652,
      "mean_token_accuracy": 0.14252583980560302,
      "num_tokens": 38864067.0,
      "step": 17955
    },
    {
      "entropy": 6.302843189239502,
      "epoch": 1.9219326876772433,
      "grad_norm": 1.125,
      "learning_rate": 0.00046365227506990137,
      "loss": 6.0284,
      "mean_token_accuracy": 0.14652848169207572,
      "num_tokens": 38875363.0,
      "step": 17960
    },
    {
      "entropy": 6.3620734214782715,
      "epoch": 1.9224677617850072,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004636314362544011,
      "loss": 6.0458,
      "mean_token_accuracy": 0.15213124379515647,
      "num_tokens": 38886414.0,
      "step": 17965
    },
    {
      "entropy": 6.348482465744018,
      "epoch": 1.923002835892771,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046361059199219457,
      "loss": 6.1193,
      "mean_token_accuracy": 0.1426404744386673,
      "num_tokens": 38897708.0,
      "step": 17970
    },
    {
      "entropy": 6.275768280029297,
      "epoch": 1.923537910000535,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004635897422838837,
      "loss": 6.0001,
      "mean_token_accuracy": 0.15499154031276702,
      "num_tokens": 38908683.0,
      "step": 17975
    },
    {
      "entropy": 6.248620080947876,
      "epoch": 1.924072984108299,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046356888713007055,
      "loss": 5.9569,
      "mean_token_accuracy": 0.15884367972612382,
      "num_tokens": 38919279.0,
      "step": 17980
    },
    {
      "entropy": 6.365066432952881,
      "epoch": 1.924608058216063,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046354802653135723,
      "loss": 6.0519,
      "mean_token_accuracy": 0.14488793760538102,
      "num_tokens": 38929353.0,
      "step": 17985
    },
    {
      "entropy": 6.312986421585083,
      "epoch": 1.9251431323238268,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004635271604883462,
      "loss": 5.9996,
      "mean_token_accuracy": 0.15333318263292312,
      "num_tokens": 38939296.0,
      "step": 17990
    },
    {
      "entropy": 6.251923274993897,
      "epoch": 1.925678206431591,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046350628900163987,
      "loss": 6.0154,
      "mean_token_accuracy": 0.15132410749793052,
      "num_tokens": 38950149.0,
      "step": 17995
    },
    {
      "entropy": 6.316252994537353,
      "epoch": 1.9262132805393546,
      "grad_norm": 1.3125,
      "learning_rate": 0.000463485412071841,
      "loss": 6.0695,
      "mean_token_accuracy": 0.14332675263285638,
      "num_tokens": 38961052.0,
      "step": 18000
    },
    {
      "epoch": 1.9262132805393546,
      "eval_entropy": 6.166329034126672,
      "eval_loss": 6.128286361694336,
      "eval_mean_token_accuracy": 0.15173338689487378,
      "eval_num_tokens": 38961052.0,
      "eval_runtime": 22.7817,
      "eval_samples_per_second": 1302.754,
      "eval_steps_per_second": 162.85,
      "step": 18000
    },
    {
      "entropy": 6.343509387969971,
      "epoch": 1.9267483546471187,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046346452969955214,
      "loss": 6.0535,
      "mean_token_accuracy": 0.153200463950634,
      "num_tokens": 38972344.0,
      "step": 18005
    },
    {
      "entropy": 6.306464624404907,
      "epoch": 1.9272834287548826,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004634436418853767,
      "loss": 6.0276,
      "mean_token_accuracy": 0.1441355124115944,
      "num_tokens": 38982877.0,
      "step": 18010
    },
    {
      "entropy": 6.32079758644104,
      "epoch": 1.9278185028626464,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046342274862991745,
      "loss": 6.0391,
      "mean_token_accuracy": 0.1504403308033943,
      "num_tokens": 38994322.0,
      "step": 18015
    },
    {
      "entropy": 6.3003448963165285,
      "epoch": 1.9283535769704105,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046340184993377793,
      "loss": 5.9548,
      "mean_token_accuracy": 0.14971892833709716,
      "num_tokens": 39003447.0,
      "step": 18020
    },
    {
      "entropy": 6.2723212242126465,
      "epoch": 1.9288886510781742,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046338094579756136,
      "loss": 6.0421,
      "mean_token_accuracy": 0.14324226900935172,
      "num_tokens": 39013760.0,
      "step": 18025
    },
    {
      "entropy": 6.161333084106445,
      "epoch": 1.9294237251859383,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046336003622187143,
      "loss": 5.8416,
      "mean_token_accuracy": 0.16033360958099366,
      "num_tokens": 39025559.0,
      "step": 18030
    },
    {
      "entropy": 6.265558528900146,
      "epoch": 1.9299587992937022,
      "grad_norm": 1.21875,
      "learning_rate": 0.000463339121207312,
      "loss": 6.0141,
      "mean_token_accuracy": 0.14988713413476945,
      "num_tokens": 39036004.0,
      "step": 18035
    },
    {
      "entropy": 6.296011114120484,
      "epoch": 1.930493873401466,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004633182007544868,
      "loss": 6.0539,
      "mean_token_accuracy": 0.14676553905010223,
      "num_tokens": 39047251.0,
      "step": 18040
    },
    {
      "entropy": 6.311164093017578,
      "epoch": 1.9310289475092302,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046329727486400015,
      "loss": 6.077,
      "mean_token_accuracy": 0.14490970373153686,
      "num_tokens": 39057255.0,
      "step": 18045
    },
    {
      "entropy": 6.347948360443115,
      "epoch": 1.9315640216169938,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004632763435364561,
      "loss": 6.0806,
      "mean_token_accuracy": 0.14535010755062103,
      "num_tokens": 39068186.0,
      "step": 18050
    },
    {
      "entropy": 6.239805030822754,
      "epoch": 1.932099095724758,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046325540677245907,
      "loss": 5.9834,
      "mean_token_accuracy": 0.1517287865281105,
      "num_tokens": 39079173.0,
      "step": 18055
    },
    {
      "entropy": 6.274848747253418,
      "epoch": 1.9326341698325218,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046323446457261365,
      "loss": 6.0639,
      "mean_token_accuracy": 0.14923022389411927,
      "num_tokens": 39089939.0,
      "step": 18060
    },
    {
      "entropy": 6.278050279617309,
      "epoch": 1.9331692439402857,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004632135169375245,
      "loss": 6.0547,
      "mean_token_accuracy": 0.14747694209218026,
      "num_tokens": 39100464.0,
      "step": 18065
    },
    {
      "entropy": 6.3848714351654055,
      "epoch": 1.9337043180480498,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004631925638677965,
      "loss": 6.0019,
      "mean_token_accuracy": 0.146662737429142,
      "num_tokens": 39111161.0,
      "step": 18070
    },
    {
      "entropy": 6.33362832069397,
      "epoch": 1.9342393921558134,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046317160536403463,
      "loss": 6.0098,
      "mean_token_accuracy": 0.15724727660417556,
      "num_tokens": 39122469.0,
      "step": 18075
    },
    {
      "entropy": 6.274511337280273,
      "epoch": 1.9347744662635775,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046315064142684416,
      "loss": 5.9836,
      "mean_token_accuracy": 0.15219168961048127,
      "num_tokens": 39131871.0,
      "step": 18080
    },
    {
      "entropy": 6.291130542755127,
      "epoch": 1.9353095403713414,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004631296720568304,
      "loss": 6.0306,
      "mean_token_accuracy": 0.1457435183227062,
      "num_tokens": 39143061.0,
      "step": 18085
    },
    {
      "entropy": 6.278008270263672,
      "epoch": 1.9358446144791053,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004631086972545987,
      "loss": 6.0247,
      "mean_token_accuracy": 0.15730181485414504,
      "num_tokens": 39152452.0,
      "step": 18090
    },
    {
      "entropy": 6.224421739578247,
      "epoch": 1.9363796885868694,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004630877170207548,
      "loss": 5.9823,
      "mean_token_accuracy": 0.15295254290103913,
      "num_tokens": 39163196.0,
      "step": 18095
    },
    {
      "entropy": 6.237130069732666,
      "epoch": 1.9369147626946333,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004630667313559045,
      "loss": 5.9717,
      "mean_token_accuracy": 0.14708348959684373,
      "num_tokens": 39173958.0,
      "step": 18100
    },
    {
      "entropy": 6.301985168457032,
      "epoch": 1.9374498368023971,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046304574026065376,
      "loss": 6.0144,
      "mean_token_accuracy": 0.15550594553351402,
      "num_tokens": 39184483.0,
      "step": 18105
    },
    {
      "entropy": 6.2686866283416744,
      "epoch": 1.937984910910161,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004630247437356087,
      "loss": 6.0585,
      "mean_token_accuracy": 0.14653006717562675,
      "num_tokens": 39196500.0,
      "step": 18110
    },
    {
      "entropy": 6.379847478866577,
      "epoch": 1.9385199850179249,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046300374178137566,
      "loss": 6.0905,
      "mean_token_accuracy": 0.14405608028173447,
      "num_tokens": 39207492.0,
      "step": 18115
    },
    {
      "entropy": 6.31580810546875,
      "epoch": 1.939055059125689,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004629827343985609,
      "loss": 6.0186,
      "mean_token_accuracy": 0.15094533562660217,
      "num_tokens": 39218204.0,
      "step": 18120
    },
    {
      "entropy": 6.332547950744629,
      "epoch": 1.9395901332334529,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046296172158777116,
      "loss": 6.0736,
      "mean_token_accuracy": 0.14471327811479567,
      "num_tokens": 39229646.0,
      "step": 18125
    },
    {
      "entropy": 6.278395366668701,
      "epoch": 1.9401252073412167,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004629407033496131,
      "loss": 6.0106,
      "mean_token_accuracy": 0.1492026813328266,
      "num_tokens": 39240151.0,
      "step": 18130
    },
    {
      "entropy": 6.346271944046021,
      "epoch": 1.9406602814489808,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004629196796846936,
      "loss": 6.0788,
      "mean_token_accuracy": 0.14997884184122084,
      "num_tokens": 39251571.0,
      "step": 18135
    },
    {
      "entropy": 6.3554517269134525,
      "epoch": 1.9411953555567445,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046289865059361983,
      "loss": 6.004,
      "mean_token_accuracy": 0.14824894890189172,
      "num_tokens": 39262937.0,
      "step": 18140
    },
    {
      "entropy": 6.364266490936279,
      "epoch": 1.9417304296645086,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046287761607699887,
      "loss": 6.0973,
      "mean_token_accuracy": 0.14132825359702111,
      "num_tokens": 39274354.0,
      "step": 18145
    },
    {
      "entropy": 6.3204405307769775,
      "epoch": 1.9422655037722725,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004628565761354382,
      "loss": 5.9815,
      "mean_token_accuracy": 0.15025858134031295,
      "num_tokens": 39284314.0,
      "step": 18150
    },
    {
      "entropy": 6.2811431884765625,
      "epoch": 1.9428005778800364,
      "grad_norm": 3.96875,
      "learning_rate": 0.00046283553076954527,
      "loss": 6.0215,
      "mean_token_accuracy": 0.1461009532213211,
      "num_tokens": 39295516.0,
      "step": 18155
    },
    {
      "entropy": 6.314745235443115,
      "epoch": 1.9433356519878004,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004628144799799277,
      "loss": 6.0232,
      "mean_token_accuracy": 0.14916462302207947,
      "num_tokens": 39307221.0,
      "step": 18160
    },
    {
      "entropy": 6.283594083786011,
      "epoch": 1.943870726095564,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004627934237671935,
      "loss": 6.0435,
      "mean_token_accuracy": 0.14676849395036698,
      "num_tokens": 39318448.0,
      "step": 18165
    },
    {
      "entropy": 6.2555793762207035,
      "epoch": 1.9444058002033282,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046277236213195064,
      "loss": 5.9351,
      "mean_token_accuracy": 0.15645020306110383,
      "num_tokens": 39329384.0,
      "step": 18170
    },
    {
      "entropy": 6.321570348739624,
      "epoch": 1.944940874311092,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046275129507480705,
      "loss": 6.0699,
      "mean_token_accuracy": 0.14776507169008254,
      "num_tokens": 39340342.0,
      "step": 18175
    },
    {
      "entropy": 6.202507734298706,
      "epoch": 1.945475948418856,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004627302225963713,
      "loss": 5.9305,
      "mean_token_accuracy": 0.15681026950478555,
      "num_tokens": 39350825.0,
      "step": 18180
    },
    {
      "entropy": 6.3702291488647464,
      "epoch": 1.94601102252662,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046270914469725174,
      "loss": 6.0806,
      "mean_token_accuracy": 0.14557316452264785,
      "num_tokens": 39361986.0,
      "step": 18185
    },
    {
      "entropy": 6.3271324634552,
      "epoch": 1.9465460966343837,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046268806137805704,
      "loss": 5.9761,
      "mean_token_accuracy": 0.15123491138219833,
      "num_tokens": 39372440.0,
      "step": 18190
    },
    {
      "entropy": 6.288694381713867,
      "epoch": 1.9470811707421478,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004626669726393958,
      "loss": 5.9973,
      "mean_token_accuracy": 0.16221890076994896,
      "num_tokens": 39383455.0,
      "step": 18195
    },
    {
      "entropy": 6.273036670684815,
      "epoch": 1.9476162448499117,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004626458784818772,
      "loss": 6.0122,
      "mean_token_accuracy": 0.1471483141183853,
      "num_tokens": 39393511.0,
      "step": 18200
    },
    {
      "entropy": 6.277576351165772,
      "epoch": 1.9481513189576756,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046262477890611026,
      "loss": 6.0035,
      "mean_token_accuracy": 0.1512005239725113,
      "num_tokens": 39404049.0,
      "step": 18205
    },
    {
      "entropy": 6.326776075363159,
      "epoch": 1.9486863930654397,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046260367391270414,
      "loss": 6.0163,
      "mean_token_accuracy": 0.15134925842285157,
      "num_tokens": 39414771.0,
      "step": 18210
    },
    {
      "entropy": 6.269438028335571,
      "epoch": 1.9492214671732033,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046258256350226835,
      "loss": 5.9818,
      "mean_token_accuracy": 0.15268274247646332,
      "num_tokens": 39424869.0,
      "step": 18215
    },
    {
      "entropy": 6.281146287918091,
      "epoch": 1.9497565412809674,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004625614476754123,
      "loss": 6.0544,
      "mean_token_accuracy": 0.1506137803196907,
      "num_tokens": 39435442.0,
      "step": 18220
    },
    {
      "entropy": 6.347705078125,
      "epoch": 1.9502916153887313,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046254032643274596,
      "loss": 5.9972,
      "mean_token_accuracy": 0.15470413863658905,
      "num_tokens": 39446704.0,
      "step": 18225
    },
    {
      "entropy": 6.303225803375244,
      "epoch": 1.9508266894964952,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046251919977487887,
      "loss": 6.0154,
      "mean_token_accuracy": 0.15310133546590804,
      "num_tokens": 39458079.0,
      "step": 18230
    },
    {
      "entropy": 6.322163820266724,
      "epoch": 1.9513617636042593,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046249806770242135,
      "loss": 6.0093,
      "mean_token_accuracy": 0.15472922325134278,
      "num_tokens": 39469867.0,
      "step": 18235
    },
    {
      "entropy": 6.289956378936767,
      "epoch": 1.9518968377120232,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004624769302159834,
      "loss": 6.0108,
      "mean_token_accuracy": 0.14659083187580108,
      "num_tokens": 39481138.0,
      "step": 18240
    },
    {
      "entropy": 6.271596002578735,
      "epoch": 1.952431911819787,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004624557873161755,
      "loss": 5.9657,
      "mean_token_accuracy": 0.15328457653522493,
      "num_tokens": 39492066.0,
      "step": 18245
    },
    {
      "entropy": 6.381879568099976,
      "epoch": 1.952966985927551,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004624346390036081,
      "loss": 6.0525,
      "mean_token_accuracy": 0.14412908926606177,
      "num_tokens": 39502576.0,
      "step": 18250
    },
    {
      "entropy": 6.277699613571167,
      "epoch": 1.9535020600353148,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004624134852788918,
      "loss": 5.9832,
      "mean_token_accuracy": 0.15188184827566148,
      "num_tokens": 39513373.0,
      "step": 18255
    },
    {
      "entropy": 6.3183245182037355,
      "epoch": 1.954037134143079,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004623923261426374,
      "loss": 6.0806,
      "mean_token_accuracy": 0.14860584884881972,
      "num_tokens": 39523970.0,
      "step": 18260
    },
    {
      "entropy": 6.325642156600952,
      "epoch": 1.9545722082508428,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000462371161595456,
      "loss": 6.0388,
      "mean_token_accuracy": 0.1458660513162613,
      "num_tokens": 39533893.0,
      "step": 18265
    },
    {
      "entropy": 6.294271993637085,
      "epoch": 1.9551072823586066,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004623499916379586,
      "loss": 5.9936,
      "mean_token_accuracy": 0.1464106485247612,
      "num_tokens": 39544276.0,
      "step": 18270
    },
    {
      "entropy": 6.320238256454468,
      "epoch": 1.9556423564663707,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004623288162707565,
      "loss": 5.9876,
      "mean_token_accuracy": 0.15372352972626685,
      "num_tokens": 39554443.0,
      "step": 18275
    },
    {
      "entropy": 6.279079008102417,
      "epoch": 1.9561774305741344,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046230763549446116,
      "loss": 5.9495,
      "mean_token_accuracy": 0.15218204110860825,
      "num_tokens": 39564508.0,
      "step": 18280
    },
    {
      "entropy": 6.306404781341553,
      "epoch": 1.9567125046818985,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004622864493096842,
      "loss": 6.0353,
      "mean_token_accuracy": 0.14407777041196823,
      "num_tokens": 39574757.0,
      "step": 18285
    },
    {
      "entropy": 6.3766943454742435,
      "epoch": 1.9572475787896624,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046226525771703733,
      "loss": 6.1135,
      "mean_token_accuracy": 0.146172334253788,
      "num_tokens": 39585911.0,
      "step": 18290
    },
    {
      "entropy": 6.363869476318359,
      "epoch": 1.9577826528974263,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004622440607171324,
      "loss": 6.0015,
      "mean_token_accuracy": 0.14791649281978608,
      "num_tokens": 39597443.0,
      "step": 18295
    },
    {
      "entropy": 6.318935441970825,
      "epoch": 1.9583177270051904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046222285831058155,
      "loss": 6.074,
      "mean_token_accuracy": 0.14525766149163247,
      "num_tokens": 39608693.0,
      "step": 18300
    },
    {
      "entropy": 6.279829454421997,
      "epoch": 1.958852801112954,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000462201650497997,
      "loss": 6.056,
      "mean_token_accuracy": 0.14851330816745759,
      "num_tokens": 39620002.0,
      "step": 18305
    },
    {
      "entropy": 6.27910966873169,
      "epoch": 1.9593878752207181,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00046218043727999105,
      "loss": 6.0235,
      "mean_token_accuracy": 0.1526221178472042,
      "num_tokens": 39629856.0,
      "step": 18310
    },
    {
      "entropy": 6.24758939743042,
      "epoch": 1.959922949328482,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046215921865717637,
      "loss": 5.9843,
      "mean_token_accuracy": 0.15295154899358748,
      "num_tokens": 39640418.0,
      "step": 18315
    },
    {
      "entropy": 6.390838289260865,
      "epoch": 1.9604580234362459,
      "grad_norm": 2.25,
      "learning_rate": 0.0004621379946301655,
      "loss": 6.1042,
      "mean_token_accuracy": 0.139867602288723,
      "num_tokens": 39652871.0,
      "step": 18320
    },
    {
      "entropy": 6.339018392562866,
      "epoch": 1.96099309754401,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004621167651995713,
      "loss": 6.0175,
      "mean_token_accuracy": 0.1491739921271801,
      "num_tokens": 39663864.0,
      "step": 18325
    },
    {
      "entropy": 6.198511219024658,
      "epoch": 1.9615281716517736,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004620955303660068,
      "loss": 6.0131,
      "mean_token_accuracy": 0.1545386165380478,
      "num_tokens": 39675835.0,
      "step": 18330
    },
    {
      "entropy": 6.334716892242431,
      "epoch": 1.9620632457595377,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046207429013008525,
      "loss": 6.0464,
      "mean_token_accuracy": 0.14758169651031494,
      "num_tokens": 39686799.0,
      "step": 18335
    },
    {
      "entropy": 6.379144525527954,
      "epoch": 1.9625983198673016,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046205304449241976,
      "loss": 5.9958,
      "mean_token_accuracy": 0.15189558565616607,
      "num_tokens": 39697564.0,
      "step": 18340
    },
    {
      "entropy": 6.299245405197143,
      "epoch": 1.9631333939750655,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004620317934536239,
      "loss": 6.0291,
      "mean_token_accuracy": 0.14984162598848344,
      "num_tokens": 39708064.0,
      "step": 18345
    },
    {
      "entropy": 6.216150093078613,
      "epoch": 1.9636684680828296,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004620105370143112,
      "loss": 5.8979,
      "mean_token_accuracy": 0.1574153944849968,
      "num_tokens": 39718076.0,
      "step": 18350
    },
    {
      "entropy": 6.295418930053711,
      "epoch": 1.9642035421905932,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004619892751750957,
      "loss": 5.9991,
      "mean_token_accuracy": 0.14951959028840064,
      "num_tokens": 39729532.0,
      "step": 18355
    },
    {
      "entropy": 6.376259756088257,
      "epoch": 1.9647386162983573,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046196800793659116,
      "loss": 6.0066,
      "mean_token_accuracy": 0.14936141222715377,
      "num_tokens": 39740372.0,
      "step": 18360
    },
    {
      "entropy": 6.267642450332642,
      "epoch": 1.9652736904061212,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004619467352994115,
      "loss": 5.8847,
      "mean_token_accuracy": 0.1573383167386055,
      "num_tokens": 39751283.0,
      "step": 18365
    },
    {
      "entropy": 6.241953372955322,
      "epoch": 1.965808764513885,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004619254572641712,
      "loss": 6.101,
      "mean_token_accuracy": 0.14580308496952057,
      "num_tokens": 39763295.0,
      "step": 18370
    },
    {
      "entropy": 6.292729616165161,
      "epoch": 1.9663438386216492,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046190417383148465,
      "loss": 6.0307,
      "mean_token_accuracy": 0.14641427397727966,
      "num_tokens": 39773849.0,
      "step": 18375
    },
    {
      "entropy": 6.269519233703614,
      "epoch": 1.966878912729413,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004618828850019663,
      "loss": 5.8426,
      "mean_token_accuracy": 0.1593016840517521,
      "num_tokens": 39784366.0,
      "step": 18380
    },
    {
      "entropy": 6.299762153625489,
      "epoch": 1.967413986837177,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046186159077623093,
      "loss": 5.9565,
      "mean_token_accuracy": 0.1473100945353508,
      "num_tokens": 39794226.0,
      "step": 18385
    },
    {
      "entropy": 6.201114273071289,
      "epoch": 1.9679490609449408,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004618402911548933,
      "loss": 5.9912,
      "mean_token_accuracy": 0.150033912807703,
      "num_tokens": 39805557.0,
      "step": 18390
    },
    {
      "entropy": 6.302155494689941,
      "epoch": 1.9684841350527047,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004618189861385686,
      "loss": 5.9598,
      "mean_token_accuracy": 0.15425646156072617,
      "num_tokens": 39816090.0,
      "step": 18395
    },
    {
      "entropy": 6.322425842285156,
      "epoch": 1.9690192091604688,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004617976757278719,
      "loss": 6.0636,
      "mean_token_accuracy": 0.14082781448960305,
      "num_tokens": 39826662.0,
      "step": 18400
    },
    {
      "entropy": 6.359599494934082,
      "epoch": 1.9695542832682327,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046177635992341855,
      "loss": 6.0609,
      "mean_token_accuracy": 0.1449219100177288,
      "num_tokens": 39837805.0,
      "step": 18405
    },
    {
      "entropy": 6.369721364974976,
      "epoch": 1.9700893573759966,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004617550387258241,
      "loss": 6.0374,
      "mean_token_accuracy": 0.1404178373515606,
      "num_tokens": 39848052.0,
      "step": 18410
    },
    {
      "entropy": 6.3376538276672365,
      "epoch": 1.9706244314837607,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004617337121357041,
      "loss": 6.035,
      "mean_token_accuracy": 0.14509089440107345,
      "num_tokens": 39859448.0,
      "step": 18415
    },
    {
      "entropy": 6.317292642593384,
      "epoch": 1.9711595055915243,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004617123801536744,
      "loss": 5.9818,
      "mean_token_accuracy": 0.15044161528348923,
      "num_tokens": 39870053.0,
      "step": 18420
    },
    {
      "entropy": 6.3356259822845455,
      "epoch": 1.9716945796992884,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000461691042780351,
      "loss": 6.1164,
      "mean_token_accuracy": 0.14240164682269096,
      "num_tokens": 39881547.0,
      "step": 18425
    },
    {
      "entropy": 6.3515886783599855,
      "epoch": 1.9722296538070523,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046166970001635,
      "loss": 6.0827,
      "mean_token_accuracy": 0.14907654747366905,
      "num_tokens": 39892974.0,
      "step": 18430
    },
    {
      "entropy": 6.367896032333374,
      "epoch": 1.9727647279148162,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046164835186228757,
      "loss": 6.0616,
      "mean_token_accuracy": 0.15225062668323516,
      "num_tokens": 39904210.0,
      "step": 18435
    },
    {
      "entropy": 6.26483473777771,
      "epoch": 1.9732998020225803,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046162699831878026,
      "loss": 6.0022,
      "mean_token_accuracy": 0.1513315573334694,
      "num_tokens": 39915464.0,
      "step": 18440
    },
    {
      "entropy": 6.28617582321167,
      "epoch": 1.973834876130344,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004616056393864446,
      "loss": 6.0364,
      "mean_token_accuracy": 0.14931846559047698,
      "num_tokens": 39927117.0,
      "step": 18445
    },
    {
      "entropy": 6.356991910934449,
      "epoch": 1.974369950238108,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004615842750658972,
      "loss": 6.0268,
      "mean_token_accuracy": 0.15534366369247438,
      "num_tokens": 39937363.0,
      "step": 18450
    },
    {
      "entropy": 6.225720453262329,
      "epoch": 1.974905024345872,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004615629053577552,
      "loss": 5.9208,
      "mean_token_accuracy": 0.1573420897126198,
      "num_tokens": 39948934.0,
      "step": 18455
    },
    {
      "entropy": 6.26023497581482,
      "epoch": 1.9754400984536358,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004615415302626355,
      "loss": 5.972,
      "mean_token_accuracy": 0.1549454376101494,
      "num_tokens": 39959435.0,
      "step": 18460
    },
    {
      "entropy": 6.358692789077759,
      "epoch": 1.9759751725613999,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046152014978115535,
      "loss": 6.0251,
      "mean_token_accuracy": 0.1486461341381073,
      "num_tokens": 39969006.0,
      "step": 18465
    },
    {
      "entropy": 6.311913394927979,
      "epoch": 1.9765102466691635,
      "grad_norm": 1.15625,
      "learning_rate": 0.000461498763913932,
      "loss": 5.9641,
      "mean_token_accuracy": 0.1449800632894039,
      "num_tokens": 39979264.0,
      "step": 18470
    },
    {
      "entropy": 6.34872989654541,
      "epoch": 1.9770453207769276,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046147737266158314,
      "loss": 6.1025,
      "mean_token_accuracy": 0.1422770544886589,
      "num_tokens": 39989791.0,
      "step": 18475
    },
    {
      "entropy": 6.333047342300415,
      "epoch": 1.9775803948846915,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004614559760247263,
      "loss": 5.9765,
      "mean_token_accuracy": 0.15406135320663453,
      "num_tokens": 39999591.0,
      "step": 18480
    },
    {
      "entropy": 6.271540927886963,
      "epoch": 1.9781154689924554,
      "grad_norm": 1.25,
      "learning_rate": 0.0004614345740039794,
      "loss": 6.1158,
      "mean_token_accuracy": 0.14870973378419877,
      "num_tokens": 40010811.0,
      "step": 18485
    },
    {
      "entropy": 6.33726315498352,
      "epoch": 1.9786505431002195,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046141316659996037,
      "loss": 6.0501,
      "mean_token_accuracy": 0.14873145669698715,
      "num_tokens": 40022190.0,
      "step": 18490
    },
    {
      "entropy": 6.385013580322266,
      "epoch": 1.9791856172079834,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046139175381328725,
      "loss": 6.0121,
      "mean_token_accuracy": 0.14732878655195236,
      "num_tokens": 40033905.0,
      "step": 18495
    },
    {
      "entropy": 6.364525985717774,
      "epoch": 1.9797206913157472,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004613703356445786,
      "loss": 6.0861,
      "mean_token_accuracy": 0.15079369097948075,
      "num_tokens": 40044923.0,
      "step": 18500
    },
    {
      "entropy": 6.283112573623657,
      "epoch": 1.9802557654235111,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004613489120944526,
      "loss": 5.9843,
      "mean_token_accuracy": 0.1534266345202923,
      "num_tokens": 40054988.0,
      "step": 18505
    },
    {
      "entropy": 6.342667818069458,
      "epoch": 1.980790839531275,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004613274831635279,
      "loss": 6.1047,
      "mean_token_accuracy": 0.1424488678574562,
      "num_tokens": 40065565.0,
      "step": 18510
    },
    {
      "entropy": 6.347809219360352,
      "epoch": 1.981325913639039,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046130604885242333,
      "loss": 6.0677,
      "mean_token_accuracy": 0.14813309907913208,
      "num_tokens": 40076282.0,
      "step": 18515
    },
    {
      "entropy": 6.324295377731323,
      "epoch": 1.981860987746803,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004612846091617578,
      "loss": 5.9865,
      "mean_token_accuracy": 0.15455236285924911,
      "num_tokens": 40087546.0,
      "step": 18520
    },
    {
      "entropy": 6.334740877151489,
      "epoch": 1.9823960618545668,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004612631640921504,
      "loss": 6.081,
      "mean_token_accuracy": 0.14651337265968323,
      "num_tokens": 40097937.0,
      "step": 18525
    },
    {
      "entropy": 6.369153165817261,
      "epoch": 1.982931135962331,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046124171364422025,
      "loss": 6.0072,
      "mean_token_accuracy": 0.1540858894586563,
      "num_tokens": 40108639.0,
      "step": 18530
    },
    {
      "entropy": 6.277071189880371,
      "epoch": 1.9834662100700946,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004612202578185868,
      "loss": 6.028,
      "mean_token_accuracy": 0.15272970944643022,
      "num_tokens": 40119533.0,
      "step": 18535
    },
    {
      "entropy": 6.331558513641357,
      "epoch": 1.9840012841778587,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004611987966158695,
      "loss": 6.0218,
      "mean_token_accuracy": 0.15207558795809745,
      "num_tokens": 40131176.0,
      "step": 18540
    },
    {
      "entropy": 6.345031881332398,
      "epoch": 1.9845363582856226,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046117733003668825,
      "loss": 6.0106,
      "mean_token_accuracy": 0.14899663925170897,
      "num_tokens": 40143293.0,
      "step": 18545
    },
    {
      "entropy": 6.343554401397705,
      "epoch": 1.9850714323933865,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004611558580816628,
      "loss": 5.9825,
      "mean_token_accuracy": 0.14394623264670373,
      "num_tokens": 40155312.0,
      "step": 18550
    },
    {
      "entropy": 6.313241481781006,
      "epoch": 1.9856065065011506,
      "grad_norm": 1.1875,
      "learning_rate": 0.000461134380751413,
      "loss": 6.0221,
      "mean_token_accuracy": 0.1522231414914131,
      "num_tokens": 40166261.0,
      "step": 18555
    },
    {
      "entropy": 6.229114294052124,
      "epoch": 1.9861415806089142,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046111289804655906,
      "loss": 5.9556,
      "mean_token_accuracy": 0.15715578198432922,
      "num_tokens": 40176641.0,
      "step": 18560
    },
    {
      "entropy": 6.302442026138306,
      "epoch": 1.9866766547166783,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004610914099677214,
      "loss": 6.0505,
      "mean_token_accuracy": 0.14907757341861724,
      "num_tokens": 40187784.0,
      "step": 18565
    },
    {
      "entropy": 6.30404372215271,
      "epoch": 1.9872117288244422,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004610699165155204,
      "loss": 6.0383,
      "mean_token_accuracy": 0.14327414333820343,
      "num_tokens": 40198219.0,
      "step": 18570
    },
    {
      "entropy": 6.271808528900147,
      "epoch": 1.987746802932206,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046104841769057677,
      "loss": 6.0053,
      "mean_token_accuracy": 0.15207861810922624,
      "num_tokens": 40208589.0,
      "step": 18575
    },
    {
      "entropy": 6.3757813453674315,
      "epoch": 1.9882818770399702,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004610269134935112,
      "loss": 5.9876,
      "mean_token_accuracy": 0.15268800109624864,
      "num_tokens": 40219587.0,
      "step": 18580
    },
    {
      "entropy": 6.316678237915039,
      "epoch": 1.9888169511477338,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004610054039249446,
      "loss": 5.996,
      "mean_token_accuracy": 0.15284471362829208,
      "num_tokens": 40230681.0,
      "step": 18585
    },
    {
      "entropy": 6.188679265975952,
      "epoch": 1.989352025255498,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046098388898549816,
      "loss": 5.9512,
      "mean_token_accuracy": 0.157837675511837,
      "num_tokens": 40241356.0,
      "step": 18590
    },
    {
      "entropy": 6.312573432922363,
      "epoch": 1.9898870993632618,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046096236867579306,
      "loss": 6.0013,
      "mean_token_accuracy": 0.1525126203894615,
      "num_tokens": 40251652.0,
      "step": 18595
    },
    {
      "entropy": 6.308320569992065,
      "epoch": 1.9904221734710257,
      "grad_norm": 1.375,
      "learning_rate": 0.00046094084299645065,
      "loss": 5.9768,
      "mean_token_accuracy": 0.15515742003917693,
      "num_tokens": 40262819.0,
      "step": 18600
    },
    {
      "entropy": 6.3292866230010985,
      "epoch": 1.9909572475787898,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046091931194809253,
      "loss": 6.1043,
      "mean_token_accuracy": 0.14111527651548386,
      "num_tokens": 40274314.0,
      "step": 18605
    },
    {
      "entropy": 6.367870903015136,
      "epoch": 1.9914923216865534,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004608977755313404,
      "loss": 6.0399,
      "mean_token_accuracy": 0.14789102971553802,
      "num_tokens": 40284733.0,
      "step": 18610
    },
    {
      "entropy": 6.294833993911743,
      "epoch": 1.9920273957943175,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004608762337468162,
      "loss": 6.0028,
      "mean_token_accuracy": 0.15236538425087928,
      "num_tokens": 40295484.0,
      "step": 18615
    },
    {
      "entropy": 6.3287547588348385,
      "epoch": 1.9925624699020814,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004608546865951418,
      "loss": 5.9936,
      "mean_token_accuracy": 0.1492601178586483,
      "num_tokens": 40305470.0,
      "step": 18620
    },
    {
      "entropy": 6.336968946456909,
      "epoch": 1.9930975440098453,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046083313407693945,
      "loss": 6.1229,
      "mean_token_accuracy": 0.1426390901207924,
      "num_tokens": 40316055.0,
      "step": 18625
    },
    {
      "entropy": 6.305480051040649,
      "epoch": 1.9936326181176094,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004608115761928315,
      "loss": 6.0052,
      "mean_token_accuracy": 0.15096798837184905,
      "num_tokens": 40326689.0,
      "step": 18630
    },
    {
      "entropy": 6.2536698341369625,
      "epoch": 1.9941676922253733,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046079001294344036,
      "loss": 6.0283,
      "mean_token_accuracy": 0.14815376400947572,
      "num_tokens": 40338511.0,
      "step": 18635
    },
    {
      "entropy": 6.281500196456909,
      "epoch": 1.9947027663331371,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004607684443293887,
      "loss": 6.0001,
      "mean_token_accuracy": 0.14639228880405425,
      "num_tokens": 40350395.0,
      "step": 18640
    },
    {
      "entropy": 6.322163009643555,
      "epoch": 1.995237840440901,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004607468703512993,
      "loss": 5.9888,
      "mean_token_accuracy": 0.15702394247055054,
      "num_tokens": 40360166.0,
      "step": 18645
    },
    {
      "entropy": 6.278965187072754,
      "epoch": 1.995772914548665,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046072529100979514,
      "loss": 6.0204,
      "mean_token_accuracy": 0.14643388539552687,
      "num_tokens": 40370655.0,
      "step": 18650
    },
    {
      "entropy": 6.324419069290161,
      "epoch": 1.996307988656429,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004607037063054993,
      "loss": 5.9881,
      "mean_token_accuracy": 0.15432731062173843,
      "num_tokens": 40381502.0,
      "step": 18655
    },
    {
      "entropy": 6.391924476623535,
      "epoch": 1.9968430627641929,
      "grad_norm": 1.1875,
      "learning_rate": 0.000460682116239035,
      "loss": 6.0788,
      "mean_token_accuracy": 0.14124054461717606,
      "num_tokens": 40391815.0,
      "step": 18660
    },
    {
      "entropy": 6.322268342971801,
      "epoch": 1.9973781368719568,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046066052081102576,
      "loss": 6.0523,
      "mean_token_accuracy": 0.14884606152772903,
      "num_tokens": 40403306.0,
      "step": 18665
    },
    {
      "entropy": 6.319442319869995,
      "epoch": 1.9979132109797209,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046063892002209505,
      "loss": 5.9653,
      "mean_token_accuracy": 0.1569172039628029,
      "num_tokens": 40414210.0,
      "step": 18670
    },
    {
      "entropy": 6.215680313110352,
      "epoch": 1.9984482850874845,
      "grad_norm": 1.5,
      "learning_rate": 0.0004606173138728666,
      "loss": 5.9476,
      "mean_token_accuracy": 0.15986442118883132,
      "num_tokens": 40424368.0,
      "step": 18675
    },
    {
      "entropy": 6.282414531707763,
      "epoch": 1.9989833591952486,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004605957023639643,
      "loss": 6.0279,
      "mean_token_accuracy": 0.14785999655723572,
      "num_tokens": 40436252.0,
      "step": 18680
    },
    {
      "entropy": 6.331739282608032,
      "epoch": 1.9995184333030125,
      "grad_norm": 1.125,
      "learning_rate": 0.00046057408549601223,
      "loss": 6.0451,
      "mean_token_accuracy": 0.15290766060352326,
      "num_tokens": 40447256.0,
      "step": 18685
    },
    {
      "entropy": 6.312280813852946,
      "epoch": 2.0,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004605524632696344,
      "loss": 5.9846,
      "mean_token_accuracy": 0.1511562036143409,
      "num_tokens": 40456516.0,
      "step": 18690
    },
    {
      "entropy": 6.3112891674041744,
      "epoch": 2.000535074107764,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004605308356854553,
      "loss": 5.9033,
      "mean_token_accuracy": 0.14962349683046341,
      "num_tokens": 40467563.0,
      "step": 18695
    },
    {
      "entropy": 6.3000099658966064,
      "epoch": 2.0010701482155278,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004605092027440994,
      "loss": 5.9649,
      "mean_token_accuracy": 0.15855505168437958,
      "num_tokens": 40478202.0,
      "step": 18700
    },
    {
      "entropy": 6.322792863845825,
      "epoch": 2.001605222323292,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004604875644461914,
      "loss": 5.9132,
      "mean_token_accuracy": 0.15457970947027205,
      "num_tokens": 40489677.0,
      "step": 18705
    },
    {
      "entropy": 6.310722827911377,
      "epoch": 2.0021402964310555,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046046592079235595,
      "loss": 5.9405,
      "mean_token_accuracy": 0.15483094304800032,
      "num_tokens": 40500694.0,
      "step": 18710
    },
    {
      "entropy": 6.231298446655273,
      "epoch": 2.0026753705388196,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004604442717832181,
      "loss": 5.9159,
      "mean_token_accuracy": 0.1547863930463791,
      "num_tokens": 40511539.0,
      "step": 18715
    },
    {
      "entropy": 6.224448585510254,
      "epoch": 2.0032104446465837,
      "grad_norm": 1.40625,
      "learning_rate": 0.000460422617419403,
      "loss": 5.8819,
      "mean_token_accuracy": 0.15421521961688994,
      "num_tokens": 40521906.0,
      "step": 18720
    },
    {
      "entropy": 6.350541257858277,
      "epoch": 2.0037455187543474,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046040095770153594,
      "loss": 5.9406,
      "mean_token_accuracy": 0.15019365549087524,
      "num_tokens": 40531454.0,
      "step": 18725
    },
    {
      "entropy": 6.376023387908935,
      "epoch": 2.0042805928621115,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004603792926302422,
      "loss": 5.9773,
      "mean_token_accuracy": 0.14757483154535295,
      "num_tokens": 40541497.0,
      "step": 18730
    },
    {
      "entropy": 6.299203109741211,
      "epoch": 2.004815666969875,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004603576222061475,
      "loss": 5.9878,
      "mean_token_accuracy": 0.15108438581228256,
      "num_tokens": 40552576.0,
      "step": 18735
    },
    {
      "entropy": 6.288552904129029,
      "epoch": 2.005350741077639,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004603359464298774,
      "loss": 5.9304,
      "mean_token_accuracy": 0.15092037469148636,
      "num_tokens": 40562906.0,
      "step": 18740
    },
    {
      "entropy": 6.268268728256226,
      "epoch": 2.0058858151854033,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046031426530205796,
      "loss": 5.8617,
      "mean_token_accuracy": 0.15695815533399582,
      "num_tokens": 40574510.0,
      "step": 18745
    },
    {
      "entropy": 6.243146562576294,
      "epoch": 2.006420889293167,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004602925788233152,
      "loss": 5.8993,
      "mean_token_accuracy": 0.16641436666250228,
      "num_tokens": 40586170.0,
      "step": 18750
    },
    {
      "entropy": 6.33104133605957,
      "epoch": 2.006955963400931,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046027088699427516,
      "loss": 5.9436,
      "mean_token_accuracy": 0.14635249152779578,
      "num_tokens": 40597613.0,
      "step": 18755
    },
    {
      "entropy": 6.360899162292481,
      "epoch": 2.007491037508695,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046024918981556434,
      "loss": 5.9889,
      "mean_token_accuracy": 0.14945476800203322,
      "num_tokens": 40608366.0,
      "step": 18760
    },
    {
      "entropy": 6.26560115814209,
      "epoch": 2.008026111616459,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004602274872878092,
      "loss": 5.931,
      "mean_token_accuracy": 0.16038887947797775,
      "num_tokens": 40618616.0,
      "step": 18765
    },
    {
      "entropy": 6.211123895645142,
      "epoch": 2.008561185724223,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046020577941163635,
      "loss": 5.8111,
      "mean_token_accuracy": 0.16040822118520737,
      "num_tokens": 40628342.0,
      "step": 18770
    },
    {
      "entropy": 6.277521419525146,
      "epoch": 2.0090962598319866,
      "grad_norm": 1.25,
      "learning_rate": 0.00046018406618767273,
      "loss": 5.8961,
      "mean_token_accuracy": 0.15277449116110803,
      "num_tokens": 40638644.0,
      "step": 18775
    },
    {
      "entropy": 6.274780750274658,
      "epoch": 2.0096313339397507,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004601623476165451,
      "loss": 5.8945,
      "mean_token_accuracy": 0.15727217644453048,
      "num_tokens": 40649945.0,
      "step": 18780
    },
    {
      "entropy": 6.383197212219239,
      "epoch": 2.010166408047515,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046014062369888084,
      "loss": 6.0349,
      "mean_token_accuracy": 0.1469561256468296,
      "num_tokens": 40660896.0,
      "step": 18785
    },
    {
      "entropy": 6.3236359596252445,
      "epoch": 2.0107014821552784,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000460118894435307,
      "loss": 5.9919,
      "mean_token_accuracy": 0.1480495199561119,
      "num_tokens": 40672072.0,
      "step": 18790
    },
    {
      "entropy": 6.36170654296875,
      "epoch": 2.0112365562630425,
      "grad_norm": 1.25,
      "learning_rate": 0.0004600971598264511,
      "loss": 5.9948,
      "mean_token_accuracy": 0.14656387865543366,
      "num_tokens": 40682622.0,
      "step": 18795
    },
    {
      "entropy": 6.249828577041626,
      "epoch": 2.011771630370806,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004600754198729407,
      "loss": 5.9052,
      "mean_token_accuracy": 0.15835212916135788,
      "num_tokens": 40693127.0,
      "step": 18800
    },
    {
      "entropy": 6.300584554672241,
      "epoch": 2.0123067044785703,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046005367457540366,
      "loss": 5.933,
      "mean_token_accuracy": 0.155922132730484,
      "num_tokens": 40705009.0,
      "step": 18805
    },
    {
      "entropy": 6.345494890213013,
      "epoch": 2.0128417785863344,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046003192393446763,
      "loss": 5.9546,
      "mean_token_accuracy": 0.14842819124460221,
      "num_tokens": 40716498.0,
      "step": 18810
    },
    {
      "entropy": 6.302134466171265,
      "epoch": 2.013376852694098,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004600101679507609,
      "loss": 5.941,
      "mean_token_accuracy": 0.15621695518493653,
      "num_tokens": 40727802.0,
      "step": 18815
    },
    {
      "entropy": 6.345720052719116,
      "epoch": 2.013911926801862,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004599884066249115,
      "loss": 5.9665,
      "mean_token_accuracy": 0.15371165573596954,
      "num_tokens": 40738866.0,
      "step": 18820
    },
    {
      "entropy": 6.298053026199341,
      "epoch": 2.014447000909626,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004599666399575479,
      "loss": 5.9424,
      "mean_token_accuracy": 0.1486460819840431,
      "num_tokens": 40750214.0,
      "step": 18825
    },
    {
      "entropy": 6.320457220077515,
      "epoch": 2.01498207501739,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004599448679492986,
      "loss": 5.9345,
      "mean_token_accuracy": 0.15878672748804093,
      "num_tokens": 40759830.0,
      "step": 18830
    },
    {
      "entropy": 6.28009524345398,
      "epoch": 2.015517149125154,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004599230906007921,
      "loss": 5.9877,
      "mean_token_accuracy": 0.14790922105312349,
      "num_tokens": 40770007.0,
      "step": 18835
    },
    {
      "entropy": 6.296482992172241,
      "epoch": 2.0160522232329177,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004599013079126575,
      "loss": 5.8985,
      "mean_token_accuracy": 0.15052998214960098,
      "num_tokens": 40781495.0,
      "step": 18840
    },
    {
      "entropy": 6.302435302734375,
      "epoch": 2.0165872973406818,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004598795198855235,
      "loss": 5.899,
      "mean_token_accuracy": 0.15590118616819382,
      "num_tokens": 40792875.0,
      "step": 18845
    },
    {
      "entropy": 6.313598775863648,
      "epoch": 2.0171223714484454,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004598577265200193,
      "loss": 5.9994,
      "mean_token_accuracy": 0.14166968315839767,
      "num_tokens": 40804484.0,
      "step": 18850
    },
    {
      "entropy": 6.26287612915039,
      "epoch": 2.0176574455562095,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004598359278167744,
      "loss": 5.9086,
      "mean_token_accuracy": 0.15280026197433472,
      "num_tokens": 40814692.0,
      "step": 18855
    },
    {
      "entropy": 6.284548759460449,
      "epoch": 2.0181925196639736,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045981412377641795,
      "loss": 5.8879,
      "mean_token_accuracy": 0.1565020576119423,
      "num_tokens": 40825443.0,
      "step": 18860
    },
    {
      "entropy": 6.3230140209198,
      "epoch": 2.0187275937717373,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045979231439957966,
      "loss": 5.9469,
      "mean_token_accuracy": 0.15315036475658417,
      "num_tokens": 40836699.0,
      "step": 18865
    },
    {
      "entropy": 6.337334251403808,
      "epoch": 2.0192626678795014,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004597704996868893,
      "loss": 5.9276,
      "mean_token_accuracy": 0.150046506524086,
      "num_tokens": 40846736.0,
      "step": 18870
    },
    {
      "entropy": 6.375490522384643,
      "epoch": 2.019797741987265,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004597486796389766,
      "loss": 6.0162,
      "mean_token_accuracy": 0.15102373510599137,
      "num_tokens": 40858362.0,
      "step": 18875
    },
    {
      "entropy": 6.3270505428314205,
      "epoch": 2.020332816095029,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045972685425647186,
      "loss": 5.9305,
      "mean_token_accuracy": 0.15315622091293335,
      "num_tokens": 40869346.0,
      "step": 18880
    },
    {
      "entropy": 6.361948871612549,
      "epoch": 2.0208678902027932,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004597050235400051,
      "loss": 6.0181,
      "mean_token_accuracy": 0.1515903055667877,
      "num_tokens": 40880003.0,
      "step": 18885
    },
    {
      "entropy": 6.360273218154907,
      "epoch": 2.021402964310557,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045968318749020675,
      "loss": 5.9474,
      "mean_token_accuracy": 0.1511355981230736,
      "num_tokens": 40891515.0,
      "step": 18890
    },
    {
      "entropy": 6.3098972797393795,
      "epoch": 2.021938038418321,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004596613461077073,
      "loss": 5.9107,
      "mean_token_accuracy": 0.15185776650905608,
      "num_tokens": 40902024.0,
      "step": 18895
    },
    {
      "entropy": 6.24690728187561,
      "epoch": 2.022473112526085,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045963949939313754,
      "loss": 5.9023,
      "mean_token_accuracy": 0.16064081639051436,
      "num_tokens": 40912160.0,
      "step": 18900
    },
    {
      "entropy": 6.337373495101929,
      "epoch": 2.0230081866338487,
      "grad_norm": 1.09375,
      "learning_rate": 0.000459617647347128,
      "loss": 6.0006,
      "mean_token_accuracy": 0.14154519885778427,
      "num_tokens": 40922516.0,
      "step": 18905
    },
    {
      "entropy": 6.366936302185058,
      "epoch": 2.023543260741613,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045959578997031,
      "loss": 5.9865,
      "mean_token_accuracy": 0.14919121861457824,
      "num_tokens": 40934148.0,
      "step": 18910
    },
    {
      "entropy": 6.388768005371094,
      "epoch": 2.0240783348493765,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045957392726331443,
      "loss": 5.9137,
      "mean_token_accuracy": 0.15576138645410537,
      "num_tokens": 40944409.0,
      "step": 18915
    },
    {
      "entropy": 6.303894472122193,
      "epoch": 2.0246134089571406,
      "grad_norm": 1.5625,
      "learning_rate": 0.00045955205922677264,
      "loss": 5.9683,
      "mean_token_accuracy": 0.1498681992292404,
      "num_tokens": 40955133.0,
      "step": 18920
    },
    {
      "entropy": 6.342651462554931,
      "epoch": 2.0251484830649047,
      "grad_norm": 1.375,
      "learning_rate": 0.0004595301858613161,
      "loss": 6.0276,
      "mean_token_accuracy": 0.15437590181827546,
      "num_tokens": 40965508.0,
      "step": 18925
    },
    {
      "entropy": 6.411085987091065,
      "epoch": 2.0256835571726683,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004595083071675763,
      "loss": 6.0123,
      "mean_token_accuracy": 0.156063212454319,
      "num_tokens": 40975931.0,
      "step": 18930
    },
    {
      "entropy": 6.344521141052246,
      "epoch": 2.0262186312804324,
      "grad_norm": 1.125,
      "learning_rate": 0.00045948642314618514,
      "loss": 5.9595,
      "mean_token_accuracy": 0.14978968650102614,
      "num_tokens": 40985797.0,
      "step": 18935
    },
    {
      "entropy": 6.267924499511719,
      "epoch": 2.026753705388196,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004594645337977744,
      "loss": 5.9549,
      "mean_token_accuracy": 0.15235372483730317,
      "num_tokens": 40996501.0,
      "step": 18940
    },
    {
      "entropy": 6.296322202682495,
      "epoch": 2.02728877949596,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004594426391229762,
      "loss": 5.921,
      "mean_token_accuracy": 0.1528517037630081,
      "num_tokens": 41005970.0,
      "step": 18945
    },
    {
      "entropy": 6.298361873626709,
      "epoch": 2.0278238536037243,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004594207391224227,
      "loss": 5.9296,
      "mean_token_accuracy": 0.14762877002358438,
      "num_tokens": 41017208.0,
      "step": 18950
    },
    {
      "entropy": 6.377904415130615,
      "epoch": 2.028358927711488,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045939883379674624,
      "loss": 6.0911,
      "mean_token_accuracy": 0.1463048666715622,
      "num_tokens": 41028435.0,
      "step": 18955
    },
    {
      "entropy": 6.2913330554962155,
      "epoch": 2.028894001819252,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045937692314657936,
      "loss": 5.9245,
      "mean_token_accuracy": 0.15806464552879335,
      "num_tokens": 41039256.0,
      "step": 18960
    },
    {
      "entropy": 6.257915115356445,
      "epoch": 2.0294290759270157,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004593550071725548,
      "loss": 5.9399,
      "mean_token_accuracy": 0.15038865208625793,
      "num_tokens": 41050527.0,
      "step": 18965
    },
    {
      "entropy": 6.3094432830810545,
      "epoch": 2.02996415003478,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045933308587530527,
      "loss": 5.9889,
      "mean_token_accuracy": 0.1501224860548973,
      "num_tokens": 41062259.0,
      "step": 18970
    },
    {
      "entropy": 6.317587757110596,
      "epoch": 2.030499224142544,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004593111592554638,
      "loss": 5.9926,
      "mean_token_accuracy": 0.1491305373609066,
      "num_tokens": 41073011.0,
      "step": 18975
    },
    {
      "entropy": 6.280509519577026,
      "epoch": 2.0310342982503076,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004592892273136635,
      "loss": 5.8665,
      "mean_token_accuracy": 0.16249484270811082,
      "num_tokens": 41083482.0,
      "step": 18980
    },
    {
      "entropy": 6.30593090057373,
      "epoch": 2.0315693723580717,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00045926729005053764,
      "loss": 5.9332,
      "mean_token_accuracy": 0.15869732201099396,
      "num_tokens": 41094028.0,
      "step": 18985
    },
    {
      "entropy": 6.26506872177124,
      "epoch": 2.0321044464658353,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004592453474667197,
      "loss": 5.9154,
      "mean_token_accuracy": 0.15455030798912048,
      "num_tokens": 41106346.0,
      "step": 18990
    },
    {
      "entropy": 6.307776689529419,
      "epoch": 2.0326395205735994,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004592233995628432,
      "loss": 5.8454,
      "mean_token_accuracy": 0.1541196435689926,
      "num_tokens": 41116130.0,
      "step": 18995
    },
    {
      "entropy": 6.276697635650635,
      "epoch": 2.0331745946813635,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000459201446339542,
      "loss": 5.8835,
      "mean_token_accuracy": 0.1549851566553116,
      "num_tokens": 41127103.0,
      "step": 19000
    },
    {
      "entropy": 6.244321727752686,
      "epoch": 2.033709668789127,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004591794877974499,
      "loss": 5.9256,
      "mean_token_accuracy": 0.1483281970024109,
      "num_tokens": 41138258.0,
      "step": 19005
    },
    {
      "entropy": 6.315855646133423,
      "epoch": 2.0342447428968913,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045915752393720094,
      "loss": 5.9613,
      "mean_token_accuracy": 0.15780217796564103,
      "num_tokens": 41148859.0,
      "step": 19010
    },
    {
      "entropy": 6.31057825088501,
      "epoch": 2.034779817004655,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045913555475942937,
      "loss": 5.8765,
      "mean_token_accuracy": 0.14937480092048644,
      "num_tokens": 41158614.0,
      "step": 19015
    },
    {
      "entropy": 6.23160400390625,
      "epoch": 2.035314891112419,
      "grad_norm": 1.25,
      "learning_rate": 0.0004591135802647695,
      "loss": 5.9184,
      "mean_token_accuracy": 0.1637422949075699,
      "num_tokens": 41170215.0,
      "step": 19020
    },
    {
      "entropy": 6.302189683914184,
      "epoch": 2.035849965220183,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004590916004538559,
      "loss": 5.9297,
      "mean_token_accuracy": 0.15089219883084298,
      "num_tokens": 41180784.0,
      "step": 19025
    },
    {
      "entropy": 6.327946710586548,
      "epoch": 2.036385039327947,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045906961532732316,
      "loss": 5.9605,
      "mean_token_accuracy": 0.15156181305646896,
      "num_tokens": 41191305.0,
      "step": 19030
    },
    {
      "entropy": 6.294243001937867,
      "epoch": 2.036920113435711,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004590476248858062,
      "loss": 5.9809,
      "mean_token_accuracy": 0.14743388891220094,
      "num_tokens": 41202388.0,
      "step": 19035
    },
    {
      "entropy": 6.246057653427124,
      "epoch": 2.037455187543475,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004590256291299399,
      "loss": 5.9047,
      "mean_token_accuracy": 0.15824347734451294,
      "num_tokens": 41212136.0,
      "step": 19040
    },
    {
      "entropy": 6.28766622543335,
      "epoch": 2.0379902616512386,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045900362806035945,
      "loss": 5.9505,
      "mean_token_accuracy": 0.15329123809933662,
      "num_tokens": 41223390.0,
      "step": 19045
    },
    {
      "entropy": 6.2713922500610355,
      "epoch": 2.0385253357590027,
      "grad_norm": 1.25,
      "learning_rate": 0.00045898162167770006,
      "loss": 5.9797,
      "mean_token_accuracy": 0.15683652609586715,
      "num_tokens": 41234775.0,
      "step": 19050
    },
    {
      "entropy": 6.310992431640625,
      "epoch": 2.0390604098667664,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004589596099825973,
      "loss": 5.8979,
      "mean_token_accuracy": 0.15252700001001357,
      "num_tokens": 41245627.0,
      "step": 19055
    },
    {
      "entropy": 6.260842227935791,
      "epoch": 2.0395954839745305,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045893759297568654,
      "loss": 5.9592,
      "mean_token_accuracy": 0.15308903455734252,
      "num_tokens": 41255441.0,
      "step": 19060
    },
    {
      "entropy": 6.193680620193481,
      "epoch": 2.0401305580822946,
      "grad_norm": 1.25,
      "learning_rate": 0.0004589155706576036,
      "loss": 5.8883,
      "mean_token_accuracy": 0.15422878861427308,
      "num_tokens": 41268686.0,
      "step": 19065
    },
    {
      "entropy": 6.374952554702759,
      "epoch": 2.0406656321900583,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045889354302898445,
      "loss": 6.0147,
      "mean_token_accuracy": 0.1451498545706272,
      "num_tokens": 41279384.0,
      "step": 19070
    },
    {
      "entropy": 6.400190448760986,
      "epoch": 2.0412007062978224,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004588715100904651,
      "loss": 6.0207,
      "mean_token_accuracy": 0.14338729083538054,
      "num_tokens": 41289923.0,
      "step": 19075
    },
    {
      "entropy": 6.277850008010864,
      "epoch": 2.041735780405586,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004588494718426818,
      "loss": 5.9166,
      "mean_token_accuracy": 0.1557366281747818,
      "num_tokens": 41300486.0,
      "step": 19080
    },
    {
      "entropy": 6.197900676727295,
      "epoch": 2.04227085451335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004588274282862708,
      "loss": 5.924,
      "mean_token_accuracy": 0.15858399122953415,
      "num_tokens": 41310507.0,
      "step": 19085
    },
    {
      "entropy": 6.281149005889892,
      "epoch": 2.042805928621114,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045880537942186855,
      "loss": 5.9458,
      "mean_token_accuracy": 0.16162229031324388,
      "num_tokens": 41321284.0,
      "step": 19090
    },
    {
      "entropy": 6.245945692062378,
      "epoch": 2.043341002728878,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004587833252501118,
      "loss": 5.8646,
      "mean_token_accuracy": 0.16073982268571854,
      "num_tokens": 41332037.0,
      "step": 19095
    },
    {
      "entropy": 6.197161769866943,
      "epoch": 2.043876076836642,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004587612657716374,
      "loss": 5.9443,
      "mean_token_accuracy": 0.15418034344911574,
      "num_tokens": 41343480.0,
      "step": 19100
    },
    {
      "entropy": 6.3141144752502445,
      "epoch": 2.0444111509444056,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004587392009870824,
      "loss": 5.9991,
      "mean_token_accuracy": 0.14724698215723037,
      "num_tokens": 41353656.0,
      "step": 19105
    },
    {
      "entropy": 6.342226314544678,
      "epoch": 2.0449462250521697,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004587171308970836,
      "loss": 5.9232,
      "mean_token_accuracy": 0.15133281722664832,
      "num_tokens": 41363640.0,
      "step": 19110
    },
    {
      "entropy": 6.314814615249634,
      "epoch": 2.045481299159934,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045869505550227857,
      "loss": 5.9639,
      "mean_token_accuracy": 0.15041980445384978,
      "num_tokens": 41374194.0,
      "step": 19115
    },
    {
      "entropy": 6.2883869171142575,
      "epoch": 2.0460163732676975,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004586729748033046,
      "loss": 5.9276,
      "mean_token_accuracy": 0.15505403578281401,
      "num_tokens": 41384968.0,
      "step": 19120
    },
    {
      "entropy": 6.269662666320801,
      "epoch": 2.0465514473754616,
      "grad_norm": 1.65625,
      "learning_rate": 0.00045865088880079924,
      "loss": 5.8144,
      "mean_token_accuracy": 0.16238731592893602,
      "num_tokens": 41395117.0,
      "step": 19125
    },
    {
      "entropy": 6.281591844558716,
      "epoch": 2.0470865214832252,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004586287974954004,
      "loss": 5.9444,
      "mean_token_accuracy": 0.15073410421609879,
      "num_tokens": 41406366.0,
      "step": 19130
    },
    {
      "entropy": 6.317609596252441,
      "epoch": 2.0476215955909893,
      "grad_norm": 1.375,
      "learning_rate": 0.0004586067008877457,
      "loss": 5.9281,
      "mean_token_accuracy": 0.15315686166286469,
      "num_tokens": 41417388.0,
      "step": 19135
    },
    {
      "entropy": 6.277280044555664,
      "epoch": 2.0481566696987534,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004585845989784734,
      "loss": 5.8915,
      "mean_token_accuracy": 0.156877364218235,
      "num_tokens": 41428085.0,
      "step": 19140
    },
    {
      "entropy": 6.342272043228149,
      "epoch": 2.048691743806517,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045856249176822155,
      "loss": 5.9915,
      "mean_token_accuracy": 0.14825173765420913,
      "num_tokens": 41439117.0,
      "step": 19145
    },
    {
      "entropy": 6.310318374633789,
      "epoch": 2.049226817914281,
      "grad_norm": 1.5,
      "learning_rate": 0.00045854037925762857,
      "loss": 5.9903,
      "mean_token_accuracy": 0.14767573177814483,
      "num_tokens": 41450042.0,
      "step": 19150
    },
    {
      "entropy": 6.293627595901489,
      "epoch": 2.049761892022045,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045851826144733296,
      "loss": 5.8864,
      "mean_token_accuracy": 0.15977600663900376,
      "num_tokens": 41460554.0,
      "step": 19155
    },
    {
      "entropy": 6.312334156036377,
      "epoch": 2.050296966129809,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045849613833797326,
      "loss": 5.9677,
      "mean_token_accuracy": 0.15383439064025878,
      "num_tokens": 41471403.0,
      "step": 19160
    },
    {
      "entropy": 6.256387233734131,
      "epoch": 2.050832040237573,
      "grad_norm": 1.125,
      "learning_rate": 0.00045847400993018835,
      "loss": 5.9556,
      "mean_token_accuracy": 0.15167587101459504,
      "num_tokens": 41482098.0,
      "step": 19165
    },
    {
      "entropy": 6.26512656211853,
      "epoch": 2.0513671143453367,
      "grad_norm": 1.375,
      "learning_rate": 0.00045845187622461724,
      "loss": 5.9272,
      "mean_token_accuracy": 0.1569909408688545,
      "num_tokens": 41493126.0,
      "step": 19170
    },
    {
      "entropy": 6.366955184936524,
      "epoch": 2.051902188453101,
      "grad_norm": 1.21875,
      "learning_rate": 0.000458429737221899,
      "loss": 6.055,
      "mean_token_accuracy": 0.14419890493154525,
      "num_tokens": 41504242.0,
      "step": 19175
    },
    {
      "entropy": 6.270892095565796,
      "epoch": 2.052437262560865,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004584075929226728,
      "loss": 5.92,
      "mean_token_accuracy": 0.15756211578845977,
      "num_tokens": 41515368.0,
      "step": 19180
    },
    {
      "entropy": 6.294704961776733,
      "epoch": 2.0529723366686286,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004583854433275782,
      "loss": 5.9554,
      "mean_token_accuracy": 0.1485486790537834,
      "num_tokens": 41525939.0,
      "step": 19185
    },
    {
      "entropy": 6.27825231552124,
      "epoch": 2.0535074107763926,
      "grad_norm": 1.671875,
      "learning_rate": 0.00045836328843725465,
      "loss": 5.979,
      "mean_token_accuracy": 0.15408600866794586,
      "num_tokens": 41536505.0,
      "step": 19190
    },
    {
      "entropy": 6.288235092163086,
      "epoch": 2.0540424848841563,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045834112825234193,
      "loss": 5.8719,
      "mean_token_accuracy": 0.1590419590473175,
      "num_tokens": 41546293.0,
      "step": 19195
    },
    {
      "entropy": 6.279095840454102,
      "epoch": 2.0545775589919204,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045831896277347985,
      "loss": 5.9228,
      "mean_token_accuracy": 0.15550676733255386,
      "num_tokens": 41556596.0,
      "step": 19200
    },
    {
      "entropy": 6.234703397750854,
      "epoch": 2.0551126330996845,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004582967920013085,
      "loss": 5.943,
      "mean_token_accuracy": 0.15806002616882325,
      "num_tokens": 41566732.0,
      "step": 19205
    },
    {
      "entropy": 6.30019326210022,
      "epoch": 2.055647707207448,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045827461593646806,
      "loss": 5.9785,
      "mean_token_accuracy": 0.14687662571668625,
      "num_tokens": 41577777.0,
      "step": 19210
    },
    {
      "entropy": 6.316343069076538,
      "epoch": 2.0561827813152123,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004582524345795989,
      "loss": 5.958,
      "mean_token_accuracy": 0.15532149523496627,
      "num_tokens": 41587978.0,
      "step": 19215
    },
    {
      "entropy": 6.283241558074951,
      "epoch": 2.056717855422976,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004582302479313413,
      "loss": 5.9071,
      "mean_token_accuracy": 0.15492736548185349,
      "num_tokens": 41598977.0,
      "step": 19220
    },
    {
      "entropy": 6.263996696472168,
      "epoch": 2.05725292953074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004582080559923362,
      "loss": 5.9577,
      "mean_token_accuracy": 0.14515373334288598,
      "num_tokens": 41609664.0,
      "step": 19225
    },
    {
      "entropy": 6.2999656200408936,
      "epoch": 2.057788003638504,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004581858587632241,
      "loss": 5.9432,
      "mean_token_accuracy": 0.15125420466065406,
      "num_tokens": 41619427.0,
      "step": 19230
    },
    {
      "entropy": 6.276629829406739,
      "epoch": 2.0583230777462678,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004581636562446461,
      "loss": 5.9371,
      "mean_token_accuracy": 0.15468958914279937,
      "num_tokens": 41629551.0,
      "step": 19235
    },
    {
      "entropy": 6.216392278671265,
      "epoch": 2.058858151854032,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004581414484372433,
      "loss": 5.9449,
      "mean_token_accuracy": 0.14881907105445863,
      "num_tokens": 41640183.0,
      "step": 19240
    },
    {
      "entropy": 6.25212254524231,
      "epoch": 2.0593932259617955,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004581192353416569,
      "loss": 5.9337,
      "mean_token_accuracy": 0.1546555995941162,
      "num_tokens": 41651934.0,
      "step": 19245
    },
    {
      "entropy": 6.20147476196289,
      "epoch": 2.0599283000695596,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004580970169585283,
      "loss": 5.8169,
      "mean_token_accuracy": 0.17226891666650773,
      "num_tokens": 41662418.0,
      "step": 19250
    },
    {
      "entropy": 6.193258333206177,
      "epoch": 2.0604633741773237,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045807479328849906,
      "loss": 5.8569,
      "mean_token_accuracy": 0.16038133352994918,
      "num_tokens": 41673419.0,
      "step": 19255
    },
    {
      "entropy": 6.195282077789306,
      "epoch": 2.0609984482850874,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045805256433221087,
      "loss": 5.8523,
      "mean_token_accuracy": 0.16209387332201003,
      "num_tokens": 41684132.0,
      "step": 19260
    },
    {
      "entropy": 6.251537418365478,
      "epoch": 2.0615335223928515,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004580303300903056,
      "loss": 5.9563,
      "mean_token_accuracy": 0.1513201877474785,
      "num_tokens": 41695051.0,
      "step": 19265
    },
    {
      "entropy": 6.312453413009644,
      "epoch": 2.062068596500615,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045800809056342534,
      "loss": 5.975,
      "mean_token_accuracy": 0.1509432926774025,
      "num_tokens": 41705986.0,
      "step": 19270
    },
    {
      "entropy": 6.306217956542969,
      "epoch": 2.0626036706083792,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00045798584575221213,
      "loss": 5.9634,
      "mean_token_accuracy": 0.15190524011850357,
      "num_tokens": 41717082.0,
      "step": 19275
    },
    {
      "entropy": 6.289776992797852,
      "epoch": 2.0631387447161433,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004579635956573084,
      "loss": 5.977,
      "mean_token_accuracy": 0.15401098281145095,
      "num_tokens": 41728783.0,
      "step": 19280
    },
    {
      "entropy": 6.257806062698364,
      "epoch": 2.063673818823907,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045794134027935646,
      "loss": 5.9712,
      "mean_token_accuracy": 0.15276308953762055,
      "num_tokens": 41740452.0,
      "step": 19285
    },
    {
      "entropy": 6.27572283744812,
      "epoch": 2.064208892931671,
      "grad_norm": 1.125,
      "learning_rate": 0.0004579190796189991,
      "loss": 5.8957,
      "mean_token_accuracy": 0.1552807003259659,
      "num_tokens": 41750966.0,
      "step": 19290
    },
    {
      "entropy": 6.329586935043335,
      "epoch": 2.064743967039435,
      "grad_norm": 1.109375,
      "learning_rate": 0.000457896813676879,
      "loss": 5.8835,
      "mean_token_accuracy": 0.1576482817530632,
      "num_tokens": 41762174.0,
      "step": 19295
    },
    {
      "entropy": 6.247367668151855,
      "epoch": 2.065279041147199,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045787454245363907,
      "loss": 6.0024,
      "mean_token_accuracy": 0.1470222771167755,
      "num_tokens": 41773396.0,
      "step": 19300
    },
    {
      "entropy": 6.300270080566406,
      "epoch": 2.065814115254963,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004578522659499225,
      "loss": 5.9494,
      "mean_token_accuracy": 0.15516361743211746,
      "num_tokens": 41783833.0,
      "step": 19305
    },
    {
      "entropy": 6.2395072937011715,
      "epoch": 2.0663491893627266,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045782998416637235,
      "loss": 5.9129,
      "mean_token_accuracy": 0.16011869609355928,
      "num_tokens": 41795363.0,
      "step": 19310
    },
    {
      "entropy": 6.3042871952056885,
      "epoch": 2.0668842634704907,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004578076971036321,
      "loss": 5.9678,
      "mean_token_accuracy": 0.15252817273139954,
      "num_tokens": 41807575.0,
      "step": 19315
    },
    {
      "entropy": 6.188302373886108,
      "epoch": 2.067419337578255,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004577854047623454,
      "loss": 5.8739,
      "mean_token_accuracy": 0.15532160997390748,
      "num_tokens": 41818496.0,
      "step": 19320
    },
    {
      "entropy": 6.299874544143677,
      "epoch": 2.0679544116860185,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004577631071431557,
      "loss": 5.9173,
      "mean_token_accuracy": 0.15724148005247116,
      "num_tokens": 41829106.0,
      "step": 19325
    },
    {
      "entropy": 6.226364040374756,
      "epoch": 2.0684894857937826,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045774080424670704,
      "loss": 5.9122,
      "mean_token_accuracy": 0.15635633319616318,
      "num_tokens": 41838095.0,
      "step": 19330
    },
    {
      "entropy": 6.202659988403321,
      "epoch": 2.069024559901546,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004577184960736433,
      "loss": 5.9143,
      "mean_token_accuracy": 0.14664157032966613,
      "num_tokens": 41847980.0,
      "step": 19335
    },
    {
      "entropy": 6.306939506530762,
      "epoch": 2.0695596340093103,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045769618262460864,
      "loss": 5.9437,
      "mean_token_accuracy": 0.15504215285182,
      "num_tokens": 41858625.0,
      "step": 19340
    },
    {
      "entropy": 6.291658735275268,
      "epoch": 2.0700947081170744,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004576738639002475,
      "loss": 5.9547,
      "mean_token_accuracy": 0.1490628845989704,
      "num_tokens": 41870939.0,
      "step": 19345
    },
    {
      "entropy": 6.31524133682251,
      "epoch": 2.070629782224838,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004576515399012041,
      "loss": 5.9893,
      "mean_token_accuracy": 0.15351814553141593,
      "num_tokens": 41882356.0,
      "step": 19350
    },
    {
      "entropy": 6.306646490097046,
      "epoch": 2.071164856332602,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004576292106281231,
      "loss": 5.9543,
      "mean_token_accuracy": 0.1505764678120613,
      "num_tokens": 41892306.0,
      "step": 19355
    },
    {
      "entropy": 6.281562805175781,
      "epoch": 2.071699930440366,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045760687608164933,
      "loss": 5.9886,
      "mean_token_accuracy": 0.14722599983215331,
      "num_tokens": 41902450.0,
      "step": 19360
    },
    {
      "entropy": 6.249219608306885,
      "epoch": 2.07223500454813,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004575845362624278,
      "loss": 5.9032,
      "mean_token_accuracy": 0.1564817577600479,
      "num_tokens": 41913319.0,
      "step": 19365
    },
    {
      "entropy": 6.299248743057251,
      "epoch": 2.072770078655894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045756219117110337,
      "loss": 5.9609,
      "mean_token_accuracy": 0.14723388403654097,
      "num_tokens": 41923598.0,
      "step": 19370
    },
    {
      "entropy": 6.323636054992676,
      "epoch": 2.0733051527636577,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045753984080832127,
      "loss": 5.9349,
      "mean_token_accuracy": 0.15037770867347716,
      "num_tokens": 41934482.0,
      "step": 19375
    },
    {
      "entropy": 6.264130878448486,
      "epoch": 2.0738402268714218,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000457517485174727,
      "loss": 5.921,
      "mean_token_accuracy": 0.1506503030657768,
      "num_tokens": 41945994.0,
      "step": 19380
    },
    {
      "entropy": 6.268437242507934,
      "epoch": 2.0743753009791854,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045749512427096595,
      "loss": 5.9357,
      "mean_token_accuracy": 0.15002575665712356,
      "num_tokens": 41956564.0,
      "step": 19385
    },
    {
      "entropy": 6.244604158401489,
      "epoch": 2.0749103750869495,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045747275809768386,
      "loss": 5.9199,
      "mean_token_accuracy": 0.15635277926921845,
      "num_tokens": 41966847.0,
      "step": 19390
    },
    {
      "entropy": 6.2994499683380125,
      "epoch": 2.0754454491947136,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004574503866555265,
      "loss": 6.0138,
      "mean_token_accuracy": 0.14326372891664504,
      "num_tokens": 41977752.0,
      "step": 19395
    },
    {
      "entropy": 6.325182485580444,
      "epoch": 2.0759805233024773,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045742800994513983,
      "loss": 5.9311,
      "mean_token_accuracy": 0.15465786159038544,
      "num_tokens": 41987837.0,
      "step": 19400
    },
    {
      "entropy": 6.32652096748352,
      "epoch": 2.0765155974102414,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045740562796717005,
      "loss": 5.9509,
      "mean_token_accuracy": 0.15155716985464096,
      "num_tokens": 41997988.0,
      "step": 19405
    },
    {
      "entropy": 6.1424243450164795,
      "epoch": 2.077050671518005,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004573832407222634,
      "loss": 5.8617,
      "mean_token_accuracy": 0.15560182929039001,
      "num_tokens": 42009643.0,
      "step": 19410
    },
    {
      "entropy": 6.266240072250366,
      "epoch": 2.077585745625769,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004573608482110663,
      "loss": 5.928,
      "mean_token_accuracy": 0.1512785881757736,
      "num_tokens": 42019972.0,
      "step": 19415
    },
    {
      "entropy": 6.3175249099731445,
      "epoch": 2.0781208197335332,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045733845043422535,
      "loss": 5.9055,
      "mean_token_accuracy": 0.15391668528318406,
      "num_tokens": 42030536.0,
      "step": 19420
    },
    {
      "entropy": 6.292443037033081,
      "epoch": 2.078655893841297,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004573160473923872,
      "loss": 5.9254,
      "mean_token_accuracy": 0.1498799204826355,
      "num_tokens": 42041217.0,
      "step": 19425
    },
    {
      "entropy": 6.129509449005127,
      "epoch": 2.079190967949061,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004572936390861988,
      "loss": 5.8204,
      "mean_token_accuracy": 0.1601771041750908,
      "num_tokens": 42051346.0,
      "step": 19430
    },
    {
      "entropy": 6.3737650394439695,
      "epoch": 2.0797260420568247,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004572712255163072,
      "loss": 6.0003,
      "mean_token_accuracy": 0.15290757939219474,
      "num_tokens": 42062718.0,
      "step": 19435
    },
    {
      "entropy": 6.2979930400848385,
      "epoch": 2.0802611161645888,
      "grad_norm": 1.25,
      "learning_rate": 0.0004572488066833596,
      "loss": 5.9025,
      "mean_token_accuracy": 0.15407718867063522,
      "num_tokens": 42073587.0,
      "step": 19440
    },
    {
      "entropy": 6.268727779388428,
      "epoch": 2.080796190272353,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045722638258800335,
      "loss": 5.95,
      "mean_token_accuracy": 0.1576678529381752,
      "num_tokens": 42083763.0,
      "step": 19445
    },
    {
      "entropy": 6.227392387390137,
      "epoch": 2.0813312643801165,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004572039532308858,
      "loss": 5.8825,
      "mean_token_accuracy": 0.1527301475405693,
      "num_tokens": 42094062.0,
      "step": 19450
    },
    {
      "entropy": 6.284812545776367,
      "epoch": 2.0818663384878806,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004571815186126549,
      "loss": 5.9515,
      "mean_token_accuracy": 0.15074780508875846,
      "num_tokens": 42106136.0,
      "step": 19455
    },
    {
      "entropy": 6.277507734298706,
      "epoch": 2.0824014125956447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045715907873395805,
      "loss": 5.9693,
      "mean_token_accuracy": 0.1462613895535469,
      "num_tokens": 42116309.0,
      "step": 19460
    },
    {
      "entropy": 6.202787017822265,
      "epoch": 2.0829364867034084,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004571366335954435,
      "loss": 5.9428,
      "mean_token_accuracy": 0.15323021709918977,
      "num_tokens": 42127747.0,
      "step": 19465
    },
    {
      "entropy": 6.258214521408081,
      "epoch": 2.0834715608111725,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004571141831977592,
      "loss": 5.8664,
      "mean_token_accuracy": 0.1650477573275566,
      "num_tokens": 42137992.0,
      "step": 19470
    },
    {
      "entropy": 6.3033544540405275,
      "epoch": 2.084006634918936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004570917275415535,
      "loss": 5.8857,
      "mean_token_accuracy": 0.15399691164493562,
      "num_tokens": 42148833.0,
      "step": 19475
    },
    {
      "entropy": 6.228662967681885,
      "epoch": 2.0845417090267,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004570692666274747,
      "loss": 5.9504,
      "mean_token_accuracy": 0.15396834909915924,
      "num_tokens": 42159276.0,
      "step": 19480
    },
    {
      "entropy": 6.296672677993774,
      "epoch": 2.0850767831344643,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004570468004561715,
      "loss": 5.9951,
      "mean_token_accuracy": 0.15441043823957443,
      "num_tokens": 42171541.0,
      "step": 19485
    },
    {
      "entropy": 6.291082763671875,
      "epoch": 2.085611857242228,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045702432902829245,
      "loss": 5.9582,
      "mean_token_accuracy": 0.1546885758638382,
      "num_tokens": 42182888.0,
      "step": 19490
    },
    {
      "entropy": 6.260376834869385,
      "epoch": 2.086146931349992,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004570018523444865,
      "loss": 5.8787,
      "mean_token_accuracy": 0.16406434327363967,
      "num_tokens": 42193405.0,
      "step": 19495
    },
    {
      "entropy": 6.264280176162719,
      "epoch": 2.0866820054577557,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045697937040540263,
      "loss": 5.9035,
      "mean_token_accuracy": 0.1571453645825386,
      "num_tokens": 42203355.0,
      "step": 19500
    },
    {
      "entropy": 6.2504435062408445,
      "epoch": 2.08721707956552,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045695688321169,
      "loss": 5.9228,
      "mean_token_accuracy": 0.14314650669693946,
      "num_tokens": 42213191.0,
      "step": 19505
    },
    {
      "entropy": 6.2893967628479,
      "epoch": 2.087752153673284,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045693439076399797,
      "loss": 5.997,
      "mean_token_accuracy": 0.14969116151332856,
      "num_tokens": 42223357.0,
      "step": 19510
    },
    {
      "entropy": 6.322054243087768,
      "epoch": 2.0882872277810476,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045691189306297595,
      "loss": 5.9011,
      "mean_token_accuracy": 0.15591805726289748,
      "num_tokens": 42234917.0,
      "step": 19515
    },
    {
      "entropy": 6.309245872497558,
      "epoch": 2.0888223018888117,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045688939010927353,
      "loss": 5.9475,
      "mean_token_accuracy": 0.1507009118795395,
      "num_tokens": 42245977.0,
      "step": 19520
    },
    {
      "entropy": 6.251159620285034,
      "epoch": 2.0893573759965753,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004568668819035406,
      "loss": 5.9213,
      "mean_token_accuracy": 0.15089723765850066,
      "num_tokens": 42257523.0,
      "step": 19525
    },
    {
      "entropy": 6.331517124176026,
      "epoch": 2.0898924501043394,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045684436844642694,
      "loss": 5.9318,
      "mean_token_accuracy": 0.1606952354311943,
      "num_tokens": 42267727.0,
      "step": 19530
    },
    {
      "entropy": 6.341193151473999,
      "epoch": 2.0904275242121035,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004568218497385828,
      "loss": 5.9124,
      "mean_token_accuracy": 0.15468440055847169,
      "num_tokens": 42278201.0,
      "step": 19535
    },
    {
      "entropy": 6.318959283828735,
      "epoch": 2.090962598319867,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004567993257806582,
      "loss": 6.0153,
      "mean_token_accuracy": 0.15113338828086853,
      "num_tokens": 42288695.0,
      "step": 19540
    },
    {
      "entropy": 6.283383703231811,
      "epoch": 2.0914976724276313,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004567767965733036,
      "loss": 6.0049,
      "mean_token_accuracy": 0.1566457286477089,
      "num_tokens": 42299418.0,
      "step": 19545
    },
    {
      "entropy": 6.252763509750366,
      "epoch": 2.092032746535395,
      "grad_norm": 1.5,
      "learning_rate": 0.00045675426211716963,
      "loss": 5.8672,
      "mean_token_accuracy": 0.1553644061088562,
      "num_tokens": 42310901.0,
      "step": 19550
    },
    {
      "entropy": 6.324523019790649,
      "epoch": 2.092567820643159,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004567317224129069,
      "loss": 5.893,
      "mean_token_accuracy": 0.15918458700180055,
      "num_tokens": 42320055.0,
      "step": 19555
    },
    {
      "entropy": 6.316038703918457,
      "epoch": 2.093102894750923,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004567091774611661,
      "loss": 6.0153,
      "mean_token_accuracy": 0.1529088646173477,
      "num_tokens": 42330393.0,
      "step": 19560
    },
    {
      "entropy": 6.2293939113616945,
      "epoch": 2.093637968858687,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045668662726259846,
      "loss": 5.9139,
      "mean_token_accuracy": 0.15792387574911118,
      "num_tokens": 42341384.0,
      "step": 19565
    },
    {
      "entropy": 6.2405201435089115,
      "epoch": 2.094173042966451,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045666407181785496,
      "loss": 5.861,
      "mean_token_accuracy": 0.16521098762750625,
      "num_tokens": 42352351.0,
      "step": 19570
    },
    {
      "entropy": 6.2072405338287355,
      "epoch": 2.094708117074215,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004566415111275869,
      "loss": 5.8773,
      "mean_token_accuracy": 0.15825552567839624,
      "num_tokens": 42363912.0,
      "step": 19575
    },
    {
      "entropy": 6.282230091094971,
      "epoch": 2.0952431911819787,
      "grad_norm": 1.125,
      "learning_rate": 0.0004566189451924457,
      "loss": 5.9656,
      "mean_token_accuracy": 0.1504607915878296,
      "num_tokens": 42374638.0,
      "step": 19580
    },
    {
      "entropy": 6.203241777420044,
      "epoch": 2.0957782652897428,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045659637401308296,
      "loss": 5.8768,
      "mean_token_accuracy": 0.1637960657477379,
      "num_tokens": 42385976.0,
      "step": 19585
    },
    {
      "entropy": 6.230372905731201,
      "epoch": 2.0963133393975064,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004565737975901505,
      "loss": 5.9496,
      "mean_token_accuracy": 0.14891730546951293,
      "num_tokens": 42395714.0,
      "step": 19590
    },
    {
      "entropy": 6.220761394500732,
      "epoch": 2.0968484135052705,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004565512159243001,
      "loss": 5.8476,
      "mean_token_accuracy": 0.16076222509145738,
      "num_tokens": 42406027.0,
      "step": 19595
    },
    {
      "entropy": 6.3459125518798825,
      "epoch": 2.0973834876130346,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004565286290161839,
      "loss": 5.998,
      "mean_token_accuracy": 0.14876420348882674,
      "num_tokens": 42417756.0,
      "step": 19600
    },
    {
      "entropy": 6.322590970993042,
      "epoch": 2.0979185617207983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045650603686645403,
      "loss": 5.9976,
      "mean_token_accuracy": 0.15048714727163315,
      "num_tokens": 42429321.0,
      "step": 19605
    },
    {
      "entropy": 6.170070934295654,
      "epoch": 2.0984536358285624,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045648343947576284,
      "loss": 5.8053,
      "mean_token_accuracy": 0.16217207312583923,
      "num_tokens": 42440003.0,
      "step": 19610
    },
    {
      "entropy": 6.283868026733399,
      "epoch": 2.098988709936326,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004564608368447628,
      "loss": 5.9307,
      "mean_token_accuracy": 0.15427326709032058,
      "num_tokens": 42450743.0,
      "step": 19615
    },
    {
      "entropy": 6.270153999328613,
      "epoch": 2.09952378404409,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004564382289741067,
      "loss": 5.9171,
      "mean_token_accuracy": 0.15128770396113395,
      "num_tokens": 42460885.0,
      "step": 19620
    },
    {
      "entropy": 6.223431015014649,
      "epoch": 2.1000588581518542,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045641561586444706,
      "loss": 5.9157,
      "mean_token_accuracy": 0.15798407793045044,
      "num_tokens": 42470854.0,
      "step": 19625
    },
    {
      "entropy": 6.228867912292481,
      "epoch": 2.100593932259618,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045639299751643714,
      "loss": 5.9826,
      "mean_token_accuracy": 0.15173570364713668,
      "num_tokens": 42482392.0,
      "step": 19630
    },
    {
      "entropy": 6.302439260482788,
      "epoch": 2.101129006367382,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045637037393072996,
      "loss": 5.9481,
      "mean_token_accuracy": 0.146514493227005,
      "num_tokens": 42494156.0,
      "step": 19635
    },
    {
      "entropy": 6.266593265533447,
      "epoch": 2.1016640804751456,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004563477451079786,
      "loss": 5.9303,
      "mean_token_accuracy": 0.15646735578775406,
      "num_tokens": 42505185.0,
      "step": 19640
    },
    {
      "entropy": 6.203398323059082,
      "epoch": 2.1021991545829097,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045632511104883657,
      "loss": 5.8849,
      "mean_token_accuracy": 0.15077874511480333,
      "num_tokens": 42516415.0,
      "step": 19645
    },
    {
      "entropy": 6.295444107055664,
      "epoch": 2.102734228690674,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004563024717539575,
      "loss": 5.958,
      "mean_token_accuracy": 0.15604271441698075,
      "num_tokens": 42526945.0,
      "step": 19650
    },
    {
      "entropy": 6.337691164016723,
      "epoch": 2.1032693027984375,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045627982722399506,
      "loss": 6.0051,
      "mean_token_accuracy": 0.14937861785292625,
      "num_tokens": 42537629.0,
      "step": 19655
    },
    {
      "entropy": 6.223173666000366,
      "epoch": 2.1038043769062016,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045625717745960306,
      "loss": 5.9535,
      "mean_token_accuracy": 0.15442239716649056,
      "num_tokens": 42547952.0,
      "step": 19660
    },
    {
      "entropy": 6.229601526260376,
      "epoch": 2.1043394510139652,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045623452246143547,
      "loss": 5.9189,
      "mean_token_accuracy": 0.15585146993398666,
      "num_tokens": 42559026.0,
      "step": 19665
    },
    {
      "entropy": 6.320555591583252,
      "epoch": 2.1048745251217293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004562118622301466,
      "loss": 5.9879,
      "mean_token_accuracy": 0.15111162513494492,
      "num_tokens": 42570046.0,
      "step": 19670
    },
    {
      "entropy": 6.338024044036866,
      "epoch": 2.1054095992294934,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004561891967663906,
      "loss": 5.9847,
      "mean_token_accuracy": 0.14832480996847153,
      "num_tokens": 42580440.0,
      "step": 19675
    },
    {
      "entropy": 6.294213342666626,
      "epoch": 2.105944673337257,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045616652607082203,
      "loss": 5.9425,
      "mean_token_accuracy": 0.14678931906819342,
      "num_tokens": 42591872.0,
      "step": 19680
    },
    {
      "entropy": 6.271905899047852,
      "epoch": 2.106479747445021,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004561438501440955,
      "loss": 5.8438,
      "mean_token_accuracy": 0.15221378803253174,
      "num_tokens": 42602843.0,
      "step": 19685
    },
    {
      "entropy": 6.246933078765869,
      "epoch": 2.107014821552785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004561211689868657,
      "loss": 5.9406,
      "mean_token_accuracy": 0.1604931741952896,
      "num_tokens": 42613476.0,
      "step": 19690
    },
    {
      "entropy": 6.178541374206543,
      "epoch": 2.107549895660549,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045609848259978766,
      "loss": 5.8501,
      "mean_token_accuracy": 0.1682644546031952,
      "num_tokens": 42624059.0,
      "step": 19695
    },
    {
      "entropy": 6.300055360794067,
      "epoch": 2.108084969768313,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004560757909835164,
      "loss": 5.9948,
      "mean_token_accuracy": 0.1509998008608818,
      "num_tokens": 42633980.0,
      "step": 19700
    },
    {
      "entropy": 6.362158727645874,
      "epoch": 2.1086200438760767,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045605309413870697,
      "loss": 5.9935,
      "mean_token_accuracy": 0.1488317593932152,
      "num_tokens": 42645504.0,
      "step": 19705
    },
    {
      "entropy": 6.335047578811645,
      "epoch": 2.109155117983841,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004560303920660151,
      "loss": 5.9497,
      "mean_token_accuracy": 0.15016166865825653,
      "num_tokens": 42655403.0,
      "step": 19710
    },
    {
      "entropy": 6.329136848449707,
      "epoch": 2.1096901920916045,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045600768476609596,
      "loss": 5.983,
      "mean_token_accuracy": 0.1526212438941002,
      "num_tokens": 42666236.0,
      "step": 19715
    },
    {
      "entropy": 6.31711711883545,
      "epoch": 2.1102252661993686,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045598497223960533,
      "loss": 6.0684,
      "mean_token_accuracy": 0.14504285529255867,
      "num_tokens": 42677365.0,
      "step": 19720
    },
    {
      "entropy": 6.303063249588012,
      "epoch": 2.1107603403071327,
      "grad_norm": 1.5625,
      "learning_rate": 0.00045596225448719913,
      "loss": 5.934,
      "mean_token_accuracy": 0.1536141477525234,
      "num_tokens": 42688518.0,
      "step": 19725
    },
    {
      "entropy": 6.316926908493042,
      "epoch": 2.1112954144148963,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045593953150953326,
      "loss": 5.9439,
      "mean_token_accuracy": 0.15064158737659455,
      "num_tokens": 42699877.0,
      "step": 19730
    },
    {
      "entropy": 6.263840675354004,
      "epoch": 2.1118304885226604,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004559168033072638,
      "loss": 5.9195,
      "mean_token_accuracy": 0.15838904082775115,
      "num_tokens": 42710191.0,
      "step": 19735
    },
    {
      "entropy": 6.2850874900817875,
      "epoch": 2.1123655626304245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045589406988104717,
      "loss": 6.0159,
      "mean_token_accuracy": 0.15054293125867843,
      "num_tokens": 42720874.0,
      "step": 19740
    },
    {
      "entropy": 6.294171094894409,
      "epoch": 2.112900636738188,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004558713312315396,
      "loss": 5.953,
      "mean_token_accuracy": 0.15830839723348616,
      "num_tokens": 42731783.0,
      "step": 19745
    },
    {
      "entropy": 6.2622315883636475,
      "epoch": 2.1134357108459523,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004558485873593978,
      "loss": 5.9247,
      "mean_token_accuracy": 0.15508714318275452,
      "num_tokens": 42741602.0,
      "step": 19750
    },
    {
      "entropy": 6.307151651382446,
      "epoch": 2.113970784953716,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045582583826527835,
      "loss": 5.8906,
      "mean_token_accuracy": 0.15512883812189102,
      "num_tokens": 42751249.0,
      "step": 19755
    },
    {
      "entropy": 6.248520946502685,
      "epoch": 2.11450585906148,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004558030839498383,
      "loss": 5.9135,
      "mean_token_accuracy": 0.15625598281621933,
      "num_tokens": 42762083.0,
      "step": 19760
    },
    {
      "entropy": 6.226223611831665,
      "epoch": 2.115040933169244,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004557803244137347,
      "loss": 5.797,
      "mean_token_accuracy": 0.17104652523994446,
      "num_tokens": 42773443.0,
      "step": 19765
    },
    {
      "entropy": 6.281828165054321,
      "epoch": 2.115576007277008,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004557575596576245,
      "loss": 5.9676,
      "mean_token_accuracy": 0.15272417813539504,
      "num_tokens": 42784126.0,
      "step": 19770
    },
    {
      "entropy": 6.250914239883423,
      "epoch": 2.116111081384772,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045573478968216526,
      "loss": 5.8636,
      "mean_token_accuracy": 0.1632809691131115,
      "num_tokens": 42795472.0,
      "step": 19775
    },
    {
      "entropy": 6.251201725006103,
      "epoch": 2.1166461554925355,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045571201448801424,
      "loss": 5.9312,
      "mean_token_accuracy": 0.1571137323975563,
      "num_tokens": 42805279.0,
      "step": 19780
    },
    {
      "entropy": 6.27225136756897,
      "epoch": 2.1171812296002996,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004556892340758293,
      "loss": 5.9649,
      "mean_token_accuracy": 0.154823037981987,
      "num_tokens": 42815680.0,
      "step": 19785
    },
    {
      "entropy": 6.29111647605896,
      "epoch": 2.1177163037080637,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004556664484462681,
      "loss": 6.011,
      "mean_token_accuracy": 0.14399922713637353,
      "num_tokens": 42826724.0,
      "step": 19790
    },
    {
      "entropy": 6.298679637908935,
      "epoch": 2.1182513778158274,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045564365759998856,
      "loss": 5.9398,
      "mean_token_accuracy": 0.15229920893907548,
      "num_tokens": 42837875.0,
      "step": 19795
    },
    {
      "entropy": 6.295600700378418,
      "epoch": 2.1187864519235915,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004556208615376488,
      "loss": 5.9294,
      "mean_token_accuracy": 0.14957500994205475,
      "num_tokens": 42848714.0,
      "step": 19800
    },
    {
      "entropy": 6.247407245635986,
      "epoch": 2.119321526031355,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004555980602599071,
      "loss": 5.9999,
      "mean_token_accuracy": 0.1533982999622822,
      "num_tokens": 42859764.0,
      "step": 19805
    },
    {
      "entropy": 6.187853193283081,
      "epoch": 2.1198566001391193,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004555752537674217,
      "loss": 5.8296,
      "mean_token_accuracy": 0.1591409020125866,
      "num_tokens": 42870346.0,
      "step": 19810
    },
    {
      "entropy": 6.233883857727051,
      "epoch": 2.1203916742468834,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004555524420608513,
      "loss": 5.9618,
      "mean_token_accuracy": 0.15640655606985093,
      "num_tokens": 42882706.0,
      "step": 19815
    },
    {
      "entropy": 6.250353813171387,
      "epoch": 2.120926748354647,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045552962514085443,
      "loss": 5.939,
      "mean_token_accuracy": 0.15857987105846405,
      "num_tokens": 42893611.0,
      "step": 19820
    },
    {
      "entropy": 6.32988715171814,
      "epoch": 2.121461822462411,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045550680300809003,
      "loss": 5.9002,
      "mean_token_accuracy": 0.15196246802806854,
      "num_tokens": 42904034.0,
      "step": 19825
    },
    {
      "entropy": 6.214957904815674,
      "epoch": 2.1219968965701748,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045548397566321706,
      "loss": 5.9339,
      "mean_token_accuracy": 0.15429795533418655,
      "num_tokens": 42915530.0,
      "step": 19830
    },
    {
      "entropy": 6.2680891990661625,
      "epoch": 2.122531970677939,
      "grad_norm": 1.25,
      "learning_rate": 0.0004554611431068947,
      "loss": 5.906,
      "mean_token_accuracy": 0.1559049293398857,
      "num_tokens": 42926696.0,
      "step": 19835
    },
    {
      "entropy": 6.293427896499634,
      "epoch": 2.123067044785703,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004554383053397822,
      "loss": 5.9621,
      "mean_token_accuracy": 0.15852337032556535,
      "num_tokens": 42936985.0,
      "step": 19840
    },
    {
      "entropy": 6.269944906234741,
      "epoch": 2.1236021188934666,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045541546236253896,
      "loss": 5.8926,
      "mean_token_accuracy": 0.15842733830213546,
      "num_tokens": 42947996.0,
      "step": 19845
    },
    {
      "entropy": 6.2426536083221436,
      "epoch": 2.1241371930012307,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045539261417582456,
      "loss": 5.837,
      "mean_token_accuracy": 0.1584152340888977,
      "num_tokens": 42959669.0,
      "step": 19850
    },
    {
      "entropy": 6.2491919040679935,
      "epoch": 2.124672267108995,
      "grad_norm": 1.125,
      "learning_rate": 0.00045536976078029874,
      "loss": 5.8221,
      "mean_token_accuracy": 0.15981436669826507,
      "num_tokens": 42969687.0,
      "step": 19855
    },
    {
      "entropy": 6.249666690826416,
      "epoch": 2.1252073412167585,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004553469021766215,
      "loss": 5.9843,
      "mean_token_accuracy": 0.14470615833997727,
      "num_tokens": 42980752.0,
      "step": 19860
    },
    {
      "entropy": 6.3244547843933105,
      "epoch": 2.1257424153245226,
      "grad_norm": 1.25,
      "learning_rate": 0.00045532403836545286,
      "loss": 6.0329,
      "mean_token_accuracy": 0.1532257989048958,
      "num_tokens": 42991949.0,
      "step": 19865
    },
    {
      "entropy": 6.204038429260254,
      "epoch": 2.1262774894322862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004553011693474528,
      "loss": 5.881,
      "mean_token_accuracy": 0.16456524282693863,
      "num_tokens": 43002689.0,
      "step": 19870
    },
    {
      "entropy": 6.278147459030151,
      "epoch": 2.1268125635400503,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004552782951232819,
      "loss": 5.9434,
      "mean_token_accuracy": 0.1511400446295738,
      "num_tokens": 43012939.0,
      "step": 19875
    },
    {
      "entropy": 6.339469003677368,
      "epoch": 2.1273476376478144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045525541569360053,
      "loss": 5.9053,
      "mean_token_accuracy": 0.15424875319004058,
      "num_tokens": 43023887.0,
      "step": 19880
    },
    {
      "entropy": 6.182108354568482,
      "epoch": 2.127882711755578,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004552325310590694,
      "loss": 5.831,
      "mean_token_accuracy": 0.1565584734082222,
      "num_tokens": 43034554.0,
      "step": 19885
    },
    {
      "entropy": 6.246485567092895,
      "epoch": 2.128417785863342,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004552096412203492,
      "loss": 6.0072,
      "mean_token_accuracy": 0.15380725115537644,
      "num_tokens": 43046322.0,
      "step": 19890
    },
    {
      "entropy": 6.298896169662475,
      "epoch": 2.128952859971106,
      "grad_norm": 1.21875,
      "learning_rate": 0.000455186746178101,
      "loss": 5.8495,
      "mean_token_accuracy": 0.15835124850273133,
      "num_tokens": 43055981.0,
      "step": 19895
    },
    {
      "entropy": 6.288560152053833,
      "epoch": 2.12948793407887,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045516384593298584,
      "loss": 5.9492,
      "mean_token_accuracy": 0.14679785594344139,
      "num_tokens": 43067217.0,
      "step": 19900
    },
    {
      "entropy": 6.313830471038818,
      "epoch": 2.130023008186634,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045514094048566486,
      "loss": 5.942,
      "mean_token_accuracy": 0.150324946641922,
      "num_tokens": 43078312.0,
      "step": 19905
    },
    {
      "entropy": 6.146958303451538,
      "epoch": 2.1305580822943977,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045511802983679955,
      "loss": 5.7504,
      "mean_token_accuracy": 0.17468599528074263,
      "num_tokens": 43090210.0,
      "step": 19910
    },
    {
      "entropy": 6.21459174156189,
      "epoch": 2.131093156402162,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004550951139870515,
      "loss": 5.874,
      "mean_token_accuracy": 0.15374426543712616,
      "num_tokens": 43101282.0,
      "step": 19915
    },
    {
      "entropy": 6.253501939773559,
      "epoch": 2.1316282305099254,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045507219293708227,
      "loss": 6.0286,
      "mean_token_accuracy": 0.14789000004529954,
      "num_tokens": 43112856.0,
      "step": 19920
    },
    {
      "entropy": 6.26109127998352,
      "epoch": 2.1321633046176895,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045504926668755384,
      "loss": 5.8572,
      "mean_token_accuracy": 0.16342170685529708,
      "num_tokens": 43123898.0,
      "step": 19925
    },
    {
      "entropy": 6.28247184753418,
      "epoch": 2.1326983787254536,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004550263352391281,
      "loss": 5.8988,
      "mean_token_accuracy": 0.15914675444364548,
      "num_tokens": 43134651.0,
      "step": 19930
    },
    {
      "entropy": 6.259369707107544,
      "epoch": 2.1332334528332173,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004550033985924672,
      "loss": 5.956,
      "mean_token_accuracy": 0.15490619987249374,
      "num_tokens": 43145385.0,
      "step": 19935
    },
    {
      "entropy": 6.298950719833374,
      "epoch": 2.1337685269409814,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004549804567482335,
      "loss": 5.9632,
      "mean_token_accuracy": 0.15268329828977584,
      "num_tokens": 43157356.0,
      "step": 19940
    },
    {
      "entropy": 6.302832078933716,
      "epoch": 2.134303601048745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045495750970708944,
      "loss": 5.9611,
      "mean_token_accuracy": 0.1575068324804306,
      "num_tokens": 43168289.0,
      "step": 19945
    },
    {
      "entropy": 6.17238221168518,
      "epoch": 2.134838675156509,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045493455746969746,
      "loss": 5.8744,
      "mean_token_accuracy": 0.1634986460208893,
      "num_tokens": 43177796.0,
      "step": 19950
    },
    {
      "entropy": 6.235850286483765,
      "epoch": 2.1353737492642733,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004549116000367205,
      "loss": 5.897,
      "mean_token_accuracy": 0.15564680993556976,
      "num_tokens": 43188846.0,
      "step": 19955
    },
    {
      "entropy": 6.339863014221192,
      "epoch": 2.135908823372037,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045488863740882133,
      "loss": 5.9821,
      "mean_token_accuracy": 0.15393128246068954,
      "num_tokens": 43199852.0,
      "step": 19960
    },
    {
      "entropy": 6.264580583572387,
      "epoch": 2.136443897479801,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045486566958666307,
      "loss": 5.8831,
      "mean_token_accuracy": 0.1546657547354698,
      "num_tokens": 43210591.0,
      "step": 19965
    },
    {
      "entropy": 6.285552263259888,
      "epoch": 2.136978971587565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045484269657090887,
      "loss": 5.8754,
      "mean_token_accuracy": 0.15188806205987931,
      "num_tokens": 43221760.0,
      "step": 19970
    },
    {
      "entropy": 6.276112174987793,
      "epoch": 2.1375140456953288,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045481971836222205,
      "loss": 5.9939,
      "mean_token_accuracy": 0.144768937677145,
      "num_tokens": 43233693.0,
      "step": 19975
    },
    {
      "entropy": 6.249089860916138,
      "epoch": 2.138049119803093,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045479673496126615,
      "loss": 5.963,
      "mean_token_accuracy": 0.1538100242614746,
      "num_tokens": 43244565.0,
      "step": 19980
    },
    {
      "entropy": 6.1829675197601315,
      "epoch": 2.1385841939108565,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045477374636870485,
      "loss": 5.8717,
      "mean_token_accuracy": 0.15872388184070588,
      "num_tokens": 43254913.0,
      "step": 19985
    },
    {
      "entropy": 6.317099332809448,
      "epoch": 2.1391192680186206,
      "grad_norm": 1.25,
      "learning_rate": 0.00045475075258520175,
      "loss": 5.9418,
      "mean_token_accuracy": 0.1559029147028923,
      "num_tokens": 43265525.0,
      "step": 19990
    },
    {
      "entropy": 6.309792804718017,
      "epoch": 2.1396543421263843,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004547277536114211,
      "loss": 6.0304,
      "mean_token_accuracy": 0.14505867809057235,
      "num_tokens": 43276954.0,
      "step": 19995
    },
    {
      "entropy": 6.286677885055542,
      "epoch": 2.1401894162341484,
      "grad_norm": 1.125,
      "learning_rate": 0.0004547047494480267,
      "loss": 5.9355,
      "mean_token_accuracy": 0.14981783032417298,
      "num_tokens": 43287491.0,
      "step": 20000
    },
    {
      "entropy": 6.218845224380493,
      "epoch": 2.1407244903419125,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004546817400956829,
      "loss": 5.9769,
      "mean_token_accuracy": 0.1453861266374588,
      "num_tokens": 43298341.0,
      "step": 20005
    },
    {
      "entropy": 6.211858320236206,
      "epoch": 2.141259564449676,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045465872555505417,
      "loss": 5.8747,
      "mean_token_accuracy": 0.1607342928647995,
      "num_tokens": 43309450.0,
      "step": 20010
    },
    {
      "entropy": 6.326441621780395,
      "epoch": 2.1417946385574402,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000454635705826805,
      "loss": 5.9226,
      "mean_token_accuracy": 0.15098298192024232,
      "num_tokens": 43320126.0,
      "step": 20015
    },
    {
      "entropy": 6.3065684795379635,
      "epoch": 2.1423297126652043,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004546126809116,
      "loss": 6.0109,
      "mean_token_accuracy": 0.14542574286460877,
      "num_tokens": 43331347.0,
      "step": 20020
    },
    {
      "entropy": 6.284770059585571,
      "epoch": 2.142864786772968,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004545896508101041,
      "loss": 5.9094,
      "mean_token_accuracy": 0.1543918713927269,
      "num_tokens": 43341351.0,
      "step": 20025
    },
    {
      "entropy": 6.3018293380737305,
      "epoch": 2.143399860880732,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004545666155229823,
      "loss": 5.9786,
      "mean_token_accuracy": 0.14938260316848756,
      "num_tokens": 43352471.0,
      "step": 20030
    },
    {
      "entropy": 6.274745321273803,
      "epoch": 2.1439349349884957,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045454357505089965,
      "loss": 5.8736,
      "mean_token_accuracy": 0.1563809722661972,
      "num_tokens": 43363041.0,
      "step": 20035
    },
    {
      "entropy": 6.2156518459320065,
      "epoch": 2.14447000909626,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004545205293945215,
      "loss": 5.9008,
      "mean_token_accuracy": 0.1533071592450142,
      "num_tokens": 43375034.0,
      "step": 20040
    },
    {
      "entropy": 6.324613189697265,
      "epoch": 2.145005083204024,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004544974785545134,
      "loss": 6.0226,
      "mean_token_accuracy": 0.1418814554810524,
      "num_tokens": 43386110.0,
      "step": 20045
    },
    {
      "entropy": 6.30292444229126,
      "epoch": 2.1455401573117876,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004544744225315407,
      "loss": 5.9228,
      "mean_token_accuracy": 0.154967238008976,
      "num_tokens": 43396150.0,
      "step": 20050
    },
    {
      "entropy": 6.288969278335571,
      "epoch": 2.1460752314195517,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004544513613262693,
      "loss": 5.9618,
      "mean_token_accuracy": 0.15213830322027205,
      "num_tokens": 43407237.0,
      "step": 20055
    },
    {
      "entropy": 6.216687345504761,
      "epoch": 2.1466103055273154,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004544282949393652,
      "loss": 5.8335,
      "mean_token_accuracy": 0.159900264441967,
      "num_tokens": 43417626.0,
      "step": 20060
    },
    {
      "entropy": 6.222168922424316,
      "epoch": 2.1471453796350795,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004544052233714941,
      "loss": 5.9254,
      "mean_token_accuracy": 0.15706650018692017,
      "num_tokens": 43427368.0,
      "step": 20065
    },
    {
      "entropy": 6.201968383789063,
      "epoch": 2.1476804537428436,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045438214662332245,
      "loss": 5.8825,
      "mean_token_accuracy": 0.15706612318754196,
      "num_tokens": 43438159.0,
      "step": 20070
    },
    {
      "entropy": 6.271171951293946,
      "epoch": 2.148215527850607,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045435906469551657,
      "loss": 5.9594,
      "mean_token_accuracy": 0.16030513793230056,
      "num_tokens": 43447403.0,
      "step": 20075
    },
    {
      "entropy": 6.234069013595581,
      "epoch": 2.1487506019583713,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045433597758874286,
      "loss": 5.8169,
      "mean_token_accuracy": 0.16663280129432678,
      "num_tokens": 43457713.0,
      "step": 20080
    },
    {
      "entropy": 6.259639692306519,
      "epoch": 2.149285676066135,
      "grad_norm": 1.46875,
      "learning_rate": 0.000454312885303668,
      "loss": 5.9565,
      "mean_token_accuracy": 0.15866110026836394,
      "num_tokens": 43468520.0,
      "step": 20085
    },
    {
      "entropy": 6.280732345581055,
      "epoch": 2.149820750173899,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045428978784095873,
      "loss": 5.9911,
      "mean_token_accuracy": 0.1513373613357544,
      "num_tokens": 43480501.0,
      "step": 20090
    },
    {
      "entropy": 6.310552597045898,
      "epoch": 2.150355824281663,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045426668520128204,
      "loss": 5.9225,
      "mean_token_accuracy": 0.1550263211131096,
      "num_tokens": 43491297.0,
      "step": 20095
    },
    {
      "entropy": 6.313686084747315,
      "epoch": 2.150890898389427,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004542435773853051,
      "loss": 5.9516,
      "mean_token_accuracy": 0.15245707482099533,
      "num_tokens": 43501212.0,
      "step": 20100
    },
    {
      "entropy": 6.246295642852783,
      "epoch": 2.151425972497191,
      "grad_norm": 1.140625,
      "learning_rate": 0.000454220464393695,
      "loss": 5.9876,
      "mean_token_accuracy": 0.14720962345600128,
      "num_tokens": 43511136.0,
      "step": 20105
    },
    {
      "entropy": 6.311372470855713,
      "epoch": 2.1519610466049546,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004541973462271192,
      "loss": 6.0018,
      "mean_token_accuracy": 0.15062142312526702,
      "num_tokens": 43522006.0,
      "step": 20110
    },
    {
      "entropy": 6.274798965454101,
      "epoch": 2.1524961207127187,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004541742228862451,
      "loss": 5.9111,
      "mean_token_accuracy": 0.15347095876932143,
      "num_tokens": 43532746.0,
      "step": 20115
    },
    {
      "entropy": 6.213828611373901,
      "epoch": 2.1530311948204828,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045415109437174064,
      "loss": 5.9223,
      "mean_token_accuracy": 0.15505475848913192,
      "num_tokens": 43543707.0,
      "step": 20120
    },
    {
      "entropy": 6.243547487258911,
      "epoch": 2.1535662689282464,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045412796068427337,
      "loss": 5.9143,
      "mean_token_accuracy": 0.15147148072719574,
      "num_tokens": 43555029.0,
      "step": 20125
    },
    {
      "entropy": 6.269501829147339,
      "epoch": 2.1541013430360105,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045410482182451153,
      "loss": 5.9669,
      "mean_token_accuracy": 0.14967311322689056,
      "num_tokens": 43565639.0,
      "step": 20130
    },
    {
      "entropy": 6.226607322692871,
      "epoch": 2.1546364171437746,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045408167779312306,
      "loss": 5.8588,
      "mean_token_accuracy": 0.16175623685121537,
      "num_tokens": 43576333.0,
      "step": 20135
    },
    {
      "entropy": 6.281616115570069,
      "epoch": 2.1551714912515383,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045405852859077626,
      "loss": 5.908,
      "mean_token_accuracy": 0.1511178880929947,
      "num_tokens": 43587012.0,
      "step": 20140
    },
    {
      "entropy": 6.255270147323609,
      "epoch": 2.1557065653593024,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045403537421813975,
      "loss": 5.8642,
      "mean_token_accuracy": 0.1586196482181549,
      "num_tokens": 43597779.0,
      "step": 20145
    },
    {
      "entropy": 6.267630863189697,
      "epoch": 2.156241639467066,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045401221467588184,
      "loss": 5.9083,
      "mean_token_accuracy": 0.1551655039191246,
      "num_tokens": 43608631.0,
      "step": 20150
    },
    {
      "entropy": 6.2630492687225345,
      "epoch": 2.15677671357483,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004539890499646714,
      "loss": 5.8889,
      "mean_token_accuracy": 0.1585201606154442,
      "num_tokens": 43620254.0,
      "step": 20155
    },
    {
      "entropy": 6.232387828826904,
      "epoch": 2.1573117876825942,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004539658800851773,
      "loss": 5.8526,
      "mean_token_accuracy": 0.15721640288829802,
      "num_tokens": 43630986.0,
      "step": 20160
    },
    {
      "entropy": 6.206603097915649,
      "epoch": 2.157846861790358,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004539427050380686,
      "loss": 5.8611,
      "mean_token_accuracy": 0.15259130746126176,
      "num_tokens": 43641260.0,
      "step": 20165
    },
    {
      "entropy": 6.2591770648956295,
      "epoch": 2.158381935898122,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045391952482401444,
      "loss": 6.0016,
      "mean_token_accuracy": 0.15266765430569648,
      "num_tokens": 43652093.0,
      "step": 20170
    },
    {
      "entropy": 6.293264770507813,
      "epoch": 2.1589170100058857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045389633944368406,
      "loss": 5.93,
      "mean_token_accuracy": 0.1554163694381714,
      "num_tokens": 43662799.0,
      "step": 20175
    },
    {
      "entropy": 6.231133937835693,
      "epoch": 2.1594520841136498,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004538731488977471,
      "loss": 5.8654,
      "mean_token_accuracy": 0.1566663548350334,
      "num_tokens": 43674198.0,
      "step": 20180
    },
    {
      "entropy": 6.233526039123535,
      "epoch": 2.159987158221414,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004538499531868731,
      "loss": 5.9247,
      "mean_token_accuracy": 0.15374590456485748,
      "num_tokens": 43684841.0,
      "step": 20185
    },
    {
      "entropy": 6.328927040100098,
      "epoch": 2.1605222323291775,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045382675231173174,
      "loss": 6.0069,
      "mean_token_accuracy": 0.1463502898812294,
      "num_tokens": 43695903.0,
      "step": 20190
    },
    {
      "entropy": 6.315093088150024,
      "epoch": 2.1610573064369416,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004538035462729931,
      "loss": 5.9085,
      "mean_token_accuracy": 0.1563501849770546,
      "num_tokens": 43706740.0,
      "step": 20195
    },
    {
      "entropy": 6.236486625671387,
      "epoch": 2.1615923805447053,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004537803350713273,
      "loss": 5.8793,
      "mean_token_accuracy": 0.15288539677858354,
      "num_tokens": 43717124.0,
      "step": 20200
    },
    {
      "entropy": 6.212529420852661,
      "epoch": 2.1621274546524694,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004537571187074043,
      "loss": 5.8855,
      "mean_token_accuracy": 0.15607571899890899,
      "num_tokens": 43727346.0,
      "step": 20205
    },
    {
      "entropy": 6.326209020614624,
      "epoch": 2.1626625287602335,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045373389718189464,
      "loss": 6.045,
      "mean_token_accuracy": 0.1398628108203411,
      "num_tokens": 43738297.0,
      "step": 20210
    },
    {
      "entropy": 6.273516035079956,
      "epoch": 2.163197602867997,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004537106704954689,
      "loss": 5.9316,
      "mean_token_accuracy": 0.15082185119390487,
      "num_tokens": 43749543.0,
      "step": 20215
    },
    {
      "entropy": 6.308429336547851,
      "epoch": 2.163732676975761,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045368743864879755,
      "loss": 5.8997,
      "mean_token_accuracy": 0.15790337771177293,
      "num_tokens": 43760324.0,
      "step": 20220
    },
    {
      "entropy": 6.304547548294067,
      "epoch": 2.164267751083525,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045366420164255156,
      "loss": 5.9847,
      "mean_token_accuracy": 0.14585561752319337,
      "num_tokens": 43770391.0,
      "step": 20225
    },
    {
      "entropy": 6.236790180206299,
      "epoch": 2.164802825191289,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004536409594774019,
      "loss": 5.87,
      "mean_token_accuracy": 0.15027831345796586,
      "num_tokens": 43781516.0,
      "step": 20230
    },
    {
      "entropy": 6.205483055114746,
      "epoch": 2.165337899299053,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045361771215401953,
      "loss": 5.9294,
      "mean_token_accuracy": 0.1617190033197403,
      "num_tokens": 43794152.0,
      "step": 20235
    },
    {
      "entropy": 6.291596508026123,
      "epoch": 2.1658729734068167,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045359445967307595,
      "loss": 5.9132,
      "mean_token_accuracy": 0.15864709615707398,
      "num_tokens": 43805712.0,
      "step": 20240
    },
    {
      "entropy": 6.3359339237213135,
      "epoch": 2.166408047514581,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004535712020352423,
      "loss": 5.9788,
      "mean_token_accuracy": 0.15329611003398896,
      "num_tokens": 43816985.0,
      "step": 20245
    },
    {
      "entropy": 6.218364953994751,
      "epoch": 2.166943121622345,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004535479392411904,
      "loss": 5.8777,
      "mean_token_accuracy": 0.15360562652349471,
      "num_tokens": 43828048.0,
      "step": 20250
    },
    {
      "entropy": 6.2735504627227785,
      "epoch": 2.1674781957301086,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045352467129159183,
      "loss": 5.9088,
      "mean_token_accuracy": 0.1542995899915695,
      "num_tokens": 43838507.0,
      "step": 20255
    },
    {
      "entropy": 6.2285665512084964,
      "epoch": 2.1680132698378727,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045350139818711846,
      "loss": 5.8427,
      "mean_token_accuracy": 0.15744375064969063,
      "num_tokens": 43849328.0,
      "step": 20260
    },
    {
      "entropy": 6.218274641036987,
      "epoch": 2.1685483439456363,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004534781199284423,
      "loss": 5.9419,
      "mean_token_accuracy": 0.15099002420902252,
      "num_tokens": 43860311.0,
      "step": 20265
    },
    {
      "entropy": 6.333108043670654,
      "epoch": 2.1690834180534004,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004534548365162354,
      "loss": 5.9894,
      "mean_token_accuracy": 0.14851120263338088,
      "num_tokens": 43871167.0,
      "step": 20270
    },
    {
      "entropy": 6.244596767425537,
      "epoch": 2.169618492161164,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045343154795117024,
      "loss": 5.8832,
      "mean_token_accuracy": 0.16107158809900285,
      "num_tokens": 43882332.0,
      "step": 20275
    },
    {
      "entropy": 6.287263870239258,
      "epoch": 2.170153566268928,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004534082542339192,
      "loss": 5.8212,
      "mean_token_accuracy": 0.1550476148724556,
      "num_tokens": 43892971.0,
      "step": 20280
    },
    {
      "entropy": 6.226340484619141,
      "epoch": 2.1706886403766923,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045338495536515497,
      "loss": 5.9148,
      "mean_token_accuracy": 0.15621641352772714,
      "num_tokens": 43904305.0,
      "step": 20285
    },
    {
      "entropy": 6.138392114639283,
      "epoch": 2.171223714484456,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004533616513455502,
      "loss": 5.8839,
      "mean_token_accuracy": 0.15544685572385789,
      "num_tokens": 43914431.0,
      "step": 20290
    },
    {
      "entropy": 6.251799583435059,
      "epoch": 2.17175878859222,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045333834217577775,
      "loss": 5.922,
      "mean_token_accuracy": 0.15863122344017028,
      "num_tokens": 43925350.0,
      "step": 20295
    },
    {
      "entropy": 6.230429840087891,
      "epoch": 2.172293862699984,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045331502785651075,
      "loss": 5.9712,
      "mean_token_accuracy": 0.15369879007339476,
      "num_tokens": 43936194.0,
      "step": 20300
    },
    {
      "entropy": 6.272718477249145,
      "epoch": 2.172828936807748,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004532917083884224,
      "loss": 5.9281,
      "mean_token_accuracy": 0.15554354786872865,
      "num_tokens": 43946692.0,
      "step": 20305
    },
    {
      "entropy": 6.2600749969482425,
      "epoch": 2.173364010915512,
      "grad_norm": 1.03125,
      "learning_rate": 0.000453268383772186,
      "loss": 5.9015,
      "mean_token_accuracy": 0.15000587701797485,
      "num_tokens": 43957493.0,
      "step": 20310
    },
    {
      "entropy": 6.266450500488281,
      "epoch": 2.1738990850232756,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045324505400847506,
      "loss": 5.9925,
      "mean_token_accuracy": 0.15225453153252602,
      "num_tokens": 43970203.0,
      "step": 20315
    },
    {
      "entropy": 6.254830408096313,
      "epoch": 2.1744341591310397,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045322171909796327,
      "loss": 5.8879,
      "mean_token_accuracy": 0.15416286736726761,
      "num_tokens": 43980778.0,
      "step": 20320
    },
    {
      "entropy": 6.25853853225708,
      "epoch": 2.1749692332388038,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045319837904132435,
      "loss": 5.9072,
      "mean_token_accuracy": 0.15834837555885314,
      "num_tokens": 43992239.0,
      "step": 20325
    },
    {
      "entropy": 6.288370752334595,
      "epoch": 2.1755043073465674,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004531750338392323,
      "loss": 5.9401,
      "mean_token_accuracy": 0.15292563587427138,
      "num_tokens": 44001882.0,
      "step": 20330
    },
    {
      "entropy": 6.224268627166748,
      "epoch": 2.1760393814543315,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004531516834923611,
      "loss": 5.9181,
      "mean_token_accuracy": 0.14842576906085014,
      "num_tokens": 44013192.0,
      "step": 20335
    },
    {
      "entropy": 6.333024692535401,
      "epoch": 2.176574455562095,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004531283280013852,
      "loss": 5.9535,
      "mean_token_accuracy": 0.14784889221191405,
      "num_tokens": 44023072.0,
      "step": 20340
    },
    {
      "entropy": 6.303611707687378,
      "epoch": 2.1771095296698593,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045310496736697875,
      "loss": 5.9593,
      "mean_token_accuracy": 0.15260110944509506,
      "num_tokens": 44034238.0,
      "step": 20345
    },
    {
      "entropy": 6.254712247848511,
      "epoch": 2.1776446037776234,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045308160158981646,
      "loss": 5.9284,
      "mean_token_accuracy": 0.1553128406405449,
      "num_tokens": 44044392.0,
      "step": 20350
    },
    {
      "entropy": 6.317963266372681,
      "epoch": 2.178179677885387,
      "grad_norm": 1.234375,
      "learning_rate": 0.000453058230670573,
      "loss": 5.9722,
      "mean_token_accuracy": 0.14989521205425263,
      "num_tokens": 44056129.0,
      "step": 20355
    },
    {
      "entropy": 6.265676403045655,
      "epoch": 2.178714751993151,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045303485460992303,
      "loss": 5.8993,
      "mean_token_accuracy": 0.15923287123441696,
      "num_tokens": 44067600.0,
      "step": 20360
    },
    {
      "entropy": 6.266202974319458,
      "epoch": 2.1792498261009148,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045301147340854177,
      "loss": 5.8078,
      "mean_token_accuracy": 0.1660426989197731,
      "num_tokens": 44077373.0,
      "step": 20365
    },
    {
      "entropy": 6.219447422027588,
      "epoch": 2.179784900208679,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045298808706710416,
      "loss": 6.0175,
      "mean_token_accuracy": 0.1538599982857704,
      "num_tokens": 44087422.0,
      "step": 20370
    },
    {
      "entropy": 6.251960039138794,
      "epoch": 2.180319974316443,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045296469558628555,
      "loss": 5.957,
      "mean_token_accuracy": 0.152763532102108,
      "num_tokens": 44098524.0,
      "step": 20375
    },
    {
      "entropy": 6.234009027481079,
      "epoch": 2.1808550484242066,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045294129896676134,
      "loss": 5.8754,
      "mean_token_accuracy": 0.1590183585882187,
      "num_tokens": 44108340.0,
      "step": 20380
    },
    {
      "entropy": 6.256120920181274,
      "epoch": 2.1813901225319707,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004529178972092071,
      "loss": 5.9239,
      "mean_token_accuracy": 0.1503912553191185,
      "num_tokens": 44118985.0,
      "step": 20385
    },
    {
      "entropy": 6.325432205200196,
      "epoch": 2.1819251966397344,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045289449031429873,
      "loss": 5.931,
      "mean_token_accuracy": 0.15761362463235856,
      "num_tokens": 44129126.0,
      "step": 20390
    },
    {
      "entropy": 6.302253341674804,
      "epoch": 2.1824602707474985,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045287107828271184,
      "loss": 5.9372,
      "mean_token_accuracy": 0.1578105390071869,
      "num_tokens": 44139731.0,
      "step": 20395
    },
    {
      "entropy": 6.220752954483032,
      "epoch": 2.1829953448552626,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004528476611151226,
      "loss": 5.8954,
      "mean_token_accuracy": 0.15240855142474174,
      "num_tokens": 44149924.0,
      "step": 20400
    },
    {
      "entropy": 6.252826690673828,
      "epoch": 2.1835304189630262,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004528242388122071,
      "loss": 5.8889,
      "mean_token_accuracy": 0.15092860758304597,
      "num_tokens": 44161155.0,
      "step": 20405
    },
    {
      "entropy": 6.270920133590698,
      "epoch": 2.1840654930707903,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004528008113746417,
      "loss": 5.9625,
      "mean_token_accuracy": 0.15139070600271226,
      "num_tokens": 44171563.0,
      "step": 20410
    },
    {
      "entropy": 6.1941750049591064,
      "epoch": 2.1846005671785544,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045277737880310284,
      "loss": 5.8676,
      "mean_token_accuracy": 0.15505518168210983,
      "num_tokens": 44182719.0,
      "step": 20415
    },
    {
      "entropy": 6.243127965927124,
      "epoch": 2.185135641286318,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045275394109826715,
      "loss": 5.8617,
      "mean_token_accuracy": 0.16277254819869996,
      "num_tokens": 44192214.0,
      "step": 20420
    },
    {
      "entropy": 6.244515132904053,
      "epoch": 2.185670715394082,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045273049826081145,
      "loss": 5.9105,
      "mean_token_accuracy": 0.15095752328634263,
      "num_tokens": 44204074.0,
      "step": 20425
    },
    {
      "entropy": 6.335301637649536,
      "epoch": 2.186205789501846,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004527070502914125,
      "loss": 6.0401,
      "mean_token_accuracy": 0.14811511635780333,
      "num_tokens": 44214402.0,
      "step": 20430
    },
    {
      "entropy": 6.294680643081665,
      "epoch": 2.18674086360961,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004526835971907475,
      "loss": 5.9523,
      "mean_token_accuracy": 0.15318475514650345,
      "num_tokens": 44224479.0,
      "step": 20435
    },
    {
      "entropy": 6.27326192855835,
      "epoch": 2.187275937717374,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045266013895949356,
      "loss": 5.9485,
      "mean_token_accuracy": 0.1522960439324379,
      "num_tokens": 44234069.0,
      "step": 20440
    },
    {
      "entropy": 6.269653367996216,
      "epoch": 2.1878110118251377,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004526366755983281,
      "loss": 5.9383,
      "mean_token_accuracy": 0.14649854600429535,
      "num_tokens": 44244978.0,
      "step": 20445
    },
    {
      "entropy": 6.274523591995239,
      "epoch": 2.188346085932902,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004526132071079286,
      "loss": 5.9008,
      "mean_token_accuracy": 0.15897367000579835,
      "num_tokens": 44255657.0,
      "step": 20450
    },
    {
      "entropy": 6.214620113372803,
      "epoch": 2.1888811600406655,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045258973348897267,
      "loss": 5.8899,
      "mean_token_accuracy": 0.1540033519268036,
      "num_tokens": 44267370.0,
      "step": 20455
    },
    {
      "entropy": 6.250532054901123,
      "epoch": 2.1894162341484296,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004525662547421382,
      "loss": 5.9232,
      "mean_token_accuracy": 0.15353896021842955,
      "num_tokens": 44277855.0,
      "step": 20460
    },
    {
      "entropy": 6.294633626937866,
      "epoch": 2.1899513082561937,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000452542770868103,
      "loss": 5.894,
      "mean_token_accuracy": 0.15279827415943145,
      "num_tokens": 44288233.0,
      "step": 20465
    },
    {
      "entropy": 6.208767032623291,
      "epoch": 2.1904863823639573,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045251928186754533,
      "loss": 5.9401,
      "mean_token_accuracy": 0.1582665979862213,
      "num_tokens": 44298054.0,
      "step": 20470
    },
    {
      "entropy": 6.248278856277466,
      "epoch": 2.1910214564717214,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004524957877411432,
      "loss": 5.9045,
      "mean_token_accuracy": 0.15315870493650435,
      "num_tokens": 44309435.0,
      "step": 20475
    },
    {
      "entropy": 6.30474967956543,
      "epoch": 2.191556530579485,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004524722884895753,
      "loss": 5.9842,
      "mean_token_accuracy": 0.14808933883905412,
      "num_tokens": 44320502.0,
      "step": 20480
    },
    {
      "entropy": 6.213052272796631,
      "epoch": 2.192091604687249,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045244878411352,
      "loss": 5.8231,
      "mean_token_accuracy": 0.16402248293161392,
      "num_tokens": 44330911.0,
      "step": 20485
    },
    {
      "entropy": 6.279637479782105,
      "epoch": 2.1926266787950133,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004524252746136559,
      "loss": 5.9406,
      "mean_token_accuracy": 0.14923323839902877,
      "num_tokens": 44342159.0,
      "step": 20490
    },
    {
      "entropy": 6.289491271972656,
      "epoch": 2.193161752902777,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045240175999066194,
      "loss": 5.8894,
      "mean_token_accuracy": 0.15697774216532706,
      "num_tokens": 44352293.0,
      "step": 20495
    },
    {
      "entropy": 6.250111436843872,
      "epoch": 2.193696827010541,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045237824024521714,
      "loss": 6.0374,
      "mean_token_accuracy": 0.1523461326956749,
      "num_tokens": 44362321.0,
      "step": 20500
    },
    {
      "entropy": 6.258454465866089,
      "epoch": 2.1942319011183047,
      "grad_norm": 1.125,
      "learning_rate": 0.0004523547153780005,
      "loss": 5.9123,
      "mean_token_accuracy": 0.15294137075543404,
      "num_tokens": 44372682.0,
      "step": 20505
    },
    {
      "entropy": 6.255133628845215,
      "epoch": 2.194766975226069,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004523311853896915,
      "loss": 5.9251,
      "mean_token_accuracy": 0.15475166887044906,
      "num_tokens": 44383482.0,
      "step": 20510
    },
    {
      "entropy": 6.173567247390747,
      "epoch": 2.195302049333833,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045230765028096925,
      "loss": 5.7481,
      "mean_token_accuracy": 0.18020534962415696,
      "num_tokens": 44395333.0,
      "step": 20515
    },
    {
      "entropy": 6.307732105255127,
      "epoch": 2.1958371234415965,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00045228411005251366,
      "loss": 6.0087,
      "mean_token_accuracy": 0.14614347517490386,
      "num_tokens": 44406210.0,
      "step": 20520
    },
    {
      "entropy": 6.296582651138306,
      "epoch": 2.1963721975493606,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004522605647050043,
      "loss": 5.9545,
      "mean_token_accuracy": 0.15185257345438002,
      "num_tokens": 44415834.0,
      "step": 20525
    },
    {
      "entropy": 6.392317104339599,
      "epoch": 2.1969072716571247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045223701423912103,
      "loss": 6.0077,
      "mean_token_accuracy": 0.15015354752540588,
      "num_tokens": 44427902.0,
      "step": 20530
    },
    {
      "entropy": 6.264021158218384,
      "epoch": 2.1974423457648884,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045221345865554386,
      "loss": 5.8668,
      "mean_token_accuracy": 0.15717575550079346,
      "num_tokens": 44437406.0,
      "step": 20535
    },
    {
      "entropy": 6.245986270904541,
      "epoch": 2.1979774198726525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045218989795495294,
      "loss": 5.8698,
      "mean_token_accuracy": 0.15797783583402633,
      "num_tokens": 44446889.0,
      "step": 20540
    },
    {
      "entropy": 6.283173036575318,
      "epoch": 2.198512493980416,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045216633213802866,
      "loss": 5.9534,
      "mean_token_accuracy": 0.1504196658730507,
      "num_tokens": 44457132.0,
      "step": 20545
    },
    {
      "entropy": 6.226572513580322,
      "epoch": 2.1990475680881802,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045214276120545137,
      "loss": 5.9426,
      "mean_token_accuracy": 0.15298772901296614,
      "num_tokens": 44468006.0,
      "step": 20550
    },
    {
      "entropy": 6.264661121368408,
      "epoch": 2.199582642195944,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004521191851579019,
      "loss": 5.8528,
      "mean_token_accuracy": 0.15686771273612976,
      "num_tokens": 44478650.0,
      "step": 20555
    },
    {
      "entropy": 6.225648021697998,
      "epoch": 2.200117716303708,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004520956039960607,
      "loss": 5.9534,
      "mean_token_accuracy": 0.1563803657889366,
      "num_tokens": 44489842.0,
      "step": 20560
    },
    {
      "entropy": 6.271835422515869,
      "epoch": 2.200652790411472,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045207201772060887,
      "loss": 5.8851,
      "mean_token_accuracy": 0.15674896836280822,
      "num_tokens": 44500444.0,
      "step": 20565
    },
    {
      "entropy": 6.216269493103027,
      "epoch": 2.2011878645192358,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004520484263322274,
      "loss": 5.8358,
      "mean_token_accuracy": 0.15618252158164977,
      "num_tokens": 44510711.0,
      "step": 20570
    },
    {
      "entropy": 6.191167688369751,
      "epoch": 2.201722938627,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004520248298315975,
      "loss": 5.8869,
      "mean_token_accuracy": 0.15784385055303574,
      "num_tokens": 44522343.0,
      "step": 20575
    },
    {
      "entropy": 6.23126187324524,
      "epoch": 2.202258012734764,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045200122821940057,
      "loss": 5.9461,
      "mean_token_accuracy": 0.15253400802612305,
      "num_tokens": 44533218.0,
      "step": 20580
    },
    {
      "entropy": 6.31988000869751,
      "epoch": 2.2027930868425276,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045197762149631795,
      "loss": 5.9801,
      "mean_token_accuracy": 0.15307527780532837,
      "num_tokens": 44543406.0,
      "step": 20585
    },
    {
      "entropy": 6.321980667114258,
      "epoch": 2.2033281609502917,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045195400966303146,
      "loss": 6.0013,
      "mean_token_accuracy": 0.1594238869845867,
      "num_tokens": 44555329.0,
      "step": 20590
    },
    {
      "entropy": 6.338374519348145,
      "epoch": 2.2038632350580554,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004519303927202228,
      "loss": 6.0024,
      "mean_token_accuracy": 0.15369459390640258,
      "num_tokens": 44565191.0,
      "step": 20595
    },
    {
      "entropy": 6.232859420776367,
      "epoch": 2.2043983091658195,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045190677066857384,
      "loss": 5.8821,
      "mean_token_accuracy": 0.16618122160434723,
      "num_tokens": 44576374.0,
      "step": 20600
    },
    {
      "entropy": 6.2201110363006595,
      "epoch": 2.2049333832735836,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004518831435087668,
      "loss": 5.8653,
      "mean_token_accuracy": 0.15962664484977723,
      "num_tokens": 44586533.0,
      "step": 20605
    },
    {
      "entropy": 6.1693439960479735,
      "epoch": 2.2054684573813472,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004518595112414838,
      "loss": 5.7794,
      "mean_token_accuracy": 0.16701733618974685,
      "num_tokens": 44596753.0,
      "step": 20610
    },
    {
      "entropy": 6.336420726776123,
      "epoch": 2.2060035314891113,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004518358738674073,
      "loss": 6.0226,
      "mean_token_accuracy": 0.14653867483139038,
      "num_tokens": 44607632.0,
      "step": 20615
    },
    {
      "entropy": 6.298748540878296,
      "epoch": 2.206538605596875,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004518122313872198,
      "loss": 5.9288,
      "mean_token_accuracy": 0.15190549194812775,
      "num_tokens": 44618786.0,
      "step": 20620
    },
    {
      "entropy": 6.199263429641723,
      "epoch": 2.207073679704639,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045178858380160396,
      "loss": 5.9078,
      "mean_token_accuracy": 0.15118952095508575,
      "num_tokens": 44628760.0,
      "step": 20625
    },
    {
      "entropy": 6.309877538681031,
      "epoch": 2.207608753812403,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004517649311112426,
      "loss": 5.9728,
      "mean_token_accuracy": 0.15167373418807983,
      "num_tokens": 44641206.0,
      "step": 20630
    },
    {
      "entropy": 6.3541497707366945,
      "epoch": 2.208143827920167,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004517412733168187,
      "loss": 5.9151,
      "mean_token_accuracy": 0.15323900133371354,
      "num_tokens": 44651923.0,
      "step": 20635
    },
    {
      "entropy": 6.228582715988159,
      "epoch": 2.208678902027931,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045171761041901533,
      "loss": 5.8907,
      "mean_token_accuracy": 0.14997525066137313,
      "num_tokens": 44662375.0,
      "step": 20640
    },
    {
      "entropy": 6.097835636138916,
      "epoch": 2.2092139761356946,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004516939424185158,
      "loss": 5.819,
      "mean_token_accuracy": 0.16312852799892424,
      "num_tokens": 44673201.0,
      "step": 20645
    },
    {
      "entropy": 6.23745059967041,
      "epoch": 2.2097490502434587,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004516702693160035,
      "loss": 5.8822,
      "mean_token_accuracy": 0.15187629461288452,
      "num_tokens": 44684300.0,
      "step": 20650
    },
    {
      "entropy": 6.342277193069458,
      "epoch": 2.210284124351223,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004516465911121621,
      "loss": 5.9425,
      "mean_token_accuracy": 0.15136655420064926,
      "num_tokens": 44694603.0,
      "step": 20655
    },
    {
      "entropy": 6.234306478500367,
      "epoch": 2.2108191984589864,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004516229078076751,
      "loss": 5.9301,
      "mean_token_accuracy": 0.1570025146007538,
      "num_tokens": 44704461.0,
      "step": 20660
    },
    {
      "entropy": 6.258256816864014,
      "epoch": 2.2113542725667505,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004515992194032265,
      "loss": 5.9977,
      "mean_token_accuracy": 0.15310155004262924,
      "num_tokens": 44715784.0,
      "step": 20665
    },
    {
      "entropy": 6.253281450271606,
      "epoch": 2.211889346674514,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045157552589950023,
      "loss": 5.8647,
      "mean_token_accuracy": 0.16289864033460616,
      "num_tokens": 44726025.0,
      "step": 20670
    },
    {
      "entropy": 6.163765573501587,
      "epoch": 2.2124244207822783,
      "grad_norm": 10.125,
      "learning_rate": 0.0004515518272971805,
      "loss": 5.8329,
      "mean_token_accuracy": 0.1713068425655365,
      "num_tokens": 44736602.0,
      "step": 20675
    },
    {
      "entropy": 6.2825109481811525,
      "epoch": 2.2129594948900424,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004515281235969515,
      "loss": 6.0215,
      "mean_token_accuracy": 0.14633079916238784,
      "num_tokens": 44746260.0,
      "step": 20680
    },
    {
      "entropy": 6.278598070144653,
      "epoch": 2.213494568997806,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004515044147994978,
      "loss": 5.8933,
      "mean_token_accuracy": 0.15752835273742677,
      "num_tokens": 44756894.0,
      "step": 20685
    },
    {
      "entropy": 6.304586505889892,
      "epoch": 2.21402964310557,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004514807009055039,
      "loss": 5.9512,
      "mean_token_accuracy": 0.15352966487407685,
      "num_tokens": 44768483.0,
      "step": 20690
    },
    {
      "entropy": 6.228226804733277,
      "epoch": 2.2145647172133343,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045145698191565456,
      "loss": 5.8976,
      "mean_token_accuracy": 0.15638413429260253,
      "num_tokens": 44778843.0,
      "step": 20695
    },
    {
      "entropy": 6.25920524597168,
      "epoch": 2.215099791321098,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045143325783063466,
      "loss": 5.8428,
      "mean_token_accuracy": 0.15880079716444015,
      "num_tokens": 44788659.0,
      "step": 20700
    },
    {
      "entropy": 6.275933265686035,
      "epoch": 2.215634865428862,
      "grad_norm": 1.25,
      "learning_rate": 0.0004514095286511293,
      "loss": 5.9311,
      "mean_token_accuracy": 0.15536387264728546,
      "num_tokens": 44799730.0,
      "step": 20705
    },
    {
      "entropy": 6.248025274276733,
      "epoch": 2.2161699395366257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004513857943778235,
      "loss": 5.9031,
      "mean_token_accuracy": 0.1671956479549408,
      "num_tokens": 44811679.0,
      "step": 20710
    },
    {
      "entropy": 6.246401071548462,
      "epoch": 2.2167050136443898,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004513620550114027,
      "loss": 5.9255,
      "mean_token_accuracy": 0.15557463467121124,
      "num_tokens": 44821996.0,
      "step": 20715
    },
    {
      "entropy": 6.295328807830811,
      "epoch": 2.217240087752154,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045133831055255234,
      "loss": 5.9605,
      "mean_token_accuracy": 0.15378516539931297,
      "num_tokens": 44833713.0,
      "step": 20720
    },
    {
      "entropy": 6.2741443634033205,
      "epoch": 2.2177751618599175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004513145610019581,
      "loss": 5.9587,
      "mean_token_accuracy": 0.15700909346342087,
      "num_tokens": 44844751.0,
      "step": 20725
    },
    {
      "entropy": 6.247795104980469,
      "epoch": 2.2183102359676816,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045129080636030554,
      "loss": 5.9953,
      "mean_token_accuracy": 0.14833834767341614,
      "num_tokens": 44856089.0,
      "step": 20730
    },
    {
      "entropy": 6.289519453048706,
      "epoch": 2.2188453100754453,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004512670466282808,
      "loss": 5.9377,
      "mean_token_accuracy": 0.1532579243183136,
      "num_tokens": 44867313.0,
      "step": 20735
    },
    {
      "entropy": 6.330736303329468,
      "epoch": 2.2193803841832094,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045124328180656993,
      "loss": 5.9286,
      "mean_token_accuracy": 0.15127617865800858,
      "num_tokens": 44878430.0,
      "step": 20740
    },
    {
      "entropy": 6.249673366546631,
      "epoch": 2.2199154582909735,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045121951189585897,
      "loss": 5.8648,
      "mean_token_accuracy": 0.15719403177499772,
      "num_tokens": 44888746.0,
      "step": 20745
    },
    {
      "entropy": 6.238415145874024,
      "epoch": 2.220450532398737,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045119573689683445,
      "loss": 5.9931,
      "mean_token_accuracy": 0.1491900146007538,
      "num_tokens": 44899412.0,
      "step": 20750
    },
    {
      "entropy": 6.287767553329468,
      "epoch": 2.2209856065065012,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045117195681018273,
      "loss": 5.9073,
      "mean_token_accuracy": 0.1556428298354149,
      "num_tokens": 44910362.0,
      "step": 20755
    },
    {
      "entropy": 6.260907459259033,
      "epoch": 2.221520680614265,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004511481716365905,
      "loss": 5.8931,
      "mean_token_accuracy": 0.14954326078295707,
      "num_tokens": 44920552.0,
      "step": 20760
    },
    {
      "entropy": 6.214408493041992,
      "epoch": 2.222055754722029,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004511243813767446,
      "loss": 5.8965,
      "mean_token_accuracy": 0.14940935969352723,
      "num_tokens": 44932090.0,
      "step": 20765
    },
    {
      "entropy": 6.331368255615234,
      "epoch": 2.222590828829793,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000451100586031332,
      "loss": 5.9336,
      "mean_token_accuracy": 0.1518632084131241,
      "num_tokens": 44944122.0,
      "step": 20770
    },
    {
      "entropy": 6.314195013046264,
      "epoch": 2.2231259029375567,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045107678560103953,
      "loss": 5.9507,
      "mean_token_accuracy": 0.15628646910190583,
      "num_tokens": 44955104.0,
      "step": 20775
    },
    {
      "entropy": 6.231810760498047,
      "epoch": 2.223660977045321,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045105298008655485,
      "loss": 5.9159,
      "mean_token_accuracy": 0.15233564525842666,
      "num_tokens": 44965495.0,
      "step": 20780
    },
    {
      "entropy": 6.222903490066528,
      "epoch": 2.2241960511530845,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045102916948856493,
      "loss": 5.9716,
      "mean_token_accuracy": 0.14904370307922363,
      "num_tokens": 44976627.0,
      "step": 20785
    },
    {
      "entropy": 6.321107196807861,
      "epoch": 2.2247311252608486,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004510053538077576,
      "loss": 5.9967,
      "mean_token_accuracy": 0.15189435184001923,
      "num_tokens": 44987955.0,
      "step": 20790
    },
    {
      "entropy": 6.329780101776123,
      "epoch": 2.2252661993686127,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004509815330448203,
      "loss": 5.9546,
      "mean_token_accuracy": 0.1528179556131363,
      "num_tokens": 44998124.0,
      "step": 20795
    },
    {
      "entropy": 6.300026559829712,
      "epoch": 2.2258012734763764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000450957707200441,
      "loss": 5.9758,
      "mean_token_accuracy": 0.1455935411155224,
      "num_tokens": 45007881.0,
      "step": 20800
    },
    {
      "entropy": 6.25696063041687,
      "epoch": 2.2263363475841405,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004509338762753076,
      "loss": 5.9162,
      "mean_token_accuracy": 0.16213326752185822,
      "num_tokens": 45018994.0,
      "step": 20805
    },
    {
      "entropy": 6.2757312774658205,
      "epoch": 2.2268714216919046,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004509100402701083,
      "loss": 5.9841,
      "mean_token_accuracy": 0.15341741144657134,
      "num_tokens": 45030109.0,
      "step": 20810
    },
    {
      "entropy": 6.188795566558838,
      "epoch": 2.227406495799668,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045088619918553136,
      "loss": 5.8654,
      "mean_token_accuracy": 0.16525416001677512,
      "num_tokens": 45039872.0,
      "step": 20815
    },
    {
      "entropy": 6.320765113830566,
      "epoch": 2.2279415699074323,
      "grad_norm": 1.25,
      "learning_rate": 0.0004508623530222651,
      "loss": 5.8558,
      "mean_token_accuracy": 0.16555993109941483,
      "num_tokens": 45049216.0,
      "step": 20820
    },
    {
      "entropy": 6.2096985340118405,
      "epoch": 2.228476644015196,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045083850178099805,
      "loss": 5.8326,
      "mean_token_accuracy": 0.16490947157144548,
      "num_tokens": 45059191.0,
      "step": 20825
    },
    {
      "entropy": 6.188499879837036,
      "epoch": 2.22901171812296,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000450814645462419,
      "loss": 5.874,
      "mean_token_accuracy": 0.15842597782611847,
      "num_tokens": 45069212.0,
      "step": 20830
    },
    {
      "entropy": 6.237273693084717,
      "epoch": 2.229546792230724,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045079078406721676,
      "loss": 5.8856,
      "mean_token_accuracy": 0.15314507484436035,
      "num_tokens": 45080248.0,
      "step": 20835
    },
    {
      "entropy": 6.216241979598999,
      "epoch": 2.230081866338488,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045076691759608035,
      "loss": 5.8832,
      "mean_token_accuracy": 0.15058448761701584,
      "num_tokens": 45091315.0,
      "step": 20840
    },
    {
      "entropy": 6.225652551651001,
      "epoch": 2.230616940446252,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045074304604969886,
      "loss": 5.9112,
      "mean_token_accuracy": 0.15427898615598679,
      "num_tokens": 45102674.0,
      "step": 20845
    },
    {
      "entropy": 6.242578220367432,
      "epoch": 2.2311520145540156,
      "grad_norm": 1.125,
      "learning_rate": 0.00045071916942876163,
      "loss": 5.908,
      "mean_token_accuracy": 0.15498548597097397,
      "num_tokens": 45113171.0,
      "step": 20850
    },
    {
      "entropy": 6.2315081596374515,
      "epoch": 2.2316870886617797,
      "grad_norm": 1.125,
      "learning_rate": 0.0004506952877339581,
      "loss": 5.8852,
      "mean_token_accuracy": 0.1577208787202835,
      "num_tokens": 45122912.0,
      "step": 20855
    },
    {
      "entropy": 6.2684753894805905,
      "epoch": 2.2322221627695438,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004506714009659778,
      "loss": 5.9284,
      "mean_token_accuracy": 0.15912978053092958,
      "num_tokens": 45133839.0,
      "step": 20860
    },
    {
      "entropy": 6.213528251647949,
      "epoch": 2.2327572368773074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004506475091255105,
      "loss": 5.9031,
      "mean_token_accuracy": 0.1578981876373291,
      "num_tokens": 45144914.0,
      "step": 20865
    },
    {
      "entropy": 6.238722562789917,
      "epoch": 2.2332923109850715,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000450623612213246,
      "loss": 5.9133,
      "mean_token_accuracy": 0.16260692179203035,
      "num_tokens": 45155194.0,
      "step": 20870
    },
    {
      "entropy": 6.311830282211304,
      "epoch": 2.233827385092835,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045059971022987446,
      "loss": 6.0241,
      "mean_token_accuracy": 0.14504720270633698,
      "num_tokens": 45166771.0,
      "step": 20875
    },
    {
      "entropy": 6.304285144805908,
      "epoch": 2.2343624592005993,
      "grad_norm": 2.796875,
      "learning_rate": 0.00045057580317608585,
      "loss": 5.8682,
      "mean_token_accuracy": 0.15785759836435317,
      "num_tokens": 45177236.0,
      "step": 20880
    },
    {
      "entropy": 6.249019479751587,
      "epoch": 2.2348975333083634,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045055189105257064,
      "loss": 5.9035,
      "mean_token_accuracy": 0.15004339218139648,
      "num_tokens": 45187382.0,
      "step": 20885
    },
    {
      "entropy": 6.260206699371338,
      "epoch": 2.235432607416127,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045052797386001936,
      "loss": 5.9248,
      "mean_token_accuracy": 0.15695529878139497,
      "num_tokens": 45197287.0,
      "step": 20890
    },
    {
      "entropy": 6.29323558807373,
      "epoch": 2.235967681523891,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004505040515991224,
      "loss": 5.8926,
      "mean_token_accuracy": 0.1614371195435524,
      "num_tokens": 45208737.0,
      "step": 20895
    },
    {
      "entropy": 6.291822290420532,
      "epoch": 2.236502755631655,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045048012427057057,
      "loss": 6.0131,
      "mean_token_accuracy": 0.1502128705382347,
      "num_tokens": 45219586.0,
      "step": 20900
    },
    {
      "entropy": 6.314738750457764,
      "epoch": 2.237037829739419,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004504561918750549,
      "loss": 5.9951,
      "mean_token_accuracy": 0.14307852685451508,
      "num_tokens": 45230730.0,
      "step": 20905
    },
    {
      "entropy": 6.294840574264526,
      "epoch": 2.237572903847183,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004504322544132663,
      "loss": 5.8278,
      "mean_token_accuracy": 0.1692100301384926,
      "num_tokens": 45241330.0,
      "step": 20910
    },
    {
      "entropy": 6.188144159317017,
      "epoch": 2.2381079779549466,
      "grad_norm": 1.25,
      "learning_rate": 0.000450408311885896,
      "loss": 5.9445,
      "mean_token_accuracy": 0.15797356218099595,
      "num_tokens": 45251439.0,
      "step": 20915
    },
    {
      "entropy": 6.114798355102539,
      "epoch": 2.2386430520627107,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045038436429363537,
      "loss": 5.7445,
      "mean_token_accuracy": 0.16476468443870546,
      "num_tokens": 45262078.0,
      "step": 20920
    },
    {
      "entropy": 6.306900882720948,
      "epoch": 2.2391781261704744,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045036041163717586,
      "loss": 5.9492,
      "mean_token_accuracy": 0.15574443265795707,
      "num_tokens": 45272466.0,
      "step": 20925
    },
    {
      "entropy": 6.293915891647339,
      "epoch": 2.2397132002782385,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004503364539172091,
      "loss": 5.8696,
      "mean_token_accuracy": 0.16004620641469955,
      "num_tokens": 45281830.0,
      "step": 20930
    },
    {
      "entropy": 6.207993078231811,
      "epoch": 2.2402482743860026,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004503124911344268,
      "loss": 5.9302,
      "mean_token_accuracy": 0.15153199359774588,
      "num_tokens": 45293026.0,
      "step": 20935
    },
    {
      "entropy": 6.272617435455322,
      "epoch": 2.2407833484937663,
      "grad_norm": 1.125,
      "learning_rate": 0.00045028852328952106,
      "loss": 5.9264,
      "mean_token_accuracy": 0.157490760833025,
      "num_tokens": 45303532.0,
      "step": 20940
    },
    {
      "entropy": 6.183876895904541,
      "epoch": 2.2413184226015304,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004502645503831838,
      "loss": 5.7983,
      "mean_token_accuracy": 0.16371826082468033,
      "num_tokens": 45314448.0,
      "step": 20945
    },
    {
      "entropy": 6.232197332382202,
      "epoch": 2.241853496709294,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045024057241610727,
      "loss": 5.9395,
      "mean_token_accuracy": 0.1480267345905304,
      "num_tokens": 45323194.0,
      "step": 20950
    },
    {
      "entropy": 6.273339366912841,
      "epoch": 2.242388570817058,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004502165893889839,
      "loss": 5.8172,
      "mean_token_accuracy": 0.16930745542049408,
      "num_tokens": 45333464.0,
      "step": 20955
    },
    {
      "entropy": 6.219166898727417,
      "epoch": 2.242923644924822,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004501926013025061,
      "loss": 5.936,
      "mean_token_accuracy": 0.15483058840036393,
      "num_tokens": 45344364.0,
      "step": 20960
    },
    {
      "entropy": 6.27386965751648,
      "epoch": 2.243458719032586,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045016860815736646,
      "loss": 5.8626,
      "mean_token_accuracy": 0.15995221138000487,
      "num_tokens": 45355202.0,
      "step": 20965
    },
    {
      "entropy": 6.238163232803345,
      "epoch": 2.24399379314035,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004501446099542579,
      "loss": 5.8325,
      "mean_token_accuracy": 0.15494266003370286,
      "num_tokens": 45365696.0,
      "step": 20970
    },
    {
      "entropy": 6.1908587455749515,
      "epoch": 2.244528867248114,
      "grad_norm": 1.125,
      "learning_rate": 0.0004501206066938734,
      "loss": 5.8234,
      "mean_token_accuracy": 0.1693602755665779,
      "num_tokens": 45376609.0,
      "step": 20975
    },
    {
      "entropy": 6.298272514343262,
      "epoch": 2.2450639413558777,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045009659837690606,
      "loss": 5.9719,
      "mean_token_accuracy": 0.14289585649967193,
      "num_tokens": 45386158.0,
      "step": 20980
    },
    {
      "entropy": 6.336481666564941,
      "epoch": 2.245599015463642,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000450072585004049,
      "loss": 5.9701,
      "mean_token_accuracy": 0.15478792786598206,
      "num_tokens": 45396824.0,
      "step": 20985
    },
    {
      "entropy": 6.262418746948242,
      "epoch": 2.2461340895714055,
      "grad_norm": 1.125,
      "learning_rate": 0.0004500485665759955,
      "loss": 5.8964,
      "mean_token_accuracy": 0.15473867505788802,
      "num_tokens": 45407219.0,
      "step": 20990
    },
    {
      "entropy": 6.277280521392822,
      "epoch": 2.2466691636791696,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004500245430934394,
      "loss": 5.9669,
      "mean_token_accuracy": 0.14645546600222586,
      "num_tokens": 45418572.0,
      "step": 20995
    },
    {
      "entropy": 6.261141157150268,
      "epoch": 2.2472042377869337,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045000051455707413,
      "loss": 5.8646,
      "mean_token_accuracy": 0.15467487275600433,
      "num_tokens": 45429643.0,
      "step": 21000
    },
    {
      "epoch": 2.2472042377869337,
      "eval_entropy": 6.122672850745065,
      "eval_loss": 6.0745368003845215,
      "eval_mean_token_accuracy": 0.15561544772619104,
      "eval_num_tokens": 45429643.0,
      "eval_runtime": 22.966,
      "eval_samples_per_second": 1292.299,
      "eval_steps_per_second": 161.543,
      "step": 21000
    }
  ],
  "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": 6.6758952651264e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}